def SDTX86FPShiftOp : SDTypeProfile<1, 2, [ SDTCisSameAs<0, 1>,
SDTCisFP<0>, SDTCisInt<2> ]>;
def SDTX86VFCMP : SDTypeProfile<1, 3, [SDTCisInt<0>, SDTCisSameAs<1, 2>,
- SDTCisFP<1>, SDTCisVT<3, i8>]>;
+ SDTCisFP<1>, SDTCisVT<3, i8>,
+ SDTCisVec<1>]>;
def X86umin : SDNode<"X86ISD::UMIN", SDTIntBinOp>;
def X86umax : SDNode<"X86ISD::UMAX", SDTIntBinOp>;
def X86hsub : SDNode<"X86ISD::HSUB", SDTIntBinOp>;
def X86comi : SDNode<"X86ISD::COMI", SDTX86CmpTest>;
def X86ucomi : SDNode<"X86ISD::UCOMI", SDTX86CmpTest>;
-def X86cmpss : SDNode<"X86ISD::FSETCCss", SDTX86Cmpss>;
-def X86cmpsd : SDNode<"X86ISD::FSETCCsd", SDTX86Cmpsd>;
+def X86cmps : SDNode<"X86ISD::FSETCC", SDTX86Cmps>;
+//def X86cmpsd : SDNode<"X86ISD::FSETCCsd", SDTX86Cmpsd>;
def X86pshufb : SDNode<"X86ISD::PSHUFB",
SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0,1>,
SDTCisSameAs<0,2>]>>;
def X86pinsrw : SDNode<"X86ISD::PINSRW",
SDTypeProfile<1, 3, [SDTCisVT<0, v8i16>, SDTCisSameAs<0,1>,
SDTCisVT<2, i32>, SDTCisPtrTy<3>]>>;
-def X86insrtps : SDNode<"X86ISD::INSERTPS",
+def X86insertps : SDNode<"X86ISD::INSERTPS",
SDTypeProfile<1, 3, [SDTCisVT<0, v4f32>, SDTCisSameAs<0,1>,
- SDTCisVT<2, v4f32>, SDTCisPtrTy<3>]>>;
+ SDTCisVT<2, v4f32>, SDTCisVT<3, i8>]>>;
def X86vzmovl : SDNode<"X86ISD::VZEXT_MOVL",
SDTypeProfile<1, 1, [SDTCisSameAs<0,1>]>>;
-def X86vzmovly : SDNode<"X86ISD::VZEXT_MOVL",
- SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisVec<1>,
- SDTCisOpSmallerThanOp<1, 0> ]>>;
-
-def X86vsmovl : SDNode<"X86ISD::VSEXT_MOVL",
- SDTypeProfile<1, 1,
- [SDTCisVec<0>, SDTCisInt<1>, SDTCisInt<0>]>>;
-
def X86vzload : SDNode<"X86ISD::VZEXT_LOAD", SDTLoad,
[SDNPHasChain, SDNPMayLoad, SDNPMemOperand]>;
def X86vzext : SDNode<"X86ISD::VZEXT",
SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisVec<1>,
- SDTCisInt<0>, SDTCisInt<1>]>>;
+ SDTCisInt<0>, SDTCisInt<1>,
+ SDTCisOpSmallerThanOp<1, 0>]>>;
def X86vsext : SDNode<"X86ISD::VSEXT",
SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisVec<1>,
- SDTCisInt<0>, SDTCisInt<1>]>>;
+ SDTCisInt<0>, SDTCisInt<1>,
+ SDTCisOpSmallerThanOp<1, 0>]>>;
+def X86vtrunc : SDNode<"X86ISD::VTRUNC",
+ SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisVec<1>,
+ SDTCisInt<0>, SDTCisInt<1>,
+ SDTCisOpSmallerThanOp<0, 1>]>>;
+def X86trunc : SDNode<"X86ISD::TRUNC",
+ SDTypeProfile<1, 1, [SDTCisInt<0>, SDTCisInt<1>,
+ SDTCisOpSmallerThanOp<0, 1>]>>;
+
+def X86vtruncm : SDNode<"X86ISD::VTRUNCM",
+ SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisVec<1>,
+ SDTCisInt<0>, SDTCisInt<1>,
+ SDTCisVec<2>, SDTCisInt<2>,
+ SDTCisOpSmallerThanOp<0, 2>]>>;
def X86vfpext : SDNode<"X86ISD::VFPEXT",
SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisVec<1>,
- SDTCisFP<0>, SDTCisFP<1>]>>;
+ SDTCisFP<0>, SDTCisFP<1>,
+ SDTCisOpSmallerThanOp<1, 0>]>>;
def X86vfpround: SDNode<"X86ISD::VFPROUND",
SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisVec<1>,
- SDTCisFP<0>, SDTCisFP<1>]>>;
+ SDTCisFP<0>, SDTCisFP<1>,
+ SDTCisOpSmallerThanOp<0, 1>]>>;
def X86vshldq : SDNode<"X86ISD::VSHLDQ", SDTIntShiftOp>;
def X86vshrdq : SDNode<"X86ISD::VSRLDQ", SDTIntShiftOp>;
def X86pcmpeq : SDNode<"X86ISD::PCMPEQ", SDTIntBinOp, [SDNPCommutative]>;
def X86pcmpgt : SDNode<"X86ISD::PCMPGT", SDTIntBinOp>;
+def X86IntCmpMask : SDTypeProfile<1, 2,
+ [SDTCisVec<0>, SDTCisSameAs<1, 2>, SDTCisInt<1>]>;
+def X86pcmpeqm : SDNode<"X86ISD::PCMPEQM", X86IntCmpMask, [SDNPCommutative]>;
+def X86pcmpgtm : SDNode<"X86ISD::PCMPGTM", X86IntCmpMask>;
+
+def X86CmpMaskCC :
+ SDTypeProfile<1, 3, [SDTCisVec<0>, SDTCisInt<0>, SDTCisVec<1>,
+ SDTCisSameAs<1, 2>, SDTCisVT<3, i8>]>;
+def X86CmpMaskCCScalar :
+ SDTypeProfile<1, 3, [SDTCisInt<0>, SDTCisSameAs<1, 2>, SDTCisVT<3, i8>]>;
+
+def X86cmpm : SDNode<"X86ISD::CMPM", X86CmpMaskCC>;
+def X86cmpmu : SDNode<"X86ISD::CMPMU", X86CmpMaskCC>;
+def X86cmpms : SDNode<"X86ISD::FSETCC", X86CmpMaskCCScalar>;
+
def X86vshl : SDNode<"X86ISD::VSHL",
SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0,1>,
SDTCisVec<2>]>>;
def X86ptest : SDNode<"X86ISD::PTEST", SDTX86CmpPTest>;
def X86testp : SDNode<"X86ISD::TESTP", SDTX86CmpPTest>;
def X86kortest : SDNode<"X86ISD::KORTEST", SDTX86CmpPTest>;
-def X86ktest : SDNode<"X86ISD::KTEST", SDTX86CmpPTest>;
+def X86testm : SDNode<"X86ISD::TESTM", SDTypeProfile<1, 2, [SDTCisVec<0>,
+ SDTCisVec<1>,
+ SDTCisSameAs<2, 1>]>>;
+def X86testnm : SDNode<"X86ISD::TESTNM", SDTypeProfile<1, 2, [SDTCisVec<0>,
+ SDTCisVec<1>,
+ SDTCisSameAs<2, 1>]>>;
+def X86select : SDNode<"X86ISD::SELECT" , SDTSelect>;
def X86pmuludq : SDNode<"X86ISD::PMULUDQ",
SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisVec<1>,
SDTCisSameAs<1,2>]>>;
+def X86pmuldq : SDNode<"X86ISD::PMULDQ",
+ SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisVec<1>,
+ SDTCisSameAs<1,2>]>>;
// Specific shuffle nodes - At some point ISD::VECTOR_SHUFFLE will always get
// translated into one of the target nodes below during lowering.
