1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-23 19:23:23 +01:00
llvm-mirror/lib/Target/X86/X86InstrXOP.td
Craig Topper 5399eeead4 [X86] Stop promoting vector and/or/xor/andn to vXi64.
These promotions add additional bitcasts to the SelectionDAG that can pessimize computeKnownBits/computeNumSignBits. It also seems to interfere with broadcast formation.

This patch removes the promotion and adds isel patterns instead.

The increased table size is more than I would like, but hopefully we can find some canonicalizations or other tricks to start pruning out patterns going forward.

Differential Revision: https://reviews.llvm.org/D53268

llvm-svn: 345408
2018-10-26 17:21:26 +00:00

489 lines
25 KiB
TableGen

//===-- X86InstrXOP.td - XOP Instruction Set ---------------*- tablegen -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file describes XOP (eXtended OPerations)
//
//===----------------------------------------------------------------------===//
multiclass xop2op<bits<8> opc, string OpcodeStr, Intrinsic Int> {
def rr : IXOP<opc, MRMSrcReg, (outs VR128:$dst), (ins VR128:$src),
!strconcat(OpcodeStr, "\t{$src, $dst|$dst, $src}"),
[(set VR128:$dst, (Int VR128:$src))]>, XOP, Sched<[SchedWritePHAdd.XMM]>;
def rm : IXOP<opc, MRMSrcMem, (outs VR128:$dst), (ins i128mem:$src),
!strconcat(OpcodeStr, "\t{$src, $dst|$dst, $src}"),
[(set VR128:$dst, (Int (load addr:$src)))]>, XOP,
Sched<[SchedWritePHAdd.XMM.Folded, SchedWritePHAdd.XMM.ReadAfterFold]>;
}
let ExeDomain = SSEPackedInt in {
defm VPHSUBWD : xop2op<0xE2, "vphsubwd", int_x86_xop_vphsubwd>;
defm VPHSUBDQ : xop2op<0xE3, "vphsubdq", int_x86_xop_vphsubdq>;
defm VPHSUBBW : xop2op<0xE1, "vphsubbw", int_x86_xop_vphsubbw>;
defm VPHADDWQ : xop2op<0xC7, "vphaddwq", int_x86_xop_vphaddwq>;
defm VPHADDWD : xop2op<0xC6, "vphaddwd", int_x86_xop_vphaddwd>;
defm VPHADDUWQ : xop2op<0xD7, "vphadduwq", int_x86_xop_vphadduwq>;
defm VPHADDUWD : xop2op<0xD6, "vphadduwd", int_x86_xop_vphadduwd>;
defm VPHADDUDQ : xop2op<0xDB, "vphaddudq", int_x86_xop_vphaddudq>;
defm VPHADDUBW : xop2op<0xD1, "vphaddubw", int_x86_xop_vphaddubw>;
defm VPHADDUBQ : xop2op<0xD3, "vphaddubq", int_x86_xop_vphaddubq>;
defm VPHADDUBD : xop2op<0xD2, "vphaddubd", int_x86_xop_vphaddubd>;
defm VPHADDDQ : xop2op<0xCB, "vphadddq", int_x86_xop_vphadddq>;
defm VPHADDBW : xop2op<0xC1, "vphaddbw", int_x86_xop_vphaddbw>;
defm VPHADDBQ : xop2op<0xC3, "vphaddbq", int_x86_xop_vphaddbq>;
defm VPHADDBD : xop2op<0xC2, "vphaddbd", int_x86_xop_vphaddbd>;
}
// Scalar load 2 addr operand instructions
multiclass xop2opsld<bits<8> opc, string OpcodeStr, Intrinsic Int,
Operand memop, ComplexPattern mem_cpat,
X86FoldableSchedWrite sched> {
def rr : IXOP<opc, MRMSrcReg, (outs VR128:$dst), (ins VR128:$src),
