mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-08-23 17:29:19 +00:00
1469e29334
The new solution is to not use this lowering if there are any dynamic allocas in the current function. We know up front if there are dynamic allocas, but we don't know if we'll need to create stack temporaries with large alignment during lowering. Conservatively assume that we will need such temporaries. Reviewed By: hans Differential Revision: http://reviews.llvm.org/D5128 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@216775 91177308-0d34-0410-b5e6-96231b3b80d8
293 lines
11 KiB
C++
293 lines
11 KiB
C++
//===-- X86SelectionDAGInfo.cpp - X86 SelectionDAG Info -------------------===//
|
|
//
|
|
// The LLVM Compiler Infrastructure
|
|
//
|
|
// This file is distributed under the University of Illinois Open Source
|
|
// License. See LICENSE.TXT for details.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
//
|
|
// This file implements the X86SelectionDAGInfo class.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include "X86InstrInfo.h"
|
|
#include "X86ISelLowering.h"
|
|
#include "X86RegisterInfo.h"
|
|
#include "X86Subtarget.h"
|
|
#include "X86SelectionDAGInfo.h"
|
|
#include "llvm/CodeGen/SelectionDAG.h"
|
|
#include "llvm/IR/DerivedTypes.h"
|
|
#include "llvm/Target/TargetLowering.h"
|
|
|
|
using namespace llvm;
|
|
|
|
#define DEBUG_TYPE "x86-selectiondag-info"
|
|
|
|
X86SelectionDAGInfo::X86SelectionDAGInfo(const DataLayout &DL)
|
|
: TargetSelectionDAGInfo(&DL) {}
|
|
|
|
X86SelectionDAGInfo::~X86SelectionDAGInfo() {}
|
|
|
|
bool X86SelectionDAGInfo::isBaseRegConflictPossible(
|
|
SelectionDAG &DAG, ArrayRef<unsigned> ClobberSet) const {
|
|
// We cannot use TRI->hasBasePointer() until *after* we select all basic
|
|
// blocks. Legalization may introduce new stack temporaries with large
|
|
// alignment requirements. Fall back to generic code if there are any
|
|
// dynamic stack adjustments (hopefully rare) and the base pointer would
|
|
// conflict if we had to use it.
|
|
MachineFrameInfo *MFI = DAG.getMachineFunction().getFrameInfo();
|
|
if (!MFI->hasVarSizedObjects() && !MFI->hasInlineAsmWithSPAdjust())
|
|
return false;
|
|
|
|
const X86RegisterInfo *TRI = static_cast<const X86RegisterInfo *>(
|
|
DAG.getSubtarget().getRegisterInfo());
|
|
unsigned BaseReg = TRI->getBaseRegister();
|
|
for (unsigned R : ClobberSet)
|
|
if (BaseReg == R)
|
|
return true;
|
|
return false;
|
|
}
|
|
|
|
SDValue
|
|
X86SelectionDAGInfo::EmitTargetCodeForMemset(SelectionDAG &DAG, SDLoc dl,
|
|
SDValue Chain,
|
|
SDValue Dst, SDValue Src,
|
|
SDValue Size, unsigned Align,
|
|
bool isVolatile,
|
|
MachinePointerInfo DstPtrInfo) const {
|
|
ConstantSDNode *ConstantSize = dyn_cast<ConstantSDNode>(Size);
|
|
const X86Subtarget &Subtarget = DAG.getTarget().getSubtarget<X86Subtarget>();
|
|
|
|
#ifndef NDEBUG
|
|
// If the base register might conflict with our physical registers, bail out.
|
|
unsigned ClobberSet[] = {X86::RCX, X86::RAX, X86::RDI,
|
|
X86::ECX, X86::EAX, X86::EDI};
|
|
assert(!isBaseRegConflictPossible(DAG, ClobberSet));
|
|
#endif
|
|
|
|
// If to a segment-relative address space, use the default lowering.
