1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-23 19:23:23 +01:00
llvm-mirror/lib/Target/PowerPC/PPCMCInstLower.cpp

188 lines
6.1 KiB
C++
Raw Normal View History

//===-- PPCMCInstLower.cpp - Convert PPC MachineInstr to an MCInst --------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file contains code to lower PPC MachineInstrs to their corresponding
// MCInst records.
//
//===----------------------------------------------------------------------===//
#include "PPC.h"
[PowerPC] Clean up generation of ha16() / lo16() markers When targeting the Darwin assembler, we need to generate markers ha16() and lo16() to designate the high and low parts of a (symbolic) immediate. This is necessary not just for plain symbols, but also for certain symbolic expression, typically along the lines of ha16(A - B). The latter doesn't work when simply using VariantKind flags on the symbol reference. This is why the current back-end uses hacks (explicitly called out as such via multiple FIXMEs) in the symbolLo/symbolHi print methods. This patch uses target-defined MCExpr codes to represent the Darwin ha16/lo16 constructs, following along the lines of the equivalent solution used by the ARM back end to handle their :upper16: / :lower16: markers. This allows us to get rid of special handling both in the symbolLo/symbolHi print method and in the common code MCExpr::print routine. Instead, the ha16 / lo16 markers are printed simply in a custom print routine for the target MCExpr types. (As a result, the symbolLo/symbolHi print methods can now replaced by a single printS16ImmOperand routine that also handles symbolic operands.) The patch also provides a EvaluateAsRelocatableImpl routine to handle ha16/lo16 constructs. This is not actually used at the moment by any in-tree code, but is provided as it makes merging into David Fang's out-of-tree Mach-O object writer simpler. Since there is no longer any need to treat VK_PPC_GAS_HA16 and VK_PPC_DARWIN_HA16 differently, they are merged into a single VK_PPC_ADDR16_HA (and likewise for the _LO16 types). llvm-svn: 182616
2013-05-24 00:26:41 +02:00
#include "MCTargetDesc/PPCMCExpr.h"
#include "PPCSubtarget.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/ADT/Twine.h"
#include "llvm/CodeGen/AsmPrinter.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineModuleInfoImpls.h"
#include "llvm/IR/DataLayout.h"
2013-02-20 01:31:54 +01:00
#include "llvm/IR/GlobalValue.h"
#include "llvm/IR/Mangler.h"
#include "llvm/MC/MCAsmInfo.h"
#include "llvm/MC/MCExpr.h"
#include "llvm/MC/MCInst.h"
#include "llvm/Target/TargetLowering.h"
#include "llvm/Target/TargetLoweringObjectFile.h"
using namespace llvm;
static MachineModuleInfoMachO &getMachOMMI(AsmPrinter &AP) {
return AP.MMI->getObjFileInfo<MachineModuleInfoMachO>();
}
static MCSymbol *GetSymbolFromOperand(const MachineOperand &MO,
AsmPrinter &AP) {
const TargetMachine &TM = AP.TM;
Mangler &Mang = TM.getObjFileLowering()->getMangler();
const DataLayout &DL = AP.getDataLayout();
MCContext &Ctx = AP.OutContext;
SmallString<128> Name;
StringRef Suffix;
if (MO.getTargetFlags() & PPCII::MO_NLP_FLAG)
Suffix = "$non_lazy_ptr";
if (!Suffix.empty())
Name += DL.getPrivateGlobalPrefix();
if (!MO.isGlobal()) {
assert(MO.isSymbol() && "Isn't a symbol reference");
Mangler::getNameWithPrefix(Name, MO.getSymbolName(), DL);
} else {
const GlobalValue *GV = MO.getGlobal();
TM.getNameWithPrefix(Name, GV, Mang);
}
Name += Suffix;
MCSymbol *Sym = Ctx.getOrCreateSymbol(Name);
// If the symbol reference is actually to a non_lazy_ptr, not to the symbol,
// then add the suffix.
