-//===- X86InstrInfo.td - Main X86 Instruction Definition ---*- tablegen -*-===//
+//===-- X86InstrInfo.td - Main X86 Instruction Definition --*- tablegen -*-===//
//
// The LLVM Compiler Infrastructure
//
def SDTX86CmpTest : SDTypeProfile<1, 2, [SDTCisVT<0, i32>, SDTCisSameAs<1, 2>]>;
+def SDTX86Cmpsd : SDTypeProfile<1, 3, [SDTCisVT<0, f64>, SDTCisSameAs<1, 2>, SDTCisVT<3, i8>]>;
+def SDTX86Cmpss : SDTypeProfile<1, 3, [SDTCisVT<0, f32>, SDTCisSameAs<1, 2>, SDTCisVT<3, i8>]>;
+
def SDTX86Cmov : SDTypeProfile<1, 4,
[SDTCisSameAs<0, 1>, SDTCisSameAs<1, 2>,
SDTCisVT<3, i8>, SDTCisVT<4, i32>]>;
[SDTCisSameAs<0, 2>,
SDTCisSameAs<0, 3>,
SDTCisInt<0>, SDTCisVT<1, i32>]>;
+
+// SDTBinaryArithWithFlagsInOut - RES1, EFLAGS = op LHS, RHS, EFLAGS
+def SDTBinaryArithWithFlagsInOut : SDTypeProfile<2, 3,
+ [SDTCisSameAs<0, 2>,
+ SDTCisSameAs<0, 3>,
+ SDTCisInt<0>,
+ SDTCisVT<1, i32>,
+ SDTCisVT<4, i32>]>;
+// RES1, RES2, FLAGS = op LHS, RHS
+def SDT2ResultBinaryArithWithFlags : SDTypeProfile<3, 2,
+ [SDTCisSameAs<0, 1>,
+ SDTCisSameAs<0, 2>,
+ SDTCisSameAs<0, 3>,
+ SDTCisInt<0>, SDTCisVT<1, i32>]>;
def SDTX86BrCond : SDTypeProfile<0, 3,
[SDTCisVT<0, OtherVT>,
SDTCisVT<1, i8>, SDTCisVT<2, i32>]>;
[SDTCisInt<0>,
SDTCisVT<1, i8>, SDTCisVT<2, i32>]>;
+def SDTX86sahf : SDTypeProfile<1, 1, [SDTCisVT<0, i32>, SDTCisVT<1, i8>]>;
+
def SDTX86cas : SDTypeProfile<0, 3, [SDTCisPtrTy<0>, SDTCisInt<1>,
SDTCisVT<2, i8>]>;
-def SDTX86cas8 : SDTypeProfile<0, 1, [SDTCisPtrTy<0>]>;
+def SDTX86caspair : SDTypeProfile<0, 1, [SDTCisPtrTy<0>]>;
def SDTX86atomicBinary : SDTypeProfile<2, 3, [SDTCisInt<0>, SDTCisInt<1>,
SDTCisPtrTy<2>, SDTCisInt<3>,SDTCisInt<4>]>;
def SDT_X86TLSADDR : SDTypeProfile<0, 1, [SDTCisInt<0>]>;
-def SDT_X86TLSCALL : SDTypeProfile<0, 1, [SDTCisPtrTy<0>]>;
+def SDT_X86TLSCALL : SDTypeProfile<0, 1, [SDTCisInt<0>]>;
+
+def SDT_X86SEG_ALLOCA : SDTypeProfile<1, 1, [SDTCisVT<0, iPTR>, SDTCisVT<1, iPTR>]>;
+
+def SDT_X86WIN_FTOL : SDTypeProfile<0, 1, [SDTCisFP<0>]>;
def SDT_X86EHRET : SDTypeProfile<0, 1, [SDTCisInt<0>]>;
def SDT_X86TCRET : SDTypeProfile<0, 2, [SDTCisPtrTy<0>, SDTCisVT<1, i32>]>;
def SDT_X86MEMBARRIER : SDTypeProfile<0, 0, []>;
-def SDT_X86MEMBARRIERNoSSE : SDTypeProfile<0, 1, [SDTCisInt<0>]>;
def X86MemBarrier : SDNode<"X86ISD::MEMBARRIER", SDT_X86MEMBARRIER,
[SDNPHasChain]>;
-def X86MemBarrierNoSSE : SDNode<"X86ISD::MEMBARRIER", SDT_X86MEMBARRIERNoSSE,
- [SDNPHasChain]>;
def X86MFence : SDNode<"X86ISD::MFENCE", SDT_X86MEMBARRIER,
[SDNPHasChain]>;
def X86SFence : SDNode<"X86ISD::SFENCE", SDT_X86MEMBARRIER,
def X86setcc : SDNode<"X86ISD::SETCC", SDTX86SetCC>;
def X86setcc_c : SDNode<"X86ISD::SETCC_CARRY", SDTX86SetCC_C>;
+def X86sahf : SDNode<"X86ISD::SAHF", SDTX86sahf>;
+
def X86cas : SDNode<"X86ISD::LCMPXCHG_DAG", SDTX86cas,
- [SDNPHasChain, SDNPInFlag, SDNPOutFlag, SDNPMayStore,
+ [SDNPHasChain, SDNPInGlue, SDNPOutGlue, SDNPMayStore,
+ SDNPMayLoad, SDNPMemOperand]>;
+def X86cas8 : SDNode<"X86ISD::LCMPXCHG8_DAG", SDTX86caspair,
+ [SDNPHasChain, SDNPInGlue, SDNPOutGlue, SDNPMayStore,
SDNPMayLoad, SDNPMemOperand]>;
-def X86cas8 : SDNode<"X86ISD::LCMPXCHG8_DAG", SDTX86cas8,
- [SDNPHasChain, SDNPInFlag, SDNPOutFlag, SDNPMayStore,
+def X86cas16 : SDNode<"X86ISD::LCMPXCHG16_DAG", SDTX86caspair,
+ [SDNPHasChain, SDNPInGlue, SDNPOutGlue, SDNPMayStore,
SDNPMayLoad, SDNPMemOperand]>;
+
def X86AtomAdd64 : SDNode<"X86ISD::ATOMADD64_DAG", SDTX86atomicBinary,
[SDNPHasChain, SDNPMayStore,
SDNPMayLoad, SDNPMemOperand]>;
[SDNPHasChain, SDNPMayStore,
SDNPMayLoad, SDNPMemOperand]>;
def X86retflag : SDNode<"X86ISD::RET_FLAG", SDTX86Ret,
- [SDNPHasChain, SDNPOptInFlag, SDNPVariadic]>;
+ [SDNPHasChain, SDNPOptInGlue, SDNPVariadic]>;
def X86vastart_save_xmm_regs :
SDNode<"X86ISD::VASTART_SAVE_XMM_REGS",
SDNPMemOperand]>;
def X86callseq_start :
SDNode<"ISD::CALLSEQ_START", SDT_X86CallSeqStart,
- [SDNPHasChain, SDNPOutFlag]>;
+ [SDNPHasChain, SDNPOutGlue]>;
def X86callseq_end :
SDNode<"ISD::CALLSEQ_END", SDT_X86CallSeqEnd,
- [SDNPHasChain, SDNPOptInFlag, SDNPOutFlag]>;
+ [SDNPHasChain, SDNPOptInGlue, SDNPOutGlue]>;
def X86call : SDNode<"X86ISD::CALL", SDT_X86Call,
- [SDNPHasChain, SDNPOutFlag, SDNPOptInFlag,
+ [SDNPHasChain, SDNPOutGlue, SDNPOptInGlue,
SDNPVariadic]>;
def X86rep_stos: SDNode<"X86ISD::REP_STOS", SDTX86RepStr,
- [SDNPHasChain, SDNPInFlag, SDNPOutFlag, SDNPMayStore]>;
+ [SDNPHasChain, SDNPInGlue, SDNPOutGlue, SDNPMayStore]>;
def X86rep_movs: SDNode<"X86ISD::REP_MOVS", SDTX86RepStr,
- [SDNPHasChain, SDNPInFlag, SDNPOutFlag, SDNPMayStore,
+ [SDNPHasChain, SDNPInGlue, SDNPOutGlue, SDNPMayStore,
SDNPMayLoad]>;
def X86rdtsc : SDNode<"X86ISD::RDTSC_DAG", SDTX86Void,
- [SDNPHasChain, SDNPOutFlag, SDNPSideEffect]>;
+ [SDNPHasChain, SDNPOutGlue, SDNPSideEffect]>;
def X86Wrapper : SDNode<"X86ISD::Wrapper", SDTX86Wrapper>;
def X86WrapperRIP : SDNode<"X86ISD::WrapperRIP", SDTX86Wrapper>;
def X86tlsaddr : SDNode<"X86ISD::TLSADDR", SDT_X86TLSADDR,
- [SDNPHasChain, SDNPOptInFlag, SDNPOutFlag]>;
+ [SDNPHasChain, SDNPOptInGlue, SDNPOutGlue]>;
def X86ehret : SDNode<"X86ISD::EH_RETURN", SDT_X86EHRET,
[SDNPHasChain]>;
def X86tcret : SDNode<"X86ISD::TC_RETURN", SDT_X86TCRET,
