|
|
|
/* Copyright (C) 1995-1998 Eric Young (eay@cryptsoft.com)
|
|
|
|
* All rights reserved.
|
|
|
|
*
|
|
|
|
* This package is an SSL implementation written
|
|
|
|
* by Eric Young (eay@cryptsoft.com).
|
|
|
|
* The implementation was written so as to conform with Netscapes SSL.
|
|
|
|
*
|
|
|
|
* This library is free for commercial and non-commercial use as long as
|
|
|
|
* the following conditions are aheared to. The following conditions
|
|
|
|
* apply to all code found in this distribution, be it the RC4, RSA,
|
|
|
|
* lhash, DES, etc., code; not just the SSL code. The SSL documentation
|
|
|
|
* included with this distribution is covered by the same copyright terms
|
|
|
|
* except that the holder is Tim Hudson (tjh@cryptsoft.com).
|
|
|
|
*
|
|
|
|
* Copyright remains Eric Young's, and as such any Copyright notices in
|
|
|
|
* the code are not to be removed.
|
|
|
|
* If this package is used in a product, Eric Young should be given attribution
|
|
|
|
* as the author of the parts of the library used.
|
|
|
|
* This can be in the form of a textual message at program startup or
|
|
|
|
* in documentation (online or textual) provided with the package.
|
|
|
|
*
|
|
|
|
* Redistribution and use in source and binary forms, with or without
|
|
|
|
* modification, are permitted provided that the following conditions
|
|
|
|
* are met:
|
|
|
|
* 1. Redistributions of source code must retain the copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer.
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
|
|
* documentation and/or other materials provided with the distribution.
|
|
|
|
* 3. All advertising materials mentioning features or use of this software
|
|
|
|
* must display the following acknowledgement:
|
|
|
|
* "This product includes cryptographic software written by
|
|
|
|
* Eric Young (eay@cryptsoft.com)"
|
|
|
|
* The word 'cryptographic' can be left out if the rouines from the library
|
|
|
|
* being used are not cryptographic related :-).
|
|
|
|
* 4. If you include any Windows specific code (or a derivative thereof) from
|
|
|
|
* the apps directory (application code) you must include an acknowledgement:
|
|
|
|
* "This product includes software written by Tim Hudson (tjh@cryptsoft.com)"
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY ERIC YOUNG ``AS IS'' AND
|
|
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
|
|
* SUCH DAMAGE.
|
|
|
|
*
|
|
|
|
* The licence and distribution terms for any publically available version or
|
|
|
|
* derivative of this code cannot be changed. i.e. this code cannot simply be
|
|
|
|
* copied and put under another distribution licence
|
|
|
|
* [including the GNU Public Licence.]
|
|
|
|
*
|
|
|
|
* The DSS routines are based on patches supplied by
|
|
|
|
* Steven Schoch <schoch@sheba.arc.nasa.gov>. */
|
|
|
|
|
|
|
|
#include <openssl/dsa.h>
|
|
|
|
|
|
|
|
#include <string.h>
|
|
|
|
|
|
|
|
#include <openssl/bn.h>
|
|
|
|
#include <openssl/dh.h>
|
|
|
|
#include <openssl/digest.h>
|
|
|
|
#include <openssl/engine.h>
|
|
|
|
#include <openssl/err.h>
|
|
|
|
#include <openssl/ex_data.h>
|
|
|
|
#include <openssl/mem.h>
|
|
|
|
#include <openssl/rand.h>
|
|
|
|
#include <openssl/sha.h>
|
|
|
|
#include <openssl/thread.h>
|
|
|
|
|
Bound RSA and DSA key sizes better.
Most asymmetric operations scale superlinearly, which makes them
potential DoS vectors. This (and other problems) are mitigated with
fixed sizes, like RSA-2048, P-256, or curve25519.
In older algorithms like RSA and DSA, these sizes are conventions rather
than well-defined algorithms. "Everyone" uses RSA-2048, but code which
imports an RSA key may see an arbitrary key size, possibly from an
untrusted source. This is commonly a public key, so we bound RSA key
sizes in check_modulus_and_exponent_sizes.
However, some applications import external private keys, and may need
tighter bounds. These typically parse the key then check the result.
However, parsing itself can perform superlinear work (RSA_check_key or
recovering the DSA public key).
This CL does the following:
- Rename check_modulus_and_exponent_sizes to rsa_check_public_key and
additionally call it from RSA_check_key.
- Fix a bug where RSA_check_key, on CRT-less keys, did not bound d, and
bound p and q before multiplying (quadratic).
- Our DSA verifier had stricter checks on q (160-, 224-, and 256-bit
only) than our DSA signer (multiple of 8 bits). Aligner the signer to
the verifier's checks.
- Validate DSA group sizes on parse, as well as priv_key < q, to bound
the running time.
Ideally these invariants would be checked exactly once at construction,
but our RSA and DSA implementations suffer from some OpenSSL's API
mistakes (https://crbug.com/boringssl/316), which means it is hard to
consistently enforce invariants. This CL focuses on the parser, but
later I'd like to better rationalize the freeze_private_key logic.
Performance of parsing RSA and DSA keys, gathered on my laptop.
Did 15130 RSA-2048 parse operations in 5022458us (3012.5 ops/sec)
Did 4888 RSA-4096 parse operations in 5060606us (965.9 ops/sec)
Did 354 RSA-16384 parse operations in 5043565us (70.2 ops/sec)
Did 88 RSA-32768 parse operations in 5038293us (17.5 ops/sec) [rejected by this CL]
Did 35000 DSA-1024/256 parse operations in 5030447us (6957.6 ops/sec)
Did 11316 DSA-2048/256 parse operations in 5094664us (2221.1 ops/sec)
Did 5488 DSA-3072/256 parse operations in 5096032us (1076.9 ops/sec)
Did 3172 DSA-4096/256 parse operations in 5041220us (629.2 ops/sec)
Did 840 DSA-8192/256 parse operations in 5070616us (165.7 ops/sec)
Did 285 DSA-10000/256 parse operations in 5004033us (57.0 ops/sec)
Did 74 DSA-20000/256 parse operations in 5066299us (14.6 ops/sec) [rejected by this CL]
Update-Note: Some invalid or overly large RSA and DSA keys may
previously have been accepted that are now rejected at parse time. For
public keys, this only moves the error from verification to parsing. In
some private key cases, we would previously allow signing with those
keys, but the resulting signatures would not be accepted by BoringSSL
anyway. This CL makes us behave more consistently.
