2
2
mirror of https://github.com/octoleo/restic.git synced 2024-05-31 08:00:48 +00:00

chunker: Refactor to take polynomial on creation

This commit is contained in:
Alexander Neumann 2015-04-05 22:24:26 +02:00
parent 3cdf3a25b9
commit 367cc75970
2 changed files with 75 additions and 51 deletions

View File

@ -1,9 +1,9 @@
package chunker
import (
"errors"
"hash"
"io"
"sync"
)
const (
@ -23,20 +23,6 @@ const (
splitmask = (1 << AverageBits) - 1
)
var (
// pol is a randomly generated irreducible polynomial of degree 53
// in Z_2[X]. All rabin fingerprints are calculated with this polynomial.
pol = uint64(0x3DA3358B4DC173)
pol_shift = deg(pol) - 8
once sync.Once
mod_table [256]uint64
out_table [256]uint64
// tables have been filled, do not allow changing the polynom afterwards
filled bool
)
// A chunk is one content-dependent chunk of bytes whose end was cut when the
// Rabin Fingerprint had the value stored in Cut.
type Chunk struct {
@ -52,6 +38,11 @@ func (c Chunk) Reader(r io.ReaderAt) io.Reader {
// A chunker internally holds everything needed to split content.
type Chunker struct {
pol Pol
pol_shift uint
mod_table [256]uint64
out_table [256]uint64
rd io.Reader
closed bool
@ -72,28 +63,24 @@ type Chunker struct {
h hash.Hash
}
// Polynomial sets the polynomial that is to be used for calculating the rabin
// fingerprints. This function must be called before the first chunker is
// created, otherwise the results are undefined.
func SetPolynomial(f uint64) {
if filled {
panic("polynomial changed after chunker has already been used")
// New returns a new Chunker based on polynomial p that reads from data from rd
// with bufsize and pass all data to hash along the way.
func New(rd io.Reader, p Pol, bufsize int, hash hash.Hash) (*Chunker, error) {
// test irreducibility of p again
if !p.Irreducible() {
return nil, errors.New("invalid polynomial")
}
pol = f
}
// New returns a new Chunker that reads from data from rd with bufsize and pass
// all data to hash along the way.
func New(rd io.Reader, bufsize int, hash hash.Hash) *Chunker {
once.Do(fill_tables)
c := &Chunker{
buf: make([]byte, bufsize),
h: hash,
pol: p,
pol_shift: uint(p.Deg() - 8),
buf: make([]byte, bufsize),
h: hash,
}
c.fill_tables()
c.Reset(rd)
return c
return c, nil
}
// Reset restarts a chunker so that it can be reused with a different reader as
@ -121,9 +108,7 @@ func (c *Chunker) Reset(rd io.Reader) {
}
// Calculate out_table and mod_table for optimization. Must be called only once.
func fill_tables() {
filled = true
func (c *Chunker) fill_tables() {
// calculate table for sliding out bytes. The byte to slide out is used as
// the index for the table, the value contains the following:
// out_table[b] = Hash(b || 0 || ... || 0)
@ -138,15 +123,15 @@ func fill_tables() {
for b := 0; b < 256; b++ {
var hash uint64
hash = append_byte(hash, byte(b), pol)
hash = append_byte(hash, byte(b), uint64(c.pol))
for i := 0; i < WindowSize-1; i++ {
hash = append_byte(hash, 0, pol)
hash = append_byte(hash, 0, uint64(c.pol))
}
out_table[b] = hash
c.out_table[b] = hash
}
// calculate table for reduction mod Polynomial
k := deg(pol)
k := c.pol.Deg()
for b := 0; b < 256; b++ {
// mod_table[b] = A | B, where A = (b(x) * x^k mod pol) and B = b(x) * x^k
//
@ -155,7 +140,7 @@ func fill_tables() {
// two parts: Part A contains the result of the modulus operation, part
// B is used to cancel out the 8 top bits so that one XOR operation is
// enough to reduce modulo Polynomial
mod_table[b] = mod(uint64(b)<<uint(k), pol) | (uint64(b) << uint(k))
c.mod_table[b] = mod(uint64(b)<<uint(k), uint64(c.pol)) | (uint64(b) << uint(k))
}
}
@ -226,15 +211,15 @@ func (c *Chunker) Next() (*Chunk, error) {
// inline c.slide(b) and append(b) to increase performance
out := c.window[c.wpos]
c.window[c.wpos] = b
c.digest ^= out_table[out]
c.digest ^= c.out_table[out]
c.wpos = (c.wpos + 1) % WindowSize
// c.append(b)
index := c.digest >> uint(pol_shift)
index := c.digest >> c.pol_shift
c.digest <<= 8
c.digest |= uint64(b)
c.digest ^= mod_table[index]
c.digest ^= c.mod_table[index]
// end inline
add++
@ -300,17 +285,17 @@ func (c *Chunker) hashDigest() []byte {
}
func (c *Chunker) append(b byte) {
index := c.digest >> uint(pol_shift)
index := c.digest >> c.pol_shift
c.digest <<= 8
c.digest |= uint64(b)
c.digest ^= mod_table[index]
c.digest ^= c.mod_table[index]
}
func (c *Chunker) slide(b byte) {
out := c.window[c.wpos]
c.window[c.wpos] = b
c.digest ^= out_table[out]
c.digest ^= c.out_table[out]
c.wpos = (c.wpos + 1) % WindowSize
c.append(b)

