[SDTCisVT<0, i8>,
SDTCisVT<1, i8>, SDTCisVT<2, i32>]>;
+def SDTX86cas : SDTypeProfile<0, 3, [SDTCisPtrTy<0>, SDTCisInt<1>,
+ SDTCisVT<2, i8>]>;
+def SDTX86cas8 : SDTypeProfile<0, 1, [SDTCisPtrTy<0>]>;
+
def SDTX86Ret : SDTypeProfile<0, 1, [SDTCisVT<0, i16>]>;
def SDT_X86CallSeqStart : SDCallSeqStart<[ SDTCisVT<0, i32> ]>;
[SDNPHasChain]>;
def X86setcc : SDNode<"X86ISD::SETCC", SDTX86SetCC>;
+def X86cas : SDNode<"X86ISD::LCMPXCHG_DAG", SDTX86cas,
+ [SDNPHasChain, SDNPInFlag, SDNPOutFlag, SDNPMayStore,
+ SDNPMayLoad]>;
+def X86cas8 : SDNode<"X86ISD::LCMPXCHG8_DAG", SDTX86cas8,
+ [SDNPHasChain, SDNPInFlag, SDNPOutFlag, SDNPMayStore,
+ SDNPMayLoad]>;
+
def X86retflag : SDNode<"X86ISD::RET_FLAG", SDTX86Ret,
[SDNPHasChain, SDNPOptInFlag]>;
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 FPStackf32 : Predicate<"!Subtarget->hasSSE1()">;
def FPStackf64 : Predicate<"!Subtarget->hasSSE2()">;
def In32BitMode : Predicate<"!Subtarget->is64Bit()">;
def In64BitMode : Predicate<"Subtarget->is64Bit()">;
-def HasLow4G : Predicate<"Subtarget->hasLow4GUserSpaceAddress()">;
def SmallCode : Predicate<"TM.getCodeModel() == CodeModel::Small">;
def NotSmallCode : Predicate<"TM.getCodeModel() != CodeModel::Small">;
def IsStatic : Predicate<"TM.getRelocationModel() == Reloc::Static">;
def loadf64 : PatFrag<(ops node:$ptr), (f64 (load node:$ptr))>;
def loadf80 : PatFrag<(ops node:$ptr), (f80 (load node:$ptr))>;
-def sextloadi16i1 : PatFrag<(ops node:$ptr), (i16 (sextloadi1 node:$ptr))>;
-def sextloadi32i1 : PatFrag<(ops node:$ptr), (i32 (sextloadi1 node:$ptr))>;
def sextloadi16i8 : PatFrag<(ops node:$ptr), (i16 (sextloadi8 node:$ptr))>;
def sextloadi32i8 : PatFrag<(ops node:$ptr), (i32 (sextloadi8 node:$ptr))>;
def sextloadi32i16 : PatFrag<(ops node:$ptr), (i32 (sextloadi16 node:$ptr))>;
def extloadi32i8 : PatFrag<(ops node:$ptr), (i32 (extloadi8 node:$ptr))>;
def extloadi32i16 : PatFrag<(ops node:$ptr), (i32 (extloadi16 node:$ptr))>;
+
+// An 'and' node with a single use.
+def and_su : PatFrag<(ops node:$lhs, node:$rhs), (and node:$lhs, node:$rhs), [{
+ return N->hasOneUse();
+}]>;
+
//===----------------------------------------------------------------------===//
// Instruction list...
//
"bswap{l}\t$dst",
[(set GR32:$dst, (bswap GR32:$src))]>, TB;
-// FIXME: Model xchg* as two address instructions?
