//
//===----------------------------------------------------------------------===//
-#include "llvm/CodeGen/Passes.h"
-#include "llvm/CodeGen/MachineFunctionPass.h"
-#include "llvm/CodeGen/MachineInstr.h"
+#define DEBUG_TYPE "pei"
+#include "PrologEpilogInserter.h"
+#include "llvm/ADT/IndexedMap.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SetVector.h"
+#include "llvm/ADT/SmallSet.h"
+#include "llvm/ADT/Statistic.h"
+#include "llvm/CodeGen/MachineDominators.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
+#include "llvm/CodeGen/MachineInstr.h"
+#include "llvm/CodeGen/MachineLoopInfo.h"
#include "llvm/CodeGen/MachineModuleInfo.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/CodeGen/RegisterScavenging.h"
-#include "llvm/Target/TargetMachine.h"
-#include "llvm/Target/MRegisterInfo.h"
-#include "llvm/Target/TargetFrameInfo.h"
-#include "llvm/Target/TargetInstrInfo.h"
+#include "llvm/CodeGen/StackProtector.h"
+#include "llvm/IR/DiagnosticInfo.h"
+#include "llvm/IR/InlineAsm.h"
+#include "llvm/IR/LLVMContext.h"
+#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Compiler.h"
-#include "llvm/ADT/STLExtras.h"
+#include "llvm/Support/Debug.h"
+#include "llvm/Support/raw_ostream.h"
+#include "llvm/Target/TargetFrameLowering.h"
+#include "llvm/Target/TargetInstrInfo.h"
+#include "llvm/Target/TargetMachine.h"
+#include "llvm/Target/TargetRegisterInfo.h"
#include <climits>
+
using namespace llvm;
-namespace {
- struct VISIBILITY_HIDDEN PEI : public MachineFunctionPass {
- static char ID;
- PEI() : MachineFunctionPass((intptr_t)&ID) {}
+char PEI::ID = 0;
+char &llvm::PrologEpilogCodeInserterID = PEI::ID;
+
+static cl::opt<unsigned>
+WarnStackSize("warn-stack-size", cl::Hidden, cl::init((unsigned)-1),
+ cl::desc("Warn for stack size bigger than the given"
+ " number"));
+
+INITIALIZE_PASS_BEGIN(PEI, "prologepilog",
+ "Prologue/Epilogue Insertion", false, false)
+INITIALIZE_PASS_DEPENDENCY(MachineLoopInfo)
+INITIALIZE_PASS_DEPENDENCY(MachineDominatorTree)
+INITIALIZE_PASS_DEPENDENCY(StackProtector)
+INITIALIZE_PASS_DEPENDENCY(TargetPassConfig)
+INITIALIZE_PASS_END(PEI, "prologepilog",
+ "Prologue/Epilogue Insertion & Frame Finalization",
+ false, false)
+
+STATISTIC(NumScavengedRegs, "Number of frame index regs scavenged");
+STATISTIC(NumBytesStackSpace,
+ "Number of bytes used for stack in all functions");
+
+void PEI::getAnalysisUsage(AnalysisUsage &AU) const {
+ AU.setPreservesCFG();
+ AU.addPreserved<MachineLoopInfo>();
+ AU.addPreserved<MachineDominatorTree>();
+ AU.addRequired<StackProtector>();
+ AU.addRequired<TargetPassConfig>();
+ MachineFunctionPass::getAnalysisUsage(AU);
+}
- const char *getPassName() const {
- return "Prolog/Epilog Insertion & Frame Finalization";
- }
+bool PEI::isReturnBlock(MachineBasicBlock* MBB) {
+ return (MBB && !MBB->empty() && MBB->back().isReturn());
+}
- /// runOnMachineFunction - Insert prolog/epilog code and replace abstract
- /// frame indexes with appropriate references.
- ///
- bool runOnMachineFunction(MachineFunction &Fn) {
- const MRegisterInfo *MRI = Fn.getTarget().getRegisterInfo();
- RS = MRI->requiresRegisterScavenging(Fn) ? new RegScavenger() : NULL;
+/// Compute the set of return blocks
+void PEI::calculateSets(MachineFunction &Fn) {
+ // Sets used to compute spill, restore placement sets.
+ const std::vector<CalleeSavedInfo> &CSI =
+ Fn.getFrameInfo()->getCalleeSavedInfo();
- // Get MachineModuleInfo so that we can track the construction of the
- // frame.
- if (MachineModuleInfo *MMI = getAnalysisToUpdate<MachineModuleInfo>())
- Fn.getFrameInfo()->setMachineModuleInfo(MMI);
-
- // Allow the target machine to make some adjustments to the function
- // e.g. UsedPhysRegs before calculateCalleeSavedRegisters.
- MRI->processFunctionBeforeCalleeSavedScan(Fn, RS);
-
- // Scan the function for modified callee saved registers and insert spill
- // code for any callee saved registers that are modified. Also calculate
- // the MaxCallFrameSize and HasCalls variables for the function's frame
- // information and eliminates call frame pseudo instructions.
- calculateCalleeSavedRegisters(Fn);
-
- // Add the code to save and restore the callee saved registers
- saveCalleeSavedRegisters(Fn);
-
- // Allow the target machine to make final modifications to the function
- // before the frame layout is finalized.
- Fn.getTarget().getRegisterInfo()->processFunctionBeforeFrameFinalized(Fn);
-
- // Calculate actual frame offsets for all of the abstract stack objects...
- calculateFrameObjectOffsets(Fn);
-
- // Add prolog and epilog code to the function. This function is required
- // to align the stack frame as necessary for any stack variables or
- // called functions. Because of this, calculateCalleeSavedRegisters
- // must be called before this function in order to set the HasCalls
- // and MaxCallFrameSize variables.
- insertPrologEpilogCode(Fn);
-
- // Replace all MO_FrameIndex operands with physical register references
- // and actual offsets.
- //
- replaceFrameIndices(Fn);
-
- delete RS;
- return true;
- }
-
- private:
- RegScavenger *RS;
-
- // MinCSFrameIndex, MaxCSFrameIndex - Keeps the range of callee saved
- // stack frame indexes.
- unsigned MinCSFrameIndex, MaxCSFrameIndex;
-
- void calculateCalleeSavedRegisters(MachineFunction &Fn);
- void saveCalleeSavedRegisters(MachineFunction &Fn);
- void calculateFrameObjectOffsets(MachineFunction &Fn);
- void replaceFrameIndices(MachineFunction &Fn);
- void insertPrologEpilogCode(MachineFunction &Fn);
- };
- char PEI::ID = 0;
+ // If no CSRs used, we are done.
+ if (CSI.empty())
+ return;
+
+ // Save refs to entry and return blocks.
+ EntryBlock = Fn.begin();
+ for (MachineFunction::iterator MBB = Fn.begin(), E = Fn.end();
+ MBB != E; ++MBB)
+ if (isReturnBlock(MBB))
+ ReturnBlocks.push_back(MBB);
+
+ return;
}
+/// StackObjSet - A set of stack object indexes
+typedef SmallSetVector<int, 8> StackObjSet;
-/// createPrologEpilogCodeInserter - This function returns a pass that inserts
-/// prolog and epilog code, and eliminates abstract frame references.
+/// runOnMachineFunction - Insert prolog/epilog code and replace abstract
+/// frame indexes with appropriate references.
///
-FunctionPass *llvm::createPrologEpilogCodeInserter() { return new PEI(); }
+bool PEI::runOnMachineFunction(MachineFunction &Fn) {
+ const Function* F = Fn.getFunction();
+ const TargetRegisterInfo *TRI = Fn.getTarget().getRegisterInfo();
+ const TargetFrameLowering *TFI = Fn.getTarget().getFrameLowering();
+
+ assert(!Fn.getRegInfo().getNumVirtRegs() && "Regalloc must assign all vregs");
+
+ RS = TRI->requiresRegisterScavenging(Fn) ? new RegScavenger() : nullptr;
+ FrameIndexVirtualScavenging = TRI->requiresFrameIndexScavenging(Fn);
+
+ // Calculate the MaxCallFrameSize and AdjustsStack variables for the
+ // function's frame information. Also eliminates call frame pseudo
+ // instructions.
+ calculateCallsInformation(Fn);
+
+ // Allow the target machine to make some adjustments to the function
+ // e.g. UsedPhysRegs before calculateCalleeSavedRegisters.
+ TFI->processFunctionBeforeCalleeSavedScan(Fn, RS);
+
+ // Scan the function for modified callee saved registers and insert spill code
+ // for any callee saved registers that are modified.
+ calculateCalleeSavedRegisters(Fn);
+
+ // Determine placement of CSR spill/restore code:
+ // place all spills in the entry block, all restores in return blocks.
+ calculateSets(Fn);
+
+ // Add the code to save and restore the callee saved registers
+ if (!F->hasFnAttribute(Attribute::Naked))
+ insertCSRSpillsAndRestores(Fn);
+
+ // Allow the target machine to make final modifications to the function
+ // before the frame layout is finalized.
+ TFI->processFunctionBeforeFrameFinalized(Fn, RS);
+
+ // Calculate actual frame offsets for all abstract stack objects...
+ calculateFrameObjectOffsets(Fn);
+
+ // Add prolog and epilog code to the function. This function is required
+ // to align the stack frame as necessary for any stack variables or
+ // called functions. Because of this, calculateCalleeSavedRegisters()
+ // must be called before this function in order to set the AdjustsStack
+ // and MaxCallFrameSize variables.
+ if (!F->hasFnAttribute(Attribute::Naked))
+ insertPrologEpilogCode(Fn);
+
+ // Replace all MO_FrameIndex operands with physical register references
+ // and actual offsets.
+ //
+ replaceFrameIndices(Fn);
+
+ // If register scavenging is needed, as we've enabled doing it as a
+ // post-pass, scavenge the virtual registers that frame index elimiation
+ // inserted.
+ if (TRI->requiresRegisterScavenging(Fn) && FrameIndexVirtualScavenging)
+ scavengeFrameVirtualRegs(Fn);
+
+ // Clear any vregs created by virtual scavenging.
+ Fn.getRegInfo().clearVirtRegs();
+
+ // Warn on stack size when we exceeds the given limit.
+ MachineFrameInfo *MFI = Fn.getFrameInfo();
+ uint64_t StackSize = MFI->getStackSize();
+ if (WarnStackSize.getNumOccurrences() > 0 && WarnStackSize < StackSize) {
+ DiagnosticInfoStackSize DiagStackSize(*F, StackSize);
+ F->getContext().diagnose(DiagStackSize);
+ }
+ delete RS;
+ ReturnBlocks.clear();
+ return true;
+}
-/// calculateCalleeSavedRegisters - Scan the function for modified callee saved
-/// registers. Also calculate the MaxCallFrameSize and HasCalls variables for
-/// the function's frame information and eliminates call frame pseudo
-/// instructions.
-///
-void PEI::calculateCalleeSavedRegisters(MachineFunction &Fn) {
- const MRegisterInfo *RegInfo = Fn.getTarget().getRegisterInfo();
- const TargetFrameInfo *TFI = Fn.getTarget().getFrameInfo();
+/// calculateCallsInformation - Calculate the MaxCallFrameSize and AdjustsStack
+/// variables for the function's frame information and eliminate call frame
+/// pseudo instructions.
+void PEI::calculateCallsInformation(MachineFunction &Fn) {
+ const TargetInstrInfo &TII = *Fn.getTarget().getInstrInfo();
+ const TargetFrameLowering *TFI = Fn.getTarget().getFrameLowering();
+ MachineFrameInfo *MFI = Fn.getFrameInfo();
- // Get the callee saved register list...
- const unsigned *CSRegs = RegInfo->getCalleeSavedRegs(&Fn);
+ unsigned MaxCallFrameSize = 0;
+ bool AdjustsStack = MFI->adjustsStack();
// Get the function call frame set-up and tear-down instruction opcode
- int FrameSetupOpcode = RegInfo->getCallFrameSetupOpcode();
- int FrameDestroyOpcode = RegInfo->getCallFrameDestroyOpcode();
-
- // These are used to keep track the callee-save area. Initialize them.
- MinCSFrameIndex = INT_MAX;
- MaxCSFrameIndex = 0;
+ int FrameSetupOpcode = TII.getCallFrameSetupOpcode();
+ int FrameDestroyOpcode = TII.getCallFrameDestroyOpcode();
- // Early exit for targets which have no callee saved registers and no call
- // frame setup/destroy pseudo instructions.
- if ((CSRegs == 0 || CSRegs[0] == 0) &&
- FrameSetupOpcode == -1 && FrameDestroyOpcode == -1)
+ // Early exit for targets which have no call frame setup/destroy pseudo
+ // instructions.
+ if (FrameSetupOpcode == -1 && FrameDestroyOpcode == -1)
return;
- unsigned MaxCallFrameSize = 0;
- bool HasCalls = false;
-
std::vector<MachineBasicBlock::iterator> FrameSDOps;
for (MachineFunction::iterator BB = Fn.begin(), E = Fn.end(); BB != E; ++BB)
for (MachineBasicBlock::iterator I = BB->begin(); I != BB->end(); ++I)
" instructions should have a single immediate argument!");
unsigned Size = I->getOperand(0).getImm();
if (Size > MaxCallFrameSize) MaxCallFrameSize = Size;
- HasCalls = true;
+ AdjustsStack = true;
FrameSDOps.push_back(I);
+ } else if (I->isInlineAsm()) {
+ // Some inline asm's need a stack frame, as indicated by operand 1.
+ unsigned ExtraInfo = I->getOperand(InlineAsm::MIOp_ExtraInfo).getImm();
+ if (ExtraInfo & InlineAsm::Extra_IsAlignStack)
+ AdjustsStack = true;
}
- MachineFrameInfo *FFI = Fn.getFrameInfo();
- FFI->setHasCalls(HasCalls);
- FFI->setMaxCallFrameSize(MaxCallFrameSize);
-
- for (unsigned i = 0, e = FrameSDOps.size(); i != e; ++i) {
- MachineBasicBlock::iterator I = FrameSDOps[i];
- // If call frames are not being included as part of the stack frame,
- // and there is no dynamic allocation (therefore referencing frame slots
- // off sp), leave the pseudo ops alone. We'll eliminate them later.
- if (RegInfo->hasReservedCallFrame(Fn) || RegInfo->hasFP(Fn))
- RegInfo->eliminateCallFramePseudoInstr(Fn, *I->getParent(), I);
+ MFI->setAdjustsStack(AdjustsStack);
+ MFI->setMaxCallFrameSize(MaxCallFrameSize);
+
+ for (std::vector<MachineBasicBlock::iterator>::iterator
+ i = FrameSDOps.begin(), e = FrameSDOps.end(); i != e; ++i) {
+ MachineBasicBlock::iterator I = *i;
+
+ // If call frames are not being included as part of the stack frame, and
+ // the target doesn't indicate otherwise, remove the call frame pseudos
+ // here. The sub/add sp instruction pairs are still inserted, but we don't
+ // need to track the SP adjustment for frame index elimination.
+ if (TFI->canSimplifyCallFramePseudos(Fn))
+ TFI->eliminateCallFramePseudoInstr(Fn, *I->getParent(), I);
}
+}
+
+
+/// calculateCalleeSavedRegisters - Scan the function for modified callee saved
+/// registers.
+void PEI::calculateCalleeSavedRegisters(MachineFunction &F) {
+ const TargetRegisterInfo *RegInfo = F.getTarget().getRegisterInfo();
+ const TargetFrameLowering *TFI = F.getTarget().getFrameLowering();
+ MachineFrameInfo *MFI = F.getFrameInfo();
+
+ // Get the callee saved register list...
+ const MCPhysReg *CSRegs = RegInfo->getCalleeSavedRegs(&F);
+
+ // These are used to keep track the callee-save area. Initialize them.
+ MinCSFrameIndex = INT_MAX;
+ MaxCSFrameIndex = 0;
+
+ // Early exit for targets which have no callee saved registers.
+ if (!CSRegs || CSRegs[0] == 0)
+ return;
+
+ // In Naked functions we aren't going to save any registers.
+ if (F.getFunction()->hasFnAttribute(Attribute::Naked))
+ return;
- // Now figure out which *callee saved* registers are modified by the current
- // function, thus needing to be saved and restored in the prolog/epilog.
- //
- const TargetRegisterClass* const *CSRegClasses =
- RegInfo->getCalleeSavedRegClasses(&Fn);
std::vector<CalleeSavedInfo> CSI;
for (unsigned i = 0; CSRegs[i]; ++i) {
unsigned Reg = CSRegs[i];
- if (Fn.getRegInfo().isPhysRegUsed(Reg)) {
- // If the reg is modified, save it!
- CSI.push_back(CalleeSavedInfo(Reg, CSRegClasses[i]));
- } else {
- for (const unsigned *AliasSet = RegInfo->getAliasSet(Reg);
- *AliasSet; ++AliasSet) { // Check alias registers too.
- if (Fn.getRegInfo().isPhysRegUsed(*AliasSet)) {
- CSI.push_back(CalleeSavedInfo(Reg, CSRegClasses[i]));
- break;
- }
- }
+ // Functions which call __builtin_unwind_init get all their registers saved.
+ if (F.getRegInfo().isPhysRegUsed(Reg) || F.getMMI().callsUnwindInit()) {
+ // If the reg is modified, save it!
+ CSI.push_back(CalleeSavedInfo(Reg));
}
}
return; // Early exit if no callee saved registers are modified!
unsigned NumFixedSpillSlots;
- const std::pair<unsigned,int> *FixedSpillSlots =
+ const TargetFrameLowering::SpillSlot *FixedSpillSlots =
TFI->getCalleeSavedSpillSlots(NumFixedSpillSlots);
// Now that we know which registers need to be saved and restored, allocate
// stack slots for them.
- for (unsigned i = 0, e = CSI.size(); i != e; ++i) {
- unsigned Reg = CSI[i].getReg();
- const TargetRegisterClass *RC = CSI[i].getRegClass();
+ for (std::vector<CalleeSavedInfo>::iterator
+ I = CSI.begin(), E = CSI.end(); I != E; ++I) {
+ unsigned Reg = I->getReg();
+ const TargetRegisterClass *RC = RegInfo->getMinimalPhysRegClass(Reg);
+
+ int FrameIdx;
+ if (RegInfo->hasReservedSpillSlot(F, Reg, FrameIdx)) {
+ I->setFrameIdx(FrameIdx);
+ continue;
+ }
// Check to see if this physreg must be spilled to a particular stack slot
// on this target.
- const std::pair<unsigned,int> *FixedSlot = FixedSpillSlots;
+ const TargetFrameLowering::SpillSlot *FixedSlot = FixedSpillSlots;
while (FixedSlot != FixedSpillSlots+NumFixedSpillSlots &&
- FixedSlot->first != Reg)
+ FixedSlot->Reg != Reg)
++FixedSlot;
- int FrameIdx;
- if (FixedSlot == FixedSpillSlots+NumFixedSpillSlots) {
+ if (FixedSlot == FixedSpillSlots + NumFixedSpillSlots) {
// Nope, just spill it anywhere convenient.
unsigned Align = RC->getAlignment();
unsigned StackAlign = TFI->getStackAlignment();
- // We may not be able to sastify the desired alignment specification of
- // the TargetRegisterClass if the stack alignment is smaller. Use the min.
+
+ // We may not be able to satisfy the desired alignment specification of
+ // the TargetRegisterClass if the stack alignment is smaller. Use the
+ // min.
Align = std::min(Align, StackAlign);
- FrameIdx = FFI->CreateStackObject(RC->getSize(), Align);
+ FrameIdx = MFI->CreateStackObject(RC->getSize(), Align, true);
if ((unsigned)FrameIdx < MinCSFrameIndex) MinCSFrameIndex = FrameIdx;
if ((unsigned)FrameIdx > MaxCSFrameIndex) MaxCSFrameIndex = FrameIdx;
} else {
// Spill it to the stack where we must.
- FrameIdx = FFI->CreateFixedObject(RC->getSize(), FixedSlot->second);
+ FrameIdx = MFI->CreateFixedObject(RC->getSize(), FixedSlot->Offset, true);
}
- CSI[i].setFrameIdx(FrameIdx);
+
+ I->setFrameIdx(FrameIdx);
}
- FFI->setCalleeSavedInfo(CSI);
+ MFI->setCalleeSavedInfo(CSI);
}
-/// saveCalleeSavedRegisters - Insert spill code for any callee saved registers
-/// that are modified in the function.
+/// insertCSRSpillsAndRestores - Insert spill and restore code for
+/// callee saved registers used in the function.
