syzkaller/prog/rand.go
2017-12-31 12:29:08 +01:00

711 lines
17 KiB
Go

// Copyright 2015/2016 syzkaller project authors. All rights reserved.
// Use of this source code is governed by Apache 2 LICENSE that can be found in the LICENSE file.
package prog
import (
"bytes"
"fmt"
"math"
"math/rand"
"strings"
"sync"
"github.com/google/syzkaller/pkg/ifuzz"
)
var pageStartPool = sync.Pool{New: func() interface{} { return new([]uint64) }}
type randGen struct {
*rand.Rand
target *Target
inCreateResource bool
recDepth map[string]int
}
func newRand(target *Target, rs rand.Source) *randGen {
return &randGen{
Rand: rand.New(rs),
target: target,
recDepth: make(map[string]int),
}
}
func (r *randGen) rand(n int) uint64 {
return uint64(r.Intn(n))
}
func (r *randGen) randRange(begin, end uint64) uint64 {
return begin + uint64(r.Intn(int(end-begin+1)))
}
func (r *randGen) bin() bool {
return r.Intn(2) == 0
}
func (r *randGen) oneOf(n int) bool {
return r.Intn(n) == 0
}
func (r *randGen) rand64() uint64 {
v := uint64(r.Int63())
if r.bin() {
v |= 1 << 63
}
return v
}
// Some potentially interesting integers.
var specialInts = []uint64{
0, 1, 31, 32, 63, 64, 127, 128,
129, 255, 256, 257, 511, 512,
1023, 1024, 1025, 2047, 2048, 4095, 4096,
(1 << 15) - 1, (1 << 15), (1 << 15) + 1,
(1 << 16) - 1, (1 << 16), (1 << 16) + 1,
(1 << 31) - 1, (1 << 31), (1 << 31) + 1,
(1 << 32) - 1, (1 << 32), (1 << 32) + 1,
}
func (r *randGen) randInt() uint64 {
v := r.rand64()
switch {
case r.nOutOf(100, 182):
v %= 10
case r.nOutOf(50, 82):
v = specialInts[r.Intn(len(specialInts))]
case r.nOutOf(10, 32):
v %= 256
case r.nOutOf(10, 22):
v %= 4 << 10
case r.nOutOf(10, 12):
v %= 64 << 10
default:
v %= 1 << 31
}
switch {
case r.nOutOf(100, 107):
case r.nOutOf(5, 7):
v = uint64(-int64(v))
default:
v <<= uint(r.Intn(63))
}
return v
}
func (r *randGen) randRangeInt(begin uint64, end uint64) uint64 {
if r.oneOf(100) {
return r.randInt()
}
return begin + (r.Uint64() % (end - begin + 1))
}
// biasedRand returns a random int in range [0..n),
// probability of n-1 is k times higher than probability of 0.
func (r *randGen) biasedRand(n, k int) int {
nf, kf := float64(n), float64(k)
rf := nf * (kf/2 + 1) * r.Float64()
bf := (-1 + math.Sqrt(1+2*kf*rf/nf)) * nf / kf
return int(bf)
}
func (r *randGen) randArrayLen() uint64 {
const maxLen = 10
// biasedRand produces: 10, 9, ..., 1, 0,
// we want: 1, 2, ..., 9, 10, 0
return uint64(maxLen-r.biasedRand(maxLen+1, 10)+1) % (maxLen + 1)
}
func (r *randGen) randBufLen() (n uint64) {
switch {
case r.nOutOf(50, 56):
n = r.rand(256)
case r.nOutOf(5, 6):
n = 4 << 10
}
return
}
func (r *randGen) randPageCount() (n uint64) {
switch {
case r.nOutOf(100, 106):
n = r.rand(4) + 1
case r.nOutOf(5, 6):
n = r.rand(20) + 1
default:
n = (r.rand(3) + 1) * 1024
}
return
}
func (r *randGen) flags(vv []uint64) (v uint64) {
switch {
case r.nOutOf(90, 111):
for stop := false; !stop; stop = r.bin() {
v |= vv[r.rand(len(vv))]
}
case r.nOutOf(10, 21):
v = vv[r.rand(len(vv))]
case r.nOutOf(10, 11):
v = 0
default:
v = r.rand64()
}
return
}
func (r *randGen) filename(s *state) string {
// TODO: support procfs and sysfs
dir := "."