-def XCHG8rr : I<0x86, MRMDestReg, // xchg GR8, GR8
- (outs), (ins GR8:$src1, GR8:$src2),
- "xchg{b}\t{$src2|$src1}, {$src1|$src2}", []>;
-def XCHG16rr : I<0x87, MRMDestReg, // xchg GR16, GR16
- (outs), (ins GR16:$src1, GR16:$src2),
- "xchg{w}\t{$src2|$src1}, {$src1|$src2}", []>, OpSize;
-def XCHG32rr : I<0x87, MRMDestReg, // xchg GR32, GR32
- (outs), (ins GR32:$src1, GR32:$src2),
- "xchg{l}\t{$src2|$src1}, {$src1|$src2}", []>;
-
-def XCHG8mr : I<0x86, MRMDestMem,
- (outs), (ins i8mem:$src1, GR8:$src2),
- "xchg{b}\t{$src2|$src1}, {$src1|$src2}", []>;
-def XCHG16mr : I<0x87, MRMDestMem,
- (outs), (ins i16mem:$src1, GR16:$src2),
- "xchg{w}\t{$src2|$src1}, {$src1|$src2}", []>, OpSize;
-def XCHG32mr : I<0x87, MRMDestMem,
- (outs), (ins i32mem:$src1, GR32:$src2),
- "xchg{l}\t{$src2|$src1}, {$src1|$src2}", []>;
-def XCHG8rm : I<0x86, MRMSrcMem,
- (outs), (ins GR8:$src1, i8mem:$src2),
- "xchg{b}\t{$src2|$src1}, {$src1|$src2}", []>;
-def XCHG16rm : I<0x87, MRMSrcMem,
- (outs), (ins GR16:$src1, i16mem:$src2),
- "xchg{w}\t{$src2|$src1}, {$src1|$src2}", []>, OpSize;
-def XCHG32rm : I<0x87, MRMSrcMem,
- (outs), (ins GR32:$src1, i32mem:$src2),
- "xchg{l}\t{$src2|$src1}, {$src1|$src2}", []>;
// Bit scan instructions.
let Defs = [EFLAGS] in {
def RDTSC : I<0x31, RawFrm, (outs), (ins), "rdtsc", [(X86rdtsc)]>,
TB;
+let isBarrier = 1, hasCtrlDep = 1 in {
+def TRAP : I<0x0B, RawFrm, (outs), (ins), "ud2", [(trap)]>, TB;
+}
+
//===----------------------------------------------------------------------===//
// Input/Output Instructions...
//
// This probably ought to be moved to a def : Pat<> if the
// syntax can be accepted.
[(set AL, (mul AL, GR8:$src))]>; // AL,AH = AL*GR8
-let Defs = [AX,DX,EFLAGS], Uses = [AX] in
+let Defs = [AX,DX,EFLAGS], Uses = [AX], neverHasSideEffects = 1 in
def MUL16r : I<0xF7, MRM4r, (outs), (ins GR16:$src), "mul{w}\t$src", []>,
OpSize; // AX,DX = AX*GR16
-let Defs = [EAX,EDX,EFLAGS], Uses = [EAX] in
+let Defs = [EAX,EDX,EFLAGS], Uses = [EAX], neverHasSideEffects = 1 in
def MUL32r : I<0xF7, MRM4r, (outs), (ins GR32:$src), "mul{l}\t$src", []>;
// EAX,EDX = EAX*GR32
let Defs = [AL,AH,EFLAGS], Uses = [AL] in
def SHL32ri : Ii8<0xC1, MRM4r, (outs GR32:$dst), (ins GR32:$src1, i8imm:$src2),
"shl{l}\t{$src2, $dst|$dst, $src2}",
[(set GR32:$dst, (shl GR32:$src1, (i8 imm:$src2)))]>;
+// NOTE: We don't use shifts of a register by one, because 'add reg,reg' is
+// cheaper.
}
-// Shift left by one. Not used because (add x, x) is slightly cheaper.
-def SHL8r1 : I<0xD0, MRM4r, (outs GR8 :$dst), (ins GR8 :$src1),
- "shl{b}\t$dst", []>;
-def SHL16r1 : I<0xD1, MRM4r, (outs GR16:$dst), (ins GR16:$src1),
- "shl{w}\t$dst", []>, OpSize;
-def SHL32r1 : I<0xD1, MRM4r, (outs GR32:$dst), (ins GR32:$src1),
- "shl{l}\t$dst", []>;
-
let isTwoAddress = 0 in {
let Uses = [CL] in {
def SHL8mCL : I<0xD2, MRM4m, (outs), (ins i8mem :$dst),
let isCommutable = 1 in { // TEST X, Y --> TEST Y, X
def TEST8rr : I<0x84, MRMDestReg, (outs), (ins GR8:$src1, GR8:$src2),
