diff --git a/src/pkg/Makefile b/src/pkg/Makefile
index 3fa6b00ed0..d6552d0658 100644
--- a/src/pkg/Makefile
+++ b/src/pkg/Makefile
@@ -40,6 +40,7 @@ DIRS=\
 	crypto/md4\
 	crypto/md5\
 	crypto/rc4\
+	crypto/ripemd160\
 	crypto/rsa\
 	crypto/sha1\
 	crypto/sha256\
diff --git a/src/pkg/crypto/ripemd160/Makefile b/src/pkg/crypto/ripemd160/Makefile
new file mode 100644
index 0000000000..109e68edac
--- /dev/null
+++ b/src/pkg/crypto/ripemd160/Makefile
@@ -0,0 +1,12 @@
+# Copyright 2010 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.
+
+include ../../../Make.$(GOARCH)
+
+TARG=crypto/ripemd160
+GOFILES=\
+	ripemd160.go\
+	ripemd160block.go\
+
+include ../../../Make.pkg
diff --git a/src/pkg/crypto/ripemd160/ripemd160.go b/src/pkg/crypto/ripemd160/ripemd160.go
new file mode 100644
index 0000000000..9eb8340b7d
--- /dev/null
+++ b/src/pkg/crypto/ripemd160/ripemd160.go
@@ -0,0 +1,116 @@
+// Copyright 2010 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.
+
+// This package implements the RIPEMD-160 hash algorithm.
+package ripemd160
+
+// RIPEMD-160 is designed by by Hans Dobbertin, Antoon Bosselaers, and Bart
+// Preneel with specifications available at:
+// http://homes.esat.kuleuven.be/~cosicart/pdf/AB-9601/AB-9601.pdf.
+
+import (
+	"hash"
+	"os"
+)
+
+// The size of the checksum in bytes.
+const Size = 20
+
+// The block size of the hash algorithm in bytes.
+const BlockSize = 64
+
+const (
+	_s0 = 0x67452301
+	_s1 = 0xefcdab89
+	_s2 = 0x98badcfe
+	_s3 = 0x10325476
+	_s4 = 0xc3d2e1f0
+)
+
+// digest represents the partial evaluation of a checksum.
+type digest struct {
+	s  [5]uint32       // running context
+	x  [BlockSize]byte // temporary buffer
+	nx int             // index into x
+	tc uint64          // total count of bytes processed
+}
+
+func (d *digest) Reset() {
+	d.s[0], d.s[1], d.s[2], d.s[3], d.s[4] = _s0, _s1, _s2, _s3, _s4
+	d.nx = 0
+	d.tc = 0
+}
+
+// New returns a new hash.Hash computing the checksum.
+func New() hash.Hash {
+	result := new(digest)
+	result.Reset()
+	return result
+}
+
+func (d *digest) Size() int { return Size }
+
+func (d *digest) Write(p []byte) (nn int, err os.Error) {
+	nn = len(p)
+	d.tc += uint64(nn)
+	if d.nx > 0 {
+		n := len(p)
+		if n > BlockSize-d.nx {
+			n = BlockSize - d.nx
+		}
+		for i := 0; i < n; i++ {
+			d.x[d.nx+i] = p[i]
+		}
+		d.nx += n
+		if d.nx == BlockSize {
+			_Block(d, &d.x)
+			d.nx = 0
+		}
+		p = p[n:]
+	}
+	n := _Block(d, p)
+	p = p[n:]
+	if len(p) > 0 {
+		for i, x := range p {
+			d.x[i] = x
+		}
+		d.nx = len(p)
+	}
+	return
+}
+
+func (d0 *digest) Sum() []byte {
+	// Make a copy of d0 so that caller can keep writing and summing.
+	d := new(digest)
+	*d = *d0
+
+	// Padding.  Add a 1 bit and 0 bits until 56 bytes mod 64.
+	tc := d.tc
+	var tmp [64]byte
+	tmp[0] = 0x80
+	if tc%64 < 56 {
+		d.Write(tmp[0 : 56-tc%64])
+	} else {
+		d.Write(tmp[0 : 64+56-tc%64])
+	}
+
+	// Length in bits.