if r.oneOf(2) && len(s.files) != 0 {
files := make([]string, 0, len(s.files))
for f := range s.files {
files = append(files, f)
}
dir = files[r.Intn(len(files))]
if len(dir) > 0 && dir[len(dir)-1] == 0 {
dir = dir[:len(dir)-1]
}
}
if len(s.files) == 0 || r.oneOf(10) {
// Generate a new name.
for i := 0; ; i++ {
f := fmt.Sprintf("%v/file%v\x00", dir, i)
if !s.files[f] {
return f
}
}
}
files := make([]string, 0, len(s.files))
for f := range s.files {
files = append(files, f)
}
return files[r.Intn(len(files))]
}
func (r *randGen) randString(s *state, vals []string, dir Dir) []byte {
data := r.randStringImpl(s, vals)
if dir == DirOut {
for i := range data {
data[i] = 0
}
}
return data
}
func (r *randGen) randStringImpl(s *state, vals []string) []byte {
if len(vals) != 0 {
return []byte(vals[r.Intn(len(vals))])
}
if len(s.strings) != 0 && r.bin() {
// Return an existing string.
strings := make([]string, 0, len(s.strings))
for s := range s.strings {
strings = append(strings, s)
}
return []byte(strings[r.Intn(len(strings))])
}
punct := []byte{'!', '@', '#', '$', '%', '^', '&', '*', '(', ')', '-', '+', '\\',
'/', ':', '.', ',', '-', '\'', '[', ']', '{', '}'}
buf := new(bytes.Buffer)
for r.nOutOf(3, 4) {
switch {
case r.nOutOf(10, 21):
dict := r.target.StringDictionary
if len(dict) != 0 {
buf.WriteString(dict[r.Intn(len(dict))])
}
case r.nOutOf(10, 11):
buf.Write([]byte{punct[r.Intn(len(punct))]})
default:
buf.Write([]byte{byte(r.Intn(256))})
}
}
if !r.oneOf(100) {
buf.Write([]byte{0})
}
return buf.Bytes()
}
func (r *randGen) addr1(s *state, typ Type, size uint64, data Arg) (Arg, []*Call) {
npages := (size + r.target.PageSize - 1) / r.target.PageSize
if npages == 0 {
npages = 1
}
if r.bin() {
return r.randPageAddr(s, typ, npages, data, false), nil
}
for i := uint64(0); i < maxPages-npages; i++ {
free := true
for j := uint64(0); j < npages; j++ {
if s.pages[i+j] {
free = false
break
}
}
if !free {
continue
}
c := r.target.MakeMmap(i, npages)
return MakePointerArg(typ, i, 0, 0, data), []*Call{c}
}
return r.randPageAddr(s, typ, npages, data, false), nil
}
func (r *randGen) addr(s *state, typ Type, size uint64, data Arg) (Arg, []*Call) {
arg, calls := r.addr1(s, typ, size, data)
a, ok := arg.(*PointerArg)
if !ok {
panic("bad")
}
// Patch offset of the address.
switch {
case r.nOutOf(50, 102):
case r.nOutOf(50, 52):
a.PageOffset = -int(size)
case r.nOutOf(1, 2):
a.PageOffset = r.Intn(int(r.target.PageSize))
default:
if size > 0 {
a.PageOffset = -r.Intn(int(size))
}
}
return arg, calls
}
func (r *randGen) randPageAddr(s *state, typ Type, npages uint64, data Arg, vma bool) Arg {
poolPtr := pageStartPool.Get().(*[]uint64)
starts := (*poolPtr)[:0]
for i := uint64(0); i < maxPages-npages; i++ {
busy := true
for j := uint64(0); j < npages; j++ {
if !s.pages[i+j] {
busy = false
break
}
}
// TODO: it does not need to be completely busy,
// for example, mmap addr arg can be new memory.
