...

Source file src/pkg/cmd/compile/internal/ssa/gen/386Ops.go

     1	// Copyright 2016 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	// L (long word) = 32 bit
    26	// W (word)      = 16 bit
    27	// B (byte)      = 8 bit
    28	
    29	// copied from ../../x86/reg.go
    30	var regNames386 = []string{
    31		"AX",
    32		"CX",
    33		"DX",
    34		"BX",
    35		"SP",
    36		"BP",
    37		"SI",
    38		"DI",
    39		"X0",
    40		"X1",
    41		"X2",
    42		"X3",
    43		"X4",
    44		"X5",
    45		"X6",
    46		"X7",
    47	
    48		// pseudo-registers
    49		"SB",
    50	}
    51	
    52	// Notes on 387 support.
    53	//  - The 387 has a weird stack-register setup for floating-point registers.
    54	//    We use these registers when SSE registers are not available (when GO386=387).
    55	//  - We use the same register names (X0-X7) but they refer to the 387
    56	//    floating-point registers. That way, most of the SSA backend is unchanged.
    57	//  - The instruction generation pass maintains an SSE->387 register mapping.
    58	//    This mapping is updated whenever the FP stack is pushed or popped so that
    59	//    we can always find a given SSE register even when the TOS pointer has changed.
    60	//  - To facilitate the mapping from SSE to 387, we enforce that
    61	//    every basic block starts and ends with an empty floating-point stack.
    62	
    63	func init() {
    64		// Make map from reg names to reg integers.
    65		if len(regNames386) > 64 {
    66			panic("too many registers")
    67		}
    68		num := map[string]int{}
    69		for i, name := range regNames386 {
    70			num[name] = i
    71		}
    72		buildReg := func(s string) regMask {
    73			m := regMask(0)
    74			for _, r := range strings.Split(s, " ") {
    75				if n, ok := num[r]; ok {
    76					m |= regMask(1) << uint(n)
    77					continue
    78				}
    79				panic("register " + r + " not found")
    80			}
    81			return m
    82		}
    83	
    84		// Common individual register masks
    85		var (
    86			ax         = buildReg("AX")
    87			cx         = buildReg("CX")
    88			dx         = buildReg("DX")
    89			bx         = buildReg("BX")
    90			si         = buildReg("SI")
    91			gp         = buildReg("AX CX DX BX BP SI DI")
    92			fp         = buildReg("X0 X1 X2 X3 X4 X5 X6 X7")
    93			gpsp       = gp | buildReg("SP")
    94			gpspsb     = gpsp | buildReg("SB")
    95			callerSave = gp | fp
    96		)
    97		// Common slices of register masks
    98		var (
    99			gponly = []regMask{gp}
   100			fponly = []regMask{fp}
   101		)
   102	
   103		// Common regInfo
   104		var (
   105			gp01      = regInfo{inputs: nil, outputs: gponly}
   106			gp11      = regInfo{inputs: []regMask{gp}, outputs: gponly}
   107			gp11sp    = regInfo{inputs: []regMask{gpsp}, outputs: gponly}
   108			gp11sb    = regInfo{inputs: []regMask{gpspsb}, outputs: gponly}
   109			gp21      = regInfo{inputs: []regMask{gp, gp}, outputs: gponly}
   110			gp11carry = regInfo{inputs: []regMask{gp}, outputs: []regMask{gp, 0}}
   111			gp21carry = regInfo{inputs: []regMask{gp, gp}, outputs: []regMask{gp, 0}}
   112			gp1carry1 = regInfo{inputs: []regMask{gp}, outputs: gponly}
   113			gp2carry1 = 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}, clobbers: dx}
   118			gp21hmul  = regInfo{inputs: []regMask{ax, gpsp}, outputs: []regMask{dx}, clobbers: ax}
   119			gp11mod   = regInfo{inputs: []regMask{ax, gpsp &^ dx}, outputs: []regMask{dx}, clobbers: ax}
   120			gp21mul   = regInfo{inputs: []regMask{ax, gpsp}, outputs: []regMask{dx, ax}}
   121	
   122			gp2flags     = regInfo{inputs: []regMask{gpsp, gpsp}}
   123			gp1flags     = regInfo{inputs: []regMask{gpsp}}
   124			gp0flagsLoad = regInfo{inputs: []regMask{gpspsb, 0}}
   125			gp1flagsLoad = regInfo{inputs: []regMask{gpspsb, gpsp, 0}}
   126			flagsgp      = regInfo{inputs: nil, outputs: gponly}
   127	
   128			readflags = regInfo{inputs: nil, outputs: gponly}
   129			flagsgpax = regInfo{inputs: nil, clobbers: ax, outputs: []regMask{gp &^ ax}}
   130	
   131			gpload      = regInfo{inputs: []regMask{gpspsb, 0}, outputs: gponly}
   132			gp21load    = regInfo{inputs: []regMask{gp, gpspsb, 0}, outputs: gponly}
   133			gploadidx   = regInfo{inputs: []regMask{gpspsb, gpsp, 0}, outputs: gponly}
   134			gp21loadidx = regInfo{inputs: []regMask{gp, gpspsb, gpsp, 0}, outputs: gponly}
   135	
   136			gpstore         = regInfo{inputs: []regMask{gpspsb, gpsp, 0}}
   137			gpstoreconst    = regInfo{inputs: []regMask{gpspsb, 0}}
   138			gpstoreidx      = regInfo{inputs: []regMask{gpspsb, gpsp, gpsp, 0}}
   139			gpstoreconstidx = regInfo{inputs: []regMask{gpspsb, gpsp, 0}}
   140	
   141			fp01     = regInfo{inputs: nil, outputs: fponly}
   142			fp21     = regInfo{inputs: []regMask{fp, fp}, outputs: fponly}
   143			fp21load = regInfo{inputs: []regMask{fp, gpspsb, 0}, outputs: fponly}
   144			fpgp     = regInfo{inputs: fponly, outputs: gponly}
   145			gpfp     = regInfo{inputs: gponly, outputs: fponly}
   146			fp11     = regInfo{inputs: fponly, outputs: fponly}
   147			fp2flags = regInfo{inputs: []regMask{fp, fp}}
   148	
   149			fpload    = regInfo{inputs: []regMask{gpspsb, 0}, outputs: fponly}
   150			fploadidx = regInfo{inputs: []regMask{gpspsb, gpsp, 0}, outputs: fponly}
   151	
   152			fpstore    = regInfo{inputs: []regMask{gpspsb, fp, 0}}
   153			fpstoreidx = regInfo{inputs: []regMask{gpspsb, gpsp, fp, 0}}
   154		)
   155	
   156		var _386ops = []opData{
   157			// fp ops
   158			{name: "ADDSS", argLength: 2, reg: fp21, asm: "ADDSS", commutative: true, resultInArg0: true, usesScratch: true}, // fp32 add
   159			{name: "ADDSD", argLength: 2, reg: fp21, asm: "ADDSD", commutative: true, resultInArg0: true},                    // fp64 add
   160			{name: "SUBSS", argLength: 2, reg: fp21, asm: "SUBSS", resultInArg0: true, usesScratch: true},                    // fp32 sub
   161			{name: "SUBSD", argLength: 2, reg: fp21, asm: "SUBSD", resultInArg0: true},                                       // fp64 sub
   162			{name: "MULSS", argLength: 2, reg: fp21, asm: "MULSS", commutative: true, resultInArg0: true, usesScratch: true}, // fp32 mul