!strconcat(OpcodeStr, "\t{$src, $dst|$dst, $src}"),
[(set VR128:$dst, (Int VR128:$src))]>, XOP, Sched<[sched]>;
def rm : IXOP<opc, MRMSrcMem, (outs VR128:$dst), (ins memop:$src),
!strconcat(OpcodeStr, "\t{$src, $dst|$dst, $src}"),
[(set VR128:$dst, (Int mem_cpat:$src))]>, XOP,
Sched<[sched.Folded, sched.ReadAfterFold]>;
}
multiclass xop2op128<bits<8> opc, string OpcodeStr, Intrinsic Int,
X86FoldableSchedWrite sched> {
def rr : IXOP<opc, MRMSrcReg, (outs VR128:$dst), (ins VR128:$src),
!strconcat(OpcodeStr, "\t{$src, $dst|$dst, $src}"),
[(set VR128:$dst, (Int VR128:$src))]>, XOP, Sched<[sched]>;
def rm : IXOP<opc, MRMSrcMem, (outs VR128:$dst), (ins f128mem:$src),
!strconcat(OpcodeStr, "\t{$src, $dst|$dst, $src}"),
[(set VR128:$dst, (Int (load addr:$src)))]>, XOP,
Sched<[sched.Folded, sched.ReadAfterFold]>;
}
multiclass xop2op256<bits<8> opc, string OpcodeStr, Intrinsic Int,
X86FoldableSchedWrite sched> {
def Yrr : IXOP<opc, MRMSrcReg, (outs VR256:$dst), (ins VR256:$src),
!strconcat(OpcodeStr, "\t{$src, $dst|$dst, $src}"),
[(set VR256:$dst, (Int VR256:$src))]>, XOP, VEX_L, Sched<[sched]>;
def Yrm : IXOP<opc, MRMSrcMem, (outs VR256:$dst), (ins f256mem:$src),
!strconcat(OpcodeStr, "\t{$src, $dst|$dst, $src}"),
[(set VR256:$dst, (Int (load addr:$src)))]>, XOP, VEX_L,
Sched<[sched.Folded, sched.ReadAfterFold]>;
}
let ExeDomain = SSEPackedSingle in {
defm VFRCZSS : xop2opsld<0x82, "vfrczss", int_x86_xop_vfrcz_ss,
ssmem, sse_load_f32, SchedWriteFRnd.Scl>;
defm VFRCZPS : xop2op128<0x80, "vfrczps", int_x86_xop_vfrcz_ps,
SchedWriteFRnd.XMM>;
defm VFRCZPS : xop2op256<0x80, "vfrczps", int_x86_xop_vfrcz_ps_256,
SchedWriteFRnd.YMM>;
}
let ExeDomain = SSEPackedDouble in {
defm VFRCZSD : xop2opsld<0x83, "vfrczsd", int_x86_xop_vfrcz_sd,
sdmem, sse_load_f64, SchedWriteFRnd.Scl>;
defm VFRCZPD : xop2op128<0x81, "vfrczpd", int_x86_xop_vfrcz_pd,
SchedWriteFRnd.XMM>;
defm VFRCZPD : xop2op256<0x81, "vfrczpd", int_x86_xop_vfrcz_pd_256,
SchedWriteFRnd.YMM>;
}
multiclass xop3op<bits<8> opc, string OpcodeStr, SDNode OpNode,
ValueType vt128, X86FoldableSchedWrite sched> {
def rr : IXOP<opc, MRMSrcReg4VOp3, (outs VR128:$dst),
(ins VR128:$src1, VR128:$src2),
!strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
[(set VR128:$dst,
(vt128 (OpNode (vt128 VR128:$src1), (vt128 VR128:$src2))))]>,
XOP, Sched<[sched]>;
def rm : IXOP<opc, MRMSrcMem, (outs VR128:$dst),
(ins VR128:$src1, i128mem:$src2),
!strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
[(set VR128:$dst,
(vt128 (OpNode (vt128 VR128:$src1),
(vt128 (load addr:$src2)))))]>,
XOP_4V, VEX_W, Sched<[sched.Folded, sched.ReadAfterFold]>;
def mr : IXOP<opc, MRMSrcMem4VOp3, (outs VR128:$dst),
(ins i128mem:$src1, VR128:$src2),
!strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
[(set VR128:$dst,
(vt128 (OpNode (vt128 (load addr:$src1)),
(vt128 VR128:$src2))))]>,
XOP, Sched<[sched.Folded, sched.ReadAfterFold]>;
// For disassembler
let isCodeGenOnly = 1, ForceDisassemble = 1, hasSideEffects = 0 in
def rr_REV : IXOP<opc, MRMSrcReg, (outs VR128:$dst),
(ins VR128:$src1, VR128:$src2),
!strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
[]>,
XOP_4V, VEX_W, Sched<[sched]>, FoldGenData<NAME#rr>;
}
let ExeDomain = SSEPackedInt in {
defm VPROTB : xop3op<0x90, "vprotb", rotl, v16i8, SchedWriteVarVecShift.XMM>;
defm VPROTD : xop3op<0x92, "vprotd", rotl, v4i32, SchedWriteVarVecShift.XMM>;
defm VPROTQ : xop3op<0x93, "vprotq", rotl, v2i64, SchedWriteVarVecShift.XMM>;
defm VPROTW : xop3op<0x91, "vprotw", rotl, v8i16, SchedWriteVarVecShift.XMM>;
defm VPSHAB : xop3op<0x98, "vpshab", X86vpsha, v16i8, SchedWriteVarVecShift.XMM>;
defm VPSHAD : xop3op<0x9A, "vpshad", X86vpsha, v4i32, SchedWriteVarVecShift.XMM>;
defm VPSHAQ : xop3op<0x9B, "vpshaq", X86vpsha, v2i64, SchedWriteVarVecShift.XMM>;
defm VPSHAW : xop3op<0x99, "vpshaw", X86vpsha, v8i16, SchedWriteVarVecShift.XMM>;
defm VPSHLB : xop3op<0x94, "vpshlb", X86vpshl, v16i8, SchedWriteVarVecShift.XMM>;
defm VPSHLD : xop3op<0x96, "vpshld", X86vpshl, v4i32, SchedWriteVarVecShift.XMM>;
defm VPSHLQ : xop3op<0x97, "vpshlq", X86vpshl, v2i64, SchedWriteVarVecShift.XMM>;
defm VPSHLW : xop3op<0x95, "vpshlw", X86vpshl, v8i16, SchedWriteVarVecShift.XMM>;
}
multiclass xop3opimm<bits<8> opc, string OpcodeStr, SDNode OpNode,
ValueType vt128, X86FoldableSchedWrite sched> {
def ri : IXOPi8<opc, MRMSrcReg, (outs VR128:$dst),
(ins VR128:$src1, u8imm:$src2),
!strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
[(set VR128:$dst,
(vt128 (OpNode (vt128 VR128:$src1), imm:$src2)))]>,
XOP, Sched<[sched]>;
def mi : IXOPi8<opc, MRMSrcMem, (outs VR128:$dst),
(ins i128mem:$src1, u8imm:$src2),
!strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
[(set VR128:$dst,
(vt128 (OpNode (vt128 (load addr:$src1)), imm:$src2)))]>,
XOP, Sched<[sched.Folded, sched.ReadAfterFold]>;
}
let ExeDomain = SSEPackedInt in {
defm VPROTB : xop3opimm<0xC0, "vprotb", X86vrotli, v16i8,
SchedWriteVecShiftImm.XMM>;
defm VPROTD : xop3opimm<0xC2, "vprotd", X86vrotli, v4i32,
SchedWriteVecShiftImm.XMM>;
defm VPROTQ : xop3opimm<0xC3, "vprotq", X86vrotli, v2i64,
SchedWriteVecShiftImm.XMM>;
defm VPROTW : xop3opimm<0xC1, "vprotw", X86vrotli, v8i16,
SchedWriteVecShiftImm.XMM>;
}
// Instruction where second source can be memory, but third must be register
multiclass xop4opm2<bits<8> opc, string OpcodeStr, Intrinsic Int,
X86FoldableSchedWrite sched> {
let isCommutable = 1 in
def rr : IXOPi8Reg<opc, MRMSrcReg, (outs VR128:$dst),
(ins VR128:$src1, VR128:$src2, VR128:$src3),
!strconcat(OpcodeStr,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[(set VR128:$dst,
(Int VR128:$src1, VR128:$src2, VR128:$src3))]>, XOP_4V,
Sched<[sched]>;
def rm : IXOPi8Reg<opc, MRMSrcMem, (outs VR128:$dst),
(ins VR128:$src1, i128mem:$src2, VR128:$src3),
!strconcat(OpcodeStr,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[(set VR128:$dst,
(Int VR128:$src1, (load addr:$src2),
VR128:$src3))]>, XOP_4V, Sched<[sched.Folded, sched.ReadAfterFold]>;
}
let ExeDomain = SSEPackedInt in {
defm VPMADCSWD : xop4opm2<0xB6, "vpmadcswd",
int_x86_xop_vpmadcswd, SchedWriteVecIMul.XMM>;
defm VPMADCSSWD : xop4opm2<0xA6, "vpmadcsswd",
int_x86_xop_vpmadcsswd, SchedWriteVecIMul.XMM>;
defm VPMACSWW : xop4opm2<0x95, "vpmacsww",
int_x86_xop_vpmacsww, SchedWriteVecIMul.XMM>;
defm VPMACSWD : xop4opm2<0x96, "vpmacswd",
int_x86_xop_vpmacswd, SchedWriteVecIMul.XMM>;
defm VPMACSSWW : xop4opm2<0x85, "vpmacssww",
int_x86_xop_vpmacssww, SchedWriteVecIMul.XMM>;
defm VPMACSSWD : xop4opm2<0x86, "vpmacsswd",
int_x86_xop_vpmacsswd, SchedWriteVecIMul.XMM>;
defm VPMACSSDQL : xop4opm2<0x87, "vpmacssdql",
int_x86_xop_vpmacssdql, SchedWritePMULLD.XMM>;
defm VPMACSSDQH : xop4opm2<0x8F, "vpmacssdqh",
int_x86_xop_vpmacssdqh, SchedWritePMULLD.XMM>;
defm VPMACSSDD : xop4opm2<0x8E, "vpmacssdd",
int_x86_xop_vpmacssdd, SchedWritePMULLD.XMM>;
defm VPMACSDQL : xop4opm2<0x97, "vpmacsdql",
int_x86_xop_vpmacsdql, SchedWritePMULLD.XMM>;
defm VPMACSDQH : xop4opm2<0x9F, "vpmacsdqh",
int_x86_xop_vpmacsdqh, SchedWritePMULLD.XMM>;
defm VPMACSDD : xop4opm2<0x9E, "vpmacsdd",
int_x86_xop_vpmacsdd, SchedWritePMULLD.XMM>;
}
// IFMA patterns - for cases where we can safely ignore the overflow bits from
// the multiply or easily match with existing intrinsics.
let Predicates = [HasXOP] in {
def : Pat<(v8i16 (add (mul (v8i16 VR128:$src1), (v8i16 VR128:$src2)),
(v8i16 VR128:$src3))),
(VPMACSWWrr VR128:$src1, VR128:$src2, VR128:$src3)>;
def : Pat<(v4i32 (add (mul (v4i32 VR128:$src1), (v4i32 VR128:$src2)),
(v4i32 VR128:$src3))),
(VPMACSDDrr VR128:$src1, VR128:$src2, VR128:$src3)>;
def : Pat<(v2i64 (add (X86pmuldq (bc_v2i64 (X86PShufd (v4i32 VR128:$src1), (i8 -11))),
(bc_v2i64 (X86PShufd (v4i32 VR128:$src2), (i8 -11)))),
(v2i64 VR128:$src3))),
(VPMACSDQHrr VR128:$src1, VR128:$src2, VR128:$src3)>;
def : Pat<(v2i64 (add (X86pmuldq (v2i64 VR128:$src1), (v2i64 VR128:$src2)),
(v2i64 VR128:$src3))),
(VPMACSDQLrr VR128:$src1, VR128:$src2, VR128:$src3)>;
def : Pat<(v4i32 (add (X86vpmaddwd (v8i16 VR128:$src1), (v8i16 VR128:$src2)),
(v4i32 VR128:$src3))),
(VPMADCSWDrr VR128:$src1, VR128:$src2, VR128:$src3)>;
}
// Transforms to swizzle an immediate to help matching memory operand in first
// operand.