def SDTShuff1Op : SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisSameAs<0,1>]>;
def SDTShuff2Op : SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0,1>,
SDTCisSameAs<0,2>]>;
+def SDTShuff3Op : SDTypeProfile<1, 3, [SDTCisVec<0>, SDTCisSameAs<0,1>,
+ SDTCisSameAs<0,2>, SDTCisSameAs<0,3>]>;
+def SDTShuff2OpM : SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0,1>,
+ SDTCisVec<2>]>;
def SDTShuff2OpI : SDTypeProfile<1, 2, [SDTCisVec<0>,
SDTCisSameAs<0,1>, SDTCisInt<2>]>;
def SDTShuff3OpI : SDTypeProfile<1, 3, [SDTCisVec<0>, SDTCisSameAs<0,1>,
def SDTVBroadcastm : SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisVec<1>]>;
def SDTBlend : SDTypeProfile<1, 3, [SDTCisVec<0>, SDTCisSameAs<0,1>,
- SDTCisSameAs<1,2>, SDTCisVT<3, i32>]>;
+ SDTCisSameAs<1,2>, SDTCisVT<3, i8>]>;
def SDTFma : SDTypeProfile<1, 3, [SDTCisSameAs<0,1>,
SDTCisSameAs<1,2>, SDTCisSameAs<1,3>]>;
def X86PAlignr : SDNode<"X86ISD::PALIGNR", SDTShuff3OpI>;
+def X86VAlign : SDNode<"X86ISD::VALIGN", SDTShuff3OpI>;
def X86PShufd : SDNode<"X86ISD::PSHUFD", SDTShuff2OpI>;
def X86PShufhw : SDNode<"X86ISD::PSHUFHW", SDTShuff2OpI>;
def X86Movlps : SDNode<"X86ISD::MOVLPS", SDTShuff2Op>;
def X86Movlpd : SDNode<"X86ISD::MOVLPD", SDTShuff2Op>;
+def SDTPack : SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisVec<1>, SDTCisSameAs<2, 1>]>;
+def X86Packss : SDNode<"X86ISD::PACKSS", SDTPack>;
+def X86Packus : SDNode<"X86ISD::PACKUS", SDTPack>;
+
def X86Unpckl : SDNode<"X86ISD::UNPCKL", SDTShuff2Op>;
def X86Unpckh : SDNode<"X86ISD::UNPCKH", SDTShuff2Op>;
-def X86VPermilp : SDNode<"X86ISD::VPERMILP", SDTShuff2OpI>;
-def X86VPermv : SDNode<"X86ISD::VPERMV", SDTShuff2Op>;
-def X86VPermi : SDNode<"X86ISD::VPERMI", SDTShuff2OpI>;
+def X86VPermilpv : SDNode<"X86ISD::VPERMILPV", SDTShuff2OpM>;
+def X86VPermilpi : SDNode<"X86ISD::VPERMILPI", SDTShuff2OpI>;
+def X86VPermv : SDNode<"X86ISD::VPERMV", SDTShuff2Op>;
+def X86VPermi : SDNode<"X86ISD::VPERMI", SDTShuff2OpI>;
+def X86VPermv3 : SDNode<"X86ISD::VPERMV3", SDTShuff3Op>;
+def X86VPermiv3 : SDNode<"X86ISD::VPERMIV3", SDTShuff3Op>;
def X86VPerm2x128 : SDNode<"X86ISD::VPERM2X128", SDTShuff3OpI>;
def X86VBroadcast : SDNode<"X86ISD::VBROADCAST", SDTVBroadcast>;
def X86VBroadcastm : SDNode<"X86ISD::VBROADCASTM", SDTVBroadcastm>;
+def X86Vinsert : SDNode<"X86ISD::VINSERT", SDTypeProfile<1, 3,
+ [SDTCisSameAs<0, 1>, SDTCisPtrTy<3>]>, []>;
+def X86Vextract : SDNode<"X86ISD::VEXTRACT", SDTypeProfile<1, 2,