   163			{name: "MULSD", argLength: 2, reg: fp21, asm: "MULSD", commutative: true, resultInArg0: true},                    // fp64 mul
   164			{name: "DIVSS", argLength: 2, reg: fp21, asm: "DIVSS", resultInArg0: true, usesScratch: true},                    // fp32 div
   165			{name: "DIVSD", argLength: 2, reg: fp21, asm: "DIVSD", resultInArg0: true},                                       // fp64 div
   166	
   167			{name: "MOVSSload", argLength: 2, reg: fpload, asm: "MOVSS", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"}, // fp32 load
   168			{name: "MOVSDload", argLength: 2, reg: fpload, asm: "MOVSD", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"}, // fp64 load
   169			{name: "MOVSSconst", reg: fp01, asm: "MOVSS", aux: "Float32", rematerializeable: true},                               // fp32 constant
   170			{name: "MOVSDconst", reg: fp01, asm: "MOVSD", aux: "Float64", rematerializeable: true},                               // fp64 constant
   171			{name: "MOVSSloadidx1", argLength: 3, reg: fploadidx, asm: "MOVSS", aux: "SymOff", symEffect: "Read"},                // fp32 load indexed by i
   172			{name: "MOVSSloadidx4", argLength: 3, reg: fploadidx, asm: "MOVSS", aux: "SymOff", symEffect: "Read"},                // fp32 load indexed by 4*i
   173			{name: "MOVSDloadidx1", argLength: 3, reg: fploadidx, asm: "MOVSD", aux: "SymOff", symEffect: "Read"},                // fp64 load indexed by i
   174			{name: "MOVSDloadidx8", argLength: 3, reg: fploadidx, asm: "MOVSD", aux: "SymOff", symEffect: "Read"},                // fp64 load indexed by 8*i
   175	
   176			{name: "MOVSSstore", argLength: 3, reg: fpstore, asm: "MOVSS", aux: "SymOff", faultOnNilArg0: true, symEffect: "Write"}, // fp32 store
   177			{name: "MOVSDstore", argLength: 3, reg: fpstore, asm: "MOVSD", aux: "SymOff", faultOnNilArg0: true, symEffect: "Write"}, // fp64 store
   178			{name: "MOVSSstoreidx1", argLength: 4, reg: fpstoreidx, asm: "MOVSS", aux: "SymOff", symEffect: "Write"},                // fp32 indexed by i store
   179			{name: "MOVSSstoreidx4", argLength: 4, reg: fpstoreidx, asm: "MOVSS", aux: "SymOff", symEffect: "Write"},                // fp32 indexed by 4i store
   180			{name: "MOVSDstoreidx1", argLength: 4, reg: fpstoreidx, asm: "MOVSD", aux: "SymOff", symEffect: "Write"},                // fp64 indexed by i store
   181			{name: "MOVSDstoreidx8", argLength: 4, reg: fpstoreidx, asm: "MOVSD", aux: "SymOff", symEffect: "Write"},                // fp64 indexed by 8i store
   182	
   183			{name: "ADDSSload", 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: "ADDSDload", argLength: 3, reg: fp21load, asm: "ADDSD", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, symEffect: "Read"}, // fp64 arg0 + tmp, tmp loaded from arg1+auxint+aux, arg2 = mem
   185			{name: "SUBSSload", 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
   186			{name: "SUBSDload", 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
   187			{name: "MULSSload", 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
   188			{name: "MULSDload", 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
   189			{name: "DIVSSload", argLength: 3, reg: fp21load, asm: "DIVSS", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, symEffect: "Read"}, // fp32 arg0 / tmp, tmp loaded from arg1+auxint+aux, arg2 = mem
   190			{name: "DIVSDload", argLength: 3, reg: fp21load, asm: "DIVSD", aux: "SymOff", resultInArg0: true, faultOnNilArg1: true, symEffect: "Read"}, // fp64 arg0 / tmp, tmp loaded from arg1+auxint+aux, arg2 = mem
   191	
   192			// binary ops
   193			{name: "ADDL", argLength: 2, reg: gp21sp, asm: "ADDL", commutative: true, clobberFlags: true},                // arg0 + arg1
   194			{name: "ADDLconst", argLength: 1, reg: gp11sp, asm: "ADDL", aux: "Int32", typ: "UInt32", clobberFlags: true}, // arg0 + auxint
   195	
   196			{name: "ADDLcarry", argLength: 2, reg: gp21carry, asm: "ADDL", commutative: true, resultInArg0: true},                // arg0 + arg1, generates <carry,result> pair
   197			{name: "ADDLconstcarry", argLength: 1, reg: gp11carry, asm: "ADDL", aux: "Int32", resultInArg0: true},                // arg0 + auxint, generates <carry,result> pair
   198			{name: "ADCL", argLength: 3, reg: gp2carry1, asm: "ADCL", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0+arg1+carry(arg2), where arg2 is flags
   199			{name: "ADCLconst", argLength: 2, reg: gp1carry1, asm: "ADCL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0+auxint+carry(arg1), where arg1 is flags
   200	
   201			{name: "SUBL", argLength: 2, reg: gp21, asm: "SUBL", resultInArg0: true, clobberFlags: true},                    // arg0 - arg1
   202			{name: "SUBLconst", argLength: 1, reg: gp11, asm: "SUBL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 - auxint
   203	
   204			{name: "SUBLcarry", argLength: 2, reg: gp21carry, asm: "SUBL", resultInArg0: true},                                   // arg0-arg1, generates <borrow,result> pair
   205			{name: "SUBLconstcarry", argLength: 1, reg: gp11carry, asm: "SUBL", aux: "Int32", resultInArg0: true},                // arg0-auxint, generates <borrow,result> pair
   206			{name: "SBBL", argLength: 3, reg: gp2carry1, asm: "SBBL", resultInArg0: true, clobberFlags: true},                    // arg0-arg1-borrow(arg2), where arg2 is flags
   207			{name: "SBBLconst", argLength: 2, reg: gp1carry1, asm: "SBBL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0-auxint-borrow(arg1), where arg1 is flags
   208	
   209			{name: "MULL", argLength: 2, reg: gp21, asm: "IMULL", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 * arg1
   210			{name: "MULLconst", argLength: 1, reg: gp11, asm: "IMUL3L", aux: "Int32", clobberFlags: true},                    // arg0 * auxint
   211	
   212			{name: "MULLU", argLength: 2, reg: regInfo{inputs: []regMask{ax, gpsp}, outputs: []regMask{ax, 0}, clobbers: dx}, typ: "(UInt32,Flags)", asm: "MULL", commutative: true, clobberFlags: true}, // Let x = arg0*arg1 (full 32x32->64  unsigned multiply). Returns uint32(x), and flags set to overflow if uint32(x) != x.
