github.com/tidwall/go@v0.0.0-20170415222209-6694a6888b7d/src/cmd/compile/internal/ssa/gen/AMD64Ops.go (about) 1 // Copyright 2015 The Go Authors. All rights reserved. 2 // Use of this source code is governed by a BSD-style 3 // license that can be found in the LICENSE file. 4 5 // +build ignore 6 7 package main 8 9 import "strings" 10 11 // Notes: 12 // - Integer types live in the low portion of registers. Upper portions are junk. 13 // - Boolean types use the low-order byte of a register. 0=false, 1=true. 14 // Upper bytes are junk. 15 // - Floating-point types live in the low natural slot of an sse2 register. 16 // Unused portions are junk. 17 // - We do not use AH,BH,CH,DH registers. 18 // - When doing sub-register operations, we try to write the whole 19 // destination register to avoid a partial-register write. 20 // - Unused portions of AuxInt (or the Val portion of ValAndOff) are 21 // filled by sign-extending the used portion. Users of AuxInt which interpret 22 // AuxInt as unsigned (e.g. shifts) must be careful. 23 24 // Suffixes encode the bit width of various instructions. 25 // Q (quad word) = 64 bit 26 // L (long word) = 32 bit 27 // W (word) = 16 bit 28 // B (byte) = 8 bit 29 30 // copied from ../../amd64/reg.go 31 var regNamesAMD64 = []string{ 32 "AX", 33 "CX", 34 "DX", 35 "BX", 36 "SP", 37 "BP", 38 "SI", 39 "DI", 40 "R8", 41 "R9", 42 "R10", 43 "R11", 44 "R12", 45 "R13", 46 "R14", 47 "R15", 48 "X0", 49 "X1", 50 "X2", 51 "X3", 52 "X4", 53 "X5", 54 "X6", 55 "X7", 56 "X8", 57 "X9", 58 "X10", 59 "X11", 60 "X12", 61 "X13", 62 "X14", 63 "X15", 64 65 // pseudo-registers 66 "SB", 67 } 68 69 func init() { 70 // Make map from reg names to reg integers. 71 if len(regNamesAMD64) > 64 { 72 panic("too many registers") 73 } 74 num := map[string]int{} 75 for i, name := range regNamesAMD64 { 76 num[name] = i 77 } 78 buildReg := func(s string) regMask { 79 m := regMask(0) 80 for _, r := range strings.Split(s, " ") { 81 if n, ok := num[r]; ok { 82 m |= regMask(1) << uint(n) 83 continue 84 } 85 panic("register " + r + " not found") 86 } 87 return m 88 } 89 90 // Common individual register masks 91 var ( 92 ax = buildReg("AX") 93 cx = buildReg("CX") 94 dx = buildReg("DX") 95 gp = buildReg("AX CX DX BX BP SI DI R8 R9 R10 R11 R12 R13 R14 R15") 96 fp = buildReg("X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15") 97 gpsp = gp | buildReg("SP") 98 gpspsb = gpsp | buildReg("SB") 99 callerSave = gp | fp 100 ) 101 // Common slices of register masks 102 var ( 103 gponly = []regMask{gp} 104 fponly = []regMask{fp} 105 ) 106 107 // Common regInfo 108 var ( 109 gp01 = regInfo{inputs: nil, outputs: gponly} 110 gp11 = regInfo{inputs: []regMask{gp}, outputs: gponly} 111 gp11sp = regInfo{inputs: []regMask{gpsp}, outputs: gponly} 112 gp11sb = regInfo{inputs: []regMask{gpspsb}, outputs: gponly} 113 gp21 = regInfo{inputs: []regMask{gp, gp}, outputs: gponly} 114 gp21sp = regInfo{inputs: []regMask{gpsp, gp}, outputs: gponly} 115 gp21sb = regInfo{inputs: []regMask{gpspsb, gpsp}, outputs: gponly} 116 gp21shift = regInfo{inputs: []regMask{gp, cx}, outputs: []regMask{gp}} 117 gp11div = regInfo{inputs: []regMask{ax, gpsp &^ dx}, outputs: []regMask{ax, dx}} 118 gp21hmul = regInfo{inputs: []regMask{ax, gpsp}, outputs: []regMask{dx}, clobbers: ax} 119 120 gp2flags = regInfo{inputs: []regMask{gpsp, gpsp}} 121 gp1flags = regInfo{inputs: []regMask{gpsp}} 122 flagsgp = regInfo{inputs: nil, outputs: gponly} 123 124 gp11flags = regInfo{inputs: []regMask{gp}, outputs: []regMask{gp, 0}} 125 126 readflags = regInfo{inputs: nil, outputs: gponly} 127 flagsgpax = regInfo{inputs: nil, clobbers: ax, outputs: []regMask{gp &^ ax}} 128 129 gpload = regInfo{inputs: []regMask{gpspsb, 0}, outputs: gponly} 130 gp21load = regInfo{inputs: []regMask{gp, gpspsb, 0}, outputs: gponly} 131 gploadidx = regInfo{inputs: []regMask{gpspsb, gpsp, 0}, outputs: gponly} 132 133 gpstore = regInfo{inputs: []regMask{gpspsb, gpsp, 0}} 134 gpstoreconst = regInfo{inputs: []regMask{gpspsb, 0}} 135 gpstoreidx = regInfo{inputs: []regMask{gpspsb, gpsp, gpsp, 0}} 136 gpstoreconstidx = regInfo{inputs: []regMask{gpspsb, gpsp, 0}} 137 gpstorexchg = regInfo{inputs: []regMask{gp, gpspsb, 0}, outputs: []regMask{gp}} 138 cmpxchg = regInfo{inputs: []regMask{gp, ax, gp, 0}, outputs: []regMask{gp, 0}, clobbers: ax} 139 140 fp01 = regInfo{inputs: nil, outputs: fponly} 141 fp21 = regInfo{inputs: []regMask{fp, fp}, outputs: fponly} 142 fp21load = regInfo{inputs: []regMask{fp, gpspsb, 0}, outputs: fponly} 143 fpgp = regInfo{inputs: fponly, outputs: gponly} 144 gpfp = regInfo{inputs: gponly, outputs: fponly} 145 fp11 = regInfo{inputs: fponly, outputs: fponly} 146 fp2flags = regInfo{inputs: []regMask{fp, fp}} 147 148 fpload = regInfo{inputs: []regMask{gpspsb, 0}, outputs: fponly} 149 fploadidx = regInfo{inputs: []regMask{gpspsb, gpsp, 0}, outputs: fponly} 150 151 fpstore = regInfo{inputs: []regMask{gpspsb, fp, 0}} 152 fpstoreidx = regInfo{inputs: []regMask{gpspsb, gpsp, fp, 0}} 153 ) 154 155 var AMD64ops = []opData{ 156 // fp ops 157 {name: "ADDSS", argLength: 2, reg: fp21, asm: "ADDSS", commutative: true, resultInArg0: true}, // fp32 add 158 {name: "ADDSD", argLength: 2, reg: fp21, asm: "ADDSD", commutative: true, resultInArg0: true}, // fp64 add 159 {name: "SUBSS", argLength: 2, reg: fp21, asm: "SUBSS", resultInArg0: true}, // fp32 sub 160 {name: "SUBSD", argLength: 2, reg: fp21, asm: "SUBSD", resultInArg0: true}, // fp64 sub 161 {name: "MULSS", argLength: 2, reg: fp21, asm: "MULSS", commutative: true, resultInArg0: true}, // fp32 mul 162 {name: "MULSD", argLength: 2, reg: fp21, asm: "MULSD", commutative: true, resultInArg0: true}, // fp64 mul 163 {name: "DIVSS", argLength: 2, reg: fp21, asm: "DIVSS", resultInArg0: true}, // fp32 div 164 {name: "DIVSD", argLength: 2, reg: fp21, asm: "DIVSD", resultInArg0: true}, // fp64 div 165 166 {name: "MOVSSload", argLength: 2, reg: fpload, asm: "MOVSS", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"}, // fp32 load 167 {name: "MOVSDload", argLength: 2, reg: fpload, asm: "MOVSD", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"}, // fp64 load 168 {name: "MOVSSconst", reg: fp01, asm: "MOVSS", aux: "Float32", rematerializeable: true}, // fp32 constant 169 {name: "MOVSDconst", reg: fp01, asm: "MOVSD", aux: "Float64", rematerializeable: true}, // fp64 constant 170 {name: "MOVSSloadidx1", argLength: 3, reg: fploadidx, asm: "MOVSS", aux: "SymOff", symEffect: "Read"}, // fp32 load indexed by i 171 {name: "MOVSSloadidx4", argLength: 3, reg: fploadidx, asm: "MOVSS", aux: "SymOff", symEffect: "Read"}, // fp32 load indexed by 4*i 172 {name: "MOVSDloadidx1", argLength: 3, reg: fploadidx, asm: "MOVSD", aux: "SymOff", symEffect: "Read"}, // fp64 load indexed by i 173 {name: "MOVSDloadidx8", argLength: 3, reg: fploadidx, asm: "MOVSD", aux: "SymOff", symEffect: "Read"}, // fp64 load indexed by 8*i 174 175 {name: "MOVSSstore", argLength: 3, reg: fpstore, asm: "MOVSS", aux: "SymOff", faultOnNilArg0: true, symEffect: "Write"}, // fp32 store 176 {name: "MOVSDstore", argLength: 3, reg: fpstore, asm: "MOVSD", aux: "SymOff", faultOnNilArg0: true, symEffect: "Write"}, // fp64 store 177 {name: "MOVSSstoreidx1", argLength: 4, reg: fpstoreidx, asm: "MOVSS", aux: "SymOff", symEffect: "Write"}, // fp32 indexed by i store 178 {name: "MOVSSstoreidx4", argLength: 4, reg: fpstoreidx, asm: "MOVSS", aux: "SymOff", symEffect: "Write"}, // fp32 indexed by 4i store 179 {name: "MOVSDstoreidx1", argLength: 4, reg: fpstoreidx, asm: "MOVSD", aux: "SymOff", symEffect: "Write"}, // fp64 indexed by i store 180 {name: "MOVSDstoreidx8", argLength: 4, reg: fpstoreidx, asm: "MOVSD", aux: "SymOff", symEffect: "Write"}, // fp64 indexed by 8i store 181 182 {name: "ADDSDmem", argLength: 3, reg: fp21load, asm: "ADDSD", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, symEffect: "Read"}, // fp32 arg0 + tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 183 {name: "ADDSSmem", argLength: 3, reg: fp21load, asm: "ADDSS", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, symEffect: "Read"}, // fp32 arg0 + tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 184 {name: "SUBSSmem", argLength: 3, reg: fp21load, asm: "SUBSS", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, symEffect: "Read"}, // fp32 arg0 - tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 185 {name: "SUBSDmem", argLength: 3, reg: fp21load, asm: "SUBSD", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, symEffect: "Read"}, // fp64 arg0 - tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 186 {name: "MULSSmem", argLength: 3, reg: fp21load, asm: "MULSS", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, symEffect: "Read"}, // fp32 arg0 * tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 187 {name: "MULSDmem", argLength: 3, reg: fp21load, asm: "MULSD", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, symEffect: "Read"}, // fp64 arg0 * tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 188 189 // binary ops 190 {name: "ADDQ", argLength: 2, reg: gp21sp, asm: "ADDQ", commutative: true, clobberFlags: true}, // arg0 + arg1 191 {name: "ADDL", argLength: 2, reg: gp21sp, asm: "ADDL", commutative: true, clobberFlags: true}, // arg0 + arg1 192 {name: "ADDQconst", argLength: 1, reg: gp11sp, asm: "ADDQ", aux: "Int64", typ: "UInt64", clobberFlags: true}, // arg0 + auxint 193 {name: "ADDLconst", argLength: 1, reg: gp11sp, asm: "ADDL", aux: "Int32", clobberFlags: true}, // arg0 + auxint 194 195 {name: "SUBQ", argLength: 2, reg: gp21, asm: "SUBQ", resultInArg0: true, clobberFlags: true}, // arg0 - arg1 196 {name: "SUBL", argLength: 2, reg: gp21, asm: "SUBL", resultInArg0: true, clobberFlags: true}, // arg0 - arg1 197 {name: "SUBQconst", argLength: 1, reg: gp11, asm: "SUBQ", aux: "Int64", resultInArg0: true, clobberFlags: true}, // arg0 - auxint 198 {name: "SUBLconst", argLength: 1, reg: gp11, asm: "SUBL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 - auxint 199 200 {name: "MULQ", argLength: 2, reg: gp21, asm: "IMULQ", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 * arg1 201 {name: "MULL", argLength: 2, reg: gp21, asm: "IMULL", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 * arg1 202 {name: "MULQconst", argLength: 1, reg: gp11, asm: "IMULQ", aux: "Int64", resultInArg0: true, clobberFlags: true}, // arg0 * auxint 203 {name: "MULLconst", argLength: 1, reg: gp11, asm: "IMULL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 * auxint 204 205 {name: "HMULQ", argLength: 2, reg: gp21hmul, commutative: true, asm: "IMULQ", clobberFlags: true}, // (arg0 * arg1) >> width 206 {name: "HMULL", argLength: 2, reg: gp21hmul, commutative: true, asm: "IMULL", clobberFlags: true}, // (arg0 * arg1) >> width 207 {name: "HMULQU", argLength: 2, reg: gp21hmul, commutative: true, asm: "MULQ", clobberFlags: true}, // (arg0 * arg1) >> width 208 {name: "HMULLU", argLength: 2, reg: gp21hmul, commutative: true, asm: "MULL", clobberFlags: true}, // (arg0 * arg1) >> width 209 210 {name: "AVGQU", argLength: 2, reg: gp21, commutative: true, resultInArg0: true, clobberFlags: true}, // (arg0 + arg1) / 2 as unsigned, all 64 result bits 211 212 {name: "DIVQ", argLength: 2, reg: gp11div, typ: "(Int64,Int64)", asm: "IDIVQ", clobberFlags: true}, // [arg0 / arg1, arg0 % arg1] 213 {name: "DIVL", argLength: 