whatcanGOwrong

This commit is contained in:
2024-09-19 21:38:24 -04:00
commit d0ae4d841d
17908 changed files with 4096831 additions and 0 deletions
@@ -0,0 +1,2 @@
tables.go: ../armmap/map.go ../arm.csv
go run ../armmap/map.go -fmt=decoder ../arm.csv >_tables.go && gofmt _tables.go >tables.go && rm _tables.go
@@ -0,0 +1,569 @@
// Copyright 2014 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package armasm
import (
"encoding/binary"
"fmt"
)
// An instFormat describes the format of an instruction encoding.
// An instruction with 32-bit value x matches the format if x&mask == value
// and the condition matches.
// The condition matches if x>>28 == 0xF && value>>28==0xF
// or if x>>28 != 0xF and value>>28 == 0.
// If x matches the format, then the rest of the fields describe how to interpret x.
// The opBits describe bits that should be extracted from x and added to the opcode.
// For example opBits = 0x1234 means that the value
//
// (2 bits at offset 1) followed by (4 bits at offset 3)
//
// should be added to op.
// Finally the args describe how to decode the instruction arguments.
// args is stored as a fixed-size array; if there are fewer than len(args) arguments,
// args[i] == 0 marks the end of the argument list.
type instFormat struct {
mask uint32
value uint32
priority int8
op Op
opBits uint64
args instArgs
}
type instArgs [4]instArg
var (
errMode = fmt.Errorf("unsupported execution mode")
errShort = fmt.Errorf("truncated instruction")
errUnknown = fmt.Errorf("unknown instruction")
)
var decoderCover []bool
// Decode decodes the leading bytes in src as a single instruction.
func Decode(src []byte, mode Mode) (inst Inst, err error) {
if mode != ModeARM {
return Inst{}, errMode
}
if len(src) < 4 {
return Inst{}, errShort
}
if decoderCover == nil {
decoderCover = make([]bool, len(instFormats))
}
x := binary.LittleEndian.Uint32(src)
// The instFormat table contains both conditional and unconditional instructions.
// Considering only the top 4 bits, the conditional instructions use mask=0, value=0,
// while the unconditional instructions use mask=f, value=f.
// Prepare a version of x with the condition cleared to 0 in conditional instructions
// and then assume mask=f during matching.
const condMask = 0xf0000000
xNoCond := x
if x&condMask != condMask {
xNoCond &^= condMask
}
var priority int8
Search:
for i := range instFormats {
f := &instFormats[i]
if xNoCond&(f.mask|condMask) != f.value || f.priority <= priority {
continue
}
delta := uint32(0)
deltaShift := uint(0)
for opBits := f.opBits; opBits != 0; opBits >>= 16 {
n := uint(opBits & 0xFF)
off := uint((opBits >> 8) & 0xFF)
delta |= (x >> off) & (1<<n - 1) << deltaShift
deltaShift += n
}
op := f.op + Op(delta)
// Special case: BKPT encodes with condition but cannot have one.
if op&^15 == BKPT_EQ && op != BKPT {
continue Search
}
var args Args
for j, aop := range f.args {
if aop == 0 {
break
}
arg := decodeArg(aop, x)
if arg == nil { // cannot decode argument
continue Search
}
args[j] = arg
}
decoderCover[i] = true
inst = Inst{
Op: op,
Args: args,
Enc: x,
Len: 4,
}
priority = f.priority
continue Search
}
if inst.Op != 0 {
return inst, nil
}
return Inst{}, errUnknown
}
// An instArg describes the encoding of a single argument.
// In the names used for arguments, _p_ means +, _m_ means -,
// _pm_ means ± (usually keyed by the U bit).
// The _W suffix indicates a general addressing mode based on the P and W bits.
// The _offset and _postindex suffixes force the given addressing mode.
// The rest should be somewhat self-explanatory, at least given
// the decodeArg function.
type instArg uint8
const (
_ instArg = iota
arg_APSR
arg_FPSCR
arg_Dn_half
arg_R1_0
arg_R1_12
arg_R2_0
arg_R2_12
arg_R_0
arg_R_12
arg_R_12_nzcv
arg_R_16
arg_R_16_WB
arg_R_8
arg_R_rotate
arg_R_shift_R
arg_R_shift_imm
arg_SP
arg_Sd
arg_Sd_Dd
arg_Dd_Sd
arg_Sm
arg_Sm_Dm
arg_Sn
arg_Sn_Dn
arg_const
arg_endian
arg_fbits
arg_fp_0
arg_imm24
arg_imm5
arg_imm5_32
arg_imm5_nz
arg_imm_12at8_4at0
arg_imm_4at16_12at0
arg_imm_vfp
arg_label24
arg_label24H
arg_label_m_12
arg_label_p_12
arg_label_pm_12
arg_label_pm_4_4
arg_lsb_width
arg_mem_R
arg_mem_R_pm_R_W
arg_mem_R_pm_R_postindex
arg_mem_R_pm_R_shift_imm_W
arg_mem_R_pm_R_shift_imm_offset
arg_mem_R_pm_R_shift_imm_postindex
arg_mem_R_pm_imm12_W
arg_mem_R_pm_imm12_offset
arg_mem_R_pm_imm12_postindex
arg_mem_R_pm_imm8_W
arg_mem_R_pm_imm8_postindex
arg_mem_R_pm_imm8at0_offset
arg_option
arg_registers
arg_registers1
arg_registers2
arg_satimm4
arg_satimm5
arg_satimm4m1
arg_satimm5m1
arg_widthm1
)
// decodeArg decodes the arg described by aop from the instruction bits x.
// It returns nil if x cannot be decoded according to aop.
func decodeArg(aop instArg, x uint32) Arg {
switch aop {
default:
return nil
case arg_APSR:
return APSR
case arg_FPSCR:
return FPSCR
case arg_R_0:
return Reg(x & (1<<4 - 1))
case arg_R_8:
return Reg((x >> 8) & (1<<4 - 1))
case arg_R_12:
return Reg((x >> 12) & (1<<4 - 1))
case arg_R_16:
return Reg((x >> 16) & (1<<4 - 1))
case arg_R_12_nzcv:
r := Reg((x >> 12) & (1<<4 - 1))
if r == R15 {
return APSR_nzcv
}
return r
case arg_R_16_WB:
mode := AddrLDM
if (x>>21)&1 != 0 {
mode = AddrLDM_WB
}
return Mem{Base: Reg((x >> 16) & (1<<4 - 1)), Mode: mode}
case arg_R_rotate:
Rm := Reg(x & (1<<4 - 1))
typ, count := decodeShift(x)
// ROR #0 here means ROR #0, but decodeShift rewrites to RRX #1.
if typ == RotateRightExt {
return Rm
}
return RegShift{Rm, typ, count}
case arg_R_shift_R:
Rm := Reg(x & (1<<4 - 1))
Rs := Reg((x >> 8) & (1<<4 - 1))
typ := Shift((x >> 5) & (1<<2 - 1))
return RegShiftReg{Rm, typ, Rs}
case arg_R_shift_imm:
Rm := Reg(x & (1<<4 - 1))
typ, count := decodeShift(x)
if typ == ShiftLeft && count == 0 {
return Reg(Rm)
}
return RegShift{Rm, typ, count}
case arg_R1_0:
return Reg((x & (1<<4 - 1)))
case arg_R1_12:
return Reg(((x >> 12) & (1<<4 - 1)))
case arg_R2_0:
return Reg((x & (1<<4 - 1)) | 1)
case arg_R2_12:
return Reg(((x >> 12) & (1<<4 - 1)) | 1)
case arg_SP:
return SP
case arg_Sd_Dd:
v := (x >> 12) & (1<<4 - 1)
vx := (x >> 22) & 1
sz := (x >> 8) & 1
if sz != 0 {
return D0 + Reg(vx<<4+v)
} else {
return S0 + Reg(v<<1+vx)
}
case arg_Dd_Sd:
return decodeArg(arg_Sd_Dd, x^(1<<8))
case arg_Sd:
v := (x >> 12) & (1<<4 - 1)
vx := (x >> 22) & 1
return S0 + Reg(v<<1+vx)
case arg_Sm_Dm:
v := (x >> 0) & (1<<4 - 1)
vx := (x >> 5) & 1
sz := (x >> 8) & 1
if sz != 0 {
return D0 + Reg(vx<<4+v)
} else {
return S0 + Reg(v<<1+vx)
}
case arg_Sm:
v := (x >> 0) & (1<<4 - 1)
vx := (x >> 5) & 1
return S0 + Reg(v<<1+vx)
case arg_Dn_half:
v := (x >> 16) & (1<<4 - 1)
vx := (x >> 7) & 1
return RegX{D0 + Reg(vx<<4+v), int((x >> 21) & 1)}
case arg_Sn_Dn:
v := (x >> 16) & (1<<4 - 1)
vx := (x >> 7) & 1
sz := (x >> 8) & 1
if sz != 0 {
return D0 + Reg(vx<<4+v)
} else {
return S0 + Reg(v<<1+vx)
}
case arg_Sn:
v := (x >> 16) & (1<<4 - 1)
vx := (x >> 7) & 1
return S0 + Reg(v<<1+vx)
case arg_const:
v := x & (1<<8 - 1)
rot := (x >> 8) & (1<<4 - 1) * 2
if rot > 0 && v&3 == 0 {
// could rotate less
return ImmAlt{uint8(v), uint8(rot)}
}
if rot >= 24 && ((v<<(32-rot))&0xFF)>>(32-rot) == v {
// could wrap around to rot==0.
