// X86 Operand Definitions.
//
-def i32imm_pcrel : Operand<i32> {
- let PrintMethod = "print_pcrel_imm";
-}
-
// A version of ptr_rc which excludes SP, ESP, and RSP. This is used for
// the index operand of an address, to conform to x86 encoding restrictions.
def ptr_rc_nosp : PointerLikeRegClass<1>;
let Name = "Mem";
let SuperClass = ?;
}
+def X86AbsMemAsmOperand : AsmOperandClass {
+ let Name = "AbsMem";
+ let SuperClass = X86MemAsmOperand;
+}
+def X86NoSegMemAsmOperand : AsmOperandClass {
+ let Name = "NoSegMem";
+ let SuperClass = X86MemAsmOperand;
+}
class X86MemOperand<string printMethod> : Operand<iPTR> {
let PrintMethod = printMethod;
let MIOperandInfo = (ops ptr_rc, i8imm, ptr_rc_nosp, i32imm, i8imm);
def opaque80mem : X86MemOperand<"printopaquemem">;
def opaque512mem : X86MemOperand<"printopaquemem">;
-def offset8 : Operand<i64> { let PrintMethod = "print_pcrel_imm"; }
-def offset16 : Operand<i64> { let PrintMethod = "print_pcrel_imm"; }
-def offset32 : Operand<i64> { let PrintMethod = "print_pcrel_imm"; }
-def offset64 : Operand<i64> { let PrintMethod = "print_pcrel_imm"; }
-
def i8mem : X86MemOperand<"printi8mem">;
def i16mem : X86MemOperand<"printi16mem">;
def i32mem : X86MemOperand<"printi32mem">;
def lea32mem : Operand<i32> {
let PrintMethod = "printlea32mem";
let MIOperandInfo = (ops GR32, i8imm, GR32_NOSP, i32imm);
- let ParserMatchClass = X86MemAsmOperand;
+ let ParserMatchClass = X86NoSegMemAsmOperand;
+}
+
+let ParserMatchClass = X86AbsMemAsmOperand,
+ PrintMethod = "print_pcrel_imm" in {
+def i32imm_pcrel : Operand<i32>;
+
+def offset8 : Operand<i64>;
+def offset16 : Operand<i64>;
+def offset32 : Operand<i64>;
+def offset64 : Operand<i64>;
+
+// Branch targets have OtherVT type and print as pc-relative values.
+def brtarget : Operand<OtherVT>;
+def brtarget8 : Operand<OtherVT>;
+
}
def SSECC : Operand<i8> {
let ParserMatchClass = ImmSExt8AsmOperand;
}
-// Branch targets have OtherVT type and print as pc-relative values.
-def brtarget : Operand<OtherVT> {
- let PrintMethod = "print_pcrel_imm";
-}
-
-def brtarget8 : Operand<OtherVT> {
- let PrintMethod = "print_pcrel_imm";
-}
-
//===----------------------------------------------------------------------===//
// X86 Complex Pattern Definitions.
//
def X86_COND_P : PatLeaf<(i8 14)>; // alt. COND_PE
def X86_COND_S : PatLeaf<(i8 15)>;
-def i16immSExt8 : PatLeaf<(i16 imm), [{
- // i16immSExt8 predicate - True if the 16-bit immediate fits in a 8-bit
- // sign extended field.
- return (int16_t)N->getZExtValue() == (int8_t)N->getZExtValue();
+def immSext8 : PatLeaf<(imm), [{
+ return N->getSExtValue() == (int8_t)N->getSExtValue();
}]>;
-def i32immSExt8 : PatLeaf<(i32 imm), [{
- // i32immSExt8 predicate - True if the 32-bit immediate fits in a 8-bit
- // sign extended field.
- return (int32_t)N->getZExtValue() == (int8_t)N->getZExtValue();
+def i16immSExt8 : PatLeaf<(i16 immSext8)>;
+def i32immSExt8 : PatLeaf<(i32 immSext8)>;
+
+/// Load patterns: these constraint the match to the right address space.
+def dsload : PatFrag<(ops node:$ptr), (load node:$ptr), [{
+ if (const Value *Src = cast<LoadSDNode>(N)->getSrcValue())
+ if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
+ if (PT->getAddressSpace() > 255)
+ return false;
+ return true;
}]>;
+def gsload : PatFrag<(ops node:$ptr), (load node:$ptr), [{
+ if (const Value *Src = cast<LoadSDNode>(N)->getSrcValue())
+ if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
+ return PT->getAddressSpace() == 256;
+ return false;
+}]>;
+
+def fsload : PatFrag<(ops node:$ptr), (load node:$ptr), [{
+ if (const Value *Src = cast<LoadSDNode>(N)->getSrcValue())
+ if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
+ return PT->getAddressSpace() == 257;
+ return false;
+}]>;
+
+
// Helper fragments for loads.
// It's always safe to treat a anyext i16 load as a i32 load if the i16 is
// known to be 32-bit aligned or better. Ditto for i8 to i16.
return false;
}]>;
-def loadi16_anyext : PatFrag<(ops node:$ptr), (i32 (unindexedload node:$ptr)),
-[{
+def loadi16_anyext : PatFrag<(ops node:$ptr), (i32 (unindexedload node:$ptr)),[{
LoadSDNode *LD = cast<LoadSDNode>(N);
if (const Value *Src = LD->getSrcValue())
if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
return false;
}]>;
-def nvloadi32 : PatFrag<(ops node:$ptr), (i32 (unindexedload node:$ptr)), [{
- LoadSDNode *LD = cast<LoadSDNode>(N);
- if (const Value *Src = LD->getSrcValue())
- if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
- if (PT->getAddressSpace() > 255)
- return false;
- if (LD->isVolatile())
- return false;
- ISD::LoadExtType ExtType = LD->getExtensionType();
- if (ExtType == ISD::NON_EXTLOAD)
- return true;
- if (ExtType == ISD::EXTLOAD)
- return LD->getAlignment() >= 4;
- return false;
-}]>;
-
-def gsload : PatFrag<(ops node:$ptr), (load node:$ptr), [{
- if (const Value *Src = cast<LoadSDNode>(N)->getSrcValue())
- if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
- return PT->getAddressSpace() == 256;
- return false;
-}]>;
-
-def fsload : PatFrag<(ops node:$ptr), (load node:$ptr), [{
