boringssl/crypto/fipsmodule/ec/simple.c

575 lines
19 KiB
C
Raw Normal View History

/* Originally written by Bodo Moeller for the OpenSSL project.
* ====================================================================
* Copyright (c) 1998-2005 The OpenSSL Project. All rights reserved.
*
* 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 above 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 acknowledgment:
* "This product includes software developed by the OpenSSL Project
* for use in the OpenSSL Toolkit. (http://www.openssl.org/)"
*
* 4. The names "OpenSSL Toolkit" and "OpenSSL Project" must not be used to
* endorse or promote products derived from this software without
* prior written permission. For written permission, please contact
* openssl-core@openssl.org.
*
* 5. Products derived from this software may not be called "OpenSSL"
* nor may "OpenSSL" appear in their names without prior written
* permission of the OpenSSL Project.
*
* 6. Redistributions of any form whatsoever must retain the following
* acknowledgment:
* "This product includes software developed by the OpenSSL Project
* for use in the OpenSSL Toolkit (http://www.openssl.org/)"
*
* THIS SOFTWARE IS PROVIDED BY THE OpenSSL PROJECT ``AS IS'' AND ANY
* EXPRESSED 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 OpenSSL PROJECT OR
* ITS 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.
* ====================================================================
*
* This product includes cryptographic software written by Eric Young
* (eay@cryptsoft.com). This product includes software written by Tim
* Hudson (tjh@cryptsoft.com).
*
*/
/* ====================================================================
* Copyright 2002 Sun Microsystems, Inc. ALL RIGHTS RESERVED.
*
* Portions of the attached software ("Contribution") are developed by
* SUN MICROSYSTEMS, INC., and are contributed to the OpenSSL project.
*
* The Contribution is licensed pursuant to the OpenSSL open source
* license provided above.
*
* The elliptic curve binary polynomial software is originally written by
* Sheueling Chang Shantz and Douglas Stebila of Sun Microsystems
* Laboratories. */
#include <openssl/ec.h>
#include <string.h>
#include <openssl/bn.h>
#include <openssl/err.h>
#include <openssl/mem.h>
#include "internal.h"
#include "../../internal.h"
// Most method functions in this file are designed to work with non-trivial
// representations of field elements if necessary (see ecp_mont.c): while
// standard modular addition and subtraction are used, the field_mul and
// field_sqr methods will be used for multiplication, and field_encode and
// field_decode (if defined) will be used for converting between
// representations.
//
// Functions here specifically assume that if a non-trivial representation is
// used, it is a Montgomery representation (i.e. 'encoding' means multiplying
// by some factor R).
int ec_GFp_simple_group_init(EC_GROUP *group) {
BN_init(&group->field);
group->a_is_minus3 = 0;
return 1;
}
void ec_GFp_simple_group_finish(EC_GROUP *group) {
BN_free(&group->field);
}
int ec_GFp_simple_group_set_curve(EC_GROUP *group, const BIGNUM *p,
const BIGNUM *a, const BIGNUM *b,
BN_CTX *ctx) {
int ret = 0;
BN_CTX *new_ctx = NULL;
// p must be a prime > 3
if (BN_num_bits(p) <= 2 || !BN_is_odd(p)) {
OPENSSL_PUT_ERROR(EC, EC_R_INVALID_FIELD);
return 0;
}
if (ctx == NULL) {
ctx = new_ctx = BN_CTX_new();
if (ctx == NULL) {
return 0;
}
}
BN_CTX_start(ctx);
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
BIGNUM *tmp = BN_CTX_get(ctx);
if (tmp == NULL) {
goto err;
}
// group->field
if (!BN_copy(&group->field, p)) {
goto err;
}
BN_set_negative(&group->field, 0);
// Store the field in minimal form, so it can be used with |BN_ULONG| arrays.