return ImmAlt{uint8(v), uint8(rot)}
}
return Imm(v>>rot | v<<(32-rot))
case arg_endian:
return Endian((x >> 9) & 1)
case arg_fbits:
return Imm((16 << ((x >> 7) & 1)) - ((x&(1<<4-1))<<1 | (x>>5)&1))
case arg_fp_0:
return Imm(0)
case arg_imm24:
return Imm(x & (1<<24 - 1))
case arg_imm5:
return Imm((x >> 7) & (1<<5 - 1))
case arg_imm5_32:
x = (x >> 7) & (1<<5 - 1)
if x == 0 {
x = 32
}
return Imm(x)
case arg_imm5_nz:
x = (x >> 7) & (1<<5 - 1)
if x == 0 {
return nil
}
return Imm(x)
case arg_imm_4at16_12at0:
return Imm((x>>16)&(1<<4-1)<<12 | x&(1<<12-1))
case arg_imm_12at8_4at0:
return Imm((x>>8)&(1<<12-1)<<4 | x&(1<<4-1))
case arg_imm_vfp:
x = (x>>16)&(1<<4-1)<<4 | x&(1<<4-1)
return Imm(x)
case arg_label24:
imm := (x & (1<<24 - 1)) << 2
return PCRel(int32(imm<<6) >> 6)
case arg_label24H:
h := (x >> 24) & 1
imm := (x&(1<<24-1))<<2 | h<<1
return PCRel(int32(imm<<6) >> 6)
case arg_label_m_12:
d := int32(x & (1<<12 - 1))
return Mem{Base: PC, Mode: AddrOffset, Offset: int16(-d)}
case arg_label_p_12:
d := int32(x & (1<<12 - 1))
return Mem{Base: PC, Mode: AddrOffset, Offset: int16(d)}
case arg_label_pm_12:
d := int32(x & (1<<12 - 1))
u := (x >> 23) & 1
if u == 0 {
d = -d
}
return Mem{Base: PC, Mode: AddrOffset, Offset: int16(d)}
case arg_label_pm_4_4:
d := int32((x>>8)&(1<<4-1)<<4 | x&(1<<4-1))
u := (x >> 23) & 1
if u == 0 {
d = -d
}
return PCRel(d)
case arg_lsb_width:
lsb := (x >> 7) & (1<<5 - 1)
msb := (x >> 16) & (1<<5 - 1)
if msb < lsb || msb >= 32 {
return nil
}
return Imm(msb + 1 - lsb)
case arg_mem_R:
Rn := Reg((x >> 16) & (1<<4 - 1))
return Mem{Base: Rn, Mode: AddrOffset}
case arg_mem_R_pm_R_postindex:
// Treat [<Rn>],+/-<Rm> like [<Rn>,+/-<Rm>{,<shift>}]{!}
// by forcing shift bits to <<0 and P=0, W=0 (postindex=true).
return decodeArg(arg_mem_R_pm_R_shift_imm_W, x&^((1<<7-1)<<5|1<<24|1<<21))
case arg_mem_R_pm_R_W:
// Treat [<Rn>,+/-<Rm>]{!} like [<Rn>,+/-<Rm>{,<shift>}]{!}
// by forcing shift bits to <<0.
return decodeArg(arg_mem_R_pm_R_shift_imm_W, x&^((1<<7-1)<<5))
case arg_mem_R_pm_R_shift_imm_offset:
// Treat [<Rn>],+/-<Rm>{,<shift>} like [<Rn>,+/-<Rm>{,<shift>}]{!}
// by forcing P=1, W=0 (index=false, wback=false).
return decodeArg(arg_mem_R_pm_R_shift_imm_W, x&^(1<<21)|1<<24)
case arg_mem_R_pm_R_shift_imm_postindex:
// Treat [<Rn>],+/-<Rm>{,<shift>} like [<Rn>,+/-<Rm>{,<shift>}]{!}
// by forcing P=0, W=0 (postindex=true).
return decodeArg(arg_mem_R_pm_R_shift_imm_W, x&^(1<<24|1<<21))
case arg_mem_R_pm_R_shift_imm_W:
Rn := Reg((x >> 16) & (1<<4 - 1))
Rm := Reg(x & (1<<4 - 1))
typ, count := decodeShift(x)
u := (x >> 23) & 1
w := (x >> 21) & 1
p := (x >> 24) & 1
if p == 0 && w == 1 {
return nil
}
sign := int8(+1)
if u == 0 {
sign = -1
}
mode := AddrMode(uint8(p<<1) | uint8(w^1))
return Mem{Base: Rn, Mode: mode, Sign: sign, Index: Rm, Shift: typ, Count: count}
case arg_mem_R_pm_imm12_offset:
// Treat [<Rn>,#+/-<imm12>] like [<Rn>{,#+/-<imm12>}]{!}
// by forcing P=1, W=0 (index=false, wback=false).
return decodeArg(arg_mem_R_pm_imm12_W, x&^(1<<21)|1<<24)
case arg_mem_R_pm_imm12_postindex:
// Treat [<Rn>],#+/-<imm12> like [<Rn>{,#+/-<imm12>}]{!}
// by forcing P=0, W=0 (postindex=true).
return decodeArg(arg_mem_R_pm_imm12_W, x&^(1<<24|1<<21))
case arg_mem_R_pm_imm12_W:
Rn := Reg((x >> 16) & (1<<4 - 1))
u := (x >> 23) & 1
w := (x >> 21) & 1
p := (x >> 24) & 1
if p == 0 && w == 1 {
return nil
}
sign := int8(+1)
if u == 0 {
sign = -1
}
imm := int16(x & (1<<12 - 1))
mode := AddrMode(uint8(p<<1) | uint8(w^1))
return Mem{Base: Rn, Mode: mode, Offset: int16(sign) * imm}
case arg_mem_R_pm_imm8_postindex:
// Treat [<Rn>],#+/-<imm8> like [<Rn>{,#+/-<imm8>}]{!}
// by forcing P=0, W=0 (postindex=true).
return decodeArg(arg_mem_R_pm_imm8_W, x&^(1<<24|1<<21))
case arg_mem_R_pm_imm8_W:
Rn := Reg((x >> 16) & (1<<4 - 1))
u := (x >> 23) & 1
w := (x >> 21) & 1
p := (x >> 24) & 1
if p == 0 && w == 1 {
return nil
}
sign := int8(+1)
if u == 0 {
sign = -1
}
imm := int16((x>>8)&(1<<4-1)<<4 | x&(1<<4-1))
mode := AddrMode(uint8(p<<1) | uint8(w^1))
return Mem{Base: Rn, Mode: mode, Offset: int16(sign) * imm}
case arg_mem_R_pm_imm8at0_offset:
Rn := Reg((x >> 16) & (1<<4 - 1))
u := (x >> 23) & 1
sign := int8(+1)
if u == 0 {
sign = -1
}
imm := int16(x&(1<<8-1)) << 2
return Mem{Base: Rn, Mode: AddrOffset, Offset: int16(sign) * imm}
case arg_option:
return Imm(x & (1<<4 - 1))
case arg_registers:
return RegList(x & (1<<16 - 1))
case arg_registers2:
x &= 1<<16 - 1
n := 0
for i := 0; i < 16; i++ {
if x>>uint(i)&1 != 0 {
n++
}
}
if n < 2 {
return nil
}
return RegList(x)
case arg_registers1:
Rt := (x >> 12) & (1<<4 - 1)
return RegList(1 << Rt)
case arg_satimm4:
return Imm((x >> 16) & (1<<4 - 1))
case arg_satimm5:
return Imm((x >> 16) & (1<<5 - 1))
case arg_satimm4m1:
return Imm((x>>16)&(1<<4-1) + 1)
case arg_satimm5m1:
return Imm((x>>16)&(1<<5-1) + 1)
case arg_widthm1:
return Imm((x>>16)&(1<<5-1) + 1)
}
}
// decodeShift decodes the shift-by-immediate encoded in x.
func decodeShift(x uint32) (Shift, uint8) {
count := (x >> 7) & (1<<5 - 1)
typ := Shift((x >> 5) & (1<<2 - 1))
switch typ {
case ShiftRight, ShiftRightSigned:
if count == 0 {
count = 32
}
case RotateRight:
if count == 0 {
typ = RotateRightExt
count = 1
}
}
return typ, uint8(count)
}
@@ -0,0 +1,69 @@
// Copyright 2014 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package armasm
import (
"encoding/hex"
"io/ioutil"
"strconv"
"strings"
"testing"
)
func TestDecode(t *testing.T) {
data, err := ioutil.ReadFile("testdata/decode.txt")
if err != nil {
t.Fatal(err)
}
all := string(data)
for strings.Contains(all, "\t\t") {
all = strings.Replace(all, "\t\t", "\t", -1)
}
for _, line := range strings.Split(all, "\n") {
line = strings.TrimSpace(line)
if line == "" || strings.HasPrefix(line, "#") {
continue
}
f := strings.SplitN(line, "\t", 4)
i := strings.Index(f[0], "|")
if i < 0 {
t.Errorf("parsing %q: missing | separator", f[0])
continue
}
if i%2 != 0 {
t.Errorf("parsing %q: misaligned | separator", f[0])
}
size := i / 2
code, err := hex.DecodeString(f[0][:i] + f[0][i+1:])
if err != nil {
t.Errorf("parsing %q: %v", f[0], err)
continue
}
mode, err := strconv.Atoi(f[1])
if err != nil {
t.Errorf("invalid mode %q in: %s", f[1], line)
continue
}
syntax, asm := f[2], f[3]
inst, err := Decode(code, Mode(mode))
var out string
if err != nil {
out = "error: " + err.Error()
} else {
switch syntax {
case "gnu":
out = GNUSyntax(inst)
case "plan9": // [sic]
out = GoSyntax(inst, 0, nil, nil)
default:
t.Errorf("unknown syntax %q", syntax)
continue
}
}
if out != asm || inst.Len != size {
t.Errorf("Decode(%s) [%s] = %s, %d, want %s, %d", f[0], syntax, out, inst.Len, asm, size)
}
}
}
@@ -0,0 +1,615 @@
// Copyright 2014 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Support for testing against external disassembler program.