   213	
   214			{name: "HMULL", argLength: 2, reg: gp21hmul, commutative: true, asm: "IMULL", clobberFlags: true}, // (arg0 * arg1) >> width
   215			{name: "HMULLU", argLength: 2, reg: gp21hmul, commutative: true, asm: "MULL", clobberFlags: true}, // (arg0 * arg1) >> width
   216	
   217			{name: "MULLQU", argLength: 2, reg: gp21mul, commutative: true, asm: "MULL", clobberFlags: true}, // arg0 * arg1, high 32 in result[0], low 32 in result[1]
   218	
   219			{name: "AVGLU", argLength: 2, reg: gp21, commutative: true, resultInArg0: true, clobberFlags: true}, // (arg0 + arg1) / 2 as unsigned, all 32 result bits
   220	
   221			// For DIVL, DIVW, MODL and MODW, AuxInt non-zero means that the divisor has been proved to be not -1.
   222			{name: "DIVL", argLength: 2, reg: gp11div, asm: "IDIVL", aux: "Bool", clobberFlags: true}, // arg0 / arg1
   223			{name: "DIVW", argLength: 2, reg: gp11div, asm: "IDIVW", aux: "Bool", clobberFlags: true}, // arg0 / arg1
   224			{name: "DIVLU", argLength: 2, reg: gp11div, asm: "DIVL", clobberFlags: true},              // arg0 / arg1
   225			{name: "DIVWU", argLength: 2, reg: gp11div, asm: "DIVW", clobberFlags: true},              // arg0 / arg1
   226	
   227			{name: "MODL", argLength: 2, reg: gp11mod, asm: "IDIVL", aux: "Bool", clobberFlags: true}, // arg0 % arg1
   228			{name: "MODW", argLength: 2, reg: gp11mod, asm: "IDIVW", aux: "Bool", clobberFlags: true}, // arg0 % arg1
   229			{name: "MODLU", argLength: 2, reg: gp11mod, asm: "DIVL", clobberFlags: true},              // arg0 % arg1
   230			{name: "MODWU", argLength: 2, reg: gp11mod, asm: "DIVW", clobberFlags: true},              // arg0 % arg1
   231	
   232			{name: "ANDL", argLength: 2, reg: gp21, asm: "ANDL", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 & arg1
   233			{name: "ANDLconst", argLength: 1, reg: gp11, asm: "ANDL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 & auxint
   234	
   235			{name: "ORL", argLength: 2, reg: gp21, asm: "ORL", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 | arg1
   236			{name: "ORLconst", argLength: 1, reg: gp11, asm: "ORL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 | auxint
   237	
   238			{name: "XORL", argLength: 2, reg: gp21, asm: "XORL", commutative: true, resultInArg0: true, clobberFlags: true}, // arg0 ^ arg1
   239			{name: "XORLconst", argLength: 1, reg: gp11, asm: "XORL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 ^ auxint
   240	
   241			{name: "CMPL", argLength: 2, reg: gp2flags, asm: "CMPL", typ: "Flags"},                    // arg0 compare to arg1
   242			{name: "CMPW", argLength: 2, reg: gp2flags, asm: "CMPW", typ: "Flags"},                    // arg0 compare to arg1
   243			{name: "CMPB", argLength: 2, reg: gp2flags, asm: "CMPB", typ: "Flags"},                    // arg0 compare to arg1
   244			{name: "CMPLconst", argLength: 1, reg: gp1flags, asm: "CMPL", typ: "Flags", aux: "Int32"}, // arg0 compare to auxint
   245			{name: "CMPWconst", argLength: 1, reg: gp1flags, asm: "CMPW", typ: "Flags", aux: "Int16"}, // arg0 compare to auxint
   246			{name: "CMPBconst", argLength: 1, reg: gp1flags, asm: "CMPB", typ: "Flags", aux: "Int8"},  // arg0 compare to auxint
   247	
   248			// compare *(arg0+auxint+aux) to arg1 (in that order). arg2=mem.
   249			{name: "CMPLload", argLength: 3, reg: gp1flagsLoad, asm: "CMPL", aux: "SymOff", typ: "Flags", symEffect: "Read", faultOnNilArg0: true},
   250			{name: "CMPWload", argLength: 3, reg: gp1flagsLoad, asm: "CMPW", aux: "SymOff", typ: "Flags", symEffect: "Read", faultOnNilArg0: true},
   251			{name: "CMPBload", argLength: 3, reg: gp1flagsLoad, asm: "CMPB", aux: "SymOff", typ: "Flags", symEffect: "Read", faultOnNilArg0: true},
   252	
   253			// compare *(arg0+ValAndOff(AuxInt).Off()+aux) to ValAndOff(AuxInt).Val() (in that order). arg1=mem.