- if (const Value *Src = cast<LoadSDNode>(N)->getSrcValue())
- if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
- return PT->getAddressSpace() == 257;
- return false;
-}]>;
-
-def loadi8 : PatFrag<(ops node:$ptr), (i8 (load node:$ptr)), [{
- if (const Value *Src = cast<LoadSDNode>(N)->getSrcValue())
- if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
- if (PT->getAddressSpace() > 255)
- return false;
- return true;
-}]>;
-def loadi64 : PatFrag<(ops node:$ptr), (i64 (load node:$ptr)), [{
- if (const Value *Src = cast<LoadSDNode>(N)->getSrcValue())
- if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
- if (PT->getAddressSpace() > 255)
- return false;
- return true;
-}]>;
-
-def loadf32 : PatFrag<(ops node:$ptr), (f32 (load node:$ptr)), [{
- if (const Value *Src = cast<LoadSDNode>(N)->getSrcValue())
- if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
- if (PT->getAddressSpace() > 255)
- return false;
- return true;
-}]>;
-def loadf64 : PatFrag<(ops node:$ptr), (f64 (load node:$ptr)), [{
- if (const Value *Src = cast<LoadSDNode>(N)->getSrcValue())
- if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
- if (PT->getAddressSpace() > 255)
- return false;
- return true;
-}]>;
-def loadf80 : PatFrag<(ops node:$ptr), (f80 (load node:$ptr)), [{
- if (const Value *Src = cast<LoadSDNode>(N)->getSrcValue())
- if (const PointerType *PT = dyn_cast<PointerType>(Src->getType()))
- if (PT->getAddressSpace() > 255)
- return false;
- return true;
-}]>;
+def loadi8 : PatFrag<(ops node:$ptr), (i8 (dsload node:$ptr))>;
+def loadi64 : PatFrag<(ops node:$ptr), (i64 (dsload node:$ptr))>;
+def loadf32 : PatFrag<(ops node:$ptr), (f32 (dsload node:$ptr))>;
+def loadf64 : PatFrag<(ops node:$ptr), (f64 (dsload node:$ptr))>;
+def loadf80 : PatFrag<(ops node:$ptr), (f80 (dsload node:$ptr))>;
def sextloadi16i8 : PatFrag<(ops node:$ptr), (i16 (sextloadi8 node:$ptr))>;
def sextloadi32i8 : PatFrag<(ops node:$ptr), (i32 (sextloadi8 node:$ptr))>;
def or_is_add : PatFrag<(ops node:$lhs, node:$rhs), (or node:$lhs, node:$rhs),[{
if (ConstantSDNode *CN = dyn_cast<ConstantSDNode>(N->getOperand(1)))
return CurDAG->MaskedValueIsZero(N->getOperand(0), CN->getAPIntValue());
- return false;
-}]>;
-def or_not_add : PatFrag<(ops node:$lhs, node:$rhs),(or node:$lhs, node:$rhs),[{
- ConstantSDNode *CN = dyn_cast<ConstantSDNode>(N->getOperand(1));
- if (!CN) return true;
- return !CurDAG->MaskedValueIsZero(N->getOperand(0), CN->getAPIntValue());
+ else {
+ unsigned BitWidth = N->getValueType(0).getScalarType().getSizeInBits();
+ APInt Mask = APInt::getAllOnesValue(BitWidth);
+ APInt KnownZero0, KnownOne0;
+ CurDAG->ComputeMaskedBits(N->getOperand(0), Mask, KnownZero0, KnownOne0, 0);
+ APInt KnownZero1, KnownOne1;
+ CurDAG->ComputeMaskedBits(N->getOperand(1), Mask, KnownZero1, KnownOne1, 0);
+ return (~KnownZero0 & ~KnownZero1) == 0;
+ }
}]>;
// 'shld' and 'shrd' instruction patterns. Note that even though these have
"", []>;
//===----------------------------------------------------------------------===//
-// Control Flow Instructions...
+// Control Flow Instructions.
//
// Return instructions.
"lret\t$amt", []>;
}
-// All branches are RawFrm, Void, Branch, and Terminators
-let isBranch = 1, isTerminator = 1 in
- class IBr<bits<8> opcode, dag ins, string asm, list<dag> pattern> :
- I<opcode, RawFrm, (outs), ins, asm, pattern>;
+// Unconditional branches.
+let isBarrier = 1, isBranch = 1, isTerminator = 1 in {
+ def JMP_4 : Ii32PCRel<0xE9, RawFrm, (outs), (ins brtarget:$dst),
+ "jmp\t$dst", [(br bb:$dst)]>;
+ def JMP_1 : Ii8PCRel<0xEB, RawFrm, (outs), (ins brtarget8:$dst),
+ "jmp\t$dst", []>;
+}
-let isBranch = 1, isBarrier = 1 in {
- def JMP : IBr<0xE9, (ins brtarget:$dst), "jmp\t$dst", [(br bb:$dst)]>;
- def JMP8 : IBr<0xEB, (ins brtarget8:$dst), "jmp\t$dst", []>;
+// Conditional Branches.
+let isBranch = 1, isTerminator = 1, Uses = [EFLAGS] in {
+ multiclass ICBr<bits<8> opc1, bits<8> opc4, string asm, PatFrag Cond> {
+ def _1 : Ii8PCRel <opc1, RawFrm, (outs), (ins brtarget8:$dst), asm, []>;
+ def _4 : Ii32PCRel<opc4, RawFrm, (outs), (ins brtarget:$dst), asm,
+ [(X86brcond bb:$dst, Cond, EFLAGS)]>, TB;
+ }
}
+defm JO : ICBr<0x70, 0x80, "jo\t$dst" , X86_COND_O>;
+defm JNO : ICBr<0x71, 0x81, "jno\t$dst" , X86_COND_NO>;
+defm JB : ICBr<0x72, 0x82, "jb\t$dst" , X86_COND_B>;
+defm JAE : ICBr<0x73, 0x83, "jae\t$dst", X86_COND_AE>;
+defm JE : ICBr<0x74, 0x84, "je\t$dst" , X86_COND_E>;
+defm JNE : ICBr<0x75, 0x85, "jne\t$dst", X86_COND_NE>;
+defm JBE : ICBr<0x76, 0x86, "jbe\t$dst", X86_COND_BE>;
+defm JA : ICBr<0x77, 0x87, "ja\t$dst" , X86_COND_A>;
+defm JS : ICBr<0x78, 0x88, "js\t$dst" , X86_COND_S>;
+defm JNS : ICBr<0x79, 0x89, "jns\t$dst", X86_COND_NS>;
+defm JP : ICBr<0x7A, 0x8A, "jp\t$dst" , X86_COND_P>;
+defm JNP : ICBr<0x7B, 0x8B, "jnp\t$dst", X86_COND_NP>;
+defm JL : ICBr<0x7C, 0x8C, "jl\t$dst" , X86_COND_L>;
+defm JGE : ICBr<0x7D, 0x8D, "jge\t$dst", X86_COND_GE>;
+defm JLE : ICBr<0x7E, 0x8E, "jle\t$dst", X86_COND_LE>;
+defm JG : ICBr<0x7F, 0x8F, "jg\t$dst" , X86_COND_G>;
+
+// FIXME: What about the CX/RCX versions of this instruction?
+let Uses = [ECX], isBranch = 1, isTerminator = 1 in
+ def JCXZ8 : Ii8PCRel<0xE3, RawFrm, (outs), (ins brtarget8:$dst),
+ "jcxz\t$dst", []>;
+
+
// Indirect branches
let isBranch = 1, isTerminator = 1, isBarrier = 1, isIndirectBranch = 1 in {
def JMP32r : I<0xFF, MRM4r, (outs), (ins GR32:$dst), "jmp{l}\t{*}$dst",
"ljmp{l}\t{*}$dst", []>;
}
-// Conditional branches
-let Uses = [EFLAGS] in {
-// Short conditional jumps
-def JO8 : IBr<0x70, (ins brtarget8:$dst), "jo\t$dst", []>;
-def JNO8 : IBr<0x71, (ins brtarget8:$dst), "jno\t$dst", []>;
-def JB8 : IBr<0x72, (ins brtarget8:$dst), "jb\t$dst", []>;
-def JAE8 : IBr<0x73, (ins brtarget8:$dst), "jae\t$dst", []>;
-def JE8 : IBr<0x74, (ins brtarget8:$dst), "je\t$dst", []>;
-def JNE8 : IBr<0x75, (ins brtarget8:$dst), "jne\t$dst", []>;
-def JBE8 : IBr<0x76, (ins brtarget8:$dst), "jbe\t$dst", []>;
-def JA8 : IBr<0x77, (ins brtarget8:$dst), "ja\t$dst", []>;
-def JS8 : IBr<0x78, (ins brtarget8:$dst), "js\t$dst", []>;
-def JNS8 : IBr<0x79, (ins brtarget8:$dst), "jns\t$dst", []>;
-def JP8 : IBr<0x7A, (ins brtarget8:$dst), "jp\t$dst", []>;
-def JNP8 : IBr<0x7B, (ins brtarget8:$dst), "jnp\t$dst", []>;
-def JL8 : IBr<0x7C, (ins brtarget8:$dst), "jl\t$dst", []>;
-def JGE8 : IBr<0x7D, (ins brtarget8:$dst), "jge\t$dst", []>;
-def JLE8 : IBr<0x7E, (ins brtarget8:$dst), "jle\t$dst", []>;
-def JG8 : IBr<0x7F, (ins brtarget8:$dst), "jg\t$dst", []>;
-
-def JCXZ8 : IBr<0xE3, (ins brtarget8:$dst), "jcxz\t$dst", []>;
-
-def JE : IBr<0x84, (ins brtarget:$dst), "je\t$dst",
- [(X86brcond bb:$dst, X86_COND_E, EFLAGS)]>, TB;
-def JNE : IBr<0x85, (ins brtarget:$dst), "jne\t$dst",
- [(X86brcond bb:$dst, X86_COND_NE, EFLAGS)]>, TB;
-def JL : IBr<0x8C, (ins brtarget:$dst), "jl\t$dst",
- [(X86brcond bb:$dst, X86_COND_L, EFLAGS)]>, TB;
-def JLE : IBr<0x8E, (ins brtarget:$dst), "jle\t$dst",
- [(X86brcond bb:$dst, X86_COND_LE, EFLAGS)]>, TB;
-def JG : IBr<0x8F, (ins brtarget:$dst), "jg\t$dst",
- [(X86brcond bb:$dst, X86_COND_G, EFLAGS)]>, TB;
-def JGE : IBr<0x8D, (ins brtarget:$dst), "jge\t$dst",
- [(X86brcond bb:$dst, X86_COND_GE, EFLAGS)]>, TB;
-
-def JB : IBr<0x82, (ins brtarget:$dst), "jb\t$dst",
- [(X86brcond bb:$dst, X86_COND_B, EFLAGS)]>, TB;
-def JBE : IBr<0x86, (ins brtarget:$dst), "jbe\t$dst",
- [(X86brcond bb:$dst, X86_COND_BE, EFLAGS)]>, TB;
-def JA : IBr<0x87, (ins brtarget:$dst), "ja\t$dst",
- [(X86brcond bb:$dst, X86_COND_A, EFLAGS)]>, TB;
-def JAE : IBr<0x83, (ins brtarget:$dst), "jae\t$dst",
- [(X86brcond bb:$dst, X86_COND_AE, EFLAGS)]>, TB;
-
-def JS : IBr<0x88, (ins brtarget:$dst), "js\t$dst",
- [(X86brcond bb:$dst, X86_COND_S, EFLAGS)]>, TB;
-def JNS : IBr<0x89, (ins brtarget:$dst), "jns\t$dst",
- [(X86brcond bb:$dst, X86_COND_NS, EFLAGS)]>, TB;
-def JP : IBr<0x8A, (ins brtarget:$dst), "jp\t$dst",
- [(X86brcond bb:$dst, X86_COND_P, EFLAGS)]>, TB;
-def JNP : IBr<0x8B, (ins brtarget:$dst), "jnp\t$dst",
- [(X86brcond bb:$dst, X86_COND_NP, EFLAGS)]>, TB;
-def JO : IBr<0x80, (ins brtarget:$dst), "jo\t$dst",
- [(X86brcond bb:$dst, X86_COND_O, EFLAGS)]>, TB;
-def JNO : IBr<0x81, (ins brtarget:$dst), "jno\t$dst",
- [(X86brcond bb:$dst, X86_COND_NO, EFLAGS)]>, TB;
-} // Uses = [EFLAGS]
// Loop instructions
XMM0, XMM1, XMM2, XMM3, XMM4, XMM5, XMM6, XMM7,
XMM8, XMM9, XMM10, XMM11, XMM12, XMM13, XMM14, XMM15, EFLAGS],
Uses = [ESP] in {
- def CALLpcrel32 : Ii32<0xE8, RawFrm,
+ def CALLpcrel32 : Ii32PCRel<0xE8, RawFrm,
(outs), (ins i32imm_pcrel:$dst,variable_ops),
"call\t$dst", []>;
def CALL32r : I<0xFF, MRM2r, (outs), (ins GR32:$dst, variable_ops),
"#TC_RETURN $dst $offset",
[]>;
-let isCall = 1, isTerminator = 1, isReturn = 1, isBarrier = 1 in
- def TAILJMPd : IBr<0xE9, (ins i32imm_pcrel:$dst), "jmp\t$dst # TAILCALL",
+// FIXME: The should be pseudo instructions that are lowered when going to
+// mcinst.
+let isCall = 1, isBranch = 1, isTerminator = 1, isReturn = 1, isBarrier = 1 in
+ def TAILJMPd : Ii32<0xE9, RawFrm, (outs),(ins i32imm_pcrel:$dst,variable_ops),
+ "jmp\t$dst # TAILCALL",
[]>;
let isCall = 1, isTerminator = 1, isReturn = 1, isBarrier = 1 in
- def TAILJMPr : I<0xFF, MRM4r, (outs), (ins GR32:$dst),
+ def TAILJMPr : I<0xFF, MRM4r, (outs), (ins GR32:$dst, variable_ops),
"jmp{l}\t{*}$dst # TAILCALL",
[]>;
let isCall = 1, isTerminator = 1, isReturn = 1, isBarrier = 1 in
- def TAILJMPm : I<0xFF, MRM4m, (outs), (ins i32mem:$dst),
+ def TAILJMPm : I<0xFF, MRM4m, (outs), (ins i32mem:$dst, variable_ops),
"jmp\t{*}$dst # TAILCALL", []>;
//===----------------------------------------------------------------------===//
"lea{l}\t{$src|$dst}, {$dst|$src}",
[(set GR32:$dst, lea32addr:$src)]>, Requires<[In32BitMode]>;
-let Defs = [ECX,EDI,ESI], Uses = [ECX,EDI,ESI] in {
+let Defs = [ECX,EDI,ESI], Uses = [ECX,EDI,ESI], isCodeGenOnly = 1 in {
def REP_MOVSB : I<0xA4, RawFrm, (outs), (ins), "{rep;movsb|rep movsb}",
[(X86rep_movs i8)]>, REP;
def REP_MOVSW : I<0xA5, RawFrm, (outs), (ins), "{rep;movsw|rep movsw}",
[(X86rep_movs i32)]>, REP;
}
-let Defs = [ECX,EDI], Uses = [AL,ECX,EDI] in
+// 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}", []>;
+}
+
+let Defs = [ECX,EDI], Uses = [AL,ECX,EDI], isCodeGenOnly = 1 in
def REP_STOSB : I<0xAA, RawFrm, (outs), (ins), "{rep;stosb|rep stosb}",
[(X86rep_stos i8)]>, REP;
-let Defs = [ECX,EDI], Uses = [AX,ECX,EDI] in
+let Defs = [ECX,EDI], Uses = [AX,ECX,EDI], isCodeGenOnly = 1 in
def REP_STOSW : I<0xAB, RawFrm, (outs), (ins), "{rep;stosw|rep stosw}",
[(X86rep_stos i16)]>, REP, OpSize;
-let Defs = [ECX,EDI], Uses = [EAX,ECX,EDI] in
+let Defs = [ECX,EDI], Uses = [EAX,ECX,EDI], isCodeGenOnly = 1 in
def REP_STOSD : I<0xAB, RawFrm, (outs), (ins), "{rep;stosl|rep stosd}",
[(X86rep_stos i32)]>, REP;
+// 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}", []>;
+let Defs = [EDI], Uses = [AX,EDI,EFLAGS] in
+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 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 RDTSC : I<0x31, RawFrm, (outs), (ins), "rdtsc", [(X86rdtsc)]>,
TB;
+let Defs = [RAX, RCX, RDX] in
+def RDTSCP : I<0x01, MRM_F9, (outs), (ins), "rdtscp", []>, TB;
+
let isBarrier = 1, hasCtrlDep = 1 in {
def TRAP : I<0x0B, RawFrm, (outs), (ins), "ud2", [(trap)]>, TB;
}
"mov{l}\t{$src, $dst|$dst, $src}",
[(set GR32:$dst, imm:$src)]>;
}
+
def MOV8mi : Ii8 <0xC6, MRM0m, (outs), (ins i8mem :$dst, i8imm :$src),
"mov{b}\t{$src, $dst|$dst, $src}",
[(store (i8 imm:$src), addr:$dst)]>;
def MOV32rr_REV : I<0x8B, MRMSrcReg, (outs GR32:$dst), (ins GR32:$src),
"mov{l}\t{$src, $dst|$dst, $src}", []>;
-let canFoldAsLoad = 1, isReMaterializable = 1, mayHaveSideEffects = 1 in {
+let canFoldAsLoad = 1, isReMaterializable = 1 in {
def MOV8rm : I<0x8A, MRMSrcMem, (outs GR8 :$dst), (ins i8mem :$src),
"mov{b}\t{$src, $dst|$dst, $src}",
[(set GR8:$dst, (loadi8 addr:$src))]>;
(outs), (ins i8mem_NOREX:$dst, GR8_NOREX:$src),
"mov{b}\t{$src, $dst|$dst, $src} # NOREX", []>;
let mayLoad = 1,
- canFoldAsLoad = 1, isReMaterializable = 1, mayHaveSideEffects = 1 in
+ canFoldAsLoad = 1, isReMaterializable = 1 in
def MOV8rm_NOREX : I<0x8A, MRMSrcMem,
(outs GR8_NOREX:$dst), (ins i8mem_NOREX:$src),
"mov{b}\t{$src, $dst|$dst, $src} # NOREX", []>;
//
// Extra precision multiplication
-let Defs = [AL,AH,EFLAGS], Uses = [AL] in
+
+// AL is really implied by AX, by the registers in Defs must match the
+// SDNode results (i8, i32).