// Copied and simplified from ../../x86/x86asm/ext_test.go.
package armasm
import (
"bufio"
"bytes"
"encoding/hex"
"flag"
"fmt"
"io"
"io/ioutil"
"log"
"math/rand"
"os"
"os/exec"
"regexp"
"runtime"
"strings"
"testing"
"time"
)
var (
printTests = flag.Bool("printtests", false, "print test cases that exercise new code paths")
dumpTest = flag.Bool("dump", false, "dump all encodings")
mismatch = flag.Bool("mismatch", false, "log allowed mismatches")
longTest = flag.Bool("long", false, "long test")
keep = flag.Bool("keep", false, "keep object files around")
debug = false
)
// An ExtInst represents a single decoded instruction parsed
// from an external disassembler's output.
type ExtInst struct {
addr uint32
enc [4]byte
nenc int
text string
}
func (r ExtInst) String() string {
return fmt.Sprintf("%#x: % x: %s", r.addr, r.enc, r.text)
}
// An ExtDis is a connection between an external disassembler and a test.
type ExtDis struct {
Arch Mode
Dec chan ExtInst
File *os.File
Size int
KeepFile bool
Cmd *exec.Cmd
}
// Run runs the given command - the external disassembler - and returns
// a buffered reader of its standard output.
func (ext *ExtDis) Run(cmd ...string) (*bufio.Reader, error) {
if *keep {
log.Printf("%s\n", strings.Join(cmd, " "))
}
ext.Cmd = exec.Command(cmd[0], cmd[1:]...)
out, err := ext.Cmd.StdoutPipe()
if err != nil {
return nil, fmt.Errorf("stdoutpipe: %v", err)
}
if err := ext.Cmd.Start(); err != nil {
return nil, fmt.Errorf("exec: %v", err)
}
b := bufio.NewReaderSize(out, 1<<20)
return b, nil
}
// Wait waits for the command started with Run to exit.
func (ext *ExtDis) Wait() error {
return ext.Cmd.Wait()
}
// testExtDis tests a set of byte sequences against an external disassembler.
// The disassembler is expected to produce the given syntax and be run
// in the given architecture mode (16, 32, or 64-bit).
// The extdis function must start the external disassembler
// and then parse its output, sending the parsed instructions on ext.Dec.
// The generate function calls its argument f once for each byte sequence
// to be tested. The generate function itself will be called twice, and it must
// make the same sequence of calls to f each time.
// When a disassembly does not match the internal decoding,
// allowedMismatch determines whether this mismatch should be
// allowed, or else considered an error.
func testExtDis(
t *testing.T,
syntax string,
arch Mode,
extdis func(ext *ExtDis) error,
generate func(f func([]byte)),
allowedMismatch func(text string, size int, inst *Inst, dec ExtInst) bool,
) {
start := time.Now()
ext := &ExtDis{
Dec: make(chan ExtInst),
Arch: arch,
}
errc := make(chan error)
// First pass: write instructions to input file for external disassembler.
file, f, size, err := writeInst(generate)
if err != nil {
t.Fatal(err)
}
ext.Size = size
ext.File = f
defer func() {
f.Close()
if !*keep {
os.Remove(file)
}
}()
// Second pass: compare disassembly against our decodings.
var (
totalTests = 0
totalSkips = 0
totalErrors = 0
errors = make([]string, 0, 100) // sampled errors, at most cap
)
go func() {
errc <- extdis(ext)
}()
generate(func(enc []byte) {
dec, ok := <-ext.Dec
if !ok {
t.Errorf("decoding stream ended early")
return
}
inst, text := disasm(syntax, arch, pad(enc))
totalTests++
if *dumpTest {
fmt.Printf("%x -> %s [%d]\n", enc[:len(enc)], dec.text, dec.nenc)
}
if text != dec.text || inst.Len != dec.nenc {
suffix := ""
if allowedMismatch(text, size, &inst, dec) {
totalSkips++
if !*mismatch {
return
}
suffix += " (allowed mismatch)"
}
totalErrors++
if len(errors) >= cap(errors) {
j := rand.Intn(totalErrors)
if j >= cap(errors) {
return
}
errors = append(errors[:j], errors[j+1:]...)
}
errors = append(errors, fmt.Sprintf("decode(%x) = %q, %d, want %q, %d%s", enc, text, inst.Len, dec.text, dec.nenc, suffix))
}
})
if *mismatch {
totalErrors -= totalSkips
}
for _, b := range errors {
t.Log(b)
}
if totalErrors > 0 {
t.Fail()
}
t.Logf("%d test cases, %d expected mismatches, %d failures; %.0f cases/second", totalTests, totalSkips, totalErrors, float64(totalTests)/time.Since(start).Seconds())
if err := <-errc; err != nil {
t.Fatalf("external disassembler: %v", err)
}
}
const start = 0x8000 // start address of text
// writeInst writes the generated byte sequences to a new file
// starting at offset start. That file is intended to be the input to
// the external disassembler.
func writeInst(generate func(func([]byte))) (file string, f *os.File, size int, err error) {
f, err = ioutil.TempFile("", "armasm")
if err != nil {
return
}
file = f.Name()
f.Seek(start, io.SeekStart)
w := bufio.NewWriter(f)
defer w.Flush()
size = 0
generate(func(x []byte) {
if len(x) > 4 {
x = x[:4]
}
if debug {
fmt.Printf("%#x: %x%x\n", start+size, x, zeros[len(x):])
}
w.Write(x)
w.Write(zeros[len(x):])
size += len(zeros)
})
return file, f, size, nil
}
var zeros = []byte{0, 0, 0, 0}
// pad pads the code sequence with pops.
func pad(enc []byte) []byte {
if len(enc) < 4 {
enc = append(enc[:len(enc):len(enc)], zeros[:4-len(enc)]...)
}
return enc
}
// disasm returns the decoded instruction and text
// for the given source bytes, using the given syntax and mode.
func disasm(syntax string, mode Mode, src []byte) (inst Inst, text string) {
// If printTests is set, we record the coverage value
// before and after, and we write out the inputs for which
// coverage went up, in the format expected in testdata/decode.text.
// This produces a fairly small set of test cases that exercise nearly
// all the code.
var cover float64
if *printTests {
cover -= coverage()
}
inst, err := Decode(src, mode)
if err != nil {
text = "error: " + err.Error()
} else {
text = inst.String()
switch syntax {
//case "arm":
// text = ARMSyntax(inst)
case "gnu":
text = GNUSyntax(inst)
//case "plan9": // [sic]
// text = GoSyntax(inst, 0, nil)
default:
text = "error: unknown syntax " + syntax
}
}
if *printTests {
cover += coverage()
if cover > 0 {
max := len(src)
if max > 4 && inst.Len <= 4 {
max = 4
}
fmt.Printf("%x|%x\t%d\t%s\t%s\n", src[:inst.Len], src[inst.Len:max], mode, syntax, text)
}
}
return
}
// coverage returns a floating point number denoting the
// test coverage until now. The number increases when new code paths are exercised,
// both in the Go program and in the decoder byte code.
func coverage() float64 {
/*
testing.Coverage is not in the main distribution.
The implementation, which must go in package testing, is:
// Coverage reports the current code coverage as a fraction in the range [0, 1].
func Coverage() float64 {
var n, d int64
for _, counters := range cover.Counters {
for _, c := range counters {
if c > 0 {
n++
}
d++
}
}
if d == 0 {
return 0
}
return float64(n) / float64(d)
}
*/
var f float64
f += testing.Coverage()
f += decodeCoverage()
return f
}
func decodeCoverage() float64 {
n := 0
for _, t := range decoderCover {
if t {
n++
}
}
return float64(1+n) / float64(1+len(decoderCover))
}
// Helpers for writing disassembler output parsers.
// hasPrefix reports whether any of the space-separated words in the text s
// begins with any of the given prefixes.
func hasPrefix(s string, prefixes ...string) bool {
for _, prefix := range prefixes {
for s := s; s != ""; {
if strings.HasPrefix(s, prefix) {
return true
}
i := strings.Index(s, " ")
if i < 0 {
break
}
s = s[i+1:]
}
}
return false
}
// contains reports whether the text s contains any of the given substrings.
func contains(s string, substrings ...string) bool {
for _, sub := range substrings {
if strings.Contains(s, sub) {
return true
}
}
return false
}
// isHex reports whether b is a hexadecimal character (0-9A-Fa-f).
func isHex(b byte) bool { return b == '0' || unhex[b] > 0 }
// parseHex parses the hexadecimal byte dump in hex,
// appending the parsed bytes to raw and returning the updated slice.
// The returned bool signals whether any invalid hex was found.
// Spaces and tabs between bytes are okay but any other non-hex is not.
func parseHex(hex []byte, raw []byte) ([]byte, bool) {
hex = trimSpace(hex)
for j := 0; j < len(hex); {
for hex[j] == ' ' || hex[j] == '\t' {
j++
}
if j >= len(hex) {
break
}
if j+2 > len(hex) || !isHex(hex[j]) || !isHex(hex[j+1]) {
return nil, false
}
raw = append(raw, unhex[hex[j]]<<4|unhex[hex[j+1]])
j += 2
}
return raw, true
}
var unhex = [256]byte{
'0': 0,
'1': 1,
'2': 2,
'3': 3,
'4': 4,
'5': 5,
'6': 6,
'7': 7,
'8': 8,
'9': 9,
'A': 10,
'B': 11,
'C': 12,
'D': 13,
'E': 14,
'F': 15,
'a': 10,
'b': 11,
'c': 12,
'd': 13,
'e': 14,
'f': 15,
}
// index is like bytes.Index(s, []byte(t)) but avoids the allocation.
func index(s []byte, t string) int {
i := 0
for {
j := bytes.IndexByte(s[i:], t[0])
if j < 0 {
return -1
}
i = i + j
if i+len(t) > len(s) {
return -1
}
for k := 1; k < len(t); k++ {
if s[i+k] != t[k] {
goto nomatch
}
}
return i
nomatch:
i++
}
}
// fixSpace rewrites runs of spaces, tabs, and newline characters into single spaces in s.