///
-void PEI::saveCalleeSavedRegisters(MachineFunction &Fn) {
+void PEI::insertCSRSpillsAndRestores(MachineFunction &Fn) {
// Get callee saved register information.
- MachineFrameInfo *FFI = Fn.getFrameInfo();
- const std::vector<CalleeSavedInfo> &CSI = FFI->getCalleeSavedInfo();
-
+ MachineFrameInfo *MFI = Fn.getFrameInfo();
+ const std::vector<CalleeSavedInfo> &CSI = MFI->getCalleeSavedInfo();
+
+ MFI->setCalleeSavedInfoValid(true);
+
// Early exit if no callee saved registers are modified!
if (CSI.empty())
return;
const TargetInstrInfo &TII = *Fn.getTarget().getInstrInfo();
-
- // Now that we have a stack slot for each register to be saved, insert spill
- // code into the entry block.
- MachineBasicBlock *MBB = Fn.begin();
- MachineBasicBlock::iterator I = MBB->begin();
- if (!TII.spillCalleeSavedRegisters(*MBB, I, CSI)) {
+ const TargetFrameLowering *TFI = Fn.getTarget().getFrameLowering();
+ const TargetRegisterInfo *TRI = Fn.getTarget().getRegisterInfo();
+ MachineBasicBlock::iterator I;
+
+ // Spill using target interface.
+ I = EntryBlock->begin();
+ if (!TFI->spillCalleeSavedRegisters(*EntryBlock, I, CSI, TRI)) {
for (unsigned i = 0, e = CSI.size(); i != e; ++i) {
- // Add the callee-saved register as live-in. It's killed at the spill.
- MBB->addLiveIn(CSI[i].getReg());
+ // Add the callee-saved register as live-in.
+ // It's killed at the spill.
+ EntryBlock->addLiveIn(CSI[i].getReg());
// Insert the spill to the stack frame.
- TII.storeRegToStackSlot(*MBB, I, CSI[i].getReg(), true,
- CSI[i].getFrameIdx(), CSI[i].getRegClass());
+ unsigned Reg = CSI[i].getReg();
+ const TargetRegisterClass *RC = TRI->getMinimalPhysRegClass(Reg);
+ TII.storeRegToStackSlot(*EntryBlock, I, Reg, true, CSI[i].getFrameIdx(),
+ RC, TRI);
}
}
- // Add code to restore the callee-save registers in each exiting block.
- for (MachineFunction::iterator FI = Fn.begin(), E = Fn.end(); FI != E; ++FI)
- // If last instruction is a return instruction, add an epilogue.
- if (!FI->empty() && FI->back().getDesc().isReturn()) {
- MBB = FI;
- I = MBB->end(); --I;
-
- // Skip over all terminator instructions, which are part of the return
- // sequence.
- MachineBasicBlock::iterator I2 = I;
- while (I2 != MBB->begin() && (--I2)->getDesc().isTerminator())
- I = I2;
-
- bool AtStart = I == MBB->begin();
- MachineBasicBlock::iterator BeforeI = I;
- if (!AtStart)
- --BeforeI;
-
- // Restore all registers immediately before the return and any terminators
- // that preceed it.
- if (!TII.restoreCalleeSavedRegisters(*MBB, I, CSI)) {
- for (unsigned i = 0, e = CSI.size(); i != e; ++i) {
- TII.loadRegFromStackSlot(*MBB, I, CSI[i].getReg(),
- CSI[i].getFrameIdx(),
- CSI[i].getRegClass());
- assert(I != MBB->begin() &&
- "loadRegFromStackSlot didn't insert any code!");
- // Insert in reverse order. loadRegFromStackSlot can insert multiple
- // instructions.
- if (AtStart)
- I = MBB->begin();
- else {
- I = BeforeI;
- ++I;
- }
+ // Restore using target interface.
+ for (unsigned ri = 0, re = ReturnBlocks.size(); ri != re; ++ri) {
+ MachineBasicBlock *MBB = ReturnBlocks[ri];
+ I = MBB->end();
+ --I;
+
+ // Skip over all terminator instructions, which are part of the return
+ // sequence.
+ MachineBasicBlock::iterator I2 = I;
+ while (I2 != MBB->begin() && (--I2)->isTerminator())
+ I = I2;
+
+ bool AtStart = I == MBB->begin();
+ MachineBasicBlock::iterator BeforeI = I;
+ if (!AtStart)
+ --BeforeI;
+
+ // Restore all registers immediately before the return and any
+ // terminators that precede it.
+ if (!TFI->restoreCalleeSavedRegisters(*MBB, I, CSI, TRI)) {
+ for (unsigned i = 0, e = CSI.size(); i != e; ++i) {
+ unsigned Reg = CSI[i].getReg();
+ const TargetRegisterClass *RC = TRI->getMinimalPhysRegClass(Reg);
+ TII.loadRegFromStackSlot(*MBB, I, Reg, CSI[i].getFrameIdx(), RC, TRI);
+ assert(I != MBB->begin() &&
+ "loadRegFromStackSlot didn't insert any code!");
+ // Insert in reverse order. loadRegFromStackSlot can insert
+ // multiple instructions.
+ if (AtStart)
+ I = MBB->begin();
+ else {
+ I = BeforeI;
+ ++I;
}
}
}
+ }
+}
+
+/// AdjustStackOffset - Helper function used to adjust the stack frame offset.
+static inline void
+AdjustStackOffset(MachineFrameInfo *MFI, int FrameIdx,
+ bool StackGrowsDown, int64_t &Offset,
+ unsigned &MaxAlign) {
+ // If the stack grows down, add the object size to find the lowest address.
+ if (StackGrowsDown)
+ Offset += MFI->getObjectSize(FrameIdx);
+
+ unsigned Align = MFI->getObjectAlignment(FrameIdx);
+
+ // If the alignment of this object is greater than that of the stack, then
+ // increase the stack alignment to match.
+ MaxAlign = std::max(MaxAlign, Align);
+
+ // Adjust to alignment boundary.
+ Offset = (Offset + Align - 1) / Align * Align;
+
+ if (StackGrowsDown) {
+ DEBUG(dbgs() << "alloc FI(" << FrameIdx << ") at SP[" << -Offset << "]\n");
+ MFI->setObjectOffset(FrameIdx, -Offset); // Set the computed offset
+ } else {
+ DEBUG(dbgs() << "alloc FI(" << FrameIdx << ") at SP[" << Offset << "]\n");
+ MFI->setObjectOffset(FrameIdx, Offset);
+ Offset += MFI->getObjectSize(FrameIdx);
+ }
}
+/// AssignProtectedObjSet - Helper function to assign large stack objects (i.e.,
+/// those required to be close to the Stack Protector) to stack offsets.