Bug: oss-fuzz:24730
Change-Id: I4ad2003ee61138b693e65d3da4c6aa00bc165251
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/42504
Reviewed-by: Adam Langley <agl@google.com>
5 years ago
|
|
|
#include "internal.h"
|
|
|
|
#include "../fipsmodule/bn/internal.h"
|
|
|
|
#include "../internal.h"
|
|
|
|
|
|
|
|
|
|
|
|
// Primality test according to FIPS PUB 186[-1], Appendix 2.1: 50 rounds of
|
|
|
|
// Miller-Rabin.
|
|
|
|
#define DSS_prime_checks 50
|
|
|
|
|
|
|
|
static int dsa_sign_setup(const DSA *dsa, BN_CTX *ctx_in, BIGNUM **out_kinv,
|
|
|
|
BIGNUM **out_r);
|
|
|
|
|
|
|
|
static CRYPTO_EX_DATA_CLASS g_ex_data_class = CRYPTO_EX_DATA_CLASS_INIT;
|
|
|
|
|
|
|
|
DSA *DSA_new(void) {
|
|
|
|
DSA *dsa = OPENSSL_malloc(sizeof(DSA));
|
|
|
|
if (dsa == NULL) {
|
|
|
|
OPENSSL_PUT_ERROR(DSA, ERR_R_MALLOC_FAILURE);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
OPENSSL_memset(dsa, 0, sizeof(DSA));
|
|
|
|
|
|
|
|
dsa->references = 1;
|
|
|
|
|
|
|
|
CRYPTO_MUTEX_init(&dsa->method_mont_lock);
|
|
|
|
CRYPTO_new_ex_data(&dsa->ex_data);
|
|
|
|
|
|
|
|
return dsa;
|
|
|
|
}
|
|
|
|
|
|
|
|
void DSA_free(DSA *dsa) {
|
|
|
|
if (dsa == NULL) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!CRYPTO_refcount_dec_and_test_zero(&dsa->references)) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
CRYPTO_free_ex_data(&g_ex_data_class, dsa, &dsa->ex_data);
|
|
|
|
|
|
|
|
BN_clear_free(dsa->p);
|
|
|
|
BN_clear_free(dsa->q);
|
|
|
|
BN_clear_free(dsa->g);
|
|
|
|
BN_clear_free(dsa->pub_key);
|
|
|
|
BN_clear_free(dsa->priv_key);
|
|
|
|
BN_MONT_CTX_free(dsa->method_mont_p);
|
|
|
|
BN_MONT_CTX_free(dsa->method_mont_q);
|
|
|
|
CRYPTO_MUTEX_cleanup(&dsa->method_mont_lock);
|
|
|
|
OPENSSL_free(dsa);
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_up_ref(DSA *dsa) {
|
|
|
|
CRYPTO_refcount_inc(&dsa->references);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
const BIGNUM *DSA_get0_pub_key(const DSA *dsa) { return dsa->pub_key; }
|
|
|
|
|
|
|
|
const BIGNUM *DSA_get0_priv_key(const DSA *dsa) { return dsa->priv_key; }
|
|
|
|
|
|
|
|
const BIGNUM *DSA_get0_p(const DSA *dsa) { return dsa->p; }
|
|
|
|
|
|
|
|
const BIGNUM *DSA_get0_q(const DSA *dsa) { return dsa->q; }
|
|
|
|
|
|
|
|
const BIGNUM *DSA_get0_g(const DSA *dsa) { return dsa->g; }
|
|
|
|
|
|
|
|
void DSA_get0_key(const DSA *dsa, const BIGNUM **out_pub_key,
|
|
|
|
const BIGNUM **out_priv_key) {
|
|
|
|
if (out_pub_key != NULL) {
|
|
|
|
*out_pub_key = dsa->pub_key;
|
|
|
|
}
|
|
|
|
if (out_priv_key != NULL) {
|
|
|
|
*out_priv_key = dsa->priv_key;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void DSA_get0_pqg(const DSA *dsa, const BIGNUM **out_p, const BIGNUM **out_q,
|
|
|
|
const BIGNUM **out_g) {
|
|
|
|
if (out_p != NULL) {
|
|
|
|
*out_p = dsa->p;
|
|
|
|
}
|
|
|
|
if (out_q != NULL) {
|
|
|
|
*out_q = dsa->q;
|
|
|
|
}
|
|
|
|
if (out_g != NULL) {
|
|
|
|
*out_g = dsa->g;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_set0_key(DSA *dsa, BIGNUM *pub_key, BIGNUM *priv_key) {
|
|
|
|
if (dsa->pub_key == NULL && pub_key == NULL) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (pub_key != NULL) {
|
|
|
|
BN_free(dsa->pub_key);
|
|
|
|
dsa->pub_key = pub_key;
|
|
|
|
}
|
|
|
|
if (priv_key != NULL) {
|
|
|
|
BN_free(dsa->priv_key);
|
|
|
|
dsa->priv_key = priv_key;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_set0_pqg(DSA *dsa, BIGNUM *p, BIGNUM *q, BIGNUM *g) {
|
|
|
|
if ((dsa->p == NULL && p == NULL) ||
|
|
|
|
(dsa->q == NULL && q == NULL) ||
|
|
|
|
(dsa->g == NULL && g == NULL)) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (p != NULL) {
|
|
|
|
BN_free(dsa->p);
|
|
|
|
dsa->p = p;
|
|
|
|
}
|
|
|
|
if (q != NULL) {
|
|
|
|
BN_free(dsa->q);
|
|
|
|
dsa->q = q;
|
|
|
|
}
|
|
|
|
if (g != NULL) {
|
|
|
|
BN_free(dsa->g);
|
|
|
|
dsa->g = g;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_generate_parameters_ex(DSA *dsa, unsigned bits, const uint8_t *seed_in,
|
|
|
|
size_t seed_len, int *out_counter,
|
|
|
|
unsigned long *out_h, BN_GENCB *cb) {
|
|
|
|
int ok = 0;
|
|
|
|
unsigned char seed[SHA256_DIGEST_LENGTH];
|
|
|
|
unsigned char md[SHA256_DIGEST_LENGTH];
|
|
|
|
unsigned char buf[SHA256_DIGEST_LENGTH], buf2[SHA256_DIGEST_LENGTH];
|
|
|
|
BIGNUM *r0, *W, *X, *c, *test;
|
|
|
|
BIGNUM *g = NULL, *q = NULL, *p = NULL;
|
|
|
|
BN_MONT_CTX *mont = NULL;
|
|
|
|
int k, n = 0, m = 0;
|
|
|
|
unsigned i;
|
|
|
|
int counter = 0;
|
|
|
|
int r = 0;
|
|
|
|
BN_CTX *ctx = NULL;
|
|
|
|
unsigned int h = 2;
|
|
|
|
unsigned qsize;
|
|
|
|
const EVP_MD *evpmd;
|
|
|
|
|
|
|
|
evpmd = (bits >= 2048) ? EVP_sha256() : EVP_sha1();
|
|
|
|
qsize = EVP_MD_size(evpmd);
|
|
|
|
|
|
|
|
if (bits < 512) {
|
|
|
|
bits = 512;
|
|
|
|
}
|
|
|
|
|
|
|
|
bits = (bits + 63) / 64 * 64;
|
|
|
|
|
|
|
|
if (seed_in != NULL) {
|
|
|
|
if (seed_len < (size_t)qsize) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
if (seed_len > (size_t)qsize) {
|
|
|
|
// Only consume as much seed as is expected.
|
|
|
|
seed_len = qsize;
|
|
|
|
}
|
|
|
|
OPENSSL_memcpy(seed, seed_in, seed_len);
|
|
|
|
}
|
|
|
|
|
|
|
|
ctx = BN_CTX_new();
|
|
|
|
if (ctx == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
BN_CTX_start(ctx);
|
|
|
|
|
|
|
|
r0 = BN_CTX_get(ctx);
|
|
|
|
g = BN_CTX_get(ctx);
|
|
|
|
W = BN_CTX_get(ctx);
|
|
|
|
q = BN_CTX_get(ctx);
|
|
|
|
X = BN_CTX_get(ctx);
|
|
|
|
c = BN_CTX_get(ctx);
|
|
|
|
p = BN_CTX_get(ctx);
|
|
|
|
test = BN_CTX_get(ctx);
|
|
|
|
|
|
|
|
if (test == NULL || !BN_lshift(test, BN_value_one(), bits - 1)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
for (;;) {
|
|
|
|
// Find q.
