Commit graph

108 commits

Author SHA1 Message Date
Gabor Mezei
ac70ad6576
Fix coding style
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-16 19:31:21 +01:00
Gabor Mezei
cf228706cd
Restrict input parameter size for ecp_mod_p521_raw
The imput mpi parameter must have twice as many limbs as the modulus.

Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-15 18:19:08 +01:00
Gabor Mezei
d10d429380
Stack usage optimization for mod_p521
Instead of creating an mpi on the stack, reuse the unused part of the input mpi.

Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-15 18:13:51 +01:00
Janos Follath
fe24e91a34
mod_p521: document reduction algorithm
Signed-off-by: Janos Follath <janos.follath@arm.com>
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-15 18:13:50 +01:00
Janos Follath
666673e83f
modp521: apply naming conventions
Apply the usual parameter name and align the local variables and
comments. This naming diverges from the standard notation, but this is
beneficial as our variable meanings diverge as well and the difference
can help avoiding confusion.

Signed-off-by: Janos Follath <janos.follath@arm.com>
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-15 18:13:50 +01:00
Janos Follath
13c3aa13af
Revert changes to mod_p521 flow
It is not necessary to save the middle limb upfront as overwriting it is
the desired result: in the first step we are reducing modulo
2^{512+biL}.

Arguably, the original flow is more intuitive and easier to see the idea
behind it.

Signed-off-by: Janos Follath <janos.follath@arm.com>
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-15 18:13:50 +01:00
Gabor Mezei
6bfbd36507
Fix coding style issues
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-15 18:13:49 +01:00
Gabor Mezei
b62ad5d569
Rename function to follow naming convention
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-15 18:13:48 +01:00
Gabor Mezei
2cb630edee
Change the ecp_mod_p521_raw to be testable
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-15 18:05:22 +01:00
Gabor Mezei
8450ab9c60
Fix Secp521r1 reduction
The prototype calculated with wrong limb size and not taken into account
the overflow in the shared limb.

Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-15 18:03:03 +01:00
Gabor Mezei
42df16c84b
Extract Secp521r1 from the prototype
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-15 18:02:57 +01:00
Gilles Peskine
e2a9f86755
Merge pull request #6971 from gabor-mezei-arm/6026_Secp192r1_fast_reduction
Extract Secp192r1 fast reduction from the prototype
2023-02-15 16:22:36 +01:00
Gabor Mezei
a264831cff
Update documentation and add comments
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-02-13 16:29:05 +01:00
Dave Rodgman
6dd757a8ba Fix use of sizeof without brackets
Signed-off-by: Dave Rodgman <dave.rodgman@arm.com>
2023-02-02 12:40:50 +00:00
Gabor Mezei
2038ce976e
Rename function to follow naming convention
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-01-31 14:40:05 +01:00
Gabor Mezei
deece2bb65
Change the ecp_mod_p192_raw to be testable
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-01-31 14:40:05 +01:00
Gabor Mezei
b5bba497fe
Extract Secp192r1 from the prototype
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
2023-01-31 14:40:05 +01:00
Minos Galanakis
e9fa7a74cd ecp_curves: Update pre-processor define guards for ecp_mpi_load().
This patch adjusts the logic, so that the method is included,
when the following components are enabled:

* MBEDTLS_ECP_DP_CURVE448_ENABLED
* MBEDTLS_ECP_DP_CURVE25519_ENABLED
* ECP_LOAD_GROUP

Signed-off-by: Minos Galanakis <minos.galanakis@arm.com>
2023-01-19 16:12:07 +00:00
Minos Galanakis
146fed9849 ecp_curves: Hardcode Montgomery const for curve448.
This patch adds two embedded constants used by `ecp_use_curve448()`.
The method has been updated to read that into an mpi instead of
calculating it on the fly.

Signed-off-by: Minos Galanakis <minos.galanakis@arm.com>
2023-01-19 16:11:50 +00:00
Minos Galanakis
bececeb0b9 ecp_curves: Hardcod Montgomery const for curve25519
This patch adds two embedded constants used by `ecp_use_curve25519()`.
The method has been updated to read that into an mpi instead of
calculating it on the fly.

