Fix some issues in WalkChainUsers dealing with

CopyToReg/CopyFromReg/INLINEASM.  These are annoying because
they have the same opcode before an after isel.  Fix this by
setting their NodeID to -1 to indicate that they are selected,
just like what automatically happens when selecting things that
end up being machine nodes.

With that done, give IsLegalToFold a new flag that causes it to
ignore chains.  This lets the HandleMergeInputChains routine be
the one place that validates chains after a match is successful,
enabling the new hotness in chain processing.  This smarter
chain processing eliminates the need for "PreprocessRMW" in the
X86 and MSP430 backends and enables MSP to start matching it's
multiple mem operand instructions more aggressively.

I currently #if out the dead code in the X86 backend and MSP 
backend, I'll remove it for real in a follow-on patch.

The testcase changes are:
  test/CodeGen/X86/sse3.ll: we generate better code
  test/CodeGen/X86/store_op_load_fold2.ll: PreprocessRMW was 
      miscompiling this before, we now generate correct code
      Convert it to filecheck while I'm at it.
  test/CodeGen/MSP430/Inst16mm.ll: Add a testcase for mem/mem
      folding to make anton happy. :)

llvm-svn: 97596

3 files changed, 39 insertions, 56 deletions

diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp
index ff8ab466d5f..8e88f16d699 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp
@@ -1376,8 +1376,8 @@ static SDNode *findFlagUse(SDNode *N) {
 /// This function recursively traverses up the operand chain, ignoring
 /// certain nodes.
 static bool findNonImmUse(SDNode *Use, SDNode* Def, SDNode *ImmedUse,
-                          SDNode *Root,
-                          SmallPtrSet<SDNode*, 16> &Visited) {
+                          SDNode *Root, SmallPtrSet<SDNode*, 16> &Visited,
+                          bool IgnoreChains) {
   // The NodeID's are given uniques ID's where a node ID is guaranteed to be
   // greater than all of its (recursive) operands.  If we scan to a point where
   // 'use' is smaller than the node we're scanning for, then we know we will
@@ -1395,6 +1395,10 @@ static bool findNonImmUse(SDNode *Use, SDNode* Def, SDNode *ImmedUse,
     return false;
 
   for (unsigned i = 0, e = Use->getNumOperands(); i != e; ++i) {
+    // Ignore chain uses, they are validated by HandleMergeInputChains.
+    if (Use->getOperand(i).getValueType() == MVT::Other && IgnoreChains)
+      continue;
+    
     SDNode *N = Use->getOperand(i).getNode();
     if (N == Def) {
       if (Use == ImmedUse || Use == Root)
@@ -1404,7 +1408,7 @@ static bool findNonImmUse(SDNode *Use, SDNode* Def, SDNode *ImmedUse,
     }
 
     // Traverse up the operand chain.
-    if (findNonImmUse(N, Def, ImmedUse, Root, Visited))
+    if (findNonImmUse(N, Def, ImmedUse, Root, Visited, IgnoreChains))
       return true;
   }
   return false;
@@ -1419,9 +1423,10 @@ static bool findNonImmUse(SDNode *Use, SDNode* Def, SDNode *ImmedUse,
 /// have one non-chain use, we only need to watch out for load/op/store
 /// and load/op/cmp case where the root (store / cmp) may reach the load via
 /// its chain operand.
-static inline bool isNonImmUse(SDNode *Root, SDNode *Def, SDNode *ImmedUse) {
+static inline bool isNonImmUse(SDNode *Root, SDNode *Def, SDNode *ImmedUse,
+                               bool IgnoreChains) {
   SmallPtrSet<SDNode*, 16> Visited;
-  return findNonImmUse(Root, Def, ImmedUse, Root, Visited);
+  return findNonImmUse(Root, Def, ImmedUse, Root, Visited, IgnoreChains);
 }
 
 /// IsProfitableToFold - Returns true if it's profitable to fold the specific
@@ -1434,7 +1439,8 @@ bool SelectionDAGISel::IsProfitableToFold(SDValue N, SDNode *U,
 