bn_set_minimal_width(&group->field);
// group->a
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
if (!BN_nnmod(tmp, a, &group->field, ctx) ||
!ec_bignum_to_felem(group, &group->a, tmp)) {
goto err;
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
// group->a_is_minus3
if (!BN_add_word(tmp, 3)) {
goto err;
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
group->a_is_minus3 = (0 == BN_cmp(tmp, &group->field));
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
// group->b
if (!BN_nnmod(tmp, b, &group->field, ctx) ||
!ec_bignum_to_felem(group, &group->b, tmp)) {
goto err;
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
if (!ec_bignum_to_felem(group, &group->one, BN_value_one())) {
goto err;
}
ret = 1;
err:
BN_CTX_end(ctx);
BN_CTX_free(new_ctx);
return ret;
}
int ec_GFp_simple_group_get_curve(const EC_GROUP *group, BIGNUM *p, BIGNUM *a,
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
BIGNUM *b) {
if ((p != NULL && !BN_copy(p, &group->field)) ||
(a != NULL && !ec_felem_to_bignum(group, a, &group->a)) ||
(b != NULL && !ec_felem_to_bignum(group, b, &group->b))) {
return 0;
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
return 1;
}
unsigned ec_GFp_simple_group_get_degree(const EC_GROUP *group) {
return BN_num_bits(&group->field);
}
void ec_GFp_simple_point_init(EC_RAW_POINT *point) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
OPENSSL_memset(&point->X, 0, sizeof(EC_FELEM));
OPENSSL_memset(&point->Y, 0, sizeof(EC_FELEM));
OPENSSL_memset(&point->Z, 0, sizeof(EC_FELEM));
}
void ec_GFp_simple_point_copy(EC_RAW_POINT *dest, const EC_RAW_POINT *src) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
OPENSSL_memcpy(&dest->X, &src->X, sizeof(EC_FELEM));
OPENSSL_memcpy(&dest->Y, &src->Y, sizeof(EC_FELEM));
OPENSSL_memcpy(&dest->Z, &src->Z, sizeof(EC_FELEM));
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
void ec_GFp_simple_point_set_to_infinity(const EC_GROUP *group,
EC_RAW_POINT *point) {
// Although it is strictly only necessary to zero Z, we zero the entire point
// in case |point| was stack-allocated and yet to be initialized.
ec_GFp_simple_point_init(point);
}
int ec_GFp_simple_point_set_affine_coordinates(const EC_GROUP *group,
EC_RAW_POINT *point,
const BIGNUM *x,
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
const BIGNUM *y) {
if (x == NULL || y == NULL) {
OPENSSL_PUT_ERROR(EC, ERR_R_PASSED_NULL_PARAMETER);
return 0;
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
if (!ec_bignum_to_felem(group, &point->X, x) ||
!ec_bignum_to_felem(group, &point->Y, y)) {
return 0;
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
OPENSSL_memcpy(&point->Z, &group->one, sizeof(EC_FELEM));
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
return 1;
}
void ec_GFp_simple_add(const EC_GROUP *group, EC_RAW_POINT *out,
const EC_RAW_POINT *a, const EC_RAW_POINT *b) {
if (a == b) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
ec_GFp_simple_dbl(group, out, a);
return;
}
// The method is taken from:
// http://hyperelliptic.org/EFD/g1p/auto-shortw-jacobian.html#addition-add-2007-bl
//
// Coq transcription and correctness proof:
// <https://github.com/davidben/fiat-crypto/blob/c7b95f62b2a54b559522573310e9b487327d219a/src/Curves/Weierstrass/Jacobian.v#L467>
// <https://github.com/davidben/fiat-crypto/blob/c7b95f62b2a54b559522573310e9b487327d219a/src/Curves/Weierstrass/Jacobian.v#L544>
void (*const felem_mul)(const EC_GROUP *, EC_FELEM *r, const EC_FELEM *a,
const EC_FELEM *b) = group->meth->felem_mul;
void (*const felem_sqr)(const EC_GROUP *, EC_FELEM *r, const EC_FELEM *a) =
group->meth->felem_sqr;
EC_FELEM x_out, y_out, z_out;
BN_ULONG z1nz = ec_felem_non_zero_mask(group, &a->Z);
BN_ULONG z2nz = ec_felem_non_zero_mask(group, &b->Z);
// z1z1 = z1z1 = z1**2
EC_FELEM z1z1;
felem_sqr(group, &z1z1, &a->Z);
// z2z2 = z2**2
EC_FELEM z2z2;
felem_sqr(group, &z2z2, &b->Z);
// u1 = x1*z2z2
EC_FELEM u1;
felem_mul(group, &u1, &a->X, &z2z2);
// two_z1z2 = (z1 + z2)**2 - (z1z1 + z2z2) = 2z1z2
EC_FELEM two_z1z2;
ec_felem_add(group, &two_z1z2, &a->Z, &b->Z);
felem_sqr(group, &two_z1z2, &two_z1z2);
ec_felem_sub(group, &two_z1z2, &two_z1z2, &z1z1);
ec_felem_sub(group, &two_z1z2, &two_z1z2, &z2z2);
// s1 = y1 * z2**3
EC_FELEM s1;
felem_mul(group, &s1, &b->Z, &z2z2);
felem_mul(group, &s1, &s1, &a->Y);
// u2 = x2*z1z1
EC_FELEM u2;
felem_mul(group, &u2, &b->X, &z1z1);
// h = u2 - u1
EC_FELEM h;
ec_felem_sub(group, &h, &u2, &u1);
BN_ULONG xneq = ec_felem_non_zero_mask(group, &h);
// z_out = two_z1z2 * h
felem_mul(group, &z_out, &h, &two_z1z2);
// z1z1z1 = z1 * z1z1
EC_FELEM z1z1z1;
felem_mul(group, &z1z1z1, &a->Z, &z1z1);
// s2 = y2 * z1**3
EC_FELEM s2;
felem_mul(group, &s2, &b->Y, &z1z1z1);
// r = (s2 - s1)*2
EC_FELEM r;
ec_felem_sub(group, &r, &s2, &s1);
ec_felem_add(group, &r, &r, &r);
BN_ULONG yneq = ec_felem_non_zero_mask(group, &r);
// TODO(davidben): Analyze how case relates to timing considerations for the
// supported curves which hit it (P-224, P-384, and P-521) and the
// to-be-written constant-time generic multiplication implementation.
if (!xneq && !yneq && z1nz && z2nz) {
ec_GFp_simple_dbl(group, out, a);
return;
}
// I = (2h)**2
EC_FELEM i;
ec_felem_add(group, &i, &h, &h);
felem_sqr(group, &i, &i);
// J = h * I
EC_FELEM j;
felem_mul(group, &j, &h, &i);
// V = U1 * I
EC_FELEM v;
felem_mul(group, &v, &u1, &i);
// x_out = r**2 - J - 2V
felem_sqr(group, &x_out, &r);
ec_felem_sub(group, &x_out, &x_out, &j);
ec_felem_sub(group, &x_out, &x_out, &v);
ec_felem_sub(group, &x_out, &x_out, &v);
// y_out = r(V-x_out) - 2 * s1 * J
ec_felem_sub(group, &y_out, &v, &x_out);
felem_mul(group, &y_out, &y_out, &r);
EC_FELEM s1j;
felem_mul(group, &s1j, &s1, &j);
ec_felem_sub(group, &y_out, &y_out, &s1j);
ec_felem_sub(group, &y_out, &y_out, &s1j);
ec_felem_select(group, &x_out, z1nz, &x_out, &b->X);
ec_felem_select(group, &out->X, z2nz, &x_out, &a->X);
ec_felem_select(group, &y_out, z1nz, &y_out, &b->Y);
ec_felem_select(group, &out->Y, z2nz, &y_out, &a->Y);
ec_felem_select(group, &z_out, z1nz, &z_out, &b->Z);
ec_felem_select(group, &out->Z, z2nz, &z_out, &a->Z);
}
void ec_GFp_simple_dbl(const EC_GROUP *group, EC_RAW_POINT *r,
const EC_RAW_POINT *a) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
void (*const felem_mul)(const EC_GROUP *, EC_FELEM *r, const EC_FELEM *a,
const EC_FELEM *b) = group->meth->felem_mul;