def CommuteVPCOMCC : SDNodeXForm<imm, [{
uint8_t Imm = N->getZExtValue() & 0x7;
Imm = X86::getSwappedVPCOMImm(Imm);
return getI8Imm(Imm, SDLoc(N));
}]>;
// Instruction where second source can be memory, third must be imm8
multiclass xopvpcom<bits<8> opc, string Suffix, SDNode OpNode, ValueType vt128,
X86FoldableSchedWrite sched> {
let ExeDomain = SSEPackedInt in { // SSE integer instructions
let isCommutable = 1 in
def ri : IXOPi8<opc, MRMSrcReg, (outs VR128:$dst),
(ins VR128:$src1, VR128:$src2, XOPCC:$cc),
!strconcat("vpcom${cc}", Suffix,
"\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
[(set VR128:$dst,
(vt128 (OpNode (vt128 VR128:$src1), (vt128 VR128:$src2),
imm:$cc)))]>,
XOP_4V, Sched<[sched]>;
def mi : IXOPi8<opc, MRMSrcMem, (outs VR128:$dst),
(ins VR128:$src1, i128mem:$src2, XOPCC:$cc),
!strconcat("vpcom${cc}", Suffix,
"\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
[(set VR128:$dst,
(vt128 (OpNode (vt128 VR128:$src1),
(vt128 (load addr:$src2)),
imm:$cc)))]>,
XOP_4V, Sched<[sched.Folded, sched.ReadAfterFold]>;
let isAsmParserOnly = 1, hasSideEffects = 0 in {
def ri_alt : IXOPi8<opc, MRMSrcReg, (outs VR128:$dst),
(ins VR128:$src1, VR128:$src2, u8imm:$src3),
!strconcat("vpcom", Suffix,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[]>, XOP_4V, Sched<[sched]>, NotMemoryFoldable;
let mayLoad = 1 in
def mi_alt : IXOPi8<opc, MRMSrcMem, (outs VR128:$dst),
(ins VR128:$src1, i128mem:$src2, u8imm:$src3),
!strconcat("vpcom", Suffix,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[]>, XOP_4V, Sched<[sched.Folded, sched.ReadAfterFold]>,
NotMemoryFoldable;
}
}
def : Pat<(OpNode (load addr:$src2),
(vt128 VR128:$src1), imm:$cc),
(!cast<Instruction>(NAME#"mi") VR128:$src1, addr:$src2,
(CommuteVPCOMCC imm:$cc))>;
}
defm VPCOMB : xopvpcom<0xCC, "b", X86vpcom, v16i8, SchedWriteVecALU.XMM>;
defm VPCOMW : xopvpcom<0xCD, "w", X86vpcom, v8i16, SchedWriteVecALU.XMM>;
defm VPCOMD : xopvpcom<0xCE, "d", X86vpcom, v4i32, SchedWriteVecALU.XMM>;
defm VPCOMQ : xopvpcom<0xCF, "q", X86vpcom, v2i64, SchedWriteVecALU.XMM>;
defm VPCOMUB : xopvpcom<0xEC, "ub", X86vpcomu, v16i8, SchedWriteVecALU.XMM>;
defm VPCOMUW : xopvpcom<0xED, "uw", X86vpcomu, v8i16, SchedWriteVecALU.XMM>;
defm VPCOMUD : xopvpcom<0xEE, "ud", X86vpcomu, v4i32, SchedWriteVecALU.XMM>;