 /// IsLegalToFold - Returns true if the specific operand node N of
 /// U can be folded during instruction selection that starts at Root.
-bool SelectionDAGISel::IsLegalToFold(SDValue N, SDNode *U, SDNode *Root) const {
+bool SelectionDAGISel::IsLegalToFold(SDValue N, SDNode *U, SDNode *Root,
+                                     bool IgnoreChains) const {
   if (OptLevel == CodeGenOpt::None) return false;
 
   // If Root use can somehow reach N through a path that that doesn't contain
@@ -1488,7 +1494,7 @@ bool SelectionDAGISel::IsLegalToFold(SDValue N, SDNode *U, SDNode *Root) const {
     VT = Root->getValueType(Root->getNumValues()-1);
   }
 
-  return !isNonImmUse(Root, N.getNode(), U);
+  return !isNonImmUse(Root, N.getNode(), U, IgnoreChains);
 }
 
 SDNode *SelectionDAGISel::Select_INLINEASM(SDNode *N) {
@@ -1500,6 +1506,7 @@ SDNode *SelectionDAGISel::Select_INLINEASM(SDNode *N) {
   VTs.push_back(MVT::Flag);
   SDValue New = CurDAG->getNode(ISD::INLINEASM, N->getDebugLoc(),
                                 VTs, &Ops[0], Ops.size());
+  New->setNodeId(-1);
   return New.getNode();
 }
 
@@ -1636,11 +1643,17 @@ WalkChainUsers(SDNode *ChainedNode,
     // pattern that we're selecting down into the already selected chunk of the
     // DAG.
     if (User->isMachineOpcode() ||
-        User->getOpcode() == ISD::CopyToReg ||
-        User->getOpcode() == ISD::CopyFromReg ||
-        User->getOpcode() == ISD::INLINEASM ||
         User->getOpcode() == ISD::HANDLENODE)  // Root of the graph.
       continue;
+    
+    if (User->getOpcode() == ISD::CopyToReg ||
+        User->getOpcode() == ISD::CopyFromReg ||
+        User->getOpcode() == ISD::INLINEASM) {
+      // If their node ID got reset to -1 then they've already been selected.
+      // Treat them like a MachineOpcode.
+      if (User->getNodeId() == -1)
+        continue;
+    }
 
     // If we have a TokenFactor, we handle it specially.
     if (User->getOpcode() != ISD::TokenFactor) {
@@ -1876,6 +1889,7 @@ SelectCodeCommon(SDNode *NodeToMatch, const unsigned char *MatcherTable,
   case ISD::TokenFactor:
   case ISD::CopyFromReg:
   case ISD::CopyToReg:
+    NodeToMatch->setNodeId(-1); // Mark selected.
     return 0;
   case ISD::AssertSext:
   case ISD::AssertZext:
@@ -2172,7 +2186,7 @@ SelectCodeCommon(SDNode *NodeToMatch, const unsigned char *MatcherTable,
       if (!IsProfitableToFold(N, NodeStack[NodeStack.size()-2].getNode(),
                               NodeToMatch) ||
           !IsLegalToFold(N, NodeStack[NodeStack.size()-2].getNode(),
-                         NodeToMatch))
+                         NodeToMatch, true/*We validate our own chains*/))
         break;
       
       continue;
diff --git a/llvm/lib/Target/MSP430/MSP430ISelDAGToDAG.cpp b/llvm/lib/Target/MSP430/MSP430ISelDAGToDAG.cpp
index 54062a00f02..2ca184e5eaf 100644
--- a/llvm/lib/Target/MSP430/MSP430ISelDAGToDAG.cpp
+++ b/llvm/lib/Target/MSP430/MSP430ISelDAGToDAG.cpp
@@ -125,7 +125,9 @@ namespace {
     bool MatchWrapper(SDValue N, MSP430ISelAddressMode &AM);
     bool MatchAddressBase(SDValue N, MSP430ISelAddressMode &AM);
 