+	tc <<= 3
+	for i := uint(0); i < 8; i++ {
+		tmp[i] = byte(tc >> (8 * i))
+	}
+	d.Write(tmp[0:8])
+
+	if d.nx != 0 {
+		panicln("oops")
+	}
+
+	p := make([]byte, 20)
+	j := 0
+	for _, s := range d.s {
+		p[j], p[j+1], p[j+2], p[j+3] = byte(s), byte(s>>8), byte(s>>16), byte(s>>24)
+		j += 4
+	}
+	return p
+}
diff --git a/src/pkg/crypto/ripemd160/ripemd160_test.go b/src/pkg/crypto/ripemd160/ripemd160_test.go
new file mode 100644
index 0000000000..eaa3d78c26
--- /dev/null
+++ b/src/pkg/crypto/ripemd160/ripemd160_test.go
@@ -0,0 +1,64 @@
+// Copyright 2010 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 ripemd160
+
+// Test vectors are from:
+// http://homes.esat.kuleuven.be/~bosselae/ripemd160.html
+
+import (
+	"fmt"
+	"io"
+	"testing"
+)
+
+type mdTest struct {
+	out string
+	in  string
+}
+
+var vectors = [...]mdTest{
+	mdTest{"9c1185a5c5e9fc54612808977ee8f548b2258d31", ""},
+	mdTest{"0bdc9d2d256b3ee9daae347be6f4dc835a467ffe", "a"},
+	mdTest{"8eb208f7e05d987a9b044a8e98c6b087f15a0bfc", "abc"},
+	mdTest{"5d0689ef49d2fae572b881b123a85ffa21595f36", "message digest"},
+	mdTest{"f71c27109c692c1b56bbdceb5b9d2865b3708dbc", "abcdefghijklmnopqrstuvwxyz"},
+	mdTest{"12a053384a9c0c88e405a06c27dcf49ada62eb2b", "abcdbcdecdefdefgefghfghighijhijkijkljklmklmnlmnomnopnopq"},
+	mdTest{"b0e20b6e3116640286ed3a87a5713079b21f5189", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789"},
+	mdTest{"9b752e45573d4b39f4dbd3323cab82bf63326bfb", "12345678901234567890123456789012345678901234567890123456789012345678901234567890"},
+}
+
+func TestVectors(t *testing.T) {
+	for i := 0; i < len(vectors); i++ {
+		tv := vectors[i]
+		md := New()
+		for j := 0; j < 3; j++ {
+			if j < 2 {
+				io.WriteString(md, tv.in)
+			} else {
+				io.WriteString(md, tv.in[0:len(tv.in)/2])
+				md.Sum()
+				io.WriteString(md, tv.in[len(tv.in)/2:])
+			}
+			s := fmt.Sprintf("%x", md.Sum())
+			if s != tv.out {
+				t.Fatalf("RIPEMD-160[%d](%s) = %s, expected %s", j, tv.in, s, tv.out)
+			}
+			md.Reset()
+		}
+	}
+}
+
+func TestMillionA(t *testing.T) {
+	md := New()
+	for i := 0; i < 100000; i++ {
+		io.WriteString(md, "aaaaaaaaaa")
+	}
+	out := "52783243c1697bdbe16d37f97f68f08325dc1528"
+	s := fmt.Sprintf("%x", md.Sum())
+	if s != out {
+		t.Fatalf("RIPEMD-160 (1 million 'a') = %s, expected %s", s, out)
+	}
+	md.Reset()
+}
diff --git a/src/pkg/crypto/ripemd160/ripemd160block.go b/src/pkg/crypto/ripemd160/ripemd160block.go
new file mode 100644
index 0000000000..7bc8e6c485
--- /dev/null
+++ b/src/pkg/crypto/ripemd160/ripemd160block.go
@@ -0,0 +1,161 @@
+// Copyright 2010 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.
+
+// RIPEMD-160 block step.
+// In its own file so that a faster assembly or C version
+// can be substituted easily.
+
+package ripemd160
+
+// work buffer indices and roll amounts for one line
+var _n = [80]uint{
+	0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
+	7, 4, 13, 1, 10, 6, 15, 3, 12, 0, 9, 5, 2, 14, 11, 8,
+	3, 10, 14, 4, 9, 15, 8, 1, 2, 7, 0, 6, 13, 11, 5, 12,
+	1, 9, 11, 10, 0, 8, 12, 4, 13, 3, 7, 15, 14, 5, 6, 2,
+	4, 0, 5, 9, 7, 12, 2, 10, 14, 1, 3, 8, 11, 6, 15, 13,
+}
+
+var _r = [80]uint{
+	11, 14, 15, 12, 5, 8, 7, 9, 11, 13, 14, 15, 6, 7, 9, 8,
+	7, 6, 8, 13, 11, 9, 7, 15, 7, 12, 15, 9, 11, 7, 13, 12,
+	11, 13, 6, 7, 14, 9, 13, 15, 14, 8, 13, 6, 5, 12, 7, 5,
+	11, 12, 14, 15, 14, 15, 9, 8, 9, 14, 5, 6, 8, 6, 5, 12,
+	9, 15, 5, 11, 6, 8, 13, 12, 5, 12, 13, 14, 11, 8, 5, 6,