"test{b}\t{$src2, $src1|$src1, $src2}",
- [(X86cmp (and GR8:$src1, GR8:$src2), 0),
+ [(X86cmp (and_su GR8:$src1, GR8:$src2), 0),
(implicit EFLAGS)]>;
def TEST16rr : I<0x85, MRMDestReg, (outs), (ins GR16:$src1, GR16:$src2),
"test{w}\t{$src2, $src1|$src1, $src2}",
- [(X86cmp (and GR16:$src1, GR16:$src2), 0),
+ [(X86cmp (and_su GR16:$src1, GR16:$src2), 0),
(implicit EFLAGS)]>,
OpSize;
def TEST32rr : I<0x85, MRMDestReg, (outs), (ins GR32:$src1, GR32:$src2),
"test{l}\t{$src2, $src1|$src1, $src2}",
- [(X86cmp (and GR32:$src1, GR32:$src2), 0),
+ [(X86cmp (and_su GR32:$src1, GR32:$src2), 0),
(implicit EFLAGS)]>;
}
def TEST8ri : Ii8 <0xF6, MRM0r, // flags = GR8 & imm8
(outs), (ins GR8:$src1, i8imm:$src2),
"test{b}\t{$src2, $src1|$src1, $src2}",
- [(X86cmp (and GR8:$src1, imm:$src2), 0),
+ [(X86cmp (and_su GR8:$src1, imm:$src2), 0),
(implicit EFLAGS)]>;
def TEST16ri : Ii16<0xF7, MRM0r, // flags = GR16 & imm16
(outs), (ins GR16:$src1, i16imm:$src2),
"test{w}\t{$src2, $src1|$src1, $src2}",
- [(X86cmp (and GR16:$src1, imm:$src2), 0),
+ [(X86cmp (and_su GR16:$src1, imm:$src2), 0),
(implicit EFLAGS)]>, OpSize;
def TEST32ri : Ii32<0xF7, MRM0r, // flags = GR32 & imm32
(outs), (ins GR32:$src1, i32imm:$src2),
"test{l}\t{$src2, $src1|$src1, $src2}",
- [(X86cmp (and GR32:$src1, imm:$src2), 0),
+ [(X86cmp (and_su GR32:$src1, imm:$src2), 0),
(implicit EFLAGS)]>;
def TEST8mi : Ii8 <0xF6, MRM0m, // flags = [mem8] & imm8
}
+//===----------------------------------------------------------------------===//
+// Atomic support
+//
+
+//FIXME: Please check the format Pseudo is certainly wrong, but the opcode and
+// prefixes should be correct
+
+let Defs = [EAX, EFLAGS], Uses = [EAX] in {
+def CMPXCHG32 : I<0xB1, Pseudo, (outs), (ins i32mem:$ptr, GR32:$swap),
+ "cmpxchgl $swap,$ptr", []>, TB;
+def LCMPXCHG32 : I<0xB1, Pseudo, (outs), (ins i32mem:$ptr, GR32:$swap),
+ "lock cmpxchgl $swap,$ptr",
+ [(X86cas addr:$ptr, GR32:$swap, 4)]>, TB, LOCK;
+}
+let Defs = [EAX, EBX, ECX, EDX, EFLAGS], Uses = [EAX, EBX, ECX, EDX] in {
+def CMPXCHG8B : I<0xC7, Pseudo, (outs), (ins i32mem:$ptr),
+ "cmpxchg8b $ptr", []>, TB;
+def LCMPXCHG8B : I<0xC7, Pseudo, (outs), (ins i32mem:$ptr),
+ "lock cmpxchg8b $ptr",
+ [(X86cas8 addr:$ptr)]>, TB, LOCK;
+}
+
+let Defs = [AX, EFLAGS], Uses = [AX] in {
+def CMPXCHG16 : I<0xB1, Pseudo, (outs), (ins i16mem:$ptr, GR16:$swap),
+ "cmpxchgw $swap,($ptr)", []>, TB, OpSize;
+def LCMPXCHG16 : I<0xB1, Pseudo, (outs), (ins i16mem:$ptr, GR16:$swap),
+ "lock cmpxchgw $swap,$ptr",
+ [(X86cas addr:$ptr, GR16:$swap, 2)]>, TB, OpSize, LOCK;
+}
+let Defs = [AL, EFLAGS], Uses = [AL] in {
+def CMPXCHG8 : I<0xB0, Pseudo, (outs), (ins i8mem:$ptr, GR8:$swap),
+ "cmpxchgb $swap,($ptr)", []>, TB;
+def LCMPXCHG8 : I<0xB0, Pseudo, (outs), (ins i8mem:$ptr, GR8:$swap),
+ "lock cmpxchgb $swap,$ptr",
+ [(X86cas addr:$ptr, GR8:$swap, 1)]>, TB, LOCK;
+}
+
+let Constraints = "$val = $dst", Defs = [EFLAGS] in {
