-//=====---- ARMSubtarget.h - Define Subtarget for the ARM -----*- C++ -*--====//
+//===-- ARMSubtarget.h - Define Subtarget for the ARM ----------*- C++ -*--===//
//
// The LLVM Compiler Infrastructure
//
#ifndef ARMSUBTARGET_H
#define ARMSUBTARGET_H
+
+#include "ARMFrameLowering.h"
+#include "ARMISelLowering.h"
+#include "ARMInstrInfo.h"
+#include "ARMSelectionDAGInfo.h"
+#include "ARMSubtarget.h"
+#include "Thumb1FrameLowering.h"
+#include "Thumb1InstrInfo.h"
+#include "Thumb2InstrInfo.h"
#include "MCTargetDesc/ARMMCTargetDesc.h"
-#include "llvm/Target/TargetSubtargetInfo.h"
-#include "llvm/MC/MCInstrItineraries.h"
#include "llvm/ADT/Triple.h"
+#include "llvm/IR/DataLayout.h"
+#include "llvm/MC/MCInstrItineraries.h"
+#include "llvm/Target/TargetSubtargetInfo.h"
#include <string>
#define GET_SUBTARGETINFO_HEADER
namespace llvm {
class GlobalValue;
class StringRef;
+class TargetOptions;
class ARMSubtarget : public ARMGenSubtargetInfo {
protected:
enum ARMProcFamilyEnum {
- Others, CortexA8, CortexA9
+ Others, CortexA5, CortexA7, CortexA8, CortexA9, CortexA12, CortexA15,
+ CortexR5, Swift, CortexA53, CortexA57, Krait
+ };
+ enum ARMProcClassEnum {
+ None, AClass, RClass, MClass
};
/// ARMProcFamily - ARM processor family: Cortex-A8, Cortex-A9, and others.
ARMProcFamilyEnum ARMProcFamily;
- /// HasV4TOps, HasV5TOps, HasV5TEOps, HasV6Ops, HasV6T2Ops, HasV7Ops -
+ /// ARMProcClass - ARM processor class: None, AClass, RClass or MClass.
+ ARMProcClassEnum ARMProcClass;
+
+ /// HasV4TOps, HasV5TOps, HasV5TEOps,
+ /// HasV6Ops, HasV6MOps, HasV6T2Ops, HasV7Ops, HasV8Ops -
/// Specify whether target support specific ARM ISA variants.
bool HasV4TOps;
bool HasV5TOps;
bool HasV5TEOps;
bool HasV6Ops;
+ bool HasV6MOps;
bool HasV6T2Ops;
bool HasV7Ops;
+ bool HasV8Ops;
- /// HasVFPv2, HasVFPv3, HasNEON - Specify what floating point ISAs are
- /// supported.
+ /// HasVFPv2, HasVFPv3, HasVFPv4, HasFPARMv8, HasNEON - Specify what
+ /// floating point ISAs are supported.
bool HasVFPv2;
bool HasVFPv3;
+ bool HasVFPv4;
+ bool HasFPARMv8;
bool HasNEON;
/// UseNEONForSinglePrecisionFP - if the NEONFP attribute has been
/// determine if NEON should actually be used.
bool UseNEONForSinglePrecisionFP;
+ /// UseMulOps - True if non-microcoded fused integer multiply-add and
+ /// multiply-subtract instructions should be used.
+ bool UseMulOps;
+
/// SlowFPVMLx - If the VFP2 / NEON instructions are available, indicates
/// whether the FP VML[AS] instructions are slow (if so, don't use them).
bool SlowFPVMLx;
/// InThumbMode - True if compiling for Thumb, false for ARM.
bool InThumbMode;
- /// InNaClMode - True if targeting Native Client
- bool InNaClMode;
-
/// HasThumb2 - True if Thumb2 instructions are supported.
bool HasThumb2;
- /// IsMClass - True if the subtarget belongs to the 'M' profile of CPUs -
- /// v6m, v7m for example.
