diff options
Diffstat (limited to 'contrib/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td')
| -rw-r--r-- | contrib/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td | 225 |
1 files changed, 195 insertions, 30 deletions
diff --git a/contrib/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td b/contrib/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td index 7069bd6..dfe58ef 100644 --- a/contrib/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td +++ b/contrib/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td @@ -12,18 +12,29 @@ //===----------------------------------------------------------------------===// //===----------------------------------------------------------------------===// +// MMX specific DAG Nodes. +//===----------------------------------------------------------------------===// + +// Low word of MMX to GPR. +def MMX_X86movd2w : SDNode<"X86ISD::MMX_MOVD2W", SDTypeProfile<1, 1, + [SDTCisVT<0, i32>, SDTCisVT<1, x86mmx>]>>; +// GPR to low word of MMX. +def MMX_X86movw2d : SDNode<"X86ISD::MMX_MOVW2D", SDTypeProfile<1, 1, + [SDTCisVT<0, x86mmx>, SDTCisVT<1, i32>]>>; + +//===----------------------------------------------------------------------===// // MMX Pattern Fragments //===----------------------------------------------------------------------===// def load_mmx : PatFrag<(ops node:$ptr), (x86mmx (load node:$ptr))>; +def load_mvmmx : PatFrag<(ops node:$ptr), + (x86mmx (MMX_X86movw2d (load node:$ptr)))>; def bc_mmx : PatFrag<(ops node:$in), (x86mmx (bitconvert node:$in))>; //===----------------------------------------------------------------------===// // SSE specific DAG Nodes. //===----------------------------------------------------------------------===// -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>, SDTCisVec<1>]>; @@ -52,7 +63,6 @@ def X86fandn : SDNode<"X86ISD::FANDN", SDTFPBinOp, [SDNPCommutative, SDNPAssociative]>; def X86frsqrt : SDNode<"X86ISD::FRSQRT", SDTFPUnaryOp>; def X86frcp : SDNode<"X86ISD::FRCP", SDTFPUnaryOp>; -def X86fsrl : SDNode<"X86ISD::FSRL", SDTX86FPShiftOp>; def X86fgetsign: SDNode<"X86ISD::FGETSIGNx86",SDTFPToIntOp>; def X86fhadd : SDNode<"X86ISD::FHADD", SDTFPBinOp>; def X86fhsub : SDNode<"X86ISD::FHSUB", SDTFPBinOp>; @@ -65,6 +75,9 @@ def X86cmps : SDNode<"X86ISD::FSETCC", SDTX86Cmps>; def X86pshufb : SDNode<"X86ISD::PSHUFB", SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0,1>, SDTCisSameAs<0,2>]>>; +def X86psadbw : SDNode<"X86ISD::PSADBW", + SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0,1>, + SDTCisSameAs<0,2>]>>; def X86andnp : SDNode<"X86ISD::ANDNP", SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0,1>, SDTCisSameAs<0,2>]>>; @@ -134,14 +147,21 @@ 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>]>; + SDTypeProfile<1, 3, [SDTCisVec<0>, SDTCVecEltisVT<0, i1>, + SDTCisVec<1>, SDTCisSameAs<2, 1>, + SDTCisSameNumEltsAs<0, 1>, SDTCisVT<3, i8>]>; +def X86CmpMaskCCRound : + SDTypeProfile<1, 4, [SDTCisVec<0>,SDTCVecEltisVT<0, i1>, + SDTCisVec<1>, SDTCisSameAs<2, 1>, + SDTCisSameNumEltsAs<0, 1>, SDTCisVT<3, i8>, + SDTCisInt<4>]>; 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 X86cmpm : SDNode<"X86ISD::CMPM", X86CmpMaskCC>; +def X86cmpmRnd : SDNode<"X86ISD::CMPM_RND", X86CmpMaskCCRound>; +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>, @@ -160,16 +180,21 @@ def X86vsrai : SDNode<"X86ISD::VSRAI", SDTIntShiftOp>; def