This commit adds four known answer tests for TLS 1.3 record protection
from the following sources:
- RFC 8448 "Example Handshake Traces for TLS 1.3"
- tls13.ulfheim.net "The New Illustrated TLS Connection"
It extends the test coverage of the existing record protection tests
in the following ways:
- The existing record protection tests hand-craft record transform
structures; the new tests use the function
mbedtls_ssl_tls13_populate_transform()
from library source to create an TLS 1.3 transform from raw
key material and connection information.
- The existing record protection tests only check that encryption
and decryption are inverse to each other; as such, they don't
catch non-compliant implementations of encryption and decryption
which happen to be inverse to each other. By adding a known answer
test for TLS 1.3 record protection, can gain confidence that our
implementation is indeed standards-compliant.
Signed-off-by: Hanno Becker <hanno.becker@arm.com>
This is necessary for the case where the public part of an EC keypair
needs to be computed from the private part - either because it was not
included (it's an optional component) or because it was compressed (a
format we can't parse).
This changes the API of two public functions: mbedtls_pk_parse_key() and
mbedtls_pk_parse_keyfile().
Tests and programs have been adapted. Some programs use a non-secure RNG
(from the test library) just to get things to compile and run; in a
future commit this should be improved in order to demonstrate best
practice.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Simple find and replace using `#include (<|")mbedtls/(.*)_internal.h(>|")`
and `#include $1$2_internal.h$3`.
Also re-generated visualc files by running
`scripts/generate_visualc_files.pl`.
Signed-off-by: Chris Jones <christopher.jones@arm.com>
Some functions were not deinitializing the PSA subsystem. This could
lead to resource leaks at the level of individual test cases, and
possibly at the level of the whole test suite depending on the order
and selection of test cases.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Make USE_PSA_INIT() and USE_PSA_DONE() available in all test suites in
all cases, doing nothing if MBEDTLS_USE_PSA_CRYPTO is disabled. Use
those in preference to having explicit
defined(MBEDTLS_USE_PSA_CRYPTO) checks (but there may still be places
left where using the new macros would be better).
Also provide PSA_INIT() by symmetry with PSA_DONE(), functional
whenver MBEDTLS_PSA_CRYPTO_C is enabled, but currently unused.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Moves the functions `test_fail`, `test_set_step`, `test_skip` and the struct
`test_info` from `tests/suites/helpers.function` to `tests/src/helpers.*`.
This is done to open these functions up to the API where they can be used by
other functions in the 'src' test infrastructure module.
As the functions are now contained within the src folder of the testing
infrastructure, the `mbedtls_` prefix has been added to the functions.
Signed-off-by: Chris Jones <christopher.jones@arm.com>
ssl_tls1_3_keys.c exports a structure containing all labels used
in the TLS 1.3 key schedule, but the TLS 1.3 key scheduling unit
tests so far replicated those labels in the test file. In particular,
wrong label values in ssl_tls1_3_keys.c wouldn't have been caught
by the unit tests.
This commit modifies the TLS 1.3 key schedule unit tests to use
the TLS 1.3 labels as exported by ssl_tls1_3_keys.c. This not only
makes sure that those labels are correct, but also avoids hardcoding
their hex-encoding in the test file.
Signed-off-by: Hanno Becker <hanno.becker@arm.com>
The tests are supposed to be failing now (in all.sh component
test_memsan_constant_flow), but they don't as apparently MemSan doesn't
complain when the src argument of memcpy() is uninitialized, see
https://github.com/google/sanitizers/issues/1296
The next commit will add an option to test constant flow with valgrind, which
will hopefully correctly flag the current non-constant-flow implementation.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
The convention from the TLS RFC is a bit unusual, so even if the test
function's introductory comment mentions that we're taking the RFC's
definition, it doesn't hurt to repeat it in crucial places.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Passing a length of 0 to it is perfectly acceptable, the macro was designed to
handle it correctly.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
We only have a single integer available for two nested loops, but the loop
sizes are small enough compared to the integer's range that we can encode both
indexes. Since the integer is displayed in decimal in case of errors, use a
power of 10 to pack the two indexes together.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Currently this breaks all.sh component test_memsan_constant_flow, just as
expected, as the current implementation is not constant flow.
This will be fixed in the next commit.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Everything works at the byte level, not bit level. Flipping the lsb is just
one convenient way to corrupt a byte, but don't really care about individual
bits.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
The test function now depends on MBEDTLS_TEST_HOOKS, which is enabled by
config.py full, and since there are already components in all.sh exercising
the full config, this test function is sill exercised even with this new
dependency.
Since this is the first time a test function depends on MBEDTLS_TEST_HOOKS,
fix a bug in check-names.sh that wasn't apparent so far: headers from
library/*.h were not considered when looking for macro definitions. This
became apparent because MBEDTLS_STATIC_TESTABLE is defined in library/common.h
and started being used in library/ssl_msg.c, so was flagged as a likely typo.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
The dummy implementation is not constant-flow at all for now, it's just
here as a starting point and a support for developing the tests and putting
the infrastructure in place.
Depending on the implementation strategy, there might be various corner cases
depending on where the lengths fall relative to block boundaries. So it seems
safer to just test all possible lengths in a given range than to use only a
few randomly-chosen values.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