2, reg: gp11div, typ: "(Int32,Int32)", asm: "IDIVL", clobberFlags: true}, // [arg0 / arg1, arg0 % arg1] 214 {name: "DIVW", argLength: 2, reg: gp11div, typ: "(Int16,Int16)", asm: "IDIVW", clobberFlags: true}, // [arg0 / arg1, arg0 % arg1] 215 {name: "DIVQU", argLength: 2, reg: gp11div, typ: "(UInt64,UInt64)", asm: "DIVQ", clobberFlags: true}, // [arg0 / arg1, arg0 % arg1] 216 {name: "DIVLU", argLength: 2, reg: gp11div, typ: "(UInt32,UInt32)", asm: "DIVL", clobberFlags: true}, // [arg0 / arg1, arg0 % arg1] 217 {name: "DIVWU", argLength: 2, reg: gp11div, typ: "(UInt16,UInt16)", asm: "DIVW", clobberFlags: true}, // [arg0 / arg1, arg0 % arg1] 218 219 {name: "MULQU2", argLength: 2, reg: regInfo{inputs: []regMask{ax, gpsp}, outputs: []regMask{dx, ax}}, commutative: true, asm: "MULQ", clobberFlags: true}, // arg0 * arg1, returns (hi, lo) 220 {name: "DIVQU2", argLength: 3, reg: regInfo{inputs: []regMask{dx, ax, gpsp}, outputs: []regMask{ax, dx}}, asm: "DIVQ", clobberFlags: true}, // arg0:arg1 / arg2 (128-bit divided by 64-bit), returns (q, r) 221 222 {name: "ANDQ", argLength: 2, reg: gp21, asm: "ANDQ", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 & arg1 223 {name: "ANDL", argLength: 2, reg: gp21, asm: "ANDL", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 & arg1 224 {name: "ANDQconst", argLength: 1, reg: gp11, asm: "ANDQ", aux: "Int64", resultInArg0: true, clobberFlags: true}, // arg0 & auxint 225 {name: "ANDLconst", argLength: 1, reg: gp11, asm: "ANDL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 & auxint 226 227 {name: "ORQ", argLength: 2, reg: gp21, asm: "ORQ", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 | arg1 228 {name: "ORL", argLength: 2, reg: gp21, asm: "ORL", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 | arg1 229 {name: "ORQconst", argLength: 1, reg: gp11, asm: "ORQ", aux: "Int64", resultInArg0: true, clobberFlags: true}, // arg0 | auxint 230 {name: "ORLconst", argLength: 1, reg: gp11, asm: "ORL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 | auxint 231 232 {name: "XORQ", argLength: 2, reg: gp21, asm: "XORQ", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 ^ arg1 233 {name: "XORL", argLength: 2, reg: gp21, asm: "XORL", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 ^ arg1 234 {name: "XORQconst", argLength: 1, reg: gp11, asm: "XORQ", aux: "Int64", resultInArg0: true, clobberFlags: true}, // arg0 ^ auxint 235 {name: "XORLconst", argLength: 1, reg: gp11, asm: "XORL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 ^ auxint 236 237 {name: "CMPQ", argLength: 2, reg: gp2flags, asm: "CMPQ", typ: "Flags"}, // arg0 compare to arg1 238 {name: "CMPL", argLength: 2, reg: gp2flags, asm: "CMPL", typ: "Flags"}, // arg0 compare to arg1 239 {name: "CMPW", argLength: 2, reg: gp2flags, asm: "CMPW", typ: "Flags"}, // arg0 compare to arg1 240 {name: "CMPB", argLength: 2, reg: gp2flags, asm: "CMPB", typ: "Flags"}, // arg0 compare to arg1 241 {name: "CMPQconst", argLength: 1, reg: gp1flags, asm: "CMPQ", typ: "Flags", aux: "Int64"}, // arg0 compare to auxint 242 {name: "CMPLconst", argLength: 1, reg: gp1flags, asm: "CMPL", typ: "Flags", aux: "Int32"}, // arg0 compare to auxint 243 {name: "CMPWconst", argLength: 1, reg: gp1flags, asm: "CMPW", typ: "Flags", aux: "Int16"}, // arg0 compare to auxint 244 {name: "CMPBconst", argLength: 1, reg: gp1flags, asm: "CMPB", typ: "Flags", aux: "Int8"}, // arg0 compare to auxint 245 246 {name: "UCOMISS", argLength: 2, reg: fp2flags, asm: "UCOMISS", typ: "Flags"}, // arg0 compare to arg1, f32 247 {name: "UCOMISD", argLength: 2, reg: fp2flags, asm: "UCOMISD", typ: "Flags"}, // arg0 compare to arg1, f64 248 249 {name: "BTL", argLength: 2, reg: gp2flags, asm: "BTL", typ: "Flags"}, // test whether bit arg0 % 32 in arg1 is set 250 {name: "BTQ", argLength: 2, reg: gp2flags, asm: "BTQ", typ: "Flags"}, // test whether bit arg0 % 64 in arg1 is set 251 {name: "BTLconst", argLength: 1, reg: gp1flags, asm: "BTL", typ: "Flags", aux: "Int8"}, // test whether bit auxint in arg0 is set, 0 <= auxint < 32 252 {name: "BTQconst", argLength: 1, reg: gp1flags, asm: "BTQ", typ: "Flags", aux: "Int8"}, // test whether bit auxint in arg0 is set, 0 <= auxint < 64 253 254 {name: "TESTQ", argLength: 2, reg: gp2flags, commutative: true, asm: "TESTQ", typ: "Flags"}, // (arg0 & arg1) compare to 0 255 {name: "TESTL", argLength: 2, reg: gp2flags, commutative: true, asm: "TESTL", typ: "Flags"}, // (arg0 & arg1) compare to 0 256 {name: "TESTW", argLength: 2, reg: gp2flags, commutative: true, asm: "TESTW", typ: "Flags"}, // (arg0 & arg1) compare to 0 257 {name: "TESTB", argLength: 2, reg: gp2flags, commutative: true, asm: "TESTB", typ: "Flags"}, // (arg0 & arg1) compare to 0 258 {name: "TESTQconst", argLength: 1, reg: gp1flags, asm: "TESTQ", typ: "Flags", aux: "Int64"}, // (arg0 & auxint) compare to 0 259 {name: "TESTLconst", argLength: 1, reg: gp1flags, asm: "TESTL", typ: "Flags", aux: "Int32"}, // (arg0 & auxint) compare to 0 260 {name: "TESTWconst", argLength: 1, reg: gp1flags, asm: "TESTW", typ: "Flags", aux: "Int16"}, // (arg0 & auxint) compare to 0 261 {name: "TESTBconst", argLength: 1, reg: gp1flags, asm: "TESTB", typ: "Flags", aux: "Int8"}, // (arg0 & auxint) compare to 0 262 263 {name: "SHLQ", argLength: 2, reg: gp21shift, asm: "SHLQ", resultInArg0: true, clobberFlags: true}, // arg0 << arg1, shift amount is mod 64 264 {name: "SHLL", argLength: 2, reg: gp21shift, asm: "SHLL", resultInArg0: true, clobberFlags: true}, // arg0 << arg1, shift amount is mod 32 265 {name: "SHLQconst", argLength: 1, reg: gp11, asm: "SHLQ", aux: "Int8", resultInArg0: true, clobberFlags: true}, // arg0 << auxint, shift amount 0-63 266 {name: "SHLLconst", argLength: 1, reg: gp11, asm: "SHLL", aux: "Int8", resultInArg0: true, clobberFlags: true}, // arg0 << auxint, shift amount 0-31 267 // Note: x86 is weird, the 16 and 8 byte shifts still use all 5 bits of shift amount! 