- [SDNPHasChain, SDNPOptInFlag, SDNPVariadic]>;
+ [SDNPHasChain, SDNPOptInGlue, SDNPVariadic]>;
def X86add_flag : SDNode<"X86ISD::ADD", SDTBinaryArithWithFlags,
[SDNPCommutative]>;
def X86sub_flag : SDNode<"X86ISD::SUB", SDTBinaryArithWithFlags>;
def X86smul_flag : SDNode<"X86ISD::SMUL", SDTBinaryArithWithFlags,
[SDNPCommutative]>;
-def X86umul_flag : SDNode<"X86ISD::UMUL", SDTUnaryArithWithFlags,
+def X86umul_flag : SDNode<"X86ISD::UMUL", SDT2ResultBinaryArithWithFlags,
[SDNPCommutative]>;
+def X86adc_flag : SDNode<"X86ISD::ADC", SDTBinaryArithWithFlagsInOut>;
+def X86sbb_flag : SDNode<"X86ISD::SBB", SDTBinaryArithWithFlagsInOut>;
def X86inc_flag : SDNode<"X86ISD::INC", SDTUnaryArithWithFlags>;
def X86dec_flag : SDNode<"X86ISD::DEC", SDTUnaryArithWithFlags>;
[SDNPCommutative]>;
def X86and_flag : SDNode<"X86ISD::AND", SDTBinaryArithWithFlags,
[SDNPCommutative]>;
+def X86andn_flag : SDNode<"X86ISD::ANDN", SDTBinaryArithWithFlags>;
+
+def X86blsi_flag : SDNode<"X86ISD::BLSI", SDTUnaryArithWithFlags>;
+def X86blsmsk_flag : SDNode<"X86ISD::BLSMSK", SDTUnaryArithWithFlags>;
+def X86blsr_flag : SDNode<"X86ISD::BLSR", SDTUnaryArithWithFlags>;
def X86mul_imm : SDNode<"X86ISD::MUL_IMM", SDTIntBinOp>;
def X86WinAlloca : SDNode<"X86ISD::WIN_ALLOCA", SDTX86Void,
- [SDNPHasChain, SDNPInFlag, SDNPOutFlag]>;
+ [SDNPHasChain, SDNPInGlue, SDNPOutGlue]>;
+
+def X86SegAlloca : SDNode<"X86ISD::SEG_ALLOCA", SDT_X86SEG_ALLOCA,
+ [SDNPHasChain]>;
def X86TLSCall : SDNode<"X86ISD::TLSCALL", SDT_X86TLSCALL,
- []>;
+ [SDNPHasChain, SDNPOptInGlue, SDNPOutGlue]>;
+
+def X86WinFTOL : SDNode<"X86ISD::WIN_FTOL", SDT_X86WIN_FTOL,
+ [SDNPHasChain, SDNPOutGlue]>;
//===----------------------------------------------------------------------===//
// X86 Operand Definitions.
// *mem - Operand definitions for the funky X86 addressing mode operands.
//
-def X86MemAsmOperand : AsmOperandClass {
- let Name = "Mem";
- let SuperClasses = [];
+def X86MemAsmOperand : AsmOperandClass {
+ let Name = "Mem"; let PredicateMethod = "isMem";
+}
+def X86Mem8AsmOperand : AsmOperandClass {
+ let Name = "Mem8"; let PredicateMethod = "isMem8";
+}
+def X86Mem16AsmOperand : AsmOperandClass {
+ let Name = "Mem16"; let PredicateMethod = "isMem16";
+}
+def X86Mem32AsmOperand : AsmOperandClass {
+ let Name = "Mem32"; let PredicateMethod = "isMem32";
}
+def X86Mem64AsmOperand : AsmOperandClass {
+ let Name = "Mem64"; let PredicateMethod = "isMem64";
+}
+def X86Mem80AsmOperand : AsmOperandClass {
+ let Name = "Mem80"; let PredicateMethod = "isMem80";
+}
+def X86Mem128AsmOperand : AsmOperandClass {
+ let Name = "Mem128"; let PredicateMethod = "isMem128";
+}
+def X86Mem256AsmOperand : AsmOperandClass {
+ let Name = "Mem256"; let PredicateMethod = "isMem256";
+}
+
def X86AbsMemAsmOperand : AsmOperandClass {
let Name = "AbsMem";
let SuperClasses = [X86MemAsmOperand];
let ParserMatchClass = X86MemAsmOperand;
}
+let OperandType = "OPERAND_MEMORY" in {
def opaque32mem : X86MemOperand<"printopaquemem">;
def opaque48mem : X86MemOperand<"printopaquemem">;
def opaque80mem : X86MemOperand<"printopaquemem">;
def opaque512mem : X86MemOperand<"printopaquemem">;
-def i8mem : X86MemOperand<"printi8mem">;
-def i16mem : X86MemOperand<"printi16mem">;
-def i32mem : X86MemOperand<"printi32mem">;
-def i64mem : X86MemOperand<"printi64mem">;
-def i128mem : X86MemOperand<"printi128mem">;
-def i256mem : X86MemOperand<"printi256mem">;
-def f32mem : X86MemOperand<"printf32mem">;
-def f64mem : X86MemOperand<"printf64mem">;
-def f80mem : X86MemOperand<"printf80mem">;
-def f128mem : X86MemOperand<"printf128mem">;
-def f256mem : X86MemOperand<"printf256mem">;
+def i8mem : X86MemOperand<"printi8mem"> {
+ let ParserMatchClass = X86Mem8AsmOperand; }
+def i16mem : X86MemOperand<"printi16mem"> {
+ let ParserMatchClass = X86Mem16AsmOperand; }
+def i32mem : X86MemOperand<"printi32mem"> {
+ let ParserMatchClass = X86Mem32AsmOperand; }
+def i64mem : X86MemOperand<"printi64mem"> {
+ let ParserMatchClass = X86Mem64AsmOperand; }
+def i128mem : X86MemOperand<"printi128mem"> {
+ let ParserMatchClass = X86Mem128AsmOperand; }
+def i256mem : X86MemOperand<"printi256mem"> {
+ let ParserMatchClass = X86Mem256AsmOperand; }
+def f32mem : X86MemOperand<"printf32mem"> {
+ let ParserMatchClass = X86Mem32AsmOperand; }
+def f64mem : X86MemOperand<"printf64mem"> {
+ let ParserMatchClass = X86Mem64AsmOperand; }
+def f80mem : X86MemOperand<"printf80mem"> {
+ let ParserMatchClass = X86Mem80AsmOperand; }
+def f128mem : X86MemOperand<"printf128mem"> {
+ let ParserMatchClass = X86Mem128AsmOperand; }
+def f256mem : X86MemOperand<"printf256mem">{
+ let ParserMatchClass = X86Mem256AsmOperand; }
+}
// A version of i8mem for use on x86-64 that uses GR64_NOREX instead of
// plain GR64, so that it doesn't potentially require a REX prefix.
def i8mem_NOREX : Operand<i64> {
let PrintMethod = "printi8mem";
let MIOperandInfo = (ops GR64_NOREX, i8imm, GR64_NOREX_NOSP, i32imm, i8imm);
- let ParserMatchClass = X86MemAsmOperand;
+ let ParserMatchClass = X86Mem8AsmOperand;
+ let OperandType = "OPERAND_MEMORY";
}
+// GPRs available for tailcall.
+// It represents GR64_TC or GR64_TCW64.