if (MO.getTargetFlags() & PPCII::MO_NLP_FLAG) {
MachineModuleInfoMachO &MachO = getMachOMMI(AP);
MachineModuleInfoImpl::StubValueTy &StubSym = MachO.getGVStubEntry(Sym);
if (!StubSym.getPointer()) {
assert(MO.isGlobal() && "Extern symbol not handled yet");
StubSym = MachineModuleInfoImpl::
StubValueTy(AP.getSymbol(MO.getGlobal()),
!MO.getGlobal()->hasInternalLinkage());
}
return Sym;
}
return Sym;
}
static MCOperand GetSymbolRef(const MachineOperand &MO, const MCSymbol *Symbol,
[PowerPC] Always use "assembler dialect" 1 A setting in MCAsmInfo defines the "assembler dialect" to use. This is used by common code to choose between alternatives in a multi-alternative GNU inline asm statement like the following: __asm__ ("{sfe|subfe} %0,%1,%2" : "=r" (out) : "r" (in1), "r" (in2)); The meaning of these dialects is platform specific, and GCC defines those for PowerPC to use dialect 0 for old-style (POWER) mnemonics and 1 for new-style (PowerPC) mnemonics, like in the example above. To be compatible with inline asm used with GCC, LLVM ought to do the same. Specifically, this means we should always use assembler dialect 1 since old-style mnemonics really aren't supported on any current platform. However, the current LLVM back-end uses: AssemblerDialect = 1; // New-Style mnemonics. in PPCMCAsmInfoDarwin, and AssemblerDialect = 0; // Old-Style mnemonics. in PPCLinuxMCAsmInfo. The Linux setting really isn't correct, we should be using new-style mnemonics everywhere. This is changed by this commit. Unfortunately, the setting of this variable is overloaded in the back-end to decide whether or not we are on a Darwin target. This is done in PPCInstPrinter (the "SyntaxVariant" is initialized from the MCAsmInfo AssemblerDialect setting), and also in PPCMCExpr. Setting AssemblerDialect to 1 for both Darwin and Linux no longer allows us to make this distinction. Instead, this patch uses the MCSubtargetInfo passed to createPPCMCInstPrinter to distinguish Darwin targets, and ignores the SyntaxVariant parameter. As to PPCMCExpr, this patch adds an explicit isDarwin argument that needs to be passed in by the caller when creating a target MCExpr. (To do so this patch implicitly also reverts commit 184441.) llvm-svn: 185858
2013-07-08 22:20:51 +02:00
AsmPrinter &Printer, bool isDarwin) {
MCContext &Ctx = Printer.OutContext;
MCSymbolRefExpr::VariantKind RefKind = MCSymbolRefExpr::VK_None;
unsigned access = MO.getTargetFlags() & PPCII::MO_ACCESS_MASK;
switch (access) {
case PPCII::MO_TPREL_LO:
RefKind = MCSymbolRefExpr::VK_PPC_TPREL_LO;
break;
case PPCII::MO_TPREL_HA:
RefKind = MCSymbolRefExpr::VK_PPC_TPREL_HA;
break;
case PPCII::MO_DTPREL_LO:
RefKind = MCSymbolRefExpr::VK_PPC_DTPREL_LO;
break;
case PPCII::MO_TLSLD_LO:
RefKind = MCSymbolRefExpr::VK_PPC_GOT_TLSLD_LO;
break;
case PPCII::MO_TOC_LO:
RefKind = MCSymbolRefExpr::VK_PPC_TOC_LO;
break;
case PPCII::MO_TLS:
RefKind = MCSymbolRefExpr::VK_PPC_TLS;
break;