|
|
|
|
for (;;) {
|
|
|
|
// step 1
|
|
|
|
if (!BN_GENCB_call(cb, BN_GENCB_GENERATED, m++)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
int use_random_seed = (seed_in == NULL);
|
|
|
|
if (use_random_seed) {
|
|
|
|
if (!RAND_bytes(seed, qsize)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
// If we come back through, use random seed next time.
|
|
|
|
seed_in = NULL;
|
|
|
|
}
|
|
|
|
OPENSSL_memcpy(buf, seed, qsize);
|
|
|
|
OPENSSL_memcpy(buf2, seed, qsize);
|
|
|
|
// precompute "SEED + 1" for step 7:
|
|
|
|
for (i = qsize - 1; i < qsize; i--) {
|
|
|
|
buf[i]++;
|
|
|
|
if (buf[i] != 0) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// step 2
|
|
|
|
if (!EVP_Digest(seed, qsize, md, NULL, evpmd, NULL) ||
|
|
|
|
!EVP_Digest(buf, qsize, buf2, NULL, evpmd, NULL)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
for (i = 0; i < qsize; i++) {
|
|
|
|
md[i] ^= buf2[i];
|
|
|
|
}
|
|
|
|
|
|
|
|
// step 3
|
|
|
|
md[0] |= 0x80;
|
|
|
|
md[qsize - 1] |= 0x01;
|
|
|
|
if (!BN_bin2bn(md, qsize, q)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// step 4
|
|
|
|
r = BN_is_prime_fasttest_ex(q, DSS_prime_checks, ctx, use_random_seed, cb);
|
|
|
|
if (r > 0) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (r != 0) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// do a callback call
|
|
|
|
// step 5
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!BN_GENCB_call(cb, 2, 0) || !BN_GENCB_call(cb, 3, 0)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// step 6
|
|
|
|
counter = 0;
|
|
|
|
// "offset = 2"
|
|
|
|
|
|
|
|
n = (bits - 1) / 160;
|
|
|
|
|
|
|
|
for (;;) {
|
|
|
|
if ((counter != 0) && !BN_GENCB_call(cb, BN_GENCB_GENERATED, counter)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// step 7
|
|
|
|
BN_zero(W);
|
|
|
|
// now 'buf' contains "SEED + offset - 1"
|
|
|
|
for (k = 0; k <= n; k++) {
|
|
|
|
// obtain "SEED + offset + k" by incrementing:
|
|
|
|
for (i = qsize - 1; i < qsize; i--) {
|
|
|
|
buf[i]++;
|
|
|
|
if (buf[i] != 0) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!EVP_Digest(buf, qsize, md, NULL, evpmd, NULL)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// step 8
|
|
|
|
if (!BN_bin2bn(md, qsize, r0) ||
|
|
|
|
!BN_lshift(r0, r0, (qsize << 3) * k) ||
|
|
|
|
!BN_add(W, W, r0)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// more of step 8
|
|
|
|
if (!BN_mask_bits(W, bits - 1) ||
|
|
|
|
!BN_copy(X, W) ||
|
|
|
|
!BN_add(X, X, test)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// step 9
|
|
|
|
if (!BN_lshift1(r0, q) ||
|
|
|
|
!BN_mod(c, X, r0, ctx) ||
|
|
|
|
!BN_sub(r0, c, BN_value_one()) ||
|
|
|
|
!BN_sub(p, X, r0)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// step 10
|
|
|
|
if (BN_cmp(p, test) >= 0) {
|
|
|
|
// step 11
|
|
|
|
r = BN_is_prime_fasttest_ex(p, DSS_prime_checks, ctx, 1, cb);
|
|
|
|
if (r > 0) {
|
|
|
|
goto end; // found it
|
|
|
|
}
|
|
|
|
if (r != 0) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// step 13
|
|
|
|
counter++;
|
|
|
|
// "offset = offset + n + 1"
|
|
|
|
|
|
|
|
// step 14
|
|
|
|
if (counter >= 4096) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
end:
|
|
|
|
if (!BN_GENCB_call(cb, 2, 1)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// We now need to generate g
|
|
|
|
// Set r0=(p-1)/q
|
|
|
|
if (!BN_sub(test, p, BN_value_one()) ||
|
|
|
|
!BN_div(r0, NULL, test, q, ctx)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
mont = BN_MONT_CTX_new_for_modulus(p, ctx);
|
|
|
|
if (mont == NULL ||
|
|
|
|
!BN_set_word(test, h)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
for (;;) {
|
|
|
|
// g=test^r0%p
|
|
|
|
if (!BN_mod_exp_mont(g, test, r0, p, ctx, mont)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
if (!BN_is_one(g)) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (!BN_add(test, test, BN_value_one())) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
h++;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!BN_GENCB_call(cb, 3, 1)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
ok = 1;
|
|
|
|
|
|
|
|
err:
|
|
|
|
if (ok) {
|
|
|
|
BN_free(dsa->p);
|
|
|
|
BN_free(dsa->q);
|
|
|
|
BN_free(dsa->g);
|
|
|
|
dsa->p = BN_dup(p);
|
|
|
|
dsa->q = BN_dup(q);
|
|
|
|
dsa->g = BN_dup(g);
|
|
|
|
if (dsa->p == NULL || dsa->q == NULL || dsa->g == NULL) {
|
|
|
|
ok = 0;
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
if (out_counter != NULL) {
|
|
|
|
*out_counter = counter;
|
|
|
|
}
|
|
|
|
if (out_h != NULL) {
|
|
|
|
*out_h = h;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ctx) {
|
|
|
|
BN_CTX_end(ctx);
|
|
|
|
BN_CTX_free(ctx);
|
|
|
|
}
|
|
|
|
|
|
|
|
BN_MONT_CTX_free(mont);
|
|
|
|
|
|
|
|
return ok;
|
|
|
|
}
|
|
|
|
|
|
|
|
DSA *DSAparams_dup(const DSA *dsa) {
|
|
|
|
DSA *ret = DSA_new();
|
|
|
|
if (ret == NULL) {
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
ret->p = BN_dup(dsa->p);
|
|
|
|
ret->q = BN_dup(dsa->q);
|
|
|
|
ret->g = BN_dup(dsa->g);
|
|
|
|
if (ret->p == NULL || ret->q == NULL || ret->g == NULL) {
|
|
|
|
DSA_free(ret);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_generate_key(DSA *dsa) {
|
|
|
|
int ok = 0;
|
|
|
|
BN_CTX *ctx = NULL;
|
|
|
|
BIGNUM *pub_key = NULL, *priv_key = NULL;
|
|
|
|
|
|
|
|
ctx = BN_CTX_new();
|
|
|
|
if (ctx == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
priv_key = dsa->priv_key;
|
|
|
|
if (priv_key == NULL) {
|
|
|
|
priv_key = BN_new();
|
|
|
|
if (priv_key == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!BN_rand_range_ex(priv_key, 1, dsa->q)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
pub_key = dsa->pub_key;
|
|
|
|
if (pub_key == NULL) {
|
|
|
|
pub_key = BN_new();
|
|
|
|
if (pub_key == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!BN_MONT_CTX_set_locked(&dsa->method_mont_p, &dsa->method_mont_lock,
|
|
|
|
dsa->p, ctx) ||
|
|
|
|
!BN_mod_exp_mont_consttime(pub_key, dsa->g, priv_key, dsa->p, ctx,
|
|
|
|
dsa->method_mont_p)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
dsa->priv_key = priv_key;
|
|
|
|
dsa->pub_key = pub_key;
|
|
|
|
ok = 1;
|
|
|
|
|
|
|
|
err:
|
|
|
|
if (dsa->pub_key == NULL) {
|
|
|
|
BN_free(pub_key);
|
|
|
|
}
|
|
|
|
if (dsa->priv_key == NULL) {
|
|
|
|
BN_free(priv_key);
|
|
|
|
}
|
|
|
|
BN_CTX_free(ctx);
|
|
|
|
|
|
|
|
return ok;
|
|
|
|
}
|
|
|
|
|
|
|
|
DSA_SIG *DSA_SIG_new(void) {
|
|
|
|
DSA_SIG *sig;
|
|
|
|
sig = OPENSSL_malloc(sizeof(DSA_SIG));
|
|
|
|
if (!sig) {
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
sig->r = NULL;
|
|
|
|
sig->s = NULL;
|
|
|
|
return sig;
|
|
|
|
}
|
|
|
|
|
|
|
|
void DSA_SIG_free(DSA_SIG *sig) {
|
|
|
|
if (!sig) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
BN_free(sig->r);
|
|
|
|
BN_free(sig->s);
|
|
|
|
OPENSSL_free(sig);
|
|
|
|
}
|
|
|
|
|
|
|
|
// mod_mul_consttime sets |r| to |a| * |b| modulo |mont->N|, treating |a| and
|
|
|
|
// |b| as secret. This function internally uses Montgomery reduction, but
|
|
|
|
// neither inputs nor outputs are in Montgomery form.
