blob: 7f8cb7f5e6ff238a16249da745891dc04532e359 [file] [log] [blame]
//===--- AArch64CallLowering.cpp - Call lowering --------------------------===//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
/// \file
/// This file implements the lowering of LLVM calls to machine code calls for
/// GlobalISel.
#include "AArch64CallLowering.h"
#include "AArch64ISelLowering.h"
#include "AArch64MachineFunctionInfo.h"
#include "AArch64Subtarget.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/CodeGen/Analysis.h"
#include "llvm/CodeGen/CallingConvLower.h"
#include "llvm/CodeGen/GlobalISel/MachineIRBuilder.h"
#include "llvm/CodeGen/GlobalISel/Utils.h"
#include "llvm/CodeGen/LowLevelType.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/CodeGen/MachineMemOperand.h"
#include "llvm/CodeGen/MachineOperand.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/CodeGen/TargetRegisterInfo.h"
#include "llvm/CodeGen/TargetSubtargetInfo.h"
#include "llvm/CodeGen/ValueTypes.h"
#include "llvm/IR/Argument.h"
#include "llvm/IR/Attributes.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/Type.h"
#include "llvm/IR/Value.h"
#include "llvm/Support/MachineValueType.h"
#include <algorithm>
#include <cassert>
#include <cstdint>
#include <iterator>
using namespace llvm;
AArch64CallLowering::AArch64CallLowering(const AArch64TargetLowering &TLI)
: CallLowering(&TLI) {}
namespace {
struct IncomingArgHandler : public CallLowering::ValueHandler {
IncomingArgHandler(MachineIRBuilder &MIRBuilder, MachineRegisterInfo &MRI,
CCAssignFn *AssignFn)
: ValueHandler(MIRBuilder, MRI, AssignFn), StackUsed(0) {}
unsigned getStackAddress(uint64_t Size, int64_t Offset,
MachinePointerInfo &MPO) override {
auto &MFI = MIRBuilder.getMF().getFrameInfo();
int FI = MFI.CreateFixedObject(Size, Offset, true);
MPO = MachinePointerInfo::getFixedStack(MIRBuilder.getMF(), FI);
unsigned AddrReg = MRI.createGenericVirtualRegister(LLT::pointer(0, 64));
MIRBuilder.buildFrameIndex(AddrReg, FI);
StackUsed = std::max(StackUsed, Size + Offset);
return AddrReg;
void assignValueToReg(unsigned ValVReg, unsigned PhysReg,
CCValAssign &VA) override {
switch (VA.getLocInfo()) {
MIRBuilder.buildCopy(ValVReg, PhysReg);
case CCValAssign::LocInfo::SExt:
case CCValAssign::LocInfo::ZExt:
case CCValAssign::LocInfo::AExt: {
auto Copy = MIRBuilder.buildCopy(LLT{VA.getLocVT()}, PhysReg);
MIRBuilder.buildTrunc(ValVReg, Copy);
void assignValueToAddress(unsigned ValVReg, unsigned Addr, uint64_t Size,
MachinePointerInfo &MPO, CCValAssign &VA) override {
// FIXME: Get alignment
auto MMO = MIRBuilder.getMF().getMachineMemOperand(
MPO, MachineMemOperand::MOLoad | MachineMemOperand::MOInvariant, Size,
MIRBuilder.buildLoad(ValVReg, Addr, *MMO);
/// How the physical register gets marked varies between formal
/// parameters (it's a basic-block live-in), and a call instruction
/// (it's an implicit-def of the BL).
virtual void markPhysRegUsed(unsigned PhysReg) = 0;
uint64_t StackUsed;
struct FormalArgHandler : public IncomingArgHandler {
FormalArgHandler(MachineIRBuilder &MIRBuilder, MachineRegisterInfo &MRI,
CCAssignFn *AssignFn)
: IncomingArgHandler(MIRBuilder, MRI, AssignFn) {}
void markPhysRegUsed(unsigned PhysReg) override {
struct CallReturnHandler : public IncomingArgHandler {
CallReturnHandler(MachineIRBuilder &MIRBuilder, MachineRegisterInfo &MRI,
MachineInstrBuilder MIB, CCAssignFn *AssignFn)
: IncomingArgHandler(MIRBuilder, MRI, AssignFn), MIB(MIB) {}
void markPhysRegUsed(unsigned PhysReg) override {
MIB.addDef(PhysReg, RegState::Implicit);
MachineInstrBuilder MIB;
struct OutgoingArgHandler : public CallLowering::ValueHandler {
OutgoingArgHandler(MachineIRBuilder &MIRBuilder, MachineRegisterInfo &MRI,
MachineInstrBuilder MIB, CCAssignFn *AssignFn,
CCAssignFn *AssignFnVarArg)
: ValueHandler(MIRBuilder, MRI, AssignFn), MIB(MIB),
AssignFnVarArg(AssignFnVarArg), StackSize(0) {}
unsigned getStackAddress(uint64_t Size, int64_t Offset,
MachinePointerInfo &MPO) override {
LLT p0 = LLT::pointer(0, 64);
LLT s64 = LLT::scalar(64);
unsigned SPReg = MRI.createGenericVirtualRegister(p0);
MIRBuilder.buildCopy(SPReg, AArch64::SP);
unsigned OffsetReg = MRI.createGenericVirtualRegister(s64);
MIRBuilder.buildConstant(OffsetReg, Offset);
unsigned AddrReg = MRI.createGenericVirtualRegister(p0);
MIRBuilder.buildGEP(AddrReg, SPReg, OffsetReg);
MPO = MachinePointerInfo::getStack(MIRBuilder.getMF(), Offset);
return AddrReg;
void assignValueToReg(unsigned ValVReg, unsigned PhysReg,
CCValAssign &VA) override {
MIB.addUse(PhysReg, RegState::Implicit);
unsigned ExtReg = extendRegister(ValVReg, VA);
MIRBuilder.buildCopy(PhysReg, ExtReg);
void assignValueToAddress(unsigned ValVReg, unsigned Addr, uint64_t Size,
MachinePointerInfo &MPO, CCValAssign &VA) override {
if (VA.getLocInfo() == CCValAssign::LocInfo::AExt) {
Size = VA.getLocVT().getSizeInBits() / 8;
ValVReg = MIRBuilder.buildAnyExt(LLT::scalar(Size * 8), ValVReg)
auto MMO = MIRBuilder.getMF().getMachineMemOperand(
MPO, MachineMemOperand::MOStore, Size, 1);
MIRBuilder.buildStore(ValVReg, Addr, *MMO);
bool assignArg(unsigned ValNo, MVT ValVT, MVT LocVT,
CCValAssign::LocInfo LocInfo,
const CallLowering::ArgInfo &Info,
CCState &State) override {
bool Res;
if (Info.IsFixed)
Res = AssignFn(ValNo, ValVT, LocVT, LocInfo, Info.Flags, State);
Res = AssignFnVarArg(ValNo, ValVT, LocVT, LocInfo, Info.Flags, State);
StackSize = State.getNextStackOffset();
return Res;
MachineInstrBuilder MIB;
CCAssignFn *AssignFnVarArg;
uint64_t StackSize;
} // namespace
void AArch64CallLowering::splitToValueTypes(
const ArgInfo &OrigArg, SmallVectorImpl<ArgInfo> &SplitArgs,
const DataLayout &DL, MachineRegisterInfo &MRI, CallingConv::ID CallConv,
const SplitArgTy &PerformArgSplit) const {
const AArch64TargetLowering &TLI = *getTLI<AArch64TargetLowering>();
LLVMContext &Ctx = OrigArg.Ty->getContext();
if (OrigArg.Ty->isVoidTy())
SmallVector<EVT, 4> SplitVTs;
SmallVector<uint64_t, 4> Offsets;
ComputeValueVTs(TLI, DL, OrigArg.Ty, SplitVTs, &Offsets, 0);
if (SplitVTs.size() == 1) {
// No splitting to do, but we want to replace the original type (e.g. [1 x
// double] -> double).