void (*const felem_sqr)(const EC_GROUP *, EC_FELEM *r, const EC_FELEM *a) =
group->meth->felem_sqr;
Make generic point arithmetic slightly less variable-time. The generic code special-cases affine points, but this leaks information. (Of course, the generic code also doesn't have a constant-time multiply and other problems, but one thing at a time.) The optimization in point doubling is not useful. Point multiplication more-or-less never doubles an affine point. The optimization in point addition *is* useful because the wNAF code converts the tables to affine. Accordingly, align with the P-256 code which adds a 'mixed' parameter. (I haven't aligned the formally-verified point formulas themselves yet; initial testing suggests that the large number of temporaries take a perf hit with BIGNUM. I'll check the results in EC_FELEM, which will be stack-allocated, to see if we still need to help the compiler out.) Strangly, it actually got a bit faster with this change. I'm guessing because now it doesn't need to bother with unnecessary comparisons and maybe was kinder to the branch predictor? Before: Did 2201 ECDH P-384 operations in 3068341us (717.3 ops/sec) Did 4092 ECDSA P-384 signing operations in 3076981us (1329.9 ops/sec) Did 3503 ECDSA P-384 verify operations in 3024753us (1158.1 ops/sec) Did 992 ECDH P-521 operations in 3017884us (328.7 ops/sec) Did 1798 ECDSA P-521 signing operations in 3059000us (587.8 ops/sec) Did 1581 ECDSA P-521 verify operations in 3033142us (521.2 ops/sec) After: Did 2310 ECDH P-384 operations in 3092648us (746.9 ops/sec) Did 4080 ECDSA P-384 signing operations in 3044588us (1340.1 ops/sec) Did 3520 ECDSA P-384 verify operations in 3056070us (1151.8 ops/sec) Did 992 ECDH P-521 operations in 3012779us (329.3 ops/sec) Did 1792 ECDSA P-521 signing operations in 3019459us (593.5 ops/sec) Did 1600 ECDSA P-521 verify operations in 3047749us (525.0 ops/sec) Bug: 239 Change-Id: If5d13825fc98e4c58bdd1580cf0245bf7ce93a82 Reviewed-on: https://boringssl-review.googlesource.com/27004 Reviewed-by: Adam Langley <agl@google.com> Commit-Queue: David Benjamin <davidben@google.com> CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
2018-04-01 19:31:44 +01:00
if (group->a_is_minus3) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
// The method is taken from:
// http://hyperelliptic.org/EFD/g1p/auto-shortw-jacobian-3.html#doubling-dbl-2001-b
//
// Coq transcription and correctness proof:
// <https://github.com/mit-plv/fiat-crypto/blob/79f8b5f39ed609339f0233098dee1a3c4e6b3080/src/Curves/Weierstrass/Jacobian.v#L93>
// <https://github.com/mit-plv/fiat-crypto/blob/79f8b5f39ed609339f0233098dee1a3c4e6b3080/src/Curves/Weierstrass/Jacobian.v#L201>
EC_FELEM delta, gamma, beta, ftmp, ftmp2, tmptmp, alpha, fourbeta;
// delta = z^2
felem_sqr(group, &delta, &a->Z);
// gamma = y^2
felem_sqr(group, &gamma, &a->Y);
// beta = x*gamma
felem_mul(group, &beta, &a->X, &gamma);
// alpha = 3*(x-delta)*(x+delta)
ec_felem_sub(group, &ftmp, &a->X, &delta);
ec_felem_add(group, &ftmp2, &a->X, &delta);
ec_felem_add(group, &tmptmp, &ftmp2, &ftmp2);
ec_felem_add(group, &ftmp2, &ftmp2, &tmptmp);
felem_mul(group, &alpha, &ftmp, &ftmp2);
// x' = alpha^2 - 8*beta
