go/src/crypto/ecdsa/ecdsa_test.go

402 lines
8.4 KiB
Go
Raw Normal View History

// Copyright 2011 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 ecdsa
import (
"bufio"
"compress/bzip2"
"crypto/elliptic"
"crypto/rand"
"crypto/sha1"
"crypto/sha256"
"crypto/sha512"
"encoding/hex"
"hash"
"io"
"math/big"
"os"
"strings"
"testing"
)
func testAllCurves(t *testing.T, f func(*testing.T, elliptic.Curve)) {
tests := []struct {
name string
curve elliptic.Curve
}{
{"P256", elliptic.P256()},
{"P224", elliptic.P224()},
{"P384", elliptic.P384()},
{"P521", elliptic.P521()},
}
if testing.Short() {
tests = tests[:1]
}
for _, test := range tests {
curve := test.curve
t.Run(test.name, func(t *testing.T) {
t.Parallel()
f(t, curve)
})
}
}
func TestKeyGeneration(t *testing.T) {
testAllCurves(t, testKeyGeneration)
}
func testKeyGeneration(t *testing.T, c elliptic.Curve) {
priv, err := GenerateKey(c, rand.Reader)
if err != nil {
t.Fatal(err)
}
if !c.IsOnCurve(priv.PublicKey.X, priv.PublicKey.Y) {
t.Errorf("public key invalid: %s", err)
}
}
func TestSignAndVerify(t *testing.T) {
testAllCurves(t, testSignAndVerify)
}
func testSignAndVerify(t *testing.T, c elliptic.Curve) {
priv, _ := GenerateKey(c, rand.Reader)
hashed := []byte("testing")
r, s, err := Sign(rand.Reader, priv, hashed)
if err != nil {
t.Errorf("error signing: %s", err)
return
}
if !Verify(&priv.PublicKey, hashed, r, s) {
t.Errorf("Verify failed")
}
hashed[0] ^= 0xff
if Verify(&priv.PublicKey, hashed, r, s) {
t.Errorf("Verify always works!")
}
}
func TestSignAndVerifyASN1(t *testing.T) {
testAllCurves(t, testSignAndVerifyASN1)
}
func testSignAndVerifyASN1(t *testing.T, c elliptic.Curve) {
priv, _ := GenerateKey(c, rand.Reader)
hashed := []byte("testing")
sig, err := SignASN1(rand.Reader, priv, hashed)
if err != nil {
t.Errorf("error signing: %s", err)
return
}
if !VerifyASN1(&priv.PublicKey, hashed, sig) {
t.Errorf("VerifyASN1 failed")
}
hashed[0] ^= 0xff
if VerifyASN1(&priv.PublicKey, hashed, sig) {
t.Errorf("VerifyASN1 always works!")
}
}
func TestNonceSafety(t *testing.T) {
testAllCurves(t, testNonceSafety)
}
func testNonceSafety(t *testing.T, c elliptic.Curve) {
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
priv, _ := GenerateKey(c, rand.Reader)
hashed := []byte("testing")
r0, s0, err := Sign(zeroReader, priv, hashed)
if err != nil {
t.Errorf("error signing: %s", err)
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
return
}
hashed = []byte("testing...")
r1, s1, err := Sign(zeroReader, priv, hashed)
if err != nil {
t.Errorf("error signing: %s", err)
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
return
}
if s0.Cmp(s1) == 0 {
// This should never happen.
