Begin implementation of an inline spiller.

InlineSpiller inserts loads and spills immediately instead of deferring to
VirtRegMap. This is possible now because SlotIndexes allows instructions to be
inserted and renumbered.

This is work in progress, and is mostly a copy of TrivialSpiller so far. It
works very well for functions that don't require spilling.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107227 91177308-0d34-0410-b5e6-96231b3b80d8

include/llvm/CodeGen/MachineRegisterInfo.h

@@ -363,7 +363,18 @@ public:
     defusechain_iterator operator++(int) {        // Postincrement
       defusechain_iterator tmp = *this; ++*this; return tmp;
     }
-    
+
+    /// skipInstruction - move forward until reaching a different instruction.
+    /// Return the skipped instruction that is no longer pointed to, or NULL if
+    /// already pointing to end().
+    MachineInstr *skipInstruction() {
+      if (!Op) return 0;
+      MachineInstr *MI = Op->getParent();
+      do ++*this;
+      while (Op && Op->getParent() == MI);
+      return MI;
+    }
+
     MachineOperand &getOperand() const {
       assert(Op && "Cannot dereference end iterator!");
       return *Op;

lib/CodeGen/CMakeLists.txt

@@ -13,6 +13,7 @@ add_llvm_library(LLVMCodeGen
   GCMetadataPrinter.cpp
   GCStrategy.cpp
   IfConversion.cpp
+  InlineSpiller.cpp
   IntrinsicLowering.cpp
   LLVMTargetMachine.cpp
   LatencyPriorityQueue.cpp

lib/CodeGen/InlineSpiller.cpp

@@ -0,0 +1,136 @@
+//===-------- InlineSpiller.cpp - Insert spills and restores inline -------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+//
+// The inline spiller modifies the machine function directly instead of
+// inserting spills and restores in VirtRegMap.
+//
+//===----------------------------------------------------------------------===//
+
+#define DEBUG_TYPE "spiller"
+#include "Spiller.h"
+#include "VirtRegMap.h"
+#include "llvm/CodeGen/LiveIntervalAnalysis.h"
+#include "llvm/CodeGen/MachineFrameInfo.h"
+#include "llvm/CodeGen/MachineFunction.h"
+#include "llvm/CodeGen/MachineRegisterInfo.h"
+#include "llvm/Target/TargetMachine.h"
+#include "llvm/Target/TargetInstrInfo.h"
+#include "llvm/Support/Debug.h"
+#include "llvm/Support/raw_ostream.h"
+
+using namespace llvm;
+
+namespace {
+class InlineSpiller : public Spiller {
+  MachineFunction &mf_;
+  LiveIntervals &lis_;
+  VirtRegMap &vrm_;
+  MachineFrameInfo &mfi_;
+  MachineRegisterInfo &mri_;
+  const TargetInstrInfo &tii_;
+  const TargetRegisterInfo &tri_;
+
+  ~InlineSpiller() {}
+
+public:
+  InlineSpiller(MachineFunction *mf, LiveIntervals *lis, VirtRegMap *vrm)
+    : mf_(*mf), lis_(*lis), vrm_(*vrm),
+      mfi_(*mf->getFrameInfo()),
+      mri_(mf->getRegInfo()),
+      tii_(*mf->getTarget().getInstrInfo()),
+      tri_(*mf->getTarget().getRegisterInfo()) {}
+
+  void spill(LiveInterval *li,
+             std::vector<LiveInterval*> &newIntervals,
+             SmallVectorImpl<LiveInterval*> &spillIs,
+             SlotIndex *earliestIndex);
+};
+}
+
+namespace llvm {
+Spiller *createInlineSpiller(MachineFunction *mf,
+                             LiveIntervals *lis,
+                             const MachineLoopInfo *mli,
+                             VirtRegMap *vrm) {
+  return new InlineSpiller(mf, lis, vrm);
+}
+}
+
+void InlineSpiller::spill(LiveInterval *li,
+                          std::vector<LiveInterval*> &newIntervals,
+                          SmallVectorImpl<LiveInterval*> &spillIs,
+                          SlotIndex *earliestIndex) {
+  DEBUG(dbgs() << "Inline spilling " << *li << "\n");
+  assert(li->isSpillable() && "Attempting to spill already spilled value.");
+  assert(!li->isStackSlot() && "Trying to spill a stack slot.");
+
+  const TargetRegisterClass *RC = mri_.getRegClass(li->reg);
+  unsigned SS = vrm_.assignVirt2StackSlot(li->reg);
+
+  for (MachineRegisterInfo::reg_iterator RI = mri_.reg_begin(li->reg);
+       MachineInstr *MI = RI.skipInstruction();) {
+    SlotIndex Idx = lis_.getInstructionIndex(MI).getDefIndex();
+
+    // Analyze instruction.
+    bool Reads, Writes;
+    SmallVector<unsigned, 8> Ops;
+    tie(Reads, Writes) = MI->readsWritesVirtualRegister(li->reg, &Ops);
+
+    // Allocate interval around instruction.
+    // FIXME: Infer regclass from instruction alone.
+    unsigned NewVReg = mri_.createVirtualRegister(RC);
+    vrm_.grow();
+    LiveInterval &NewLI = lis_.getOrCreateInterval(NewVReg);
+    NewLI.markNotSpillable();
+
+    // Reload if instruction reads register.
+    if (Reads) {
+      MachineBasicBlock::iterator MII = MI;
+      tii_.loadRegFromStackSlot(*MI->getParent(), MII, NewVReg, SS, RC, &tri_);
+      --MII; // Point to load instruction.
+      SlotIndex LoadIdx = lis_.InsertMachineInstrInMaps(MII).getDefIndex();
+      DEBUG(dbgs() << "\treload:  " << LoadIdx << '\t' << *MII);
+      VNInfo *LoadVNI = NewLI.getNextValue(LoadIdx, 0, true,
+                                           lis_.getVNInfoAllocator());
+      NewLI.addRange(LiveRange(LoadIdx, Idx, LoadVNI));
+    }
+
+    // Rewrite instruction operands.
+    bool hasLiveDef = false;
+    for (unsigned i = 0, e = Ops.size(); i != e; ++i) {
+      MachineOperand &MO = MI->getOperand(Ops[i]);
+      MO.setReg(NewVReg);
+      if (MO.isUse()) {
+        if (!MI->isRegTiedToDefOperand(Ops[i]))
+          MO.setIsKill();
+      } else {
+        if (!MO.isDead())
+          hasLiveDef = true;
+      }
+    }
+    DEBUG(dbgs() << "\trewrite: " << Idx << '\t' << *MI);
+
+    // Spill is instruction writes register.
+    // FIXME: Use a second vreg if instruction has no tied ops.
+    if (Writes && hasLiveDef) {
+      MachineBasicBlock::iterator MII = MI;
+      tii_.storeRegToStackSlot(*MI->getParent(), ++MII, NewVReg, true, SS, RC,
+                               &tri_);
+      --MII; // Point to store instruction.
+      SlotIndex StoreIdx = lis_.InsertMachineInstrInMaps(MII).getDefIndex();
+      DEBUG(dbgs() << "\tspilled: " << StoreIdx << '\t' << *MII);
+      VNInfo *StoreVNI = NewLI.getNextValue(Idx, 0, true,
+                                            lis_.getVNInfoAllocator());
+      NewLI.addRange(LiveRange(Idx, StoreIdx, StoreVNI));
+    }
+
+    DEBUG(dbgs() << "\tinterval: " << NewLI << '\n');
+    newIntervals.push_back(&NewLI);
+  }
+}