SDTX86CmpPTest : SDTypeProfile<1, 2, [SDTCisVT<0, i32>, SDTCisVec<1>, SDTCisSameAs<2, 1>]>; +def X86addus : SDNode<"X86ISD::ADDUS", SDTIntBinOp>; def X86subus : SDNode<"X86ISD::SUBUS", SDTIntBinOp>; +def X86adds : SDNode<"X86ISD::ADDS", SDTIntBinOp>; +def X86subs : SDNode<"X86ISD::SUBS", SDTIntBinOp>; def X86ptest : SDNode<"X86ISD::PTEST", SDTX86CmpPTest>; def X86testp : SDNode<"X86ISD::TESTP", SDTX86CmpPTest>; def X86kortest : SDNode<"X86ISD::KORTEST", SDTX86CmpPTest>; def X86testm : SDNode<"X86ISD::TESTM", SDTypeProfile<1, 2, [SDTCisVec<0>, - SDTCisVec<1>, - SDTCisSameAs<2, 1>]>>; + SDTCisVec<1>, SDTCisSameAs<2, 1>, + SDTCVecEltisVT<0, i1>, + SDTCisSameNumEltsAs<0, 1>]>>; def X86testnm : SDNode<"X86ISD::TESTNM", SDTypeProfile<1, 2, [SDTCisVec<0>, - SDTCisVec<1>, - SDTCisSameAs<2, 1>]>>; + SDTCisVec<1>, SDTCisSameAs<2, 1>, + SDTCVecEltisVT<0, i1>, + SDTCisSameNumEltsAs<0, 1>]>>; def X86select : SDNode<"X86ISD::SELECT" , SDTSelect>; def X86pmuludq : SDNode<"X86ISD::PMULUDQ", @@ -194,6 +219,8 @@ def SDTShuff2OpI : SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0,1>, SDTCisInt<2>]>; def SDTShuff3OpI : SDTypeProfile<1, 3, [SDTCisVec<0>, SDTCisSameAs<0,1>, SDTCisSameAs<0,2>, SDTCisInt<3>]>; +def SDTFPBinOpImmRound: SDTypeProfile<1, 4, [SDTCisVec<0>, SDTCisSameAs<0,1>, + SDTCisSameAs<0,2>, SDTCisInt<3>, SDTCisInt<4>]>; def SDTVBroadcast : SDTypeProfile<1, 1, [SDTCisVec<0>]>; def SDTVBroadcastm : SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisVec<1>]>; @@ -201,12 +228,22 @@ def SDTVBroadcastm : SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisVec<1>]>; def SDTBlend : SDTypeProfile<1, 3, [SDTCisVec<0>, SDTCisSameAs<0,1>, SDTCisSameAs<1,2>, SDTCisVT<3, i8>]>; +def SDTFPBinOpRound : SDTypeProfile<1, 3, [ // fadd_round, fmul_round, etc. + SDTCisSameAs<0, 1>, SDTCisSameAs<0, 2>, SDTCisFP<0>, SDTCisInt<3>]>; + +def SDTFPUnaryOpRound : SDTypeProfile<1, 2, [ // fsqrt_round, fgetexp_round, etc. + SDTCisSameAs<0, 1>, SDTCisFP<0>, SDTCisInt<2>]>; + def SDTFma : SDTypeProfile<1, 3, [SDTCisSameAs<0,1>, SDTCisSameAs<1,2>, SDTCisSameAs<1,3>]>; +def SDTFmaRound : SDTypeProfile<1, 4, [SDTCisSameAs<0,1>, + SDTCisSameAs<1,2>, SDTCisSameAs<1,3>, SDTCisInt<4>]>; def STDFp1SrcRm : SDTypeProfile<1, 2, [SDTCisSameAs<0,1>, SDTCisVec<0>, SDTCisInt<2>]>; def STDFp2SrcRm : SDTypeProfile<1, 3, [SDTCisSameAs<0,1>, SDTCisVec<0>, SDTCisInt<3>]>; +def STDFp3SrcRm : SDTypeProfile<1, 4, [SDTCisSameAs<0,1>, + SDTCisVec<0>, SDTCisInt<3>, SDTCisInt<4>]>; def X86PAlignr : SDNode<"X86ISD::PALIGNR", SDTShuff3OpI>; def X86VAlign : SDNode<"X86ISD::VALIGN", SDTShuff3OpI>; @@ -215,7 +252,8 @@ def X86PShufd : SDNode<"X86ISD::PSHUFD", SDTShuff2OpI>; def X86PShufhw : SDNode<"X86ISD::PSHUFHW", SDTShuff2OpI>; def X86PShuflw : SDNode<"X86ISD::PSHUFLW", SDTShuff2OpI>; -def X86Shufp : SDNode<"X86ISD::SHUFP", SDTShuff3OpI>; +def X86Shufp : SDNode<"X86ISD::SHUFP", SDTShuff3OpI>; +def X86Shuf128 : SDNode<"X86ISD::SHUF128", SDTShuff3OpI>; def X86Movddup : SDNode<"X86ISD::MOVDDUP", SDTShuff1Op>; def X86Movshdup : SDNode<"X86ISD::MOVSHDUP", SDTShuff1Op>; @@ -247,8 +285,13 @@ def X86VPermiv3 : SDNode<"X86ISD::VPERMIV3", SDTShuff3Op>; def X86VPerm2x128 : SDNode<"X86ISD::VPERM2X128", SDTShuff3OpI>; +def X86VFixupimm : SDNode<"X86ISD::VFIXUPIMM", SDTFPBinOpImmRound>; +def X86VRange : SDNode<"X86ISD::VRANGE", SDTFPBinOpImmRound>; + +def X86SubVBroadcast : SDNode<"X86ISD::SUBV_BROADCAST", + SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisVec<1>, + SDTCisSubVecOfVec<1, 0>]>, []>; 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, @@ -258,6 +301,15 @@ def X86Blendi : SDNode<"X86ISD::BLENDI", SDTBlend>; def X86Addsub : SDNode<"X86ISD::ADDSUB", SDTFPBinOp>; +def X86faddRnd : SDNode<"X86ISD::FADD_RND", SDTFPBinOpRound>; +def X86fsubRnd : SDNode<"X86ISD::FSUB_RND", SDTFPBinOpRound>; +def X86fmulRnd : SDNode<"X86ISD::FMUL_RND", SDTFPBinOpRound>; +def X86fdivRnd : SDNode<"X86ISD::FDIV_RND", SDTFPBinOpRound>; +def X86fmaxRnd : SDNode<"X86ISD::FMAX_RND", SDTFPBinOpRound>; +def X86fminRnd : SDNode<"X86ISD::FMIN_RND", SDTFPBinOpRound>; +def X86fsqrtRnd : SDNode<"X86ISD::FSQRT_RND", SDTFPUnaryOpRound>; +def X86fgetexpRnd : SDNode<"X86ISD::FGETEXP_RND", SDTFPUnaryOpRound>; + def X86Fmadd : SDNode<"X86ISD::FMADD", SDTFma>; def X86Fnmadd : SDNode<"X86ISD::FNMADD", SDTFma>; def X86Fmsub : SDNode<"X86ISD::FMSUB", SDTFma>; @@ -265,12 +317,20 @@ def X86Fnmsub : SDNode<"X86ISD::FNMSUB", SDTFma>; def X86Fmaddsub : SDNode<"X86ISD::FMADDSUB", SDTFma>; def X86Fmsubadd : SDNode<"X86ISD::FMSUBADD", SDTFma>; +def X86FmaddRnd : SDNode<"X86ISD::FMADD_RND", SDTFmaRound>; +def X86FnmaddRnd : SDNode<"X86ISD::FNMADD_RND", SDTFmaRound>; +def X86FmsubRnd : SDNode<"X86ISD::FMSUB_RND", SDTFmaRound>; +def X86FnmsubRnd : SDNode<"X86ISD::FNMSUB_RND", SDTFmaRound>; +def X86FmaddsubRnd : SDNode<"X86ISD::FMADDSUB_RND", SDTFmaRound>; +def X86FmsubaddRnd : SDNode<"X86ISD::FMSUBADD_RND", SDTFmaRound>; + def X86rsqrt28 : SDNode<"X86ISD::RSQRT28", STDFp1SrcRm>; def X86rcp28 : SDNode<"X86ISD::RCP28", STDFp1SrcRm>; def X86exp2 : SDNode<"X86ISD::EXP2", STDFp1SrcRm>; def X86rsqrt28s : SDNode<"X86ISD::RSQRT28", STDFp2SrcRm>; def X86rcp28s : SDNode<"X86ISD::RCP28", STDFp2SrcRm>; +def X86RndScale : SDNode<"X86ISD::RNDSCALE", STDFp3SrcRm>; def SDT_PCMPISTRI : SDTypeProfile<2, 3, [SDTCisVT<0, i32>, SDTCisVT<1, i32>, SDTCisVT<2, v16i8>, SDTCisVT<3, v16i8>, @@ -346,6 +406,15 @@ 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))>; +// These are needed to match a scalar load that is used in a vector-only +// math instruction such as the FP logical ops: andps, andnps, orps, xorps. +// The memory operand is required to be a 128-bit load, so it must be converted +// from a vector to a scalar. +def loadf32_128 : PatFrag<(ops node:$ptr), + (f32 (vector_extract (loadv4f32 node:$ptr), (iPTR 0)))>; +def loadf64_128 : PatFrag<(ops node:$ptr), + (f64 (vector_extract (loadv2f64 node:$ptr), (iPTR 0)))>; + // Like 'store', but always requires 128-bit vector alignment. def alignedstore : PatFrag<(ops node:$val, node:$ptr), (store node:$val, node:$ptr), [{ @@ -437,17 +506,15 @@ def memopv4f32 : PatFrag<(ops node:$ptr), (v4f32 (memop node:$ptr))>; def memopv2f64 : PatFrag<(ops node:$ptr), (v2f64 (memop node:$ptr))>; def memopv2i64 : PatFrag<(ops node:$ptr), (v2i64 (memop node:$ptr))>; -// 256-bit memop pattern fragments -// NOTE: all 256-bit integer vector loads are promoted to v4i64 -def memopv8f32 : PatFrag<(ops node:$ptr), (v8f32 (memop node:$ptr))>; -def memopv4f64 : PatFrag<(ops node:$ptr), (v4f64 (memop node:$ptr))>; -def memopv4i64 : PatFrag<(ops node:$ptr), (v4i64 (memop node:$ptr))>; +// These are needed to match a scalar memop that is used in a vector-only +// math instruction such as the FP logical ops: andps, andnps, orps, xorps. +// The