t.Errorf("the signatures on two different messages were the same")
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
}
if r0.Cmp(r1) == 0 {
t.Errorf("the nonce used for two different messages was the same")
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
}
}
func TestINDCCA(t *testing.T) {
testAllCurves(t, testINDCCA)
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
}
func testINDCCA(t *testing.T, c elliptic.Curve) {
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
priv, _ := GenerateKey(c, rand.Reader)
hashed := []byte("testing")
r0, s0, err := Sign(rand.Reader, priv, hashed)
if err != nil {
t.Errorf("error signing: %s", err)
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
return
}
r1, s1, err := Sign(rand.Reader, priv, hashed)
if err != nil {
t.Errorf("error signing: %s", err)
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
return
}
if s0.Cmp(s1) == 0 {
t.Errorf("two signatures of the same message produced the same result")
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
}
if r0.Cmp(r1) == 0 {
t.Errorf("two signatures of the same message produced the same nonce")
crypto/ecdsa: make Sign safe with broken entropy sources ECDSA is unsafe to use if an entropy source produces predictable output for the ephemeral nonces. E.g., [Nguyen]. A simple countermeasure is to hash the secret key, the message, and entropy together to seed a CSPRNG, from which the ephemeral key is derived. Fixes #9452 -- This is a minimalist (in terms of patch size) solution, though not the most parsimonious in its use of primitives: - csprng_key = ChopMD-256(SHA2-512(priv.D||entropy||hash)) - reader = AES-256-CTR(k=csprng_key) This, however, provides at most 128-bit collision-resistance, so that Adv will have a term related to the number of messages signed that is significantly worse than plain ECDSA. This does not seem to be of any practical importance. ChopMD-256(SHA2-512(x)) is used, rather than SHA2-256(x), for two sets of reasons: *Practical:* SHA2-512 has a larger state and 16 more rounds; it is likely non-generically stronger than SHA2-256. And, AFAIK, cryptanalysis backs this up. (E.g., [Biryukov] gives a distinguisher on 47-round SHA2-256 with cost < 2^85.) This is well below a reasonable security-strength target. *Theoretical:* [Coron] and [Chang] show that Chop-MD(F(x)) is indifferentiable from a random oracle for slightly beyond the birthday barrier. It seems likely that this makes a generic security proof that this construction remains UF-CMA is possible in the indifferentiability framework. -- Many thanks to Payman Mohassel for reviewing this construction; any mistakes are mine, however. And, as he notes, reusing the private key in this way means that the generic-group (non-RO) proof of ECDSA's security given in [Brown] no longer directly applies. -- [Brown]: http://www.cacr.math.uwaterloo.ca/techreports/2000/corr2000-54.ps "Brown. The exact security of ECDSA. 2000" [Coron]: https://www.cs.nyu.edu/~puniya/papers/merkle.pdf "Coron et al. Merkle-Damgard revisited. 2005" [Chang]: https://www.iacr.org/archive/fse2008/50860436/50860436.pdf "Chang and Nandi. Improved indifferentiability security analysis of chopMD hash function. 2008" [Biryukov]: http://www.iacr.org/archive/asiacrypt2011/70730269/70730269.pdf "Biryukov et al. Second-order differential collisions for reduced SHA-256. 2011" [Nguyen]: ftp://ftp.di.ens.fr/pub/users/pnguyen/PubECDSA.ps "Nguyen and Shparlinski. The insecurity of the elliptic curve digital signature algorithm with partially known nonces. 2003" New tests: TestNonceSafety: Check that signatures are safe even with a broken entropy source. TestINDCCA: Check that signatures remain non-deterministic with a functional entropy source. Updated "golden" KATs in crypto/tls/testdata that use ECDSA suites. Change-Id: I55337a2fbec2e42a36ce719bd2184793682d678a Reviewed-on: https://go-review.googlesource.com/3340 Reviewed-by: Adam Langley <agl@golang.org>
2015-01-26 23:00:21 -08:00
}
}
func fromHex(s string) *big.Int {
r, ok := new(big.Int).SetString(s, 16)
if !ok {
panic("bad hex")
}
return r
}
func TestVectors(t *testing.T) {
// This test runs the full set of NIST test vectors from
// https://csrc.nist.gov/groups/STM/cavp/documents/dss/186-3ecdsatestvectors.zip
//
// The SigVer.rsp file has been edited to remove test vectors for
// unsupported algorithms and has been compressed.
if testing.Short() {
return
}
f, err := os.Open("testdata/SigVer.rsp.bz2")
if err != nil {
t.Fatal(err)
}
buf := bufio.NewReader(bzip2.NewReader(f))
lineNo := 1
var h hash.Hash
var msg []byte
var hashed []byte
var r, s *big.Int
pub := new(PublicKey)
for {
line, err := buf.ReadString('\n')
if len(line) == 0 {
if err == io.EOF {
break
}
t.Fatalf("error reading from input: %s", err)
}
lineNo++
// Need to remove \r\n from the end of the line.
