[DagCombiner] Allow shuffles to merge through bitcasts

Currently shuffles may only be combined if they are of the same type, despite the fact that bitcasts are often introduced in between shuffle nodes (e.g. x86 shuffle type widening).

This patch allows a single input shuffle to peek through bitcasts and if the input is another shuffle will merge them, shuffling using the smallest sized type, and re-applying the bitcasts at the inputs and output instead.

Dropped old ShuffleToZext test - this patch removes the use of the zext and vector-zext.ll covers these anyhow.

Differential Revision: http://reviews.llvm.org/D7939

llvm-svn: 231380

1 files changed, 83 insertions, 0 deletions

diff --git a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
index 4a074a03d9a..4cf34346184 100644
--- a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -11877,6 +11877,89 @@ SDValue DAGCombiner::visitVECTOR_SHUFFLE(SDNode *N) {
       return V;
   }
 
+  // If this shuffle only has a single input that is a bitcasted shuffle,
+  // attempt to merge the 2 shuffles and suitably bitcast the inputs/output
+  // back to their original types.
+  if (N0.getOpcode() == ISD::BITCAST && N0.hasOneUse() &&
+      N1.getOpcode() == ISD::UNDEF && Level < AfterLegalizeVectorOps &&
+      TLI.isTypeLegal(VT)) {
+
+    // Peek through the bitcast only if there is one user.
+    SDValue BC0 = N0;
+    while (BC0.getOpcode() == ISD::BITCAST) {
+      if (!BC0.hasOneUse())
+        break;
+      BC0 = BC0.getOperand(0);
+    }
+
+    auto ScaleShuffleMask = [](ArrayRef<int> Mask, int Scale) {
+      if (Scale == 1)
+        return SmallVector<int, 8>(Mask.begin(), Mask.end());
+
+      SmallVector<int, 8> NewMask;
+      for (int M : Mask)
+        for (int s = 0; s != Scale; ++s)
+          NewMask.push_back(M < 0 ? -1 : Scale * M + s);
+      return NewMask;
+    };
+
+    if (BC0.getOpcode() == ISD::VECTOR_SHUFFLE && BC0.hasOneUse()) {
+      EVT SVT = VT.getScalarType();
+      EVT InnerVT = BC0->getValueType(0);
+      EVT InnerSVT = InnerVT.getScalarType();
+
+      // Determine which shuffle works with the smaller scalar type.
+      EVT ScaleVT = SVT.bitsLT(InnerSVT) ? VT : InnerVT;
+      EVT ScaleSVT = ScaleVT.getScalarType();
+
+      if (TLI.isTypeLegal(ScaleVT) &&
+          0 == (InnerSVT.getSizeInBits() % ScaleSVT.getSizeInBits()) &&
+          0 == (SVT.getSizeInBits() % ScaleSVT.getSizeInBits())) {
+
+        int InnerScale = InnerSVT.getSizeInBits() / ScaleSVT.getSizeInBits();
+        int OuterScale = SVT.getSizeInBits() / ScaleSVT.getSizeInBits();
+
+        // Scale the shuffle masks to the smaller scalar type.
+        ShuffleVectorSDNode *InnerSVN = cast<ShuffleVectorSDNode>(BC0);
+        SmallVector<int, 8> InnerMask =
+            ScaleShuffleMask(InnerSVN->getMask(), InnerScale);
+        SmallVector<int, 8> OuterMask =
+            ScaleShuffleMask(SVN->getMask(), OuterScale);
+
+        // Merge the shuffle masks.
+        SmallVector<int, 8> NewMask;
+        for (int M : OuterMask)
+          NewMask.push_back(M < 0 ? -1 : InnerMask[M]);
+
+        // Test for shuffle mask legality over both commutations.
+        SDValue SV0 = BC0->getOperand(0);
+        SDValue SV1 = BC0->getOperand(1);
+        bool LegalMask = TLI.isShuffleMaskLegal(NewMask, ScaleVT);
+        if (!LegalMask) {
+          for (int i = 0, e = (int)NewMask.size(); i != e; ++i) {
+            int idx = NewMask[i];
+            if (idx < 0)
+              continue;
+            else if (idx < e)
+              NewMask[i] = idx + e;
+            else
+              NewMask[i] = idx - e;
+          }
+          std::swap(SV0, SV1);
+          LegalMask = TLI.isShuffleMaskLegal(NewMask, ScaleVT);
+        }
+
+        if (LegalMask) {
+          SV0 = DAG.getNode(ISD::BITCAST, SDLoc(N), ScaleVT, SV0);
+          SV1 = DAG.getNode(ISD::BITCAST, SDLoc(N), ScaleVT, SV1);
+          return DAG.getNode(
+              ISD::BITCAST, SDLoc(N), VT,
+              DAG.getVectorShuffle(ScaleVT, SDLoc(N), SV0, SV1, NewMask));
+        }
+      }
+    }
+  }
+
   // Canonicalize shuffles according to rules:
   //  shuffle(A, shuffle(A, B)) -> shuffle(shuffle(A,B), A)
   //  shuffle(B, shuffle(A, B)) -> shuffle(shuffle(A,B), B)