+let Defs = [AL,EFLAGS,AX], Uses = [AL] in
def MUL8r : I<0xF6, MRM4r, (outs), (ins GR8:$src), "mul{b}\t$src",
// FIXME: Used for 8-bit mul, ignore result upper 8 bits.
// This probably ought to be moved to a def : Pat<> if the
"mul{l}\t$src",
[]>; // EAX,EDX = EAX*GR32
-let Defs = [AL,AH,EFLAGS], Uses = [AL] in
+let Defs = [AL,EFLAGS,AX], Uses = [AL] in
def MUL8m : I<0xF6, MRM4m, (outs), (ins i8mem :$src),
"mul{b}\t$src",
// FIXME: Used for 8-bit mul, ignore result upper 8 bits.
}
let neverHasSideEffects = 1 in {
-let Defs = [AL,AH,EFLAGS], Uses = [AL] in
+let Defs = [AL,EFLAGS,AX], Uses = [AL] in
def IMUL8r : I<0xF6, MRM5r, (outs), (ins GR8:$src), "imul{b}\t$src", []>;
// AL,AH = AL*GR8
let Defs = [AX,DX,EFLAGS], Uses = [AX] in
def IMUL32r : I<0xF7, MRM5r, (outs), (ins GR32:$src), "imul{l}\t$src", []>;
// EAX,EDX = EAX*GR32
let mayLoad = 1 in {
-let Defs = [AL,AH,EFLAGS], Uses = [AL] in
+let Defs = [AL,EFLAGS,AX], Uses = [AL] in
def IMUL8m : I<0xF6, MRM5m, (outs), (ins i8mem :$src),
"imul{b}\t$src", []>; // AL,AH = AL*[mem8]
let Defs = [AX,DX,EFLAGS], Uses = [AX] in
} // neverHasSideEffects
// unsigned division/remainder
-let Defs = [AL,AH,EFLAGS], Uses = [AX] in
+let Defs = [AL,EFLAGS,AX], Uses = [AX] in
def DIV8r : I<0xF6, MRM6r, (outs), (ins GR8:$src), // AX/r8 = AL,AH
"div{b}\t$src", []>;
let Defs = [AX,DX,EFLAGS], Uses = [AX,DX] in
def DIV32r : I<0xF7, MRM6r, (outs), (ins GR32:$src), // EDX:EAX/r32 = EAX,EDX
"div{l}\t$src", []>;
let mayLoad = 1 in {
-let Defs = [AL,AH,EFLAGS], Uses = [AX] in
+let Defs = [AL,EFLAGS,AX], Uses = [AX] in
def DIV8m : I<0xF6, MRM6m, (outs), (ins i8mem:$src), // AX/[mem8] = AL,AH
"div{b}\t$src", []>;
let Defs = [AX,DX,EFLAGS], Uses = [AX,DX] in
}
// Signed division/remainder.
-let Defs = [AL,AH,EFLAGS], Uses = [AX] in
+let Defs = [AL,EFLAGS,AX], Uses = [AX] in
def IDIV8r : I<0xF6, MRM7r, (outs), (ins GR8:$src), // AX/r8 = AL,AH
"idiv{b}\t$src", []>;
let Defs = [AX,DX,EFLAGS], Uses = [AX,DX] in
def IDIV32r: I<0xF7, MRM7r, (outs), (ins GR32:$src), // EDX:EAX/r32 = EAX,EDX
"idiv{l}\t$src", []>;
let mayLoad = 1, mayLoad = 1 in {
-let Defs = [AL,AH,EFLAGS], Uses = [AX] in
+let Defs = [AL,EFLAGS,AX], Uses = [AX] in
def IDIV8m : I<0xF6, MRM7m, (outs), (ins i8mem:$src), // AX/[mem8] = AL,AH
"idiv{b}\t$src", []>;
let Defs = [AX,DX,EFLAGS], Uses = [AX,DX] in
def OR8ri : Ii8 <0x80, MRM1r, (outs GR8 :$dst),
(ins GR8 :$src1, i8imm:$src2),
"or{b}\t{$src2, $dst|$dst, $src2}",
- [(set GR8:$dst, (or_not_add GR8:$src1, imm:$src2)),
+ [(set GR8:$dst, (or GR8:$src1, imm:$src2)),
(implicit EFLAGS)]>;
def OR16ri : Ii16<0x81, MRM1r, (outs GR16:$dst),
(ins GR16:$src1, i16imm:$src2),
"or{w}\t{$src2, $dst|$dst, $src2}",
- [(set GR16:$dst, (or_not_add GR16:$src1, imm:$src2)),
+ [(set GR16:$dst, (or GR16:$src1, imm:$src2)),
(implicit EFLAGS)]>, OpSize;
def OR32ri : Ii32<0x81, MRM1r, (outs GR32:$dst),
(ins GR32:$src1, i32imm:$src2),
"or{l}\t{$src2, $dst|$dst, $src2}",
- [(set GR32:$dst, (or_not_add GR32:$src1, imm:$src2)),
+ [(set GR32:$dst, (or GR32:$src1, imm:$src2)),
(implicit EFLAGS)]>;
def OR16ri8 : Ii8<0x83, MRM1r, (outs GR16:$dst),
(ins GR16:$src1, i16i8imm:$src2),
"or{w}\t{$src2, $dst|$dst, $src2}",
- [(set GR16:$dst, (or_not_add GR16:$src1, i16immSExt8:$src2)),
+ [(set GR16:$dst, (or GR16:$src1, i16immSExt8:$src2)),
(implicit EFLAGS)]>, OpSize;
def OR32ri8 : Ii8<0x83, MRM1r, (outs GR32:$dst),
(ins GR32:$src1, i32i8imm:$src2),
"or{l}\t{$src2, $dst|$dst, $src2}",
- [(set GR32:$dst, (or_not_add GR32:$src1, i32immSExt8:$src2)),
+ [(set GR32:$dst, (or GR32:$src1, i32immSExt8:$src2)),
(implicit EFLAGS)]>;
let isTwoAddress = 0 in {
def OR8mr : I<0x08, MRMDestMem, (outs), (ins i8mem:$dst, GR8:$src),
def RCL8r1 : I<0xD0, MRM2r, (outs GR8:$dst), (ins GR8:$src),
"rcl{b}\t{1, $dst|$dst, 1}", []>;
-def RCL8m1 : I<0xD0, MRM2m, (outs i8mem:$dst), (ins i8mem:$src),
- "rcl{b}\t{1, $dst|$dst, 1}", []>;
let Uses = [CL] in {
def RCL8rCL : I<0xD2, MRM2r, (outs GR8:$dst), (ins GR8:$src),
"rcl{b}\t{%cl, $dst|$dst, CL}", []>;
-def RCL8mCL : I<0xD2, MRM2m, (outs i8mem:$dst), (ins i8mem:$src),
- "rcl{b}\t{%cl, $dst|$dst, CL}", []>;
}
def RCL8ri : Ii8<0xC0, MRM2r, (outs GR8:$dst), (ins GR8:$src, i8imm:$cnt),
