Commit graph

11673 commits

Author SHA1 Message Date
Gilles Peskine bbdf310c66
Merge pull request #221 from gilles-peskine-arm/annotate_todo_comments-20190813
SE keys: fix psa_destroy_key; add Github issue numbers for missing code
2019-08-14 13:59:53 +02:00
k-stachowiak 012dcc4b87 Remove PSA_PRE_1_0_KEY_DERIVATION and the corresponding code 2019-08-13 18:42:40 +02:00
k-stachowiak 0b74cf85ea Remove psa_key_derivation() and associated static functions 2019-08-13 18:42:27 +02:00
Gilles Peskine 26f3e2800d Honor MBEDTLS_CONFIG_FILE in fuzz tests 2019-08-13 18:00:02 +02:00
Gilles Peskine 56c0161b68 Test that a shared library build produces a dynamically linked executable 2019-08-13 17:54:26 +02:00
Gilles Peskine cf74050fea Test that the shared library build with CMake works 2019-08-13 17:54:26 +02:00
Gilles Peskine abf9b4dee8 Add a test of MBEDTLS_CONFIG_FILE
configs/README.txt documents that you can use an alternative
configuration file by defining the preprocessor symbol
MBEDTLS_CONFIG_FILE. Test this.
2019-08-13 17:54:26 +02:00
Gilles Peskine c9d7f94a65 Add issue numbers for some missing parts of secure element support 2019-08-13 16:17:16 +02:00
Gilles Peskine 5da7b3e55c Drivers must have a psa_destroy_key method
Drivers that allow destroying a key must have a destroy method. This
test bug was previously not caught because of an implementation bug
that lost the error triggered by the missing destroy method.
2019-08-13 16:04:34 +02:00
Gilles Peskine 4b7f340fbf Clean up status code handling inside psa_destroy_key
Adopt a simple method for tracking whether there was a failure: each
fallible operation sets overall_status, unless overall_status is
already non-successful. Thus in case of multiple failures, the
function always reports whatever failed first. This may not always be
the right thing, but it's simple.

This revealed a bug whereby if the only failure was the call to
psa_destroy_se_key(), i.e. if the driver reported a failure or if the
driver lacked support for destroying keys, psa_destroy_key() would
ignore that failure.

For a key in a secure element, if creating a transaction file fails,
don't touch storage, but close the key in memory. This may not be
right, but it's no wronger than it was before. Tracked in
https://github.com/ARMmbed/mbed-crypto/issues/215
2019-08-13 15:58:36 +02:00
Gilles Peskine 9ce31c466d Note about destroying a key with other open handles
https://github.com/ARMmbed/mbed-crypto/issues/214
2019-08-13 15:14:20 +02:00
Gilles Peskine caec27821f SE keys: make psa_destroy_key remove the key from storage 2019-08-13 15:11:49 +02:00
Gilles Peskine 8fe253ae4a SE keys: test that psa_destroy_key removes the key from storage 2019-08-13 15:11:25 +02:00
Gilles Peskine 3f7cd62ff5 Document better what wiping a key slot does not do
When a key slot is wiped, a copy of the key material may remain in
operations. This is undesirable, but does not violate the safety of
the code. Tracked in https://github.com/ARMmbed/mbed-crypto/issues/86
2019-08-13 15:01:08 +02:00
Gilles Peskine b4e73e9747 Add some design notes about multipart operation structures 2019-08-13 15:00:57 +02:00
Ron Eldor 40244bc348 Fix the license header of hkdf
Change the license header of `hkdf.h` to a format the that script
`apache_to_gpl.pl` knows how to parse.
2019-08-13 14:51:29 +03:00
Janos Follath 1d57a20cbe Make TODO comments consistent 2019-08-13 12:15:34 +01:00
Gilles Peskine 6eece5b666 Exclude DTLS 1.2 only with older OpenSSL
compat.sh used to skip OpenSSL altogether for DTLS 1.2, because older
versions of OpenSSL didn't support it. But these days it is supported.

We don't want to use DTLS 1.2 with OpenSSL unconditionally, because we
still use legacy versions of OpenSSL to test with legacy ciphers. So
check whether the version we're using supports it.
2019-08-13 10:36:14 +02:00
Gilles Peskine 30e13eb1c6
Merge pull request #211 from gilles-peskine-arm/psa-se_driver-generate_key
Add hooks for generate and sign in a secure element
2019-08-12 11:45:02 +02:00
Gilles Peskine 8df72f271f SE generate/sign/verify tests: also test export_public
Add a flow where the key is imported or fake-generated in the secure
element, then call psa_export_public_key and do the software
verification with the public key.
2019-08-09 16:43:36 +02:00
Gilles Peskine af906f852c RAM test driver: implement export_public 2019-08-09 16:43:36 +02:00
Gilles Peskine c068ded015 RAM test driver: improve key creation
Factor common code of ram_import and ram_fake_generate into a common
auxiliary function.

Reject key types that aren't supported by this test code.

Report the bit size correctly for EC key pairs.
2019-08-09 16:43:36 +02:00
Gilles Peskine eecadde6ad SE support: Test sign and verify hooks with a passthrough driver 2019-08-09 16:43:35 +02:00
Gilles Peskine edc6424d77 SE driver support: Implement sign and verify hooks 2019-08-09 16:43:35 +02:00
Gilles Peskine 11792086cc SE keys: implement and smoke-test p_generate 2019-08-09 16:43:35 +02:00
Gilles Peskine f3801fff77 Update import_key and generate_key SE methods to the current API
The methods to import and generate a key in a secure element drivers
were written for an earlier version of the application-side interface.
Now that there is a psa_key_attributes_t structure that combines all
key metadata including its lifetime (location), type, size, policy and
extra type-specific data (domain parameters), pass that to drivers
instead of separate arguments for each piece of metadata. This makes
the interface less cluttered.