268 269 {name: "SHRQ", argLength: 2, reg: gp21shift, asm: "SHRQ", resultInArg0: true, clobberFlags: true}, // unsigned arg0 >> arg1, shift amount is mod 64 270 {name: "SHRL", argLength: 2, reg: gp21shift, asm: "SHRL", resultInArg0: true, clobberFlags: true}, // unsigned arg0 >> arg1, shift amount is mod 32 271 {name: "SHRW", argLength: 2, reg: gp21shift, asm: "SHRW", resultInArg0: true, clobberFlags: true}, // unsigned arg0 >> arg1, shift amount is mod 32 272 {name: "SHRB", argLength: 2, reg: gp21shift, asm: "SHRB", resultInArg0: true, clobberFlags: true}, // unsigned arg0 >> arg1, shift amount is mod 32 273 {name: "SHRQconst", argLength: 1, reg: gp11, asm: "SHRQ", aux: "Int8", resultInArg0: true, clobberFlags: true}, // unsigned arg0 >> auxint, shift amount 0-63 274 {name: "SHRLconst", argLength: 1, reg: gp11, asm: "SHRL", aux: "Int8", resultInArg0: true, clobberFlags: true}, // unsigned arg0 >> auxint, shift amount 0-31 275 {name: "SHRWconst", argLength: 1, reg: gp11, asm: "SHRW", aux: "Int8", resultInArg0: true, clobberFlags: true}, // unsigned arg0 >> auxint, shift amount 0-15 276 {name: "SHRBconst", argLength: 1, reg: gp11, asm: "SHRB", aux: "Int8", resultInArg0: true, clobberFlags: true}, // unsigned arg0 >> auxint, shift amount 0-7 277 278 {name: "SARQ", argLength: 2, reg: gp21shift, asm: "SARQ", resultInArg0: true, clobberFlags: true}, // signed arg0 >> arg1, shift amount is mod 64 279 {name: "SARL", argLength: 2, reg: gp21shift, asm: "SARL", resultInArg0: true, clobberFlags: true}, // signed arg0 >> arg1, shift amount is mod 32 280 {name: "SARW", argLength: 2, reg: gp21shift, asm: "SARW", resultInArg0: true, clobberFlags: true}, // signed arg0 >> arg1, shift amount is mod 32 281 {name: "SARB", argLength: 2, reg: gp21shift, asm: "SARB", resultInArg0: true, clobberFlags: true}, // signed arg0 >> arg1, shift amount is mod 32 282 {name: "SARQconst", argLength: 1, reg: gp11, asm: "SARQ", aux: "Int8", resultInArg0: true, clobberFlags: true}, // signed arg0 >> auxint, shift amount 0-63 283 {name: "SARLconst", argLength: 1, reg: gp11, asm: "SARL", aux: "Int8", resultInArg0: true, clobberFlags: true}, // signed arg0 >> auxint, shift amount 0-31 284 {name: "SARWconst", argLength: 1, reg: gp11, asm: "SARW", aux: "Int8", resultInArg0: true, clobberFlags: true}, // signed arg0 >> auxint, shift amount 0-15 285 {name: "SARBconst", argLength: 1, reg: gp11, asm: "SARB", aux: "Int8", resultInArg0: true, clobberFlags: true}, // signed arg0 >> auxint, shift amount 0-7 286 287 {name: "ROLQconst", argLength: 1, reg: gp11, asm: "ROLQ", aux: "Int8", resultInArg0: true, clobberFlags: true}, // arg0 rotate left auxint, rotate amount 0-63 288 {name: "ROLLconst", argLength: 1, reg: gp11, asm: "ROLL", aux: "Int8", resultInArg0: true, clobberFlags: true}, // arg0 rotate left auxint, rotate amount 0-31 289 {name: "ROLWconst", argLength: 1, reg: gp11, asm: "ROLW", aux: "Int8", resultInArg0: true, clobberFlags: true}, // arg0 rotate left auxint, rotate amount 0-15 290 {name: "ROLBconst", argLength: 1, reg: gp11, asm: "ROLB", aux: "Int8", resultInArg0: true, clobberFlags: true}, // arg0 rotate left auxint, rotate amount 0-7 291 292 {name: "ADDLmem", argLength: 3, reg: gp21load, asm: "ADDL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 + tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 293 {name: "ADDQmem", argLength: 3, reg: gp21load, asm: "ADDQ", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 + tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 294 {name: "SUBQmem", argLength: 3, reg: gp21load, asm: "SUBQ", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 - tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 295 {name: "SUBLmem", argLength: 3, reg: gp21load, asm: "SUBL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 - tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 296 {name: "ANDLmem", argLength: 3, reg: gp21load, asm: "ANDL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 & tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 297 {name: "ANDQmem", argLength: 3, reg: gp21load, asm: "ANDQ", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 & tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 298 {name: "ORQmem", argLength: 3, reg: gp21load, asm: "ORQ", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 | tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 299 {name: "ORLmem", argLength: 3, reg: gp21load, asm: "ORL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 | tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 300 {name: "XORQmem", argLength: 3, reg: gp21load, asm: "XORQ", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 ^ tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 301 {name: "XORLmem", argLength: 3, reg: gp21load, asm: "XORL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 ^ tmp, tmp loaded from arg1+auxint+aux, arg2 = mem 302 303 // unary ops 304 {name: "NEGQ", argLength: 1, reg: gp11, asm: "NEGQ", resultInArg0: true, clobberFlags: true}, // -arg0 305 {name: "NEGL", argLength: 1, reg: gp11, asm: "NEGL", resultInArg0: true, clobberFlags: true}, // -arg0 306 307 {name: "NOTQ", argLength: 1, reg: gp11, asm: "NOTQ", resultInArg0: true, clobberFlags: true}, // ^arg0 308 {name: "NOTL", argLength: 1, reg: gp11, asm: "NOTL", resultInArg0: true, clobberFlags: true}, // ^arg0 309 310 // BSF{L,Q} returns a tuple [result, flags] 311 // result is undefined if the input is zero. 312 // flags are set to "equal" if the input is zero, "not equal" otherwise. 