"rcl{b}\t{$cnt, $dst|$dst, $cnt}", []>;
-def RCL8mi : Ii8<0xC0, MRM2m, (outs i8mem:$dst), (ins i8mem:$src, i8imm:$cnt),
- "rcl{b}\t{$cnt, $dst|$dst, $cnt}", []>;
def RCL16r1 : I<0xD1, MRM2r, (outs GR16:$dst), (ins GR16:$src),
"rcl{w}\t{1, $dst|$dst, 1}", []>, OpSize;
-def RCL16m1 : I<0xD1, MRM2m, (outs i16mem:$dst), (ins i16mem:$src),
- "rcl{w}\t{1, $dst|$dst, 1}", []>, OpSize;
let Uses = [CL] in {
def RCL16rCL : I<0xD3, MRM2r, (outs GR16:$dst), (ins GR16:$src),
"rcl{w}\t{%cl, $dst|$dst, CL}", []>, OpSize;
-def RCL16mCL : I<0xD3, MRM2m, (outs i16mem:$dst), (ins i16mem:$src),
- "rcl{w}\t{%cl, $dst|$dst, CL}", []>, OpSize;
}
def RCL16ri : Ii8<0xC1, MRM2r, (outs GR16:$dst), (ins GR16:$src, i8imm:$cnt),
"rcl{w}\t{$cnt, $dst|$dst, $cnt}", []>, OpSize;
-def RCL16mi : Ii8<0xC1, MRM2m, (outs i16mem:$dst),
- (ins i16mem:$src, i8imm:$cnt),
- "rcl{w}\t{$cnt, $dst|$dst, $cnt}", []>, OpSize;
def RCL32r1 : I<0xD1, MRM2r, (outs GR32:$dst), (ins GR32:$src),
"rcl{l}\t{1, $dst|$dst, 1}", []>;
-def RCL32m1 : I<0xD1, MRM2m, (outs i32mem:$dst), (ins i32mem:$src),
- "rcl{l}\t{1, $dst|$dst, 1}", []>;
let Uses = [CL] in {
def RCL32rCL : I<0xD3, MRM2r, (outs GR32:$dst), (ins GR32:$src),
"rcl{l}\t{%cl, $dst|$dst, CL}", []>;
-def RCL32mCL : I<0xD3, MRM2m, (outs i32mem:$dst), (ins i32mem:$src),
- "rcl{l}\t{%cl, $dst|$dst, CL}", []>;
}
def RCL32ri : Ii8<0xC1, MRM2r, (outs GR32:$dst), (ins GR32:$src, i8imm:$cnt),
"rcl{l}\t{$cnt, $dst|$dst, $cnt}", []>;
-def RCL32mi : Ii8<0xC1, MRM2m, (outs i32mem:$dst),
- (ins i32mem:$src, i8imm:$cnt),
- "rcl{l}\t{$cnt, $dst|$dst, $cnt}", []>;
def RCR8r1 : I<0xD0, MRM3r, (outs GR8:$dst), (ins GR8:$src),
"rcr{b}\t{1, $dst|$dst, 1}", []>;
-def RCR8m1 : I<0xD0, MRM3m, (outs i8mem:$dst), (ins i8mem:$src),
- "rcr{b}\t{1, $dst|$dst, 1}", []>;
let Uses = [CL] in {
def RCR8rCL : I<0xD2, MRM3r, (outs GR8:$dst), (ins GR8:$src),
"rcr{b}\t{%cl, $dst|$dst, CL}", []>;
-def RCR8mCL : I<0xD2, MRM3m, (outs i8mem:$dst), (ins i8mem:$src),
- "rcr{b}\t{%cl, $dst|$dst, CL}", []>;
}
def RCR8ri : Ii8<0xC0, MRM3r, (outs GR8:$dst), (ins GR8:$src, i8imm:$cnt),
"rcr{b}\t{$cnt, $dst|$dst, $cnt}", []>;
-def RCR8mi : Ii8<0xC0, MRM3m, (outs i8mem:$dst), (ins i8mem:$src, i8imm:$cnt),
- "rcr{b}\t{$cnt, $dst|$dst, $cnt}", []>;
def RCR16r1 : I<0xD1, MRM3r, (outs GR16:$dst), (ins GR16:$src),
"rcr{w}\t{1, $dst|$dst, 1}", []>, OpSize;
-def RCR16m1 : I<0xD1, MRM3m, (outs i16mem:$dst), (ins i16mem:$src),
- "rcr{w}\t{1, $dst|$dst, 1}", []>, OpSize;
let Uses = [CL] in {
def RCR16rCL : I<0xD3, MRM3r, (outs GR16:$dst), (ins GR16:$src),
"rcr{w}\t{%cl, $dst|$dst, CL}", []>, OpSize;
-def RCR16mCL : I<0xD3, MRM3m, (outs i16mem:$dst), (ins i16mem:$src),
- "rcr{w}\t{%cl, $dst|$dst, CL}", []>, OpSize;
}
def RCR16ri : Ii8<0xC1, MRM3r, (outs GR16:$dst), (ins GR16:$src, i8imm:$cnt),
"rcr{w}\t{$cnt, $dst|$dst, $cnt}", []>, OpSize;
-def RCR16mi : Ii8<0xC1, MRM3m, (outs i16mem:$dst),
- (ins i16mem:$src, i8imm:$cnt),
- "rcr{w}\t{$cnt, $dst|$dst, $cnt}", []>, OpSize;
def RCR32r1 : I<0xD1, MRM3r, (outs GR32:$dst), (ins GR32:$src),
"rcr{l}\t{1, $dst|$dst, 1}", []>;
-def RCR32m1 : I<0xD1, MRM3m, (outs i32mem:$dst), (ins i32mem:$src),
- "rcr{l}\t{1, $dst|$dst, 1}", []>;
let Uses = [CL] in {
def RCR32rCL : I<0xD3, MRM3r, (outs GR32:$dst), (ins GR32:$src),
"rcr{l}\t{%cl, $dst|$dst, CL}", []>;
-def RCR32mCL : I<0xD3, MRM3m, (outs i32mem:$dst), (ins i32mem:$src),
- "rcr{l}\t{%cl, $dst|$dst, CL}", []>;
}
def RCR32ri : Ii8<0xC1, MRM3r, (outs GR32:$dst), (ins GR32:$src, i8imm:$cnt),
"rcr{l}\t{$cnt, $dst|$dst, $cnt}", []>;
-def RCR32mi : Ii8<0xC1, MRM3m, (outs i32mem:$dst),
- (ins i32mem:$src, i8imm:$cnt),
+
+let isTwoAddress = 0 in {
+def RCL8m1 : I<0xD0, MRM2m, (outs), (ins i8mem:$dst),
+ "rcl{b}\t{1, $dst|$dst, 1}", []>;
+def RCL8mi : Ii8<0xC0, MRM2m, (outs), (ins i8mem:$dst, i8imm:$cnt),
+ "rcl{b}\t{$cnt, $dst|$dst, $cnt}", []>;
+def RCL16m1 : I<0xD1, MRM2m, (outs), (ins i16mem:$dst),
+ "rcl{w}\t{1, $dst|$dst, 1}", []>, OpSize;
+def RCL16mi : Ii8<0xC1, MRM2m, (outs), (ins i16mem:$dst, i8imm:$cnt),
+ "rcl{w}\t{$cnt, $dst|$dst, $cnt}", []>, OpSize;
+def RCL32m1 : I<0xD1, MRM2m, (outs), (ins i32mem:$dst),
+ "rcl{l}\t{1, $dst|$dst, 1}", []>;
+def RCL32mi : Ii8<0xC1, MRM2m, (outs), (ins i32mem:$dst, i8imm:$cnt),