// If s must be rewritten, it is rewritten in place.
func fixSpace(s []byte) []byte {
s = trimSpace(s)
for i := 0; i < len(s); i++ {
if s[i] == '\t' || s[i] == '\n' || i > 0 && s[i] == ' ' && s[i-1] == ' ' {
goto Fix
}
}
return s
Fix:
b := s
w := 0
for i := 0; i < len(s); i++ {
c := s[i]
if c == '\t' || c == '\n' {
c = ' '
}
if c == ' ' && w > 0 && b[w-1] == ' ' {
continue
}
b[w] = c
w++
}
if w > 0 && b[w-1] == ' ' {
w--
}
return b[:w]
}
// trimSpace trims leading and trailing space from s, returning a subslice of s.
func trimSpace(s []byte) []byte {
j := len(s)
for j > 0 && (s[j-1] == ' ' || s[j-1] == '\t' || s[j-1] == '\n') {
j--
}
i := 0
for i < j && (s[i] == ' ' || s[i] == '\t') {
i++
}
return s[i:j]
}
// pcrel matches instructions using relative addressing mode.
var (
pcrel = regexp.MustCompile(`^((?:.* )?(?:b|bl)x?(?:eq|ne|cs|cc|mi|pl|vs|vc|hi|ls|ge|lt|gt|le)?) 0x([0-9a-f]+)$`)
)
// Generators.
//
// The test cases are described as functions that invoke a callback repeatedly,
// with a new input sequence each time. These helpers make writing those
// a little easier.
// condCases generates conditional instructions.
func condCases(t *testing.T) func(func([]byte)) {
return func(try func([]byte)) {
// All the strides are relatively prime to 2 and therefore to 2²⁸,
// so we will not repeat any instructions until we have tried all 2²⁸.
// Using a stride other than 1 is meant to visit the instructions in a
// pseudorandom order, which gives better variety in the set of
// test cases chosen by -printtests.
stride := uint32(10007)
n := 1 << 28 / 7
if testing.Short() {
stride = 100003
n = 1 << 28 / 1001
} else if *longTest {
stride = 200000033
n = 1 << 28
}
x := uint32(0)
for i := 0; i < n; i++ {
enc := (x%15)<<28 | x&(1<<28-1)
try([]byte{byte(enc), byte(enc >> 8), byte(enc >> 16), byte(enc >> 24)})
x += stride
}
}
}
// uncondCases generates unconditional instructions.
func uncondCases(t *testing.T) func(func([]byte)) {
return func(try func([]byte)) {
condCases(t)(func(enc []byte) {
enc[3] |= 0xF0
try(enc)
})
}
}
func countBits(x uint32) int {
n := 0
for ; x != 0; x >>= 1 {
n += int(x & 1)
}
return n
}
func expandBits(x, m uint32) uint32 {
var out uint32
for i := uint(0); i < 32; i++ {
out >>= 1
if m&1 != 0 {
out |= (x & 1) << 31
x >>= 1
}
m >>= 1
}
return out
}
func tryCondMask(mask, val uint32, try func([]byte)) {
n := countBits(^mask)
bits := uint32(0)
for i := 0; i < 1<<uint(n); i++ {
bits += 848251 // arbitrary prime
x := val | expandBits(bits, ^mask) | uint32(i)%15<<28
try([]byte{byte(x), byte(x >> 8), byte(x >> 16), byte(x >> 24)})
}
}
// vfpCases generates VFP instructions.
func vfpCases(t *testing.T) func(func([]byte)) {
const (
vfpmask uint32 = 0xFF00FE10
vfp uint32 = 0x0E009A00
)
return func(try func([]byte)) {
tryCondMask(0xff00fe10, 0x0e009a00, try) // standard VFP instruction space
tryCondMask(0xffc00f7f, 0x0e000b10, try) // VFP MOV core reg to/from float64 half
tryCondMask(0xffe00f7f, 0x0e000a10, try) // VFP MOV core reg to/from float32
tryCondMask(0xffef0fff, 0x0ee10a10, try) // VFP MOV core reg to/from cond codes
}
}
// hexCases generates the cases written in hexadecimal in the encoded string.
// Spaces in 'encoded' separate entire test cases, not individual bytes.
func hexCases(t *testing.T, encoded string) func(func([]byte)) {
return func(try func([]byte)) {
for _, x := range strings.Fields(encoded) {
src, err := hex.DecodeString(x)
if err != nil {
t.Errorf("parsing %q: %v", x, err)
}
try(src)
}
}
}
// testdataCases generates the test cases recorded in testdata/decode.txt.
// It only uses the inputs; it ignores the answers recorded in that file.
func testdataCases(t *testing.T) func(func([]byte)) {
var codes [][]byte
data, err := ioutil.ReadFile("testdata/decode.txt")
if err != nil {
t.Fatal(err)
}
for _, line := range strings.Split(string(data), "\n") {
line = strings.TrimSpace(line)
if line == "" || strings.HasPrefix(line, "#") {
continue
}
f := strings.Fields(line)[0]
i := strings.Index(f, "|")
if i < 0 {
t.Errorf("parsing %q: missing | separator", f)
continue
}
if i%2 != 0 {
t.Errorf("parsing %q: misaligned | separator", f)
}
code, err := hex.DecodeString(f[:i] + f[i+1:])
if err != nil {
t.Errorf("parsing %q: %v", f, err)
continue
}
codes = append(codes, code)
}
return func(try func([]byte)) {
for _, code := range codes {
try(code)
}
}
}
func caller(skip int) string {
pc, _, _, _ := runtime.Caller(skip)
f := runtime.FuncForPC(pc)
name := "?"
if f != nil {
name = f.Name()
if i := strings.LastIndex(name, "."); i >= 0 {
name = name[i+1:]
}
}
return name
}
@@ -0,0 +1,164 @@
// Copyright 2014 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package armasm
import (
"bytes"
"fmt"
"strings"
)
var saveDot = strings.NewReplacer(
".F16", "_dot_F16",
".F32", "_dot_F32",
".F64", "_dot_F64",
".S32", "_dot_S32",
".U32", "_dot_U32",
".FXS", "_dot_S",
".FXU", "_dot_U",
".32", "_dot_32",
)
// GNUSyntax returns the GNU assembler syntax for the instruction, as defined by GNU binutils.
// This form typically matches the syntax defined in the ARM Reference Manual.
func GNUSyntax(inst Inst) string {
var buf bytes.Buffer
op := inst.Op.String()
op = saveDot.Replace(op)
op = strings.Replace(op, ".", "", -1)
op = strings.Replace(op, "_dot_", ".", -1)
op = strings.ToLower(op)
buf.WriteString(op)
sep := " "
for i, arg := range inst.Args {
if arg == nil {
break
}
text := gnuArg(&inst, i, arg)
if text == "" {
continue
}
buf.WriteString(sep)
sep = ", "
buf.WriteString(text)
}
return buf.String()
}
func gnuArg(inst *Inst, argIndex int, arg Arg) string {
switch inst.Op &^ 15 {
case LDRD_EQ, LDREXD_EQ, STRD_EQ:
if argIndex == 1 {
// second argument in consecutive pair not printed
return ""
}
case STREXD_EQ:
if argIndex == 2 {
// second argument in consecutive pair not printed
return ""
}
}
switch arg := arg.(type) {
case Imm:
switch inst.Op &^ 15 {
case BKPT_EQ:
return fmt.Sprintf("%#04x", uint32(arg))
case SVC_EQ:
return fmt.Sprintf("%#08x", uint32(arg))
}
return fmt.Sprintf("#%d", int32(arg))
case ImmAlt:
return fmt.Sprintf("#%d, %d", arg.Val, arg.Rot)
case Mem:
R := gnuArg(inst, -1, arg.Base)
X := ""
if arg.Sign != 0 {
X = ""
if arg.Sign < 0 {
X = "-"
}
X += gnuArg(inst, -1, arg.Index)
if arg.Shift == ShiftLeft && arg.Count == 0 {
// nothing
} else if arg.Shift == RotateRightExt {
X += ", rrx"
} else {
X += fmt.Sprintf(", %s #%d", strings.ToLower(arg.Shift.String()), arg.Count)
}
} else {
X = fmt.Sprintf("#%d", arg.Offset)
}
switch arg.Mode {
case AddrOffset:
if X == "#0" {
return fmt.Sprintf("[%s]", R)
}
return fmt.Sprintf("[%s, %s]", R, X)
case AddrPreIndex:
return fmt.Sprintf("[%s, %s]!", R, X)
case AddrPostIndex:
return fmt.Sprintf("[%s], %s", R, X)
case AddrLDM:
if X == "#0" {
return R
}
case AddrLDM_WB:
if X == "#0" {
return R + "!"
}
}
return fmt.Sprintf("[%s Mode(%d) %s]", R, int(arg.Mode), X)
case PCRel:
return fmt.Sprintf(".%+#x", int32(arg)+4)
case Reg:
switch inst.Op &^ 15 {
case LDREX_EQ:
if argIndex == 0 {
return fmt.Sprintf("r%d", int32(arg))
}
}
switch arg {
case R10:
return "sl"
case R11:
return "fp"
case R12:
return "ip"
}
case RegList:
var buf bytes.Buffer
fmt.Fprintf(&buf, "{")
sep := ""
for i := 0; i < 16; i++ {
if arg&(1<<uint(i)) != 0 {
fmt.Fprintf(&buf, "%s%s", sep, gnuArg(inst, -1, Reg(i)))
sep = ", "
}
}
fmt.Fprintf(&buf, "}")
return buf.String()
case RegShift:
if arg.Shift == ShiftLeft && arg.Count == 0 {
return gnuArg(inst, -1, arg.Reg)
}
if arg.Shift == RotateRightExt {
return gnuArg(inst, -1, arg.Reg) + ", rrx"
}
return fmt.Sprintf("%s, %s #%d", gnuArg(inst, -1, arg.Reg), strings.ToLower(arg.Shift.String()), arg.Count)
case RegShiftReg:
return fmt.Sprintf("%s, %s %s", gnuArg(inst, -1, arg.Reg), strings.ToLower(arg.Shift.String()), gnuArg(inst, -1, arg.RegCount))
}
return strings.ToLower(arg.String())
}
@@ -0,0 +1,438 @@
// Copyright 2014 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package armasm
import (
"bytes"
"fmt"
)
// A Mode is an instruction execution mode.
type Mode int
const (
_ Mode = iota
ModeARM
ModeThumb
)
func (m Mode) String() string {
switch m {
case ModeARM:
return "ARM"
case ModeThumb:
return "Thumb"
}
return fmt.Sprintf("Mode(%d)", int(m))
}
// An Op is an ARM opcode.
type Op uint16
// NOTE: The actual Op values are defined in tables.go.
