Commit graph

29730 commits

Author SHA1 Message Date
Gilles Peskine
04ae479b04 mbedtls_ecp_write_key_ext: document error for no private key set
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-28 13:19:42 +01:00
Gilles Peskine
c0f7a8680f mbedtls_ecp_write_key(): deprecate the old function
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-28 13:19:42 +01:00
Gilles Peskine
84b9f1b039 mbedtls_ecp_write_key_ext(): migrate internally
Stop using mbedtls_ecp_write_key() except to test it.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-28 13:19:42 +01:00
Gilles Peskine
acdc52e154 mbedtls_ecp_write_key_ext(): recommend over the old function in documentation
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-28 13:19:42 +01:00
Gilles Peskine
e3fb4ccabf mbedtls_ecp_write_key_ext(): new function
Same as mbedtls_ecp_write_key(), but doesn't require the caller to figure out
the length of the output and possibly distinguish between Weierstrass and
Montgomery curves.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-28 13:19:42 +01:00
Manuel Pégourié-Gonnard
7f523bf9eb
Merge pull request #8845 from gilles-peskine-arm/ecp-write-doc-3.6
Document ECP write functions
2024-02-28 11:04:38 +00:00
Dave Rodgman
09e6fb42eb
Merge pull request #8865 from daverodgman/iar-fixes-feb27
Add missing casts to fix IAR warnings
2024-02-27 14:41:53 +00:00
Manuel Pégourié-Gonnard
5b5faf0898
Merge pull request #8844 from davidhorstmann-arm/restore-x509-functions-to-public
Restore some X509 functions to public headers
2024-02-27 10:55:16 +00:00
Dave Rodgman
6a3da2d5ed Add missing casts
Signed-off-by: Dave Rodgman <dave.rodgman@arm.com>
2024-02-27 10:41:53 +00:00
Tom Cosgrove
ca21b241bd
Merge pull request #8840 from gilles-peskine-arm/domain_parameters-remove
Remove domain parameters
2024-02-27 10:36:51 +00:00
Gilles Peskine
e22f6a9610 Finish cleaning up override that's no longer needed
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-26 16:57:31 +01:00
Gilles Peskine
97c0b2f393 Remove domain parameters from psa_key_attributes_t
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-26 16:57:30 +01:00
David Horstmann
ef950ccb1d Un-unrestore mbedtls_x509_string_to_names()
Re-restore mbedtls_x509_string_to_names() to public as our example
programs use it, and it is the reverse of mbedtls_x509_dn_gets().

Add a docstring, so that it is a properly documented public function.

Signed-off-by: David Horstmann <david.horstmann@arm.com>
2024-02-26 13:59:43 +00:00
Tom Cosgrove
f4a200f106
Merge pull request #8838 from paul-elliott-arm/improve_test_data_accessors
Improve test info data accessors
2024-02-26 11:22:20 +00:00
Gilles Peskine
ae5eb64705 Remove domain parameters from the public API
Only leave deprecated, minimal non-linkable functions.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-26 08:50:53 +01:00
Gilles Peskine
bb6f3ff394 Rename variables
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-26 08:50:53 +01:00
Gilles Peskine
4c32b69f37 Ignore domain parameters in RSA key generation
Remove the ability to select a custom public exponent via domain parameters
in RSA key generation. The only way to select a custom public exponent is
now to pass custom production parameters to psa_generate_key_ext().