   254			{name: "CMPLconstload", argLength: 2, reg: gp0flagsLoad, asm: "CMPL", aux: "SymValAndOff", typ: "Flags", symEffect: "Read", faultOnNilArg0: true},
   255			{name: "CMPWconstload", argLength: 2, reg: gp0flagsLoad, asm: "CMPW", aux: "SymValAndOff", typ: "Flags", symEffect: "Read", faultOnNilArg0: true},
   256			{name: "CMPBconstload", argLength: 2, reg: gp0flagsLoad, asm: "CMPB", aux: "SymValAndOff", typ: "Flags", symEffect: "Read", faultOnNilArg0: true},
   257	
   258			{name: "UCOMISS", argLength: 2, reg: fp2flags, asm: "UCOMISS", typ: "Flags", usesScratch: true}, // arg0 compare to arg1, f32
   259			{name: "UCOMISD", argLength: 2, reg: fp2flags, asm: "UCOMISD", typ: "Flags", usesScratch: true}, // arg0 compare to arg1, f64
   260	
   261			{name: "TESTL", argLength: 2, reg: gp2flags, commutative: true, asm: "TESTL", typ: "Flags"}, // (arg0 & arg1) compare to 0
   262			{name: "TESTW", argLength: 2, reg: gp2flags, commutative: true, asm: "TESTW", typ: "Flags"}, // (arg0 & arg1) compare to 0
   263			{name: "TESTB", argLength: 2, reg: gp2flags, commutative: true, asm: "TESTB", typ: "Flags"}, // (arg0 & arg1) compare to 0
   264			{name: "TESTLconst", argLength: 1, reg: gp1flags, asm: "TESTL", typ: "Flags", aux: "Int32"}, // (arg0 & auxint) compare to 0
   265			{name: "TESTWconst", argLength: 1, reg: gp1flags, asm: "TESTW", typ: "Flags", aux: "Int16"}, // (arg0 & auxint) compare to 0
   266			{name: "TESTBconst", argLength: 1, reg: gp1flags, asm: "TESTB", typ: "Flags", aux: "Int8"},  // (arg0 & auxint) compare to 0
   267	
   268			{name: "SHLL", argLength: 2, reg: gp21shift, asm: "SHLL", resultInArg0: true, clobberFlags: true},               // arg0 << arg1, shift amount is mod 32
   269			{name: "SHLLconst", argLength: 1, reg: gp11, asm: "SHLL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 << auxint, shift amount 0-31
   270			// Note: x86 is weird, the 16 and 8 byte shifts still use all 5 bits of shift amount!
   271	
   272			{name: "SHRL", argLength: 2, reg: gp21shift, asm: "SHRL", resultInArg0: true, clobberFlags: true},               // unsigned arg0 >> arg1, shift amount is mod 32
   273			{name: "SHRW", argLength: 2, reg: gp21shift, asm: "SHRW", resultInArg0: true, clobberFlags: true},               // unsigned arg0 >> arg1, shift amount is mod 32
   274			{name: "SHRB", argLength: 2, reg: gp21shift, asm: "SHRB", resultInArg0: true, clobberFlags: true},               // unsigned arg0 >> arg1, shift amount is mod 32
   275			{name: "SHRLconst", argLength: 1, reg: gp11, asm: "SHRL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // unsigned arg0 >> auxint, shift amount 0-31
   276			{name: "SHRWconst", argLength: 1, reg: gp11, asm: "SHRW", aux: "Int16", resultInArg0: true, clobberFlags: true}, // unsigned arg0 >> auxint, shift amount 0-15
   277			{name: "SHRBconst", argLength: 1, reg: gp11, asm: "SHRB", aux: "Int8", resultInArg0: true, clobberFlags: true},  // unsigned arg0 >> auxint, shift amount 0-7
   278	
   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: "SARLconst", argLength: 1, reg: gp11, asm: "SARL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // signed arg0 >> auxint, shift amount 0-31
   283			{name: "SARWconst", argLength: 1, reg: gp11, asm: "SARW", aux: "Int16", resultInArg0: true, clobberFlags: true}, // signed arg0 >> auxint, shift amount 0-15
   284			{name: "SARBconst", argLength: 1, reg: gp11, asm: "SARB", aux: "Int8", resultInArg0: true, clobberFlags: true},  // signed arg0 >> auxint, shift amount 0-7
   285	
   286			{name: "ROLLconst", argLength: 1, reg: gp11, asm: "ROLL", aux: "Int32", resultInArg0: true, clobberFlags: true}, // arg0 rotate left auxint, rotate amount 0-31
   287			{name: "ROLWconst", argLength: 1, reg: gp11, asm: "ROLW", aux: "Int16", resultInArg0: true, clobberFlags: true}, // arg0 rotate left auxint, rotate amount 0-15
   288			{name: "ROLBconst", argLength: 1, reg: gp11, asm: "ROLB", aux: "Int8", resultInArg0: true, clobberFlags: true},  // arg0 rotate left auxint, rotate amount 0-7
   289	
   290			// binary-op with a memory source operand
   291			{name: "ADDLload", 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
   292			{name: "SUBLload", 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
   293			{name: "MULLload", argLength: 3, reg: gp21load, asm: "IMULL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 * tmp, tmp loaded from  arg1+auxint+aux, arg2 = mem
   294			{name: "ANDLload", 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
   295			{name: "ORLload", 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
   296			{name: "XORLload", 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
   297	
   298			// binary-op with an indexed memory source operand
   299			{name: "ADDLloadidx4", argLength: 4, reg: gp21loadidx, asm: "ADDL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"},  // arg0 + tmp, tmp loaded from  arg1+arg2*4+auxint+aux, arg3 = mem
   300			{name: "SUBLloadidx4", argLength: 4, reg: gp21loadidx, asm: "SUBL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"},  // arg0 - tmp, tmp loaded from  arg1+arg2*4+auxint+aux, arg3 = mem
   301			{name: "MULLloadidx4", argLength: 4, reg: gp21loadidx, asm: "IMULL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"}, // arg0 * tmp, tmp loaded from  arg1+arg2*4+auxint+aux, arg3 = mem
   302			{name: "ANDLloadidx4", argLength: 4, reg: gp21loadidx, asm: "ANDL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"},  // arg0 & tmp, tmp loaded from  arg1+arg2*4+auxint+aux, arg3 = mem
   303			{name: "ORLloadidx4", argLength: 4, reg: gp21loadidx, asm: "ORL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"},    // arg0 | tmp, tmp loaded from  arg1+arg2*4+auxint+aux, arg3 = mem
   304			{name: "XORLloadidx4", argLength: 4, reg: gp21loadidx, asm: "XORL", aux: "SymOff", resultInArg0: true, clobberFlags: true, faultOnNilArg1: true, symEffect: "Read"},  // arg0 ^ tmp, tmp loaded from  arg1+arg2*4+auxint+aux, arg3 = mem
   305	
   306			// unary ops
   307			{name: "NEGL", argLength: 1, reg: gp11, asm: "NEGL", resultInArg0: true, clobberFlags: true}, // -arg0
   308	
   309			{name: "NOTL", argLength: 1, reg: gp11, asm: "NOTL", resultInArg0: true, clobberFlags: true}, // ^arg0
   310	
   311			{name: "BSFL", argLength: 1, reg: gp11, asm: "BSFL", clobberFlags: true}, // arg0 # of low-order zeroes ; undef if zero
   312			{name: "BSFW", argLength: 1, reg: gp11, asm: "BSFW", clobberFlags: true}, // arg0 # of low-order zeroes ; undef if zero
   313	
   314			{name: "BSRL", argLength: 1, reg: gp11, asm: "BSRL", clobberFlags: true}, // arg0 # of high-order zeroes ; undef if zero
   315			{name: "BSRW", argLength: 1, reg: gp11, asm: "BSRW", clobberFlags: true}, // arg0 # of high-order zeroes ; undef if zero
   316	
   317			{name: "BSWAPL", argLength: 1, reg: gp11, asm: "BSWAPL", resultInArg0: true, clobberFlags: true}, // arg0 swap bytes
   318	
   319			{name: "SQRTSD", argLength: 1, reg: fp11, asm: "SQRTSD"}, // sqrt(arg0)
   320	
   321			{name: "SBBLcarrymask", argLength: 1, reg: flagsgp, asm: "SBBL"}, // (int32)(-1) if carry is set, 0 if carry is clear.