+def ptr_rc_tailcall : PointerLikeRegClass<2>;
+
// Special i32mem for addresses of load folding tail calls. These are not
// allowed to use callee-saved registers since they must be scheduled
// after callee-saved register are popped.
def i32mem_TC : Operand<i32> {
let PrintMethod = "printi32mem";
let MIOperandInfo = (ops GR32_TC, i8imm, GR32_TC, i32imm, i8imm);
- let ParserMatchClass = X86MemAsmOperand;
+ let ParserMatchClass = X86Mem32AsmOperand;
+ let OperandType = "OPERAND_MEMORY";
}
// Special i64mem for addresses of load folding tail calls. These are not
// after callee-saved register are popped.
def i64mem_TC : Operand<i64> {
let PrintMethod = "printi64mem";
- let MIOperandInfo = (ops GR64_TC, i8imm, GR64_TC, i32imm, i8imm);
- let ParserMatchClass = X86MemAsmOperand;
+ let MIOperandInfo = (ops ptr_rc_tailcall, i8imm,
+ ptr_rc_tailcall, i32imm, i8imm);
+ let ParserMatchClass = X86Mem64AsmOperand;
+ let OperandType = "OPERAND_MEMORY";
}
-let ParserMatchClass = X86AbsMemAsmOperand,
+let OperandType = "OPERAND_PCREL",
+ ParserMatchClass = X86AbsMemAsmOperand,
PrintMethod = "print_pcrel_imm" in {
def i32imm_pcrel : Operand<i32>;
def i16imm_pcrel : Operand<i16>;
def SSECC : Operand<i8> {
let PrintMethod = "printSSECC";
+ let OperandType = "OPERAND_IMMEDIATE";
+}
+
+def AVXCC : Operand<i8> {
+ let PrintMethod = "printSSECC";
+ let OperandType = "OPERAND_IMMEDIATE";
}
class ImmSExtAsmOperandClass : AsmOperandClass {
let RenderMethod = "addImmOperands";
}
+class ImmZExtAsmOperandClass : AsmOperandClass {
+ let SuperClasses = [ImmAsmOperand];
+ let RenderMethod = "addImmOperands";
+}
+
// Sign-extended immediate classes. We don't need to define the full lattice
// here because there is no instruction with an ambiguity between ImmSExti64i32
// and ImmSExti32i8.
let Name = "ImmSExti32i8";
}
+// [0, 0x000000FF]
+def ImmZExtu32u8AsmOperand : ImmZExtAsmOperandClass {
+ let Name = "ImmZExtu32u8";
+}
+
+
// [0, 0x0000007F] |
// [0xFFFFFFFFFFFFFF80, 0xFFFFFFFFFFFFFFFF]
def ImmSExti64i8AsmOperand : ImmSExtAsmOperandClass {
// 16-bits but only 8 bits are significant.
def i16i8imm : Operand<i16> {
let ParserMatchClass = ImmSExti16i8AsmOperand;
+ let OperandType = "OPERAND_IMMEDIATE";
}
// 32-bits but only 8 bits are significant.
def i32i8imm : Operand<i32> {
let ParserMatchClass = ImmSExti32i8AsmOperand;
+ let OperandType = "OPERAND_IMMEDIATE";
+}
+// 32-bits but only 8 bits are significant, and those 8 bits are unsigned.
+def u32u8imm : Operand<i32> {
+ let ParserMatchClass = ImmZExtu32u8AsmOperand;
+ let OperandType = "OPERAND_IMMEDIATE";
}
// 64-bits but only 32 bits are significant.
def i64i32imm : Operand<i64> {
let ParserMatchClass = ImmSExti64i32AsmOperand;
+ let OperandType = "OPERAND_IMMEDIATE";
}
// 64-bits but only 32 bits are significant, and those bits are treated as being
def i64i32imm_pcrel : Operand<i64> {
let PrintMethod = "print_pcrel_imm";
let ParserMatchClass = X86AbsMemAsmOperand;
+ let OperandType = "OPERAND_PCREL";
}
// 64-bits but only 8 bits are significant.
def i64i8imm : Operand<i64> {
let ParserMatchClass = ImmSExti64i8AsmOperand;
+ let OperandType = "OPERAND_IMMEDIATE";
}
def lea64_32mem : Operand<i32> {
def HasCMov : Predicate<"Subtarget->hasCMov()">;
def NoCMov : Predicate<"!Subtarget->hasCMov()">;
-// FIXME: temporary hack to let codegen assert or generate poor code in case
-// no AVX version of the desired intructions is present, this is better for
-// incremental dev (without fallbacks it's easier to spot what's missing)
-def HasMMX : Predicate<"Subtarget->hasMMX() && !Subtarget->hasAVX()">;
+def HasMMX : Predicate<"Subtarget->hasMMX()">;
def Has3DNow : Predicate<"Subtarget->has3DNow()">;
def Has3DNowA : Predicate<"Subtarget->has3DNowA()">;
-def HasSSE1 : Predicate<"Subtarget->hasSSE1() && !Subtarget->hasAVX()">;
-def HasSSE2 : Predicate<"Subtarget->hasSSE2() && !Subtarget->hasAVX()">;
-def HasSSE3 : Predicate<"Subtarget->hasSSE3() && !Subtarget->hasAVX()">;
-def HasSSSE3 : Predicate<"Subtarget->hasSSSE3() && !Subtarget->hasAVX()">;
-def HasSSE41 : Predicate<"Subtarget->hasSSE41() && !Subtarget->hasAVX()">;
-def HasSSE42 : Predicate<"Subtarget->hasSSE42() && !Subtarget->hasAVX()">;
-def HasSSE4A : Predicate<"Subtarget->hasSSE4A() && !Subtarget->hasAVX()">;
-
+def HasSSE1 : Predicate<"Subtarget->hasSSE1()">;
+def HasSSE2 : Predicate<"Subtarget->hasSSE2()">;
+def HasSSE3 : Predicate<"Subtarget->hasSSE3()">;
+def HasSSSE3 : Predicate<"Subtarget->hasSSSE3()">;
+def HasSSE41 : Predicate<"Subtarget->hasSSE41()">;
+def HasSSE42 : Predicate<"Subtarget->hasSSE42()">;
+def HasSSE4A : Predicate<"Subtarget->hasSSE4A()">;
def HasAVX : Predicate<"Subtarget->hasAVX()">;
+def HasAVX2 : Predicate<"Subtarget->hasAVX2()">;
+
+def HasPOPCNT : Predicate<"Subtarget->hasPOPCNT()">;
+def HasAES : Predicate<"Subtarget->hasAES()">;
def HasCLMUL : Predicate<"Subtarget->hasCLMUL()">;
def HasFMA3 : Predicate<"Subtarget->hasFMA3()">;
def HasFMA4 : Predicate<"Subtarget->hasFMA4()">;
+def HasXOP : Predicate<"Subtarget->hasXOP()">;
+def HasMOVBE : Predicate<"Subtarget->hasMOVBE()">;
+def HasRDRAND : Predicate<"Subtarget->hasRDRAND()">;
+def HasF16C : Predicate<"Subtarget->hasF16C()">;
+def HasFSGSBase : Predicate<"Subtarget->hasFSGSBase()">;
+def HasLZCNT : Predicate<"Subtarget->hasLZCNT()">;
+def HasBMI : Predicate<"Subtarget->hasBMI()">;
+def HasBMI2 : Predicate<"Subtarget->hasBMI2()">;
def FPStackf32 : Predicate<"!Subtarget->hasSSE1()">;
def FPStackf64 : Predicate<"!Subtarget->hasSSE2()">;
-def In32BitMode : Predicate<"!Subtarget->is64Bit()">, AssemblerPredicate;
-def In64BitMode : Predicate<"Subtarget->is64Bit()">, AssemblerPredicate;
+def HasCmpxchg16b: Predicate<"Subtarget->hasCmpxchg16b()">;
+def In32BitMode : Predicate<"!Subtarget->is64Bit()">,
+ AssemblerPredicate<"!Mode64Bit">;
+def In64BitMode : Predicate<"Subtarget->is64Bit()">,
+ AssemblerPredicate<"Mode64Bit">;
def IsWin64 : Predicate<"Subtarget->isTargetWin64()">;
-def NotWin64 : Predicate<"!Subtarget->isTargetWin64()">;
+def IsNaCl : Predicate<"Subtarget->isTargetNaCl()">;
+def NotNaCl : Predicate<"!Subtarget->isTargetNaCl()">;
def SmallCode : Predicate<"TM.getCodeModel() == CodeModel::Small">;
def KernelCode : Predicate<"TM.getCodeModel() == CodeModel::Kernel">;
def FarData : Predicate<"TM.getCodeModel() != CodeModel::Small &&"
def OptForSpeed : Predicate<"!OptForSize">;
def FastBTMem : Predicate<"!Subtarget->isBTMemSlow()">;
def CallImmAddr : Predicate<"Subtarget->IsLegalToCallImmediateAddr(TM)">;
-def HasAES : Predicate<"Subtarget->hasAES()">;
//===----------------------------------------------------------------------===//
// X86 Instruction Format Definitions.
include "X86InstrFormats.td"
//===----------------------------------------------------------------------===//
-// Pattern fragments...
+// Pattern fragments.
//
// X86 specific condition code. These correspond to CondCode in
def X86_COND_P : PatLeaf<(i8 14)>; // alt. COND_PE
def X86_COND_S : PatLeaf<(i8 15)>;
-def immSext8 : PatLeaf<(imm), [{ return immSext8(N); }]>;
+let FastIselShouldIgnore = 1 in { // FastIsel should ignore all simm8 instrs.
+ def i16immSExt8 : ImmLeaf<i16, [{ return Imm == (int8_t)Imm; }]>;
+ def i32immSExt8 : ImmLeaf<i32, [{ return Imm == (int8_t)Imm; }]>;
+ def i64immSExt8 : ImmLeaf<i64, [{ return Imm == (int8_t)Imm; }]>;
+}
-def i16immSExt8 : PatLeaf<(i16 immSext8)>;
-def i32immSExt8 : PatLeaf<(i32 immSext8)>;
-def i64immSExt8 : PatLeaf<(i64 immSext8)>;
-def i64immSExt32 : PatLeaf<(i64 imm), [{ return i64immSExt32(N); }]>;
-def i64immZExt32 : PatLeaf<(i64 imm), [{
- // i64immZExt32 predicate - True if the 64-bit immediate fits in a 32-bit
- // unsignedsign extended field.