[PowerPC] Clean up generation of ha16() / lo16() markers When targeting the Darwin assembler, we need to generate markers ha16() and lo16() to designate the high and low parts of a (symbolic) immediate. This is necessary not just for plain symbols, but also for certain symbolic expression, typically along the lines of ha16(A - B). The latter doesn't work when simply using VariantKind flags on the symbol reference. This is why the current back-end uses hacks (explicitly called out as such via multiple FIXMEs) in the symbolLo/symbolHi print methods. This patch uses target-defined MCExpr codes to represent the Darwin ha16/lo16 constructs, following along the lines of the equivalent solution used by the ARM back end to handle their :upper16: / :lower16: markers. This allows us to get rid of special handling both in the symbolLo/symbolHi print method and in the common code MCExpr::print routine. Instead, the ha16 / lo16 markers are printed simply in a custom print routine for the target MCExpr types. (As a result, the symbolLo/symbolHi print methods can now replaced by a single printS16ImmOperand routine that also handles symbolic operands.) The patch also provides a EvaluateAsRelocatableImpl routine to handle ha16/lo16 constructs. This is not actually used at the moment by any in-tree code, but is provided as it makes merging into David Fang's out-of-tree Mach-O object writer simpler. Since there is no longer any need to treat VK_PPC_GAS_HA16 and VK_PPC_DARWIN_HA16 differently, they are merged into a single VK_PPC_ADDR16_HA (and likewise for the _LO16 types). llvm-svn: 182616
2013-05-24 00:26:41 +02:00
}
if (MO.getTargetFlags() == PPCII::MO_PLT)
RefKind = MCSymbolRefExpr::VK_PLT;
const MCExpr *Expr = MCSymbolRefExpr::create(Symbol, RefKind, Ctx);
if (!MO.isJTI() && MO.getOffset())
Expr = MCBinaryExpr::createAdd(Expr,
MCConstantExpr::create(MO.getOffset(), Ctx),
Ctx);
// Subtract off the PIC base if required.
if (MO.getTargetFlags() & PPCII::MO_PIC_FLAG) {
const MachineFunction *MF = MO.getParent()->getParent()->getParent();
const MCExpr *PB = MCSymbolRefExpr::create(MF->getPICBaseSymbol(), Ctx);
Expr = MCBinaryExpr::createSub(Expr, PB, Ctx);
}
[PowerPC] Clean up generation of ha16() / lo16() markers When targeting the Darwin assembler, we need to generate markers ha16() and lo16() to designate the high and low parts of a (symbolic) immediate. This is necessary not just for plain symbols, but also for certain symbolic expression, typically along the lines of ha16(A - B). The latter doesn't work when simply using VariantKind flags on the symbol reference. This is why the current back-end uses hacks (explicitly called out as such via multiple FIXMEs) in the symbolLo/symbolHi print methods. This patch uses target-defined MCExpr codes to represent the Darwin ha16/lo16 constructs, following along the lines of the equivalent solution used by the ARM back end to handle their :upper16: / :lower16: markers. This allows us to get rid of special handling both in the symbolLo/symbolHi print method and in the common code MCExpr::print routine. Instead, the ha16 / lo16 markers are printed simply in a custom print routine for the target MCExpr types. (As a result, the symbolLo/symbolHi print methods can now replaced by a single printS16ImmOperand routine that also handles symbolic operands.) The patch also provides a EvaluateAsRelocatableImpl routine to handle ha16/lo16 constructs. This is not actually used at the moment by any in-tree code, but is provided as it makes merging into David Fang's out-of-tree Mach-O object writer simpler. Since there is no longer any need to treat VK_PPC_GAS_HA16 and VK_PPC_DARWIN_HA16 differently, they are merged into a single VK_PPC_ADDR16_HA (and likewise for the _LO16 types). llvm-svn: 182616
2013-05-24 00:26:41 +02:00
// Add ha16() / lo16() markers if required.