   322			// Note: SBBW and SBBB are subsumed by SBBL
   323	
   324			{name: "SETEQ", argLength: 1, reg: readflags, asm: "SETEQ"}, // extract == condition from arg0
   325			{name: "SETNE", argLength: 1, reg: readflags, asm: "SETNE"}, // extract != condition from arg0
   326			{name: "SETL", argLength: 1, reg: readflags, asm: "SETLT"},  // extract signed < condition from arg0
   327			{name: "SETLE", argLength: 1, reg: readflags, asm: "SETLE"}, // extract signed <= condition from arg0
   328			{name: "SETG", argLength: 1, reg: readflags, asm: "SETGT"},  // extract signed > condition from arg0
   329			{name: "SETGE", argLength: 1, reg: readflags, asm: "SETGE"}, // extract signed >= condition from arg0
   330			{name: "SETB", argLength: 1, reg: readflags, asm: "SETCS"},  // extract unsigned < condition from arg0
   331			{name: "SETBE", argLength: 1, reg: readflags, asm: "SETLS"}, // extract unsigned <= condition from arg0
   332			{name: "SETA", argLength: 1, reg: readflags, asm: "SETHI"},  // extract unsigned > condition from arg0
   333			{name: "SETAE", argLength: 1, reg: readflags, asm: "SETCC"}, // extract unsigned >= condition from arg0
   334			{name: "SETO", argLength: 1, reg: readflags, asm: "SETOS"},  // extract if overflow flag is set from arg0
   335			// Need different opcodes for floating point conditions because
   336			// any comparison involving a NaN is always FALSE and thus
   337			// the patterns for inverting conditions cannot be used.
   338			{name: "SETEQF", argLength: 1, reg: flagsgpax, asm: "SETEQ", clobberFlags: true}, // extract == condition from arg0
   339			{name: "SETNEF", argLength: 1, reg: flagsgpax, asm: "SETNE", clobberFlags: true}, // extract != condition from arg0
   340			{name: "SETORD", argLength: 1, reg: flagsgp, asm: "SETPC"},                       // extract "ordered" (No Nan present) condition from arg0
   341			{name: "SETNAN", argLength: 1, reg: flagsgp, asm: "SETPS"},                       // extract "unordered" (Nan present) condition from arg0
   342	
   343			{name: "SETGF", argLength: 1, reg: flagsgp, asm: "SETHI"},  // extract floating > condition from arg0
   344			{name: "SETGEF", argLength: 1, reg: flagsgp, asm: "SETCC"}, // extract floating >= condition from arg0
   345	
   346			{name: "MOVBLSX", argLength: 1, reg: gp11, asm: "MOVBLSX"}, // sign extend arg0 from int8 to int32
   347			{name: "MOVBLZX", argLength: 1, reg: gp11, asm: "MOVBLZX"}, // zero extend arg0 from int8 to int32
   348			{name: "MOVWLSX", argLength: 1, reg: gp11, asm: "MOVWLSX"}, // sign extend arg0 from int16 to int32
   349			{name: "MOVWLZX", argLength: 1, reg: gp11, asm: "MOVWLZX"}, // zero extend arg0 from int16 to int32
   350	
   351			{name: "MOVLconst", reg: gp01, asm: "MOVL", typ: "UInt32", aux: "Int32", rematerializeable: true}, // 32 low bits of auxint
   352	
   353			{name: "CVTTSD2SL", argLength: 1, reg: fpgp, asm: "CVTTSD2SL", usesScratch: true}, // convert float64 to int32
   354			{name: "CVTTSS2SL", argLength: 1, reg: fpgp, asm: "CVTTSS2SL", usesScratch: true}, // convert float32 to int32
   355			{name: "CVTSL2SS", argLength: 1, reg: gpfp, asm: "CVTSL2SS", usesScratch: true},   // convert int32 to float32
   356			{name: "CVTSL2SD", argLength: 1, reg: gpfp, asm: "CVTSL2SD", usesScratch: true},   // convert int32 to float64
   357			{name: "CVTSD2SS", argLength: 1, reg: fp11, asm: "CVTSD2SS", usesScratch: true},   // convert float64 to float32
   358			{name: "CVTSS2SD", argLength: 1, reg: fp11, asm: "CVTSS2SD"},                      // convert float32 to float64
   359	
   360			{name: "PXOR", argLength: 2, reg: fp21, asm: "PXOR", commutative: true, resultInArg0: true}, // exclusive or, applied to X regs for float negation.
