Revert "[NFC][ScheduleDAG] Remove unused EntrySU SUnit"

This reverts commit 0345d88de654259ae90494bf9b015416e2cccacb.

Google internal backend uses EntrySU, we are looking into removing
dependency on it.

Differential Revision: https://reviews.llvm.org/D88018

include/llvm/CodeGen/MachineScheduler.h

@@ -234,7 +234,7 @@ public:
   virtual void leaveMBB() {};
 
   /// Notify this strategy that all roots have been released (including those
-  /// that depend on ExitSU).
+  /// that depend on EntrySU or ExitSU).
   virtual void registerRoots() {}
 
   /// Pick the next node to schedule, or return NULL. Set IsTopNode to true to

include/llvm/CodeGen/ScheduleDAG.h

@@ -560,6 +560,7 @@ class TargetRegisterInfo;
     MachineFunction &MF;                ///< Machine function
     MachineRegisterInfo &MRI;           ///< Virtual/real register map
     std::vector<SUnit> SUnits;          ///< The scheduling units.
+    SUnit EntrySU;                      ///< Special node for the region entry.
     SUnit ExitSU;                       ///< Special node for the region exit.
 
 #ifdef NDEBUG

lib/CodeGen/MachineScheduler.cpp

@@ -680,7 +680,7 @@ void ScheduleDAGMI::releasePred(SUnit *SU, SDep *PredEdge) {
     PredSU->BotReadyCycle = SU->BotReadyCycle + PredEdge->getLatency();
 
   --PredSU->NumSuccsLeft;
-  if (PredSU->NumSuccsLeft == 0)
+  if (PredSU->NumSuccsLeft == 0 && PredSU != &EntrySU)
     SchedImpl->releaseBottomNode(PredSU);
 }
 
@@ -853,7 +853,7 @@ void ScheduleDAGMI::initQueues(ArrayRef<SUnit*> TopRoots,
   NextClusterSucc = nullptr;
   NextClusterPred = nullptr;
 
-  // Release all DAG roots for scheduling, not including ExitSU.
+  // Release all DAG roots for scheduling, not including EntrySU/ExitSU.
   //
   // Nodes with unreleased weak edges can still be roots.
   // Release top roots in forward order.
@@ -867,6 +867,7 @@ void ScheduleDAGMI::initQueues(ArrayRef<SUnit*> TopRoots,
     SchedImpl->releaseBottomNode(*I);
   }
 
+  releaseSuccessors(&EntrySU);
   releasePredecessors(&ExitSU);
 
   SchedImpl->registerRoots();
@@ -1167,6 +1168,8 @@ void ScheduleDAGMILive::updatePressureDiffs(
 
 void ScheduleDAGMILive::dump() const {
 #if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+  if (EntrySU.getInstr() != nullptr)
+    dumpNodeAll(EntrySU);
   for (const SUnit &SU : SUnits) {
     dumpNodeAll(SU);
     if (ShouldTrackPressure) {

lib/CodeGen/MacroFusion.cpp

@@ -109,21 +109,23 @@ static bool fuseInstructionPair(ScheduleDAGInstrs &DAG, SUnit &FirstSU,
 