+ "rcl{l}\t{$cnt, $dst|$dst, $cnt}", []>;
+def RCR8m1 : I<0xD0, MRM3m, (outs), (ins i8mem:$dst),
+ "rcr{b}\t{1, $dst|$dst, 1}", []>;
+def RCR8mi : Ii8<0xC0, MRM3m, (outs), (ins i8mem:$dst, i8imm:$cnt),
+ "rcr{b}\t{$cnt, $dst|$dst, $cnt}", []>;
+def RCR16m1 : I<0xD1, MRM3m, (outs), (ins i16mem:$dst),
+ "rcr{w}\t{1, $dst|$dst, 1}", []>, OpSize;
+def RCR16mi : Ii8<0xC1, MRM3m, (outs), (ins i16mem:$dst, i8imm:$cnt),
+ "rcr{w}\t{$cnt, $dst|$dst, $cnt}", []>, OpSize;
+def RCR32m1 : I<0xD1, MRM3m, (outs), (ins i32mem:$dst),
+ "rcr{l}\t{1, $dst|$dst, 1}", []>;
+def RCR32mi : Ii8<0xC1, MRM3m, (outs), (ins i32mem:$dst, i8imm:$cnt),
"rcr{l}\t{$cnt, $dst|$dst, $cnt}", []>;
+let Uses = [CL] in {
+def RCL8mCL : I<0xD2, MRM2m, (outs), (ins i8mem:$dst),
+ "rcl{b}\t{%cl, $dst|$dst, CL}", []>;
+def RCL16mCL : I<0xD3, MRM2m, (outs), (ins i16mem:$dst),
+ "rcl{w}\t{%cl, $dst|$dst, CL}", []>, OpSize;
+def RCL32mCL : I<0xD3, MRM2m, (outs), (ins i32mem:$dst),
+ "rcl{l}\t{%cl, $dst|$dst, CL}", []>;
+def RCR8mCL : I<0xD2, MRM3m, (outs), (ins i8mem:$dst),
+ "rcr{b}\t{%cl, $dst|$dst, CL}", []>;
+def RCR16mCL : I<0xD3, MRM3m, (outs), (ins i16mem:$dst),
+ "rcr{w}\t{%cl, $dst|$dst, CL}", []>, OpSize;
+def RCR32mCL : I<0xD3, MRM3m, (outs), (ins i32mem:$dst),
+ "rcr{l}\t{%cl, $dst|$dst, CL}", []>;
+}
+}
+
// FIXME: provide shorter instructions when imm8 == 1
let Uses = [CL] in {
def ROL8rCL : I<0xD2, MRM0r, (outs GR8 :$dst), (ins GR8 :$src),
def SBB32mr : I<0x19, MRMDestMem, (outs), (ins i32mem:$dst, GR32:$src2),
"sbb{l}\t{$src2, $dst|$dst, $src2}",
[(store (sube (load addr:$dst), GR32:$src2), addr:$dst)]>;
- def SBB8mi : Ii32<0x80, MRM3m, (outs), (ins i8mem:$dst, i8imm:$src2),
- "sbb{b}\t{$src2, $dst|$dst, $src2}",
+ def SBB8mi : Ii8<0x80, MRM3m, (outs), (ins i8mem:$dst, i8imm:$src2),
+ "sbb{b}\t{$src2, $dst|$dst, $src2}",
[(store (sube (loadi8 addr:$dst), imm:$src2), addr:$dst)]>;
def SBB16mi : Ii16<0x81, MRM3m, (outs), (ins i16mem:$dst, i16imm:$src2),
"sbb{w}\t{$src2, $dst|$dst, $src2}",
let Uses = [EFLAGS] in {
// Use sbb to materialize carry bit.
-
let Defs = [EFLAGS], isCodeGenOnly = 1 in {
-def SETB_C8r : I<0x18, MRMInitReg, (outs GR8:$dst), (ins),
- "sbb{b}\t$dst, $dst",
+// FIXME: These are pseudo ops that should be replaced with Pat<> patterns.
+// However, Pat<> can't replicate the destination reg into the inputs of the
+// result.
+// FIXME: Change these to have encoding Pseudo when X86MCCodeEmitter replaces
+// X86CodeEmitter.
+def SETB_C8r : I<0x18, MRMInitReg, (outs GR8:$dst), (ins), "",
[(set GR8:$dst, (X86setcc_c X86_COND_B, EFLAGS))]>;
-def SETB_C16r : I<0x19, MRMInitReg, (outs GR16:$dst), (ins),
- "sbb{w}\t$dst, $dst",
+def SETB_C16r : I<0x19, MRMInitReg, (outs GR16:$dst), (ins), "",
[(set GR16:$dst, (X86setcc_c X86_COND_B, EFLAGS))]>,
OpSize;
-def SETB_C32r : I<0x19, MRMInitReg, (outs GR32:$dst), (ins),
- "sbb{l}\t$dst, $dst",
+def SETB_C32r : I<0x19, MRMInitReg, (outs GR32:$dst), (ins), "",
[(set GR32:$dst, (X86setcc_c X86_COND_B, EFLAGS))]>;
} // isCodeGenOnly
"movz{wl|x}\t{$src, $dst|$dst, $src}",
[(set GR32:$dst, (zextloadi32i16 addr:$src))]>, TB;
-// These are the same as the regular regular MOVZX32rr8 and MOVZX32rm8
+// These are the same as the regular MOVZX32rr8 and MOVZX32rm8
// except that they use GR32_NOREX for the output operand register class
// instead of GR32. This allows them to operate on h registers on x86-64.
def MOVZX32_NOREXrr8 : I<0xB6, MRMSrcReg,
// Alias instructions that map movr0 to xor.
// FIXME: remove when we can teach regalloc that xor reg, reg is ok.
+// FIXME: Set encoding to pseudo.
let Defs = [EFLAGS], isReMaterializable = 1, isAsCheapAsAMove = 1,
isCodeGenOnly = 1 in {
-def MOV8r0 : I<0x30, MRMInitReg, (outs GR8 :$dst), (ins),
- "xor{b}\t$dst, $dst",
+def MOV8r0 : I<0x30, MRMInitReg, (outs GR8 :$dst), (ins), "",
[(set GR8:$dst, 0)]>;
+
+// We want to rewrite MOV16r0 in terms of MOV32r0, because it's a smaller
+// encoding and avoids a partial-register update sometimes, but doing so
+// at isel time interferes with rematerialization in the current register
+// allocator. For now, this is rewritten when the instruction is lowered
+// to an MCInst.