defm VPCOMUQ : xopvpcom<0xEF, "uq", X86vpcomu, v2i64, SchedWriteVecALU.XMM>;
multiclass xop4op<bits<8> opc, string OpcodeStr, SDNode OpNode,
ValueType vt128, X86FoldableSchedWrite sched> {
def rrr : IXOPi8Reg<opc, MRMSrcReg, (outs VR128:$dst),
(ins VR128:$src1, VR128:$src2, VR128:$src3),
!strconcat(OpcodeStr,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[(set VR128:$dst,
(vt128 (OpNode (vt128 VR128:$src1), (vt128 VR128:$src2),
(vt128 VR128:$src3))))]>,
XOP_4V, Sched<[sched]>;
def rrm : IXOPi8Reg<opc, MRMSrcMemOp4, (outs VR128:$dst),
(ins VR128:$src1, VR128:$src2, i128mem:$src3),
!strconcat(OpcodeStr,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[(set VR128:$dst,
(vt128 (OpNode (vt128 VR128:$src1), (vt128 VR128:$src2),
(vt128 (load addr:$src3)))))]>,
XOP_4V, VEX_W, Sched<[sched.Folded, sched.ReadAfterFold, sched.ReadAfterFold]>;
def rmr : IXOPi8Reg<opc, MRMSrcMem, (outs VR128:$dst),
(ins VR128:$src1, i128mem:$src2, VR128:$src3),
!strconcat(OpcodeStr,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[(set VR128:$dst,
(v16i8 (OpNode (vt128 VR128:$src1), (vt128 (load addr:$src2)),
(vt128 VR128:$src3))))]>,
XOP_4V, Sched<[sched.Folded, sched.ReadAfterFold,
// 128mem:$src2
ReadDefault, ReadDefault, ReadDefault, ReadDefault,
ReadDefault,
// VR128:$src3
sched.ReadAfterFold]>;
// For disassembler
let isCodeGenOnly = 1, ForceDisassemble = 1, hasSideEffects = 0 in
def rrr_REV : IXOPi8Reg<opc, MRMSrcRegOp4, (outs VR128:$dst),
(ins VR128:$src1, VR128:$src2, VR128:$src3),
!strconcat(OpcodeStr,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[]>, XOP_4V, VEX_W, Sched<[sched]>, FoldGenData<NAME#rrr>;
}
let ExeDomain = SSEPackedInt in {
defm VPPERM : xop4op<0xA3, "vpperm", X86vpperm, v16i8,
SchedWriteVarShuffle.XMM>;
}
// Instruction where either second or third source can be memory
multiclass xop4op_int<bits<8> opc, string OpcodeStr, RegisterClass RC,
X86MemOperand x86memop, ValueType VT,
X86FoldableSchedWrite sched> {
def rrr : IXOPi8Reg<opc, MRMSrcReg, (outs RC:$dst),
(ins RC:$src1, RC:$src2, RC:$src3),
!strconcat(OpcodeStr,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[(set RC:$dst, (VT (or (and RC:$src3, RC:$src1),
(X86andnp RC:$src3, RC:$src2))))]>, XOP_4V,
Sched<[sched]>;
// FIXME: This pattern can't match.