+ [SDTCisVec<1>, SDTCisPtrTy<2>]>, []>;
def X86Blendi : SDNode<"X86ISD::BLENDI", SDTBlend>;
+
+def X86Addsub : SDNode<"X86ISD::ADDSUB", SDTFPBinOp>;
+
def X86Fmadd : SDNode<"X86ISD::FMADD", SDTFma>;
def X86Fnmadd : SDNode<"X86ISD::FNMADD", SDTFma>;
def X86Fmsub : SDNode<"X86ISD::FMSUB", SDTFma>;
def ssmem : Operand<v4f32> {
let PrintMethod = "printf32mem";
let MIOperandInfo = (ops ptr_rc, i8imm, ptr_rc_nosp, i32imm, i8imm);
- let ParserMatchClass = X86MemAsmOperand;
+ let ParserMatchClass = X86Mem32AsmOperand;
let OperandType = "OPERAND_MEMORY";
}
def sdmem : Operand<v2f64> {
let PrintMethod = "printf64mem";
let MIOperandInfo = (ops ptr_rc, i8imm, ptr_rc_nosp, i32imm, i8imm);
- let ParserMatchClass = X86MemAsmOperand;
+ let ParserMatchClass = X86Mem64AsmOperand;
let OperandType = "OPERAND_MEMORY";
}
def loadv4f64 : PatFrag<(ops node:$ptr), (v4f64 (load node:$ptr))>;
def loadv4i64 : PatFrag<(ops node:$ptr), (v4i64 (load node:$ptr))>;
-// 128-/256-bit extload pattern fragments
+// 512-bit load pattern fragments
+def loadv16f32 : PatFrag<(ops node:$ptr), (v16f32 (load node:$ptr))>;
+def loadv8f64 : PatFrag<(ops node:$ptr), (v8f64 (load node:$ptr))>;
+def loadv64i8 : PatFrag<(ops node:$ptr), (v64i8 (load node:$ptr))>;
+def loadv32i16 : PatFrag<(ops node:$ptr), (v32i16 (load node:$ptr))>;
+def loadv16i32 : PatFrag<(ops node:$ptr), (v16i32 (load node:$ptr))>;
+def loadv8i64 : PatFrag<(ops node:$ptr), (v8i64 (load node:$ptr))>;
+
+// 128-/256-/512-bit extload pattern fragments
def extloadv2f32 : PatFrag<(ops node:$ptr), (v2f64 (extloadvf32 node:$ptr))>;
def extloadv4f32 : PatFrag<(ops node:$ptr), (v4f64 (extloadvf32 node:$ptr))>;
+def extloadv8f32 : PatFrag<(ops node:$ptr), (v8f64 (extloadvf32 node:$ptr))>;
// Like 'store', but always requires 128-bit vector alignment.
def alignedstore : PatFrag<(ops node:$val, node:$ptr),
return cast<StoreSDNode>(N)->getAlignment() >= 32;
}]>;
+// Like 'store', but always requires 512-bit vector alignment.
+def alignedstore512 : PatFrag<(ops node:$val, node:$ptr),
+ (store node:$val, node:$ptr), [{
+ return cast<StoreSDNode>(N)->getAlignment() >= 64;
+}]>;
+
// Like 'load', but always requires 128-bit vector alignment.
def alignedload : PatFrag<(ops node:$ptr), (load node:$ptr), [{
return cast<LoadSDNode>(N)->getAlignment() >= 16;
return cast<LoadSDNode>(N)->getAlignment() >= 32;
}]>;
+// Like 'load', but always requires 512-bit vector alignment.