felem_sqr(group, &r->X, &alpha);
ec_felem_add(group, &fourbeta, &beta, &beta);
ec_felem_add(group, &fourbeta, &fourbeta, &fourbeta);
ec_felem_add(group, &tmptmp, &fourbeta, &fourbeta);
ec_felem_sub(group, &r->X, &r->X, &tmptmp);
// z' = (y + z)^2 - gamma - delta
ec_felem_add(group, &delta, &gamma, &delta);
ec_felem_add(group, &ftmp, &a->Y, &a->Z);
felem_sqr(group, &r->Z, &ftmp);
ec_felem_sub(group, &r->Z, &r->Z, &delta);
// y' = alpha*(4*beta - x') - 8*gamma^2
ec_felem_sub(group, &r->Y, &fourbeta, &r->X);
ec_felem_add(group, &gamma, &gamma, &gamma);
felem_sqr(group, &gamma, &gamma);
felem_mul(group, &r->Y, &alpha, &r->Y);
ec_felem_add(group, &gamma, &gamma, &gamma);
ec_felem_sub(group, &r->Y, &r->Y, &gamma);
} else {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
// The method is taken from:
// http://www.hyperelliptic.org/EFD/g1p/auto-shortw-jacobian.html#doubling-dbl-2007-bl
//
// Coq transcription and correctness proof:
// <https://github.com/davidben/fiat-crypto/blob/c7b95f62b2a54b559522573310e9b487327d219a/src/Curves/Weierstrass/Jacobian.v#L102>
// <https://github.com/davidben/fiat-crypto/blob/c7b95f62b2a54b559522573310e9b487327d219a/src/Curves/Weierstrass/Jacobian.v#L534>
EC_FELEM xx, yy, yyyy, zz;
felem_sqr(group, &xx, &a->X);
felem_sqr(group, &yy, &a->Y);
felem_sqr(group, &yyyy, &yy);
felem_sqr(group, &zz, &a->Z);
// s = 2*((x_in + yy)^2 - xx - yyyy)
EC_FELEM s;
ec_felem_add(group, &s, &a->X, &yy);
felem_sqr(group, &s, &s);
ec_felem_sub(group, &s, &s, &xx);
ec_felem_sub(group, &s, &s, &yyyy);
ec_felem_add(group, &s, &s, &s);
// m = 3*xx + a*zz^2
EC_FELEM m;
felem_sqr(group, &m, &zz);
felem_mul(group, &m, &group->a, &m);
ec_felem_add(group, &m, &m, &xx);
ec_felem_add(group, &m, &m, &xx);
ec_felem_add(group, &m, &m, &xx);
// x_out = m^2 - 2*s
felem_sqr(group, &r->X, &m);
ec_felem_sub(group, &r->X, &r->X, &s);
ec_felem_sub(group, &r->X, &r->X, &s);
// z_out = (y_in + z_in)^2 - yy - zz
ec_felem_add(group, &r->Z, &a->Y, &a->Z);
felem_sqr(group, &r->Z, &r->Z);
ec_felem_sub(group, &r->Z, &r->Z, &yy);
ec_felem_sub(group, &r->Z, &r->Z, &zz);
// y_out = m*(s-x_out) - 8*yyyy
ec_felem_add(group, &yyyy, &yyyy, &yyyy);
ec_felem_add(group, &yyyy, &yyyy, &yyyy);
ec_felem_add(group, &yyyy, &yyyy, &yyyy);
ec_felem_sub(group, &r->Y, &s, &r->X);
felem_mul(group, &r->Y, &r->Y, &m);
ec_felem_sub(group, &r->Y, &r->Y, &yyyy);
}
}
void ec_GFp_simple_invert(const EC_GROUP *group, EC_RAW_POINT *point) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
ec_felem_neg(group, &point->Y, &point->Y);
}
int ec_GFp_simple_is_at_infinity(const EC_GROUP *group,
const EC_RAW_POINT *point) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
return ec_felem_non_zero_mask(group, &point->Z) == 0;
}
int ec_GFp_simple_is_on_curve(const EC_GROUP *group,
const EC_RAW_POINT *point) {
if (ec_GFp_simple_is_at_infinity(group, point)) {
return 1;
}
// We have a curve defined by a Weierstrass equation
// y^2 = x^3 + a*x + b.
// The point to consider is given in Jacobian projective coordinates
// where (X, Y, Z) represents (x, y) = (X/Z^2, Y/Z^3).
// Substituting this and multiplying by Z^6 transforms the above equation
// into
// Y^2 = X^3 + a*X*Z^4 + b*Z^6.