Update parameter names and descriptions to follow general conventions.

Document the public-key output on key generation more precisely.
Explain that it is optional in a driver, and when a driver would
implement it. Declare that it is optional in the core, too (which
means that a crypto core might not support drivers for secure elements
that do need this feature).

Update the implementation and the tests accordingly.
2019-08-09 16:43:35 +02:00
Gilles Peskine 5a2d15256d
Merge pull request #183 from gilles-peskine-arm/psa-se_driver-register_key
Secure element key registration
2019-08-09 16:22:52 +02:00
Gilles Peskine 2c897d76ff Document the rationale for the armel build
Call the component xxx_arm5vte, because that's what it does. Explain
"armel", and more generally why this component exists, in a comment.
2019-08-09 16:05:05 +02:00
Gilles Peskine 49bd58274e Test the call to p_validate_slot_number when registering a key 2019-08-09 15:05:32 +02:00
Gilles Peskine a5f8749812 SE key registration: call p_validate_slot_number
When registering a key in a secure element, if the driver has a
p_validate_slot_number method, call it.
2019-08-09 15:05:32 +02:00
Gilles Peskine d772958ffc New function mbedtls_psa_register_se_key
Register an existing key in a secure element.

Minimal implementation that doesn't call any driver method and just
lets the application declare whatever it wants.
2019-08-09 15:05:21 +02:00
Gilles Peskine e88c2c1338 Pass the key creation method to drivers
Pass the key creation method (import/generate/derive/copy) to the
driver methods to allocate or validate a slot number. This allows
drivers to enforce policies such as "this key slot can only be used
for keys generated inside the secure element".
2019-08-09 14:59:31 +02:00
Gilles Peskine df17914e01 psa_start_key_creation: take the method as a parameter
Let psa_start_key_creation know what type of key creation this is. This
will be used at least for key registration in a secure element, which
is a peculiar kind of creation since it uses existing key material.
2019-08-09 14:54:03 +02:00
Gilles Peskine b231d996a0
Merge pull request #202 from gilles-peskine-arm/psa-se_driver-choose_key_slot_number
Let applications create a key in a specific secure element slot
2019-08-09 14:29:54 +02:00
Gilles Peskine fdbc544b22
Merge pull request #192 from gilles-peskine-arm/psa-test-policy_bad_algorithm
Add some negative tests for policy checks
2019-08-09 13:32:45 +02:00
Gilles Peskine 9d75202efb Clarify and expand the documentation of the allocate/create sequence 2019-08-09 11:33:48 +02:00
Janos Follath 849b05afb8 Fix PSA tests
The test framework has changed, but it did not cause any merge
conflicts. Still it affected new code in the tests.
2019-08-09 10:22:32 +01:00
Gilles Peskine 311809a49d
Merge pull request #201 from gilles-peskine-arm/psa-se_driver-set_key_slot_number
Add slot number attribute
2019-08-08 19:19:29 +02:00
Gilles Peskine 059d9d3fc5
Merge pull request #182 from gilles-peskine-arm/ecp-factorize-reduction
Replace some macros by functions in ecp
2019-08-08 19:16:27 +02:00
Gilles Peskine 8a52af9b77 Switch armel build to -Os
Without any -O option, the default is -O0, and then the assembly code
is not used, so this would not be a non-regression test for the
assembly code that doesn't build.
2019-08-08 16:09:02 +02:00
Janos Follath 8aa7e9bc56 Merge branch 'psa-api-1.0-beta' into merge-psa-api-branch-into-development 2019-08-08 14:40:23 +01:00
Gilles Peskine 0a1104474b Test restarting after creating a key in a specific slot 2019-08-08 11:02:30 +02:00
Gilles Peskine 46d9439a5e Support slot_number attribute when creating a key
Allow the application to choose the slot number in a secure element,
rather than always letting the driver choose.

With this commit, any application may request any slot. In an
implementation with isolation, it's up to the service to filter key
creation requests and apply policies to limit which applications can
request which slot.
2019-08-08 11:02:30 +02:00
Gilles Peskine ae9964d3ef Add validate_slot_number method to SE drivers
Pave the way for allowing the application to choose the slot number in
a secure element, rather than always letting the driver choose.
2019-08-08 11:02:30 +02:00
Gilles Peskine 0a23322431 Improve documentation of the allocate method 2019-08-08 11:02:30 +02:00
Gilles Peskine edbed5670a Rename psa_internal_allocate_key_slot to psa_get_empty_key_slot
This function no longer modifies anything, so it doesn't actually
allocate the slot. Now, it just returns the empty key slot, and it's
up to the caller to cause the slot to be in use (or not).
2019-08-08 10:58:09 +02:00
Gilles Peskine 094dac1d12 Fix copypasta 2019-08-08 10:58:09 +02:00
Gilles Peskine 013f5474cf Fix erasure of external flags
This didn't break anything now, but would have broken things once we
start to add internal flags.
2019-08-08 10:58:09 +02:00
Gilles Peskine 5a68056755 Rename internal macro to pass check-names.sh
check-names.sh rejects MBEDTLS_XXX identifiers that are not defined in
a public header.
2019-08-08 10:58:09 +02:00
Gilles Peskine 5fe5e27591 Test slot_number attribute
Test the behavior of the getter/setter functions.

Test that psa_get_key_slot_number() reports a slot number for a key in
a secure element, and doesn't report a slot number for a key that is
not in a secure element.

Test that psa_get_key_slot_number() reports the correct slot number
for a key in a secure element.
2019-08-08 10:58:09 +02:00