A subsequent commit will remove domain parameters altogether from the API,
thus this commit does not bother to update the documentation.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-26 08:50:38 +01:00
Gilles Peskine
6a2c400b8c typo
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-26 08:34:06 +01:00
Gilles Peskine
0f63028809
Merge pull request #8815 from gilles-peskine-arm/psa_generate_key_ext-prototype
Introduce psa_generate_key_ext
2024-02-26 07:16:49 +00:00
Paul Elliott
9011dae0c1 Improve documentation / comments
Signed-off-by: Paul Elliott <paul.elliott@arm.com>
2024-02-24 10:57:22 +00:00
Paul Elliott
665cf928d9
Merge pull request #8856 from Ryan-Everett-arm/threadsafe-openkey
Make psa_open_key threadsafe
2024-02-24 10:29:57 +00:00
Paul Elliott
5a4a6e44ef
Merge pull request #8833 from Ryan-Everett-arm/threadsafe-multiparts
Make multi-part operations thread-safe
2024-02-24 10:29:20 +00:00
Tom Cosgrove
817772a6ca
Merge pull request #8716 from mschulz-at-hilscher/feature/gcm_largetable
Use large GCM tables
2024-02-23 16:25:38 +00:00
Manuel Pégourié-Gonnard
81c322329e
Merge pull request #8855 from gilles-peskine-arm/benchmark-ecdh-no-legacy
Remove most uses of MBEDTLS_ALLOW_PRIVATE_ACCESS in test programs
2024-02-23 09:16:46 +00:00
Gilles Peskine
7f72a06e02 Remove cruft
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-22 18:41:25 +01:00
Gilles Peskine
f6eb0b8ab0 Changelog entry for benchmark improvement
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-22 18:40:10 +01:00
Manuel Pégourié-Gonnard
dd9cbf99c2 Benchmark only one side of ECDH, both static and ephemeral
Static ECDH is of interest to us as developers because it's a generic
scalar multiplication (as opposed to using the standard base point) and
it's useful to have that handy.

For reference the other operations of interest to developers are:
- multiplication of the conventional base point: ECDSA signing is almost
exactly that (just a few field ops on top, notably 1 inversion);
- linear combination: ECDSA verification is almost exactly that too.

Including ephemeral as well, because it's hopefully what's of interest
to most users.

Compared to the previous version, include only one side of the
operations. I don't think including both sides is of interest to anyone.

Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
2024-02-22 12:29:06 +01:00
Gilles Peskine
74589ba31c ssl_context_info: explicitly note accesses to private fields
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-22 12:12:01 +01:00
Gilles Peskine
72da8b3521 Don't authorize private access to fields where not actually needed
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-22 12:07:29 +01:00
Gilles Peskine
abf0be392a fuzz_dtlsserver: explicitly note the one access to a private field
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-22 12:07:09 +01:00
Gilles Peskine
d5f68976e8 fuzz_pubkey, fuzz_privkey: no real need to access private fields
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-22 12:05:35 +01:00
Ryan Everett
e110a4c900 Make psa_open_key threadsafe
This is a simple case of register then unregister

Signed-off-by: Ryan Everett <ryan.everett@arm.com>
2024-02-22 10:43:24 +00:00
Ryan Everett
9af70e51c1 Make multi-part AEAD operations thread-safe
The setup calls are the only calls to use a key ID.
The key is then copied into the operation object,
all future API calls use the copy instead of the key in the slot.

Simultaneous API calls on the same operation object are not thread-safe.

Signed-off-by: Ryan Everett <ryan.everett@arm.com>
2024-02-22 10:27:52 +00:00
Ryan Everett
bbedfcec2e Make multi-part PAKE operations thread-safe
The only interaction with key IDs here is in the changed function.

Simultaneous API calls on the same operation object are not thread-safe.

Signed-off-by: Ryan Everett <ryan.everett@arm.com>
2024-02-22 10:27:52 +00:00
Ryan Everett
5ac6fa7aae Make multi-part key derivation operations thread-safe
One can input a key using a key identifier through the two changed functions.
Inputted keys are copied into the operation object.
Any material inputted in byte form is separate to the key slot system.
Outputting a key is threadsafe as per the key loading work.
The verification API is yet to be implemented.

Simultaneous API calls on the same operation object are not thread-safe.

Signed-off-by: Ryan Everett <ryan.everett@arm.com>
2024-02-22 10:27:52 +00:00
Ryan Everett
c0053cc499 Make multi-part cipher operations thread-safe
Within setup we create a copy of the key and put it in the operation field.
After setup, we only ever use the new copy,
and do not interact with any key slots.
Therefore we need only register as a reader of the key during setup,
then unregister after we stop accessing the key.

Simultaneous API calls on the same operation object are not thread-safe.

Signed-off-by: Ryan Everett <ryan.everett@arm.com>
2024-02-22 10:27:52 +00:00
Ryan Everett
291267f486 Make restartable signature verifications thread-safe
We copy the key from a slot to the operation object in _start.
_complete and _abort do not access any key slots, instead using
the local copy.

Concurrently using the same operation object is not thread-safe.