   // Make the FirstSU also dependent on the dependencies of the SecondSU to
   // prevent them from being scheduled between the FirstSU and the SecondSU.
-  for (const SDep &SI : SecondSU.Preds) {
-    SUnit *SU = SI.getSUnit();
-    if (SI.isWeak() || isHazard(SI) || &FirstSU == SU || FirstSU.isSucc(SU))
-      continue;
-    LLVM_DEBUG(dbgs() << "  Bind "; DAG.dumpNodeName(*SU); dbgs() << " - ";
-               DAG.dumpNodeName(FirstSU); dbgs() << '\n';);
-    DAG.addEdge(&FirstSU, SDep(SU, SDep::Artificial));
-  }
-  // ExitSU comes last by design, which acts like an implicit dependency
-  // between ExitSU and any bottom root in the graph. We should transfer
-  // this to FirstSU as well.
-  if (&SecondSU == &DAG.ExitSU) {
-    for (SUnit &SU : DAG.SUnits) {
-      if (SU.Succs.empty())
-        DAG.addEdge(&FirstSU, SDep(&SU, SDep::Artificial));
+  if (&FirstSU != &DAG.EntrySU) {
+    for (const SDep &SI : SecondSU.Preds) {
+      SUnit *SU = SI.getSUnit();
+      if (SI.isWeak() || isHazard(SI) || &FirstSU == SU || FirstSU.isSucc(SU))
+        continue;
+      LLVM_DEBUG(dbgs() << "  Bind "; DAG.dumpNodeName(*SU); dbgs() << " - ";
+                 DAG.dumpNodeName(FirstSU); dbgs() << '\n';);
+      DAG.addEdge(&FirstSU, SDep(SU, SDep::Artificial));
+    }
+    // ExitSU comes last by design, which acts like an implicit dependency
+    // between ExitSU and any bottom root in the graph. We should transfer
+    // this to FirstSU as well.
+    if (&SecondSU == &DAG.ExitSU) {
+      for (SUnit &SU : DAG.SUnits) {
+        if (SU.Succs.empty())
+          DAG.addEdge(&FirstSU, SDep(&SU, SDep::Artificial));
+      }
     }
   }
 

lib/CodeGen/PostRASchedulerList.cpp

@@ -527,6 +527,9 @@ void SchedulePostRATDList::ListScheduleTopDown() {
   // blocks are a single region).
   HazardRec->Reset();
 
+  // Release any successors of the special Entry node.
+  ReleaseSuccessors(&EntrySU);
+
   // Add all leaves to Available queue.
   for (unsigned i = 0, e = SUnits.size(); i != e; ++i) {
     // It is available if it has no predecessors.

lib/CodeGen/ScheduleDAG.cpp

@@ -63,6 +63,7 @@ ScheduleDAG::~ScheduleDAG() = default;
 
 void ScheduleDAG::clearDAG() {
   SUnits.clear();
+  EntrySU = SUnit();
   ExitSU = SUnit();
 }
 
@@ -351,7 +352,9 @@ LLVM_DUMP_METHOD void SUnit::dumpAttributes() const {
 }
 
 LLVM_DUMP_METHOD void ScheduleDAG::dumpNodeName(const SUnit &SU) const {
-  if (&SU == &ExitSU)
+  if (&SU == &EntrySU)
+    dbgs() << "EntrySU";
+  else if (&SU == &ExitSU)
     dbgs() << "ExitSU";
   else
     dbgs() << "SU(" << SU.NodeNum << ")";
@@ -653,7 +656,7 @@ std::vector<int> ScheduleDAGTopologicalSort::GetSubGraph(const SUnit &StartSU,
     for (int I = SU->Preds.size()-1; I >= 0; --I) {
       const SUnit *Pred = SU->Preds[I].getSUnit();
       unsigned s = Pred->NodeNum;
-      // Edges to non-SUnits are allowed but ignored (e.g. ExitSU).
+      // Edges to non-SUnits are allowed but ignored (e.g. EntrySU).
       if (Pred->isBoundaryNode())
         continue;
       if (Node2Index[s] == LowerBound) {

lib/CodeGen/ScheduleDAGInstrs.cpp

@@ -1167,6 +1167,8 @@ void ScheduleDAGInstrs::dumpNode(const SUnit &SU) const {
 
 void ScheduleDAGInstrs::dump() const {
 #if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+  if (EntrySU.getInstr() != nullptr)
+    dumpNodeAll(EntrySU);
   for (const SUnit &SU : SUnits)
     dumpNodeAll(SU);
   if (ExitSU.getInstr() != nullptr)
@@ -1177,7 +1179,9 @@ void ScheduleDAGInstrs::dump() const {
 std::string ScheduleDAGInstrs::getGraphNodeLabel(const SUnit *SU) const {
   std::string s;
   raw_string_ostream oss(s);
-  if (SU == &ExitSU)
+  if (SU == &EntrySU)
+    oss << "<entry>";
+  else if (SU == &ExitSU)
     oss << "<exit>";
   else
     SU->getInstr()->print(oss, /*IsStandalone=*/true);