if !busy {
continue
}
starts = append(starts, i)
}
var page uint64
if len(starts) != 0 {
page = starts[r.rand(len(starts))]
} else {
page = r.rand(int(maxPages - npages))
}
if !vma {
npages = 0
}
*poolPtr = starts
pageStartPool.Put(poolPtr)
return MakePointerArg(typ, page, 0, npages, data)
}
func (r *randGen) createResource(s *state, res *ResourceType) (arg Arg, calls []*Call) {
if r.inCreateResource {
special := res.SpecialValues()
return MakeResultArg(res, nil, special[r.Intn(len(special))]), nil
}
r.inCreateResource = true
defer func() { r.inCreateResource = false }()
kind := res.Desc.Name
if r.oneOf(1000) {
// Spoof resource subkind.
var all []string
for kind1 := range r.target.resourceMap {
if r.target.isCompatibleResource(res.Desc.Kind[0], kind1) {
all = append(all, kind1)
}
}
kind = all[r.Intn(len(all))]
}
// Find calls that produce the necessary resources.
metas0 := r.target.resourceCtors[kind]
// TODO: reduce priority of less specialized ctors.
var metas []*Syscall
for _, meta := range metas0 {
if s.ct == nil || s.ct.run[meta.ID] == nil {
continue
}
metas = append(metas, meta)
}
if len(metas) == 0 {
return MakeResultArg(res, nil, res.Default()), nil
}
// Now we have a set of candidate calls that can create the necessary resource.
for i := 0; i < 1e3; i++ {
// Generate one of them.
meta := metas[r.Intn(len(metas))]
calls := r.generateParticularCall(s, meta)
s1 := newState(r.target, s.ct)
s1.analyze(calls[len(calls)-1])
// Now see if we have what we want.
var allres []Arg
for kind1, res1 := range s1.resources {
if r.target.isCompatibleResource(kind, kind1) {
allres = append(allres, res1...)
}
}
if len(allres) != 0 {
// Bingo!
arg := MakeResultArg(res, allres[r.Intn(len(allres))], 0)
return arg, calls
}
// Discard unsuccessful calls.
for _, c := range calls {
foreachArg(c, func(arg, _ Arg, _ *[]Arg) {
if a, ok := arg.(*ResultArg); ok && a.Res != nil {
delete(*a.Res.(ArgUsed).Used(), arg)
}
})
}
}
// Generally we can loop several times, e.g. when we choose a call that returns
// the resource in an array, but then generateArg generated that array of zero length.
// But we must succeed eventually.
panic("failed to create a resource")
}
func (r *randGen) generateText(kind TextKind) []byte {
switch kind {
case Text_arm64:
// Just a stub, need something better.
text := make([]byte, 50)
for i := range text {
text[i] = byte(r.Intn(256))
}
return text
default:
cfg := createIfuzzConfig(kind)
return ifuzz.Generate(cfg, r.Rand)
}
}
func (r *randGen) mutateText(kind TextKind, text []byte) []byte {
switch kind {
case Text_arm64:
return mutateData(r, text, 40, 60)
default:
cfg := createIfuzzConfig(kind)
return ifuzz.Mutate(cfg, r.Rand, text)
}
}
func createIfuzzConfig(kind TextKind) *ifuzz.Config {
cfg := &ifuzz.Config{
Len: 10,
Priv: true,
Exec: true,
MemRegions: []ifuzz.MemRegion{
{0 << 12, 1 << 12},
{1 << 12, 1 << 12},
{2 << 12, 1 << 12},
{3 << 12, 1 << 12},
{4 << 12, 1 << 12},
{5 << 12, 1 << 12},
{6 << 12, 1 << 12},
{7 << 12, 1 << 12},
{8 << 12, 1 << 12},
{9 << 12, 1 << 12},
{0xfec00000, 0x100}, // ioapic
},
}
switch kind {
case Text_x86_real:
cfg.Mode = ifuzz.ModeReal16
case Text_x86_16:
cfg.Mode = ifuzz.ModeProt16
case Text_x86_32:
cfg.Mode = ifuzz.ModeProt32
case Text_x86_64:
cfg.Mode = ifuzz.ModeLong64
}
return cfg
}
// nOutOf returns true n out of outOf times.
func (r *randGen) nOutOf(n, outOf int) bool {
if n <= 0 || n >= outOf {
panic("bad probability")
}
v := r.Intn(outOf)
return v < n
}
func (r *randGen) generateCall(s *state, p *Prog) []*Call {
call := -1
if len(p.Calls) != 0 {
for i := 0; i < 5; i++ {
c := p.Calls[r.Intn(len(p.Calls))].Meta
call = c.ID
// There is roughly half of mmap's so ignore them.