- return (uint64_t)N->getZExtValue() == (uint32_t)N->getZExtValue();
-}]>;
+def i64immSExt32 : ImmLeaf<i64, [{ return Imm == (int32_t)Imm; }]>;
-def i64immZExt32SExt8 : PatLeaf<(i64 imm), [{
- uint64_t v = N->getZExtValue();
- return v == (uint32_t)v && (int32_t)v == (int8_t)v;
+
+// i64immZExt32 predicate - True if the 64-bit immediate fits in a 32-bit
+// unsigned field.
+def i64immZExt32 : ImmLeaf<i64, [{ return (uint64_t)Imm == (uint32_t)Imm; }]>;
+
+def i64immZExt32SExt8 : ImmLeaf<i64, [{
+ return (uint64_t)Imm == (uint32_t)Imm && (int32_t)Imm == (int8_t)Imm;
}]>;
// Helper fragments for loads.
}
let Defs = [RSP], Uses = [RSP], neverHasSideEffects = 1, mayStore = 1 in {
-def PUSH64i8 : Ii8<0x6a, RawFrm, (outs), (ins i8imm:$imm),
+def PUSH64i8 : Ii8<0x6a, RawFrm, (outs), (ins i64i8imm:$imm),
"push{q}\t$imm", []>;
def PUSH64i16 : Ii16<0x68, RawFrm, (outs), (ins i16imm:$imm),
"push{q}\t$imm", []>;
// These uses the DF flag in the EFLAGS register to inc or dec EDI and ESI
let Defs = [EDI,ESI], Uses = [EDI,ESI,EFLAGS] in {
-def MOVSB : I<0xA4, RawFrm, (outs), (ins), "{movsb}", []>;
-def MOVSW : I<0xA5, RawFrm, (outs), (ins), "{movsw}", []>, OpSize;
-def MOVSD : I<0xA5, RawFrm, (outs), (ins), "{movsl|movsd}", []>;
+def MOVSB : I<0xA4, RawFrm, (outs), (ins), "movsb", []>;
+def MOVSW : I<0xA5, RawFrm, (outs), (ins), "movsw", []>, OpSize;
+def MOVSD : I<0xA5, RawFrm, (outs), (ins), "movs{l|d}", []>;
def MOVSQ : RI<0xA5, RawFrm, (outs), (ins), "movsq", []>;
}
// These uses the DF flag in the EFLAGS register to inc or dec EDI and ESI
let Defs = [EDI], Uses = [AL,EDI,EFLAGS] in
-def STOSB : I<0xAA, RawFrm, (outs), (ins), "{stosb}", []>;
+def STOSB : I<0xAA, RawFrm, (outs), (ins), "stosb", []>;
let Defs = [EDI], Uses = [AX,EDI,EFLAGS] in
-def STOSW : I<0xAB, RawFrm, (outs), (ins), "{stosw}", []>, OpSize;
+def STOSW : I<0xAB, RawFrm, (outs), (ins), "stosw", []>, OpSize;
let Defs = [EDI], Uses = [EAX,EDI,EFLAGS] in
-def STOSD : I<0xAB, RawFrm, (outs), (ins), "{stosl|stosd}", []>;
+def STOSD : I<0xAB, RawFrm, (outs), (ins), "stos{l|d}", []>;
let Defs = [RCX,RDI], Uses = [RAX,RCX,RDI,EFLAGS] in
def STOSQ : RI<0xAB, RawFrm, (outs), (ins), "stosq", []>;
-def SCAS8 : I<0xAE, RawFrm, (outs), (ins), "scas{b}", []>;
-def SCAS16 : I<0xAF, RawFrm, (outs), (ins), "scas{w}", []>, OpSize;
-def SCAS32 : I<0xAF, RawFrm, (outs), (ins), "scas{l}", []>;
+def SCAS8 : I<0xAE, RawFrm, (outs), (ins), "scasb", []>;
+def SCAS16 : I<0xAF, RawFrm, (outs), (ins), "scasw", []>, OpSize;
+def SCAS32 : I<0xAF, RawFrm, (outs), (ins), "scas{l|d}", []>;
def SCAS64 : RI<0xAF, RawFrm, (outs), (ins), "scasq", []>;
-def CMPS8 : I<0xA6, RawFrm, (outs), (ins), "cmps{b}", []>;
-def CMPS16 : I<0xA7, RawFrm, (outs), (ins), "cmps{w}", []>, OpSize;
-def CMPS32 : I<0xA7, RawFrm, (outs), (ins), "cmps{l}", []>;
+def CMPS8 : I<0xA6, RawFrm, (outs), (ins), "cmpsb", []>;
+def CMPS16 : I<0xA7, RawFrm, (outs), (ins), "cmpsw", []>, OpSize;
+def CMPS32 : I<0xA7, RawFrm, (outs), (ins), "cmps{l|d}", []>;
def CMPS64 : RI<0xA7, RawFrm, (outs), (ins), "cmpsq", []>;
/// moffs8, moffs16 and moffs32 versions of moves. The immediate is a
/// 32-bit offset from the PC. These are only valid in x86-32 mode.
def MOV8o8a : Ii32 <0xA0, RawFrm, (outs), (ins offset8:$src),
- "mov{b}\t{$src, %al|%al, $src}", []>,
+ "mov{b}\t{$src, %al|AL, $src}", []>,
Requires<[In32BitMode]>;
def MOV16o16a : Ii32 <0xA1, RawFrm, (outs), (ins offset16:$src),
- "mov{w}\t{$src, %ax|%ax, $src}", []>, OpSize,
+ "mov{w}\t{$src, %ax|AL, $src}", []>, OpSize,
Requires<[In32BitMode]>;
def MOV32o32a : Ii32 <0xA1, RawFrm, (outs), (ins offset32:$src),
- "mov{l}\t{$src, %eax|%eax, $src}", []>,
+ "mov{l}\t{$src, %eax|EAX, $src}", []>,
Requires<[In32BitMode]>;
def MOV8ao8 : Ii32 <0xA2, RawFrm, (outs offset8:$dst), (ins),
- "mov{b}\t{%al, $dst|$dst, %al}", []>,
+ "mov{b}\t{%al, $dst|$dst, AL}", []>,
Requires<[In32BitMode]>;
def MOV16ao16 : Ii32 <0xA3, RawFrm, (outs offset16:$dst), (ins),
- "mov{w}\t{%ax, $dst|$dst, %ax}", []>, OpSize,
+ "mov{w}\t{%ax, $dst|$dst, AL}", []>, OpSize,
Requires<[In32BitMode]>;
def MOV32ao32 : Ii32 <0xA3, RawFrm, (outs offset32:$dst), (ins),
- "mov{l}\t{%eax, $dst|$dst, %eax}", []>,
+ "mov{l}\t{%eax, $dst|$dst, EAX}", []>,
Requires<[In32BitMode]>;
// FIXME: These definitions are utterly broken
// in question.
/*
def MOV64o8a : RIi8<0xA0, RawFrm, (outs), (ins offset8:$src),
- "mov{q}\t{$src, %rax|%rax, $src}", []>;
+ "mov{q}\t{$src, %rax|RAX, $src}", []>;
def MOV64o64a : RIi32<0xA1, RawFrm, (outs), (ins offset64:$src),
- "mov{q}\t{$src, %rax|%rax, $src}", []>;
+ "mov{q}\t{$src, %rax|RAX, $src}", []>;
def MOV64ao8 : RIi8<0xA2, RawFrm, (outs offset8:$dst), (ins),
- "mov{q}\t{%rax, $dst|$dst, %rax}", []>;
+ "mov{q}\t{%rax, $dst|$dst, RAX}", []>;
def MOV64ao64 : RIi32<0xA3, RawFrm, (outs offset64:$dst), (ins),
- "mov{q}\t{%rax, $dst|$dst, %rax}", []>;
+ "mov{q}\t{%rax, $dst|$dst, RAX}", []>;
*/
def MOV8mr_NOREX : I<0x88, MRMDestMem,
(outs), (ins i8mem_NOREX:$dst, GR8_NOREX:$src),
"mov{b}\t{$src, $dst|$dst, $src} # NOREX", []>;
-let mayLoad = 1,
+let mayLoad = 1, neverHasSideEffects = 1,
canFoldAsLoad = 1, isReMaterializable = 1 in
def MOV8rm_NOREX : I<0x8A, MRMSrcMem,
(outs GR8_NOREX:$dst), (ins i8mem_NOREX:$src),
// Condition code ops, incl. set if equal/not equal/...
