MI-Sched: schedule physreg copies.

The register allocator expects minimal physreg live ranges. Schedule
physreg copies accordingly. This is slightly tricky when they occur in
the middle of the scheduling region. For now, this is handled by
rescheduling the copy when its associated instruction is
scheduled. Eventually we may instead bundle them, but only if we can
preserve the bundles as parallel copies during regalloc.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179449 91177308-0d34-0410-b5e6-96231b3b80d8

5 files changed, 141 insertions, 14 deletions

diff --git a/include/llvm/CodeGen/MachineScheduler.h b/include/llvm/CodeGen/MachineScheduler.h
index 57febe7746..99cbd870ec 100644
--- a/include/llvm/CodeGen/MachineScheduler.h
+++ b/include/llvm/CodeGen/MachineScheduler.h
@@ -297,6 +297,10 @@ public:
   /// reorderable instructions.
   virtual void schedule();
 
+  /// Change the position of an instruction within the basic block and update
+  /// live ranges and region boundary iterators.
+  void moveInstruction(MachineInstr *MI, MachineBasicBlock::iterator InsertPos);
+
   /// Get current register pressure for the top scheduled instructions.
   const IntervalPressure &getTopPressure() const { return TopPressure; }
   const RegPressureTracker &getTopRPTracker() const { return TopRPTracker; }
@@ -362,7 +366,6 @@ protected:
 
   void updateScheduledPressure(const std::vector<unsigned> &NewMaxPressure);
 
-  void moveInstruction(MachineInstr *MI, MachineBasicBlock::iterator InsertPos);
   bool checkSchedLimit();
 
   void findRootsAndBiasEdges(SmallVectorImpl<SUnit*> &TopRoots,
diff --git a/include/llvm/CodeGen/ScheduleDAG.h b/include/llvm/CodeGen/ScheduleDAG.h
index 8c959da696..e13636c250 100644
--- a/include/llvm/CodeGen/ScheduleDAG.h
+++ b/include/llvm/CodeGen/ScheduleDAG.h
@@ -302,6 +302,7 @@ namespace llvm {
     bool isCallOp         : 1;          // Is a function call operand.
     bool isTwoAddress     : 1;          // Is a two-address instruction.
     bool isCommutable     : 1;          // Is a commutable instruction.
+    bool hasPhysRegUses   : 1;          // Has physreg uses.
     bool hasPhysRegDefs   : 1;          // Has physreg defs that are being used.
     bool hasPhysRegClobbers : 1;        // Has any physreg defs, used or not.
     bool isPending        : 1;          // True once pending.
@@ -331,10 +332,10 @@ namespace llvm {
         NodeQueueId(0), NumPreds(0), NumSuccs(0), NumPredsLeft(0),
         NumSuccsLeft(0), WeakPredsLeft(0), WeakSuccsLeft(0), NumRegDefsLeft(0),
         Latency(0), isVRegCycle(false), isCall(false), isCallOp(false),
-        isTwoAddress(false), isCommutable(false), hasPhysRegDefs(false),
-        hasPhysRegClobbers(false), isPending(false), isAvailable(false),
-        isScheduled(false), isScheduleHigh(false), isScheduleLow(false),
-        isCloned(false), SchedulingPref(Sched::None),
+        isTwoAddress(false), isCommutable(false), hasPhysRegUses(false),
+        hasPhysRegDefs(false), hasPhysRegClobbers(false), isPending(false),
+        isAvailable(false), isScheduled(false), isScheduleHigh(false),
+        isScheduleLow(false), isCloned(false), SchedulingPref(Sched::None),
         isDepthCurrent(false), isHeightCurrent(false), Depth(0), Height(0),
         TopReadyCycle(0), BotReadyCycle(0), CopyDstRC(NULL), CopySrcRC(NULL) {}
 
