2 files changed, 59 insertions, 7 deletions

diff --git a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
index de715e4791e..26013cdb0b8 100644
--- a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -167,6 +167,7 @@ namespace {
 
     bool CombineToPreIndexedLoadStore(SDNode *N);
     bool CombineToPostIndexedLoadStore(SDNode *N);
+    SDValue SplitIndexingFromLoad(LoadSDNode *LD);
     bool SliceUpLoad(SDNode *N);
 
     void ReplaceLoadWithPromotedLoad(SDNode *Load, SDNode *ExtLoad);
@@ -761,10 +762,14 @@ CommitTargetLoweringOpt(const TargetLowering::TargetLoweringOpt &TLO) {
 
     // If the operands of this node are only used by the node, they will now
     // be dead.  Make sure to visit them first to delete dead nodes early.
-    for (unsigned i = 0, e = TLO.Old.getNode()->getNumOperands(); i != e; ++i)
-      if (TLO.Old.getNode()->getOperand(i).getNode()->hasOneUse())
-        AddToWorkList(TLO.Old.getNode()->getOperand(i).getNode());
-
+    for (unsigned i = 0, e = TLO.Old.getNode()->getNumOperands(); i != e; ++i) {
+      SDNode *Op = TLO.Old.getNode()->getOperand(i).getNode();
+      // For an operand generating multiple values, one of the values may
+      // become dead allowing further simplification (e.g. split index
+      // arithmetic from an indexed load).
+      if (Op->hasOneUse() || Op->getNumValues() > 1)
+        AddToWorkList(Op);
+    }
     DAG.DeleteNode(TLO.Old.getNode());
   }
 }
@@ -7844,6 +7849,17 @@ bool DAGCombiner::CombineToPostIndexedLoadStore(SDNode *N) {
   return false;
 }
 
+/// \brief Return the base-pointer arithmetic from an indexed \p LD.
+SDValue DAGCombiner::SplitIndexingFromLoad(LoadSDNode *LD) {
+  ISD::MemIndexedMode AM = LD->getAddressingMode();
+  assert(AM != ISD::UNINDEXED);
+  SDValue BP = LD->getOperand(1);
+  SDValue Inc = LD->getOperand(2);
+  unsigned Opc =
+      (AM == ISD::PRE_INC || AM == ISD::POST_INC ? ISD::ADD : ISD::SUB);
+  return DAG.getNode(Opc, SDLoc(LD), BP.getSimpleValueType(), BP, Inc);
+}
+
 SDValue DAGCombiner::visitLOAD(SDNode *N) {
   LoadSDNode *LD  = cast<LoadSDNode>(N);
   SDValue Chain = LD->getChain();
@@ -7880,8 +7896,16 @@ SDValue DAGCombiner::visitLOAD(SDNode *N) {
     } else {
       // Indexed loads.
       assert(N->getValueType(2) == MVT::Other && "Malformed indexed loads?");
-      if (!N->hasAnyUseOfValue(0) && !N->hasAnyUseOfValue(1)) {
+      if (!N->hasAnyUseOfValue(0)) {
         SDValue Undef = DAG.getUNDEF(N->getValueType(0));
+        SDValue Index;
+        if (N->hasAnyUseOfValue(1)) {
+          Index = SplitIndexingFromLoad(LD);
+          // Try to fold the base pointer arithmetic into subsequent loads and
+          // stores.
+          AddUsersToWorkList(N);
+        } else
+          Index = DAG.getUNDEF(N->getValueType(1));
         DEBUG(dbgs() << "\nReplacing.7 ";
               N->dump(&DAG);
               dbgs() << "\nWith: ";
@@ -7889,8 +7913,7 @@ SDValue DAGCombiner::visitLOAD(SDNode *N) {
               dbgs() << " and 2 other values\n");
         WorkListRemover DeadNodes(*this);
         DAG.ReplaceAllUsesOfValueWith(SDValue(N, 0), Undef);
-        DAG.ReplaceAllUsesOfValueWith(SDValue(N, 1),
-                                      DAG.getUNDEF(N->getValueType(1)));
+        DAG.ReplaceAllUsesOfValueWith(SDValue(N, 1), Index);
         DAG.ReplaceAllUsesOfValueWith(SDValue(N, 2), Chain);
         removeFromWorkList(N);
         DAG.DeleteNode(N);
diff --git a/llvm/test/CodeGen/ARM64/dagcombiner-dead-indexed-load.ll b/llvm/test/CodeGen/ARM64/dagcombiner-dead-indexed-load.ll
new file mode 100644
index 00000000000..2cf01357324
--- /dev/null
+++ b/llvm/test/CodeGen/ARM64/dagcombiner-dead-indexed-load.ll
@@ -0,0 +1,29 @@
+; RUN: llc -mcpu=cyclone < %s | FileCheck %s
+
+target datalayout = "e-i64:64-n32:64-S128"
+target triple = "arm64-apple-ios"
+
+%"struct.SU" = type { i32, %"struct.SU"*, i32*, i32, i32, %"struct.BO", i32, [5 x i8] }
+%"struct.BO" = type { %"struct.RE" }
+
+%"struct.RE" = type { i32, i32, i32, i32 }
+
+; This is a read-modify-write of some bifields combined into an i48.  It gets
+; legalized into i32 and i16 accesses.  Only a single store of zero to the low
+; i32 part should be live.
+
+; CHECK-LABEL: test:
+; CHECK-NOT: ldr
+; CHECK: str wzr
+; CHECK-NOT: str
+define void @test(%"struct.SU"* nocapture %su) {
+entry:
+  %r1 = getelementptr inbounds %"struct.SU"* %su, i64 1, i32 5
+  %r2 = bitcast %"struct.BO"* %r1 to i48*
+  %r3 = load i48* %r2, align 8
+  %r4 = and i48 %r3, -4294967296
+  %r5 = or i48 0, %r4
+  store i48 %r5, i48* %r2, align 8
+
+  ret void
+}