-let Defs = [EFLAGS], Uses = [AH], neverHasSideEffects = 1 in
-def SAHF : I<0x9E, RawFrm, (outs), (ins), "sahf", []>; // flags = AH
+let Defs = [EFLAGS], Uses = [AH] in
+def SAHF : I<0x9E, RawFrm, (outs), (ins), "sahf",
+ [(set EFLAGS, (X86sahf AH))]>;
let Defs = [AH], Uses = [EFLAGS], neverHasSideEffects = 1 in
def LAHF : I<0x9F, RawFrm, (outs), (ins), "lahf", []>; // AH = flags
}
def XCHG16ar : I<0x90, AddRegFrm, (outs), (ins GR16:$src),
- "xchg{w}\t{$src, %ax|%ax, $src}", []>, OpSize;
+ "xchg{w}\t{$src, %ax|AX, $src}", []>, OpSize;
def XCHG32ar : I<0x90, AddRegFrm, (outs), (ins GR32:$src),
- "xchg{l}\t{$src, %eax|%eax, $src}", []>;
+ "xchg{l}\t{$src, %eax|EAX, $src}", []>, Requires<[In32BitMode]>;
+// Uses GR32_NOAX in 64-bit mode to prevent encoding using the 0x90 NOP encoding.
+// xchg %eax, %eax needs to clear upper 32-bits of RAX so is not a NOP.
+def XCHG32ar64 : I<0x90, AddRegFrm, (outs), (ins GR32_NOAX:$src),
+ "xchg{l}\t{$src, %eax|EAX, $src}", []>, Requires<[In64BitMode]>;
def XCHG64ar : RI<0x90, AddRegFrm, (outs), (ins GR64:$src),
- "xchg{q}\t{$src, %rax|%rax, $src}", []>;
+ "xchg{q}\t{$src, %rax|RAX, $src}", []>;
let Defs = [RAX, RDX, EFLAGS], Uses = [RAX, RBX, RCX, RDX] in
def CMPXCHG16B : RI<0xC7, MRM1m, (outs), (ins i128mem:$dst),
- "cmpxchg16b\t$dst", []>, TB;
+ "cmpxchg16b\t$dst", []>, TB, Requires<[HasCmpxchg16b]>;
// Lock instruction prefix
def LOCK_PREFIX : I<0xF0, RawFrm, (outs), (ins), "lock", []>;
+// Rex64 instruction prefix
+def REX64_PREFIX : I<0x48, RawFrm, (outs), (ins), "rex64", []>;
+
+// Data16 instruction prefix
+def DATA16_PREFIX : I<0x66, RawFrm, (outs), (ins), "data16", []>;
+
// Repeat string operation instruction prefixes
// These uses the DF flag in the EFLAGS register to inc or dec ECX
let Defs = [ECX], Uses = [ECX,EFLAGS] in {
def ARPL16mr : I<0x63, MRMSrcMem, (outs GR16:$src), (ins i16mem:$dst),
"arpl\t{$src, $dst|$dst, $src}", []>, Requires<[In32BitMode]>;
+//===----------------------------------------------------------------------===//
+// MOVBE Instructions
+//
+let Predicates = [HasMOVBE] in {
+ def MOVBE16rm : I<0xF0, MRMSrcMem, (outs GR16:$dst), (ins i16mem:$src),
+ "movbe{w}\t{$src, $dst|$dst, $src}",
+ [(set GR16:$dst, (bswap (loadi16 addr:$src)))]>, OpSize, T8;
+ def MOVBE32rm : I<0xF0, MRMSrcMem, (outs GR32:$dst), (ins i32mem:$src),
+ "movbe{l}\t{$src, $dst|$dst, $src}",
+ [(set GR32:$dst, (bswap (loadi32 addr:$src)))]>, T8;
+ def MOVBE64rm : RI<0xF0, MRMSrcMem, (outs GR64:$dst), (ins i64mem:$src),
+ "movbe{q}\t{$src, $dst|$dst, $src}",
+ [(set GR64:$dst, (bswap (loadi64 addr:$src)))]>, T8;
+ def MOVBE16mr : I<0xF1, MRMDestMem, (outs), (ins i16mem:$dst, GR16:$src),
+ "movbe{w}\t{$src, $dst|$dst, $src}",
+ [(store (bswap GR16:$src), addr:$dst)]>, OpSize, T8;
+ def MOVBE32mr : I<0xF1, MRMDestMem, (outs), (ins i32mem:$dst, GR32:$src),
+ "movbe{l}\t{$src, $dst|$dst, $src}",
+ [(store (bswap GR32:$src), addr:$dst)]>, T8;
+ def MOVBE64mr : RI<0xF1, MRMDestMem, (outs), (ins i64mem:$dst, GR64:$src),
+ "movbe{q}\t{$src, $dst|$dst, $src}",
+ [(store (bswap GR64:$src), addr:$dst)]>, T8;
+}
+
+//===----------------------------------------------------------------------===//
+// RDRAND Instruction
+//
+let Predicates = [HasRDRAND], Defs = [EFLAGS] in {
+ def RDRAND16r : I<0xC7, MRM6r, (outs GR16:$dst), (ins),
+ "rdrand{w}\t$dst", []>, OpSize, TB;
+ def RDRAND32r : I<0xC7, MRM6r, (outs GR32:$dst), (ins),
+ "rdrand{l}\t$dst", []>, TB;
+ def RDRAND64r : RI<0xC7, MRM6r, (outs GR64:$dst), (ins),
+ "rdrand{q}\t$dst", []>, TB;
+}
+
+//===----------------------------------------------------------------------===//
+// LZCNT Instruction
+//
+let Predicates = [HasLZCNT], Defs = [EFLAGS] in {
+ def LZCNT16rr : I<0xBD, MRMSrcReg, (outs GR16:$dst), (ins GR16:$src),
+ "lzcnt{w}\t{$src, $dst|$dst, $src}",
+ [(set GR16:$dst, (ctlz GR16:$src)), (implicit EFLAGS)]>, XS,
+ OpSize;
+ def LZCNT16rm : I<0xBD, MRMSrcMem, (outs GR16:$dst), (ins i16mem:$src),
+ "lzcnt{w}\t{$src, $dst|$dst, $src}",
+ [(set GR16:$dst, (ctlz (loadi16 addr:$src))),
+ (implicit EFLAGS)]>, XS, OpSize;
+
+ def LZCNT32rr : I<0xBD, MRMSrcReg, (outs GR32:$dst), (ins GR32:$src),
+ "lzcnt{l}\t{$src, $dst|$dst, $src}",
+ [(set GR32:$dst, (ctlz GR32:$src)), (implicit EFLAGS)]>, XS;
+ def LZCNT32rm : I<0xBD, MRMSrcMem, (outs GR32:$dst), (ins i32mem:$src),
+ "lzcnt{l}\t{$src, $dst|$dst, $src}",
+ [(set GR32:$dst, (ctlz (loadi32 addr:$src))),
+ (implicit EFLAGS)]>, XS;
+
+ def LZCNT64rr : RI<0xBD, MRMSrcReg, (outs GR64:$dst), (ins GR64:$src),
+ "lzcnt{q}\t{$src, $dst|$dst, $src}",
+ [(set GR64:$dst, (ctlz GR64:$src)), (implicit EFLAGS)]>,
+ XS;
+ def LZCNT64rm : RI<0xBD, MRMSrcMem, (outs GR64:$dst), (ins i64mem:$src),
+ "lzcnt{q}\t{$src, $dst|$dst, $src}",
+ [(set GR64:$dst, (ctlz (loadi64 addr:$src))),
+ (implicit EFLAGS)]>, XS;
+}
+
+//===----------------------------------------------------------------------===//
+// BMI Instructions
+//
+let Predicates = [HasBMI], Defs = [EFLAGS] in {
+ def TZCNT16rr : I<0xBC, MRMSrcReg, (outs GR16:$dst), (ins GR16:$src),
+ "tzcnt{w}\t{$src, $dst|$dst, $src}",
+ [(set GR16:$dst, (cttz GR16:$src)), (implicit EFLAGS)]>, XS,
+ OpSize;
+ def TZCNT16rm : I<0xBC, MRMSrcMem, (outs GR16:$dst), (ins i16mem:$src),
+ "tzcnt{w}\t{$src, $dst|$dst, $src}",
+ [(set GR16:$dst, (cttz (loadi16 addr:$src))),
+ (implicit EFLAGS)]>, XS, OpSize;
+
+ def TZCNT32rr : I<0xBC, MRMSrcReg, (outs GR32:$dst), (ins GR32:$src),
+ "tzcnt{l}\t{$src, $dst|$dst, $src}",
+ [(set GR32:$dst, (cttz GR32:$src)), (implicit EFLAGS)]>, XS;
+ def TZCNT32rm : I<0xBC, MRMSrcMem, (outs GR32:$dst), (ins i32mem:$src),
+ "tzcnt{l}\t{$src, $dst|$dst, $src}",
+ [(set GR32:$dst, (cttz (loadi32 addr:$src))),
+ (implicit EFLAGS)]>, XS;
+
+ def TZCNT64rr : RI<0xBC, MRMSrcReg, (outs GR64:$dst), (ins GR64:$src),
+ "tzcnt{q}\t{$src, $dst|$dst, $src}",