- bool IsMClass;
-
/// NoARM - True if subtarget does not support ARM mode execution.
bool NoARM;
- /// PostRAScheduler - True if using post-register-allocation scheduler.
- bool PostRAScheduler;
-
/// IsR9Reserved - True if R9 is a not available as general purpose register.
bool IsR9Reserved;
/// imms (including global addresses).
bool UseMovt;
+ /// SupportsTailCall - True if the OS supports tail call. The dynamic linker
+ /// must be able to synthesize call stubs for interworking between ARM and
+ /// Thumb.
+ bool SupportsTailCall;
+
/// HasFP16 - True if subtarget supports half-precision FP (We support VFP+HF
/// only so far)
bool HasFP16;
/// HasHardwareDivide - True if subtarget supports [su]div
bool HasHardwareDivide;
+ /// HasHardwareDivideInARM - True if subtarget supports [su]div in ARM mode
+ bool HasHardwareDivideInARM;
+
/// HasT2ExtractPack - True if subtarget supports thumb2 extract/pack
/// instructions.
bool HasT2ExtractPack;
/// CPSR setting instruction.
bool AvoidCPSRPartialUpdate;
+ /// AvoidMOVsShifterOperand - If true, codegen should avoid using flag setting
+ /// movs with shifter operand (i.e. asr, lsl, lsr).
+ bool AvoidMOVsShifterOperand;
+
+ /// HasRAS - Some processors perform return stack prediction. CodeGen should
+ /// avoid issue "normal" call instructions to callees which do not return.
+ bool HasRAS;
+
/// HasMPExtension - True if the subtarget supports Multiprocessing
/// extension (ARMv7 only).
bool HasMPExtension;
+ /// HasVirtualization - True if the subtarget supports the Virtualization
+ /// extension.
+ bool HasVirtualization;
+
/// FPOnlySP - If true, the floating point unit only supports single
/// precision.
bool FPOnlySP;
+ /// If true, the processor supports the Performance Monitor Extensions. These
+ /// include a generic cycle-counter as well as more fine-grained (often
+ /// implementation-specific) events.
+ bool HasPerfMon;
+
+ /// HasTrustZone - if true, processor supports TrustZone security extensions
+ bool HasTrustZone;
+
+ /// HasCrypto - if true, processor supports Cryptography extensions
+ bool HasCrypto;
+
+ /// HasCRC - if true, processor supports CRC instructions
+ bool HasCRC;
+
+ /// If true, the instructions "vmov.i32 d0, #0" and "vmov.i32 q0, #0" are
+ /// particularly effective at zeroing a VFP register.
+ bool HasZeroCycleZeroing;
+
/// AllowsUnalignedMem - If true, the subtarget allows unaligned memory
/// accesses for some types. For details, see
- /// ARMTargetLowering::allowsUnalignedMemoryAccesses().
+ /// ARMTargetLowering::allowsMisalignedMemoryAccesses().
bool AllowsUnalignedMem;
+ /// RestrictIT - If true, the subtarget disallows generation of deprecated IT
+ /// blocks to conform to ARMv8 rule.
+ bool RestrictIT;
+
/// Thumb2DSP - If true, the subtarget supports the v7 DSP (saturating arith
/// and such) instructions in Thumb2 code.
bool Thumb2DSP;
+ /// NaCl TRAP instruction is generated instead of the regular TRAP.
+ bool UseNaClTrap;
+
+ /// Target machine allowed unsafe FP math (such as use of NEON fp)
+ bool UnsafeFPMath;
+
/// stackAlignment - The minimum alignment known to hold of the stack frame on
/// entry to the function and which must be maintained by every function.
unsigned stackAlignment;
/// CPUString - String name of used CPU.
std::string CPUString;
+ /// IsLittle - The target is Little Endian
+ bool IsLittle;
+
/// TargetTriple - What processor and OS we're targeting.
Triple TargetTriple;
+ /// SchedModel - Processor specific instruction costs.
+ const MCSchedModel *SchedModel;
+
/// Selected instruction itineraries (one entry per itinerary class.)