// They are chosen to simplify instruction decoding and
// are not a dense packing from 0 to N, although the
// density is high, probably at least 90%.
func (op Op) String() string {
if op >= Op(len(opstr)) || opstr[op] == "" {
return fmt.Sprintf("Op(%d)", int(op))
}
return opstr[op]
}
// An Inst is a single instruction.
type Inst struct {
Op Op // Opcode mnemonic
Enc uint32 // Raw encoding bits.
Len int // Length of encoding in bytes.
Args Args // Instruction arguments, in ARM manual order.
}
func (i Inst) String() string {
var buf bytes.Buffer
buf.WriteString(i.Op.String())
for j, arg := range i.Args {
if arg == nil {
break
}
if j == 0 {
buf.WriteString(" ")
} else {
buf.WriteString(", ")
}
buf.WriteString(arg.String())
}
return buf.String()
}
// An Args holds the instruction arguments.
// If an instruction has fewer than 4 arguments,
// the final elements in the array are nil.
type Args [4]Arg
// An Arg is a single instruction argument, one of these types:
// Endian, Imm, Mem, PCRel, Reg, RegList, RegShift, RegShiftReg.
type Arg interface {
IsArg()
String() string
}
type Float32Imm float32
func (Float32Imm) IsArg() {}
func (f Float32Imm) String() string {
return fmt.Sprintf("#%v", float32(f))
}
type Float64Imm float32
func (Float64Imm) IsArg() {}
func (f Float64Imm) String() string {
return fmt.Sprintf("#%v", float64(f))
}
// An Imm is an integer constant.
type Imm uint32
func (Imm) IsArg() {}
func (i Imm) String() string {
return fmt.Sprintf("#%#x", uint32(i))
}
// An ImmAlt is an alternate encoding of an integer constant.
type ImmAlt struct {
Val uint8
Rot uint8
}
func (ImmAlt) IsArg() {}
func (i ImmAlt) Imm() Imm {
v := uint32(i.Val)
r := uint(i.Rot)
return Imm(v>>r | v<<(32-r))
}
func (i ImmAlt) String() string {
return fmt.Sprintf("#%#x, %d", i.Val, i.Rot)
}
// A Label is a text (code) address.
type Label uint32
func (Label) IsArg() {}
func (i Label) String() string {
return fmt.Sprintf("%#x", uint32(i))
}
// A Reg is a single register.
// The zero value denotes R0, not the absence of a register.
type Reg uint8
const (
R0 Reg = iota
R1
R2
R3
R4
R5
R6
R7
R8
R9
R10
R11
R12
R13
R14
R15
S0
S1
S2
S3
S4
S5
S6
S7
S8
S9
S10
S11
S12
S13
S14
S15
S16
S17
S18
S19
S20
S21
S22
S23
S24
S25
S26
S27
S28
S29
S30
S31
D0
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
D15
D16
D17
D18
D19
D20
D21
D22
D23
D24
D25
D26
D27
D28
D29
D30
D31
APSR
APSR_nzcv
FPSCR
SP = R13
LR = R14
PC = R15
)
func (Reg) IsArg() {}
func (r Reg) String() string {
switch r {
case APSR:
return "APSR"
case APSR_nzcv:
return "APSR_nzcv"
case FPSCR:
return "FPSCR"
case SP:
return "SP"
case PC:
return "PC"
case LR:
return "LR"
}
if R0 <= r && r <= R15 {
return fmt.Sprintf("R%d", int(r-R0))
}
if S0 <= r && r <= S31 {
return fmt.Sprintf("S%d", int(r-S0))
}
if D0 <= r && r <= D31 {
return fmt.Sprintf("D%d", int(r-D0))
}
return fmt.Sprintf("Reg(%d)", int(r))
}
// A RegX represents a fraction of a multi-value register.
// The Index field specifies the index number,
// but the size of the fraction is not specified.
// It must be inferred from the instruction and the register type.
// For example, in a VMOV instruction, RegX{D5, 1} represents
// the top 32 bits of the 64-bit D5 register.
type RegX struct {
Reg Reg
Index int
}
func (RegX) IsArg() {}
func (r RegX) String() string {
return fmt.Sprintf("%s[%d]", r.Reg, r.Index)
}
// A RegList is a register list.
// Bits at indexes x = 0 through 15 indicate whether the corresponding Rx register is in the list.
type RegList uint16
func (RegList) IsArg() {}
func (r RegList) String() string {
var buf bytes.Buffer
fmt.Fprintf(&buf, "{")
sep := ""
for i := 0; i < 16; i++ {
if r&(1<<uint(i)) != 0 {
fmt.Fprintf(&buf, "%s%s", sep, Reg(i).String())
sep = ","
}
}
fmt.Fprintf(&buf, "}")
return buf.String()
}
// An Endian is the argument to the SETEND instruction.
type Endian uint8
const (
LittleEndian Endian = 0
BigEndian Endian = 1
)
func (Endian) IsArg() {}
func (e Endian) String() string {
if e != 0 {
return "BE"
}
return "LE"
}
// A Shift describes an ARM shift operation.
type Shift uint8
const (
ShiftLeft Shift = 0 // left shift
ShiftRight Shift = 1 // logical (unsigned) right shift
ShiftRightSigned Shift = 2 // arithmetic (signed) right shift
RotateRight Shift = 3 // right rotate
RotateRightExt Shift = 4 // right rotate through carry (Count will always be 1)
)
var shiftName = [...]string{
"LSL", "LSR", "ASR", "ROR", "RRX",
}
func (s Shift) String() string {
if s < 5 {
return shiftName[s]
}
return fmt.Sprintf("Shift(%d)", int(s))
}
// A RegShift is a register shifted by a constant.
type RegShift struct {
Reg Reg
Shift Shift
Count uint8
}
func (RegShift) IsArg() {}
func (r RegShift) String() string {
return fmt.Sprintf("%s %s #%d", r.Reg, r.Shift, r.Count)
}
// A RegShiftReg is a register shifted by a register.
type RegShiftReg struct {
Reg Reg
Shift Shift
RegCount Reg
}
func (RegShiftReg) IsArg() {}
func (r RegShiftReg) String() string {
return fmt.Sprintf("%s %s %s", r.Reg, r.Shift, r.RegCount)
}
// A PCRel describes a memory address (usually a code label)
// as a distance relative to the program counter.
// TODO(rsc): Define which program counter (PC+4? PC+8? PC?).
type PCRel int32
func (PCRel) IsArg() {}
func (r PCRel) String() string {
return fmt.Sprintf("PC%+#x", int32(r))
}
// An AddrMode is an ARM addressing mode.
type AddrMode uint8
const (
_ AddrMode = iota
AddrPostIndex // [R], X use address R, set R = R + X
AddrPreIndex // [R, X]! use address R + X, set R = R + X
AddrOffset // [R, X] use address R + X
AddrLDM // R [R] but formats as R, for LDM/STM only
AddrLDM_WB // R! - [R], X where X is instruction-specific amount, for LDM/STM only
)
// A Mem is a memory reference made up of a base R and index expression X.
// The effective memory address is R or R+X depending on AddrMode.
// The index expression is X = Sign*(Index Shift Count) + Offset,
// but in any instruction either Sign = 0 or Offset = 0.
type Mem struct {
Base Reg
Mode AddrMode
Sign int8
Index Reg
Shift Shift
Count uint8
Offset int16
}
func (Mem) IsArg() {}
func (m Mem) String() string {
R := m.Base.String()
X := ""
if m.Sign != 0 {
X = "+"
if m.Sign < 0 {
X = "-"
}
X += m.Index.String()
if m.Shift != ShiftLeft || m.Count != 0 {
X += fmt.Sprintf(", %s #%d", m.Shift, m.Count)
}
} else {
X = fmt.Sprintf("#%d", m.Offset)
}
switch m.Mode {
case AddrOffset:
if X == "#0" {
return fmt.Sprintf("[%s]", R)
}
return fmt.Sprintf("[%s, %s]", R, X)
case AddrPreIndex:
return fmt.Sprintf("[%s, %s]!", R, X)
case AddrPostIndex:
return fmt.Sprintf("[%s], %s", R, X)
case AddrLDM:
if X == "#0" {
return R
}
case AddrLDM_WB:
if X == "#0" {
return R + "!"
}
}
return fmt.Sprintf("[%s Mode(%d) %s]", R, int(m.Mode), X)
}
@@ -0,0 +1,268 @@
// Copyright 2014 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package armasm
import (
"encoding/binary"
"strings"
"testing"
)
func TestObjdumpARMTestdata(t *testing.T) { testObjdumpARM(t, testdataCases(t)) }
func TestObjdumpARMManual(t *testing.T) { testObjdumpARM(t, hexCases(t, objdumpManualTests)) }
func TestObjdumpARMCond(t *testing.T) { testObjdumpARM(t, condCases(t)) }
func TestObjdumpARMUncond(t *testing.T) { testObjdumpARM(t, uncondCases(t)) }
func TestObjdumpARMVFP(t *testing.T) { testObjdumpARM(t, vfpCases(t)) }
// objdumpManualTests holds test cases that will be run by TestObjdumpARMManual.
// If you are debugging a few cases that turned up in a longer run, it can be useful
// to list them here and then use -run=Manual, particularly with tracing enabled.
// Note that these are byte sequences, so they must be reversed from the usual
// word presentation.
var objdumpManualTests = `
002a9b1d
001b9bed
020b8ded
003a9b1d
060b8ded
fcde1100
b4de1100
bc480000
0b008de7
0b00ade7
fdbcfaf7
`
// allowedMismatchObjdump reports whether the mismatch between text and dec
// should be allowed by the test.
func allowedMismatchObjdump(text string, size int, inst *Inst, dec ExtInst) bool {
if hasPrefix(text, "error:") {
if hasPrefix(dec.text, unsupported...) || strings.Contains(dec.text, "invalid:") || strings.HasSuffix(dec.text, "^") || strings.Contains(dec.text, "f16.f64") || strings.Contains(dec.text, "f64.f16") {
return true
}
// word 4320F02C: libopcodes says 'nopmi {44}'.
if hasPrefix(dec.text, "nop") && strings.Contains(dec.text, "{") {
return true
}
}
if hasPrefix(dec.text, "error:") && text == "undef" && inst.Enc == 0xf7fabcfd {
return true
}
// word 00f02053: libopcodes says 'noppl {0}'.
if hasPrefix(dec.text, "nop") && hasPrefix(text, "nop") && dec.text == text+" {0}" {
return true
}
// word F57FF04F. we say 'dsb #15', libopcodes says 'dsb sy'.
if hasPrefix(text, "dsb") && hasPrefix(dec.text, "dsb") {
return true
}
// word F57FF06F. we say 'isb #15', libopcodes says 'isb sy'.
if hasPrefix(text, "isb") && hasPrefix(dec.text, "isb") {
return true
}
// word F57FF053. we say 'dmb #3', libopcodes says 'dmb osh'.
if hasPrefix(text, "dmb") && hasPrefix(dec.text, "dmb") {
return true
}
// word 992D0000. push/stmdb with no registers (undefined).