+ [(set GR64:$dst, (cttz GR64:$src)), (implicit EFLAGS)]>,
+ XS;
+ def TZCNT64rm : RI<0xBC, MRMSrcMem, (outs GR64:$dst), (ins i64mem:$src),
+ "tzcnt{q}\t{$src, $dst|$dst, $src}",
+ [(set GR64:$dst, (cttz (loadi64 addr:$src))),
+ (implicit EFLAGS)]>, XS;
+}
+
+multiclass bmi_bls<string mnemonic, Format RegMRM, Format MemMRM,
+ RegisterClass RC, X86MemOperand x86memop, SDNode OpNode,
+ PatFrag ld_frag> {
+ def rr : I<0xF3, RegMRM, (outs RC:$dst), (ins RC:$src),
+ !strconcat(mnemonic, "\t{$src, $dst|$dst, $src}"),
+ [(set RC:$dst, EFLAGS, (OpNode RC:$src))]>, T8, VEX_4V;
+ def rm : I<0xF3, MemMRM, (outs RC:$dst), (ins x86memop:$src),
+ !strconcat(mnemonic, "\t{$src, $dst|$dst, $src}"),
+ [(set RC:$dst, EFLAGS, (OpNode (ld_frag addr:$src)))]>,
+ T8, VEX_4V;
+}
+
+let Predicates = [HasBMI], Defs = [EFLAGS] in {
+ defm BLSR32 : bmi_bls<"blsr{l}", MRM1r, MRM1m, GR32, i32mem,
+ X86blsr_flag, loadi32>;
+ defm BLSR64 : bmi_bls<"blsr{q}", MRM1r, MRM1m, GR64, i64mem,
+ X86blsr_flag, loadi64>, VEX_W;
+ defm BLSMSK32 : bmi_bls<"blsmsk{l}", MRM2r, MRM2m, GR32, i32mem,
+ X86blsmsk_flag, loadi32>;
+ defm BLSMSK64 : bmi_bls<"blsmsk{q}", MRM2r, MRM2m, GR64, i64mem,
+ X86blsmsk_flag, loadi64>, VEX_W;
+ defm BLSI32 : bmi_bls<"blsi{l}", MRM3r, MRM3m, GR32, i32mem,
+ X86blsi_flag, loadi32>;
+ defm BLSI64 : bmi_bls<"blsi{q}", MRM3r, MRM3m, GR64, i64mem,
+ X86blsi_flag, loadi64>, VEX_W;
+}
+
+multiclass bmi_bextr_bzhi<bits<8> opc, string mnemonic, RegisterClass RC,
+ X86MemOperand x86memop, Intrinsic Int,
+ PatFrag ld_frag> {
+ def rr : I<opc, MRMSrcReg, (outs RC:$dst), (ins RC:$src1, RC:$src2),
+ !strconcat(mnemonic, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
+ [(set RC:$dst, (Int RC:$src1, RC:$src2)), (implicit EFLAGS)]>,
+ T8, VEX_4VOp3;
+ def rm : I<opc, MRMSrcMem, (outs RC:$dst), (ins x86memop:$src1, RC:$src2),
+ !strconcat(mnemonic, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
+ [(set RC:$dst, (Int (ld_frag addr:$src1), RC:$src2)),
+ (implicit EFLAGS)]>, T8, VEX_4VOp3;
+}
+
+let Predicates = [HasBMI], Defs = [EFLAGS] in {
+ defm BEXTR32 : bmi_bextr_bzhi<0xF7, "bextr{l}", GR32, i32mem,
+ int_x86_bmi_bextr_32, loadi32>;
+ defm BEXTR64 : bmi_bextr_bzhi<0xF7, "bextr{q}", GR64, i64mem,
+ int_x86_bmi_bextr_64, loadi64>, VEX_W;
+}
+
+let Predicates = [HasBMI2], Defs = [EFLAGS] in {
+ defm BZHI32 : bmi_bextr_bzhi<0xF5, "bzhi{l}", GR32, i32mem,
+ int_x86_bmi_bzhi_32, loadi32>;
+ defm BZHI64 : bmi_bextr_bzhi<0xF5, "bzhi{q}", GR64, i64mem,
+ int_x86_bmi_bzhi_64, loadi64>, VEX_W;
+}
+
+multiclass bmi_pdep_pext<string mnemonic, RegisterClass RC,
+ X86MemOperand x86memop, Intrinsic Int,
+ PatFrag ld_frag> {
+ def rr : I<0xF5, MRMSrcReg, (outs RC:$dst), (ins RC:$src1, RC:$src2),
+ !strconcat(mnemonic, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
+ [(set RC:$dst, (Int RC:$src1, RC:$src2))]>,
+ VEX_4V;
+ def rm : I<0xF5, MRMSrcMem, (outs RC:$dst), (ins RC:$src1, x86memop:$src2),
+ !strconcat(mnemonic, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
+ [(set RC:$dst, (Int RC:$src1, (ld_frag addr:$src2)))]>, VEX_4V;
+}
+
+let Predicates = [HasBMI2] in {
+ defm PDEP32 : bmi_pdep_pext<"pdep{l}", GR32, i32mem,
+ int_x86_bmi_pdep_32, loadi32>, T8XD;
+ defm PDEP64 : bmi_pdep_pext<"pdep{q}", GR64, i64mem,
+ int_x86_bmi_pdep_64, loadi64>, T8XD, VEX_W;
+ defm PEXT32 : bmi_pdep_pext<"pext{l}", GR32, i32mem,
+ int_x86_bmi_pext_32, loadi32>, T8XS;
+ defm PEXT64 : bmi_pdep_pext<"pext{q}", GR64, i64mem,
+ int_x86_bmi_pext_64, loadi64>, T8XS, VEX_W;
+}
+
//===----------------------------------------------------------------------===//
// Subsystems.
//===----------------------------------------------------------------------===//
// FMA - Fused Multiply-Add support (requires FMA)
include "X86InstrFMA.td"
+// XOP
+include "X86InstrXOP.td"
+
// SSE, MMX and 3DNow! vector support.
include "X86InstrSSE.td"
include "X86InstrMMX.td"
include "X86Instr3DNow.td"
include "X86InstrVMX.td"
+include "X86InstrSVM.td"
// System instructions.
include "X86InstrSystem.td"
def : MnemonicAlias<"call", "callq">, Requires<[In64BitMode]>;
def : MnemonicAlias<"cbw", "cbtw">;
+def : MnemonicAlias<"cwde", "cwtl">;
def : MnemonicAlias<"cwd", "cwtd">;
def : MnemonicAlias<"cdq", "cltd">;
-def : MnemonicAlias<"cwde", "cwtl">;
def : MnemonicAlias<"cdqe", "cltq">;
+def : MnemonicAlias<"cqo", "cqto">;
+
+// lret maps to lretl, it is not ambiguous with lretq.
+def : MnemonicAlias<"lret", "lretl">;
+
+def : MnemonicAlias<"leavel", "leave">, Requires<[In32BitMode]>;
+def : MnemonicAlias<"leaveq", "leave">, Requires<[In64BitMode]>;
+
+def : MnemonicAlias<"loopz", "loope">;
+def : MnemonicAlias<"loopnz", "loopne">;
def : MnemonicAlias<"pop", "popl">, Requires<[In32BitMode]>;
def : MnemonicAlias<"pop", "popq">, Requires<[In64BitMode]>;
// System instruction aliases.
def : MnemonicAlias<"iret", "iretl">;
def : MnemonicAlias<"sysret", "sysretl">;
+def : MnemonicAlias<"sysexit", "sysexitl">;
def : MnemonicAlias<"lgdtl", "lgdt">, Requires<[In32BitMode]>;
def : MnemonicAlias<"lgdtq", "lgdt">, Requires<[In64BitMode]>;
def : MnemonicAlias<"fcmovnae", "fcmovb">;
def : MnemonicAlias<"fcmovna", "fcmovbe">;
def : MnemonicAlias<"fcmovae", "fcmovnb">;
-def : MnemonicAlias<"fcompi", "fcomip">;
+def : MnemonicAlias<"fcomip", "fcompi">;
def : MnemonicAlias<"fildq", "fildll">;
+def : MnemonicAlias<"fistpq", "fistpll">;
+def : MnemonicAlias<"fisttpq", "fisttpll">;
def : MnemonicAlias<"fldcww", "fldcw">;
def : MnemonicAlias<"fnstcww", "fnstcw">;
def : MnemonicAlias<"fnstsww", "fnstsw">;
-def : MnemonicAlias<"fucompi", "fucomip">;
+def : MnemonicAlias<"fucomip", "fucompi">;
def : MnemonicAlias<"fwait", "wait">;
// Assembler Instruction Aliases
//===----------------------------------------------------------------------===//
+// aad/aam default to base 10 if no operand is specified.