+static void
+AssignProtectedObjSet(const StackObjSet &UnassignedObjs,
+ SmallSet<int, 16> &ProtectedObjs,
+ MachineFrameInfo *MFI, bool StackGrowsDown,
+ int64_t &Offset, unsigned &MaxAlign) {
+
+ for (StackObjSet::const_iterator I = UnassignedObjs.begin(),
+ E = UnassignedObjs.end(); I != E; ++I) {
+ int i = *I;
+ AdjustStackOffset(MFI, i, StackGrowsDown, Offset, MaxAlign);
+ ProtectedObjs.insert(i);
+ }
+}
/// calculateFrameObjectOffsets - Calculate actual frame offsets for all of the
/// abstract stack objects.
///
void PEI::calculateFrameObjectOffsets(MachineFunction &Fn) {
- const TargetFrameInfo &TFI = *Fn.getTarget().getFrameInfo();
+ const TargetFrameLowering &TFI = *Fn.getTarget().getFrameLowering();
+ StackProtector *SP = &getAnalysis<StackProtector>();
bool StackGrowsDown =
- TFI.getStackGrowthDirection() == TargetFrameInfo::StackGrowsDown;
+ TFI.getStackGrowthDirection() == TargetFrameLowering::StackGrowsDown;
// Loop over all of the stack objects, assigning sequential addresses...
- MachineFrameInfo *FFI = Fn.getFrameInfo();
-
- unsigned MaxAlign = 0;
+ MachineFrameInfo *MFI = Fn.getFrameInfo();
// Start at the beginning of the local area.
// The Offset is the distance from the stack top in the direction
- // of stack growth -- so it's always positive.
- int64_t Offset = TFI.getOffsetOfLocalArea();
+ // of stack growth -- so it's always nonnegative.
+ int LocalAreaOffset = TFI.getOffsetOfLocalArea();
if (StackGrowsDown)
- Offset = -Offset;
- assert(Offset >= 0
+ LocalAreaOffset = -LocalAreaOffset;
+ assert(LocalAreaOffset >= 0
&& "Local area offset should be in direction of stack growth");
+ int64_t Offset = LocalAreaOffset;
// If there are fixed sized objects that are preallocated in the local area,
// non-fixed objects can't be allocated right at the start of local area.
- // We currently don't support filling in holes in between fixed sized objects,
- // so we adjust 'Offset' to point to the end of last fixed sized
+ // We currently don't support filling in holes in between fixed sized
+ // objects, so we adjust 'Offset' to point to the end of last fixed sized
// preallocated object.
- for (int i = FFI->getObjectIndexBegin(); i != 0; ++i) {
+ for (int i = MFI->getObjectIndexBegin(); i != 0; ++i) {
int64_t FixedOff;
if (StackGrowsDown) {
// The maximum distance from the stack pointer is at lower address of
// the object -- which is given by offset. For down growing stack
// the offset is negative, so we negate the offset to get the distance.
- FixedOff = -FFI->getObjectOffset(i);
+ FixedOff = -MFI->getObjectOffset(i);
} else {
// The maximum distance from the start pointer is at the upper
// address of the object.
- FixedOff = FFI->getObjectOffset(i) + FFI->getObjectSize(i);
+ FixedOff = MFI->getObjectOffset(i) + MFI->getObjectSize(i);
}
if (FixedOff > Offset) Offset = FixedOff;
}
// callee saved registers.
if (StackGrowsDown) {
for (unsigned i = MinCSFrameIndex; i <= MaxCSFrameIndex; ++i) {
- // If stack grows down, we need to add size of find the lowest
+ // If the stack grows down, we need to add the size to find the lowest
// address of the object.
- Offset += FFI->getObjectSize(i);
+ Offset += MFI->getObjectSize(i);
- unsigned Align = FFI->getObjectAlignment(i);
- // If the alignment of this object is greater than that of the stack, then
- // increase the stack alignment to match.
- MaxAlign = std::max(MaxAlign, Align);
+ unsigned Align = MFI->getObjectAlignment(i);
// Adjust to alignment boundary
Offset = (Offset+Align-1)/Align*Align;
- FFI->setObjectOffset(i, -Offset); // Set the computed offset
+ MFI->setObjectOffset(i, -Offset); // Set the computed offset
}
} else {
- for (unsigned i = MaxCSFrameIndex; i >= MinCSFrameIndex; --i) {
- unsigned Align = FFI->getObjectAlignment(i);
- // If the alignment of this object is greater than that of the stack, then
- // increase the stack alignment to match.
- MaxAlign = std::max(MaxAlign, Align);
+ int MaxCSFI = MaxCSFrameIndex, MinCSFI = MinCSFrameIndex;
+ for (int i = MaxCSFI; i >= MinCSFI ; --i) {
+ unsigned Align = MFI->getObjectAlignment(i);
// Adjust to alignment boundary
Offset = (Offset+Align-1)/Align*Align;
- FFI->setObjectOffset(i, Offset);
- Offset += FFI->getObjectSize(i);
+ MFI->setObjectOffset(i, Offset);
+ Offset += MFI->getObjectSize(i);
}
}
+ unsigned MaxAlign = MFI->getMaxAlignment();
+
// Make sure the special register scavenging spill slot is closest to the
- // frame pointer if a frame pointer is required.
- const MRegisterInfo *RegInfo = Fn.getTarget().getRegisterInfo();
- if (RS && RegInfo->hasFP(Fn)) {
- int SFI = RS->getScavengingFrameIndex();
- if (SFI >= 0) {
- // If stack grows down, we need to add size of the lowest
- // address of the object.
- if (StackGrowsDown)
- Offset += FFI->getObjectSize(SFI);
+ // incoming stack pointer if a frame pointer is required and is closer
+ // to the incoming rather than the final stack pointer.
+ const TargetRegisterInfo *RegInfo = Fn.getTarget().getRegisterInfo();
+ bool EarlyScavengingSlots = (TFI.hasFP(Fn) &&
+ TFI.isFPCloseToIncomingSP() &&
+ RegInfo->useFPForScavengingIndex(Fn) &&
+ !RegInfo->needsStackRealignment(Fn));
+ if (RS && EarlyScavengingSlots) {
+ SmallVector<int, 2> SFIs;
+ RS->getScavengingFrameIndices(SFIs);
+ for (SmallVectorImpl<int>::iterator I = SFIs.begin(),
+ IE = SFIs.end(); I != IE; ++I)
+ AdjustStackOffset(MFI, *I, StackGrowsDown, Offset, MaxAlign);
+ }
- unsigned Align = FFI->getObjectAlignment(SFI);
- // Adjust to alignment boundary
- Offset = (Offset+Align-1)/Align*Align;
+ // FIXME: Once this is working, then enable flag will change to a target
+ // check for whether the frame is large enough to want to use virtual
+ // frame index registers. Functions which don't want/need this optimization
+ // will continue to use the existing code path.
+ if (MFI->getUseLocalStackAllocationBlock()) {
+ unsigned Align = MFI->getLocalFrameMaxAlign();
+
+ // Adjust to alignment boundary.
+ Offset = (Offset + Align - 1) / Align * Align;
+
+ DEBUG(dbgs() << "Local frame base offset: " << Offset << "\n");
+
+ // Resolve offsets for objects in the local block.