switch (access) {
case PPCII::MO_LO:
Expr = PPCMCExpr::createLo(Expr, isDarwin, Ctx);
break;
case PPCII::MO_HA:
Expr = PPCMCExpr::createHa(Expr, isDarwin, Ctx);
break;
[PowerPC] Clean up generation of ha16() / lo16() markers When targeting the Darwin assembler, we need to generate markers ha16() and lo16() to designate the high and low parts of a (symbolic) immediate. This is necessary not just for plain symbols, but also for certain symbolic expression, typically along the lines of ha16(A - B). The latter doesn't work when simply using VariantKind flags on the symbol reference. This is why the current back-end uses hacks (explicitly called out as such via multiple FIXMEs) in the symbolLo/symbolHi print methods. This patch uses target-defined MCExpr codes to represent the Darwin ha16/lo16 constructs, following along the lines of the equivalent solution used by the ARM back end to handle their :upper16: / :lower16: markers. This allows us to get rid of special handling both in the symbolLo/symbolHi print method and in the common code MCExpr::print routine. Instead, the ha16 / lo16 markers are printed simply in a custom print routine for the target MCExpr types. (As a result, the symbolLo/symbolHi print methods can now replaced by a single printS16ImmOperand routine that also handles symbolic operands.) The patch also provides a EvaluateAsRelocatableImpl routine to handle ha16/lo16 constructs. This is not actually used at the moment by any in-tree code, but is provided as it makes merging into David Fang's out-of-tree Mach-O object writer simpler. Since there is no longer any need to treat VK_PPC_GAS_HA16 and VK_PPC_DARWIN_HA16 differently, they are merged into a single VK_PPC_ADDR16_HA (and likewise for the _LO16 types). llvm-svn: 182616
2013-05-24 00:26:41 +02:00
}
return MCOperand::createExpr(Expr);
}
void llvm::LowerPPCMachineInstrToMCInst(const MachineInstr *MI, MCInst &OutMI,
[PowerPC] Always use "assembler dialect" 1 A setting in MCAsmInfo defines the "assembler dialect" to use. This is used by common code to choose between alternatives in a multi-alternative GNU inline asm statement like the following: __asm__ ("{sfe|subfe} %0,%1,%2" : "=r" (out) : "r" (in1), "r" (in2)); The meaning of these dialects is platform specific, and GCC defines those for PowerPC to use dialect 0 for old-style (POWER) mnemonics and 1 for new-style (PowerPC) mnemonics, like in the example above. To be compatible with inline asm used with GCC, LLVM ought to do the same. Specifically, this means we should always use assembler dialect 1 since old-style mnemonics really aren't supported on any current platform. However, the current LLVM back-end uses: AssemblerDialect = 1; // New-Style mnemonics. in PPCMCAsmInfoDarwin, and AssemblerDialect = 0; // Old-Style mnemonics. in PPCLinuxMCAsmInfo. The Linux setting really isn't correct, we should be using new-style mnemonics everywhere. This is changed by this commit. Unfortunately, the setting of this variable is overloaded in the back-end to decide whether or not we are on a Darwin target. This is done in PPCInstPrinter (the "SyntaxVariant" is initialized from the MCAsmInfo AssemblerDialect setting), and also in PPCMCExpr. Setting AssemblerDialect to 1 for both Darwin and Linux no longer allows us to make this distinction. Instead, this patch uses the MCSubtargetInfo passed to createPPCMCInstPrinter to distinguish Darwin targets, and ignores the SyntaxVariant parameter. As to PPCMCExpr, this patch adds an explicit isDarwin argument that needs to be passed in by the caller when creating a target MCExpr. (To do so this patch implicitly also reverts commit 184441.) llvm-svn: 185858
2013-07-08 22:20:51 +02:00
AsmPrinter &AP, bool isDarwin) {
OutMI.setOpcode(MI->getOpcode());
for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
const MachineOperand &MO = MI->getOperand(i);
MCOperand MCOp;
switch (MO.getType()) {
default:
MI->dump();
llvm_unreachable("unknown operand type");
case MachineOperand::MO_Register:
assert(!MO.getSubReg() && "Subregs should be eliminated!");
