The pkcs7_get_signed_data function should be made global #1
Comments
naynajain
pushed a commit
that referenced
this issue
Nov 17, 2020
The previous code used comparison operators >= and == that are quite likely to
be compiled to branches by some compilers on some architectures (with some
optimisation levels).
For example, take the following function:
void old_update( size_t data_len, size_t *padlen )
{
*padlen *= ( data_len >= *padlen + 1 );
}
With Clang 3.8, let's compile it for the Arm v6-M architecture:
% clang --target=arm-none-eabi -march=armv6-m -Os foo.c -S -o - |
sed -n '/^old_update:$/,/\.size/p'
old_update:
.fnstart
@ BB#0:
.save {r4, lr}
push {r4, lr}
ldr r2, [r1]
adds r4, r2, #1
movs r3, #0
cmp r4, r0
bls .LBB0_2
@ BB#1:
mov r2, r3
.LBB0_2:
str r2, [r1]
pop {r4, pc}
.Lfunc_end0:
.size old_update, .Lfunc_end0-old_update
We can see an unbalanced secret-dependant branch, resulting in a total
execution time depends on the value of the secret (here padlen) in a
straightforward way.
The new version, based on bit operations, doesn't have this issue:
new_update:
.fnstart
@ BB#0:
ldr r2, [r1]
subs r0, r0, #1
subs r0, r0, r2
asrs r0, r0, Mbed-TLS#31
bics r2, r0
str r2, [r1]
bx lr
.Lfunc_end1:
.size new_update, .Lfunc_end1-new_update
(As a bonus, it's smaller and uses less stack.)
While there's no formal guarantee that the version based on bit operations in
C won't be translated using branches by the compiler, experiments tend to show
that's the case [1], and it is commonly accepted knowledge in the practical
crypto community that if we want to sick to C, bit operations are the safest
bet [2].
[1] https://github.com/mpg/ct/blob/master/results
[2] https://github.com/veorq/cryptocoding
Signed-off-by: Manuel Pégourié-Gonnard <[email protected]>
nick-child-ibm
pushed a commit
to nick-child-ibm/mbedtls-1
that referenced
this issue
May 25, 2021
Update my fork to Github repo
nick-child-ibm
pushed a commit
that referenced
this issue
May 12, 2022
The PSA Crypto API uses 0 as the initial counter value, but the test vector
in RFC 7539 uses 1. So the unit tests here include an extra leading block.
The expected data for this leading block was calculated with Cryptodome.
#!/usr/bin/env python3
import re
from Cryptodome.Cipher import ChaCha20
key = bytes.fromhex('000102030405060708090a0b0c0d0e0f101112131415161718191a1b1c1d1e1f')
nonce = bytes.fromhex('000000000000004a00000000')
encrypt = lambda pt: ChaCha20.new(key=key, nonce=nonce).encrypt(pt)
# Cryptodome uses counter=0, like PSA Crypto. Prepend a 64-byte input block #0
# so that the plaintext from RFC 7539 starts exactly at block #1.
header = b'The RFC 7539 test vector uses counter=1, but PSA uses counter=0.'
assert(len(header) == 64)
sunscreen = b"Ladies and Gentlemen of the class of '99: If I could offer you only one tip for the future, sunscreen would be it."
plaintext = header + sunscreen
zeros = b'\x00' * len(plaintext)
keystream = encrypt(zeros)
ciphertext = encrypt(plaintext)
print('RFC 7539 §2.4.2')
print('Keystream:')
print(re.sub(r'(..)', r'\1:', keystream[64:].hex()))
print('Ciphertext Subscreen:')
print(re.sub(r'(..)', r'\1 ', ciphertext[64:].hex()))
print('')
print(f"""\
PSA symmetric decrypt: ChaCha20, RFC7539 keystream
depends_on:PSA_WANT_ALG_STREAM_CIPHER:PSA_WANT_KEY_TYPE_CHACHA20
# Keystream from RFC 7539 §2.4.2, with an extra 64-byte output block prepended
# because the test vector starts at counter=1 but our API starts at counter=0.
cipher_decrypt:PSA_ALG_STREAM_CIPHER:PSA_KEY_TYPE_CHACHA20:"{key.hex()}":"{nonce.hex()}":"{zeros.hex()}":"{keystream.hex()}"
PSA symmetric decrypt: ChaCha20, RFC7539 sunscreen
depends_on:PSA_WANT_ALG_STREAM_CIPHER:PSA_WANT_KEY_TYPE_CHACHA20
# Test vector from RFC 7539 §2.4.2, with an extra 64-byte block prepended
# because the test vector starts at counter=1 but our API starts at counter=0.
cipher_decrypt:PSA_ALG_STREAM_CIPHER:PSA_KEY_TYPE_CHACHA20:"{key.hex()}":"{nonce.hex()}":"{ciphertext.hex()}":"{plaintext.hex()}"
""")
Signed-off-by: Gilles Peskine <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Description
Enhancement
Due to some signed files being in
mbedtls_pkcs7_signed_dataformat (rather thanpkcs7), it would be nice to be able to call thepkcs7_get_signed_datafunction directly (right now it is static). This would allow for outside programs to use the function to parse the new file format.Suggested enhancement
Add
pkcs7_get_signed_datato pkcs7.hThe text was updated successfully, but these errors were encountered: