- removed the check on saltlen > 0 and added tests
positive test cases for this.
- added negative test cases when even saltlen == 0
is not enough. This allowed to uncover an underflow bu
in the slen check (when olen-slen-2 is negative)
- fixed the saltlen check to avoid underflow
- added more test cases where saltlen is the maximum
possible value and one above the maximum possible value
(different hash, different key size)
Signed-off-by: Cédric Meuter <cedric.meuter@gmail.com>
When an Mbed TLS error code combines a low-level error and a
high-level error, the low-level error is usually closer to the root
cause (for example HW_ACCEL_FAILED or ENTROPY_SOURCE_FAILED is more
informative than RSA_PRIVATE_FAILED). So prioritize the low-level code
when converting to a PSA error code, rather than the high-level code
as was (rather arbitrarily) done before.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
mbedtls_rsa_private() could return the sum of two RSA error codes
instead of a valid error code in some rare circumstances:
* If rsa_prepare_blinding() returned MBEDTLS_ERR_RSA_RNG_FAILED
(indicating a misbehaving or misconfigured RNG).
* If the comparison with the public value failed (typically indicating
a glitch attack).
Make sure not to add two high-level error codes.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
- all positibe test cases were sampled from the CAVP test suite
(SigGenPSS_186-2.txt, SigGenPSS_186-3.txt)
Only kept one representative for each triple (modlen, sha, saltlen)
- two extra test cases were added to cover the maximum salt length
(slen=olen-slen-2 and slen=(olen-slen-2)-1)
- in rsa.c, the salt intermediate buffer was too small to cover cases
where slen > hlen. So reworked the code to generate the salt in the
encoded message directly. This has the advantage to remove a memcpy
and a memset.
Signed-off-by: Cédric Meuter <cedric.meuter@gmail.com>
- Fixed code style.
- Clarified the documentation of what happens when saltlen is set to
MBEDTLS_RSA_SALT_LEN_ANY.
- Added range check on saltlen to reject out of range values.
(Code review done by @gilles-peskine-arm)
Signed-off-by: Cédric Meuter <cedric.meuter@gmail.com>
extension of mbedtls_rsa_rsassa_pss_sign() with an extra argument
'saltlen' which allows to inject the length of the salt to the function,
as opposed to the original function which internally computes the
maximum possible salt length. If MBEDTLS_RSA_SALT_LEN_ANY is passed
the function falls back to the the original behaviour. The original
function mbedtls_rsa_rsassa_pss_sign() can simply defer to it.
This allows to make some CAVP PSS generation tests that require the use
of a salt length which is smaller that the hash length.
Signed-off-by: Cédric Meuter <cedric.meuter@gmail.com>
Closing a wrapped key with the new SE driver interface while
MBEDTLS_PSA_CRYPTO_SE_C is also enabled leads to the key material not
being freed, even though an old SE driver is not in use, leading to a
memory leak. This is because a wrapped key is also considered external.
This commit extends the check for skipping by checking whether an
old-style SE driver is registered with the provided slot, in addition to
checking whether the key is external.
Signed-off-by: Fredrik Strupe <fredrik.strupe@silabs.com>
Make it clear that this is an abstraction of the random generator
abstraction, and not an abstraction of the PSA random generator.
mbedtls_psa_get_random and MBEDTLS_PSA_RANDOM_STATE are public-facing
definitions and will be moved in a subsequent commit.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
In the external RNG case, don't make mbedtls_psa_get_random() a
static inline function: this would likely result in identical
instances of this function in every module that uses it. Instead, make
it a single function with external linkage.
In the non-external case, instead of a trivial wrapper function, make
mbedtls_psa_get_random a constant pointer to whichever DRBG function
is being used.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Hide the obtention of the pointer to the RNG state behind a macro.
To make it possible to use this macro in a module other than
psa_crypto.c, which will happen in the future, make sure that the
definition of the macro does not reference internal variables of
psa_crypto.c. For this purpose, in the internal-DRBG case, export a
symbol containing the address of the DRBG state.
When the RNG state is a pointer a DRBG state, just keep this pointer
in a variable: there's no need to store a pointer to a larger structure.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Executed ./scripts/bump_version.sh --version 2.25.0 --so-crypto 6
Increasing the SO version of the crypto library, because the openless
API improvement came with API/ABI incompatibilities. For example
- the size of psa_key_handle_t changed
- the type of a parameter in 18 public functions has changed from
psa_key_handle_t to mbedtls_svc_key_id_t
Signed-off-by: Janos Follath <janos.follath@arm.com>
T1 is set to a 2-limb value. The first operation that takes it as
input is mbedtls_mpi_mul_int, which makes it grow to 3 limbs. Later it
is shifted left, which causes it to grow again. Set its size to the
final size from the start. This saves two calls to calloc(), at the
expense of a slowdown in some operations involving T1 as input since
it now has more leading zeros.
Setting T1 to 3 limbs initially instead of 2 saves about 6% of the
calloc() calls in test_suite_ecp and does not incur a performance
penalty. Setting T1 to A->n + 2 limbs instead of 2 saves about 20% of
the calloc calls and does not cause a measurable performance
difference on my Linux PC.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Note a possible microoptimization in mbedtls_mpi_mul_hlp that I tried
in the hope of reducing the number of allocations, but turned out to
be counterproductive.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Rewrite mbedtls_mpi_mul_int to call mpi_mul_hlp directly rather than
create a temporary mpi object. This has the benefit of not performing
an allocation when the multiplication is in place (mpi operand aliased
with the result) and the result mpi is large enough.
This saves about 40% of the calloc() calls in test_suite_ecp. There is
no measurable performance difference on my Linux PC.
The cost is a few bytes in bignum.o.
When there is no aliasing, or when there is aliasing but the mpi
object needs to be enlarged, the performance difference is negligible.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
mbedtls_mpi_sub_abs systematically allocated a new mpi when the result
was aliased with the right operand (i.e. X = A - X). This aliasing
very commonly happens during ECP operations. Rewrite the function to
allocate only if the result might not fit otherwise.
This costs a few bytes of code size in bignum.o, and might make
mbedtls_mpi_sub_abs very very slightly slower when no reallocation is
done. However, there is a substantial performance gain in ECP
operations with Montgomery curves (10-20% on my PC).
test_suite_ecp drops from 1422794 to 1271506 calls to calloc().
This commit also fixes a bug whereby mbedtls_mpi_sub_abs would leak
memory when X == B (so TB was in use) and the result was negative.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
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>
This algorithm replaces the pre-existing stream cipher algorithms.
The underlying stream cipher is determined by the key type.
Signed-off-by: Bence Szépkúti <bence.szepkuti@arm.com>
In order to remove large buffers from the stack, the der data is written
into the same buffer that the pem is eventually written into, however
although the pem data is zero terminated, there is now data left in the
buffer after the zero termination, which can cause
mbedtls_x509_crt_parse to fail to parse the same buffer if passed back
in. Patches also applied to mbedtls_pk_write_pubkey_pem, and
mbedtls_pk_write_key_pem, which use similar methods of writing der data
to the same buffer, and tests modified to hopefully catch any future
regression on this.
Signed-off-by: Paul Elliott <paul.elliott@arm.com>