Signed-off-by: Ryan Everett <ryan.everett@arm.com>
2024-02-22 10:27:52 +00:00
Ryan Everett
dcc03d552d Make restartable signature operations thread-safe
We copy the key from a slot to the operation object in _start.
_complete and _abort do not access any key slots, instead using
the local copy.

Concurrently using the same operation object is not thread-safe.

Signed-off-by: Ryan Everett <ryan.everett@arm.com>
2024-02-22 10:27:52 +00:00
Ryan Everett
fb9857ff9e Make multi-part MAC operations thread-safe
Within setup we create a copy of the key and put it in the operation field.
After setup, we only ever use the new copy - and do not interact with any key slots.
Therefore we need only register as a reader of the key during setup,
then unregister after we stop accessing the key.

Simultaneous API calls on the same operation object are not thread-safe.

Signed-off-by: Ryan Everett <ryan.everett@arm.com>
2024-02-22 10:27:52 +00:00
Gilles Peskine
7a3059ac92 Benchmark: not using private fields anymore
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-22 10:53:31 +01:00
Gilles Peskine
984352d6f1 Benchmark: remove the legacy-context ECDH block
We have two blocks of code to benchmark ECDH. One uses the legacy context
structure, which is only enabled when MBEDTLS_ECP_RESTARTABLE is enabled.
That block doesn't convey any information that's specific to restartable
ECC, it exists only for historical reasons (it came first). The other block
uses only the implementation-agnostic API.

Remove the block that uses the legacy context. It doesn't provide much
extra information and most users won't even see it.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-22 10:50:12 +01:00
Gilles Peskine
0aab69d2eb
Merge pull request #8807 from gilles-peskine-arm/pk_import_into_psa-implement_import
Implement mbedtls_pk_import_into_psa
2024-02-21 15:45:17 +00:00
Paul Elliott
d237190f04
Merge pull request #8773 from Ryan-Everett-arm/threadsafe-key-locking
Make key locking and one-shot operations thread safe
2024-02-21 13:55:12 +00:00
Gilles Peskine
dd49c739f0 Merge remote-tracking branch 'development' into pk_import_into_psa-implement_import
Conflicts:
* tests/suites/test_suite_pk.function: consecutive changes to the
  depends_on line of pk_sign_verify and its argument list.
2024-02-21 12:10:40 +01:00
David Horstmann
8ed3c60504 Remove unnecessary ChangeLog entry
This bug was never present in a release, so no ChangeLog entry is
needed.

Signed-off-by: David Horstmann <david.horstmann@arm.com>
2024-02-21 10:36:54 +00:00
David Horstmann
946491360b Unrestore mbedtls_x509_string_to_names()
This function should be internal, being exposed only via functions like
mbedtls_x509write_crt_set_subject_name().

Signed-off-by: David Horstmann <david.horstmann@arm.com>
2024-02-21 10:35:33 +00:00
Manuel Pégourié-Gonnard
0ecb5fd6f5
Merge pull request #8574 from ronald-cron-arm/ssl-tickets
Fix and align ticket age check in ssl_ticket.c for TLS 1.2 and TLS 1.3
2024-02-21 09:38:46 +00:00
Ryan Everett
93cea578b9 Clarify which unregister operation needs to be used
Signed-off-by: Ryan Everett <ryan.everett@arm.com>
2024-02-20 18:01:29 +00:00
Gilles Peskine
23605d19d9 More renaming: method -> production parameters
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-20 17:10:56 +01:00
Gilles Peskine
092ce51c47 Rename "key generation method" to "key production parameters"
"Key generation method" was misleading since it also applies to key
derivation. Change "key generation" to "key production", which we aren't
using yet and has roughly the right intuition. Change "method" to
"parameters" which there seems to be a slight preference for. Discussion
thread: https://github.com/Mbed-TLS/mbedtls/pull/8815#discussion_r1486524295

Identifiers renamed:
psa_key_generation_method_t → psa_key_production_parameters_t
psa_key_generation_method_s → psa_key_production_parameters_s
PSA_KEY_GENERATION_METHOD_INIT → PSA_KEY_PRODUCTION_PARAMETERS_INIT
method → params
method_data_length → params_data_length
default_method → default_production_parameters
psa_key_generation_method_is_default → psa_key_production_parameters_are_default
setup_key_generation_method → setup_key_production_parameters
key_generation_method_init → key_production_parameters_init

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2024-02-20 16:18:13 +01:00