def rrm : IXOPi8Reg<opc, MRMSrcMemOp4, (outs RC:$dst),
(ins RC:$src1, RC:$src2, x86memop:$src3),
!strconcat(OpcodeStr,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[(set RC:$dst, (VT (or (and (load addr:$src3), RC:$src1),
(X86andnp (load addr:$src3), RC:$src2))))]>,
XOP_4V, VEX_W, Sched<[sched.Folded, sched.ReadAfterFold, sched.ReadAfterFold]>;
def rmr : IXOPi8Reg<opc, MRMSrcMem, (outs RC:$dst),
(ins RC:$src1, x86memop:$src2, RC:$src3),
!strconcat(OpcodeStr,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[(set RC:$dst, (VT (or (and RC:$src3, RC:$src1),
(X86andnp RC:$src3, (load addr:$src2)))))]>,
XOP_4V, Sched<[sched.Folded, sched.ReadAfterFold,
// x86memop:$src2
ReadDefault, ReadDefault, ReadDefault, ReadDefault,
ReadDefault,
// RC::$src3
sched.ReadAfterFold]>;
// For disassembler
let isCodeGenOnly = 1, ForceDisassemble = 1, hasSideEffects = 0 in
def rrr_REV : IXOPi8Reg<opc, MRMSrcRegOp4, (outs RC:$dst),
(ins RC:$src1, RC:$src2, RC:$src3),
!strconcat(OpcodeStr,
"\t{$src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3}"),
[]>, XOP_4V, VEX_W, Sched<[sched]>, FoldGenData<NAME#rrr>;
}
let ExeDomain = SSEPackedInt in {
defm VPCMOV : xop4op_int<0xA2, "vpcmov", VR128, i128mem, v2i64,
SchedWriteShuffle.XMM>;
defm VPCMOVY : xop4op_int<0xA2, "vpcmov", VR256, i256mem, v4i64,
SchedWriteShuffle.YMM>, VEX_L;
}
let Predicates = [HasXOP] in {
def : Pat<(v16i8 (or (and VR128:$src3, VR128:$src1),
(X86andnp VR128:$src3, VR128:$src2))),
(VPCMOVrrr VR128:$src1, VR128:$src2, VR128:$src3)>;
def : Pat<(v8i16 (or (and VR128:$src3, VR128:$src1),
(X86andnp VR128:$src3, VR128:$src2))),
(VPCMOVrrr VR128:$src1, VR128:$src2, VR128:$src3)>;
def : Pat<(v4i32 (or (and VR128:$src3, VR128:$src1),
(X86andnp VR128:$src3, VR128:$src2))),
(VPCMOVrrr VR128:$src1, VR128:$src2, VR128:$src3)>;
def : Pat<(or (and VR128:$src3, VR128:$src1),
(X86andnp VR128:$src3, (bc_v16i8 (loadv2i64 addr:$src2)))),
(VPCMOVrmr VR128:$src1, addr:$src2, VR128:$src3)>;
def : Pat<(or (and VR128:$src3, VR128:$src1),
(X86andnp VR128:$src3, (bc_v8i16 (loadv2i64 addr:$src2)))),
(VPCMOVrmr VR128:$src1, addr:$src2, VR128:$src3)>;
def : Pat<(or (and VR128:$src3, VR128:$src1),
(X86andnp VR128:$src3, (bc_v4i32 (loadv2i64 addr:$src2)))),
(VPCMOVrmr VR128:$src1, addr:$src2, VR128:$src3)>;
def : Pat<(v32i8 (or (and VR256:$src3, VR256:$src1),
(X86andnp VR256:$src3, VR256:$src2))),
(VPCMOVYrrr VR256:$src1, VR256:$src2, VR256:$src3)>;
def : Pat<(v16i16 (or (and VR256:$src3, VR256:$src1),
(X86andnp VR256:$src3, VR256:$src2))),
(VPCMOVYrrr VR256:$src1, VR256:$src2, VR256:$src3)>;
def : Pat<(v8i32 (or (and VR256:$src3, VR256:$src1),
(X86andnp VR256:$src3, VR256:$src2))),
(VPCMOVYrrr VR256:$src1, VR256:$src2, VR256:$src3)>;
def : Pat<(or (and VR256:$src3, VR256:$src1),