+def LXADD32 : I<0xC1, Pseudo, (outs GR32:$dst), (ins i32mem:$ptr, GR32:$val),
+ "lock xadd $val, $ptr",
+ [(set GR32:$dst, (atomic_las_32 addr:$ptr, GR32:$val))]>,
+ TB, LOCK;
+def LXADD16 : I<0xC1, Pseudo, (outs GR16:$dst), (ins i16mem:$ptr, GR16:$val),
+ "lock xadd $val, $ptr",
+ [(set GR16:$dst, (atomic_las_16 addr:$ptr, GR16:$val))]>,
+ TB, OpSize, LOCK;
+def LXADD8 : I<0xC0, Pseudo, (outs GR8:$dst), (ins i8mem:$ptr, GR8:$val),
+ "lock xadd $val, $ptr",
+ [(set GR8:$dst, (atomic_las_8 addr:$ptr, GR8:$val))]>,
+ TB, LOCK;
+def XADD32 : I<0xC1, Pseudo, (outs GR32:$dst), (ins i32mem:$ptr, GR32:$val),
+ "xadd $val, $ptr", []>, TB;
+def XADD16 : I<0xC1, Pseudo, (outs GR16:$dst), (ins i16mem:$ptr, GR16:$val),
+ "xadd $val, $ptr", []>, TB, OpSize;
+def XADD8 : I<0xC0, Pseudo, (outs GR8:$dst), (ins i8mem:$ptr, GR8:$val),
+ "xadd $val, $ptr", []>, TB;
+
+def LXCHG32 : I<0x87, Pseudo, (outs GR32:$dst), (ins i32mem:$ptr, GR32:$val),
+ "lock xchg $val, $ptr",
+ [(set GR32:$dst, (atomic_swap_32 addr:$ptr, GR32:$val))]>, LOCK;
+def LXCHG16 : I<0x87, Pseudo, (outs GR16:$dst), (ins i16mem:$ptr, GR16:$val),
+ "lock xchg $val, $ptr",
+ [(set GR16:$dst, (atomic_swap_16 addr:$ptr, GR16:$val))]>,
+ OpSize, LOCK;
+def LXCHG8 : I<0x86, Pseudo, (outs GR8:$dst), (ins i8mem:$ptr, GR8:$val),
+ "lock xchg $val, $ptr",
+ [(set GR8:$dst, (atomic_swap_8 addr:$ptr, GR8:$val))]>, LOCK;
+def XCHG32 : I<0x87, Pseudo, (outs GR32:$dst), (ins i32mem:$ptr, GR32:$val),
+ "xchg $val, $ptr", []>;
+def XCHG16 : I<0x87, Pseudo, (outs GR16:$dst), (ins i16mem:$ptr, GR16:$val),
+ "xchg $val, $ptr", []>, OpSize;
+def XCHG8 : I<0x86, Pseudo, (outs GR8:$dst), (ins i8mem:$ptr, GR8:$val),
+ "xchg $val, $ptr", []>;
+}
+
//===----------------------------------------------------------------------===//
// Non-Instruction Patterns
//===----------------------------------------------------------------------===//
def : Pat<(subc GR32:$src1, i32immSExt8:$src2),
(SUB32ri8 GR32:$src1, i32immSExt8:$src2)>;
-def : Pat<(truncstorei1 (i8 imm:$src), addr:$dst),
- (MOV8mi addr:$dst, imm:$src)>;
-def : Pat<(truncstorei1 GR8:$src, addr:$dst),
- (MOV8mr addr:$dst, GR8:$src)>;
-
// Comparisons.
// TEST R,R is smaller than CMP R,0
def : Pat<(parallel (X86cmp GR32:$src1, 0), (implicit EFLAGS)),
(TEST32rr GR32:$src1, GR32:$src1)>;
-// {s|z}extload bool -> {s|z}extload byte
-def : Pat<(sextloadi16i1 addr:$src), (MOVSX16rm8 addr:$src)>;
-def : Pat<(sextloadi32i1 addr:$src), (MOVSX32rm8 addr:$src)>;
+// zextload bool -> zextload byte
def : Pat<(zextloadi8i1 addr:$src), (MOV8rm addr:$src)>;
def : Pat<(zextloadi16i1 addr:$src), (MOVZX16rm8 addr:$src)>;
def : Pat<(zextloadi32i1 addr:$src), (MOVZX32rm8 addr:$src)>;
(srl GR16:$src2, (sub 16, CL:$amt))), addr:$dst),
(SHLD16mrCL addr:$dst, GR16:$src2)>;
-
//===----------------------------------------------------------------------===//
// Floating Point Stack Support
//===----------------------------------------------------------------------===//
projects / oota-llvm.git / blobdiff