+ for (unsigned i = 0, e = MFI->getLocalFrameObjectCount(); i != e; ++i) {
+ std::pair<int, int64_t> Entry = MFI->getLocalFrameObjectMap(i);
+ int64_t FIOffset = (StackGrowsDown ? -Offset : Offset) + Entry.second;
+ DEBUG(dbgs() << "alloc FI(" << Entry.first << ") at SP[" <<
+ FIOffset << "]\n");
+ MFI->setObjectOffset(Entry.first, FIOffset);
+ }
+ // Allocate the local block
+ Offset += MFI->getLocalFrameSize();
+
+ MaxAlign = std::max(Align, MaxAlign);
+ }
- if (StackGrowsDown) {
- FFI->setObjectOffset(SFI, -Offset); // Set the computed offset
- } else {
- FFI->setObjectOffset(SFI, Offset);
- Offset += FFI->getObjectSize(SFI);
+ // Make sure that the stack protector comes before the local variables on the
+ // stack.
+ SmallSet<int, 16> ProtectedObjs;
+ if (MFI->getStackProtectorIndex() >= 0) {
+ StackObjSet LargeArrayObjs;
+ StackObjSet SmallArrayObjs;
+ StackObjSet AddrOfObjs;
+
+ AdjustStackOffset(MFI, MFI->getStackProtectorIndex(), StackGrowsDown,
+ Offset, MaxAlign);
+
+ // Assign large stack objects first.
+ for (unsigned i = 0, e = MFI->getObjectIndexEnd(); i != e; ++i) {
+ if (MFI->isObjectPreAllocated(i) &&
+ MFI->getUseLocalStackAllocationBlock())
+ continue;
+ if (i >= MinCSFrameIndex && i <= MaxCSFrameIndex)
+ continue;
+ if (RS && RS->isScavengingFrameIndex((int)i))
+ continue;
+ if (MFI->isDeadObjectIndex(i))
+ continue;
+ if (MFI->getStackProtectorIndex() == (int)i)
+ continue;
+
+ switch (SP->getSSPLayout(MFI->getObjectAllocation(i))) {
+ case StackProtector::SSPLK_None:
+ continue;
+ case StackProtector::SSPLK_SmallArray:
+ SmallArrayObjs.insert(i);
+ continue;
+ case StackProtector::SSPLK_AddrOf:
+ AddrOfObjs.insert(i);
+ continue;
+ case StackProtector::SSPLK_LargeArray:
+ LargeArrayObjs.insert(i);
+ continue;
}
+ llvm_unreachable("Unexpected SSPLayoutKind.");
}
+
+ AssignProtectedObjSet(LargeArrayObjs, ProtectedObjs, MFI, StackGrowsDown,
+ Offset, MaxAlign);
+ AssignProtectedObjSet(SmallArrayObjs, ProtectedObjs, MFI, StackGrowsDown,
+ Offset, MaxAlign);
+ AssignProtectedObjSet(AddrOfObjs, ProtectedObjs, MFI, StackGrowsDown,
+ Offset, MaxAlign);
}
// Then assign frame offsets to stack objects that are not used to spill
// callee saved registers.
- for (unsigned i = 0, e = FFI->getObjectIndexEnd(); i != e; ++i) {
+ for (unsigned i = 0, e = MFI->getObjectIndexEnd(); i != e; ++i) {
+ if (MFI->isObjectPreAllocated(i) &&
+ MFI->getUseLocalStackAllocationBlock())
+ continue;
if (i >= MinCSFrameIndex && i <= MaxCSFrameIndex)
continue;
- if (RS && (int)i == RS->getScavengingFrameIndex())
+ if (RS && RS->isScavengingFrameIndex((int)i))
+ continue;
+ if (MFI->isDeadObjectIndex(i))
+ continue;
+ if (MFI->getStackProtectorIndex() == (int)i)
+ continue;
+ if (ProtectedObjs.count(i))
continue;
- // If stack grows down, we need to add size of find the lowest
- // address of the object.
- if (StackGrowsDown)
- Offset += FFI->getObjectSize(i);
-
- unsigned Align = FFI->getObjectAlignment(i);
- // If the alignment of this object is greater than that of the stack, then
- // increase the stack alignment to match.
- MaxAlign = std::max(MaxAlign, Align);
- // Adjust to alignment boundary
- Offset = (Offset+Align-1)/Align*Align;
-
- if (StackGrowsDown) {
- FFI->setObjectOffset(i, -Offset); // Set the computed offset
- } else {
- FFI->setObjectOffset(i, Offset);
- Offset += FFI->getObjectSize(i);
- }
+ AdjustStackOffset(MFI, i, StackGrowsDown, Offset, MaxAlign);
}
// Make sure the special register scavenging spill slot is closest to the
// stack pointer.
- if (RS && !RegInfo->hasFP(Fn)) {
- int SFI = RS->getScavengingFrameIndex();
- if (SFI >= 0) {
- // If stack grows down, we need to add size of find the lowest
- // address of the object.
- if (StackGrowsDown)
- Offset += FFI->getObjectSize(SFI);
-
- unsigned Align = FFI->getObjectAlignment(SFI);
- // Adjust to alignment boundary
- Offset = (Offset+Align-1)/Align*Align;
-
- if (StackGrowsDown) {
- FFI->setObjectOffset(SFI, -Offset); // Set the computed offset
- } else {
- FFI->setObjectOffset(SFI, Offset);
- Offset += FFI->getObjectSize(SFI);
- }
- }
+ if (RS && !EarlyScavengingSlots) {
+ SmallVector<int, 2> SFIs;
+ RS->getScavengingFrameIndices(SFIs);
+ for (SmallVectorImpl<int>::iterator I = SFIs.begin(),
+ IE = SFIs.end(); I != IE; ++I)
+ AdjustStackOffset(MFI, *I, StackGrowsDown, Offset, MaxAlign);
}
- // Round up the size to a multiple of the alignment, but only if there are
- // calls or alloca's in the function. This ensures that any calls to
- // subroutines have their stack frames suitable aligned.
- if (!RegInfo->targetHandlesStackFrameRounding() &&
- (FFI->hasCalls() || FFI->hasVarSizedObjects())) {
+ if (!TFI.targetHandlesStackFrameRounding()) {
// If we have reserved argument space for call sites in the function
// immediately on entry to the current function, count it as part of the
// overall stack size.
- if (RegInfo->hasReservedCallFrame(Fn))
- Offset += FFI->getMaxCallFrameSize();
-
- unsigned AlignMask = TFI.getStackAlignment() - 1;
+ if (MFI->adjustsStack() && TFI.hasReservedCallFrame(Fn))
+ Offset += MFI->getMaxCallFrameSize();
+
+ // Round up the size to a multiple of the alignment. If the function has
+ // any calls or alloca's, align to the target's StackAlignment value to
+ // ensure that the callee's frame or the alloca data is suitably aligned;
+ // otherwise, for leaf functions, align to the TransientStackAlignment
+ // value.
