X-Git-Url: http://demsky.eecs.uci.edu/git/?a=blobdiff_plain;f=lib%2FTarget%2FX86%2FX86InstrFragmentsSIMD.td;h=70c3d076d501d8932f2c85ffb1e405a1c4209132;hb=beb6898df8f96ccea4ae147587479b507bb3e491;hp=f5d28eda6d7de3149982e7fea7bf49d5a2286928;hpb=045573ce21282ee7d1c58e57d00a77ede8c732da;p=oota-llvm.git diff --git a/lib/Target/X86/X86InstrFragmentsSIMD.td b/lib/Target/X86/X86InstrFragmentsSIMD.td index f5d28eda6d7..70c3d076d50 100644 --- a/lib/Target/X86/X86InstrFragmentsSIMD.td +++ b/lib/Target/X86/X86InstrFragmentsSIMD.td @@ -15,51 +15,8 @@ // MMX Pattern Fragments //===----------------------------------------------------------------------===// -def load_mmx : PatFrag<(ops node:$ptr), (v1i64 (load node:$ptr))>; - -def bc_v8i8 : PatFrag<(ops node:$in), (v8i8 (bitconvert node:$in))>; -def bc_v4i16 : PatFrag<(ops node:$in), (v4i16 (bitconvert node:$in))>; -def bc_v2i32 : PatFrag<(ops node:$in), (v2i32 (bitconvert node:$in))>; -def bc_v1i64 : PatFrag<(ops node:$in), (v1i64 (bitconvert node:$in))>; - -//===----------------------------------------------------------------------===// -// MMX Masks -//===----------------------------------------------------------------------===// - -// MMX_SHUFFLE_get_shuf_imm xform function: convert vector_shuffle mask to -// PSHUFW imm. -def MMX_SHUFFLE_get_shuf_imm : SDNodeXForm; - -// Patterns for: vector_shuffle v1, v2, <2, 6, 3, 7, ...> -def mmx_unpckh : PatFrag<(ops node:$lhs, node:$rhs), - (vector_shuffle node:$lhs, node:$rhs), [{ - return X86::isUNPCKHMask(cast(N)); -}]>; - -// Patterns for: vector_shuffle v1, v2, <0, 4, 2, 5, ...> -def mmx_unpckl : PatFrag<(ops node:$lhs, node:$rhs), - (vector_shuffle node:$lhs, node:$rhs), [{ - return X86::isUNPCKLMask(cast(N)); -}]>; - -// Patterns for: vector_shuffle v1, , <0, 0, 1, 1, ...> -def mmx_unpckh_undef : PatFrag<(ops node:$lhs, node:$rhs), - (vector_shuffle node:$lhs, node:$rhs), [{ - return X86::isUNPCKH_v_undef_Mask(cast(N)); -}]>; - -// Patterns for: vector_shuffle v1, , <2, 2, 3, 3, ...> -def mmx_unpckl_undef : PatFrag<(ops node:$lhs, node:$rhs), - (vector_shuffle node:$lhs, node:$rhs), [{ - return X86::isUNPCKL_v_undef_Mask(cast(N)); -}]>; - -def mmx_pshufw : PatFrag<(ops node:$lhs, node:$rhs), - (vector_shuffle node:$lhs, node:$rhs), [{ - return X86::isPSHUFDMask(cast(N)); -}], MMX_SHUFFLE_get_shuf_imm>; +def load_mmx : PatFrag<(ops node:$ptr), (x86mmx (load node:$ptr))>; +def bc_mmx : PatFrag<(ops node:$in), (x86mmx (bitconvert node:$in))>; //===----------------------------------------------------------------------===// // SSE specific DAG Nodes. @@ -102,7 +59,7 @@ def X86insrtps : SDNode<"X86ISD::INSERTPS", def X86vzmovl : SDNode<"X86ISD::VZEXT_MOVL", SDTypeProfile<1, 1, [SDTCisSameAs<0,1>]>>; def X86vzload : SDNode<"X86ISD::VZEXT_LOAD", SDTLoad, - [SDNPHasChain, SDNPMayLoad]>; + [SDNPHasChain, SDNPMayLoad, SDNPMemOperand]>; def X86vshl : SDNode<"X86ISD::VSHL", SDTIntShiftOp>; def X86vshr : SDNode<"X86ISD::VSRL", SDTIntShiftOp>; def X86cmpps : SDNode<"X86ISD::CMPPS", SDTX86VFCMP>; @@ -122,6 +79,57 @@ def SDTX86CmpPTest : SDTypeProfile<1, 2, [SDTCisVT<0, i32>, def X86ptest : SDNode<"X86ISD::PTEST", SDTX86CmpPTest>; def X86testp : SDNode<"X86ISD::TESTP", SDTX86CmpPTest>; +// Specific shuffle nodes - At some point ISD::VECTOR_SHUFFLE will always get +// translated into one of the target nodes below during lowering. +// Note: this is a work in progress... +def SDTShuff1Op : SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisSameAs<0,1>]>; +def SDTShuff2Op : SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0,1>, + SDTCisSameAs<0,2>]>; + +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 X86PAlign : SDNode<"X86ISD::PALIGN", SDTShuff3OpI>; + +def X86PShufd : SDNode<"X86ISD::PSHUFD", SDTShuff2OpI>; +def X86PShufhw : SDNode<"X86ISD::PSHUFHW", SDTShuff2OpI>; +def X86PShuflw : SDNode<"X86ISD::PSHUFLW", SDTShuff2OpI>; + +def X86Shufpd : SDNode<"X86ISD::SHUFPD", SDTShuff3OpI>; +def X86Shufps : SDNode<"X86ISD::SHUFPS", SDTShuff3OpI>; + +def X86Movddup : SDNode<"X86ISD::MOVDDUP", SDTShuff1Op>; +def X86Movshdup : SDNode<"X86ISD::MOVSHDUP", SDTShuff1Op>; +def X86Movsldup : SDNode<"X86ISD::MOVSLDUP", SDTShuff1Op>; + +def X86Movsd : SDNode<"X86ISD::MOVSD", SDTShuff2Op>; +def X86Movss : SDNode<"X86ISD::MOVSS", SDTShuff2Op>; + +def X86Movlhps : SDNode<"X86ISD::MOVLHPS", SDTShuff2Op>; +def X86Movlhpd : SDNode<"X86ISD::MOVLHPD", SDTShuff2Op>; +def X86Movhlps : SDNode<"X86ISD::MOVHLPS", SDTShuff2Op>; +def X86Movhlpd : SDNode<"X86ISD::MOVHLPD", SDTShuff2Op>; + +def X86Movlps : SDNode<"X86ISD::MOVLPS", SDTShuff2Op>; +def X86Movlpd : SDNode<"X86ISD::MOVLPD", SDTShuff2Op>; + +def X86Unpcklps : SDNode<"X86ISD::UNPCKLPS", SDTShuff2Op>; +def X86Unpcklpd : SDNode<"X86ISD::UNPCKLPD", SDTShuff2Op>; +def X86Unpckhps : SDNode<"X86ISD::UNPCKHPS", SDTShuff2Op>; +def X86Unpckhpd : SDNode<"X86ISD::UNPCKHPD", SDTShuff2Op>; + +def X86Punpcklbw : SDNode<"X86ISD::PUNPCKLBW", SDTShuff2Op>; +def X86Punpcklwd : SDNode<"X86ISD::PUNPCKLWD", SDTShuff2Op>; +def X86Punpckldq : SDNode<"X86ISD::PUNPCKLDQ", SDTShuff2Op>; +def X86Punpcklqdq : SDNode<"X86ISD::PUNPCKLQDQ", SDTShuff2Op>; + +def X86Punpckhbw : SDNode<"X86ISD::PUNPCKHBW", SDTShuff2Op>; +def X86Punpckhwd : SDNode<"X86ISD::PUNPCKHWD", SDTShuff2Op>; +def X86Punpckhdq : SDNode<"X86ISD::PUNPCKHDQ", SDTShuff2Op>; +def X86Punpckhqdq : SDNode<"X86ISD::PUNPCKHQDQ", SDTShuff2Op>; + //===----------------------------------------------------------------------===// // SSE Complex Patterns //===----------------------------------------------------------------------===// @@ -130,9 +138,11 @@ def X86testp : SDNode<"X86ISD::TESTP", SDTX86CmpPTest>; // the top elements. These are used for the SSE 'ss' and 'sd' instruction // forms. def sse_load_f32 : ComplexPattern; + [SDNPHasChain, SDNPMayLoad, SDNPMemOperand, + SDNPWantRoot]>; def sse_load_f64 : ComplexPattern; + [SDNPHasChain, SDNPMayLoad, SDNPMemOperand, + SDNPWantRoot]>; def ssmem : Operand { let PrintMethod = "printf32mem"; @@ -149,12 +159,13 @@ def sdmem : Operand { // SSE pattern fragments //===----------------------------------------------------------------------===// +// 128-bit load pattern fragments def loadv4f32 : PatFrag<(ops node:$ptr), (v4f32 (load node:$ptr))>; def loadv2f64 : PatFrag<(ops node:$ptr), (v2f64 (load node:$ptr))>; def loadv4i32 : PatFrag<(ops node:$ptr), (v4i32 (load node:$ptr))>; def loadv2i64 : PatFrag<(ops node:$ptr), (v2i64 (load node:$ptr))>; -// FIXME: move this to a more appropriate place after all AVX is done. +// 256-bit load pattern fragments def loadv8f32 : PatFrag<(ops node:$ptr), (v8f32 (load node:$ptr))>; def loadv4f64 : PatFrag<(ops node:$ptr), (v4f64 (load node:$ptr))>; def loadv8i32 : PatFrag<(ops node:$ptr), (v8i32 (load node:$ptr))>; @@ -175,6 +186,8 @@ def alignedloadfsf32 : PatFrag<(ops node:$ptr), (f32 (alignedload node:$ptr))>; def alignedloadfsf64 : PatFrag<(ops node:$ptr), (f64 (alignedload node:$ptr))>; + +// 128-bit aligned load pattern fragments def alignedloadv4f32 : PatFrag<(ops node:$ptr), (v4f32 (alignedload node:$ptr))>; def alignedloadv2f64 : PatFrag<(ops node:$ptr), @@ -184,7 +197,7 @@ def alignedloadv4i32 : PatFrag<(ops node:$ptr), def alignedloadv2i64 : PatFrag<(ops node:$ptr), (v2i64 (alignedload node:$ptr))>; -// FIXME: move this to a more appropriate place after all AVX is done. +// 256-bit aligned load pattern fragments def alignedloadv8f32 : PatFrag<(ops node:$ptr), (v8f32 (alignedload node:$ptr))>; def alignedloadv4f64 : PatFrag<(ops node:$ptr), @@ -207,13 +220,16 @@ def memop : PatFrag<(ops node:$ptr), (load node:$ptr), [{ def memopfsf32 : PatFrag<(ops node:$ptr), (f32 (memop node:$ptr))>; def memopfsf64 : PatFrag<(ops node:$ptr), (f64 (memop node:$ptr))>; + +// 128-bit memop pattern fragments def memopv4f32 : PatFrag<(ops node:$ptr), (v4f32 (memop node:$ptr))>; def memopv2f64 : PatFrag<(ops node:$ptr), (v2f64 (memop node:$ptr))>; def memopv4i32 : PatFrag<(ops node:$ptr), (v4i32 (memop node:$ptr))>; def memopv2i64 : PatFrag<(ops node:$ptr), (v2i64 (memop node:$ptr))>; +def memopv8i16 : PatFrag<(ops node:$ptr), (v8i16 (memop node:$ptr))>; def memopv16i8 : PatFrag<(ops node:$ptr), (v16i8 (memop node:$ptr))>; -// FIXME: move this to a more appropriate place after all AVX is done. +// 256-bit memop pattern fragments def memopv32i8 : PatFrag<(ops node:$ptr), (v32i8 (memop node:$ptr))>; def memopv8f32 : PatFrag<(ops node:$ptr), (v8f32 (memop node:$ptr))>; def memopv4f64 : PatFrag<(ops node:$ptr), (v4f64 (memop node:$ptr))>; @@ -227,10 +243,7 @@ def memop64 : PatFrag<(ops node:$ptr), (load node:$ptr), [{ return cast(N)->getAlignment() >= 8; }]>; -def memopv8i8 : PatFrag<(ops node:$ptr), (v8i8 (memop64 node:$ptr))>; -def memopv4i16 : PatFrag<(ops node:$ptr), (v4i16 (memop64 node:$ptr))>; -def memopv8i16 : PatFrag<(ops node:$ptr), (v8i16 (memop64 node:$ptr))>; -def memopv2i32 : PatFrag<(ops node:$ptr), (v2i32 (memop64 node:$ptr))>; +def memopmmx : PatFrag<(ops node:$ptr), (x86mmx (memop64 node:$ptr))>; // MOVNT Support // Like 'store', but requires the non-temporal bit to be set @@ -258,6 +271,7 @@ def unalignednontemporalstore : PatFrag<(ops node:$val, node:$ptr), return false; }]>; +// 128-bit bitconvert pattern fragments def bc_v4f32 : PatFrag<(ops node:$in), (v4f32 (bitconvert node:$in))>; def bc_v2f64 : PatFrag<(ops node:$in), (v2f64 (bitconvert node:$in))>; def bc_v16i8 : PatFrag<(ops node:$in), (v16i8 (bitconvert node:$in))>; @@ -265,7 +279,7 @@ def bc_v8i16 : PatFrag<(ops node:$in), (v8i16 (bitconvert node:$in))>; def bc_v4i32 : PatFrag<(ops node:$in), (v4i32 (bitconvert node:$in))>; def bc_v2i64 : PatFrag<(ops node:$in), (v2i64 (bitconvert node:$in))>; -// FIXME: move this to a more appropriate place after all AVX is done. +// 256-bit bitconvert pattern fragments def bc_v8i32 : PatFrag<(ops node:$in), (v8i32 (bitconvert node:$in))>; def vzmovl_v2i64 : PatFrag<(ops node:$src),