313 {name: "BSFQ", argLength: 1, reg: gp11flags, asm: "BSFQ", typ: "(UInt64,Flags)"}, // # of low-order zeroes in 64-bit arg 314 {name: "BSFL", argLength: 1, reg: gp11flags, asm: "BSFL", typ: "(UInt32,Flags)"}, // # of low-order zeroes in 32-bit arg 315 {name: "BSRQ", argLength: 1, reg: gp11flags, asm: "BSRQ", typ: "(UInt64,Flags)"}, // # of high-order zeroes in 64-bit arg 316 {name: "BSRL", argLength: 1, reg: gp11flags, asm: "BSRL", typ: "(UInt32,Flags)"}, // # of high-order zeroes in 32-bit arg 317 318 // Note ASM for ops moves whole register 319 // 320 {name: "CMOVQEQ", argLength: 3, reg: gp21, asm: "CMOVQEQ", resultInArg0: true}, // if arg2 encodes "equal" return arg1 else arg0 321 {name: "CMOVLEQ", argLength: 3, reg: gp21, asm: "CMOVLEQ", resultInArg0: true}, // if arg2 encodes "equal" return arg1 else arg0 322 323 {name: "BSWAPQ", argLength: 1, reg: gp11, asm: "BSWAPQ", resultInArg0: true, clobberFlags: true}, // arg0 swap bytes 324 {name: "BSWAPL", argLength: 1, reg: gp11, asm: "BSWAPL", resultInArg0: true, clobberFlags: true}, // arg0 swap bytes 325 326 // POPCNT instructions aren't guaranteed to be on the target platform (they are SSE4). 327 // Any use must be preceded by a successful check of runtime.support_popcnt. 328 {name: "POPCNTQ", argLength: 1, reg: gp11, asm: "POPCNTQ", clobberFlags: true}, // count number of set bits in arg0 329 {name: "POPCNTL", argLength: 1, reg: gp11, asm: "POPCNTL", clobberFlags: true}, // count number of set bits in arg0 330 331 {name: "SQRTSD", argLength: 1, reg: fp11, asm: "SQRTSD"}, // sqrt(arg0) 332 333 {name: "SBBQcarrymask", argLength: 1, reg: flagsgp, asm: "SBBQ"}, // (int64)(-1) if carry is set, 0 if carry is clear. 334 {name: "SBBLcarrymask", argLength: 1, reg: flagsgp, asm: "SBBL"}, // (int32)(-1) if carry is set, 0 if carry is clear. 335 // Note: SBBW and SBBB are subsumed by SBBL 336 337 {name: "SETEQ", argLength: 1, reg: readflags, asm: "SETEQ"}, // extract == condition from arg0 338 {name: "SETNE", argLength: 1, reg: readflags, asm: "SETNE"}, // extract != condition from arg0 339 {name: "SETL", argLength: 1, reg: readflags, asm: "SETLT"}, // extract signed < condition from arg0 340 {name: "SETLE", argLength: 1, reg: readflags, asm: "SETLE"}, // extract signed <= condition from arg0 341 {name: "SETG", argLength: 1, reg: readflags, asm: "SETGT"}, // extract signed > condition from arg0 342 {name: "SETGE", argLength: 1, reg: readflags, asm: "SETGE"}, // extract signed >= condition from arg0 343 {name: "SETB", argLength: 1, reg: readflags, asm: "SETCS"}, // extract unsigned < condition from arg0 344 {name: "SETBE", argLength: 1, reg: readflags, asm: "SETLS"}, // extract unsigned <= condition from arg0 345 {name: "SETA", argLength: 1, reg: readflags, asm: "SETHI"}, // extract unsigned > condition from arg0 346 {name: "SETAE", argLength: 1, reg: readflags, asm: "SETCC"}, // extract unsigned >= condition from arg0 347 // Need different opcodes for floating point conditions because 348 // any comparison involving a NaN is always FALSE and thus 349 // the patterns for inverting conditions cannot be used. 350 {name: "SETEQF", argLength: 1, reg: flagsgpax, asm: "SETEQ", clobberFlags: true}, // extract == condition from arg0 351 {name: "SETNEF", argLength: 1, reg: flagsgpax, asm: "SETNE", clobberFlags: true}, // extract != condition from arg0 352 {name: "SETORD", argLength: 1, reg: flagsgp, asm: "SETPC"}, // extract "ordered" (No Nan present) condition from arg0 353 {name: "SETNAN", argLength: 1, reg: flagsgp, asm: "SETPS"}, // extract "unordered" (Nan present) condition from arg0 354 355 {name: "SETGF", argLength: 1, reg: flagsgp, asm: "SETHI"}, // extract floating > condition from arg0 356 {name: "SETGEF", argLength: 1, reg: flagsgp, asm: "SETCC"}, // extract floating >= condition from arg0 357 358 {name: "MOVBQSX", argLength: 1, reg: gp11, asm: "MOVBQSX"}, // sign extend arg0 from int8 to int64 359 {name: "MOVBQZX", argLength: 1, reg: gp11, asm: "MOVBLZX"}, // zero extend arg0 from int8 to int64 360 {name: "MOVWQSX", argLength: 1, reg: gp11, asm: "MOVWQSX"}, // sign extend arg0 from int16 to int64 361 {name: "MOVWQZX", argLength: 1, reg: gp11, asm: "MOVWLZX"}, // zero extend arg0 from int16 to int64 362 {name: "MOVLQSX", argLength: 1, reg: gp11, asm: "MOVLQSX"}, // sign extend arg0 from int32 to int64 363 {name: "MOVLQZX", argLength: 1, reg: gp11, asm: "MOVL"}, // zero extend arg0 from int32 to int64 364 365 {name: "MOVLconst", reg: gp01, asm: "MOVL", typ: "UInt32", aux: "Int32", rematerializeable: true}, // 32 low bits of auxint 366 {name: "MOVQconst", reg: gp01, asm: "MOVQ", typ: "UInt64", aux: "Int64", rematerializeable: true}, // auxint 367 368 {name: "CVTTSD2SL", argLength: 1, reg: fpgp, asm: "CVTTSD2SL"}, // convert float64 to int32 369 {name: "CVTTSD2SQ", argLength: 1, reg: fpgp, asm: "CVTTSD2SQ"}, // convert float64 to int64 370 {name: "CVTTSS2SL", argLength: 1, reg: fpgp, asm: "CVTTSS2SL"}, // convert float32 to int32 371 {name: "CVTTSS2SQ", argLength: 1, reg: fpgp, asm: "CVTTSS2SQ"}, // convert float32 to int64 372 {name: "CVTSL2SS", argLength: 1, reg: gpfp, asm: "CVTSL2SS"}, // convert int32 to float32 373 {name: "CVTSL2SD", argLength: 1, reg: gpfp, asm: "CVTSL2SD"}, // convert int32 to float64 374 {name: "CVTSQ2SS", argLength: 1, reg: gpfp, asm: "CVTSQ2SS"}, // convert int64 to float32 375 {name: "CVTSQ2SD", argLength: 1, reg: gpfp, asm: "CVTSQ2SD"}, // convert int64 to float64 376 {name: "CVTSD2SS", argLength: 1, reg: fp11, asm: "CVTSD2SS"}, // convert float64 to float32 377 {name: "CVTSS2SD", argLength: 1, reg: fp11, asm: "CVTSS2SD"}, // convert float32 to float64 378 379 {name: "PXOR", argLength: 2, reg: fp21, asm: "PXOR", commutative: true, resultInArg0: true}, // exclusive or, applied to X regs for float negation. 