+}
+
+// same for the other parallel one
+var n_ = [80]uint{
+	5, 14, 7, 0, 9, 2, 11, 4, 13, 6, 15, 8, 1, 10, 3, 12,
+	6, 11, 3, 7, 0, 13, 5, 10, 14, 15, 8, 12, 4, 9, 1, 2,
+	15, 5, 1, 3, 7, 14, 6, 9, 11, 8, 12, 2, 10, 0, 4, 13,
+	8, 6, 4, 1, 3, 11, 15, 0, 5, 12, 2, 13, 9, 7, 10, 14,
+	12, 15, 10, 4, 1, 5, 8, 7, 6, 2, 13, 14, 0, 3, 9, 11,
+}
+
+var r_ = [80]uint{
+	8, 9, 9, 11, 13, 15, 15, 5, 7, 7, 8, 11, 14, 14, 12, 6,
+	9, 13, 15, 7, 12, 8, 9, 11, 7, 7, 12, 7, 6, 15, 13, 11,
+	9, 7, 15, 11, 8, 6, 6, 14, 12, 13, 5, 14, 13, 13, 7, 5,
+	15, 5, 8, 11, 14, 14, 6, 14, 6, 9, 12, 9, 12, 5, 15, 8,
+	8, 5, 12, 9, 12, 5, 14, 6, 8, 13, 6, 5, 15, 13, 11, 11,
+}
+
+func _Block(md *digest, p []byte) int {
+	n := 0
+	var x [16]uint32
+	var alpha, beta uint32
+	for len(p) >= BlockSize {
+		a, b, c, d, e := md.s[0], md.s[1], md.s[2], md.s[3], md.s[4]
+		aa, bb, cc, dd, ee := a, b, c, d, e
+		j := 0
+		for i := 0; i < 16; i++ {
+			x[i] = uint32(p[j]) | uint32(p[j+1])<<8 | uint32(p[j+2])<<16 | uint32(p[j+3])<<24
+			j += 4
+		}
+
+		// round 1
+		i := 0
+		for i < 16 {
+			alpha = a + (b ^ c ^ d) + x[_n[i]]
+			s := _r[i]
+			alpha = (alpha<<s | alpha>>(32-s)) + e
+			beta = c<<10 | c>>22
+			a, b, c, d, e = e, alpha, b, beta, d
+
+			// parallel line
+			alpha = aa + (bb ^ (cc | ^dd)) + x[n_[i]] + 0x50a28be6
+			s = r_[i]
+			alpha = (alpha<<s | alpha>>(32-s)) + ee
+			beta = cc<<10 | cc>>22
+			aa, bb, cc, dd, ee = ee, alpha, bb, beta, dd
+
+			i++
+		}
+
+		// round 2
+		for i < 32 {
+			alpha = a + (b&c | ^b&d) + x[_n[i]] + 0x5a827999
+			s := _r[i]
+			alpha = (alpha<<s | alpha>>(32-s)) + e
+			beta = c<<10 | c>>22
+			a, b, c, d, e = e, alpha, b, beta, d
+
+			// parallel line
+			alpha = aa + (bb&dd | cc&^dd) + x[n_[i]] + 0x5c4dd124
+			s = r_[i]
+			alpha = (alpha<<s | alpha>>(32-s)) + ee
+			beta = cc<<10 | cc>>22
+			aa, bb, cc, dd, ee = ee, alpha, bb, beta, dd
+
+			i++
+		}
+
+		// round 3
+		for i < 48 {
+			alpha = a + (b | ^c ^ d) + x[_n[i]] + 0x6ed9eba1
+			s := _r[i]
+			alpha = (alpha<<s | alpha>>(32-s)) + e
+			beta = c<<10 | c>>22
+			a, b, c, d, e = e, alpha, b, beta, d
+
+			// parallel line
+			alpha = aa + (bb | ^cc ^ dd) + x[n_[i]] + 0x6d703ef3
+			s = r_[i]
+			alpha = (alpha<<s | alpha>>(32-s)) + ee
+			beta = cc<<10 | cc>>22
+			aa, bb, cc, dd, ee = ee, alpha, bb, beta, dd
+
+			i++
+		}
+
+		// round 4
+		for i < 64 {
+			alpha = a + (b&d | c&^d) + x[_n[i]] + 0x8f1bbcdc
+			s := _r[i]
+			alpha = (alpha<<s | alpha>>(32-s)) + e
+			beta = c<<10 | c>>22
+			a, b, c, d, e = e, alpha, b, beta, d
+
+			// parallel line
+			alpha = aa + (bb&cc | ^bb&dd) + x[n_[i]] + 0x7a6d76e9
+			s = r_[i]
+			alpha = (alpha<<s | alpha>>(32-s)) + ee
+			beta = cc<<10 | cc>>22
+			aa, bb, cc, dd, ee = ee, alpha, bb, beta, dd
+
+			i++
+		}
+
+		// round 5
+		for i < 80 {
+			alpha = a + (b ^ (c | ^d)) + x[_n[i]] + 0xa953fd4e
+			s := _r[i]
+			alpha = (alpha<<s | alpha>>(32-s)) + e
+			beta = c<<10 | c>>22
+			a, b, c, d, e = e, alpha, b, beta, d
+
+			// parallel line
+			alpha = aa + (bb ^ cc ^ dd) + x[n_[i]]
+			s = r_[i]
+			alpha = (alpha<<s | alpha>>(32-s)) + ee
+			beta = cc<<10 | cc>>22
+			aa, bb, cc, dd, ee = ee, alpha, bb, beta, dd
+
+			i++
+		}
+
+		// combine results
+		dd += c + md.s[1]
+		md.s[1] = md.s[2] + d + ee
+		md.s[2] = md.s[3] + e + aa
+		md.s[3] = md.s[4] + a + bb
+		md.s[4] = md.s[0] + b + cc
+		md.s[0] = dd
+
+		p = p[BlockSize:]
+		n += BlockSize
+	}
+	return n
+}