|
|
if (DstPtrInfo.getAddrSpace() >= 256)
|
|
return SDValue();
|
|
|
|
// If not DWORD aligned or size is more than the threshold, call the library.
|
|
// The libc version is likely to be faster for these cases. It can use the
|
|
// address value and run time information about the CPU.
|
|
if ((Align & 3) != 0 || !ConstantSize ||
|
|
ConstantSize->getZExtValue() > Subtarget.getMaxInlineSizeThreshold()) {
|
|
// Check to see if there is a specialized entry-point for memory zeroing.
|
|
ConstantSDNode *V = dyn_cast<ConstantSDNode>(Src);
|
|
|
|
if (const char *bzeroEntry = V &&
|
|
V->isNullValue() ? Subtarget.getBZeroEntry() : nullptr) {
|
|
EVT IntPtr = DAG.getTargetLoweringInfo().getPointerTy();
|
|
Type *IntPtrTy = getDataLayout()->getIntPtrType(*DAG.getContext());
|
|
TargetLowering::ArgListTy Args;
|
|
TargetLowering::ArgListEntry Entry;
|
|
Entry.Node = Dst;
|
|
Entry.Ty = IntPtrTy;
|
|
Args.push_back(Entry);
|
|
Entry.Node = Size;
|
|
Args.push_back(Entry);
|
|
|
|
TargetLowering::CallLoweringInfo CLI(DAG);
|
|
CLI.setDebugLoc(dl).setChain(Chain)
|
|
.setCallee(CallingConv::C, Type::getVoidTy(*DAG.getContext()),
|
|
DAG.getExternalSymbol(bzeroEntry, IntPtr), std::move(Args),
|
|
0)
|
|
.setDiscardResult();
|
|
|
|
std::pair<SDValue,SDValue> CallResult = DAG.getTargetLoweringInfo().LowerCallTo(CLI);
|
|
return CallResult.second;
|
|
}
|
|
|
|
// Otherwise have the target-independent code call memset.
|
|
return SDValue();
|
|
}
|
|
|
|
uint64_t SizeVal = ConstantSize->getZExtValue();
|
|
SDValue InFlag;
|
|
EVT AVT;
|
|
SDValue Count;
|
|
ConstantSDNode *ValC = dyn_cast<ConstantSDNode>(Src);
|
|
unsigned BytesLeft = 0;
|
|
bool TwoRepStos = false;
|
|
if (ValC) {
|
|
unsigned ValReg;
|
|
uint64_t Val = ValC->getZExtValue() & 255;
|
|
|
|
// If the value is a constant, then we can potentially use larger sets.
|
|
switch (Align & 3) {
|
|
case 2: // WORD aligned
|
|
AVT = MVT::i16;
|
|
ValReg = X86::AX;
|
|
Val = (Val << 8) | Val;
|
|
break;
|
|
case 0: // DWORD aligned
|
|
AVT = MVT::i32;
|
|
ValReg = X86::EAX;
|
|
Val = (Val << 8) | Val;
|
|
Val = (Val << 16) | Val;
|
|
if (Subtarget.is64Bit() && ((Align & 0x7) == 0)) { // QWORD aligned
|
|
AVT = MVT::i64;
|
|
ValReg = X86::RAX;
|
|
Val = (Val << 32) | Val;
|
|
}
|
|
break;
|
|
default: // Byte aligned
|
|
AVT = MVT::i8;
|
|
ValReg = X86::AL;
|
|
Count = DAG.getIntPtrConstant(SizeVal);
|
|
break;
|
|
}
|
|
|
|
if (AVT.bitsGT(MVT::i8)) {
|
|
unsigned UBytes = AVT.getSizeInBits() / 8;
|
|
Count = DAG.getIntPtrConstant(SizeVal / UBytes);
|
|
BytesLeft = SizeVal % UBytes;
|
|
}
|
|
|
|
Chain = DAG.getCopyToReg(Chain, dl, ValReg, DAG.getConstant(Val, AVT),
|
|
InFlag);
|
|
InFlag = Chain.getValue(1);