   361	
   362			{name: "LEAL", argLength: 1, reg: gp11sb, aux: "SymOff", rematerializeable: true, symEffect: "Addr"}, // arg0 + auxint + offset encoded in aux
   363			{name: "LEAL1", argLength: 2, reg: gp21sb, commutative: true, aux: "SymOff", symEffect: "Addr"},      // arg0 + arg1 + auxint + aux
   364			{name: "LEAL2", argLength: 2, reg: gp21sb, aux: "SymOff", symEffect: "Addr"},                         // arg0 + 2*arg1 + auxint + aux
   365			{name: "LEAL4", argLength: 2, reg: gp21sb, aux: "SymOff", symEffect: "Addr"},                         // arg0 + 4*arg1 + auxint + aux
   366			{name: "LEAL8", argLength: 2, reg: gp21sb, aux: "SymOff", symEffect: "Addr"},                         // arg0 + 8*arg1 + auxint + aux
   367			// Note: LEAL{1,2,4,8} must not have OpSB as either argument.
   368	
   369			// auxint+aux == add auxint and the offset of the symbol in aux (if any) to the effective address
   370			{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.
   371			{name: "MOVBLSXload", argLength: 2, reg: gpload, asm: "MOVBLSX", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"},             // ditto, sign extend to int32
   372			{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.
   373			{name: "MOVWLSXload", argLength: 2, reg: gpload, asm: "MOVWLSX", aux: "SymOff", faultOnNilArg0: true, symEffect: "Read"},             // ditto, sign extend to int32
   374			{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.
   375			{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
   376			{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
   377			{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
   378	
   379			// direct binary-op on memory (read-modify-write)
   380			{name: "ADDLmodify", argLength: 3, reg: gpstore, asm: "ADDL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, clobberFlags: true, symEffect: "Read,Write"}, // *(arg0+auxint+aux) += arg1, arg2=mem
   381			{name: "SUBLmodify", argLength: 3, reg: gpstore, asm: "SUBL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, clobberFlags: true, symEffect: "Read,Write"}, // *(arg0+auxint+aux) -= arg1, arg2=mem
   382			{name: "ANDLmodify", argLength: 3, reg: gpstore, asm: "ANDL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, clobberFlags: true, symEffect: "Read,Write"}, // *(arg0+auxint+aux) &= arg1, arg2=mem
   383			{name: "ORLmodify", argLength: 3, reg: gpstore, asm: "ORL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, clobberFlags: true, symEffect: "Read,Write"},   // *(arg0+auxint+aux) |= arg1, arg2=mem
   384			{name: "XORLmodify", argLength: 3, reg: gpstore, asm: "XORL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, clobberFlags: true, symEffect: "Read,Write"}, // *(arg0+auxint+aux) ^= arg1, arg2=mem
   385	
   386			// direct binary-op on indexed memory (read-modify-write)
   387			{name: "ADDLmodifyidx4", argLength: 4, reg: gpstoreidx, asm: "ADDL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, clobberFlags: true, symEffect: "Read,Write"}, // *(arg0+arg1*4+auxint+aux) += arg2, arg3=mem
   388			{name: "SUBLmodifyidx4", argLength: 4, reg: gpstoreidx, asm: "SUBL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, clobberFlags: true, symEffect: "Read,Write"}, // *(arg0+arg1*4+auxint+aux) -= arg2, arg3=mem
   389			{name: "ANDLmodifyidx4", argLength: 4, reg: gpstoreidx, asm: "ANDL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, clobberFlags: true, symEffect: "Read,Write"}, // *(arg0+arg1*4+auxint+aux) &= arg2, arg3=mem
   390			{name: "ORLmodifyidx4", argLength: 4, reg: gpstoreidx, asm: "ORL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, clobberFlags: true, symEffect: "Read,Write"},   // *(arg0+arg1*4+auxint+aux) |= arg2, arg3=mem
   391			{name: "XORLmodifyidx4", argLength: 4, reg: gpstoreidx, asm: "XORL", aux: "SymOff", typ: "Mem", faultOnNilArg0: true, clobberFlags: true, symEffect: "Read,Write"}, // *(arg0+arg1*4+auxint+aux) ^= arg2, arg3=mem
   392	
   393			// direct binary-op on memory with a constant (read-modify-write)
   394			{name: "ADDLconstmodify", argLength: 2, reg: gpstoreconst, asm: "ADDL", aux: "SymValAndOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true, symEffect: "Read,Write"}, // add ValAndOff(AuxInt).Val() to arg0+ValAndOff(AuxInt).Off()+aux, arg1=mem
   395			{name: "ANDLconstmodify", argLength: 2, reg: gpstoreconst, asm: "ANDL", aux: "SymValAndOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true, symEffect: "Read,Write"}, // and ValAndOff(AuxInt).Val() to arg0+ValAndOff(AuxInt).Off()+aux, arg1=mem
   396			{name: "ORLconstmodify", argLength: 2, reg: gpstoreconst, asm: "ORL", aux: "SymValAndOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true, symEffect: "Read,Write"},   // or  ValAndOff(AuxInt).Val() to arg0+ValAndOff(AuxInt).Off()+aux, arg1=mem
   397			{name: "XORLconstmodify", argLength: 2, reg: gpstoreconst, asm: "XORL", aux: "SymValAndOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true, symEffect: "Read,Write"}, // xor ValAndOff(AuxInt).Val() to arg0+ValAndOff(AuxInt).Off()+aux, arg1=mem
   398	
   399			// direct binary-op on indexed memory with a constant (read-modify-write)
   400			{name: "ADDLconstmodifyidx4", argLength: 3, reg: gpstoreconstidx, asm: "ADDL", aux: "SymValAndOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true, symEffect: "Read,Write"}, // add ValAndOff(AuxInt).Val() to arg0+arg1*4+ValAndOff(AuxInt).Off()+aux, arg2=mem
   401			{name: "ANDLconstmodifyidx4", argLength: 3, reg: gpstoreconstidx, asm: "ANDL", aux: "SymValAndOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true, symEffect: "Read,Write"}, // and ValAndOff(AuxInt).Val() to arg0+arg1*4+ValAndOff(AuxInt).Off()+aux, arg2=mem
   402			{name: "ORLconstmodifyidx4", argLength: 3, reg: gpstoreconstidx, asm: "ORL", aux: "SymValAndOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true, symEffect: "Read,Write"},   // or  ValAndOff(AuxInt).Val() to arg0+arg1*4+ValAndOff(AuxInt).Off()+aux, arg2=mem
   403			{name: "XORLconstmodifyidx4", argLength: 3, reg: gpstoreconstidx, asm: "XORL", aux: "SymValAndOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true, symEffect: "Read,Write"}, // xor ValAndOff(AuxInt).Val() to arg0+arg1*4+ValAndOff(AuxInt).Off()+aux, arg2=mem
   404	
   405			// indexed loads/stores
   406			{name: "MOVBloadidx1", argLength: 3, reg: gploadidx, commutative: true, asm: "MOVBLZX", aux: "SymOff", 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", symEffect: "Read"}, // load 2 bytes from arg0+arg1+auxint+aux. arg2=mem
   408			{name: "MOVWloadidx2", argLength: 3, reg: gploadidx, asm: "MOVWLZX", aux: "SymOff", 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", symEffect: "Read"},    // load 4 bytes from arg0+arg1+auxint+aux. arg2=mem
   410			{name: "MOVLloadidx4", argLength: 3, reg: gploadidx, asm: "MOVL", aux: "SymOff", symEffect: "Read"},                       // load 4 bytes from arg0+4*arg1+auxint+aux. arg2=mem
   411			// TODO: sign-extending indexed loads
   412			{name: "MOVBstoreidx1", argLength: 4, reg: gpstoreidx, commutative: true, asm: "MOVB", aux: "SymOff", symEffect: "Write"}, // store byte in arg2 to arg0+arg1+auxint+aux. arg3=mem
   413			{name: "MOVWstoreidx1", argLength: 4, reg: gpstoreidx, commutative: true, asm: "MOVW", aux: "SymOff", symEffect: "Write"}, // store 2 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem
   414			{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
   415			{name: "MOVLstoreidx1", argLength: 4, reg: gpstoreidx, commutative: true, asm: "MOVL", aux: "SymOff", symEffect: "Write"}, // store 4 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem
   416			{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
   417			// TODO: add size-mismatched indexed loads, like MOVBstoreidx4.