if !strings.HasSuffix(line, "\r\n") {
t.Fatalf("bad line ending (expected \\r\\n) on line %d", lineNo)
}
line = line[:len(line)-2]
if len(line) == 0 || line[0] == '#' {
continue
}
if line[0] == '[' {
line = line[1 : len(line)-1]
curve, hash, _ := strings.Cut(line, ",")
switch curve {
case "P-224":
pub.Curve = elliptic.P224()
case "P-256":
pub.Curve = elliptic.P256()
case "P-384":
pub.Curve = elliptic.P384()
case "P-521":
pub.Curve = elliptic.P521()
default:
pub.Curve = nil
}
switch hash {
case "SHA-1":
h = sha1.New()
case "SHA-224":
h = sha256.New224()
case "SHA-256":
h = sha256.New()
case "SHA-384":
h = sha512.New384()
case "SHA-512":
h = sha512.New()
default:
h = nil
}
continue
}
if h == nil || pub.Curve == nil {
continue
}
switch {
case strings.HasPrefix(line, "Msg = "):
if msg, err = hex.DecodeString(line[6:]); err != nil {
t.Fatalf("failed to decode message on line %d: %s", lineNo, err)
}
case strings.HasPrefix(line, "Qx = "):
pub.X = fromHex(line[5:])
case strings.HasPrefix(line, "Qy = "):
pub.Y = fromHex(line[5:])
case strings.HasPrefix(line, "R = "):
r = fromHex(line[4:])
case strings.HasPrefix(line, "S = "):
s = fromHex(line[4:])
case strings.HasPrefix(line, "Result = "):
expected := line[9] == 'P'
h.Reset()
h.Write(msg)
hashed := h.Sum(hashed[:0])
if Verify(pub, hashed, r, s) != expected {
t.Fatalf("incorrect result on line %d", lineNo)
}
default:
t.Fatalf("unknown variable on line %d: %s", lineNo, line)
}
}
}
func TestNegativeInputs(t *testing.T) {
testAllCurves(t, testNegativeInputs)
}
func testNegativeInputs(t *testing.T, curve elliptic.Curve) {
key, err := GenerateKey(curve, rand.Reader)
if err != nil {
t.Errorf("failed to generate key")
}
var hash [32]byte
r := new(big.Int).SetInt64(1)
r.Lsh(r, 550 /* larger than any supported curve */)
r.Neg(r)
if Verify(&key.PublicKey, hash[:], r, r) {
t.Errorf("bogus signature accepted")
}
}
func TestZeroHashSignature(t *testing.T) {
testAllCurves(t, testZeroHashSignature)
}
func testZeroHashSignature(t *testing.T, curve elliptic.Curve) {
zeroHash := make([]byte, 64)
privKey, err := GenerateKey(curve, rand.Reader)
if err != nil {
panic(err)
}
// Sign a hash consisting of all zeros.
r, s, err := Sign(rand.Reader, privKey, zeroHash)
if err != nil {
panic(err)
}
// Confirm that it can be verified.
if !Verify(&privKey.PublicKey, zeroHash, r, s) {
t.Errorf("zero hash signature verify failed for %T", curve)
}
}
func benchmarkAllCurves(t *testing.B, f func(*testing.B, elliptic.Curve)) {
tests := []struct {
name string
curve elliptic.Curve
}{
{"P256", elliptic.P256()},
{"P224", elliptic.P224()},
{"P384", elliptic.P384()},
{"P521", elliptic.P521()},
}
for _, test := range tests {
curve := test.curve
t.Run(test.name, func(t *testing.B) {
f(t, curve)
})
}
}
func BenchmarkSign(b *testing.B) {
benchmarkAllCurves(b, func(b *testing.B, curve elliptic.Curve) {
priv, err := GenerateKey(curve, rand.Reader)
if err != nil {
b.Fatal(err)
}
hashed := []byte("testing")
b.ReportAllocs()
b.ResetTimer()
for i := 0; i < b.N; i++ {
sig, err := SignASN1(rand.Reader, priv, hashed)
if err != nil {
b.Fatal(err)
}
// Prevent the compiler from optimizing out the operation.
hashed[0] = sig[0]
}
})
}
func BenchmarkVerify(b *testing.B) {
benchmarkAllCurves(b, func(b *testing.B, curve elliptic.Curve) {
priv, err := GenerateKey(curve, rand.Reader)
if err != nil {
b.Fatal(err)
}
hashed := []byte("testing")
r, s, err := Sign(rand.Reader, priv, hashed)
if err != nil {
b.Fatal(err)
}
b.ReportAllocs()
b.ResetTimer()
for i := 0; i < b.N; i++ {
if !Verify(&priv.PublicKey, hashed, r, s) {
b.Fatal("verify failed")
}
}
})
}
func BenchmarkGenerateKey(b *testing.B) {
benchmarkAllCurves(b, func(b *testing.B, curve elliptic.Curve) {
b.ReportAllocs()
b.ResetTimer()
for i := 0; i < b.N; i++ {
if _, err := GenerateKey(curve, rand.Reader); err != nil {
b.Fatal(err)
}
}
})
}