380 381 {name: "LEAQ", argLength: 1, reg: gp11sb, asm: "LEAQ", aux: "SymOff", rematerializeable: true, symEffect: "Addr"}, // arg0 + auxint + offset encoded in aux 382 {name: "LEAQ1", argLength: 2, reg: gp21sb, commutative: true, aux: "SymOff", symEffect: "Addr"}, // arg0 + arg1 + auxint + aux 383 {name: "LEAQ2", argLength: 2, reg: gp21sb, aux: "SymOff", symEffect: "Addr"}, // arg0 + 2*arg1 + auxint + aux 384 {name: "LEAQ4", argLength: 2, reg: gp21sb, aux: "SymOff", symEffect: "Addr"}, // arg0 + 4*arg1 + auxint + aux 385 {name: "LEAQ8", argLength: 2, reg: gp21sb, aux: "SymOff", symEffect: "Addr"}, // arg0 + 8*arg1 + auxint + aux 386 // Note: LEAQ{1,2,4,8} must not have OpSB as either argument. 387 388 {name: "LEAL", argLength: 1, reg: gp11sb, asm: "LEAL", aux: "SymOff", rematerializeable: true, symEffect: "Addr"}, // arg0 + auxint + offset encoded in aux 389 390 // auxint+aux == add auxint and the offset of the symbol in aux (if any) to the effective address 391 {name: "MOVBload", argLength: 2, reg: gpload, asm: "MOVBLZX", aux: "SymOff", typ: "UInt8", faultOnNilArg0: true, symEffect: "Read"}, // load byte from arg0+auxint+aux. arg1=mem. Zero extend. 392 {name: "MOVBQSXload", argLength: 2, reg: gpload, asm: "MOVBQSX", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"}, // ditto, sign extend to int64 393 {name: "MOVWload", argLength: 2, reg: gpload, asm: "MOVWLZX", aux: "SymOff", typ: "UInt16", faultOnNilArg0: true, symEffect: "Read"}, // load 2 bytes from arg0+auxint+aux. arg1=mem. Zero extend. 394 {name: "MOVWQSXload", argLength: 2, reg: gpload, asm: "MOVWQSX", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"}, // ditto, sign extend to int64 395 {name: "MOVLload", argLength: 2, reg: gpload, asm: "MOVL", aux: "SymOff", typ: "UInt32", faultOnNilArg0: true, symEffect: "Read"}, // load 4 bytes from arg0+auxint+aux. arg1=mem. Zero extend. 396 {name: "MOVLQSXload", argLength: 2, reg: gpload, asm: "MOVLQSX", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"}, // ditto, sign extend to int64 397 {name: "MOVQload", argLength: 2, reg: gpload, asm: "MOVQ", aux: "SymOff", typ: "UInt64", faultOnNilArg0: true, symEffect: "Read"}, // load 8 bytes from arg0+auxint+aux. arg1=mem 398 {name: "MOVBstore", argLength: 3, reg: gpstore, asm: "MOVB", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store byte in arg1 to arg0+auxint+aux. arg2=mem 399 {name: "MOVWstore", argLength: 3, reg: gpstore, asm: "MOVW", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store 2 bytes in arg1 to arg0+auxint+aux. arg2=mem 400 {name: "MOVLstore", argLength: 3, reg: gpstore, asm: "MOVL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store 4 bytes in arg1 to arg0+auxint+aux. arg2=mem 401 {name: "MOVQstore", argLength: 3, reg: gpstore, asm: "MOVQ", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store 8 bytes in arg1 to arg0+auxint+aux. arg2=mem 402 {name: "MOVOload", argLength: 2, reg: fpload, asm: "MOVUPS", aux: "SymOff", typ: "Int128", faultOnNilArg0: true, symEffect: "Read"}, // load 16 bytes from arg0+auxint+aux. arg1=mem 403 {name: "MOVOstore", argLength: 3, reg: fpstore, asm: "MOVUPS", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store 16 bytes in arg1 to arg0+auxint+aux. arg2=mem 404 405 // indexed loads/stores 406 {name: "MOVBloadidx1", argLength: 3, reg: gploadidx, commutative: true, asm: "MOVBLZX", aux: "SymOff", typ: "UInt8", symEffect: "Read"}, // load a byte from arg0+arg1+auxint+aux. arg2=mem 407 {name: "MOVWloadidx1", argLength: 3, reg: gploadidx, commutative: true, asm: "MOVWLZX", aux: "SymOff", typ: "UInt16", symEffect: "Read"}, // load 2 bytes from arg0+arg1+auxint+aux. arg2=mem 408 {name: "MOVWloadidx2", argLength: 3, reg: gploadidx, asm: "MOVWLZX", aux: "SymOff", typ: "UInt16", symEffect: "Read"}, // load 2 bytes from arg0+2*arg1+auxint+aux. arg2=mem 409 {name: "MOVLloadidx1", argLength: 3, reg: gploadidx, commutative: true, asm: "MOVL", aux: "SymOff", typ: "UInt32", symEffect: "Read"}, // load 4 bytes from arg0+arg1+auxint+aux. arg2=mem 410 {name: "MOVLloadidx4", argLength: 3, reg: gploadidx, asm: "MOVL", aux: "SymOff", typ: "UInt32", symEffect: "Read"}, // load 4 bytes from arg0+4*arg1+auxint+aux. arg2=mem 411 {name: "MOVQloadidx1", argLength: 3, reg: gploadidx, commutative: true, asm: "MOVQ", aux: "SymOff", typ: "UInt64", symEffect: "Read"}, // load 8 bytes from arg0+arg1+auxint+aux. arg2=mem 412 {name: "MOVQloadidx8", argLength: 3, reg: gploadidx, asm: "MOVQ", aux: "SymOff", typ: "UInt64", symEffect: "Read"}, // load 8 bytes from arg0+8*arg1+auxint+aux. arg2=mem 413 // TODO: sign-extending indexed loads 414 // TODO: mark the MOVXstoreidx1 ops as commutative. Generates too many rewrite rules at the moment. 415 {name: "MOVBstoreidx1", argLength: 4, reg: gpstoreidx, asm: "MOVB", aux: "SymOff", symEffect: "Write"}, // store byte in arg2 to arg0+arg1+auxint+aux. arg3=mem 416 {name: "MOVWstoreidx1", argLength: 4, reg: gpstoreidx, asm: "MOVW", aux: "SymOff", symEffect: "Write"}, // store 2 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem 417 {name: "MOVWstoreidx2", argLength: 4, reg: gpstoreidx, asm: "MOVW", aux: "SymOff", symEffect: "Write"}, // store 2 bytes in arg2 to arg0+2*arg1+auxint+aux. arg3=mem 418 {name: "MOVLstoreidx1", argLength: 4, reg: gpstoreidx, asm: "MOVL", aux: "SymOff", symEffect: "Write"}, // store 4 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem 419 {name: "MOVLstoreidx4", argLength: 4, reg: gpstoreidx, asm: "MOVL", aux: "SymOff", symEffect: "Write"}, // store 4 bytes in arg2 to arg0+4*arg1+auxint+aux. arg3=mem 420 {name: "MOVQstoreidx1", argLength: 4, reg: gpstoreidx, asm: "MOVQ", aux: "SymOff", symEffect: "Write"}, // store 8 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem 421 {name: "MOVQstoreidx8", argLength: 4, reg: gpstoreidx, asm: "MOVQ", aux: "SymOff", symEffect: "Write"}, // store 8 bytes in arg2 to arg0+8*arg1+auxint+aux. arg3=mem 422 // TODO: add size-mismatched indexed loads, like MOVBstoreidx4. 