// To test this, we add up the right-hand side in 'rh'.
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
void (*const felem_mul)(const EC_GROUP *, EC_FELEM *r, const EC_FELEM *a,
const EC_FELEM *b) = group->meth->felem_mul;
void (*const felem_sqr)(const EC_GROUP *, EC_FELEM *r, const EC_FELEM *a) =
group->meth->felem_sqr;
// rh := X^2
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
EC_FELEM rh;
felem_sqr(group, &rh, &point->X);
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
EC_FELEM tmp, Z4, Z6;
if (!ec_felem_equal(group, &point->Z, &group->one)) {
felem_sqr(group, &tmp, &point->Z);
felem_sqr(group, &Z4, &tmp);
felem_mul(group, &Z6, &Z4, &tmp);
// rh := (rh + a*Z^4)*X
if (group->a_is_minus3) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
ec_felem_add(group, &tmp, &Z4, &Z4);
ec_felem_add(group, &tmp, &tmp, &Z4);
ec_felem_sub(group, &rh, &rh, &tmp);
felem_mul(group, &rh, &rh, &point->X);
} else {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
felem_mul(group, &tmp, &Z4, &group->a);
ec_felem_add(group, &rh, &rh, &tmp);
felem_mul(group, &rh, &rh, &point->X);
}
// rh := rh + b*Z^6
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
felem_mul(group, &tmp, &group->b, &Z6);
ec_felem_add(group, &rh, &rh, &tmp);
} else {
// rh := (rh + a)*X
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
ec_felem_add(group, &rh, &rh, &group->a);
felem_mul(group, &rh, &rh, &point->X);
// rh := rh + b
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
ec_felem_add(group, &rh, &rh, &group->b);
}
// 'lh' := Y^2
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
felem_sqr(group, &tmp, &point->Y);
return ec_felem_equal(group, &tmp, &rh);
}
int ec_GFp_simple_cmp(const EC_GROUP *group, const EC_RAW_POINT *a,
const EC_RAW_POINT *b) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
// Note this function returns zero if |a| and |b| are equal and 1 if they are
// not equal.
if (ec_GFp_simple_is_at_infinity(group, a)) {
return ec_GFp_simple_is_at_infinity(group, b) ? 0 : 1;
}
if (ec_GFp_simple_is_at_infinity(group, b)) {
return 1;
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
int a_Z_is_one = ec_felem_equal(group, &a->Z, &group->one);
int b_Z_is_one = ec_felem_equal(group, &b->Z, &group->one);
if (a_Z_is_one && b_Z_is_one) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
return !ec_felem_equal(group, &a->X, &b->X) ||
!ec_felem_equal(group, &a->Y, &b->Y);
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
void (*const felem_mul)(const EC_GROUP *, EC_FELEM *r, const EC_FELEM *a,
const EC_FELEM *b) = group->meth->felem_mul;
void (*const felem_sqr)(const EC_GROUP *, EC_FELEM *r, const EC_FELEM *a) =
group->meth->felem_sqr;
// We have to decide whether
// (X_a/Z_a^2, Y_a/Z_a^3) = (X_b/Z_b^2, Y_b/Z_b^3),
// or equivalently, whether
// (X_a*Z_b^2, Y_a*Z_b^3) = (X_b*Z_a^2, Y_b*Z_a^3).
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
EC_FELEM tmp1, tmp2, Za23, Zb23;
const EC_FELEM *tmp1_, *tmp2_;
if (!b_Z_is_one) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
felem_sqr(group, &Zb23, &b->Z);
felem_mul(group, &tmp1, &a->X, &Zb23);
tmp1_ = &tmp1;
} else {
tmp1_ = &a->X;
}
if (!a_Z_is_one) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
felem_sqr(group, &Za23, &a->Z);
felem_mul(group, &tmp2, &b->X, &Za23);
tmp2_ = &tmp2;
} else {
tmp2_ = &b->X;
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
// Compare X_a*Z_b^2 with X_b*Z_a^2.
if (!ec_felem_equal(group, tmp1_, tmp2_)) {
return 1; // The points differ.