SplitArgs.emplace_back(OrigArg.Reg, SplitVTs[0].getTypeForEVT(Ctx),
OrigArg.Flags, OrigArg.IsFixed);
unsigned FirstRegIdx = SplitArgs.size();
bool NeedsRegBlock = TLI.functionArgumentNeedsConsecutiveRegisters(
OrigArg.Ty, CallConv, false);
for (auto SplitVT : SplitVTs) {
Type *SplitTy = SplitVT.getTypeForEVT(Ctx);
ArgInfo{MRI.createGenericVirtualRegister(getLLTForType(*SplitTy, DL)),
SplitTy, OrigArg.Flags, OrigArg.IsFixed});
if (NeedsRegBlock)
for (unsigned i = 0; i < Offsets.size(); ++i)
PerformArgSplit(SplitArgs[FirstRegIdx + i].Reg, Offsets[i] * 8);
bool AArch64CallLowering::lowerReturn(MachineIRBuilder &MIRBuilder,
const Value *Val,
ArrayRef<unsigned> VRegs) const {
auto MIB = MIRBuilder.buildInstrNoInsert(AArch64::RET_ReallyLR);
assert(((Val && !VRegs.empty()) || (!Val && VRegs.empty())) &&
"Return value without a vreg");
bool Success = true;
if (!VRegs.empty()) {
MachineFunction &MF = MIRBuilder.getMF();
const Function &F = MF.getFunction();
MachineRegisterInfo &MRI = MF.getRegInfo();
const AArch64TargetLowering &TLI = *getTLI<AArch64TargetLowering>();
CCAssignFn *AssignFn = TLI.CCAssignFnForReturn(F.getCallingConv());
auto &DL = F.getParent()->getDataLayout();
LLVMContext &Ctx = Val->getType()->getContext();
SmallVector<EVT, 4> SplitEVTs;
ComputeValueVTs(TLI, DL, Val->getType(), SplitEVTs);
assert(VRegs.size() == SplitEVTs.size() &&
"For each split Type there should be exactly one VReg.");
SmallVector<ArgInfo, 8> SplitArgs;
for (unsigned i = 0; i < SplitEVTs.size(); ++i) {
// We zero-extend i1s to i8.
unsigned CurVReg = VRegs[i];
if (MRI.getType(VRegs[i]).getSizeInBits() == 1) {
CurVReg = MIRBuilder.buildZExt(LLT::scalar(8), CurVReg)
ArgInfo CurArgInfo = ArgInfo{CurVReg, SplitEVTs[i].getTypeForEVT(Ctx)};
setArgFlags(CurArgInfo, AttributeList::ReturnIndex, DL, F);
splitToValueTypes(CurArgInfo, SplitArgs, DL, MRI, F.getCallingConv(),
[&](unsigned Reg, uint64_t Offset) {
MIRBuilder.buildExtract(Reg, CurVReg, Offset);
OutgoingArgHandler Handler(MIRBuilder, MRI, MIB, AssignFn, AssignFn);
Success = handleAssignments(MIRBuilder, SplitArgs, Handler);
return Success;
bool AArch64CallLowering::lowerFormalArguments(MachineIRBuilder &MIRBuilder,
const Function &F,
ArrayRef<unsigned> VRegs) const {
MachineFunction &MF = MIRBuilder.getMF();
MachineBasicBlock &MBB = MIRBuilder.getMBB();
MachineRegisterInfo &MRI = MF.getRegInfo();
auto &DL = F.getParent()->getDataLayout();
SmallVector<ArgInfo, 8> SplitArgs;
unsigned i = 0;
for (auto &Arg : F.args()) {
if (DL.getTypeStoreSize(Arg.getType()) == 0)
ArgInfo OrigArg{VRegs[i], Arg.getType()};
setArgFlags(OrigArg, i + AttributeList::FirstArgIndex, DL, F);
bool Split = false;
LLT Ty = MRI.getType(VRegs[i]);
unsigned Dst = VRegs[i];
splitToValueTypes(OrigArg, SplitArgs, DL, MRI, F.getCallingConv(),
[&](unsigned Reg, uint64_t Offset) {
if (!Split) {
Split = true;
Dst = MRI.createGenericVirtualRegister(Ty);
unsigned Tmp = MRI.createGenericVirtualRegister(Ty);
MIRBuilder.buildInsert(Tmp, Dst, Reg, Offset);
Dst = Tmp;
if (Dst != VRegs[i])
MIRBuilder.buildCopy(VRegs[i], Dst);
if (!MBB.empty())
const AArch64TargetLowering &TLI = *getTLI<AArch64TargetLowering>();
CCAssignFn *AssignFn =
TLI.CCAssignFnForCall(F.getCallingConv(), /*IsVarArg=*/false);
FormalArgHandler Handler(MIRBuilder, MRI, AssignFn);
if (!handleAssignments(MIRBuilder, SplitArgs, Handler))
return false;
if (F.isVarArg()) {
if (!MF.getSubtarget<AArch64Subtarget>().isTargetDarwin()) {
// FIXME: we need to reimplement saveVarArgsRegisters from
// AArch64ISelLowering.