+def MOV16r0 : I<0x31, MRMInitReg, (outs GR16:$dst), (ins),
+ "",
+ [(set GR16:$dst, 0)]>, OpSize;
-def MOV32r0 : I<0x31, MRMInitReg, (outs GR32:$dst), (ins),
- "xor{l}\t$dst, $dst",
+// FIXME: Set encoding to pseudo.
+def MOV32r0 : I<0x31, MRMInitReg, (outs GR32:$dst), (ins), "",
[(set GR32:$dst, 0)]>;
}
-// Use xorl instead of xorw since we don't care about the high 16 bits,
-// it's smaller, and it avoids a partial-register update.
-let AddedComplexity = 1 in
-def : Pat<(i16 0),
- (EXTRACT_SUBREG (MOV32r0), x86_subreg_16bit)>;
-
//===----------------------------------------------------------------------===//
// Thread Local Storage Instructions
//
def LSL32rr : I<0x03, MRMSrcReg, (outs GR32:$dst), (ins GR32:$src),
"lsl{l}\t{$src, $dst|$dst, $src}", []>, TB;
-def INVLPG : I<0x01, RawFrm, (outs), (ins), "invlpg", []>, TB;
+def INVLPG : I<0x01, MRM7m, (outs), (ins i8mem:$addr), "invlpg\t$addr", []>, TB;
def STRr : I<0x00, MRM1r, (outs GR16:$dst), (ins),
"str{w}\t{$dst}", []>, TB;
def LLDT16m : I<0x00, MRM2m, (outs), (ins i16mem:$src),
"lldt{w}\t$src", []>, TB;
+// Lock instruction prefix
+def LOCK_PREFIX : I<0xF0, RawFrm, (outs), (ins), "lock", []>;
+
+// Repeat string operation instruction prefixes
+// These uses the DF flag in the EFLAGS register to inc or dec ECX
+let Defs = [ECX], Uses = [ECX,EFLAGS] in {
+// Repeat (used with INS, OUTS, MOVS, LODS and STOS)
+def REP_PREFIX : I<0xF3, RawFrm, (outs), (ins), "rep", []>;
+// Repeat while not equal (used with CMPS and SCAS)
+def REPNE_PREFIX : I<0xF2, RawFrm, (outs), (ins), "repne", []>;
+}
+
+// Segment override instruction prefixes
+def CS_PREFIX : I<0x2E, RawFrm, (outs), (ins), "cs", []>;
+def SS_PREFIX : I<0x36, RawFrm, (outs), (ins), "ss", []>;
+def DS_PREFIX : I<0x3E, RawFrm, (outs), (ins), "ds", []>;
+def ES_PREFIX : I<0x26, RawFrm, (outs), (ins), "es", []>;
+def FS_PREFIX : I<0x64, RawFrm, (outs), (ins), "fs", []>;
+def GS_PREFIX : I<0x65, RawFrm, (outs), (ins), "gs", []>;
+
// String manipulation instructions
def LODSB : I<0xAC, RawFrm, (outs), (ins), "lodsb", []>;
// VMX instructions
// 66 0F 38 80
-def INVEPT : I<0x38, RawFrm, (outs), (ins), "invept", []>, OpSize, TB;
+def INVEPT : I<0x80, RawFrm, (outs), (ins), "invept", []>, OpSize, T8;
// 66 0F 38 81
-def INVVPID : I<0x38, RawFrm, (outs), (ins), "invvpid", []>, OpSize, TB;
+def INVVPID : I<0x81, RawFrm, (outs), (ins), "invvpid", []>, OpSize, T8;
// 0F 01 C1
-def VMCALL : I<0x01, RawFrm, (outs), (ins), "vmcall", []>, TB;
+def VMCALL : I<0x01, MRM_C1, (outs), (ins), "vmcall", []>, TB;
def VMCLEARm : I<0xC7, MRM6m, (outs), (ins i64mem:$vmcs),
"vmclear\t$vmcs", []>, OpSize, TB;
// 0F 01 C2
-def VMLAUNCH : I<0x01, RawFrm, (outs), (ins), "vmlaunch", []>, TB;
+def VMLAUNCH : I<0x01, MRM_C2, (outs), (ins), "vmlaunch", []>, TB;
// 0F 01 C3
-def VMRESUME : I<0x01, RawFrm, (outs), (ins), "vmresume", []>, TB;
+def VMRESUME : I<0x01, MRM_C3, (outs), (ins), "vmresume", []>, TB;
def VMPTRLDm : I<0xC7, MRM6m, (outs), (ins i64mem:$vmcs),
"vmptrld\t$vmcs", []>, TB;
def VMPTRSTm : I<0xC7, MRM7m, (outs i64mem:$vmcs), (ins),
def VMWRITE32rr : I<0x79, MRMSrcReg, (outs GR32:$dst), (ins GR32:$src),
"vmwrite{l}\t{$src, $dst|$dst, $src}", []>, TB;
// 0F 01 C4
-def VMXOFF : I<0x01, RawFrm, (outs), (ins), "vmxoff", []>, OpSize;
+def VMXOFF : I<0x01, MRM_C4, (outs), (ins), "vmxoff", []>, TB;
def VMXON : I<0xC7, MRM6m, (outs), (ins i64mem:$vmxon),
"vmxon\t{$vmxon}", []>, XD;
def : Pat<(i32 (anyext GR8 :$src)), (MOVZX32rr8 GR8 :$src)>;
def : Pat<(i32 (anyext GR16:$src)), (MOVZX32rr16 GR16:$src)>;
-// (and (i32 load), 255) -> (zextload i8)
-def : Pat<(i32 (and (nvloadi32 addr:$src), (i32 255))),
- (MOVZX32rm8 addr:$src)>;
-def : Pat<(i32 (and (nvloadi32 addr:$src), (i32 65535))),
- (MOVZX32rm16 addr:$src)>;
-
//===----------------------------------------------------------------------===//
// Some peepholes
//===----------------------------------------------------------------------===//
def : Pat<(shl GR32:$src1, (i8 1)), (ADD32rr GR32:$src1, GR32:$src1)>;
// (shl x (and y, 31)) ==> (shl x, y)
-def : Pat<(shl GR8:$src1, (and CL:$amt, 31)),
+def : Pat<(shl GR8:$src1, (and CL, 31)),
(SHL8rCL GR8:$src1)>;
-def : Pat<(shl GR16:$src1, (and CL:$amt, 31)),
+def : Pat<(shl GR16:$src1, (and CL, 31)),
(SHL16rCL GR16:$src1)>;
-def : Pat<(shl GR32:$src1, (and CL:$amt, 31)),
+def : Pat<(shl GR32:$src1, (and CL, 31)),
(SHL32rCL GR32:$src1)>;
-def : Pat<(store (shl (loadi8 addr:$dst), (and CL:$amt, 31)), addr:$dst),
+def : Pat<(store (shl (loadi8 addr:$dst), (and CL, 31)), addr:$dst),
(SHL8mCL addr:$dst)>;
-def : Pat<(store (shl (loadi16 addr:$dst), (and CL:$amt, 31)), addr:$dst),
+def : Pat<(store (shl (loadi16 addr:$dst), (and CL, 31)), addr:$dst),
(SHL16mCL addr:$dst)>;
-def : Pat<(store (shl (loadi32 addr:$dst), (and CL:$amt, 31)), addr:$dst),
+def : Pat<(store (shl (loadi32 addr:$dst), (and CL, 31)), addr:$dst),
(SHL32mCL addr:$dst)>;
-def : Pat<(srl GR8:$src1, (and CL:$amt, 31)),
+def : Pat<(srl GR8:$src1, (and CL, 31)),
(SHR8rCL GR8:$src1)>;
-def : Pat<(srl GR16:$src1, (and CL:$amt, 31)),
+def : Pat<(srl GR16:$src1, (and CL, 31)),
