chacha20: expose internal/chacha20 package

const KeySize = 32
const NonceSize = 12
func HChaCha20(key, nonce []byte) ([]byte, error)
type Cipher struct {}
func NewUnauthenticatedCipher(key, nonce []byte) (*Cipher, error)
func (s *Cipher) XORKeyStream(dst, src []byte)

Small performance hit in chacha20poly1305, probably due to the loss
of the Advance API, which we might consider adding later. No new
allocations, thanks to the mid-stack inliner.

name                            old time/op    new time/op    delta
Chacha20Poly1305/Open-64-8        1.60µs ± 0%    1.68µs ± 1%  +4.94%  (p=0.000 n=9+10)
Chacha20Poly1305/Seal-64-8        1.56µs ± 0%    1.64µs ± 1%  +5.21%  (p=0.000 n=8+10)
Chacha20Poly1305/Open-64-X-8      2.10µs ± 1%    2.22µs ± 1%  +5.81%  (p=0.000 n=10+10)
Chacha20Poly1305/Seal-64-X-8      2.07µs ± 1%    2.17µs ± 0%  +4.88%  (p=0.000 n=10+10)
Chacha20Poly1305/Open-1350-8      15.4µs ± 0%    15.7µs ± 1%  +1.65%  (p=0.000 n=10+10)
Chacha20Poly1305/Seal-1350-8      15.6µs ± 2%    15.9µs ± 1%  +1.58%  (p=0.028 n=10+9)
Chacha20Poly1305/Open-1350-X-8    16.0µs ± 1%    16.3µs ± 2%  +2.00%  (p=0.000 n=10+10)
Chacha20Poly1305/Seal-1350-X-8    15.9µs ± 0%    16.3µs ± 1%  +1.91%  (p=0.000 n=10+8)
Chacha20Poly1305/Open-8192-8      85.6µs ± 0%    86.6µs ± 1%  +1.21%  (p=0.000 n=10+10)
Chacha20Poly1305/Seal-8192-8      85.7µs ± 0%    86.3µs ± 0%  +0.68%  (p=0.001 n=9+9)
Chacha20Poly1305/Open-8192-X-8    86.4µs ± 1%    87.1µs ± 1%  +0.76%  (p=0.035 n=10+9)
Chacha20Poly1305/Seal-8192-X-8    86.0µs ± 0%    87.0µs ± 1%  +1.14%  (p=0.000 n=9+9)

Updates golang/go#24485

Change-Id: I2ec2ef487a03f013049915d9063751c75a78408b
Reviewed-on: https://go-review.googlesource.com/c/crypto/+/185980
Reviewed-by: Michael Munday <mike.munday@ibm.com>

ssh/cipher.go

@@ -16,9 +16,8 @@ import (
 	"hash"
 	"io"
 	"io/ioutil"
-	"math/bits"
 
-	"golang.org/x/crypto/internal/chacha20"
+	"golang.org/x/crypto/chacha20"
 	"golang.org/x/crypto/poly1305"
 )
 
@@ -642,8 +641,8 @@ const chacha20Poly1305ID = "chacha20-poly1305@openssh.com"
 // the methods here also implement padding, which RFC4253 Section 6
 // also requires of stream ciphers.
 type chacha20Poly1305Cipher struct {
-	lengthKey  [8]uint32
-	contentKey [8]uint32
+	lengthKey  [32]byte
+	contentKey [32]byte
 	buf        []byte
 }
 
@@ -656,21 +655,21 @@ func newChaCha20Cipher(key, unusedIV, unusedMACKey []byte, unusedAlgs directionA
 		buf: make([]byte, 256),
 	}
 
-	for i := range c.contentKey {
-		c.contentKey[i] = binary.LittleEndian.Uint32(key[i*4 : (i+1)*4])
-	}
-	for i := range c.lengthKey {
-		c.lengthKey[i] = binary.LittleEndian.Uint32(key[(i+8)*4 : (i+9)*4])
-	}
+	copy(c.contentKey[:], key[:32])
+	copy(c.lengthKey[:], key[32:])
 	return c, nil
 }
 
 func (c *chacha20Poly1305Cipher) readCipherPacket(seqNum uint32, r io.Reader) ([]byte, error) {
-	nonce := [3]uint32{0, 0, bits.ReverseBytes32(seqNum)}
-	s := chacha20.New(c.contentKey, nonce)
-	var polyKey [32]byte
+	nonce := make([]byte, 12)
+	binary.BigEndian.PutUint32(nonce[8:], seqNum)
+	s, err := chacha20.NewUnauthenticatedCipher(c.contentKey[:], nonce)
+	if err != nil {
+		return nil, err
+	}
+	var polyKey, discardBuf [32]byte
 	s.XORKeyStream(polyKey[:], polyKey[:])
-	s.Advance() // skip next 32 bytes
+	s.XORKeyStream(discardBuf[:], discardBuf[:]) // skip the next 32 bytes
 
 	encryptedLength := c.buf[:4]
 	if _, err := io.ReadFull(r, encryptedLength); err != nil {
@@ -678,7 +677,11 @@ func (c *chacha20Poly1305Cipher) readCipherPacket(seqNum uint32, r io.Reader) ([
 	}
 
 	var lenBytes [4]byte
-	chacha20.New(c.lengthKey, nonce).XORKeyStream(lenBytes[:], encryptedLength)
+	ls, err := chacha20.NewUnauthenticatedCipher(c.lengthKey[:], nonce)
+	if err != nil {
+		return nil, err
+	}
+	ls.XORKeyStream(lenBytes[:], encryptedLength)
 
 	length := binary.BigEndian.Uint32(lenBytes[:])
 	if length > maxPacket {
@@ -724,11 +727,15 @@ func (c *chacha20Poly1305Cipher) readCipherPacket(seqNum uint32, r io.Reader) ([
 }
 
 func (c *chacha20Poly1305Cipher) writeCipherPacket(seqNum uint32, w io.Writer, rand io.Reader, payload []byte) error {
-	nonce := [3]uint32{0, 0, bits.ReverseBytes32(seqNum)}
-	s := chacha20.New(c.contentKey, nonce)
-	var polyKey [32]byte
+	nonce := make([]byte, 12)
+	binary.BigEndian.PutUint32(nonce[8:], seqNum)
+	s, err := chacha20.NewUnauthenticatedCipher(c.contentKey[:], nonce)
+	if err != nil {
+		return err
+	}
+	var polyKey, discardBuf [32]byte
 	s.XORKeyStream(polyKey[:], polyKey[:])
-	s.Advance() // skip next 32 bytes
+	s.XORKeyStream(discardBuf[:], discardBuf[:]) // skip the next 32 bytes
 
 	// There is no blocksize, so fall back to multiple of 8 byte
 	// padding, as described in RFC 4253, Sec 6.
@@ -748,7 +755,11 @@ func (c *chacha20Poly1305Cipher) writeCipherPacket(seqNum uint32, w io.Writer, r
 	}
 
 	binary.BigEndian.PutUint32(c.buf, uint32(1+len(payload)+padding))
-	chacha20.New(c.lengthKey, nonce).XORKeyStream(c.buf, c.buf[:4])
+	ls, err := chacha20.NewUnauthenticatedCipher(c.lengthKey[:], nonce)
+	if err != nil {
+		return err
+	}
+	ls.XORKeyStream(c.buf, c.buf[:4])
 	c.buf[4] = byte(padding)
 	copy(c.buf[5:], payload)
 	packetEnd := 5 + len(payload) + padding