InstrItineraryData InstrItins;
- public:
- enum {
- isELF, isDarwin
- } TargetType;
+ /// Options passed via command line that could influence the target
+ const TargetOptions &Options;
+ public:
enum {
+ ARM_ABI_UNKNOWN,
ARM_ABI_APCS,
ARM_ABI_AAPCS // ARM EABI
} TargetABI;
/// of the specified triple.
///
ARMSubtarget(const std::string &TT, const std::string &CPU,
- const std::string &FS);
+ const std::string &FS, TargetMachine &TM, bool IsLittle,
+ const TargetOptions &Options);
/// getMaxInlineSizeThreshold - Returns the maximum memset / memcpy size
/// that still makes it profitable to inline the call.
unsigned getMaxInlineSizeThreshold() const {
- // FIXME: For now, we don't lower memcpy's to loads / stores for Thumb1.
- // Change this once Thumb1 ldmia / stmia support is added.
- return isThumb1Only() ? 0 : 64;
+ return 64;
}
/// ParseSubtargetFeatures - Parses features string setting specified
/// subtarget options. Definition of function is auto generated by tblgen.
void ParseSubtargetFeatures(StringRef CPU, StringRef FS);
+ /// \brief Reset the features for the ARM target.
+ void resetSubtargetFeatures(const MachineFunction *MF) override;
+
+ /// initializeSubtargetDependencies - Initializes using a CPU and feature string
+ /// so that we can use initializer lists for subtarget initialization.
+ ARMSubtarget &initializeSubtargetDependencies(StringRef CPU, StringRef FS);
+
+ const DataLayout *getDataLayout() const override { return &DL; }
+ const ARMSelectionDAGInfo *getSelectionDAGInfo() const override {
+ return &TSInfo;
+ }
+ const ARMBaseInstrInfo *getInstrInfo() const override {
+ return InstrInfo.get();
+ }
+ const ARMTargetLowering *getTargetLowering() const override {
+ return &TLInfo;
+ }
+ const ARMFrameLowering *getFrameLowering() const override {
+ return FrameLowering.get();
+ }
+ const ARMBaseRegisterInfo *getRegisterInfo() const override {
+ return &InstrInfo->getRegisterInfo();
+ }
+
+private:
+ const DataLayout DL;
+ ARMSelectionDAGInfo TSInfo;
+ // Either Thumb1InstrInfo or Thumb2InstrInfo.
+ std::unique_ptr<ARMBaseInstrInfo> InstrInfo;
+ ARMTargetLowering TLInfo;
+ // Either Thumb1FrameLowering or ARMFrameLowering.
+ std::unique_ptr<ARMFrameLowering> FrameLowering;
+
+ void initializeEnvironment();
+ void resetSubtargetFeatures(StringRef CPU, StringRef FS);
+public:
void computeIssueWidth();
bool hasV4TOps() const { return HasV4TOps; }
bool hasV5TOps() const { return HasV5TOps; }
bool hasV5TEOps() const { return HasV5TEOps; }
bool hasV6Ops() const { return HasV6Ops; }
+ bool hasV6MOps() const { return HasV6MOps; }
bool hasV6T2Ops() const { return HasV6T2Ops; }
bool hasV7Ops() const { return HasV7Ops; }
+ bool hasV8Ops() const { return HasV8Ops; }
+ bool isCortexA5() const { return ARMProcFamily == CortexA5; }
+ bool isCortexA7() const { return ARMProcFamily == CortexA7; }
bool isCortexA8() const { return ARMProcFamily == CortexA8; }
bool isCortexA9() const { return ARMProcFamily == CortexA9; }
+ bool isCortexA15() const { return ARMProcFamily == CortexA15; }
+ bool isSwift() const { return ARMProcFamily == Swift; }
+ bool isCortexM3() const { return CPUString == "cortex-m3"; }
+ bool isLikeA9() const { return isCortexA9() || isCortexA15() || isKrait(); }
+ bool isCortexR5() const { return ARMProcFamily == CortexR5; }
+ bool isKrait() const { return ARMProcFamily == Krait; }
