Fix UMULO support for 2x register width to allow the full

range without a libcall to a new mulo<mode> libcall
that we'd have to create.

Finishes the rest of rdar://9090077 and rdar://9210061


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133318 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 21 insertions, 0 deletions

diff --git a/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp b/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp
index 6797bece13..cbb8da8f18 100644
--- a/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp
+++ b/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp
@@ -2160,6 +2160,27 @@ void DAGTypeLegalizer::ExpandIntRes_XMULO(SDNode *N,
   const Type *PtrTy = PtrVT.getTypeForEVT(*DAG.getContext());
   DebugLoc dl = N->getDebugLoc();
 
+  // A divide for UMULO should be faster than a function call.
+  if (N->getOpcode() == ISD::UMULO) {
+    SDValue LHS = N->getOperand(0), RHS = N->getOperand(1);
+    DebugLoc DL = N->getDebugLoc();
+
+    SDValue MUL = DAG.getNode(ISD::MUL, DL, LHS.getValueType(), LHS, RHS);
+    SplitInteger(MUL, Lo, Hi);
+
+    // A divide for UMULO will be faster than a function call. Select to
+    // make sure we aren't using 0.
+    SDValue isZero = DAG.getSetCC(dl, TLI.getSetCCResultType(VT),
+				  RHS, DAG.getConstant(0, VT), ISD::SETNE);
+    SDValue NotZero = DAG.getNode(ISD::SELECT, dl, VT, isZero,
+				  DAG.getConstant(1, VT), RHS);
+    SDValue DIV = DAG.getNode(ISD::UDIV, DL, LHS.getValueType(), MUL, NotZero);
+    SDValue Overflow;
+    Overflow = DAG.getSetCC(DL, N->getValueType(1), DIV, LHS, ISD::SETNE);
+    ReplaceValueWith(SDValue(N, 1), Overflow);
+    return;
+  }
+
   // Replace this with a libcall that will check overflow.
   RTLIB::Libcall LC = RTLIB::UNKNOWN_LIBCALL;
   if (VT == MVT::i32)