Signed-off-by: Minos Galanakis <minos.galanakis@arm.com>
2023-01-19 11:38:19 +00:00
Gilles Peskine
449bd8303e Switch to the new code style
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2023-01-11 14:50:10 +01:00
Gilles Peskine
5a34b36bbd Remove more now-redundant definitions of inline
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2022-11-25 13:26:44 +01:00
Tom Cosgrove
82d3f1e824 Remove bignum_internal.h, moving contents to bignum_core.h
Signed-off-by: Tom Cosgrove <tom.cosgrove@arm.com>
2022-08-23 12:01:39 +01:00
Pol Henarejos
b101cb6111
Since the group is unloaded for all curves, it is better to initialize the group also for all curves.
Signed-off-by: Pol Henarejos <pol.henarejos@cttc.es>
2022-05-06 18:43:58 +02:00
Pol Henarejos
aa68d36234
Fix order value for curve x448.
Signed-off-by: Pol Henarejos <pol.henarejos@cttc.es>
2022-05-05 19:22:29 +02:00
Hanno Becker
0dbf04a9a6 Remove unnecessary memory operations in p25519 quasireduction
Signed-off-by: Hanno Becker <hanno.becker@arm.com>
2022-04-13 07:26:53 +01:00
Hanno Becker
127fcabb21 Fail gracefully upon unexpectedly large input to p25519 reduction
Signed-off-by: Hanno Becker <hanno.becker@arm.com>
2022-04-12 22:18:36 +01:00
Hanno Becker
bb04cb992f Fix check in p25519 quasi-reduction
Signed-off-by: Hanno Becker <hanno.becker@arm.com>
2022-04-12 11:18:11 +01:00
Hanno Becker
d830feb256 Simplify check in p25519 quasi-reduction
Signed-off-by: Hanno Becker <hanno.becker@arm.com>
2022-04-12 11:10:19 +01:00
Hanno Becker
2ef0cff6c3 Fix size check in p25519 modular reduction
The check was meant to precisely catch an underflow.

Signed-off-by: Hanno Becker <hanno.becker@arm.com>
2022-04-12 11:02:05 +01:00
Hanno Becker
53b3c607a0 Move const keyword prior to type name
Signed-off-by: Hanno Becker <hanno.becker@arm.com>
2022-04-11 13:46:30 +01:00
Hanno Becker
e9dd9a1f31 Use size_t for number of limbs
Signed-off-by: Hanno Becker <hanno.becker@arm.com>
2022-04-11 09:06:27 +01:00
Hanno Becker
6454993e2e Safeguard against calling p255 reduction with single-width MPI
(In this case, there's nothing to do anyway since we only do a
quasi-reduction to N+1 limbs)

Signed-off-by: Hanno Becker <hanno.becker@arm.com>
2022-04-11 07:35:58 +01:00
Hanno Becker
25bb732ea7 Simplify x25519 reduction using internal bignum MLA helper
Signed-off-by: Hanno Becker <hanno.becker@arm.com>
2022-04-11 07:03:48 +01:00
Janos Follath
1107ee4e44 Add prefix to BYTES_TO_T_UINT_*
These macros were moved into a header and now check-names.sh is failing.
Add an MBEDTL_ prefix to the macro names to make it pass.

Signed-off-by: Janos Follath <janos.follath@arm.com>
2021-06-25 12:46:40 +01:00
Janos Follath
8b8b781524 Use mbedtls_mpi_lset() more
Signed-off-by: Janos Follath <janos.follath@arm.com>
2021-06-24 15:00:33 +01:00
Janos Follath
8c70e815dd Move mpi constant macros to bn_mul.h
Signed-off-by: Janos Follath <janos.follath@arm.com>
2021-06-24 14:48:38 +01:00
Manuel Pégourié-Gonnard
06215eaa3e Avoid complaints about undeclared non-static symbols
Clang was complaining and check-names.sh too

This only duplicates macros, so no impact on code size. In 3.0 we can
probably avoid the duplication by using an internal header under
library/ but this won't work for 2.16.

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2021-06-23 12:59:02 +02:00
Manuel Pégourié-Gonnard
2d457b8fca Use more compact encoding of Montgomery curve constants
Base 256 beats base 16.

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2021-06-23 12:44:21 +02:00
Manuel Pégourié-Gonnard
2389a6000e Use a more compact encoding of bad points
Base 10 is horrible, base 256 is much better.