|
|
|
|
static int mod_mul_consttime(BIGNUM *r, const BIGNUM *a, const BIGNUM *b,
|
|
|
|
const BN_MONT_CTX *mont, BN_CTX *ctx) {
|
|
|
|
BN_CTX_start(ctx);
|
|
|
|
BIGNUM *tmp = BN_CTX_get(ctx);
|
|
|
|
// |BN_mod_mul_montgomery| removes a factor of R, so we cancel it with a
|
|
|
|
// single |BN_to_montgomery| which adds one factor of R.
|
|
|
|
int ok = tmp != NULL &&
|
|
|
|
BN_to_montgomery(tmp, a, mont, ctx) &&
|
|
|
|
BN_mod_mul_montgomery(r, tmp, b, mont, ctx);
|
|
|
|
BN_CTX_end(ctx);
|
|
|
|
return ok;
|
|
|
|
}
|
|
|
|
|
|
|
|
DSA_SIG *DSA_do_sign(const uint8_t *digest, size_t digest_len, const DSA *dsa) {
|
Bound RSA and DSA key sizes better.
Most asymmetric operations scale superlinearly, which makes them
potential DoS vectors. This (and other problems) are mitigated with
fixed sizes, like RSA-2048, P-256, or curve25519.
In older algorithms like RSA and DSA, these sizes are conventions rather
than well-defined algorithms. "Everyone" uses RSA-2048, but code which
imports an RSA key may see an arbitrary key size, possibly from an
untrusted source. This is commonly a public key, so we bound RSA key
sizes in check_modulus_and_exponent_sizes.
However, some applications import external private keys, and may need
tighter bounds. These typically parse the key then check the result.
However, parsing itself can perform superlinear work (RSA_check_key or
recovering the DSA public key).
This CL does the following:
- Rename check_modulus_and_exponent_sizes to rsa_check_public_key and
additionally call it from RSA_check_key.
- Fix a bug where RSA_check_key, on CRT-less keys, did not bound d, and
bound p and q before multiplying (quadratic).
- Our DSA verifier had stricter checks on q (160-, 224-, and 256-bit
only) than our DSA signer (multiple of 8 bits). Aligner the signer to
the verifier's checks.
- Validate DSA group sizes on parse, as well as priv_key < q, to bound
the running time.
Ideally these invariants would be checked exactly once at construction,
but our RSA and DSA implementations suffer from some OpenSSL's API
mistakes (https://crbug.com/boringssl/316), which means it is hard to
consistently enforce invariants. This CL focuses on the parser, but
later I'd like to better rationalize the freeze_private_key logic.
Performance of parsing RSA and DSA keys, gathered on my laptop.
Did 15130 RSA-2048 parse operations in 5022458us (3012.5 ops/sec)
Did 4888 RSA-4096 parse operations in 5060606us (965.9 ops/sec)
Did 354 RSA-16384 parse operations in 5043565us (70.2 ops/sec)
Did 88 RSA-32768 parse operations in 5038293us (17.5 ops/sec) [rejected by this CL]
Did 35000 DSA-1024/256 parse operations in 5030447us (6957.6 ops/sec)
Did 11316 DSA-2048/256 parse operations in 5094664us (2221.1 ops/sec)
Did 5488 DSA-3072/256 parse operations in 5096032us (1076.9 ops/sec)
Did 3172 DSA-4096/256 parse operations in 5041220us (629.2 ops/sec)
Did 840 DSA-8192/256 parse operations in 5070616us (165.7 ops/sec)
Did 285 DSA-10000/256 parse operations in 5004033us (57.0 ops/sec)
Did 74 DSA-20000/256 parse operations in 5066299us (14.6 ops/sec) [rejected by this CL]
Update-Note: Some invalid or overly large RSA and DSA keys may
previously have been accepted that are now rejected at parse time. For
public keys, this only moves the error from verification to parsing. In
some private key cases, we would previously allow signing with those
keys, but the resulting signatures would not be accepted by BoringSSL
anyway. This CL makes us behave more consistently.
Bug: oss-fuzz:24730
Change-Id: I4ad2003ee61138b693e65d3da4c6aa00bc165251
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/42504
Reviewed-by: Adam Langley <agl@google.com>
5 years ago
|
|
|
if (!dsa_check_parameters(dsa)) {
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
BIGNUM *kinv = NULL, *r = NULL, *s = NULL;
|
|
|
|
BIGNUM m;
|
|
|
|
BIGNUM xr;
|
|
|
|
BN_CTX *ctx = NULL;
|
|
|
|
DSA_SIG *ret = NULL;
|
|
|
|
|
|
|
|
BN_init(&m);
|
|
|
|
BN_init(&xr);
|
|
|
|
s = BN_new();
|
|
|
|
if (s == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
ctx = BN_CTX_new();
|
|
|
|
if (ctx == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
redo:
|
|
|
|
if (!dsa_sign_setup(dsa, ctx, &kinv, &r)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (digest_len > BN_num_bytes(dsa->q)) {
|
|
|
|
// If the digest length is greater than the size of |dsa->q| use the
|
|
|
|
// BN_num_bits(dsa->q) leftmost bits of the digest, see FIPS 186-3, 4.2.
|
|
|
|
// Note the above check that |dsa->q| is a multiple of 8 bits.