@@ -345,10 +346,10 @@ namespace llvm {
         NodeQueueId(0), NumPreds(0), NumSuccs(0), NumPredsLeft(0),
         NumSuccsLeft(0), WeakPredsLeft(0), WeakSuccsLeft(0), NumRegDefsLeft(0),
         Latency(0), isVRegCycle(false), isCall(false), isCallOp(false),
-        isTwoAddress(false), isCommutable(false), hasPhysRegDefs(false),
-        hasPhysRegClobbers(false), isPending(false), isAvailable(false),
-        isScheduled(false), isScheduleHigh(false), isScheduleLow(false),
-        isCloned(false), SchedulingPref(Sched::None),
+        isTwoAddress(false), isCommutable(false), hasPhysRegUses(false),
+        hasPhysRegDefs(false), hasPhysRegClobbers(false), isPending(false),
+        isAvailable(false), isScheduled(false), isScheduleHigh(false),
+        isScheduleLow(false), isCloned(false), SchedulingPref(Sched::None),
         isDepthCurrent(false), isHeightCurrent(false), Depth(0), Height(0),
         TopReadyCycle(0), BotReadyCycle(0), CopyDstRC(NULL), CopySrcRC(NULL) {}
 
@@ -358,10 +359,10 @@ namespace llvm {
         NodeQueueId(0), NumPreds(0), NumSuccs(0), NumPredsLeft(0),
         NumSuccsLeft(0), WeakPredsLeft(0), WeakSuccsLeft(0), NumRegDefsLeft(0),
         Latency(0), isVRegCycle(false), isCall(false), isCallOp(false),
-        isTwoAddress(false), isCommutable(false), hasPhysRegDefs(false),
-        hasPhysRegClobbers(false), isPending(false), isAvailable(false),
-        isScheduled(false), isScheduleHigh(false), isScheduleLow(false),
-        isCloned(false), SchedulingPref(Sched::None),
+        isTwoAddress(false), isCommutable(false), hasPhysRegUses(false),
+        hasPhysRegDefs(false), hasPhysRegClobbers(false), isPending(false),
+        isAvailable(false), isScheduled(false), isScheduleHigh(false),
+        isScheduleLow(false), isCloned(false), SchedulingPref(Sched::None),
         isDepthCurrent(false), isHeightCurrent(false), Depth(0), Height(0),
         TopReadyCycle(0), BotReadyCycle(0), CopyDstRC(NULL), CopySrcRC(NULL) {}
 
diff --git a/lib/CodeGen/MachineScheduler.cpp b/lib/CodeGen/MachineScheduler.cpp
index 5bd2349b50..736f6deae9 100644
--- a/lib/CodeGen/MachineScheduler.cpp
+++ b/lib/CodeGen/MachineScheduler.cpp
@@ -404,6 +404,8 @@ void ScheduleDAGMI::releasePredecessors(SUnit *SU) {
   }
 }
 
+/// This is normally called from the main scheduler loop but may also be invoked
+/// by the scheduling strategy to perform additional code motion.
 void ScheduleDAGMI::moveInstruction(MachineInstr *MI,
                                     MachineBasicBlock::iterator InsertPos) {
   // Advance RegionBegin if the first instruction moves down.
@@ -916,7 +918,7 @@ public:
   /// Represent the type of SchedCandidate found within a single queue.
   /// pickNodeBidirectional depends on these listed by decreasing priority.
   enum CandReason {
-    NoCand, SingleExcess, SingleCritical, Cluster,
+    NoCand, PhysRegCopy, SingleExcess, SingleCritical, Cluster,
     ResourceReduce, ResourceDemand, BotHeightReduce, BotPathReduce,
     TopDepthReduce, TopPathReduce, SingleMax, MultiPressure, NextDefUse,
     NodeOrder};
@@ -1191,6 +1193,8 @@ protected:
                          const RegPressureTracker &RPTracker,
                          SchedCandidate &Candidate);
 
+  void reschedulePhysRegCopies(SUnit *SU, bool isTop);
+
 #ifndef NDEBUG
   void traceCandidate(const SchedCandidate &Cand);
 #endif
@@ -1696,6 +1700,34 @@ static unsigned getWeakLeft(const SUnit *SU, bool isTop) {
   return (isTop) ? SU->WeakPredsLeft : SU->WeakSuccsLeft;
 }
 
