The key derivation algorithm PSA_ALG_TLS12_ECJPAKE_TO_PMS cannot be
used on a shared secret from a key agreement since its input must be
an ECC public key. Reject this properly.
This is tested by test_suite_psa_crypto_op_fail.generated.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
The following shell command lists features that seem to be supported, but
are missing from include/psa/crypto_config.h:
```
for x in $(grep -ho -Ew '(PSA_WANT|MBEDTLS_PSA_BUILTIN)_\w+_\w+' library/psa_crypto*.c | sed 's/^MBEDTLS_PSA_BUILTIN/PSA_WANT/' | sort -u); do grep -qw $x include/psa/crypto_config.h || echo $x; done
```
This looks for PSA_WANT_<kind>_<thing> macros that gate a part of the
library, as well as their MBEDTLS_PSA_BUILTIN_<kind>_<thing> counterparts.
This is not necessarily a complete list of identifiers that must appear
in the config file, since a few features are not gated.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
In some contexts, the output pointer may equal the first input
pointer, in which case copying is not only superfluous but results in
"Source and destination overlap in memcpy" errors from Valgrind (as I
observed in the context of ecp_double_jac) and a diagnostic message
from TrustInSoft Analyzer (as Pascal Cuoq reported in the context of
other ECP functions called by cert-app with a suitable certificate).
Signed-off-by: Aaron M. Ucko <ucko@ncbi.nlm.nih.gov>
Now it can accept serial both as decimal and hex number (only one format
at a time, of course, not simultaneously).
Signed-off-by: Valerio Setti <vsetti@baylibre.com>
"version 1.1 of #5137" is not meaningful to users, only as an internal
project milestone. Explain what this means from a user's point of view.
Announce the requirement for jsonschema in the proper section, which is
"Requirement changes". Mention jinja2 and basic.requirements.txt which
had not previously been explicitly mentioned in the changelog.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
If my understanding is correct (to be confirmed in review), this is a new
feature which was not particularly desired on its own but was the simplest
way to fix an interoperability issue in TLS 1.2 caused accidentally by
the work on TLS 1.3.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Mention the name of the new functions in the "Features" entry. Clarify what
they're for (there's no structure called mbedtls_x509_named_data, it's
mbedtls_asn1_named_data, but that name isn't so important here since we've
mentioned the names of the functions).
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
We removed internal code left over after removing a feature in Mbed TLS 3.0.
The removal of the internal code is not user-visible.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