   418	
   419			// For storeconst ops, the AuxInt field encodes both
   420			// the value to store and an address offset of the store.
   421			// Cast AuxInt to a ValAndOff to extract Val and Off fields.
   422			{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
   423			{name: "MOVWstoreconst", argLength: 2, reg: gpstoreconst, asm: "MOVW", aux: "SymValAndOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store low 2 bytes of ...
   424			{name: "MOVLstoreconst", argLength: 2, reg: gpstoreconst, asm: "MOVL", aux: "SymValAndOff", typ: "Mem", faultOnNilArg0: true, symEffect: "Write"}, // store low 4 bytes of ...
   425	
   426			{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
   427			{name: "MOVWstoreconstidx1", argLength: 3, reg: gpstoreconstidx, asm: "MOVW", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store low 2 bytes of ... arg1 ...
   428			{name: "MOVWstoreconstidx2", argLength: 3, reg: gpstoreconstidx, asm: "MOVW", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store low 2 bytes of ... 2*arg1 ...
   429			{name: "MOVLstoreconstidx1", argLength: 3, reg: gpstoreconstidx, asm: "MOVL", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store low 4 bytes of ... arg1 ...
   430			{name: "MOVLstoreconstidx4", argLength: 3, reg: gpstoreconstidx, asm: "MOVL", aux: "SymValAndOff", typ: "Mem", symEffect: "Write"}, // store low 4 bytes of ... 4*arg1 ...
   431	
   432			// arg0 = pointer to start of memory to zero
   433			// arg1 = value to store (will always be zero)
   434			// arg2 = mem
   435			// auxint = offset into duffzero code to start executing
   436			// returns mem
   437			{
   438				name:      "DUFFZERO",
   439				aux:       "Int64",
   440				argLength: 3,
   441				reg: regInfo{
   442					inputs:   []regMask{buildReg("DI"), buildReg("AX")},
   443					clobbers: buildReg("DI CX"),
   444					// Note: CX is only clobbered when dynamic linking.
   445				},
   446				faultOnNilArg0: true,
   447			},
   448	
   449			// arg0 = address of memory to zero
   450			// arg1 = # of 4-byte words to zero
   451			// arg2 = value to store (will always be zero)
   452			// arg3 = mem
   453			// returns mem
   454			{
   455				name:      "REPSTOSL",
   456				argLength: 4,
   457				reg: regInfo{
   458					inputs:   []regMask{buildReg("DI"), buildReg("CX"), buildReg("AX")},
   459					clobbers: buildReg("DI CX"),
   460				},
   461				faultOnNilArg0: true,
   462			},
   463	
   464			{name: "CALLstatic", argLength: 1, reg: regInfo{clobbers: callerSave}, aux: "SymOff", clobberFlags: true, call: true, symEffect: "None"},                          // call static function aux.(*obj.LSym).  arg0=mem, auxint=argsize, returns mem
   465			{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
   466			{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
   467	
   468			// arg0 = destination pointer
   469			// arg1 = source pointer
   470			// arg2 = mem
   471			// auxint = offset from duffcopy symbol to call
   472			// returns memory
   473			{
   474				name:      "DUFFCOPY",
   475				aux:       "Int64",
   476				argLength: 3,
   477				reg: regInfo{
   478					inputs:   []regMask{buildReg("DI"), buildReg("SI")},
   479					clobbers: buildReg("DI SI CX"), // uses CX as a temporary
   480				},
   481				clobberFlags:   true,
   482				faultOnNilArg0: true,
   483				faultOnNilArg1: true,
   484			},
   485	
   486			// arg0 = destination pointer
   487			// arg1 = source pointer
   488			// arg2 = # of 8-byte words to copy
   489			// arg3 = mem
   490			// returns memory
   491			{
   492				name:      "REPMOVSL",
   493				argLength: 4,
   494				reg: regInfo{
   495					inputs:   []regMask{buildReg("DI"), buildReg("SI"), buildReg("CX")},
   496					clobbers: buildReg("DI SI CX"),
   497				},
   498				faultOnNilArg0: true,
   499				faultOnNilArg1: true,
   500			},
   501	
   502			// (InvertFlags (CMPL a b)) == (CMPL b a)
   503			// So if we want (SETL (CMPL a b)) but we can't do that because a is a constant,
   504			// then we do (SETL (InvertFlags (CMPL b a))) instead.
   505			// Rewrites will convert this to (SETG (CMPL b a)).
   506			// InvertFlags is a pseudo-op which can't appear in assembly output.
   507			{name: "InvertFlags", argLength: 1}, // reverse direction of arg0
   508	
   509			// Pseudo-ops
   510			{name: "LoweredGetG", argLength: 1, reg: gp01}, // arg0=mem
   511			// Scheduler ensures LoweredGetClosurePtr occurs only in entry block,
   512			// and sorts it to the very beginning of the block to prevent other
   513			// use of DX (the closure pointer)
   514			{name: "LoweredGetClosurePtr", reg: regInfo{outputs: []regMask{buildReg("DX")}}, zeroWidth: true},
   515			// LoweredGetCallerPC evaluates to the PC to which its "caller" will return.