lib/CodeGen/SelectionDAG/ScheduleDAGFast.cpp

@@ -150,8 +150,8 @@ void ScheduleDAGFast::ReleasePred(SUnit *SU, SDep *PredEdge) {
   --PredSU->NumSuccsLeft;
 
   // If all the node's successors are scheduled, this node is ready
-  // to be scheduled.
-  if (PredSU->NumSuccsLeft == 0) {
+  // to be scheduled. Ignore the special EntrySU node.
+  if (PredSU->NumSuccsLeft == 0 && PredSU != &EntrySU) {
     PredSU->isAvailable = true;
     AvailableQueue.push(PredSU);
   }

lib/CodeGen/SelectionDAG/ScheduleDAGRRList.cpp

@@ -415,8 +415,8 @@ void ScheduleDAGRRList::ReleasePred(SUnit *SU, const SDep *PredEdge) {
   }
 
   // If all the node's successors are scheduled, this node is ready
-  // to be scheduled.
-  if (PredSU->NumSuccsLeft == 0) {
+  // to be scheduled. Ignore the special EntrySU node.
+  if (PredSU->NumSuccsLeft == 0 && PredSU != &EntrySU) {
     PredSU->isAvailable = true;
 
     unsigned Height = PredSU->getHeight();

lib/CodeGen/SelectionDAG/ScheduleDAGSDNodes.cpp

@@ -696,6 +696,8 @@ void ScheduleDAGSDNodes::dumpNode(const SUnit &SU) const {
 
 void ScheduleDAGSDNodes::dump() const {
 #if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+  if (EntrySU.getNode() != nullptr)
+    dumpNodeAll(EntrySU);
   for (const SUnit &SU : SUnits)
     dumpNodeAll(SU);
   if (ExitSU.getNode() != nullptr)

lib/CodeGen/SelectionDAG/ScheduleDAGVLIW.cpp

@@ -166,6 +166,9 @@ void ScheduleDAGVLIW::scheduleNodeTopDown(SUnit *SU, unsigned CurCycle) {
 void ScheduleDAGVLIW::listScheduleTopDown() {
   unsigned CurCycle = 0;
 
+  // Release any successors of the special Entry node.
+  releaseSuccessors(&EntrySU);
+
   // All leaves to AvailableQueue.
   for (unsigned i = 0, e = SUnits.size(); i != e; ++i) {
     // It is available if it has no predecessors.

lib/Target/AMDGPU/GCNMinRegStrategy.cpp

@@ -248,6 +248,7 @@ GCNMinRegScheduler::schedule(ArrayRef<const SUnit*> TopRoots,
   for (auto SU : TopRoots) {
     RQ.push_back(*new (Alloc.Allocate()) Candidate(SU, StepNo));
   }
+  releaseSuccessors(&DAG.EntrySU, StepNo);
 
   while (!RQ.empty()) {
     LLVM_DEBUG(dbgs() << "\n=== Picking candidate, Step = " << StepNo

lib/Target/AMDGPU/SIMachineScheduler.h

@@ -455,6 +455,7 @@ public:
   MachineRegisterInfo *getMRI() { return &MRI; }
   const TargetRegisterInfo *getTRI() { return TRI; }
   ScheduleDAGTopologicalSort *GetTopo() { return &Topo; }
+  SUnit &getEntrySU() { return EntrySU; }
   SUnit& getExitSU() { return ExitSU; }
 
   void restoreSULinksLeft();