+ unsigned StackAlign;
+ if (MFI->adjustsStack() || MFI->hasVarSizedObjects() ||
+ (RegInfo->needsStackRealignment(Fn) && MFI->getObjectIndexEnd() != 0))
+ StackAlign = TFI.getStackAlignment();
+ else
+ StackAlign = TFI.getTransientStackAlignment();
+
+ // If the frame pointer is eliminated, all frame offsets will be relative to
+ // SP not FP. Align to MaxAlign so this works.
+ StackAlign = std::max(StackAlign, MaxAlign);
+ unsigned AlignMask = StackAlign - 1;
Offset = (Offset + AlignMask) & ~uint64_t(AlignMask);
}
// Update frame info to pretend that this is part of the stack...
- FFI->setStackSize(Offset+TFI.getOffsetOfLocalArea());
-
- // Remember the required stack alignment in case targets need it to perform
- // dynamic stack alignment.
- assert(FFI->getMaxAlignment() == MaxAlign &&
- "Stack alignment calculation broken!");
+ int64_t StackSize = Offset - LocalAreaOffset;
+ MFI->setStackSize(StackSize);
+ NumBytesStackSpace += StackSize;
}
-
/// insertPrologEpilogCode - Scan the function for modified callee saved
/// registers, insert spill code for these callee saved registers, then add
/// prolog and epilog code to the function.
///
void PEI::insertPrologEpilogCode(MachineFunction &Fn) {
+ const TargetFrameLowering &TFI = *Fn.getTarget().getFrameLowering();
+
// Add prologue to the function...
- Fn.getTarget().getRegisterInfo()->emitPrologue(Fn);
+ TFI.emitPrologue(Fn);
// Add epilogue to restore the callee-save registers in each exiting block
for (MachineFunction::iterator I = Fn.begin(), E = Fn.end(); I != E; ++I) {
// If last instruction is a return instruction, add an epilogue
- if (!I->empty() && I->back().getDesc().isReturn())
- Fn.getTarget().getRegisterInfo()->emitEpilogue(Fn, *I);
+ if (!I->empty() && I->back().isReturn())
+ TFI.emitEpilogue(Fn, *I);
}
-}
+ // Emit additional code that is required to support segmented stacks, if
+ // we've been asked for it. This, when linked with a runtime with support
+ // for segmented stacks (libgcc is one), will result in allocating stack
+ // space in small chunks instead of one large contiguous block.
+ if (Fn.shouldSplitStack())
+ TFI.adjustForSegmentedStacks(Fn);
+
+ // Emit additional code that is required to explicitly handle the stack in
+ // HiPE native code (if needed) when loaded in the Erlang/OTP runtime. The
+ // approach is rather similar to that of Segmented Stacks, but it uses a
+ // different conditional check and another BIF for allocating more stack
+ // space.
+ if (Fn.getFunction()->getCallingConv() == CallingConv::HiPE)
+ TFI.adjustForHiPEPrologue(Fn);
+}
/// replaceFrameIndices - Replace all MO_FrameIndex operands with physical
/// register references and actual offsets.
void PEI::replaceFrameIndices(MachineFunction &Fn) {
if (!Fn.getFrameInfo()->hasStackObjects()) return; // Nothing to do?
+ // Store SPAdj at exit of a basic block.
+ SmallVector<int, 8> SPState;
+ SPState.resize(Fn.getNumBlockIDs());
+ SmallPtrSet<MachineBasicBlock*, 8> Reachable;
+
+ // Iterate over the reachable blocks in DFS order.
+ for (df_ext_iterator<MachineFunction*, SmallPtrSet<MachineBasicBlock*, 8> >
+ DFI = df_ext_begin(&Fn, Reachable), DFE = df_ext_end(&Fn, Reachable);
+ DFI != DFE; ++DFI) {
+ int SPAdj = 0;
+ // Check the exit state of the DFS stack predecessor.
+ if (DFI.getPathLength() >= 2) {
+ MachineBasicBlock *StackPred = DFI.getPath(DFI.getPathLength() - 2);
+ assert(Reachable.count(StackPred) &&
+ "DFS stack predecessor is already visited.\n");
+ SPAdj = SPState[StackPred->getNumber()];
+ }
+ MachineBasicBlock *BB = *DFI;
+ replaceFrameIndices(BB, Fn, SPAdj);
+ SPState[BB->getNumber()] = SPAdj;
+ }
+
+ // Handle the unreachable blocks.
+ for (MachineFunction::iterator BB = Fn.begin(), E = Fn.end(); BB != E; ++BB) {
+ if (Reachable.count(BB))
+ // Already handled in DFS traversal.
+ continue;
+ int SPAdj = 0;
+ replaceFrameIndices(BB, Fn, SPAdj);
+ }
+}
+
+void PEI::replaceFrameIndices(MachineBasicBlock *BB, MachineFunction &Fn,
+ int &SPAdj) {
const TargetMachine &TM = Fn.getTarget();
assert(TM.getRegisterInfo() && "TM::getRegisterInfo() must be implemented!");
- const MRegisterInfo &MRI = *TM.getRegisterInfo();
- const TargetFrameInfo *TFI = TM.getFrameInfo();
+ const TargetInstrInfo &TII = *Fn.getTarget().getInstrInfo();
+ const TargetRegisterInfo &TRI = *TM.getRegisterInfo();
+ const TargetFrameLowering *TFI = TM.getFrameLowering();
bool StackGrowsDown =
- TFI->getStackGrowthDirection() == TargetFrameInfo::StackGrowsDown;
- int FrameSetupOpcode = MRI.getCallFrameSetupOpcode();
- int FrameDestroyOpcode = MRI.getCallFrameDestroyOpcode();
+ TFI->getStackGrowthDirection() == TargetFrameLowering::StackGrowsDown;
+ int FrameSetupOpcode = TII.getCallFrameSetupOpcode();
+ int FrameDestroyOpcode = TII.getCallFrameDestroyOpcode();
- for (MachineFunction::iterator BB = Fn.begin(), E = Fn.end(); BB != E; ++BB) {
- int SPAdj = 0; // SP offset due to call frame setup / destroy.
- if (RS) RS->enterBasicBlock(BB);
+ if (RS && !FrameIndexVirtualScavenging) RS->enterBasicBlock(BB);
+
+ for (MachineBasicBlock::iterator I = BB->begin(); I != BB->end(); ) {
+
+ if (I->getOpcode() == FrameSetupOpcode ||
+ I->getOpcode() == FrameDestroyOpcode) {
+ // Remember how much SP has been adjusted to create the call
+ // frame.
+ int Size = I->getOperand(0).getImm();
+
+ if ((!StackGrowsDown && I->getOpcode() == FrameSetupOpcode) ||
+ (StackGrowsDown && I->getOpcode() == FrameDestroyOpcode))
+ Size = -Size;
+
+ SPAdj += Size;
+
+ MachineBasicBlock::iterator PrevI = BB->end();
+ if (I != BB->begin()) PrevI = std::prev(I);
+ TFI->eliminateCallFramePseudoInstr(Fn, *BB, I);
+
+ // Visit the instructions created by eliminateCallFramePseudoInstr().