   516			// I.e., if f calls g "calls" getcallerpc,
   517			// the result should be the PC within f that g will return to.
   518			// See runtime/stubs.go for a more detailed discussion.
   519			{name: "LoweredGetCallerPC", reg: gp01, rematerializeable: true},
   520			// LoweredGetCallerSP returns the SP of the caller of the current function.
   521			{name: "LoweredGetCallerSP", reg: gp01, rematerializeable: true},
   522			//arg0=ptr,arg1=mem, returns void.  Faults if ptr is nil.
   523			{name: "LoweredNilCheck", argLength: 2, reg: regInfo{inputs: []regMask{gpsp}}, clobberFlags: true, nilCheck: true, faultOnNilArg0: true},
   524	
   525			// LoweredWB invokes runtime.gcWriteBarrier. arg0=destptr, arg1=srcptr, arg2=mem, aux=runtime.gcWriteBarrier
   526			// It saves all GP registers if necessary, but may clobber others.
   527			{name: "LoweredWB", argLength: 3, reg: regInfo{inputs: []regMask{buildReg("DI"), ax}, clobbers: callerSave &^ gp}, clobberFlags: true, aux: "Sym", symEffect: "None"},
   528	
   529			// There are three of these functions so that they can have three different register inputs.
   530			// When we check 0 <= c <= cap (A), then 0 <= b <= c (B), then 0 <= a <= b (C), we want the
   531			// default registers to match so we don't need to copy registers around unnecessarily.
   532			{name: "LoweredPanicBoundsA", argLength: 3, aux: "Int64", reg: regInfo{inputs: []regMask{dx, bx}}, typ: "Mem"}, // arg0=idx, arg1=len, arg2=mem, returns memory. AuxInt contains report code (see PanicBounds in genericOps.go).
   533			{name: "LoweredPanicBoundsB", argLength: 3, aux: "Int64", reg: regInfo{inputs: []regMask{cx, dx}}, typ: "Mem"}, // arg0=idx, arg1=len, arg2=mem, returns memory. AuxInt contains report code (see PanicBounds in genericOps.go).
   534			{name: "LoweredPanicBoundsC", argLength: 3, aux: "Int64", reg: regInfo{inputs: []regMask{ax, cx}}, typ: "Mem"}, // arg0=idx, arg1=len, arg2=mem, returns memory. AuxInt contains report code (see PanicBounds in genericOps.go).
   535			// Extend ops are the same as Bounds ops except the indexes are 64-bit.
   536			{name: "LoweredPanicExtendA", argLength: 4, aux: "Int64", reg: regInfo{inputs: []regMask{si, dx, bx}}, typ: "Mem"}, // arg0=idxHi, arg1=idxLo, arg2=len, arg3=mem, returns memory. AuxInt contains report code (see PanicExtend in genericOps.go).
   537			{name: "LoweredPanicExtendB", argLength: 4, aux: "Int64", reg: regInfo{inputs: []regMask{si, cx, dx}}, typ: "Mem"}, // arg0=idxHi, arg1=idxLo, arg2=len, arg3=mem, returns memory. AuxInt contains report code (see PanicExtend in genericOps.go).
   538			{name: "LoweredPanicExtendC", argLength: 4, aux: "Int64", reg: regInfo{inputs: []regMask{si, ax, cx}}, typ: "Mem"}, // arg0=idxHi, arg1=idxLo, arg2=len, arg3=mem, returns memory. AuxInt contains report code (see PanicExtend in genericOps.go).
   539	
   540			// Constant flag values. For any comparison, there are 5 possible
   541			// outcomes: the three from the signed total order (<,==,>) and the
   542			// three from the unsigned total order. The == cases overlap.
   543			// Note: there's a sixth "unordered" outcome for floating-point
   544			// comparisons, but we don't use such a beast yet.
   545			// These ops are for temporary use by rewrite rules. They
   546			// cannot appear in the generated assembly.
   547			{name: "FlagEQ"},     // equal
   548			{name: "FlagLT_ULT"}, // signed < and unsigned <
   549			{name: "FlagLT_UGT"}, // signed < and unsigned >
   550			{name: "FlagGT_UGT"}, // signed > and unsigned <
   551			{name: "FlagGT_ULT"}, // signed > and unsigned >
   552	
   553			// Special op for -x on 387
   554			{name: "FCHS", argLength: 1, reg: fp11},
   555	
   556			// Special ops for PIC floating-point constants.
   557			// MOVSXconst1 loads the address of the constant-pool entry into a register.
   558			// MOVSXconst2 loads the constant from that address.
   559			// MOVSXconst1 returns a pointer, but we type it as uint32 because it can never point to the Go heap.
   560			{name: "MOVSSconst1", reg: gp01, typ: "UInt32", aux: "Float32"},
   561			{name: "MOVSDconst1", reg: gp01, typ: "UInt32", aux: "Float64"},
   562			{name: "MOVSSconst2", argLength: 1, reg: gpfp, asm: "MOVSS"},
   563			{name: "MOVSDconst2", argLength: 1, reg: gpfp, asm: "MOVSD"},
   564		}
   565	
   566		var _386blocks = []blockData{
   567			{name: "EQ"},
   568			{name: "NE"},
   569			{name: "LT"},
   570			{name: "LE"},
   571			{name: "GT"},
   572			{name: "GE"},
   573			{name: "OS"},
   574			{name: "OC"},
   575			{name: "ULT"},
   576			{name: "ULE"},
   577			{name: "UGT"},
   578			{name: "UGE"},
   579			{name: "EQF"},
   580			{name: "NEF"},
   581			{name: "ORD"}, // FP, ordered comparison (parity zero)
   582			{name: "NAN"}, // FP, unordered comparison (parity one)
   583		}
   584	
   585		archs = append(archs, arch{
   586			name:            "386",
   587			pkg:             "cmd/internal/obj/x86",
   588			genfile:         "../../x86/ssa.go",
   589			ops:             _386ops,
   590			blocks:          _386blocks,
   591			regnames:        regNames386,
   592			gpregmask:       gp,
   593			fpregmask:       fp,
   594			framepointerreg: int8(num["BP"]),
   595			linkreg:         -1, // not used
   596		})
   597	}
   598	

View as plain text