|
|
|
|
digest_len = BN_num_bytes(dsa->q);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (BN_bin2bn(digest, digest_len, &m) == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// |m| is bounded by 2^(num_bits(q)), which is slightly looser than q. This
|
|
|
|
// violates |bn_mod_add_consttime| and |mod_mul_consttime|'s preconditions.
|
|
|
|
// (The underlying algorithms could accept looser bounds, but we reduce for
|
|
|
|
// simplicity.)
|
|
|
|
size_t q_width = bn_minimal_width(dsa->q);
|
|
|
|
if (!bn_resize_words(&m, q_width) ||
|
|
|
|
!bn_resize_words(&xr, q_width)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
bn_reduce_once_in_place(m.d, 0 /* no carry word */, dsa->q->d,
|
|
|
|
xr.d /* scratch space */, q_width);
|
|
|
|
|
|
|
|
// Compute s = inv(k) (m + xr) mod q. Note |dsa->method_mont_q| is
|
|
|
|
// initialized by |dsa_sign_setup|.
|
|
|
|
if (!mod_mul_consttime(&xr, dsa->priv_key, r, dsa->method_mont_q, ctx) ||
|
|
|
|
!bn_mod_add_consttime(s, &xr, &m, dsa->q, ctx) ||
|
|
|
|
!mod_mul_consttime(s, s, kinv, dsa->method_mont_q, ctx)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Redo if r or s is zero as required by FIPS 186-3: this is
|
|
|
|
// very unlikely.
|
|
|
|
if (BN_is_zero(r) || BN_is_zero(s)) {
|
|
|
|
goto redo;
|
|
|
|
}
|
|
|
|
ret = DSA_SIG_new();
|
|
|
|
if (ret == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
ret->r = r;
|
|
|
|
ret->s = s;
|
|
|
|
|
|
|
|
err:
|
|
|
|
if (ret == NULL) {
|
|
|
|
OPENSSL_PUT_ERROR(DSA, ERR_R_BN_LIB);
|
|
|
|
BN_free(r);
|
|
|
|
BN_free(s);
|
|
|
|
}
|
|
|
|
BN_CTX_free(ctx);
|
|
|
|
BN_clear_free(&m);
|
|
|
|
BN_clear_free(&xr);
|
|
|
|
BN_clear_free(kinv);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_do_verify(const uint8_t *digest, size_t digest_len, DSA_SIG *sig,
|
|
|
|
const DSA *dsa) {
|
|
|
|
int valid;
|
|
|
|
if (!DSA_do_check_signature(&valid, digest, digest_len, sig, dsa)) {
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return valid;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_do_check_signature(int *out_valid, const uint8_t *digest,
|
|
|
|
size_t digest_len, DSA_SIG *sig, const DSA *dsa) {
|
|
|
|
*out_valid = 0;
|
Bound RSA and DSA key sizes better.
Most asymmetric operations scale superlinearly, which makes them
potential DoS vectors. This (and other problems) are mitigated with
fixed sizes, like RSA-2048, P-256, or curve25519.
In older algorithms like RSA and DSA, these sizes are conventions rather
than well-defined algorithms. "Everyone" uses RSA-2048, but code which
imports an RSA key may see an arbitrary key size, possibly from an
untrusted source. This is commonly a public key, so we bound RSA key
sizes in check_modulus_and_exponent_sizes.
However, some applications import external private keys, and may need
tighter bounds. These typically parse the key then check the result.
However, parsing itself can perform superlinear work (RSA_check_key or
recovering the DSA public key).
This CL does the following:
- Rename check_modulus_and_exponent_sizes to rsa_check_public_key and
additionally call it from RSA_check_key.
- Fix a bug where RSA_check_key, on CRT-less keys, did not bound d, and
bound p and q before multiplying (quadratic).
- Our DSA verifier had stricter checks on q (160-, 224-, and 256-bit
only) than our DSA signer (multiple of 8 bits). Aligner the signer to
the verifier's checks.
- Validate DSA group sizes on parse, as well as priv_key < q, to bound
the running time.
Ideally these invariants would be checked exactly once at construction,
but our RSA and DSA implementations suffer from some OpenSSL's API
mistakes (https://crbug.com/boringssl/316), which means it is hard to
consistently enforce invariants. This CL focuses on the parser, but
later I'd like to better rationalize the freeze_private_key logic.
Performance of parsing RSA and DSA keys, gathered on my laptop.
Did 15130 RSA-2048 parse operations in 5022458us (3012.5 ops/sec)
Did 4888 RSA-4096 parse operations in 5060606us (965.9 ops/sec)
Did 354 RSA-16384 parse operations in 5043565us (70.2 ops/sec)
Did 88 RSA-32768 parse operations in 5038293us (17.5 ops/sec) [rejected by this CL]
Did 35000 DSA-1024/256 parse operations in 5030447us (6957.6 ops/sec)
Did 11316 DSA-2048/256 parse operations in 5094664us (2221.1 ops/sec)
Did 5488 DSA-3072/256 parse operations in 5096032us (1076.9 ops/sec)
Did 3172 DSA-4096/256 parse operations in 5041220us (629.2 ops/sec)
Did 840 DSA-8192/256 parse operations in 5070616us (165.7 ops/sec)
Did 285 DSA-10000/256 parse operations in 5004033us (57.0 ops/sec)
Did 74 DSA-20000/256 parse operations in 5066299us (14.6 ops/sec) [rejected by this CL]
Update-Note: Some invalid or overly large RSA and DSA keys may
previously have been accepted that are now rejected at parse time. For
public keys, this only moves the error from verification to parsing. In
some private key cases, we would previously allow signing with those
keys, but the resulting signatures would not be accepted by BoringSSL
anyway. This CL makes us behave more consistently.
Bug: oss-fuzz:24730
Change-Id: I4ad2003ee61138b693e65d3da4c6aa00bc165251
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/42504
Reviewed-by: Adam Langley <agl@google.com>
5 years ago
|
|
|
if (!dsa_check_parameters(dsa)) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
Bound RSA and DSA key sizes better.
Most asymmetric operations scale superlinearly, which makes them
potential DoS vectors. This (and other problems) are mitigated with
fixed sizes, like RSA-2048, P-256, or curve25519.
In older algorithms like RSA and DSA, these sizes are conventions rather
than well-defined algorithms. "Everyone" uses RSA-2048, but code which
imports an RSA key may see an arbitrary key size, possibly from an
untrusted source. This is commonly a public key, so we bound RSA key
sizes in check_modulus_and_exponent_sizes.
However, some applications import external private keys, and may need
tighter bounds. These typically parse the key then check the result.
However, parsing itself can perform superlinear work (RSA_check_key or
recovering the DSA public key).
This CL does the following:
- Rename check_modulus_and_exponent_sizes to rsa_check_public_key and
additionally call it from RSA_check_key.
- Fix a bug where RSA_check_key, on CRT-less keys, did not bound d, and
bound p and q before multiplying (quadratic).
- Our DSA verifier had stricter checks on q (160-, 224-, and 256-bit
only) than our DSA signer (multiple of 8 bits). Aligner the signer to
the verifier's checks.
- Validate DSA group sizes on parse, as well as priv_key < q, to bound
the running time.
Ideally these invariants would be checked exactly once at construction,
but our RSA and DSA implementations suffer from some OpenSSL's API
mistakes (https://crbug.com/boringssl/316), which means it is hard to
consistently enforce invariants. This CL focuses on the parser, but
later I'd like to better rationalize the freeze_private_key logic.