423 424 // For storeconst ops, the AuxInt field encodes both 425 // the value to store and an address offset of the store. 426 // Cast AuxInt to a ValAndOff to extract Val and Off fields. 427 {name: "MOVBstoreconst", argLength: 2, reg: gpstoreconst, asm: "MOVB", aux: "SymValAndOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store low byte of ValAndOff(AuxInt).Val() to arg0+ValAndOff(AuxInt).Off()+aux. arg1=mem 428 {name: "MOVWstoreconst", argLength: 2, reg: gpstoreconst, asm: "MOVW", aux: "SymValAndOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store low 2 bytes of ... 429 {name: "MOVLstoreconst", argLength: 2, reg: gpstoreconst, asm: "MOVL", aux: "SymValAndOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store low 4 bytes of ... 430 {name: "MOVQstoreconst", argLength: 2, reg: gpstoreconst, asm: "MOVQ", aux: "SymValAndOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store 8 bytes of ... 431 432 {name: "MOVBstoreconstidx1", argLength: 3, reg: gpstoreconstidx, asm: "MOVB", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store low byte of ValAndOff(AuxInt).Val() to arg0+1*arg1+ValAndOff(AuxInt).Off()+aux. arg2=mem 433 {name: "MOVWstoreconstidx1", argLength: 3, reg: gpstoreconstidx, asm: "MOVW", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store low 2 bytes of ... arg1 ... 434 {name: "MOVWstoreconstidx2", argLength: 3, reg: gpstoreconstidx, asm: "MOVW", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store low 2 bytes of ... 2*arg1 ... 435 {name: "MOVLstoreconstidx1", argLength: 3, reg: gpstoreconstidx, asm: "MOVL", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store low 4 bytes of ... arg1 ... 436 {name: "MOVLstoreconstidx4", argLength: 3, reg: gpstoreconstidx, asm: "MOVL", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store low 4 bytes of ... 4*arg1 ... 437 {name: "MOVQstoreconstidx1", argLength: 3, reg: gpstoreconstidx, asm: "MOVQ", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store 8 bytes of ... arg1 ... 438 {name: "MOVQstoreconstidx8", argLength: 3, reg: gpstoreconstidx, asm: "MOVQ", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store 8 bytes of ... 8*arg1 ... 439 440 // arg0 = pointer to start of memory to zero 441 // arg1 = value to store (will always be zero) 442 // arg2 = mem 443 // auxint = # of bytes to zero 444 // returns mem 445 { 446 name: "DUFFZERO", 447 aux: "Int64", 448 argLength: 3, 449 reg: regInfo{ 450 inputs: []regMask{buildReg("DI"), buildReg("X0")}, 451 clobbers: buildReg("DI"), 452 }, 453 clobberFlags: true, 454 faultOnNilArg0: true, 455 }, 456 {name: "MOVOconst", reg: regInfo{nil, 0, []regMask{fp}}, typ: "Int128", aux: "Int128", rematerializeable: true}, 457 458 // arg0 = address of memory to zero 459 // arg1 = # of 8-byte words to zero 460 // arg2 = value to store (will always be zero) 461 // arg3 = mem 462 // returns mem 463 { 464 name: "REPSTOSQ", 465 argLength: 4, 466 reg: regInfo{ 467 inputs: []regMask{buildReg("DI"), buildReg("CX"), buildReg("AX")}, 468 clobbers: buildReg("DI CX"), 469 }, 470 faultOnNilArg0: true, 471 }, 472 473 {name: "CALLstatic", argLength: 1, reg: regInfo{clobbers: callerSave}, aux: "SymOff", clobberFlags: true, call: true, symEffect: "None"}, // call static function aux.(*gc.Sym). arg0=mem, auxint=argsize, returns mem 474 {name: "CALLclosure", argLength: 3, reg: regInfo{inputs: []regMask{gpsp, buildReg("DX"), 0}, clobbers: callerSave}, aux: "Int64", clobberFlags: true, call: true}, // call function via closure. arg0=codeptr, arg1=closure, arg2=mem, auxint=argsize, returns mem 475 {name: "CALLinter", argLength: 2, reg: regInfo{inputs: []regMask{gp}, clobbers: callerSave}, aux: "Int64", clobberFlags: true, call: true}, // call fn by pointer. arg0=codeptr, arg1=mem, auxint=argsize, returns mem 476 477 // arg0 = destination pointer 478 // arg1 = source pointer 479 // arg2 = mem 480 // auxint = offset from duffcopy symbol to call 481 // returns memory 482 { 483 name: "DUFFCOPY", 484 aux: "Int64", 485 argLength: 3, 486 reg: regInfo{ 487 inputs: []regMask{buildReg("DI"), buildReg("SI")}, 488 clobbers: buildReg("DI SI X0"), // uses X0 as a temporary 489 }, 490 clobberFlags: true, 491 faultOnNilArg0: true, 492 faultOnNilArg1: true, 493 }, 494 495 // arg0 = destination pointer 496 // arg1 = source pointer 497 // arg2 = # of 8-byte words to copy 498 // arg3 = mem 499 // returns memory 500 { 501 name: "REPMOVSQ", 502 argLength: 4, 503 reg: regInfo{ 504 inputs: []regMask{buildReg("DI"), buildReg("SI"), buildReg("CX")}, 505 clobbers: buildReg("DI SI CX"), 506 }, 507 faultOnNilArg0: true, 508 faultOnNilArg1: true, 509 }, 510 511 // (InvertFlags (CMPQ a b)) == (CMPQ b a) 512 // So if we want (SETL (CMPQ a b)) but we can't do that because a is a constant, 513 // then we do (SETL (InvertFlags (CMPQ b a))) instead. 514 // Rewrites will convert this to (SETG (CMPQ b a)). 515 // InvertFlags is a pseudo-op which can't appear in assembly output. 