+/// Minimize physical register live ranges. Regalloc wants them adjacent to
+/// their physreg def/use.
+///
+/// FIXME: This is an unnecessary check on the critical path. Most are root/leaf
+/// copies which can be prescheduled. The rest (e.g. x86 MUL) could be bundled
+/// with the operation that produces or consumes the physreg. We'll do this when
+/// regalloc has support for parallel copies.
+static int biasPhysRegCopy(const SUnit *SU, bool isTop) {
+  const MachineInstr *MI = SU->getInstr();
+  if (!MI->isCopy())
+    return 0;
+
+  unsigned ScheduledOper = isTop ? 1 : 0;
+  unsigned UnscheduledOper = isTop ? 0 : 1;
+  // If we have already scheduled the physreg produce/consumer, immediately
+  // schedule the copy.
+  if (TargetRegisterInfo::isPhysicalRegister(
+        MI->getOperand(ScheduledOper).getReg()))
+    return 1;
+  // If the physreg is at the boundary, defer it. Otherwise schedule it
+  // immediately to free the dependent. We can hoist the copy later.
+  bool AtBoundary = isTop ? !SU->NumSuccsLeft : !SU->NumPredsLeft;
+  if (TargetRegisterInfo::isPhysicalRegister(
+        MI->getOperand(UnscheduledOper).getReg()))
+    return AtBoundary ? -1 : 1;
+  return 0;
+}
+
 /// Apply a set of heursitics to a new candidate. Heuristics are currently
 /// hierarchical. This may be more efficient than a graduated cost model because
 /// we don't need to evaluate all aspects of the model for each node in the
@@ -1723,6 +1755,12 @@ void ConvergingScheduler::tryCandidate(SchedCandidate &Cand,
     TryCand.Reason = NodeOrder;
     return;
   }
+
+  if (tryGreater(biasPhysRegCopy(TryCand.SU, Zone.isTop()),
+                 biasPhysRegCopy(Cand.SU, Zone.isTop()),
+                 TryCand, Cand, PhysRegCopy))
+    return;
+
   // Avoid exceeding the target's limit.
   if (tryLess(TryCand.RPDelta.Excess.UnitIncrease,
               Cand.RPDelta.Excess.UnitIncrease, TryCand, Cand, SingleExcess))
@@ -1851,6 +1889,7 @@ const char *ConvergingScheduler::getReasonStr(
   ConvergingScheduler::CandReason Reason) {
   switch (Reason) {
   case NoCand:         return "NOCAND    ";
+  case PhysRegCopy:    return "PREG-COPY";
   case SingleExcess:   return "REG-EXCESS";
   case SingleCritical: return "REG-CRIT  ";
   case Cluster:        return "CLUSTER   ";
@@ -2069,17 +2108,49 @@ SUnit *ConvergingScheduler::pickNode(bool &IsTopNode) {
   return SU;
 }
 
+void ConvergingScheduler::reschedulePhysRegCopies(SUnit *SU, bool isTop) {
+
+  MachineBasicBlock::iterator InsertPos = SU->getInstr();
+  if (!isTop)
+    ++InsertPos;
+  SmallVectorImpl<SDep> &Deps = isTop ? SU->Preds : SU->Succs;
+
+  // Find already scheduled copies with a single physreg dependence and move
+  // them just above the scheduled instruction.
+  for (SmallVectorImpl<SDep>::iterator I = Deps.begin(), E = Deps.end();
+       I != E; ++I) {
+    if (I->getKind() != SDep::Data || !TRI->isPhysicalRegister(I->getReg()))
+      continue;
+    SUnit *DepSU = I->getSUnit();
+    if (isTop ? DepSU->Succs.size() > 1 : DepSU->Preds.size() > 1)
+      continue;
+    MachineInstr *Copy = DepSU->getInstr();
+    if (!Copy->isCopy())
+      continue;
+    DEBUG(dbgs() << "  Rescheduling physreg copy ";
+          I->getSUnit()->dump(DAG));
+    DAG->moveInstruction(Copy, InsertPos);
+  }
+}
+
 /// Update the scheduler's state after scheduling a node. This is the same node
 /// that was just returned by pickNode(). However, ScheduleDAGMI needs to update
 /// it's state based on the current cycle before MachineSchedStrategy does.
+///
+/// FIXME: Eventually, we may bundle physreg copies rather than rescheduling
+/// them here. See comments in biasPhysRegCopy.
 void ConvergingScheduler::schedNode(SUnit *SU, bool IsTopNode) {
   if (IsTopNode) {
     SU->TopReadyCycle = Top.CurrCycle;
     Top.bumpNode(SU);
+    if (SU->hasPhysRegUses)
+      reschedulePhysRegCopies(SU, true);
   }
   else {
     SU->BotReadyCycle = Bot.CurrCycle;
     Bot.bumpNode(SU);
+    if (SU->hasPhysRegDefs)
+      reschedulePhysRegCopies(SU, false);
   }
 }
 