if c != r.target.MmapSyscall {
break
}
}
}
idx := 0
if s.ct == nil {
idx = r.Intn(len(r.target.Syscalls))
} else {
idx = s.ct.Choose(r.Rand, call)
}
meta := r.target.Syscalls[idx]
return r.generateParticularCall(s, meta)
}
func (r *randGen) generateParticularCall(s *state, meta *Syscall) (calls []*Call) {
c := &Call{
Meta: meta,
Ret: MakeReturnArg(meta.Ret),
}
c.Args, calls = r.generateArgs(s, meta.Args)
r.target.assignSizesCall(c)
calls = append(calls, c)
for _, c1 := range calls {
r.target.SanitizeCall(c1)
}
return calls
}
// GenerateAllSyzProg generates a program that contains all pseudo syz_ calls for testing.
func (target *Target) GenerateAllSyzProg(rs rand.Source) *Prog {
p := &Prog{
Target: target,
}
r := newRand(target, rs)
s := newState(target, nil)
handled := make(map[string]bool)
for _, meta := range target.Syscalls {
if !strings.HasPrefix(meta.CallName, "syz_") || handled[meta.CallName] {
continue
}
handled[meta.CallName] = true
calls := r.generateParticularCall(s, meta)
for _, c := range calls {
s.analyze(c)
p.Calls = append(p.Calls, c)
}
}
if err := p.validate(); err != nil {
panic(err)
}
return p
}
// GenerateSimpleProg generates the simplest non-empty program for testing
// (e.g. containing a single mmap).
func (target *Target) GenerateSimpleProg() *Prog {
return &Prog{
Target: target,
Calls: []*Call{target.MakeMmap(0, 1)},
}
}
func (r *randGen) generateArgs(s *state, types []Type) ([]Arg, []*Call) {
var calls []*Call
args := make([]Arg, len(types))
// Generate all args. Size args have the default value 0 for now.
for i, typ := range types {
arg, calls1 := r.generateArg(s, typ)
if arg == nil {
panic(fmt.Sprintf("generated arg is nil for type '%v', types: %+v", typ.Name(), types))
}
args[i] = arg
calls = append(calls, calls1...)
}
return args, calls
}
func (r *randGen) generateArg(s *state, typ Type) (arg Arg, calls []*Call) {
if typ.Dir() == DirOut {
// No need to generate something interesting for output scalar arguments.
// But we still need to generate the argument itself so that it can be referenced
// in subsequent calls. For the same reason we do generate pointer/array/struct
// output arguments (their elements can be referenced in subsequent calls).
switch typ.(type) {
case *IntType, *FlagsType, *ConstType, *ProcType,
*VmaType, *ResourceType:
return defaultArg(typ), nil
}
}
if typ.Optional() && r.oneOf(5) {
return defaultArg(typ), nil
}
// Allow infinite recursion for optional pointers.
if pt, ok := typ.(*PtrType); ok && typ.Optional() {
if str, ok := pt.Type.(*StructType); ok {
r.recDepth[str.Name()] += 1
defer func() {
r.recDepth[str.Name()] -= 1
if r.recDepth[str.Name()] == 0 {
delete(r.recDepth, str.Name())
}
}()
if r.recDepth[str.Name()] >= 3 {
return MakePointerArg(typ, 0, 0, 0, nil), nil
}
}
}
switch a := typ.(type) {
case *ResourceType:
switch {
case r.nOutOf(1000, 1011):
// Get an existing resource.
var allres []Arg
for name1, res1 := range s.resources {
if name1 == "iocbptr" {
continue
}
if r.target.isCompatibleResource(a.Desc.Name, name1) ||
r.oneOf(20) && r.target.isCompatibleResource(a.Desc.Kind[0], name1) {
allres = append(allres, res1...)