lib/CodeGen/Spiller.cpp

@@ -26,7 +26,7 @@
 using namespace llvm;
 
 namespace {
-  enum SpillerName { trivial, standard, splitting };
+  enum SpillerName { trivial, standard, splitting, inline_ };
 }
 
 static cl::opt<SpillerName>
@@ -36,6 +36,7 @@ spillerOpt("spiller",
            cl::values(clEnumVal(trivial,   "trivial spiller"),
                       clEnumVal(standard,  "default spiller"),
                       clEnumVal(splitting, "splitting spiller"),
+                      "inline", inline_,   "inline spiller",
                       clEnumValEnd),
            cl::init(standard));
 
@@ -506,6 +507,13 @@ private:
 } // end anonymous namespace
 
 
+namespace llvm {
+Spiller *createInlineSpiller(MachineFunction*,
+                             LiveIntervals*,
+                             const MachineLoopInfo*,
+                             VirtRegMap*);
+}
+
 llvm::Spiller* llvm::createSpiller(MachineFunction *mf, LiveIntervals *lis,
                                    const MachineLoopInfo *loopInfo,
                                    VirtRegMap *vrm) {
@@ -514,5 +522,6 @@ llvm::Spiller* llvm::createSpiller(MachineFunction *mf, LiveIntervals *lis,
   case trivial: return new TrivialSpiller(mf, lis, vrm);
   case standard: return new StandardSpiller(lis, loopInfo, vrm);
   case splitting: return new SplittingSpiller(mf, lis, loopInfo, vrm);
+  case inline_: return createInlineSpiller(mf, lis, loopInfo, vrm);
   }
 }