bool hasARMOps() const { return !NoARM; }
bool hasVFP2() const { return HasVFPv2; }
bool hasVFP3() const { return HasVFPv3; }
+ bool hasVFP4() const { return HasVFPv4; }
+ bool hasFPARMv8() const { return HasFPARMv8; }
bool hasNEON() const { return HasNEON; }
+ bool hasCrypto() const { return HasCrypto; }
+ bool hasCRC() const { return HasCRC; }
+ bool hasVirtualization() const { return HasVirtualization; }
bool useNEONForSinglePrecisionFP() const {
return hasNEON() && UseNEONForSinglePrecisionFP; }
bool hasDivide() const { return HasHardwareDivide; }
+ bool hasDivideInARMMode() const { return HasHardwareDivideInARM; }
bool hasT2ExtractPack() const { return HasT2ExtractPack; }
bool hasDataBarrier() const { return HasDataBarrier; }
+ bool hasAnyDataBarrier() const {
+ return HasDataBarrier || (hasV6Ops() && !isThumb());
+ }
+ bool useMulOps() const { return UseMulOps; }
bool useFPVMLx() const { return !SlowFPVMLx; }
bool hasVMLxForwarding() const { return HasVMLxForwarding; }
bool isFPBrccSlow() const { return SlowFPBrcc; }
bool isFPOnlySP() const { return FPOnlySP; }
+ bool hasPerfMon() const { return HasPerfMon; }
+ bool hasTrustZone() const { return HasTrustZone; }
+ bool hasZeroCycleZeroing() const { return HasZeroCycleZeroing; }
bool prefers32BitThumb() const { return Pref32BitThumb; }
bool avoidCPSRPartialUpdate() const { return AvoidCPSRPartialUpdate; }
+ bool avoidMOVsShifterOperand() const { return AvoidMOVsShifterOperand; }
+ bool hasRAS() const { return HasRAS; }
bool hasMPExtension() const { return HasMPExtension; }
bool hasThumb2DSP() const { return Thumb2DSP; }
+ bool useNaClTrap() const { return UseNaClTrap; }
bool hasFP16() const { return HasFP16; }
bool hasD16() const { return HasD16; }
const Triple &getTargetTriple() const { return TargetTriple; }
bool isTargetDarwin() const { return TargetTriple.isOSDarwin(); }
- bool isTargetNaCl() const {
- return TargetTriple.getOS() == Triple::NativeClient;
+ bool isTargetIOS() const { return TargetTriple.isiOS(); }
+ bool isTargetLinux() const { return TargetTriple.isOSLinux(); }
+ bool isTargetNaCl() const { return TargetTriple.isOSNaCl(); }
+ bool isTargetNetBSD() const { return TargetTriple.getOS() == Triple::NetBSD; }
+ bool isTargetWindows() const { return TargetTriple.isOSWindows(); }
+
+ bool isTargetCOFF() const { return TargetTriple.isOSBinFormatCOFF(); }
+ bool isTargetELF() const { return TargetTriple.isOSBinFormatELF(); }
+ bool isTargetMachO() const { return TargetTriple.isOSBinFormatMachO(); }
+
+ // ARM EABI is the bare-metal EABI described in ARM ABI documents and
+ // can be accessed via -target arm-none-eabi. This is NOT GNUEABI.
+ // FIXME: Add a flag for bare-metal for that target and set Triple::EABI
+ // even for GNUEABI, so we can make a distinction here and still conform to
+ // the EABI on GNU (and Android) mode. This requires change in Clang, too.
+ // FIXME: The Darwin exception is temporary, while we move users to
+ // "*-*-*-macho" triples as quickly as possible.
+ bool isTargetAEABI() const {
+ return (TargetTriple.getEnvironment() == Triple::EABI ||
+ TargetTriple.getEnvironment() == Triple::EABIHF) &&
+ !isTargetDarwin() && !isTargetWindows();
+ }
+
+ // ARM Targets that support EHABI exception handling standard
+ // Darwin uses SjLj. Other targets might need more checks.