// we say 'stmdbls sp!, {}', libopcodes says 'pushls {}'.
if hasPrefix(text, "stmdb") && hasPrefix(dec.text, "push") && strings.Contains(text, "{}") && strings.Contains(dec.text, "{}") {
return true
}
// word 28BD0000. pop/ldm with no registers (undefined).
// we say 'ldmcs sp!, {}', libopcodes says 'popcs {}'.
if hasPrefix(text, "ldm") && hasPrefix(dec.text, "pop") && strings.Contains(text, "{}") && strings.Contains(dec.text, "{}") {
return true
}
// word 014640F0.
// libopcodes emits #-0 for negative zero; we don't.
if strings.Replace(dec.text, "#-0", "#0", -1) == text || strings.Replace(dec.text, ", #-0", "", -1) == text {
return true
}
// word 91EF90F0. we say 'strdls r9, [pc, #0]!' but libopcodes says 'strdls r9, [pc]'.
// word D16F60F0. we say 'strdle r6, [pc, #0]!' but libopcodes says 'strdle r6, [pc, #-0]'.
if strings.Replace(text, ", #0]!", "]", -1) == strings.Replace(dec.text, ", #-0]", "]", -1) {
return true
}
// word 510F4000. we say apsr, libopcodes says CPSR.
if strings.Replace(dec.text, "CPSR", "apsr", -1) == text {
return true
}
// word 06A4B059.
// for ssat and usat, libopcodes decodes asr #0 as asr #0 but the manual seems to say it should be asr #32.
// There is never an asr #0.
if strings.Replace(dec.text, ", asr #0", ", asr #32", -1) == text {
return true
}
if len(dec.enc) >= 4 {
raw := binary.LittleEndian.Uint32(dec.enc[:4])
// word 21FFF0B5.
// the manual is clear that this is pre-indexed mode (with !) but libopcodes generates post-index (without !).
if raw&0x01200000 == 0x01200000 && strings.Replace(text, "!", "", -1) == dec.text {
return true
}
// word C100543E: libopcodes says tst, but no evidence for that.
if strings.HasPrefix(dec.text, "tst") && raw&0x0ff00000 != 0x03100000 && raw&0x0ff00000 != 0x01100000 {
return true
}
// word C3203CE8: libopcodes says teq, but no evidence for that.
if strings.HasPrefix(dec.text, "teq") && raw&0x0ff00000 != 0x03300000 && raw&0x0ff00000 != 0x01300000 {
return true
}
// word D14C552E: libopcodes says cmp but no evidence for that.
if strings.HasPrefix(dec.text, "cmp") && raw&0x0ff00000 != 0x03500000 && raw&0x0ff00000 != 0x01500000 {
return true
}
// word 2166AA4A: libopcodes says cmn but no evidence for that.
if strings.HasPrefix(dec.text, "cmn") && raw&0x0ff00000 != 0x03700000 && raw&0x0ff00000 != 0x01700000 {
return true
}
// word E70AEEEF: libopcodes says str but no evidence for that.
if strings.HasPrefix(dec.text, "str") && len(dec.text) >= 5 && (dec.text[3] == ' ' || dec.text[5] == ' ') && raw&0x0e500018 != 0x06000000 && raw&0x0e500000 != 0x0400000 {
return true
}
// word B0AF48F4: libopcodes says strd but P=0,W=1 which is unpredictable.
if hasPrefix(dec.text, "ldr", "str") && raw&0x01200000 == 0x00200000 {
return true
}
// word B6CC1C76: libopcodes inexplicably says 'uxtab16lt r1, ip, r6, ROR #24' instead of 'uxtab16lt r1, ip, r6, ror #24'
if strings.ToLower(dec.text) == text {
return true
}
// word F410FDA1: libopcodes says PLDW but the manual is clear that PLDW is F5/F7, not F4.
// word F7D0FB17: libopcodes says PLDW but the manual is clear that PLDW has 0x10 clear
if hasPrefix(dec.text, "pld") && raw&0xfd000010 != 0xf5000000 {
return true
}
// word F650FE14: libopcodes says PLI but the manual is clear that PLI has 0x10 clear
if hasPrefix(dec.text, "pli") && raw&0xff000010 != 0xf6000000 {
return true
}
}
return false
}
// Instructions known to libopcodes (or xed) but not to us.
// Most of these are floating point coprocessor instructions.
var unsupported = strings.Fields(`
abs
acs
adf
aes
asn
atn
cdp
cf
cmf
cnf
cos
cps
crc32
dvf
eret
exp
fadd
fcmp
fcpy
fcvt
fdiv
fdv
fix
fld
flt
fmac
fmd
fml
fmr
fms
fmul
fmx
fneg
fnm
frd
fsit
fsq
fst
fsu
fto
fui
hlt
hvc
lda
ldc
ldf
lfm
lgn
log
mar
mcr
mcrr
mia
mnf
mra
mrc
mrrc
mrs
msr
msr
muf
mvf
nrm
pol
pow
rdf
rfc
rfe
rfs
rmf
rnd
rpw
rsf
sdiv
sev
sfm
sha1
sha256
sin
smc
sqt
srs
stc
stf
stl
suf
tan
udf
udiv
urd
vfma
vfms
vfnma
vfnms
vrint
wfc
wfs
`)
@@ -0,0 +1,259 @@
// Copyright 2014 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Copied and simplified from ../../x86/x86asm/objdumpext_test.go.
package armasm
import (
"bytes"
"debug/elf"
"encoding/binary"
"fmt"
"io"
"log"
"os"
"strconv"
"strings"
"testing"
)
const objdumpPath = "/usr/local/bin/arm-linux-elf-objdump"
func testObjdumpARM(t *testing.T, generate func(func([]byte))) {
testObjdumpArch(t, generate, ModeARM)
}
func testObjdumpArch(t *testing.T, generate func(func([]byte)), arch Mode) {
if testing.Short() {
t.Skip("skipping objdump test in short mode")
}
if _, err := os.Stat(objdumpPath); err != nil {
t.Skip(err)
}
testExtDis(t, "gnu", arch, objdump, generate, allowedMismatchObjdump)
}
func objdump(ext *ExtDis) error {
// File already written with instructions; add ELF header.
if ext.Arch == ModeARM {
if err := writeELF32(ext.File, ext.Size); err != nil {
return err
}
} else {
panic("unknown arch")
}
b, err := ext.Run(objdumpPath, "-d", "-z", ext.File.Name())
if err != nil {
return err
}
var (
nmatch int
reading bool
next uint32 = start
addr uint32
encbuf [4]byte
enc []byte
text string
)
flush := func() {
if addr == next {
if m := pcrel.FindStringSubmatch(text); m != nil {
targ, _ := strconv.ParseUint(m[2], 16, 64)
text = fmt.Sprintf("%s .%+#x", m[1], int32(uint32(targ)-addr-uint32(len(enc))))
}
if strings.HasPrefix(text, "stmia") {
text = "stm" + text[5:]
}
if strings.HasPrefix(text, "stmfd") {
text = "stmdb" + text[5:]
}
if strings.HasPrefix(text, "ldmfd") {
text = "ldm" + text[5:]
}
text = strings.Replace(text, "#0.0", "#0", -1)
if text == "undefined" && len(enc) == 4 {
text = "error: unknown instruction"
enc = nil
}
if len(enc) == 4 {
// prints as word but we want to record bytes
enc[0], enc[3] = enc[3], enc[0]
enc[1], enc[2] = enc[2], enc[1]
}
ext.Dec <- ExtInst{addr, encbuf, len(enc), text}
encbuf = [4]byte{}
enc = nil
next += 4
}
}
var textangle = []byte("<.text>:")
for {
line, err := b.ReadSlice('\n')
if err != nil {
if err == io.EOF {
break
}
return fmt.Errorf("reading objdump output: %v", err)
}
if bytes.Contains(line, textangle) {
reading = true
continue
}
if !reading {
continue
}
if debug {
os.Stdout.Write(line)
}
if enc1 := parseContinuation(line, encbuf[:len(enc)]); enc1 != nil {
enc = enc1
continue
}
flush()
nmatch++
addr, enc, text = parseLine(line, encbuf[:0])
if addr > next {
return fmt.Errorf("address out of sync expected <= %#x at %q in:\n%s", next, line, line)
}
}
flush()
if next != start+uint32(ext.Size) {
return fmt.Errorf("not enough results found [%d %d]", next, start+ext.Size)
}
if err := ext.Wait(); err != nil {
return fmt.Errorf("exec: %v", err)
}
return nil
}
var (
undefined = []byte("<UNDEFINED>")
unpredictable = []byte("<UNPREDICTABLE>")
illegalShifter = []byte("<illegal shifter operand>")
)
func parseLine(line []byte, encstart []byte) (addr uint32, enc []byte, text string) {
oline := line
i := index(line, ":\t")
if i < 0 {
log.Fatalf("cannot parse disassembly: %q", oline)
}
x, err := strconv.ParseUint(string(trimSpace(line[:i])), 16, 32)
if err != nil {
log.Fatalf("cannot parse disassembly: %q", oline)
}
addr = uint32(x)
line = line[i+2:]
i = bytes.IndexByte(line, '\t')
if i < 0 {
log.Fatalf("cannot parse disassembly: %q", oline)
}
enc, ok := parseHex(line[:i], encstart)
if !ok {
log.Fatalf("cannot parse disassembly: %q", oline)
}
line = trimSpace(line[i:])
if bytes.Contains(line, undefined) {
text = "undefined"
return
}
if bytes.Contains(line, illegalShifter) {
text = "undefined"
return
}
if false && bytes.Contains(line, unpredictable) {
text = "unpredictable"
return
}
if i := bytes.IndexByte(line, ';'); i >= 0 {
line = trimSpace(line[:i])
}
text = string(fixSpace(line))
return
}
func parseContinuation(line []byte, enc []byte) []byte {
i := index(line, ":\t")
if i < 0 {
return nil
}
line = line[i+1:]
enc, _ = parseHex(line, enc)
return enc
}
// writeELF32 writes an ELF32 header to the file,
// describing a text segment that starts at start
// and extends for size bytes.