memory operand is required to be a 128-bit load, so it must be converted +// from a vector to a scalar. +def memopfsf32_128 : PatFrag<(ops node:$ptr), + (f32 (vector_extract (memopv4f32 node:$ptr), (iPTR 0)))>; +def memopfsf64_128 : PatFrag<(ops node:$ptr), + (f64 (vector_extract (memopv2f64 node:$ptr), (iPTR 0)))>; -// 512-bit memop pattern fragments -def memopv16f32 : PatFrag<(ops node:$ptr), (v16f32 (memop node:$ptr))>; -def memopv8f64 : PatFrag<(ops node:$ptr), (v8f64 (memop node:$ptr))>; -def memopv16i32 : PatFrag<(ops node:$ptr), (v16i32 (memop node:$ptr))>; -def memopv8i64 : PatFrag<(ops node:$ptr), (v8i64 (memop node:$ptr))>; // SSSE3 uses MMX registers for some instructions. They aren't aligned on a // 16-byte boundary. @@ -484,6 +551,52 @@ def unalignednontemporalstore : PatFrag<(ops node:$val, node:$ptr), return false; }]>; +def mgatherv8i32 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_gather node:$src1, node:$src2, node:$src3) , [{ + if (MaskedGatherSDNode *Mgt = dyn_cast<MaskedGatherSDNode>(N)) + return (Mgt->getIndex().getValueType() == MVT::v8i32 || + Mgt->getBasePtr().getValueType() == MVT::v8i32); + return false; +}]>; + +def mgatherv8i64 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_gather node:$src1, node:$src2, node:$src3) , [{ + if (MaskedGatherSDNode *Mgt = dyn_cast<MaskedGatherSDNode>(N)) + return (Mgt->getIndex().getValueType() == MVT::v8i64 || + Mgt->getBasePtr().getValueType() == MVT::v8i64); + return false; +}]>; +def mgatherv16i32 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_gather node:$src1, node:$src2, node:$src3) , [{ + if (MaskedGatherSDNode *Mgt = dyn_cast<MaskedGatherSDNode>(N)) + return (Mgt->getIndex().getValueType() == MVT::v16i32 || + Mgt->getBasePtr().getValueType() == MVT::v16i32); + return false; +}]>; + +def mscatterv8i32 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_scatter node:$src1, node:$src2, node:$src3) , [{ + if (MaskedScatterSDNode *Sc = dyn_cast<MaskedScatterSDNode>(N)) + return (Sc->getIndex().getValueType() == MVT::v8i32 || + Sc->getBasePtr().getValueType() == MVT::v8i32); + return false; +}]>; + +def mscatterv8i64 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_scatter node:$src1, node:$src2, node:$src3) , [{ + if (MaskedScatterSDNode *Sc = dyn_cast<MaskedScatterSDNode>(N)) + return (Sc->getIndex().getValueType() == MVT::v8i64 || + Sc->getBasePtr().getValueType() == MVT::v8i64); + return false; +}]>; +def mscatterv16i32 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_scatter node:$src1, node:$src2, node:$src3) , [{ + if (MaskedScatterSDNode *Sc = dyn_cast<MaskedScatterSDNode>(N)) + return (Sc->getIndex().getValueType() == MVT::v16i32 || + Sc->getBasePtr().getValueType() == MVT::v16i32); + return false; +}]>; + // 128-bit bitconvert pattern fragments def bc_v4f32 : PatFrag<(ops node:$in), (v4f32 (bitconvert node:$in))>; def bc_v2f64 : PatFrag<(ops node:$in), (v2f64 (bitconvert node:$in))>; @@ -522,7 +635,7 @@ def fp32imm0 : PatLeaf<(f32 fpimm), [{ def I8Imm : SDNodeXForm<imm, [{ // Transformation function: get the low 8 bits. - return getI8Imm((uint8_t)N->getZExtValue()); + return getI8Imm((uint8_t)N->getZExtValue(), SDLoc(N)); }]>; def FROUND_NO_EXC : ImmLeaf<i32, [{ return Imm == 8; }]>; @@ -533,31 +646,31 @@ def FROUND_CURRENT : ImmLeaf<i32, [{ // BYTE_imm - Transform bit immediates into byte immediates. def BYTE_imm : SDNodeXForm<imm, [{ // Transformation function: imm >> 3 - return getI32Imm(N->getZExtValue() >> 3); + return getI32Imm(N->getZExtValue() >> 3, SDLoc(N)); }]>; // EXTRACT_get_vextract128_imm xform function: convert extract_subvector index // to VEXTRACTF128/VEXTRACTI128 imm. def EXTRACT_get_vextract128_imm : SDNodeXForm<extract_subvector, [{ - return getI8Imm(X86::getExtractVEXTRACT128Immediate(N)); + return getI8Imm(X86::getExtractVEXTRACT128Immediate(N), SDLoc(N)); }]>; // INSERT_get_vinsert128_imm xform function: convert insert_subvector index to // VINSERTF128/VINSERTI128 imm. def INSERT_get_vinsert128_imm : SDNodeXForm<insert_subvector, [{ - return getI8Imm(X86::getInsertVINSERT128Immediate(N)); + return getI8Imm(X86::getInsertVINSERT128Immediate(N), SDLoc(N)); }]>; // EXTRACT_get_vextract256_imm xform function: convert extract_subvector index // to VEXTRACTF64x4 imm. def EXTRACT_get_vextract256_imm : SDNodeXForm<extract_subvector, [{ - return getI8Imm(X86::getExtractVEXTRACT256Immediate(N)); + return getI8Imm(X86::getExtractVEXTRACT256Immediate(N), SDLoc(N)); }]>; // INSERT_get_vinsert256_imm xform function: convert insert_subvector index to // VINSERTF64x4 imm. def INSERT_get_vinsert256_imm : SDNodeXForm<insert_subvector, [{ - return getI8Imm(X86::getInsertVINSERT256Immediate(N)); + return getI8Imm(X86::getInsertVINSERT256Immediate(N), SDLoc(N)); }]>; def vextract128_extract : PatFrag<(ops node:$bigvec, node:$index), @@ -587,3 +700,55 @@ def vinsert256_insert : PatFrag<(ops node:$bigvec, node:$smallvec, return X86::isVINSERT256Index(N); }], INSERT_get_vinsert256_imm>; +def masked_load_aligned128 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_load node:$src1, node:$src2, node:$src3), [{ + if (auto *Load = dyn_cast<MaskedLoadSDNode>(N)) + return Load->getAlignment() >= 16; + return false; +}]>; + +def masked_load_aligned256 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_load node:$src1, node:$src2, node:$src3), [{ + if (auto *Load = dyn_cast<MaskedLoadSDNode>(N)) + return Load->getAlignment() >= 32; + return false; +}]>; + +def masked_load_aligned512 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_load node:$src1, node:$src2, node:$src3), [{ + if (auto *Load = dyn_cast<MaskedLoadSDNode>(N)) + return Load->getAlignment() >= 64; + return false; +}]>; + +def masked_load_unaligned : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_load node:$src1, node:$src2, node:$src3), [{ + return isa<MaskedLoadSDNode>(N); +}]>; + +def masked_store_aligned128 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_store node:$src1, node:$src2, node:$src3), [{ + if (auto *Store = dyn_cast<MaskedStoreSDNode>(N)) + return Store->getAlignment() >= 16; + return false; +}]>; + +def masked_store_aligned256 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_store node:$src1, node:$src2, node:$src3), [{ + if (auto *Store = dyn_cast<MaskedStoreSDNode>(N)) + return Store->getAlignment() >= 32; + return false; +}]>; + +def masked_store_aligned512 : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_store node:$src1, node:$src2, node:$src3), [{ + if (auto *Store = dyn_cast<MaskedStoreSDNode>(N)) + return Store->getAlignment() >= 64; + return false; +}]>; + +def masked_store_unaligned : PatFrag<(ops node:$src1, node:$src2, node:$src3), + (masked_store node:$src1, node:$src2, node:$src3), [{ + return isa<MaskedStoreSDNode>(N); +}]>; + |