+ bool isTargetEHABICompatible() const {
+ return (TargetTriple.getEnvironment() == Triple::EABI ||
+ TargetTriple.getEnvironment() == Triple::GNUEABI ||
+ TargetTriple.getEnvironment() == Triple::EABIHF ||
+ TargetTriple.getEnvironment() == Triple::GNUEABIHF ||
+ TargetTriple.getEnvironment() == Triple::Android) &&
+ !isTargetDarwin() && !isTargetWindows();
}
- bool isTargetELF() const { return !isTargetDarwin(); }
- bool isAPCS_ABI() const { return TargetABI == ARM_ABI_APCS; }
- bool isAAPCS_ABI() const { return TargetABI == ARM_ABI_AAPCS; }
+ bool isTargetHardFloat() const {
+ // FIXME: this is invalid for WindowsCE
+ return TargetTriple.getEnvironment() == Triple::GNUEABIHF ||
+ TargetTriple.getEnvironment() == Triple::EABIHF ||
+ isTargetWindows();
+ }
+ bool isTargetAndroid() const {
+ return TargetTriple.getEnvironment() == Triple::Android;
+ }
+
+ bool isAPCS_ABI() const {
+ assert(TargetABI != ARM_ABI_UNKNOWN);
+ return TargetABI == ARM_ABI_APCS;
+ }
+ bool isAAPCS_ABI() const {
+ assert(TargetABI != ARM_ABI_UNKNOWN);
+ return TargetABI == ARM_ABI_AAPCS;
+ }
bool isThumb() const { return InThumbMode; }
bool isThumb1Only() const { return InThumbMode && !HasThumb2; }
bool isThumb2() const { return InThumbMode && HasThumb2; }
bool hasThumb2() const { return HasThumb2; }
- bool isMClass() const { return IsMClass; }
- bool isARClass() const { return !IsMClass; }
+ bool isMClass() const { return ARMProcClass == MClass; }
+ bool isRClass() const { return ARMProcClass == RClass; }
+ bool isAClass() const { return ARMProcClass == AClass; }
bool isR9Reserved() const { return IsR9Reserved; }
- bool useMovt() const { return UseMovt && hasV6T2Ops(); }
+ bool useMovt(const MachineFunction &MF) const;
+
+ bool supportsTailCall() const { return SupportsTailCall; }
bool allowsUnalignedMem() const { return AllowsUnalignedMem; }
+ bool restrictIT() const { return RestrictIT; }
+
const std::string & getCPUString() const { return CPUString; }
+ bool isLittle() const { return IsLittle; }
+
unsigned getMispredictionPenalty() const;
- /// enablePostRAScheduler - True at 'More' optimization.
- bool enablePostRAScheduler(CodeGenOpt::Level OptLevel,
- TargetSubtargetInfo::AntiDepBreakMode& Mode,
- RegClassVector& CriticalPathRCs) const;
+ /// This function returns true if the target has sincos() routine in its
+ /// compiler runtime or math libraries.
+ bool hasSinCos() const;
+
+ /// True for some subtargets at > -O0.
+ bool enablePostMachineScheduler() const override;
+
+ // enableAtomicExpandLoadLinked - True if we need to expand our atomics.
+ bool enableAtomicExpandLoadLinked() const override;
/// getInstrItins - Return the instruction itineraies based on subtarget
/// selection.
- const InstrItineraryData &getInstrItineraryData() const { return InstrItins; }
+ const InstrItineraryData *getInstrItineraryData() const {
+ return &InstrItins;
+ }
/// getStackAlignment - Returns the minimum alignment known to hold of the
/// stack frame on entry to the function and which must be maintained by every
/// GVIsIndirectSymbol - true if the GV will be accessed via an indirect
/// symbol.
bool GVIsIndirectSymbol(const GlobalValue *GV, Reloc::Model RelocM) const;
+
};
} // End llvm namespace