}
if (!b_Z_is_one) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
felem_mul(group, &Zb23, &Zb23, &b->Z);
felem_mul(group, &tmp1, &a->Y, &Zb23);
// tmp1_ = &tmp1
} else {
tmp1_ = &a->Y;
}
if (!a_Z_is_one) {
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
felem_mul(group, &Za23, &Za23, &a->Z);
felem_mul(group, &tmp2, &b->Y, &Za23);
// tmp2_ = &tmp2
} else {
tmp2_ = &b->Y;
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
// Compare Y_a*Z_b^3 with Y_b*Z_a^3.
if (!ec_felem_equal(group, tmp1_, tmp2_)) {
return 1; // The points differ.
}
Add EC_FELEM for EC_POINTs and related temporaries. This introduces EC_FELEM, which is analogous to EC_SCALAR. It is used for EC_POINT's representation in the generic EC_METHOD, as well as random operations on tuned EC_METHODs that still are implemented genericly. Unlike EC_SCALAR, EC_FELEM's exact representation is awkwardly specific to the EC_METHOD, analogous to how the old values were BIGNUMs but may or may not have been in Montgomery form. This is kind of a nuisance, but no more than before. (If p224-64.c were easily convertable to Montgomery form, we could say |EC_FELEM| is always in Montgomery form. If we exposed the internal add and double implementations in each of the curves, we could give |EC_POINT| an |EC_METHOD|-specific representation and |EC_FELEM| is purely a |EC_GFp_mont_method| type. I'll leave this for later.) The generic add and doubling formulas are aligned with the formulas proved in fiat-crypto. Those only applied to a = -3, so I've proved a generic one in https://github.com/mit-plv/fiat-crypto/pull/356, in case someone uses a custom curve. The new formulas are verified, constant-time, and swap a multiply for a square. As expressed in fiat-crypto they do use more temporaries, but this seems to be fine with stack-allocated EC_FELEMs. (We can try to help the compiler later, but benchamrks below suggest this isn't necessary.) Unlike BIGNUM, EC_FELEM can be stack-allocated. It also captures the bounds in the type system and, in particular, that the width is correct, which will make it easier to select a point in constant-time in the future. (Indeed the old code did not always have the correct width. Its point formula involved halving and implemented this in variable time and variable width.) Before: Did 77274 ECDH P-256 operations in 10046087us (7692.0 ops/sec) Did 5959 ECDH P-384 operations in 10031701us (594.0 ops/sec) Did 10815 ECDSA P-384 signing operations in 10087892us (1072.1 ops/sec) Did 8976 ECDSA P-384 verify operations in 10071038us (891.3 ops/sec) Did 2600 ECDH P-521 operations in 10091688us (257.6 ops/sec) Did 4590 ECDSA P-521 signing operations in 10055195us (456.5 ops/sec) Did 3811 ECDSA P-521 verify operations in 10003574us (381.0 ops/sec) After: Did 77736 ECDH P-256 operations in 10029858us (7750.5 ops/sec) [+0.8%] Did 7519 ECDH P-384 operations in 10068076us (746.8 ops/sec) [+25.7%] Did 13335 ECDSA P-384 signing operations in 10029962us (1329.5 ops/sec) [+24.0%] Did 11021 ECDSA P-384 verify operations in 10088600us (1092.4 ops/sec) [+22.6%] Did 2912 ECDH P-521 operations in 10001325us (291.2 ops/sec) [+13.0%] Did 5150 ECDSA P-521 signing operations in 10027462us (513.6 ops/sec) [+12.5%] Did 4264 ECDSA P-521 verify operations in 10069694us (423.4 ops/sec) [+11.1%] This more than pays for removing points_make_affine previously and even speeds up ECDH P-256 slightly. (The point-on-curve check uses the generic code.) Next is to push the stack-allocating up to ec_wNAF_mul, followed by a constant-time single-point multiplication. Bug: 239 Change-Id: I44a2dff7c52522e491d0f8cffff64c4ab5cd353c Reviewed-on: https://boringssl-review.googlesource.com/27668 Reviewed-by: Adam Langley <agl@google.com>
2018-04-23 02:39:34 +01:00
// The points are equal.
return 0;
}