|
|
} else {
|
|
AVT = MVT::i8;
|
|
Count = DAG.getIntPtrConstant(SizeVal);
|
|
Chain = DAG.getCopyToReg(Chain, dl, X86::AL, Src, InFlag);
|
|
InFlag = Chain.getValue(1);
|
|
}
|
|
|
|
Chain = DAG.getCopyToReg(Chain, dl, Subtarget.is64Bit() ? X86::RCX : X86::ECX,
|
|
Count, InFlag);
|
|
InFlag = Chain.getValue(1);
|
|
Chain = DAG.getCopyToReg(Chain, dl, Subtarget.is64Bit() ? X86::RDI : X86::EDI,
|
|
Dst, InFlag);
|
|
InFlag = Chain.getValue(1);
|
|
|
|
SDVTList Tys = DAG.getVTList(MVT::Other, MVT::Glue);
|
|
SDValue Ops[] = { Chain, DAG.getValueType(AVT), InFlag };
|
|
Chain = DAG.getNode(X86ISD::REP_STOS, dl, Tys, Ops);
|
|
|
|
if (TwoRepStos) {
|
|
InFlag = Chain.getValue(1);
|
|
Count = Size;
|
|
EVT CVT = Count.getValueType();
|
|
SDValue Left = DAG.getNode(ISD::AND, dl, CVT, Count,
|
|
DAG.getConstant((AVT == MVT::i64) ? 7 : 3, CVT));
|
|
Chain = DAG.getCopyToReg(Chain, dl, (CVT == MVT::i64) ? X86::RCX :
|
|
X86::ECX,
|
|
Left, InFlag);
|
|
InFlag = Chain.getValue(1);
|
|
Tys = DAG.getVTList(MVT::Other, MVT::Glue);
|
|
SDValue Ops[] = { Chain, DAG.getValueType(MVT::i8), InFlag };
|
|
Chain = DAG.getNode(X86ISD::REP_STOS, dl, Tys, Ops);
|
|
} else if (BytesLeft) {
|
|
// Handle the last 1 - 7 bytes.
|
|
unsigned Offset = SizeVal - BytesLeft;
|
|
EVT AddrVT = Dst.getValueType();
|
|
EVT SizeVT = Size.getValueType();
|
|
|
|
Chain = DAG.getMemset(Chain, dl,
|
|
DAG.getNode(ISD::ADD, dl, AddrVT, Dst,
|
|
DAG.getConstant(Offset, AddrVT)),
|
|
Src,
|
|
DAG.getConstant(BytesLeft, SizeVT),
|
|
Align, isVolatile, DstPtrInfo.getWithOffset(Offset));
|
|
}
|
|
|
|
// TODO: Use a Tokenfactor, as in memcpy, instead of a single chain.
|
|
return Chain;
|
|
}
|
|
|
|
SDValue
|
|
X86SelectionDAGInfo::EmitTargetCodeForMemcpy(SelectionDAG &DAG, SDLoc dl,
|
|
SDValue Chain, SDValue Dst, SDValue Src,
|
|
SDValue Size, unsigned Align,
|
|
bool isVolatile, bool AlwaysInline,
|
|
MachinePointerInfo DstPtrInfo,
|
|
MachinePointerInfo SrcPtrInfo) const {
|
|
// This requires the copy size to be a constant, preferably
|
|
// within a subtarget-specific limit.
|
|
ConstantSDNode *ConstantSize = dyn_cast<ConstantSDNode>(Size);
|
|
const X86Subtarget &Subtarget = DAG.getTarget().getSubtarget<X86Subtarget>();
|
|
if (!ConstantSize)
|
|
return SDValue();
|
|
uint64_t SizeVal = ConstantSize->getZExtValue();
|
|
if (!AlwaysInline && SizeVal > Subtarget.getMaxInlineSizeThreshold())
|
|
return SDValue();
|
|
|
|
/// If not DWORD aligned, it is more efficient to call the library. However
|
|
/// if calling the library is not allowed (AlwaysInline), then soldier on as
|
|
/// the code generated here is better than the long load-store sequence we
|
|
/// would otherwise get.