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2021-06-23 12:25:48 +02:00
kXuan
ba9cb76e9f
static initialize comb table
MBEDTLS_ECP_FIXED_POINT_OPTIM aims to speed up ecc multiplication performance.

We compute the comb table in runtime now. It is a costly operation.

This patch add a pre-computed table to initialize well-known curves. It speed up ECDSA signature verify process in runtime by using more ROM size.

Signed-off-by: kXuan <kxuanobj@gmail.com>
2021-06-01 10:02:13 +08:00
Gilles Peskine
ff6a32d79c Fix low-probability arithmetic error in ECC
Fix the subtraction in fix_negative, which was incorrectly not looking
for a carry. This caused the result to be wrong when the least
significant limb of N was 0. Fix #4296.

The bug was introduced by d10e8fae9e
"Optimize fix_negative". Thanks to Philippe Antoine (catenacyber) for
reporting the bug which was found by his EC differential fuzzer.
Credit to OSS-Fuzz.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2021-04-04 01:10:41 +02:00
Gilles Peskine
349b37273e Fix an incorrect comment about fix_negative
We're subtracting multiples of 2^bits, not 2^(bits+32).

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2021-04-04 01:10:41 +02:00
Gilles Peskine
618be2ec41 Add unit tests for fix_negative
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2021-04-04 01:10:41 +02:00
Gilles Peskine
b76517b764 Cosmetic improvement
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2021-03-10 23:44:28 +01:00
Gilles Peskine
d10e8fae9e Optimize fix_negative
Reduce the code size, stack consumption and heap consumption in
fix_negative by encoding the special-case subtraction manually.

* Code size: ecp_curves.o goes down from 7837B down to 7769 in a
  sample Cortex-M0 build with all curves enabled. The savings come
  from not having to set up C in INIT (which is used many times) and
  from not having to catch errors in fix_negative.
* Stack consumption: get rid of C on the stack.
* Heap: mbedtls_mpi_sub_abs with destination == second operand would
  make a heap allocation. The new code doesn't do any heap allocation.
* Performance: no measurable difference.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-12-07 22:51:25 +01:00
Gilles Peskine
e54be3a9c3 Fix uncaught error if fix_negative fails
fix_negative allocates memory for its result. The calling site didn't
check the return value, so an out-of-memory error could lead to an
incorrect calculation. Fix this.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-12-06 22:32:02 +01:00
Bence Szépkúti
1e14827beb Update copyright notices to use Linux Foundation guidance
As a result, the copyright of contributors other than Arm is now
acknowledged, and the years of publishing are no longer tracked in the
source files.

Also remove the now-redundant lines declaring that the files are part of
MbedTLS.

This commit was generated using the following script:

# ========================
#!/bin/sh

# Find files
find '(' -path './.git' -o -path './3rdparty' ')' -prune -o -type f -print | xargs sed -bi '

# Replace copyright attribution line
s/Copyright.*Arm.*/Copyright The Mbed TLS Contributors/I

# Remove redundant declaration and the preceding line
$!N
/This file is part of Mbed TLS/Id
P
D
'
# ========================

Signed-off-by: Bence Szépkúti <bence.szepkuti@arm.com>
2020-08-19 10:35:41 +02:00
Gilles Peskine
aa9493a411 Add guards around code that is specific to dynamically-loaded groups
For some curves (semi-coincidentally, short Weierstrass curves), the
ECP module calculates some group parameters dynamically. Build the
code to calculate the parameters only if a relevant curve is enabled.
This fixes an unused function warning when building with only
Montgomery curves.

Signed-off-by: Gilles Peskine <gilles.peskine@arm.com>
2020-07-22 03:17:23 +02:00
Gilles Peskine
db09ef6d22 Include common.h instead of config.h in library source files
In library source files, include "common.h", which takes care of
including "mbedtls/config.h" (or the alternative MBEDTLS_CONFIG_FILE)
and other things that are used throughout the library.

FROM=$'#if !defined(MBEDTLS_CONFIG_FILE)\n#include "mbedtls/config.h"\n#else\n#include MBEDTLS_CONFIG_FILE\n#endif' perl -i -0777 -pe 's~\Q$ENV{FROM}~#include "common.h"~' library/*.c 3rdparty/*/library/*.c scripts/data_files/error.fmt scripts/data_files/version_features.fmt

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-07-02 11:26:57 +02:00