(SHR16rCL GR16:$src1)>;
-def : Pat<(srl GR32:$src1, (and CL:$amt, 31)),
+def : Pat<(srl GR32:$src1, (and CL, 31)),
(SHR32rCL GR32:$src1)>;
-def : Pat<(store (srl (loadi8 addr:$dst), (and CL:$amt, 31)), addr:$dst),
+def : Pat<(store (srl (loadi8 addr:$dst), (and CL, 31)), addr:$dst),
(SHR8mCL addr:$dst)>;
-def : Pat<(store (srl (loadi16 addr:$dst), (and CL:$amt, 31)), addr:$dst),
+def : Pat<(store (srl (loadi16 addr:$dst), (and CL, 31)), addr:$dst),
(SHR16mCL addr:$dst)>;
-def : Pat<(store (srl (loadi32 addr:$dst), (and CL:$amt, 31)), addr:$dst),
+def : Pat<(store (srl (loadi32 addr:$dst), (and CL, 31)), addr:$dst),
(SHR32mCL addr:$dst)>;
-def : Pat<(sra GR8:$src1, (and CL:$amt, 31)),
+def : Pat<(sra GR8:$src1, (and CL, 31)),
(SAR8rCL GR8:$src1)>;
-def : Pat<(sra GR16:$src1, (and CL:$amt, 31)),
+def : Pat<(sra GR16:$src1, (and CL, 31)),
(SAR16rCL GR16:$src1)>;
-def : Pat<(sra GR32:$src1, (and CL:$amt, 31)),
+def : Pat<(sra GR32:$src1, (and CL, 31)),
(SAR32rCL GR32:$src1)>;
-def : Pat<(store (sra (loadi8 addr:$dst), (and CL:$amt, 31)), addr:$dst),
+def : Pat<(store (sra (loadi8 addr:$dst), (and CL, 31)), addr:$dst),
(SAR8mCL addr:$dst)>;
-def : Pat<(store (sra (loadi16 addr:$dst), (and CL:$amt, 31)), addr:$dst),
+def : Pat<(store (sra (loadi16 addr:$dst), (and CL, 31)), addr:$dst),
(SAR16mCL addr:$dst)>;
-def : Pat<(store (sra (loadi32 addr:$dst), (and CL:$amt, 31)), addr:$dst),
+def : Pat<(store (sra (loadi32 addr:$dst), (and CL, 31)), addr:$dst),
(SAR32mCL addr:$dst)>;
// (or (x >> c) | (y << (32 - c))) ==> (shrd32 x, y, c)
addr:$dst),
(SHRD32mrCL addr:$dst, GR32:$src2)>;
-def : Pat<(shrd GR32:$src1, (i8 imm:$amt1), GR32:$src2, (i8 imm:$amt2)),
+def : Pat<(shrd GR32:$src1, (i8 imm:$amt1), GR32:$src2, (i8 imm/*:$amt2*/)),
(SHRD32rri8 GR32:$src1, GR32:$src2, (i8 imm:$amt1))>;
def : Pat<(store (shrd (loadi32 addr:$dst), (i8 imm:$amt1),
- GR32:$src2, (i8 imm:$amt2)), addr:$dst),
+ GR32:$src2, (i8 imm/*:$amt2*/)), addr:$dst),
(SHRD32mri8 addr:$dst, GR32:$src2, (i8 imm:$amt1))>;
// (or (x << c) | (y >> (32 - c))) ==> (shld32 x, y, c)
addr:$dst),
(SHLD32mrCL addr:$dst, GR32:$src2)>;
-def : Pat<(shld GR32:$src1, (i8 imm:$amt1), GR32:$src2, (i8 imm:$amt2)),
+def : Pat<(shld GR32:$src1, (i8 imm:$amt1), GR32:$src2, (i8 imm/*:$amt2*/)),
(SHLD32rri8 GR32:$src1, GR32:$src2, (i8 imm:$amt1))>;
def : Pat<(store (shld (loadi32 addr:$dst), (i8 imm:$amt1),
- GR32:$src2, (i8 imm:$amt2)), addr:$dst),
+ GR32:$src2, (i8 imm/*:$amt2*/)), addr:$dst),
(SHLD32mri8 addr:$dst, GR32:$src2, (i8 imm:$amt1))>;
// (or (x >> c) | (y << (16 - c))) ==> (shrd16 x, y, c)
addr:$dst),
(SHRD16mrCL addr:$dst, GR16:$src2)>;
-def : Pat<(shrd GR16:$src1, (i8 imm:$amt1), GR16:$src2, (i8 imm:$amt2)),
+def : Pat<(shrd GR16:$src1, (i8 imm:$amt1), GR16:$src2, (i8 imm/*:$amt2*/)),
(SHRD16rri8 GR16:$src1, GR16:$src2, (i8 imm:$amt1))>;
def : Pat<(store (shrd (loadi16 addr:$dst), (i8 imm:$amt1),
- GR16:$src2, (i8 imm:$amt2)), addr:$dst),
+ GR16:$src2, (i8 imm/*:$amt2*/)), addr:$dst),
(SHRD16mri8 addr:$dst, GR16:$src2, (i8 imm:$amt1))>;
// (or (x << c) | (y >> (16 - c))) ==> (shld16 x, y, c)
addr:$dst),
(SHLD16mrCL addr:$dst, GR16:$src2)>;
-def : Pat<(shld GR16:$src1, (i8 imm:$amt1), GR16:$src2, (i8 imm:$amt2)),
+def : Pat<(shld GR16:$src1, (i8 imm:$amt1), GR16:$src2, (i8 imm/*:$amt2*/)),
(SHLD16rri8 GR16:$src1, GR16:$src2, (i8 imm:$amt1))>;
def : Pat<(store (shld (loadi16 addr:$dst), (i8 imm:$amt1),
- GR16:$src2, (i8 imm:$amt2)), addr:$dst),
+ GR16:$src2, (i8 imm/*:$amt2*/)), addr:$dst),
(SHLD16mri8 addr:$dst, GR16:$src2, (i8 imm:$amt1))>;
// (anyext (setcc_carry)) -> (setcc_carry)
def : Pat<(i32 (anyext (i8 (X86setcc_c X86_COND_B, EFLAGS)))),
(SETB_C32r)>;
-// (or x, c) -> (add x, c) if masked bits are known zero.
-def : Pat<(parallel (or_is_add GR8:$src1, imm:$src2),
- (implicit EFLAGS)),
- (ADD8ri GR8:$src1, imm:$src2)>;
+// (or x1, x2) -> (add x1, x2) if two operands are known not to share bits.
+let AddedComplexity = 5 in { // Try this before the selecting to OR
def : Pat<(parallel (or_is_add GR16:$src1, imm:$src2),
(implicit EFLAGS)),
(ADD16ri GR16:$src1, imm:$src2)>;
def : Pat<(parallel (or_is_add GR32:$src1, i32immSExt8:$src2),
(implicit EFLAGS)),
(ADD32ri8 GR32:$src1, i32immSExt8:$src2)>;
+def : Pat<(parallel (or_is_add GR16:$src1, GR16:$src2),
+ (implicit EFLAGS)),
+ (ADD16rr GR16:$src1, GR16:$src2)>;
+def : Pat<(parallel (or_is_add GR32:$src1, GR32:$src2),
+ (implicit EFLAGS)),
+ (ADD32rr GR32:$src1, GR32:$src2)>;
+} // AddedComplexity
//===----------------------------------------------------------------------===//
// EFLAGS-defining Patterns
include "X86Instr64bit.td"
+//===----------------------------------------------------------------------===//
+// SIMD support (SSE, MMX and AVX)
+//===----------------------------------------------------------------------===//
+
+include "X86InstrFragmentsSIMD.td"
+
//===----------------------------------------------------------------------===//
// XMM Floating point support (requires SSE / SSE2)
//===----------------------------------------------------------------------===//