assert(MO.getReg() > PPC::NoRegister &&
MO.getReg() < PPC::NUM_TARGET_REGS &&
"Invalid register for this target!");
MCOp = MCOperand::createReg(MO.getReg());
break;
case MachineOperand::MO_Immediate:
MCOp = MCOperand::createImm(MO.getImm());
break;
case MachineOperand::MO_MachineBasicBlock:
MCOp = MCOperand::createExpr(MCSymbolRefExpr::create(
MO.getMBB()->getSymbol(), AP.OutContext));
break;
case MachineOperand::MO_GlobalAddress:
case MachineOperand::MO_ExternalSymbol:
[PowerPC] Always use "assembler dialect" 1 A setting in MCAsmInfo defines the "assembler dialect" to use. This is used by common code to choose between alternatives in a multi-alternative GNU inline asm statement like the following: __asm__ ("{sfe|subfe} %0,%1,%2" : "=r" (out) : "r" (in1), "r" (in2)); The meaning of these dialects is platform specific, and GCC defines those for PowerPC to use dialect 0 for old-style (POWER) mnemonics and 1 for new-style (PowerPC) mnemonics, like in the example above. To be compatible with inline asm used with GCC, LLVM ought to do the same. Specifically, this means we should always use assembler dialect 1 since old-style mnemonics really aren't supported on any current platform. However, the current LLVM back-end uses: AssemblerDialect = 1; // New-Style mnemonics. in PPCMCAsmInfoDarwin, and AssemblerDialect = 0; // Old-Style mnemonics. in PPCLinuxMCAsmInfo. The Linux setting really isn't correct, we should be using new-style mnemonics everywhere. This is changed by this commit. Unfortunately, the setting of this variable is overloaded in the back-end to decide whether or not we are on a Darwin target. This is done in PPCInstPrinter (the "SyntaxVariant" is initialized from the MCAsmInfo AssemblerDialect setting), and also in PPCMCExpr. Setting AssemblerDialect to 1 for both Darwin and Linux no longer allows us to make this distinction. Instead, this patch uses the MCSubtargetInfo passed to createPPCMCInstPrinter to distinguish Darwin targets, and ignores the SyntaxVariant parameter. As to PPCMCExpr, this patch adds an explicit isDarwin argument that needs to be passed in by the caller when creating a target MCExpr. (To do so this patch implicitly also reverts commit 184441.) llvm-svn: 185858
2013-07-08 22:20:51 +02:00
MCOp = GetSymbolRef(MO, GetSymbolFromOperand(MO, AP), AP, isDarwin);
break;
case MachineOperand::MO_JumpTableIndex:
[PowerPC] Always use "assembler dialect" 1 A setting in MCAsmInfo defines the "assembler dialect" to use. This is used by common code to choose between alternatives in a multi-alternative GNU inline asm statement like the following: __asm__ ("{sfe|subfe} %0,%1,%2" : "=r" (out) : "r" (in1), "r" (in2)); The meaning of these dialects is platform specific, and GCC defines those for PowerPC to use dialect 0 for old-style (POWER) mnemonics and 1 for new-style (PowerPC) mnemonics, like in the example above. To be compatible with inline asm used with GCC, LLVM ought to do the same. Specifically, this means we should always use assembler dialect 1 since old-style mnemonics really aren't supported on any current platform. However, the current LLVM back-end uses: AssemblerDialect = 1; // New-Style mnemonics. in PPCMCAsmInfoDarwin, and AssemblerDialect = 0; // Old-Style mnemonics. in PPCLinuxMCAsmInfo. The Linux setting really isn't correct, we should be using new-style mnemonics everywhere. This is changed by this commit. Unfortunately, the setting of this variable is overloaded in the back-end to decide whether or not we are on a Darwin target. This is done in PPCInstPrinter (the "SyntaxVariant" is initialized from the MCAsmInfo AssemblerDialect setting), and also in PPCMCExpr. Setting AssemblerDialect to 1 for both Darwin and Linux no longer allows us to make this distinction. Instead, this patch uses the MCSubtargetInfo passed to createPPCMCInstPrinter to distinguish Darwin targets, and ignores the SyntaxVariant parameter. As to PPCMCExpr, this patch adds an explicit isDarwin argument that needs to be passed in by the caller when creating a target MCExpr. (To do so this patch implicitly also reverts commit 184441.) llvm-svn: 185858