func writeELF32(f *os.File, size int) error {
f.Seek(0, io.SeekStart)
var hdr elf.Header32
var prog elf.Prog32
var sect elf.Section32
var buf bytes.Buffer
binary.Write(&buf, binary.LittleEndian, &hdr)
off1 := buf.Len()
binary.Write(&buf, binary.LittleEndian, &prog)
off2 := buf.Len()
binary.Write(&buf, binary.LittleEndian, &sect)
off3 := buf.Len()
buf.Reset()
data := byte(elf.ELFDATA2LSB)
hdr = elf.Header32{
Ident: [16]byte{0x7F, 'E', 'L', 'F', 1, data, 1},
Type: 2,
Machine: uint16(elf.EM_ARM),
Version: 1,
Entry: start,
Phoff: uint32(off1),
Shoff: uint32(off2),
Flags: 0x05000002,
Ehsize: uint16(off1),
Phentsize: uint16(off2 - off1),
Phnum: 1,
Shentsize: uint16(off3 - off2),
Shnum: 3,
Shstrndx: 2,
}
binary.Write(&buf, binary.LittleEndian, &hdr)
prog = elf.Prog32{
Type: 1,
Off: start,
Vaddr: start,
Paddr: start,
Filesz: uint32(size),
Memsz: uint32(size),
Flags: 5,
Align: start,
}
binary.Write(&buf, binary.LittleEndian, &prog)
binary.Write(&buf, binary.LittleEndian, &sect) // NULL section
sect = elf.Section32{
Name: 1,
Type: uint32(elf.SHT_PROGBITS),
Addr: start,
Off: start,
Size: uint32(size),
Flags: uint32(elf.SHF_ALLOC | elf.SHF_EXECINSTR),
Addralign: 4,
}
binary.Write(&buf, binary.LittleEndian, &sect) // .text
sect = elf.Section32{
Name: uint32(len("\x00.text\x00")),
Type: uint32(elf.SHT_STRTAB),
Addr: 0,
Off: uint32(off2 + (off3-off2)*3),
Size: uint32(len("\x00.text\x00.shstrtab\x00")),
Addralign: 1,
}
binary.Write(&buf, binary.LittleEndian, &sect)
buf.WriteString("\x00.text\x00.shstrtab\x00")
f.Write(buf.Bytes())
return nil
}
@@ -0,0 +1,401 @@
// Copyright 2014 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package armasm
import (
"bytes"
"encoding/binary"
"fmt"
"io"
"math"
"strings"
)
// GoSyntax returns the Go assembler syntax for the instruction.
// The syntax was originally defined by Plan 9.
// The pc is the program counter of the instruction, used for expanding
// PC-relative addresses into absolute ones.
// The symname function queries the symbol table for the program
// being disassembled. Given a target address it returns the name and base
// address of the symbol containing the target, if any; otherwise it returns "", 0.
// The reader r should read from the text segment using text addresses
// as offsets; it is used to display pc-relative loads as constant loads.
func GoSyntax(inst Inst, pc uint64, symname func(uint64) (string, uint64), text io.ReaderAt) string {
if symname == nil {
symname = func(uint64) (string, uint64) { return "", 0 }
}
var args []string
for _, a := range inst.Args {
if a == nil {
break
}
args = append(args, plan9Arg(&inst, pc, symname, a))
}
op := inst.Op.String()
switch inst.Op &^ 15 {
case LDR_EQ, LDRB_EQ, LDRH_EQ, LDRSB_EQ, LDRSH_EQ, VLDR_EQ:
// Check for RET
reg, _ := inst.Args[0].(Reg)
mem, _ := inst.Args[1].(Mem)
if inst.Op&^15 == LDR_EQ && reg == R15 && mem.Base == SP && mem.Sign == 0 && mem.Mode == AddrPostIndex {
return fmt.Sprintf("RET%s #%d", op[3:], mem.Offset)
}
// Check for PC-relative load.
if mem.Base == PC && mem.Sign == 0 && mem.Mode == AddrOffset && text != nil {
addr := uint32(pc) + 8 + uint32(mem.Offset)
buf := make([]byte, 8)
switch inst.Op &^ 15 {
case LDRB_EQ, LDRSB_EQ:
if _, err := text.ReadAt(buf[:1], int64(addr)); err != nil {
break
}
args[1] = fmt.Sprintf("$%#x", buf[0])
case LDRH_EQ, LDRSH_EQ:
if _, err := text.ReadAt(buf[:2], int64(addr)); err != nil {
break
}
args[1] = fmt.Sprintf("$%#x", binary.LittleEndian.Uint16(buf))
case LDR_EQ:
if _, err := text.ReadAt(buf[:4], int64(addr)); err != nil {
break
}
x := binary.LittleEndian.Uint32(buf)
if s, base := symname(uint64(x)); s != "" && uint64(x) == base {
args[1] = fmt.Sprintf("$%s(SB)", s)
} else {
args[1] = fmt.Sprintf("$%#x", x)
}
case VLDR_EQ:
switch {
case strings.HasPrefix(args[0], "D"): // VLDR.F64
if _, err := text.ReadAt(buf, int64(addr)); err != nil {
break
}
args[1] = fmt.Sprintf("$%f", math.Float64frombits(binary.LittleEndian.Uint64(buf)))
case strings.HasPrefix(args[0], "S"): // VLDR.F32
if _, err := text.ReadAt(buf[:4], int64(addr)); err != nil {
break
}
args[1] = fmt.Sprintf("$%f", math.Float32frombits(binary.LittleEndian.Uint32(buf)))
default:
panic(fmt.Sprintf("wrong FP register: %v", inst))
}
}
}
}
// Move addressing mode into opcode suffix.
suffix := ""
switch inst.Op &^ 15 {
case PLD, PLI, PLD_W:
if mem, ok := inst.Args[0].(Mem); ok {
args[0], suffix = memOpTrans(mem)
} else {
panic(fmt.Sprintf("illegal instruction: %v", inst))
}
case LDR_EQ, LDRB_EQ, LDRSB_EQ, LDRH_EQ, LDRSH_EQ, STR_EQ, STRB_EQ, STRH_EQ, VLDR_EQ, VSTR_EQ, LDREX_EQ, LDREXH_EQ, LDREXB_EQ:
if mem, ok := inst.Args[1].(Mem); ok {
args[1], suffix = memOpTrans(mem)
} else {
panic(fmt.Sprintf("illegal instruction: %v", inst))
}
case SWP_EQ, SWP_B_EQ, STREX_EQ, STREXB_EQ, STREXH_EQ:
if mem, ok := inst.Args[2].(Mem); ok {
args[2], suffix = memOpTrans(mem)
} else {
panic(fmt.Sprintf("illegal instruction: %v", inst))
}
}
// Reverse args, placing dest last.
for i, j := 0, len(args)-1; i < j; i, j = i+1, j-1 {
args[i], args[j] = args[j], args[i]
}
// For MLA-like instructions, the addend is the third operand.
switch inst.Op &^ 15 {
case SMLAWT_EQ, SMLAWB_EQ, MLA_EQ, MLA_S_EQ, MLS_EQ, SMMLA_EQ, SMMLS_EQ, SMLABB_EQ, SMLATB_EQ, SMLABT_EQ, SMLATT_EQ, SMLAD_EQ, SMLAD_X_EQ, SMLSD_EQ, SMLSD_X_EQ:
args = []string{args[1], args[2], args[0], args[3]}
}
// For STREX like instructions, the memory operands comes first.
switch inst.Op &^ 15 {
case STREX_EQ, STREXB_EQ, STREXH_EQ, SWP_EQ, SWP_B_EQ:
args = []string{args[1], args[0], args[2]}
}
// special process for FP instructions
op, args = fpTrans(&inst, op, args)
// LDR/STR like instructions -> MOV like
switch inst.Op &^ 15 {
case MOV_EQ:
op = "MOVW" + op[3:]
case LDR_EQ, MSR_EQ, MRS_EQ:
op = "MOVW" + op[3:] + suffix
case VMRS_EQ, VMSR_EQ:
op = "MOVW" + op[4:] + suffix
case LDRB_EQ, UXTB_EQ:
op = "MOVBU" + op[4:] + suffix
case LDRSB_EQ:
op = "MOVBS" + op[5:] + suffix
case SXTB_EQ:
op = "MOVBS" + op[4:] + suffix
case LDRH_EQ, UXTH_EQ:
op = "MOVHU" + op[4:] + suffix
case LDRSH_EQ:
op = "MOVHS" + op[5:] + suffix
case SXTH_EQ:
op = "MOVHS" + op[4:] + suffix
case STR_EQ:
op = "MOVW" + op[3:] + suffix
args[0], args[1] = args[1], args[0]
case STRB_EQ:
op = "MOVB" + op[4:] + suffix
args[0], args[1] = args[1], args[0]
case STRH_EQ:
op = "MOVH" + op[4:] + suffix
args[0], args[1] = args[1], args[0]
case VSTR_EQ:
args[0], args[1] = args[1], args[0]
default:
op = op + suffix
}
if args != nil {
op += " " + strings.Join(args, ", ")
}
return op
}
// assembler syntax for the various shifts.