}
}
if len(allres) != 0 {
arg = MakeResultArg(a, allres[r.Intn(len(allres))], 0)
} else {
arg, calls = r.createResource(s, a)
}
case r.nOutOf(10, 11):
// Create a new resource.
arg, calls = r.createResource(s, a)
default:
special := a.SpecialValues()
arg = MakeResultArg(a, nil, special[r.Intn(len(special))])
}
return arg, calls
case *BufferType:
switch a.Kind {
case BufferBlobRand, BufferBlobRange:
sz := r.randBufLen()
if a.Kind == BufferBlobRange {
sz = r.randRange(a.RangeBegin, a.RangeEnd)
}
if a.Dir() == DirOut {
return MakeOutDataArg(a, sz), nil
}
data := make([]byte, sz)
for i := range data {
data[i] = byte(r.Intn(256))
}
return MakeDataArg(a, data), nil
case BufferString:
data := r.randString(s, a.Values, a.Dir())
if a.Dir() == DirOut {
return MakeOutDataArg(a, uint64(len(data))), nil
}
return MakeDataArg(a, data), nil
case BufferFilename:
if a.Dir() == DirOut {
sz := 0
switch {
case r.nOutOf(1, 3):
sz = r.Intn(100)
case r.nOutOf(1, 2):
sz = 108 // UNIX_PATH_MAX
default:
sz = 4096 // PATH_MAX
}
return MakeOutDataArg(a, uint64(sz)), nil
}
return MakeDataArg(a, []byte(r.filename(s))), nil
case BufferText:
if a.Dir() == DirOut {
return MakeOutDataArg(a, uint64(r.Intn(100))), nil
}
return MakeDataArg(a, r.generateText(a.Text)), nil
default:
panic("unknown buffer kind")
}
case *VmaType:
npages := r.randPageCount()
if a.RangeBegin != 0 || a.RangeEnd != 0 {
npages = a.RangeBegin + uint64(r.Intn(int(a.RangeEnd-a.RangeBegin+1)))
}
arg := r.randPageAddr(s, a, npages, nil, true)
return arg, nil
case *FlagsType:
return MakeConstArg(a, r.flags(a.Vals)), nil
case *ConstType:
return MakeConstArg(a, a.Val), nil
case *IntType:
v := r.randInt()
switch a.Kind {
case IntFileoff:
switch {
case r.nOutOf(90, 101):
v = 0
case r.nOutOf(10, 11):
v = r.rand(100)
default:
v = r.randInt()
}
case IntRange:
v = r.randRangeInt(a.RangeBegin, a.RangeEnd)
}
return MakeConstArg(a, v), nil
case *ProcType:
return MakeConstArg(a, r.rand(int(a.ValuesPerProc))), nil
case *ArrayType:
var count uint64
switch a.Kind {
case ArrayRandLen:
count = r.randArrayLen()
case ArrayRangeLen:
count = r.randRange(a.RangeBegin, a.RangeEnd)
}
var inner []Arg
var calls []*Call
for i := uint64(0); i < count; i++ {
arg1, calls1 := r.generateArg(s, a.Type)
inner = append(inner, arg1)
calls = append(calls, calls1...)
}
return MakeGroupArg(a, inner), calls
case *StructType:
if gen := r.target.SpecialStructs[a.Name()]; gen != nil && a.Dir() != DirOut {
arg, calls = gen(&Gen{r, s}, a, nil)
return
}
args, calls := r.generateArgs(s, a.Fields)
group := MakeGroupArg(a, args)
return group, calls
case *UnionType:
optType := a.Fields[r.Intn(len(a.Fields))]
opt, calls := r.generateArg(s, optType)
return MakeUnionArg(a, opt, optType), calls
case *PtrType:
inner, calls := r.generateArg(s, a.Type)
if a.Type.Name() == "iocb" && len(s.resources["iocbptr"]) != 0 {
// It is weird, but these are actually identified by kernel by address.
// So try to reuse a previously used address.
addrs := s.resources["iocbptr"]
addr := addrs[r.Intn(len(addrs))].(*PointerArg)
arg = MakePointerArg(a, addr.PageIndex, addr.PageOffset, addr.PagesNum, inner)
return arg, calls
}
arg, calls1 := r.addr(s, a, inner.Size(), inner)
calls = append(calls, calls1...)
return arg, calls
case *LenType:
// Return placeholder value of 0 while generating len arg.
return MakeConstArg(a, 0), nil
case *CsumType:
return MakeConstArg(a, 0), nil
default:
panic("unknown argument type")
}
}