+def : InstAlias<"aad", (AAD8i8 10)>;
+def : InstAlias<"aam", (AAM8i8 10)>;
+
+// Disambiguate the mem/imm form of bt-without-a-suffix as btl.
+def : InstAlias<"bt $imm, $mem", (BT32mi8 i32mem:$mem, i32i8imm:$imm)>;
+
// clr aliases.
def : InstAlias<"clrb $reg", (XOR8rr GR8 :$reg, GR8 :$reg)>;
def : InstAlias<"clrw $reg", (XOR16rr GR16:$reg, GR16:$reg)>;
def : InstAlias<"clrl $reg", (XOR32rr GR32:$reg, GR32:$reg)>;
def : InstAlias<"clrq $reg", (XOR64rr GR64:$reg, GR64:$reg)>;
+// div and idiv aliases for explicit A register.
+def : InstAlias<"divb $src, %al", (DIV8r GR8 :$src)>;
+def : InstAlias<"divw $src, %ax", (DIV16r GR16:$src)>;
+def : InstAlias<"divl $src, %eax", (DIV32r GR32:$src)>;
+def : InstAlias<"divq $src, %rax", (DIV64r GR64:$src)>;
+def : InstAlias<"divb $src, %al", (DIV8m i8mem :$src)>;
+def : InstAlias<"divw $src, %ax", (DIV16m i16mem:$src)>;
+def : InstAlias<"divl $src, %eax", (DIV32m i32mem:$src)>;
+def : InstAlias<"divq $src, %rax", (DIV64m i64mem:$src)>;
+def : InstAlias<"idivb $src, %al", (IDIV8r GR8 :$src)>;
+def : InstAlias<"idivw $src, %ax", (IDIV16r GR16:$src)>;
+def : InstAlias<"idivl $src, %eax", (IDIV32r GR32:$src)>;
+def : InstAlias<"idivq $src, %rax", (IDIV64r GR64:$src)>;
+def : InstAlias<"idivb $src, %al", (IDIV8m i8mem :$src)>;
+def : InstAlias<"idivw $src, %ax", (IDIV16m i16mem:$src)>;
+def : InstAlias<"idivl $src, %eax", (IDIV32m i32mem:$src)>;
+def : InstAlias<"idivq $src, %rax", (IDIV64m i64mem:$src)>;
+
+
+
+// Various unary fpstack operations default to operating on on ST1.
+// For example, "fxch" -> "fxch %st(1)"
+def : InstAlias<"faddp", (ADD_FPrST0 ST1), 0>;
+def : InstAlias<"fsubp", (SUBR_FPrST0 ST1)>;
+def : InstAlias<"fsubrp", (SUB_FPrST0 ST1)>;
+def : InstAlias<"fmulp", (MUL_FPrST0 ST1)>;
+def : InstAlias<"fdivp", (DIVR_FPrST0 ST1)>;
+def : InstAlias<"fdivrp", (DIV_FPrST0 ST1)>;
+def : InstAlias<"fxch", (XCH_F ST1)>;
+def : InstAlias<"fcomi", (COM_FIr ST1)>;
+def : InstAlias<"fcompi", (COM_FIPr ST1)>;
+def : InstAlias<"fucom", (UCOM_Fr ST1)>;
+def : InstAlias<"fucomp", (UCOM_FPr ST1)>;
+def : InstAlias<"fucomi", (UCOM_FIr ST1)>;
+def : InstAlias<"fucompi", (UCOM_FIPr ST1)>;
+
+// Handle fmul/fadd/fsub/fdiv instructions with explicitly written st(0) op.
+// For example, "fadd %st(4), %st(0)" -> "fadd %st(4)". We also disambiguate
+// instructions like "fadd %st(0), %st(0)" as "fadd %st(0)" for consistency with
+// gas.
+multiclass FpUnaryAlias<string Mnemonic, Instruction Inst, bit EmitAlias = 1> {
+ def : InstAlias<!strconcat(Mnemonic, " $op, %st(0)"),
+ (Inst RST:$op), EmitAlias>;
+ def : InstAlias<!strconcat(Mnemonic, " %st(0), %st(0)"),
+ (Inst ST0), EmitAlias>;
+}
+
+defm : FpUnaryAlias<"fadd", ADD_FST0r>;
+defm : FpUnaryAlias<"faddp", ADD_FPrST0, 0>;
+defm : FpUnaryAlias<"fsub", SUB_FST0r>;
+defm : FpUnaryAlias<"fsubp", SUBR_FPrST0>;
+defm : FpUnaryAlias<"fsubr", SUBR_FST0r>;
+defm : FpUnaryAlias<"fsubrp", SUB_FPrST0>;
+defm : FpUnaryAlias<"fmul", MUL_FST0r>;
+defm : FpUnaryAlias<"fmulp", MUL_FPrST0>;
+defm : FpUnaryAlias<"fdiv", DIV_FST0r>;
+defm : FpUnaryAlias<"fdivp", DIVR_FPrST0>;
+defm : FpUnaryAlias<"fdivr", DIVR_FST0r>;
+defm : FpUnaryAlias<"fdivrp", DIV_FPrST0>;
+defm : FpUnaryAlias<"fcomi", COM_FIr, 0>;
+defm : FpUnaryAlias<"fucomi", UCOM_FIr, 0>;
+defm : FpUnaryAlias<"fcompi", COM_FIPr>;
+defm : FpUnaryAlias<"fucompi", UCOM_FIPr>;
+
+
+// Handle "f{mulp,addp} st(0), $op" the same as "f{mulp,addp} $op", since they
+// commute. We also allow fdiv[r]p/fsubrp even though they don't commute,
+// solely because gas supports it.
+def : InstAlias<"faddp %st(0), $op", (ADD_FPrST0 RST:$op), 0>;
+def : InstAlias<"fmulp %st(0), $op", (MUL_FPrST0 RST:$op)>;
+def : InstAlias<"fsubp %st(0), $op", (SUBR_FPrST0 RST:$op)>;
+def : InstAlias<"fsubrp %st(0), $op", (SUB_FPrST0 RST:$op)>;
+def : InstAlias<"fdivp %st(0), $op", (DIVR_FPrST0 RST:$op)>;
+def : InstAlias<"fdivrp %st(0), $op", (DIV_FPrST0 RST:$op)>;
+
+// We accept "fnstsw %eax" even though it only writes %ax.
+def : InstAlias<"fnstsw %eax", (FNSTSW16r)>;
+def : InstAlias<"fnstsw %al" , (FNSTSW16r)>;
+def : InstAlias<"fnstsw" , (FNSTSW16r)>;
+
// lcall and ljmp aliases. This seems to be an odd mapping in 64-bit mode, but
// this is compatible with what GAS does.
def : InstAlias<"lcall $seg, $off", (FARCALL32i i32imm:$off, i16imm:$seg)>;
def : InstAlias<"imulq $imm, $r",(IMUL64rri32 GR64:$r, GR64:$r,i64i32imm:$imm)>;
def : InstAlias<"imulq $imm, $r", (IMUL64rri8 GR64:$r, GR64:$r, i64i8imm:$imm)>;
+// inb %dx -> inb %al, %dx
+def : InstAlias<"inb %dx", (IN8rr)>;
+def : InstAlias<"inw %dx", (IN16rr)>;
+def : InstAlias<"inl %dx", (IN32rr)>;
+def : InstAlias<"inb $port", (IN8ri i8imm:$port)>;
+def : InstAlias<"inw $port", (IN16ri i8imm:$port)>;
+def : InstAlias<"inl $port", (IN32ri i8imm:$port)>;
+
// jmp and call aliases for lcall and ljmp. jmp $42,$5 -> ljmp
def : InstAlias<"call $seg, $off", (FARCALL32i i32imm:$off, i16imm:$seg)>;
def : InstAlias<"calll $seg, $off", (FARCALL32i i32imm:$off, i16imm:$seg)>;
def : InstAlias<"jmpl $seg, $off", (FARJMP32i i32imm:$off, i16imm:$seg)>;
+// Force mov without a suffix with a segment and mem to prefer the 'l' form of
+// the move. All segment/mem forms are equivalent, this has the shortest
+// encoding.
+def : InstAlias<"mov $mem, $seg", (MOV32sm SEGMENT_REG:$seg, i32mem:$mem)>;
+def : InstAlias<"mov $seg, $mem", (MOV32ms i32mem:$mem, SEGMENT_REG:$seg)>;
// Match 'movq <largeimm>, <reg>' as an alias for movabsq.
def : InstAlias<"movq $imm, $reg", (MOV64ri GR64:$reg, i64imm:$imm)>;
+// Match 'movq GR64, MMX' as an alias for movd.