+#if 0
     bool IsLegalToFold(SDValue N, SDNode *U, SDNode *Root) const;
+#endif
 
     virtual bool
     SelectInlineAsmMemoryOperand(const SDValue &Op, char ConstraintCode,
@@ -323,6 +325,7 @@ SelectInlineAsmMemoryOperand(const SDValue &Op, char ConstraintCode,
   return false;
 }
 
+#if 0
 bool MSP430DAGToDAGISel::IsLegalToFold(SDValue N, SDNode *U,
                                        SDNode *Root) const {
   if (OptLevel == CodeGenOpt::None) return false;
@@ -357,6 +360,7 @@ bool MSP430DAGToDAGISel::IsLegalToFold(SDValue N, SDNode *U,
   // Proceed to 'generic' cycle finder code
   return SelectionDAGISel::IsLegalToFold(N, U, Root);
 }
+#endif
 
 
 /// MoveBelowTokenFactor - Replace TokenFactor operand with load's chain operand
@@ -516,6 +520,7 @@ static bool isRMWLoad(SDValue N, SDValue Chain, SDValue Address,
 /// This allows selection of mem-mem instructions. Yay!
 
 void MSP430DAGToDAGISel::PreprocessForRMW() {
+  return;
   for (SelectionDAG::allnodes_iterator I = CurDAG->allnodes_begin(),
          E = CurDAG->allnodes_end(); I != E; ++I) {
     if (!ISD::isNON_TRUNCStore(I))
diff --git a/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp b/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
index e162f4e4175..5ec50650b6d 100644
--- a/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
+++ b/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
@@ -467,46 +467,6 @@ static bool isCalleeLoad(SDValue Callee, SDValue &Chain) {
 }
 
 
-/// PreprocessForRMW - Preprocess the DAG to make instruction selection better.
-/// This is only run if not in -O0 mode.
-/// This allows the instruction selector to pick more read-modify-write
-/// instructions. This is a common case:
-///
-///     [Load chain]
-///         ^
-///         |
-///       [Load]
-///       ^    ^
-///       |    |
-///      /      \-
-///     /         |
-/// [TokenFactor] [Op]
-///     ^          ^
-///     |          |
-///      \        /
-///       \      /
-///       [Store]
-///
-/// The fact the store's chain operand != load's chain will prevent the
-/// (store (op (load))) instruction from being selected. We can transform it to:
-///
-///     [Load chain]
-///         ^
-///         |
-///    [TokenFactor]
-///         ^
-///         |
-///       [Load]
-///       ^    ^
-///       |    |
-///       |     \- 
-///       |       | 
-///       |     [Op]
-///       |       ^
-///       |       |
-///       \      /
-///        \    /
-///       [Store]
 void X86DAGToDAGISel::PreprocessForRMW() {
   for (SelectionDAG::allnodes_iterator I = CurDAG->allnodes_begin(),
          E = CurDAG->allnodes_end(); I != E; ++I) {
@@ -538,6 +498,9 @@ void X86DAGToDAGISel::PreprocessForRMW() {
       ++NumLoadMoved;
       continue;
     }
+    
+    continue;
+    
 
     if (!ISD::isNON_TRUNCStore(I))
       continue;
@@ -1415,11 +1378,12 @@ bool X86DAGToDAGISel::TryFoldLoad(SDNode *P, SDValue N,
                                   SDValue &Base, SDValue &Scale,
                                   SDValue &Index, SDValue &Disp,
                                   SDValue &Segment) {
-  if (ISD::isNON_EXTLoad(N.getNode()) &&
-      IsProfitableToFold(N, P, P) &&
-      IsLegalToFold(N, P, P))
-    return SelectAddr(P, N.getOperand(1), Base, Scale, Index, Disp, Segment);
-  return false;
+  if (!ISD::isNON_EXTLoad(N.getNode()) ||
+      !IsProfitableToFold(N, P, P) ||
+      !IsLegalToFold(N, P, P))
+    return false;
+  
+  return SelectAddr(P, N.getOperand(1), Base, Scale, Index, Disp, Segment);
 }
 
 /// getGlobalBaseReg - Return an SDNode that returns the value of