Performance of parsing RSA and DSA keys, gathered on my laptop.
Did 15130 RSA-2048 parse operations in 5022458us (3012.5 ops/sec)
Did 4888 RSA-4096 parse operations in 5060606us (965.9 ops/sec)
Did 354 RSA-16384 parse operations in 5043565us (70.2 ops/sec)
Did 88 RSA-32768 parse operations in 5038293us (17.5 ops/sec) [rejected by this CL]
Did 35000 DSA-1024/256 parse operations in 5030447us (6957.6 ops/sec)
Did 11316 DSA-2048/256 parse operations in 5094664us (2221.1 ops/sec)
Did 5488 DSA-3072/256 parse operations in 5096032us (1076.9 ops/sec)
Did 3172 DSA-4096/256 parse operations in 5041220us (629.2 ops/sec)
Did 840 DSA-8192/256 parse operations in 5070616us (165.7 ops/sec)
Did 285 DSA-10000/256 parse operations in 5004033us (57.0 ops/sec)
Did 74 DSA-20000/256 parse operations in 5066299us (14.6 ops/sec) [rejected by this CL]
Update-Note: Some invalid or overly large RSA and DSA keys may
previously have been accepted that are now rejected at parse time. For
public keys, this only moves the error from verification to parsing. In
some private key cases, we would previously allow signing with those
keys, but the resulting signatures would not be accepted by BoringSSL
anyway. This CL makes us behave more consistently.
Bug: oss-fuzz:24730
Change-Id: I4ad2003ee61138b693e65d3da4c6aa00bc165251
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/42504
Reviewed-by: Adam Langley <agl@google.com>
5 years ago
|
|
|
int ret = 0;
|
|
|
|
BIGNUM u1, u2, t1;
|
|
|
|
BN_init(&u1);
|
|
|
|
BN_init(&u2);
|
|
|
|
BN_init(&t1);
|
Bound RSA and DSA key sizes better.
Most asymmetric operations scale superlinearly, which makes them
potential DoS vectors. This (and other problems) are mitigated with
fixed sizes, like RSA-2048, P-256, or curve25519.
In older algorithms like RSA and DSA, these sizes are conventions rather
than well-defined algorithms. "Everyone" uses RSA-2048, but code which
imports an RSA key may see an arbitrary key size, possibly from an
untrusted source. This is commonly a public key, so we bound RSA key
sizes in check_modulus_and_exponent_sizes.
However, some applications import external private keys, and may need
tighter bounds. These typically parse the key then check the result.
However, parsing itself can perform superlinear work (RSA_check_key or
recovering the DSA public key).
This CL does the following:
- Rename check_modulus_and_exponent_sizes to rsa_check_public_key and
additionally call it from RSA_check_key.
- Fix a bug where RSA_check_key, on CRT-less keys, did not bound d, and
bound p and q before multiplying (quadratic).
- Our DSA verifier had stricter checks on q (160-, 224-, and 256-bit
only) than our DSA signer (multiple of 8 bits). Aligner the signer to
the verifier's checks.
- Validate DSA group sizes on parse, as well as priv_key < q, to bound
the running time.
Ideally these invariants would be checked exactly once at construction,
but our RSA and DSA implementations suffer from some OpenSSL's API
mistakes (https://crbug.com/boringssl/316), which means it is hard to
consistently enforce invariants. This CL focuses on the parser, but
later I'd like to better rationalize the freeze_private_key logic.
Performance of parsing RSA and DSA keys, gathered on my laptop.
Did 15130 RSA-2048 parse operations in 5022458us (3012.5 ops/sec)
Did 4888 RSA-4096 parse operations in 5060606us (965.9 ops/sec)
Did 354 RSA-16384 parse operations in 5043565us (70.2 ops/sec)
Did 88 RSA-32768 parse operations in 5038293us (17.5 ops/sec) [rejected by this CL]
Did 35000 DSA-1024/256 parse operations in 5030447us (6957.6 ops/sec)
Did 11316 DSA-2048/256 parse operations in 5094664us (2221.1 ops/sec)
Did 5488 DSA-3072/256 parse operations in 5096032us (1076.9 ops/sec)
Did 3172 DSA-4096/256 parse operations in 5041220us (629.2 ops/sec)
Did 840 DSA-8192/256 parse operations in 5070616us (165.7 ops/sec)
Did 285 DSA-10000/256 parse operations in 5004033us (57.0 ops/sec)
Did 74 DSA-20000/256 parse operations in 5066299us (14.6 ops/sec) [rejected by this CL]
Update-Note: Some invalid or overly large RSA and DSA keys may
previously have been accepted that are now rejected at parse time. For
public keys, this only moves the error from verification to parsing. In
some private key cases, we would previously allow signing with those
keys, but the resulting signatures would not be accepted by BoringSSL
anyway. This CL makes us behave more consistently.
Bug: oss-fuzz:24730
Change-Id: I4ad2003ee61138b693e65d3da4c6aa00bc165251
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/42504
Reviewed-by: Adam Langley <agl@google.com>
5 years ago
|
|
|
BN_CTX *ctx = BN_CTX_new();
|
|
|
|
if (ctx == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (BN_is_zero(sig->r) || BN_is_negative(sig->r) ||
|
|
|
|
BN_ucmp(sig->r, dsa->q) >= 0) {
|
|
|
|
ret = 1;
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
if (BN_is_zero(sig->s) || BN_is_negative(sig->s) ||
|
|
|
|
BN_ucmp(sig->s, dsa->q) >= 0) {
|
|
|
|
ret = 1;
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Calculate W = inv(S) mod Q
|
|
|
|
// save W in u2
|
|
|
|
if (BN_mod_inverse(&u2, sig->s, dsa->q, ctx) == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// save M in u1
|
Bound RSA and DSA key sizes better.
Most asymmetric operations scale superlinearly, which makes them
potential DoS vectors. This (and other problems) are mitigated with
fixed sizes, like RSA-2048, P-256, or curve25519.
In older algorithms like RSA and DSA, these sizes are conventions rather
than well-defined algorithms. "Everyone" uses RSA-2048, but code which
imports an RSA key may see an arbitrary key size, possibly from an
untrusted source. This is commonly a public key, so we bound RSA key
sizes in check_modulus_and_exponent_sizes.
However, some applications import external private keys, and may need
tighter bounds. These typically parse the key then check the result.
However, parsing itself can perform superlinear work (RSA_check_key or
recovering the DSA public key).
This CL does the following:
- Rename check_modulus_and_exponent_sizes to rsa_check_public_key and
additionally call it from RSA_check_key.
- Fix a bug where RSA_check_key, on CRT-less keys, did not bound d, and
bound p and q before multiplying (quadratic).
- Our DSA verifier had stricter checks on q (160-, 224-, and 256-bit
only) than our DSA signer (multiple of 8 bits). Aligner the signer to
the verifier's checks.
- Validate DSA group sizes on parse, as well as priv_key < q, to bound
the running time.
Ideally these invariants would be checked exactly once at construction,
but our RSA and DSA implementations suffer from some OpenSSL's API
mistakes (https://crbug.com/boringssl/316), which means it is hard to
consistently enforce invariants. This CL focuses on the parser, but
later I'd like to better rationalize the freeze_private_key logic.
Performance of parsing RSA and DSA keys, gathered on my laptop.