(X86andnp VR256:$src3, (bc_v32i8 (loadv4i64 addr:$src2)))),
(VPCMOVYrmr VR256:$src1, addr:$src2, VR256:$src3)>;
def : Pat<(or (and VR256:$src3, VR256:$src1),
(X86andnp VR256:$src3, (bc_v16i16 (loadv4i64 addr:$src2)))),
(VPCMOVYrmr VR256:$src1, addr:$src2, VR256:$src3)>;
def : Pat<(or (and VR256:$src3, VR256:$src1),
(X86andnp VR256:$src3, (bc_v8i32 (loadv4i64 addr:$src2)))),
(VPCMOVYrmr VR256:$src1, addr:$src2, VR256:$src3)>;
}
multiclass xop_vpermil2<bits<8> Opc, string OpcodeStr, RegisterClass RC,
X86MemOperand intmemop, X86MemOperand fpmemop,
ValueType VT, PatFrag FPLdFrag, PatFrag IntLdFrag,
X86FoldableSchedWrite sched> {
def rr : IXOP5<Opc, MRMSrcReg, (outs RC:$dst),
(ins RC:$src1, RC:$src2, RC:$src3, u8imm:$src4),
!strconcat(OpcodeStr,
"\t{$src4, $src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3, $src4}"),
[(set RC:$dst,
(VT (X86vpermil2 RC:$src1, RC:$src2, RC:$src3, (i8 imm:$src4))))]>,
Sched<[sched]>;
def rm : IXOP5<Opc, MRMSrcMemOp4, (outs RC:$dst),
(ins RC:$src1, RC:$src2, intmemop:$src3, u8imm:$src4),
!strconcat(OpcodeStr,
"\t{$src4, $src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3, $src4}"),
[(set RC:$dst,
(VT (X86vpermil2 RC:$src1, RC:$src2, (IntLdFrag addr:$src3),
(i8 imm:$src4))))]>, VEX_W,
Sched<[sched.Folded, sched.ReadAfterFold, sched.ReadAfterFold]>;
def mr : IXOP5<Opc, MRMSrcMem, (outs RC:$dst),
(ins RC:$src1, fpmemop:$src2, RC:$src3, u8imm:$src4),
!strconcat(OpcodeStr,
"\t{$src4, $src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3, $src4}"),
[(set RC:$dst,
(VT (X86vpermil2 RC:$src1, (FPLdFrag addr:$src2),
RC:$src3, (i8 imm:$src4))))]>,
Sched<[sched.Folded, sched.ReadAfterFold,
// fpmemop:$src2
ReadDefault, ReadDefault, ReadDefault, ReadDefault, ReadDefault,
// RC:$src3
sched.ReadAfterFold]>;
// For disassembler
let isCodeGenOnly = 1, ForceDisassemble = 1, hasSideEffects = 0 in
def rr_REV : IXOP5<Opc, MRMSrcRegOp4, (outs RC:$dst),
(ins RC:$src1, RC:$src2, RC:$src3, u8imm:$src4),
!strconcat(OpcodeStr,
"\t{$src4, $src3, $src2, $src1, $dst|$dst, $src1, $src2, $src3, $src4}"),
[]>, VEX_W, Sched<[sched]>, FoldGenData<NAME#rr>;
}
let ExeDomain = SSEPackedDouble in {
defm VPERMIL2PD : xop_vpermil2<0x49, "vpermil2pd", VR128, i128mem, f128mem,
v2f64, loadv2f64, loadv2i64,
SchedWriteFVarShuffle.XMM>;
defm VPERMIL2PDY : xop_vpermil2<0x49, "vpermil2pd", VR256, i256mem, f256mem,
v4f64, loadv4f64, loadv4i64,
SchedWriteFVarShuffle.YMM>, VEX_L;
}
let ExeDomain = SSEPackedSingle in {
defm VPERMIL2PS : xop_vpermil2<0x48, "vpermil2ps", VR128, i128mem, f128mem,
v4f32, loadv4f32, loadv4i32,
SchedWriteFVarShuffle.XMM>;
defm VPERMIL2PSY : xop_vpermil2<0x48, "vpermil2ps", VR256, i256mem, f256mem,
v8f32, loadv8f32, loadv8i32,
SchedWriteFVarShuffle.YMM>, VEX_L;
}