// @x> is a lie; the assembler uses @> 0
// instead of @x> 1, but i wanted to be clear that it
// was a different operation (rotate right extended, not rotate right).
var plan9Shift = []string{"<<", ">>", "->", "@>", "@x>"}
func plan9Arg(inst *Inst, pc uint64, symname func(uint64) (string, uint64), arg Arg) string {
switch a := arg.(type) {
case Endian:
case Imm:
return fmt.Sprintf("$%d", uint32(a))
case Mem:
case PCRel:
addr := uint32(pc) + 8 + uint32(a)
if s, base := symname(uint64(addr)); s != "" && uint64(addr) == base {
return fmt.Sprintf("%s(SB)", s)
}
return fmt.Sprintf("%#x", addr)
case Reg:
if a < 16 {
return fmt.Sprintf("R%d", int(a))
}
case RegList:
var buf bytes.Buffer
start := -2
end := -2
fmt.Fprintf(&buf, "[")
flush := func() {
if start >= 0 {
if buf.Len() > 1 {
fmt.Fprintf(&buf, ",")
}
if start == end {
fmt.Fprintf(&buf, "R%d", start)
} else {
fmt.Fprintf(&buf, "R%d-R%d", start, end)
}
start = -2
end = -2
}
}
for i := 0; i < 16; i++ {
if a&(1<<uint(i)) != 0 {
if i == end+1 {
end++
continue
}
start = i
end = i
} else {
flush()
}
}
flush()
fmt.Fprintf(&buf, "]")
return buf.String()
case RegShift:
return fmt.Sprintf("R%d%s$%d", int(a.Reg), plan9Shift[a.Shift], int(a.Count))
case RegShiftReg:
return fmt.Sprintf("R%d%sR%d", int(a.Reg), plan9Shift[a.Shift], int(a.RegCount))
}
return strings.ToUpper(arg.String())
}
// convert memory operand from GNU syntax to Plan 9 syntax, for example,
// [r5] -> (R5)
// [r6, #4080] -> 0xff0(R6)
// [r2, r0, ror #1] -> (R2)(R0@>1)
// inst [r2, -r0, ror #1] -> INST.U (R2)(R0@>1)
// input:
//
// a memory operand
//
// return values:
//
// corresponding memory operand in Plan 9 syntax
// .W/.P/.U suffix
func memOpTrans(mem Mem) (string, string) {
suffix := ""
switch mem.Mode {
case AddrOffset, AddrLDM:
// no suffix
case AddrPreIndex, AddrLDM_WB:
suffix = ".W"
case AddrPostIndex:
suffix = ".P"
}
off := ""
if mem.Offset != 0 {
off = fmt.Sprintf("%#x", mem.Offset)
}
base := fmt.Sprintf("(R%d)", int(mem.Base))
index := ""
if mem.Sign != 0 {
sign := ""
if mem.Sign < 0 {
suffix += ".U"
}
shift := ""
if mem.Count != 0 {
shift = fmt.Sprintf("%s%d", plan9Shift[mem.Shift], mem.Count)
}
index = fmt.Sprintf("(%sR%d%s)", sign, int(mem.Index), shift)
}
return off + base + index, suffix
}
type goFPInfo struct {
op Op
transArgs []int // indexes of arguments which need transformation
gnuName string // instruction name in GNU syntax
goName string // instruction name in Plan 9 syntax
}
var fpInst []goFPInfo = []goFPInfo{
{VADD_EQ_F32, []int{2, 1, 0}, "VADD", "ADDF"},
{VADD_EQ_F64, []int{2, 1, 0}, "VADD", "ADDD"},
{VSUB_EQ_F32, []int{2, 1, 0}, "VSUB", "SUBF"},
{VSUB_EQ_F64, []int{2, 1, 0}, "VSUB", "SUBD"},
{VMUL_EQ_F32, []int{2, 1, 0}, "VMUL", "MULF"},
{VMUL_EQ_F64, []int{2, 1, 0}, "VMUL", "MULD"},
{VNMUL_EQ_F32, []int{2, 1, 0}, "VNMUL", "NMULF"},
{VNMUL_EQ_F64, []int{2, 1, 0}, "VNMUL", "NMULD"},
{VMLA_EQ_F32, []int{2, 1, 0}, "VMLA", "MULAF"},
{VMLA_EQ_F64, []int{2, 1, 0}, "VMLA", "MULAD"},
{VMLS_EQ_F32, []int{2, 1, 0}, "VMLS", "MULSF"},
{VMLS_EQ_F64, []int{2, 1, 0}, "VMLS", "MULSD"},
{VNMLA_EQ_F32, []int{2, 1, 0}, "VNMLA", "NMULAF"},
{VNMLA_EQ_F64, []int{2, 1, 0}, "VNMLA", "NMULAD"},
{VNMLS_EQ_F32, []int{2, 1, 0}, "VNMLS", "NMULSF"},
{VNMLS_EQ_F64, []int{2, 1, 0}, "VNMLS", "NMULSD"},
{VDIV_EQ_F32, []int{2, 1, 0}, "VDIV", "DIVF"},
{VDIV_EQ_F64, []int{2, 1, 0}, "VDIV", "DIVD"},
{VNEG_EQ_F32, []int{1, 0}, "VNEG", "NEGF"},
{VNEG_EQ_F64, []int{1, 0}, "VNEG", "NEGD"},
{VABS_EQ_F32, []int{1, 0}, "VABS", "ABSF"},
{VABS_EQ_F64, []int{1, 0}, "VABS", "ABSD"},
{VSQRT_EQ_F32, []int{1, 0}, "VSQRT", "SQRTF"},
{VSQRT_EQ_F64, []int{1, 0}, "VSQRT", "SQRTD"},
{VCMP_EQ_F32, []int{1, 0}, "VCMP", "CMPF"},
{VCMP_EQ_F64, []int{1, 0}, "VCMP", "CMPD"},
{VCMP_E_EQ_F32, []int{1, 0}, "VCMP.E", "CMPF"},
{VCMP_E_EQ_F64, []int{1, 0}, "VCMP.E", "CMPD"},
{VLDR_EQ, []int{1}, "VLDR", "MOV"},
{VSTR_EQ, []int{1}, "VSTR", "MOV"},
{VMOV_EQ_F32, []int{1, 0}, "VMOV", "MOVF"},
{VMOV_EQ_F64, []int{1, 0}, "VMOV", "MOVD"},
{VMOV_EQ_32, []int{1, 0}, "VMOV", "MOVW"},
{VMOV_EQ, []int{1, 0}, "VMOV", "MOVW"},
{VCVT_EQ_F64_F32, []int{1, 0}, "VCVT", "MOVFD"},
{VCVT_EQ_F32_F64, []int{1, 0}, "VCVT", "MOVDF"},
{VCVT_EQ_F32_U32, []int{1, 0}, "VCVT", "MOVWF.U"},
{VCVT_EQ_F32_S32, []int{1, 0}, "VCVT", "MOVWF"},
{VCVT_EQ_S32_F32, []int{1, 0}, "VCVT", "MOVFW"},
{VCVT_EQ_U32_F32, []int{1, 0}, "VCVT", "MOVFW.U"},
{VCVT_EQ_F64_U32, []int{1, 0}, "VCVT", "MOVWD.U"},
{VCVT_EQ_F64_S32, []int{1, 0}, "VCVT", "MOVWD"},
{VCVT_EQ_S32_F64, []int{1, 0}, "VCVT", "MOVDW"},
{VCVT_EQ_U32_F64, []int{1, 0}, "VCVT", "MOVDW.U"},
}
// convert FP instructions from GNU syntax to Plan 9 syntax, for example,
// vadd.f32 s0, s3, s4 -> ADDF F0, S3, F2
// vsub.f64 d0, d2, d4 -> SUBD F0, F2, F4
// vldr s2, [r11] -> MOVF (R11), F1
// inputs: instruction name and arguments in GNU syntax
// return values: corresponding instruction name and arguments in Plan 9 syntax
func fpTrans(inst *Inst, op string, args []string) (string, []string) {
for _, fp := range fpInst {
if inst.Op&^15 == fp.op {
// remove gnu syntax suffixes
op = strings.Replace(op, ".F32", "", -1)
op = strings.Replace(op, ".F64", "", -1)
op = strings.Replace(op, ".S32", "", -1)
op = strings.Replace(op, ".U32", "", -1)
op = strings.Replace(op, ".32", "", -1)
// compose op name
if fp.op == VLDR_EQ || fp.op == VSTR_EQ {
switch {
case strings.HasPrefix(args[fp.transArgs[0]], "D"):
op = "MOVD" + op[len(fp.gnuName):]
case strings.HasPrefix(args[fp.transArgs[0]], "S"):
op = "MOVF" + op[len(fp.gnuName):]
default:
panic(fmt.Sprintf("wrong FP register: %v", inst))
}
} else {
op = fp.goName + op[len(fp.gnuName):]
}
// transform registers
for ix, ri := range fp.transArgs {
switch {
case strings.HasSuffix(args[ri], "[1]"): // MOVW Rx, Dy[1]
break
case strings.HasSuffix(args[ri], "[0]"): // Dx[0] -> Fx
args[ri] = strings.Replace(args[ri], "[0]", "", -1)
fallthrough
case strings.HasPrefix(args[ri], "D"): // Dx -> Fx
args[ri] = "F" + args[ri][1:]
case strings.HasPrefix(args[ri], "S"):
if inst.Args[ix].(Reg)&1 == 0 { // Sx -> Fy, y = x/2, if x is even
args[ri] = fmt.Sprintf("F%d", (inst.Args[ix].(Reg)-S0)/2)
}
case strings.HasPrefix(args[ri], "$"): // CMPF/CMPD $0, Fx
break
case strings.HasPrefix(args[ri], "R"): // MOVW Rx, Dy[1]
break
default:
panic(fmt.Sprintf("wrong FP register: %v", inst))
}
}
break
}
}
return op, args
}
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,5 @@
newdecode.txt:
cd ..; go test -cover -run 'ObjdumpARMCond' -v -timeout 10h -printtests -long 2>&1 | tee log
cd ..; go test -cover -run 'ObjdumpARMUncond' -v -timeout 10h -printtests -long 2>&1 | tee -a log
egrep ' (gnu|plan9) ' ../log |sort >newdecode.txt
File diff suppressed because it is too large Load Diff