return false;
// We currently pass all varargs at 8-byte alignment.
uint64_t StackOffset = alignTo(Handler.StackUsed, 8);
auto &MFI = MIRBuilder.getMF().getFrameInfo();
AArch64FunctionInfo *FuncInfo = MF.getInfo<AArch64FunctionInfo>();
FuncInfo->setVarArgsStackIndex(MFI.CreateFixedObject(4, StackOffset, true));
auto &Subtarget = MF.getSubtarget<AArch64Subtarget>();
if (Subtarget.hasCustomCallingConv())
// Move back to the end of the basic block.
return true;
bool AArch64CallLowering::lowerCall(MachineIRBuilder &MIRBuilder,
CallingConv::ID CallConv,
const MachineOperand &Callee,
const ArgInfo &OrigRet,
ArrayRef<ArgInfo> OrigArgs) const {
MachineFunction &MF = MIRBuilder.getMF();
const Function &F = MF.getFunction();
MachineRegisterInfo &MRI = MF.getRegInfo();
auto &DL = F.getParent()->getDataLayout();
SmallVector<ArgInfo, 8> SplitArgs;
for (auto &OrigArg : OrigArgs) {
splitToValueTypes(OrigArg, SplitArgs, DL, MRI, CallConv,
[&](unsigned Reg, uint64_t Offset) {
MIRBuilder.buildExtract(Reg, OrigArg.Reg, Offset);
// Find out which ABI gets to decide where things go.
const AArch64TargetLowering &TLI = *getTLI<AArch64TargetLowering>();
CCAssignFn *AssignFnFixed =
TLI.CCAssignFnForCall(CallConv, /*IsVarArg=*/false);
CCAssignFn *AssignFnVarArg =
TLI.CCAssignFnForCall(CallConv, /*IsVarArg=*/true);
auto CallSeqStart = MIRBuilder.buildInstr(AArch64::ADJCALLSTACKDOWN);
// Create a temporarily-floating call instruction so we can add the implicit
// uses of arg registers.
auto MIB = MIRBuilder.buildInstrNoInsert(Callee.isReg() ? AArch64::BLR
: AArch64::BL);
// Tell the call which registers are clobbered.
auto TRI = MF.getSubtarget<AArch64Subtarget>().getRegisterInfo();
const uint32_t *Mask = TRI->getCallPreservedMask(MF, F.getCallingConv());
if (MF.getSubtarget<AArch64Subtarget>().hasCustomCallingConv())
TRI->UpdateCustomCallPreservedMask(MF, &Mask);
if (TRI->isAnyArgRegReserved(MF))
// Do the actual argument marshalling.
SmallVector<unsigned, 8> PhysRegs;
OutgoingArgHandler Handler(MIRBuilder, MRI, MIB, AssignFnFixed,
if (!handleAssignments(MIRBuilder, SplitArgs, Handler))
return false;
// Now we can add the actual call instruction to the correct basic block.
// If Callee is a reg, since it is used by a target specific
// instruction, it must have a register class matching the
// constraint of that instruction.
if (Callee.isReg())
MF, *TRI, MRI, *MF.getSubtarget().getInstrInfo(),
*MF.getSubtarget().getRegBankInfo(), *MIB, MIB->getDesc(), Callee, 0));
// Finally we can copy the returned value back into its virtual-register. In
// symmetry with the arugments, the physical register must be an
// implicit-define of the call instruction.
CCAssignFn *RetAssignFn = TLI.CCAssignFnForReturn(F.getCallingConv());
if (OrigRet.Reg) {
SmallVector<uint64_t, 8> RegOffsets;
SmallVector<unsigned, 8> SplitRegs;
splitToValueTypes(OrigRet, SplitArgs, DL, MRI, F.getCallingConv(),
[&](unsigned Reg, uint64_t Offset) {
CallReturnHandler Handler(MIRBuilder, MRI, MIB, RetAssignFn);
if (!handleAssignments(MIRBuilder, SplitArgs, Handler))
return false;
if (!RegOffsets.empty())
MIRBuilder.buildSequence(OrigRet.Reg, SplitRegs, RegOffsets);
return true;