516 {name: "InvertFlags", argLength: 1}, // reverse direction of arg0 517 518 // Pseudo-ops 519 {name: "LoweredGetG", argLength: 1, reg: gp01}, // arg0=mem 520 // Scheduler ensures LoweredGetClosurePtr occurs only in entry block, 521 // and sorts it to the very beginning of the block to prevent other 522 // use of DX (the closure pointer) 523 {name: "LoweredGetClosurePtr", reg: regInfo{outputs: []regMask{buildReg("DX")}}}, 524 //arg0=ptr,arg1=mem, returns void. Faults if ptr is nil. 525 {name: "LoweredNilCheck", argLength: 2, reg: regInfo{inputs: []regMask{gpsp}}, clobberFlags: true, nilCheck: true, faultOnNilArg0: true}, 526 527 // MOVQconvert converts between pointers and integers. 528 // We have a special op for this so as to not confuse GC 529 // (particularly stack maps). It takes a memory arg so it 530 // gets correctly ordered with respect to GC safepoints. 531 // arg0=ptr/int arg1=mem, output=int/ptr 532 {name: "MOVQconvert", argLength: 2, reg: gp11, asm: "MOVQ"}, 533 {name: "MOVLconvert", argLength: 2, reg: gp11, asm: "MOVL"}, // amd64p32 equivalent 534 535 // Constant flag values. For any comparison, there are 5 possible 536 // outcomes: the three from the signed total order (<,==,>) and the 537 // three from the unsigned total order. The == cases overlap. 538 // Note: there's a sixth "unordered" outcome for floating-point 539 // comparisons, but we don't use such a beast yet. 540 // These ops are for temporary use by rewrite rules. They 541 // cannot appear in the generated assembly. 542 {name: "FlagEQ"}, // equal 543 {name: "FlagLT_ULT"}, // signed < and unsigned < 544 {name: "FlagLT_UGT"}, // signed < and unsigned > 545 {name: "FlagGT_UGT"}, // signed > and unsigned < 546 {name: "FlagGT_ULT"}, // signed > and unsigned > 547 548 // Atomic loads. These are just normal loads but return <value,memory> tuples 549 // so they can be properly ordered with other loads. 550 // load from arg0+auxint+aux. arg1=mem. 551 {name: "MOVLatomicload", argLength: 2, reg: gpload, asm: "MOVL", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"}, 552 {name: "MOVQatomicload", argLength: 2, reg: gpload, asm: "MOVQ", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"}, 553 554 // Atomic stores and exchanges. Stores use XCHG to get the right memory ordering semantics. 555 // store arg0 to arg1+auxint+aux, arg2=mem. 556 // These ops return a tuple of <old contents of *(arg1+auxint+aux), memory>. 557 // Note: arg0 and arg1 are backwards compared to MOVLstore (to facilitate resultInArg0)! 558 {name: "XCHGL", argLength: 3, reg: gpstorexchg, asm: "XCHGL", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, hasSideEffects: true, symEffect: "RdWr"}, 559 {name: "XCHGQ", argLength: 3, reg: gpstorexchg, asm: "XCHGQ", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, hasSideEffects: true, symEffect: "RdWr"}, 560 561 // Atomic adds. 562 // *(arg1+auxint+aux) += arg0. arg2=mem. 563 // Returns a tuple of <old contents of *(arg1+auxint+aux), memory>. 564 // Note: arg0 and arg1 are backwards compared to MOVLstore (to facilitate resultInArg0)! 565 {name: "XADDLlock", argLength: 3, reg: gpstorexchg, asm: "XADDL", typ: "(UInt32,Mem)", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, hasSideEffects: true, symEffect: "RdWr"}, 566 {name: "XADDQlock", argLength: 3, reg: gpstorexchg, asm: "XADDQ", typ: "(UInt64,Mem)", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, hasSideEffects: true, symEffect: "RdWr"}, 567 {name: "AddTupleFirst32", argLength: 2}, // arg0=tuple <x,y>. Returns <x+arg1,y>. 568 {name: "AddTupleFirst64", argLength: 2}, // arg0=tuple <x,y>. Returns <x+arg1,y>. 569 570 // Compare and swap. 571 // arg0 = pointer, arg1 = old value, arg2 = new value, arg3 = memory. 572 // if *(arg0+auxint+aux) == arg1 { 573 // *(arg0+auxint+aux) = arg2 574 // return (true, memory) 575 // } else { 576 // return (false, memory) 577 // } 578 // Note that these instructions also return the old value in AX, but we ignore it. 579 // TODO: have these return flags instead of bool. The current system generates: 580 // CMPXCHGQ ... 581 // SETEQ AX 582 // CMPB AX, $0 583 // JNE ... 584 // instead of just 585 // CMPXCHGQ ... 586 // JEQ ... 587 // but we can't do that because memory-using ops can't generate flags yet 588 // (flagalloc wants to move flag-generating instructions around). 589 {name: "CMPXCHGLlock", argLength: 4, reg: cmpxchg, asm: "CMPXCHGL", aux: "SymOff", clobberFlags: true, faultOnNilArg0: true, hasSideEffects: true, symEffect: "RdWr"}, 590 {name: "CMPXCHGQlock", argLength: 4, reg: cmpxchg, asm: "CMPXCHGQ", aux: "SymOff", clobberFlags: true, faultOnNilArg0: true, hasSideEffects: true, symEffect: "RdWr"}, 591 592 // Atomic memory updates. 593 {name: "ANDBlock", argLength: 3, reg: gpstore, asm: "ANDB", aux: "SymOff", clobberFlags: true, faultOnNilArg0: true, hasSideEffects: true, symEffect: "RdWr"}, // *(arg0+auxint+aux) &= arg1 594 {name: "ORBlock", argLength: 3, reg: gpstore, asm: "ORB", aux: "SymOff", clobberFlags: true, faultOnNilArg0: true, hasSideEffects: true, symEffect: "RdWr"}, // *(arg0+auxint+aux) |= arg1 595 } 596 597 var AMD64blocks = []blockData{ 598 {name: "EQ"}, 599 {name: "NE"}, 600 {name: "LT"}, 601 {name: "LE"}, 602 {name: "GT"}, 603 {name: "GE"}, 604 {name: "ULT"}, 605 {name: "ULE"}, 606 {name: "UGT"}, 607 {name: "UGE"}, 608 {name: "EQF"}, 609 {name: "NEF"}, 610 {name: "ORD"}, // FP, ordered comparison (parity zero) 611 {name: "NAN"}, // FP, unordered comparison (parity one) 612 } 613 614 archs = append(archs, arch{ 615 name: "AMD64", 616 pkg: "cmd/internal/obj/x86", 617 genfile: "../../amd64/ssa.go", 618 ops: AMD64ops, 619 blocks: AMD64blocks, 620 regnames: regNamesAMD64, 621 gpregmask: gp, 622 fpregmask: fp, 623 framepointerreg: int8(num["BP"]), 624 linkreg: -1, // not used 625 }) 626 }