+def alignedload512 : PatFrag<(ops node:$ptr), (load node:$ptr), [{
+ return cast<LoadSDNode>(N)->getAlignment() >= 64;
+}]>;
+
def alignedloadfsf32 : PatFrag<(ops node:$ptr),
(f32 (alignedload node:$ptr))>;
def alignedloadfsf64 : PatFrag<(ops node:$ptr),
def alignedloadv4i64 : PatFrag<(ops node:$ptr),
(v4i64 (alignedload256 node:$ptr))>;
+// 512-bit aligned load pattern fragments
+def alignedloadv16f32 : PatFrag<(ops node:$ptr),
+ (v16f32 (alignedload512 node:$ptr))>;
+def alignedloadv16i32 : PatFrag<(ops node:$ptr),
+ (v16i32 (alignedload512 node:$ptr))>;
+def alignedloadv8f64 : PatFrag<(ops node:$ptr),
+ (v8f64 (alignedload512 node:$ptr))>;
+def alignedloadv8i64 : PatFrag<(ops node:$ptr),
+ (v8i64 (alignedload512 node:$ptr))>;
+
// Like 'load', but uses special alignment checks suitable for use in
// memory operands in most SSE instructions, which are required to
// be naturally aligned on some targets but not on others. If the subtarget
|| cast<LoadSDNode>(N)->getAlignment() >= 16;
}]>;
+def memop4 : PatFrag<(ops node:$ptr), (load node:$ptr), [{
+ return Subtarget->hasVectorUAMem()
+ || cast<LoadSDNode>(N)->getAlignment() >= 4;
+}]>;
+
+def memop8 : PatFrag<(ops node:$ptr), (load node:$ptr), [{
+ return Subtarget->hasVectorUAMem()
+ || cast<LoadSDNode>(N)->getAlignment() >= 8;
+}]>;
+
def memopfsf32 : PatFrag<(ops node:$ptr), (f32 (memop node:$ptr))>;
def memopfsf64 : PatFrag<(ops node:$ptr), (f64 (memop node:$ptr))>;
def memopv4f64 : PatFrag<(ops node:$ptr), (v4f64 (memop node:$ptr))>;
def memopv4i64 : PatFrag<(ops node:$ptr), (v4i64 (memop node:$ptr))>;
+// 512-bit memop pattern fragments
+def memopv16f32 : PatFrag<(ops node:$ptr), (v16f32 (memop4 node:$ptr))>;
+def memopv8f64 : PatFrag<(ops node:$ptr), (v8f64 (memop8 node:$ptr))>;
+def memopv16i32 : PatFrag<(ops node:$ptr), (v16i32 (memop4 node:$ptr))>;
+def memopv8i64 : PatFrag<(ops node:$ptr), (v8i64 (memop8 node:$ptr))>;
+
// SSSE3 uses MMX registers for some instructions. They aren't aligned on a
// 16-byte boundary.
// FIXME: 8 byte alignment for mmx reads is not required
def bc_v16i16 : PatFrag<(ops node:$in), (v16i16 (bitconvert node:$in))>;
def bc_v8i32 : PatFrag<(ops node:$in), (v8i32 (bitconvert node:$in))>;
def bc_v4i64 : PatFrag<(ops node:$in), (v4i64 (bitconvert node:$in))>;
+def bc_v8f32 : PatFrag<(ops node:$in), (v8f32 (bitconvert node:$in))>;
+
+// 512-bit bitconvert pattern fragments
+def bc_v16i32 : PatFrag<(ops node:$in), (v16i32 (bitconvert node:$in))>;
+def bc_v8i64 : PatFrag<(ops node:$in), (v8i64 (bitconvert node:$in))>;
+def bc_v8f64 : PatFrag<(ops node:$in), (v8f64 (bitconvert node:$in))>;
+def bc_v16f32 : PatFrag<(ops node:$in), (v16f32 (bitconvert node:$in))>;
def vzmovl_v2i64 : PatFrag<(ops node:$src),
(bitconvert (v2i64 (X86vzmovl
return N->isExactlyValue(+0.0);
}]>;
+def I8Imm : SDNodeXForm<imm, [{
+ // Transformation function: get the low 8 bits.
+ return getI8Imm((uint8_t)N->getZExtValue());
+}]>;
+
+def FROUND_NO_EXC : ImmLeaf<i32, [{ return Imm == 8; }]>;
+def FROUND_CURRENT : ImmLeaf<i32, [{
+ return Imm == X86::STATIC_ROUNDING::CUR_DIRECTION;
+}]>;
+
// BYTE_imm - Transform bit immediates into byte immediates.
def BYTE_imm : SDNodeXForm<imm, [{
// Transformation function: imm >> 3
projects / oota-llvm.git / blobdiff