2013-07-08 22:20:51 +02:00
MCOp = GetSymbolRef(MO, AP.GetJTISymbol(MO.getIndex()), AP, isDarwin);
break;
case MachineOperand::MO_ConstantPoolIndex:
[PowerPC] Always use "assembler dialect" 1 A setting in MCAsmInfo defines the "assembler dialect" to use. This is used by common code to choose between alternatives in a multi-alternative GNU inline asm statement like the following: __asm__ ("{sfe|subfe} %0,%1,%2" : "=r" (out) : "r" (in1), "r" (in2)); The meaning of these dialects is platform specific, and GCC defines those for PowerPC to use dialect 0 for old-style (POWER) mnemonics and 1 for new-style (PowerPC) mnemonics, like in the example above. To be compatible with inline asm used with GCC, LLVM ought to do the same. Specifically, this means we should always use assembler dialect 1 since old-style mnemonics really aren't supported on any current platform. However, the current LLVM back-end uses: AssemblerDialect = 1; // New-Style mnemonics. in PPCMCAsmInfoDarwin, and AssemblerDialect = 0; // Old-Style mnemonics. in PPCLinuxMCAsmInfo. The Linux setting really isn't correct, we should be using new-style mnemonics everywhere. This is changed by this commit. Unfortunately, the setting of this variable is overloaded in the back-end to decide whether or not we are on a Darwin target. This is done in PPCInstPrinter (the "SyntaxVariant" is initialized from the MCAsmInfo AssemblerDialect setting), and also in PPCMCExpr. Setting AssemblerDialect to 1 for both Darwin and Linux no longer allows us to make this distinction. Instead, this patch uses the MCSubtargetInfo passed to createPPCMCInstPrinter to distinguish Darwin targets, and ignores the SyntaxVariant parameter. As to PPCMCExpr, this patch adds an explicit isDarwin argument that needs to be passed in by the caller when creating a target MCExpr. (To do so this patch implicitly also reverts commit 184441.) llvm-svn: 185858
2013-07-08 22:20:51 +02:00
MCOp = GetSymbolRef(MO, AP.GetCPISymbol(MO.getIndex()), AP, isDarwin);
break;
case MachineOperand::MO_BlockAddress:
[PowerPC] Always use "assembler dialect" 1 A setting in MCAsmInfo defines the "assembler dialect" to use. This is used by common code to choose between alternatives in a multi-alternative GNU inline asm statement like the following: __asm__ ("{sfe|subfe} %0,%1,%2" : "=r" (out) : "r" (in1), "r" (in2)); The meaning of these dialects is platform specific, and GCC defines those for PowerPC to use dialect 0 for old-style (POWER) mnemonics and 1 for new-style (PowerPC) mnemonics, like in the example above. To be compatible with inline asm used with GCC, LLVM ought to do the same. Specifically, this means we should always use assembler dialect 1 since old-style mnemonics really aren't supported on any current platform. However, the current LLVM back-end uses: AssemblerDialect = 1; // New-Style mnemonics. in PPCMCAsmInfoDarwin, and AssemblerDialect = 0; // Old-Style mnemonics. in PPCLinuxMCAsmInfo. The Linux setting really isn't correct, we should be using new-style mnemonics everywhere. This is changed by this commit. Unfortunately, the setting of this variable is overloaded in the back-end to decide whether or not we are on a Darwin target. This is done in PPCInstPrinter (the "SyntaxVariant" is initialized from the MCAsmInfo AssemblerDialect setting), and also in PPCMCExpr. Setting AssemblerDialect to 1 for both Darwin and Linux no longer allows us to make this distinction. Instead, this patch uses the MCSubtargetInfo passed to createPPCMCInstPrinter to distinguish Darwin targets, and ignores the SyntaxVariant parameter. As to PPCMCExpr, this patch adds an explicit isDarwin argument that needs to be passed in by the caller when creating a target MCExpr. (To do so this patch implicitly also reverts commit 184441.) llvm-svn: 185858
2013-07-08 22:20:51 +02:00
MCOp = GetSymbolRef(MO,AP.GetBlockAddressSymbol(MO.getBlockAddress()),AP,
isDarwin);
break;
case MachineOperand::MO_RegisterMask:
continue;
}
OutMI.addOperand(MCOp);
}
}