|
|
if (!AlwaysInline && (Align & 3) != 0)
|
|
return SDValue();
|
|
|
|
// If to a segment-relative address space, use the default lowering.
|
|
if (DstPtrInfo.getAddrSpace() >= 256 ||
|
|
SrcPtrInfo.getAddrSpace() >= 256)
|
|
return SDValue();
|
|
|
|
// If the base register might conflict with our physical registers, bail out.
|
|
unsigned ClobberSet[] = {X86::RCX, X86::RSI, X86::RDI,
|
|
X86::ECX, X86::ESI, X86::EDI};
|
|
if (isBaseRegConflictPossible(DAG, ClobberSet))
|
|
return SDValue();
|
|
|
|
MVT AVT;
|
|
if (Align & 1)
|
|
AVT = MVT::i8;
|
|
else if (Align & 2)
|
|
AVT = MVT::i16;
|
|
else if (Align & 4)
|
|
// DWORD aligned
|
|
AVT = MVT::i32;
|
|
else
|
|
// QWORD aligned
|
|
AVT = Subtarget.is64Bit() ? MVT::i64 : MVT::i32;
|
|
|
|
unsigned UBytes = AVT.getSizeInBits() / 8;
|
|
unsigned CountVal = SizeVal / UBytes;
|
|
SDValue Count = DAG.getIntPtrConstant(CountVal);
|
|
unsigned BytesLeft = SizeVal % UBytes;
|
|
|
|
SDValue InFlag;
|
|
Chain = DAG.getCopyToReg(Chain, dl, Subtarget.is64Bit() ? X86::RCX :
|
|
X86::ECX,
|
|
Count, InFlag);
|
|
InFlag = Chain.getValue(1);
|
|
Chain = DAG.getCopyToReg(Chain, dl, Subtarget.is64Bit() ? X86::RDI :
|
|
X86::EDI,
|
|
Dst, InFlag);
|
|
InFlag = Chain.getValue(1);
|
|
Chain = DAG.getCopyToReg(Chain, dl, Subtarget.is64Bit() ? X86::RSI :
|
|
X86::ESI,
|
|
Src, InFlag);
|
|
InFlag = Chain.getValue(1);
|
|
|
|
SDVTList Tys = DAG.getVTList(MVT::Other, MVT::Glue);
|
|
SDValue Ops[] = { Chain, DAG.getValueType(AVT), InFlag };
|
|
SDValue RepMovs = DAG.getNode(X86ISD::REP_MOVS, dl, Tys, Ops);
|
|
|
|
SmallVector<SDValue, 4> Results;
|
|
Results.push_back(RepMovs);
|
|
if (BytesLeft) {
|
|
// Handle the last 1 - 7 bytes.
|
|
unsigned Offset = SizeVal - BytesLeft;
|
|
EVT DstVT = Dst.getValueType();
|
|
EVT SrcVT = Src.getValueType();
|
|
EVT SizeVT = Size.getValueType();
|
|
Results.push_back(DAG.getMemcpy(Chain, dl,
|
|
DAG.getNode(ISD::ADD, dl, DstVT, Dst,
|
|
DAG.getConstant(Offset, DstVT)),
|
|
DAG.getNode(ISD::ADD, dl, SrcVT, Src,
|
|
DAG.getConstant(Offset, SrcVT)),
|
|
DAG.getConstant(BytesLeft, SizeVT),
|
|
Align, isVolatile, AlwaysInline,
|
|
DstPtrInfo.getWithOffset(Offset),
|
|
SrcPtrInfo.getWithOffset(Offset)));
|
|
}
|
|
|
|
return DAG.getNode(ISD::TokenFactor, dl, MVT::Other, Results);
|
|
}
|