diff --git a/lib/CodeGen/ScheduleDAGInstrs.cpp b/lib/CodeGen/ScheduleDAGInstrs.cpp
index 71e7a21ef2..e4da6a41ee 100644
--- a/lib/CodeGen/ScheduleDAGInstrs.cpp
+++ b/lib/CodeGen/ScheduleDAGInstrs.cpp
@@ -262,6 +262,9 @@ void ScheduleDAGInstrs::addPhysRegDataDeps(SUnit *SU, unsigned OperIdx) {
       if (UseOp < 0)
         Dep = SDep(SU, SDep::Artificial);
       else {
+        // Set the hasPhysRegDefs only for physreg defs that have a use within
+        // the scheduling region.
+        SU->hasPhysRegDefs = true;
         Dep = SDep(SU, SDep::Data, *Alias);
         RegUse = UseSU->getInstr();
         Dep.setMinLatency(
@@ -318,6 +321,7 @@ void ScheduleDAGInstrs::addPhysRegDeps(SUnit *SU, unsigned OperIdx) {
   }
 
   if (!MO.isDef()) {
+    SU->hasPhysRegUses = true;
     // Either insert a new Reg2SUnits entry with an empty SUnits list, or
     // retrieve the existing SUnits list for this register's uses.
     // Push this SUnit on the use list.
diff --git a/test/CodeGen/X86/misched-copy.ll b/test/CodeGen/X86/misched-copy.ll
new file mode 100644
index 0000000000..d04df999b9
--- /dev/null
+++ b/test/CodeGen/X86/misched-copy.ll
@@ -0,0 +1,48 @@
+; RUN: llc %s -march=x86 -mcpu=core2 -pre-RA-sched=source -enable-misched -verify-misched -debug-only=misched 2>&1 | FileCheck %s
+;
+; Test scheduling of copy instructions.
+;
+; Argument copies should be hoisted to the top of the block.
+; Return copies should be sunk to the end.
+; MUL_HiLo PhysReg use copies should be just above the mul.
+; MUL_HiLo PhysReg def copies should be just below the mul.
+;
+; CHECK:      *** Final schedule for BB#1 ***
+; CHECK-NEXT: %EAX<def> = COPY
+; CHECK:      MUL32r %vreg{{[0-6]+}}, %EAX<imp-def>, %EDX<imp-def>, %EFLAGS<imp-def,dead>, %EAX<imp-use>;
+; CHECK-NEXT: COPY %EAX;
+; CHECK-NEXT: COPY %EDX;
+; CHECK:      DIVSSrm
+define i64 @mulhoist(i32 %a, i32 %b) #0 {
+entry:
+  br label %body
+
+body:
+  %convb = sitofp i32 %b to float
+  ; Generates an iMUL64r to legalize types.
+  %aa = zext i32 %a to i64
+  %mul = mul i64 %aa, 74383
+  ; Do some dependent long latency stuff.
+  %trunc = trunc i64 %mul to i32
+  %convm = sitofp i32 %trunc to float
+  %divm = fdiv float %convm, 0.75
+  ;%addmb = fadd float %divm, %convb
+  ;%divmb = fdiv float %addmb, 0.125
+  ; Do some independent long latency stuff.
+  %conva = sitofp i32 %a to float
+  %diva = fdiv float %conva, 0.75
+  %addab = fadd float %diva, %convb
+  %divab = fdiv float %addab, 0.125
+  br label %end
+
+end:
+  %val = fptosi float %divab to i64
+  %add = add i64 %mul, %val
+  ret i64 %add
+}
+
+attributes #0 = { nounwind ssp uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-frame-pointer-elim-non-leaf"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+!0 = metadata !{metadata !"float", metadata !1}
+!1 = metadata !{metadata !"omnipotent char", metadata !2}
+!2 = metadata !{metadata !"Simple C/C++ TBAA"}