Did 15130 RSA-2048 parse operations in 5022458us (3012.5 ops/sec)
Did 4888 RSA-4096 parse operations in 5060606us (965.9 ops/sec)
Did 354 RSA-16384 parse operations in 5043565us (70.2 ops/sec)
Did 88 RSA-32768 parse operations in 5038293us (17.5 ops/sec) [rejected by this CL]
Did 35000 DSA-1024/256 parse operations in 5030447us (6957.6 ops/sec)
Did 11316 DSA-2048/256 parse operations in 5094664us (2221.1 ops/sec)
Did 5488 DSA-3072/256 parse operations in 5096032us (1076.9 ops/sec)
Did 3172 DSA-4096/256 parse operations in 5041220us (629.2 ops/sec)
Did 840 DSA-8192/256 parse operations in 5070616us (165.7 ops/sec)
Did 285 DSA-10000/256 parse operations in 5004033us (57.0 ops/sec)
Did 74 DSA-20000/256 parse operations in 5066299us (14.6 ops/sec) [rejected by this CL]
Update-Note: Some invalid or overly large RSA and DSA keys may
previously have been accepted that are now rejected at parse time. For
public keys, this only moves the error from verification to parsing. In
some private key cases, we would previously allow signing with those
keys, but the resulting signatures would not be accepted by BoringSSL
anyway. This CL makes us behave more consistently.
Bug: oss-fuzz:24730
Change-Id: I4ad2003ee61138b693e65d3da4c6aa00bc165251
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/42504
Reviewed-by: Adam Langley <agl@google.com>
5 years ago
|
|
|
unsigned q_bits = BN_num_bits(dsa->q);
|
|
|
|
if (digest_len > (q_bits >> 3)) {
|
|
|
|
// if the digest length is greater than the size of q use the
|
|
|
|
// BN_num_bits(dsa->q) leftmost bits of the digest, see
|
|
|
|
// fips 186-3, 4.2
|
Bound RSA and DSA key sizes better.
Most asymmetric operations scale superlinearly, which makes them
potential DoS vectors. This (and other problems) are mitigated with
fixed sizes, like RSA-2048, P-256, or curve25519.
In older algorithms like RSA and DSA, these sizes are conventions rather
than well-defined algorithms. "Everyone" uses RSA-2048, but code which
imports an RSA key may see an arbitrary key size, possibly from an
untrusted source. This is commonly a public key, so we bound RSA key
sizes in check_modulus_and_exponent_sizes.
However, some applications import external private keys, and may need
tighter bounds. These typically parse the key then check the result.
However, parsing itself can perform superlinear work (RSA_check_key or
recovering the DSA public key).
This CL does the following:
- Rename check_modulus_and_exponent_sizes to rsa_check_public_key and
additionally call it from RSA_check_key.
- Fix a bug where RSA_check_key, on CRT-less keys, did not bound d, and
bound p and q before multiplying (quadratic).
- Our DSA verifier had stricter checks on q (160-, 224-, and 256-bit
only) than our DSA signer (multiple of 8 bits). Aligner the signer to
the verifier's checks.
- Validate DSA group sizes on parse, as well as priv_key < q, to bound
the running time.
Ideally these invariants would be checked exactly once at construction,
but our RSA and DSA implementations suffer from some OpenSSL's API
mistakes (https://crbug.com/boringssl/316), which means it is hard to
consistently enforce invariants. This CL focuses on the parser, but
later I'd like to better rationalize the freeze_private_key logic.
Performance of parsing RSA and DSA keys, gathered on my laptop.
Did 15130 RSA-2048 parse operations in 5022458us (3012.5 ops/sec)
Did 4888 RSA-4096 parse operations in 5060606us (965.9 ops/sec)
Did 354 RSA-16384 parse operations in 5043565us (70.2 ops/sec)
Did 88 RSA-32768 parse operations in 5038293us (17.5 ops/sec) [rejected by this CL]
Did 35000 DSA-1024/256 parse operations in 5030447us (6957.6 ops/sec)
Did 11316 DSA-2048/256 parse operations in 5094664us (2221.1 ops/sec)
Did 5488 DSA-3072/256 parse operations in 5096032us (1076.9 ops/sec)
Did 3172 DSA-4096/256 parse operations in 5041220us (629.2 ops/sec)
Did 840 DSA-8192/256 parse operations in 5070616us (165.7 ops/sec)
Did 285 DSA-10000/256 parse operations in 5004033us (57.0 ops/sec)
Did 74 DSA-20000/256 parse operations in 5066299us (14.6 ops/sec) [rejected by this CL]
Update-Note: Some invalid or overly large RSA and DSA keys may
previously have been accepted that are now rejected at parse time. For
public keys, this only moves the error from verification to parsing. In
some private key cases, we would previously allow signing with those
keys, but the resulting signatures would not be accepted by BoringSSL
anyway. This CL makes us behave more consistently.
Bug: oss-fuzz:24730
Change-Id: I4ad2003ee61138b693e65d3da4c6aa00bc165251
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/42504
Reviewed-by: Adam Langley <agl@google.com>
5 years ago
|
|
|
digest_len = (q_bits >> 3);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (BN_bin2bn(digest, digest_len, &u1) == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// u1 = M * w mod q
|
|
|
|
if (!BN_mod_mul(&u1, &u1, &u2, dsa->q, ctx)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// u2 = r * w mod q
|
|
|
|
if (!BN_mod_mul(&u2, sig->r, &u2, dsa->q, ctx)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!BN_MONT_CTX_set_locked((BN_MONT_CTX **)&dsa->method_mont_p,
|
|
|
|
(CRYPTO_MUTEX *)&dsa->method_mont_lock, dsa->p,
|
|
|
|
ctx)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!BN_mod_exp2_mont(&t1, dsa->g, &u1, dsa->pub_key, &u2, dsa->p, ctx,
|
|
|
|
dsa->method_mont_p)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// BN_copy(&u1,&t1);
|
|
|
|
// let u1 = u1 mod q
|
|
|
|
if (!BN_mod(&u1, &t1, dsa->q, ctx)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// V is now in u1. If the signature is correct, it will be
|
|
|
|
// equal to R.
|
|
|
|
*out_valid = BN_ucmp(&u1, sig->r) == 0;
|
|
|
|
ret = 1;
|
|
|
|
|
|
|
|
err:
|
|
|
|
if (ret != 1) {
|
|
|
|
OPENSSL_PUT_ERROR(DSA, ERR_R_BN_LIB);
|
|
|
|
}
|
|
|
|
BN_CTX_free(ctx);
|
|
|
|
BN_free(&u1);
|
|
|
|
BN_free(&u2);
|
|
|
|
BN_free(&t1);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_sign(int type, const uint8_t *digest, size_t digest_len,
|
|
|
|
uint8_t *out_sig, unsigned int *out_siglen, const DSA *dsa) {
|
|
|
|
DSA_SIG *s;
|
|
|
|
|
|
|
|
s = DSA_do_sign(digest, digest_len, dsa);
|
|
|
|
if (s == NULL) {
|
|
|
|
*out_siglen = 0;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
*out_siglen = i2d_DSA_SIG(s, &out_sig);
|
|
|
|
DSA_SIG_free(s);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_verify(int type, const uint8_t *digest, size_t digest_len,
|
|
|
|
const uint8_t *sig, size_t sig_len, const DSA *dsa) {
|
|
|
|
int valid;
|
|
|
|
if (!DSA_check_signature(&valid, digest, digest_len, sig, sig_len, dsa)) {
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return valid;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_check_signature(int *out_valid, const uint8_t *digest,
|
|
|
|
size_t digest_len, const uint8_t *sig, size_t sig_len,
|
|
|
|
const DSA *dsa) {
|
|
|
|
DSA_SIG *s = NULL;
|
|
|
|
int ret = 0;
|
|
|
|
uint8_t *der = NULL;
|
|
|
|
|
|
|
|
s = DSA_SIG_new();
|
|
|
|
if (s == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
const uint8_t *sigp = sig;
|
|
|
|
if (d2i_DSA_SIG(&s, &sigp, sig_len) == NULL || sigp != sig + sig_len) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Ensure that the signature uses DER and doesn't have trailing garbage.