+ if (PrevI == BB->end())
+ I = BB->begin(); // The replaced instr was the first in the block.
+ else
+ I = std::next(PrevI);
+ continue;
+ }
+
+ MachineInstr *MI = I;
+ bool DoIncr = true;
+ for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
+ if (!MI->getOperand(i).isFI())
+ continue;
+
+ // Frame indicies in debug values are encoded in a target independent
+ // way with simply the frame index and offset rather than any
+ // target-specific addressing mode.
+ if (MI->isDebugValue()) {
+ assert(i == 0 && "Frame indicies can only appear as the first "
+ "operand of a DBG_VALUE machine instruction");
+ unsigned Reg;
+ MachineOperand &Offset = MI->getOperand(1);
+ Offset.setImm(Offset.getImm() +
+ TFI->getFrameIndexReference(
+ Fn, MI->getOperand(0).getIndex(), Reg));
+ MI->getOperand(0).ChangeToRegister(Reg, false /*isDef*/);
+ continue;
+ }
+
+ // Some instructions (e.g. inline asm instructions) can have
+ // multiple frame indices and/or cause eliminateFrameIndex
+ // to insert more than one instruction. We need the register
+ // scavenger to go through all of these instructions so that
+ // it can update its register information. We keep the
+ // iterator at the point before insertion so that we can
+ // revisit them in full.
+ bool AtBeginning = (I == BB->begin());
+ if (!AtBeginning) --I;
+
+ // If this instruction has a FrameIndex operand, we need to
+ // use that target machine register info object to eliminate
+ // it.
+ TRI.eliminateFrameIndex(MI, SPAdj, i,
+ FrameIndexVirtualScavenging ? nullptr : RS);
+
+ // Reset the iterator if we were at the beginning of the BB.
+ if (AtBeginning) {
+ I = BB->begin();
+ DoIncr = false;
+ }
+
+ MI = nullptr;
+ break;
+ }
+
+ if (DoIncr && I != BB->end()) ++I;
+
+ // Update register states.
+ if (RS && !FrameIndexVirtualScavenging && MI) RS->forward(MI);
+ }
+}
+
+/// scavengeFrameVirtualRegs - Replace all frame index virtual registers
+/// with physical registers. Use the register scavenger to find an
+/// appropriate register to use.
+///
+/// FIXME: Iterating over the instruction stream is unnecessary. We can simply
+/// iterate over the vreg use list, which at this point only contains machine
+/// operands for which eliminateFrameIndex need a new scratch reg.
+void PEI::scavengeFrameVirtualRegs(MachineFunction &Fn) {
+ // Run through the instructions and find any virtual registers.
+ for (MachineFunction::iterator BB = Fn.begin(),
+ E = Fn.end(); BB != E; ++BB) {
+ RS->enterBasicBlock(BB);
+
+ int SPAdj = 0;
+
+ // The instruction stream may change in the loop, so check BB->end()
+ // directly.
for (MachineBasicBlock::iterator I = BB->begin(); I != BB->end(); ) {
- MachineInstr *MI = I;
+ // We might end up here again with a NULL iterator if we scavenged a
+ // register for which we inserted spill code for definition by what was
+ // originally the first instruction in BB.
+ if (I == MachineBasicBlock::iterator(nullptr))
+ I = BB->begin();
- // Remember how much SP has been adjustment to create the call frame.
- if (I->getOpcode() == FrameSetupOpcode ||
- I->getOpcode() == FrameDestroyOpcode) {
- int Size = I->getOperand(0).getImm();
- if ((!StackGrowsDown && I->getOpcode() == FrameSetupOpcode) ||
- (StackGrowsDown && I->getOpcode() == FrameDestroyOpcode))
- Size = -Size;
- SPAdj += Size;
- MachineBasicBlock::iterator PrevI = prior(I);
- MRI.eliminateCallFramePseudoInstr(Fn, *BB, I);
- // Visit the instructions created by eliminateCallFramePseudoInstr().
- I = next(PrevI);
- MI = NULL;
- } else {
- I++;
- for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i)
- if (MI->getOperand(i).isFrameIndex()) {
- // If this instruction has a FrameIndex operand, we need to use that
- // target machine register info object to eliminate it.
- MRI.eliminateFrameIndex(MI, SPAdj, RS);
-
- // Revisit the instruction in full. Some instructions (e.g. inline
- // asm instructions) can have multiple frame indices.
- --I;
- MI = 0;
- break;
- }
+ MachineInstr *MI = I;
+ MachineBasicBlock::iterator J = std::next(I);
+ MachineBasicBlock::iterator P =
+ I == BB->begin() ? MachineBasicBlock::iterator(nullptr)
+ : std::prev(I);
+
+ // RS should process this instruction before we might scavenge at this
+ // location. This is because we might be replacing a virtual register
+ // defined by this instruction, and if so, registers killed by this
+ // instruction are available, and defined registers are not.
+ RS->forward(I);
+
+ for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
+ if (MI->getOperand(i).isReg()) {
+ MachineOperand &MO = MI->getOperand(i);
+ unsigned Reg = MO.getReg();
+ if (Reg == 0)
+ continue;
+ if (!TargetRegisterInfo::isVirtualRegister(Reg))
+ continue;
+
+ // When we first encounter a new virtual register, it
+ // must be a definition.
+ assert(MI->getOperand(i).isDef() &&
+ "frame index virtual missing def!");
+ // Scavenge a new scratch register
+ const TargetRegisterClass *RC = Fn.getRegInfo().getRegClass(Reg);
+ unsigned ScratchReg = RS->scavengeRegister(RC, J, SPAdj);
+
+ ++NumScavengedRegs;
+
+ // Replace this reference to the virtual register with the
+ // scratch register.
+ assert (ScratchReg && "Missing scratch register!");
+ Fn.getRegInfo().replaceRegWith(Reg, ScratchReg);
+
+ // Because this instruction was processed by the RS before this
+ // register was allocated, make sure that the RS now records the
+ // register as being used.
+ RS->setUsed(ScratchReg);
+ }
}
- // Update register states.
- if (RS && MI) RS->forward(MI);
+
+ // If the scavenger needed to use one of its spill slots, the
+ // spill code will have been inserted in between I and J. This is a
+ // problem because we need the spill code before I: Move I to just
+ // prior to J.
+ if (I != std::prev(J)) {
+ BB->splice(J, BB, I);
+
+ // Before we move I, we need to prepare the RS to visit I again.
+ // Specifically, RS will assert if it sees uses of registers that
+ // it believes are undefined. Because we have already processed
+ // register kills in I, when it visits I again, it will believe that
+ // those registers are undefined. To avoid this situation, unprocess
+ // the instruction I.
+ assert(RS->getCurrentPosition() == I &&
+ "The register scavenger has an unexpected position");
+ I = P;
+ RS->unprocess(P);
+ } else
+ ++I;
}
- assert(SPAdj == 0 && "Unbalanced call frame setup / destroy pairs?");
}
}