View File

@ -12,6 +12,7 @@ import (
"math/rand"
"os"
"testing"
"time"
"github.com/restic/restic/chunker"
)
@ -34,6 +35,9 @@ type chunk struct {
Digest []byte
}
// polynomial used for all the tests below
const testPol = chunker.Pol(0x3DA3358B4DC173)
// created for 32MB of random data out of math/rand's Uint32() seeded by
// constant 23
//
@ -146,7 +150,8 @@ func get_random(seed, count int) []byte {
func TestChunker(t *testing.T) {
// setup data source
buf := get_random(23, 32*1024*1024)
ch := chunker.New(bytes.NewReader(buf), *testBufSize, sha256.New())
ch, err := chunker.New(bytes.NewReader(buf), testPol, *testBufSize, sha256.New())
ok(t, err)
chunks := test_with_data(t, ch, chunks1)
// test reader
@ -173,15 +178,43 @@ func TestChunker(t *testing.T) {
// setup nullbyte data source
buf = bytes.Repeat([]byte{0}, len(chunks2)*chunker.MinSize)
ch = chunker.New(bytes.NewReader(buf), *testBufSize, sha256.New())
ch, err = chunker.New(bytes.NewReader(buf), testPol, *testBufSize, sha256.New())
ok(t, err)
test_with_data(t, ch, chunks2)
}
func TestChunkerWithRandomPolynomial(t *testing.T) {
// setup data source
buf := get_random(23, 32*1024*1024)
// generate a new random polynomial
start := time.Now()
p, err := chunker.RandomPolynomial()
ok(t, err)
t.Logf("generating random polynomial took %v", time.Since(start))
start = time.Now()
ch, err := chunker.New(bytes.NewReader(buf), p, *testBufSize, sha256.New())
ok(t, err)
t.Logf("creating chunker took %v", time.Since(start))
// make sure that first chunk is different
c, err := ch.Next()
assert(t, c.Cut != chunks1[0].CutFP,
"Cut point is the same")
assert(t, c.Length != chunks1[0].Length,
"Length is the same")
assert(t, !bytes.Equal(c.Digest, chunks1[0].Digest),
"Digest is the same")
}
func TestChunkerWithoutHash(t *testing.T) {
// setup data source
buf := get_random(23, 32*1024*1024)
ch := chunker.New(bytes.NewReader(buf), *testBufSize, nil)
ch, err := chunker.New(bytes.NewReader(buf), testPol, *testBufSize, nil)
ok(t, err)
chunks := test_with_data(t, ch, chunks1)
// test reader
@ -211,14 +244,16 @@ func TestChunkerWithoutHash(t *testing.T) {
// setup nullbyte data source
buf = bytes.Repeat([]byte{0}, len(chunks2)*chunker.MinSize)
ch = chunker.New(bytes.NewReader(buf), *testBufSize, sha256.New())
ch, err = chunker.New(bytes.NewReader(buf), testPol, *testBufSize, sha256.New())
ok(t, err)
test_with_data(t, ch, chunks2)
}
func TestChunkerReuse(t *testing.T) {
// test multiple uses of the same chunker
ch := chunker.New(nil, *testBufSize, sha256.New())
ch, err := chunker.New(nil, testPol, *testBufSize, sha256.New())
ok(t, err)
buf := get_random(23, 32*1024*1024)
for i := 0; i < 4; i++ {
@ -254,9 +289,13 @@ func benchmarkChunker(b *testing.B, hash hash.Hash) {
rd = bytes.NewReader(get_random(23, size))
}
t1 := time.Now()
ch, err := chunker.New(rd, testPol, *testBufSize, hash)
ok(b, err)
b.Logf("generating tables took %v", time.Since(t1))
b.ResetTimer()
b.SetBytes(int64(size))
ch := chunker.New(rd, *testBufSize, hash)
var chunks int
for i := 0; i < b.N; i++ {