|
|
|
|
int der_len = i2d_DSA_SIG(s, &der);
|
|
|
|
if (der_len < 0 || (size_t)der_len != sig_len ||
|
|
|
|
OPENSSL_memcmp(sig, der, sig_len)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
ret = DSA_do_check_signature(out_valid, digest, digest_len, s, dsa);
|
|
|
|
|
|
|
|
err:
|
|
|
|
OPENSSL_free(der);
|
|
|
|
DSA_SIG_free(s);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
// der_len_len returns the number of bytes needed to represent a length of |len|
|
|
|
|
// in DER.
|
|
|
|
static size_t der_len_len(size_t len) {
|
|
|
|
if (len < 0x80) {
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
size_t ret = 1;
|
|
|
|
while (len > 0) {
|
|
|
|
ret++;
|
|
|
|
len >>= 8;
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_size(const DSA *dsa) {
|
|
|
|
size_t order_len = BN_num_bytes(dsa->q);
|
|
|
|
// Compute the maximum length of an |order_len| byte integer. Defensively
|
|
|
|
// assume that the leading 0x00 is included.
|
|
|
|
size_t integer_len = 1 /* tag */ + der_len_len(order_len + 1) + 1 + order_len;
|
|
|
|
if (integer_len < order_len) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
// A DSA signature is two INTEGERs.
|
|
|
|
size_t value_len = 2 * integer_len;
|
|
|
|
if (value_len < integer_len) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
// Add the header.
|
|
|
|
size_t ret = 1 /* tag */ + der_len_len(value_len) + value_len;
|
|
|
|
if (ret < value_len) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int dsa_sign_setup(const DSA *dsa, BN_CTX *ctx, BIGNUM **out_kinv,
|
|
|
|
BIGNUM **out_r) {
|
|
|
|
if (!dsa->p || !dsa->q || !dsa->g) {
|
|
|
|
OPENSSL_PUT_ERROR(DSA, DSA_R_MISSING_PARAMETERS);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
int ret = 0;
|
|
|
|
BIGNUM k;
|
|
|
|
BN_init(&k);
|
|
|
|
BIGNUM *r = BN_new();
|
|
|
|
BIGNUM *kinv = BN_new();
|
|
|
|
if (r == NULL || kinv == NULL ||
|
|
|
|
// Get random k
|
|
|
|
!BN_rand_range_ex(&k, 1, dsa->q) ||
|
|
|
|
!BN_MONT_CTX_set_locked((BN_MONT_CTX **)&dsa->method_mont_p,
|
|
|
|
(CRYPTO_MUTEX *)&dsa->method_mont_lock, dsa->p,
|
|
|
|
ctx) ||
|
|
|
|
!BN_MONT_CTX_set_locked((BN_MONT_CTX **)&dsa->method_mont_q,
|
|
|
|
(CRYPTO_MUTEX *)&dsa->method_mont_lock, dsa->q,
|
|
|
|
ctx) ||
|
|
|
|
// Compute r = (g^k mod p) mod q
|
|
|
|
!BN_mod_exp_mont_consttime(r, dsa->g, &k, dsa->p, ctx,
|
|
|
|
dsa->method_mont_p) ||
|
|
|
|
// Note |BN_mod| below is not constant-time and may leak information about
|
|
|
|
// |r|. |dsa->p| may be significantly larger than |dsa->q|, so this is not
|
|
|
|
// easily performed in constant-time with Montgomery reduction.
|
|
|
|
//
|
|
|
|
// However, |r| at this point is g^k (mod p). It is almost the value of
|
|
|
|
// |r| revealed in the signature anyway (g^k (mod p) (mod q)), going from
|
|
|
|
// it to |k| would require computing a discrete log.
|
|
|
|
!BN_mod(r, r, dsa->q, ctx) ||
|
|
|
|
// Compute part of 's = inv(k) (m + xr) mod q' using Fermat's Little
|
|
|
|
// Theorem.
|
|
|
|
!bn_mod_inverse_prime(kinv, &k, dsa->q, ctx, dsa->method_mont_q)) {
|
|
|
|
OPENSSL_PUT_ERROR(DSA, ERR_R_BN_LIB);
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
BN_clear_free(*out_kinv);
|
|
|
|
*out_kinv = kinv;
|
|
|
|
kinv = NULL;
|
|
|
|
|
|
|
|
BN_clear_free(*out_r);
|
|
|
|
*out_r = r;
|
|
|
|
r = NULL;
|
|
|
|
|
|
|
|
ret = 1;
|
|
|
|
|
|
|
|
err:
|
|
|
|
BN_clear_free(&k);
|
|
|
|
BN_clear_free(r);
|
|
|
|
BN_clear_free(kinv);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_get_ex_new_index(long argl, void *argp, CRYPTO_EX_unused *unused,
|
|
|
|
CRYPTO_EX_dup *dup_unused, CRYPTO_EX_free *free_func) {
|
|
|
|
int index;
|
|
|
|
if (!CRYPTO_get_ex_new_index(&g_ex_data_class, &index, argl, argp,
|
|
|
|
free_func)) {
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return index;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DSA_set_ex_data(DSA *dsa, int idx, void *arg) {
|
|
|
|
return CRYPTO_set_ex_data(&dsa->ex_data, idx, arg);
|
|
|
|
}
|
|
|
|
|
|
|
|
void *DSA_get_ex_data(const DSA *dsa, int idx) {
|
|
|
|
return CRYPTO_get_ex_data(&dsa->ex_data, idx);
|
|
|
|
}
|
|
|
|
|
|
|
|
DH *DSA_dup_DH(const DSA *dsa) {
|
|
|
|
if (dsa == NULL) {
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
DH *ret = DH_new();
|
|
|
|
if (ret == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
if (dsa->q != NULL) {
|
|
|
|
ret->priv_length = BN_num_bits(dsa->q);
|
|
|
|
if ((ret->q = BN_dup(dsa->q)) == NULL) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if ((dsa->p != NULL && (ret->p = BN_dup(dsa->p)) == NULL) ||
|
|
|
|
(dsa->g != NULL && (ret->g = BN_dup(dsa->g)) == NULL) ||
|
|
|
|
(dsa->pub_key != NULL && (ret->pub_key = BN_dup(dsa->pub_key)) == NULL) ||
|
|
|
|
(dsa->priv_key != NULL &&
|
|
|
|
(ret->priv_key = BN_dup(dsa->priv_key)) == NULL)) {
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
err:
|
|
|
|
DH_free(ret);
|
|
|
|
return NULL;
|
|
|
|
}
|