+def : InstAlias<"movq $src, $dst",
+ (MMX_MOVD64to64rr VR64:$dst, GR64:$src), 0>;
+def : InstAlias<"movq $src, $dst",
+ (MMX_MOVD64from64rr GR64:$dst, VR64:$src), 0>;
+
// movsd with no operands (as opposed to the SSE scalar move of a double) is an
// alias for movsl. (as in rep; movsd)
def : InstAlias<"movsd", (MOVSD)>;
// movsx aliases
-def : InstAlias<"movsx $src, $dst", (MOVSX16rr8W GR16:$dst, GR8:$src)>;
-def : InstAlias<"movsx $src, $dst", (MOVSX16rm8W GR16:$dst, i8mem:$src)>;
-def : InstAlias<"movsx $src, $dst", (MOVSX32rr8 GR32:$dst, GR8:$src)>;
-def : InstAlias<"movsx $src, $dst", (MOVSX32rr16 GR32:$dst, GR16:$src)>;
-def : InstAlias<"movsx $src, $dst", (MOVSX64rr8 GR64:$dst, GR8:$src)>;
-def : InstAlias<"movsx $src, $dst", (MOVSX64rr16 GR64:$dst, GR16:$src)>;
-def : InstAlias<"movsx $src, $dst", (MOVSX64rr32 GR64:$dst, GR32:$src)>;
+def : InstAlias<"movsx $src, $dst", (MOVSX16rr8 GR16:$dst, GR8:$src), 0>;
+def : InstAlias<"movsx $src, $dst", (MOVSX16rm8 GR16:$dst, i8mem:$src), 0>;
+def : InstAlias<"movsx $src, $dst", (MOVSX32rr8 GR32:$dst, GR8:$src), 0>;
+def : InstAlias<"movsx $src, $dst", (MOVSX32rr16 GR32:$dst, GR16:$src), 0>;
+def : InstAlias<"movsx $src, $dst", (MOVSX64rr8 GR64:$dst, GR8:$src), 0>;
+def : InstAlias<"movsx $src, $dst", (MOVSX64rr16 GR64:$dst, GR16:$src), 0>;
+def : InstAlias<"movsx $src, $dst", (MOVSX64rr32 GR64:$dst, GR32:$src), 0>;
// movzx aliases
-def : InstAlias<"movzx $src, $dst", (MOVZX16rr8W GR16:$dst, GR8:$src)>;
-def : InstAlias<"movzx $src, $dst", (MOVZX16rm8W GR16:$dst, i8mem:$src)>;
-def : InstAlias<"movzx $src, $dst", (MOVZX32rr8 GR32:$dst, GR8:$src)>;
-def : InstAlias<"movzx $src, $dst", (MOVZX32rr16 GR32:$dst, GR16:$src)>;
-def : InstAlias<"movzx $src, $dst", (MOVZX64rr8_Q GR64:$dst, GR8:$src)>;
-def : InstAlias<"movzx $src, $dst", (MOVZX64rr16_Q GR64:$dst, GR16:$src)>;
+def : InstAlias<"movzx $src, $dst", (MOVZX16rr8 GR16:$dst, GR8:$src), 0>;
+def : InstAlias<"movzx $src, $dst", (MOVZX16rm8 GR16:$dst, i8mem:$src), 0>;
+def : InstAlias<"movzx $src, $dst", (MOVZX32rr8 GR32:$dst, GR8:$src), 0>;
+def : InstAlias<"movzx $src, $dst", (MOVZX32rr16 GR32:$dst, GR16:$src), 0>;
+def : InstAlias<"movzx $src, $dst", (MOVZX64rr8_Q GR64:$dst, GR8:$src), 0>;
+def : InstAlias<"movzx $src, $dst", (MOVZX64rr16_Q GR64:$dst, GR16:$src), 0>;
// Note: No GR32->GR64 movzx form.
+// outb %dx -> outb %al, %dx
+def : InstAlias<"outb %dx", (OUT8rr)>;
+def : InstAlias<"outw %dx", (OUT16rr)>;
+def : InstAlias<"outl %dx", (OUT32rr)>;
+def : InstAlias<"outb $port", (OUT8ir i8imm:$port)>;
+def : InstAlias<"outw $port", (OUT16ir i8imm:$port)>;
+def : InstAlias<"outl $port", (OUT32ir i8imm:$port)>;
+
// 'sldt <mem>' can be encoded with either sldtw or sldtq with the same
// effect (both store to a 16-bit mem). Force to sldtw to avoid ambiguity
// errors, since its encoding is the most compact.
def : InstAlias<"sldt $mem", (SLDT16m i16mem:$mem)>;
+// shld/shrd op,op -> shld op, op, CL
+def : InstAlias<"shldw $r2, $r1", (SHLD16rrCL GR16:$r1, GR16:$r2)>;
+def : InstAlias<"shldl $r2, $r1", (SHLD32rrCL GR32:$r1, GR32:$r2)>;
+def : InstAlias<"shldq $r2, $r1", (SHLD64rrCL GR64:$r1, GR64:$r2)>;
+def : InstAlias<"shrdw $r2, $r1", (SHRD16rrCL GR16:$r1, GR16:$r2)>;
+def : InstAlias<"shrdl $r2, $r1", (SHRD32rrCL GR32:$r1, GR32:$r2)>;
+def : InstAlias<"shrdq $r2, $r1", (SHRD64rrCL GR64:$r1, GR64:$r2)>;
+
+def : InstAlias<"shldw $reg, $mem", (SHLD16mrCL i16mem:$mem, GR16:$reg)>;
+def : InstAlias<"shldl $reg, $mem", (SHLD32mrCL i32mem:$mem, GR32:$reg)>;
+def : InstAlias<"shldq $reg, $mem", (SHLD64mrCL i64mem:$mem, GR64:$reg)>;
+def : InstAlias<"shrdw $reg, $mem", (SHRD16mrCL i16mem:$mem, GR16:$reg)>;
+def : InstAlias<"shrdl $reg, $mem", (SHRD32mrCL i32mem:$mem, GR32:$reg)>;
+def : InstAlias<"shrdq $reg, $mem", (SHRD64mrCL i64mem:$mem, GR64:$reg)>;
+
+/* FIXME: This is disabled because the asm matcher is currently incapable of
+ * matching a fixed immediate like $1.
+// "shl X, $1" is an alias for "shl X".
+multiclass ShiftRotateByOneAlias<string Mnemonic, string Opc> {
+ def : InstAlias<!strconcat(Mnemonic, "b $op, $$1"),
+ (!cast<Instruction>(!strconcat(Opc, "8r1")) GR8:$op)>;
+ def : InstAlias<!strconcat(Mnemonic, "w $op, $$1"),
+ (!cast<Instruction>(!strconcat(Opc, "16r1")) GR16:$op)>;
+ def : InstAlias<!strconcat(Mnemonic, "l $op, $$1"),
+ (!cast<Instruction>(!strconcat(Opc, "32r1")) GR32:$op)>;
+ def : InstAlias<!strconcat(Mnemonic, "q $op, $$1"),
+ (!cast<Instruction>(!strconcat(Opc, "64r1")) GR64:$op)>;
+ def : InstAlias<!strconcat(Mnemonic, "b $op, $$1"),
+ (!cast<Instruction>(!strconcat(Opc, "8m1")) i8mem:$op)>;
+ def : InstAlias<!strconcat(Mnemonic, "w $op, $$1"),
+ (!cast<Instruction>(!strconcat(Opc, "16m1")) i16mem:$op)>;
+ def : InstAlias<!strconcat(Mnemonic, "l $op, $$1"),
+ (!cast<Instruction>(!strconcat(Opc, "32m1")) i32mem:$op)>;
+ def : InstAlias<!strconcat(Mnemonic, "q $op, $$1"),
+ (!cast<Instruction>(!strconcat(Opc, "64m1")) i64mem:$op)>;
+}
+
+defm : ShiftRotateByOneAlias<"rcl", "RCL">;
+defm : ShiftRotateByOneAlias<"rcr", "RCR">;
+defm : ShiftRotateByOneAlias<"rol", "ROL">;
+defm : ShiftRotateByOneAlias<"ror", "ROR">;
+FIXME */
// test: We accept "testX <reg>, <mem>" and "testX <mem>, <reg>" as synonyms.
def : InstAlias<"testb $val, $mem", (TEST8rm GR8 :$val, i8mem :$mem)>;
def : InstAlias<"xchgl $mem, $val", (XCHG32rm GR32:$val, i32mem:$mem)>;
def : InstAlias<"xchgq $mem, $val", (XCHG64rm GR64:$val, i64mem:$mem)>;
+// xchg: We accept "xchgX <reg>, %eax" and "xchgX %eax, <reg>" as synonyms.
+def : InstAlias<"xchgw %ax, $src", (XCHG16ar GR16:$src)>;
+def : InstAlias<"xchgl %eax, $src", (XCHG32ar GR32:$src)>, Requires<[In32BitMode]>;
+def : InstAlias<"xchgl %eax, $src", (XCHG32ar64 GR32_NOAX:$src)>, Requires<[In64BitMode]>;
+def : InstAlias<"xchgq %rax, $src", (XCHG64ar GR64:$src)>;