Reworked the validation of MAC algorithm with the used key type by
introducing psa_mac_key_can_do, which guarantees that PSA_MAC_LENGTH can
be called successfully after validation of the algorithm and key type.
This means psa_get_mac_output_length is no longer required.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
Comparing algorithm with its FULL_LENGTH_MAC version doesn't work in
cases where algorithm is a wildcard. Wildcard input is not specified in
the documentation of the function, but in order to test the function
using the same test as PSA_MAC_LENGTH we're mimicking that behaviour here.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
Avoid code duplication. Also update the guarantees made by the function
doc to match the guarantees given by PSA_MAC_LENGTH.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
This makes it more in-line with how psa_key_policy_permits works. It
also adds consistency: the intersection of MAC with default length and
MAC with exact-length is now computed correctly in case the exact length
equals the default length of the algorithm when used with the given
key type.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
Change psa_ecdsa_sign/verify signature to that of
a sign/verify_hash driver entry point before to
move them to the psa_crypto_ecp.c ECP specific file.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Change psa_rsa_sign/verify signature to that of
a sign/verify_hash driver entry point before to
move them to the psa_crypto_rsa.c RSA specific file.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Change psa_driver_wrapper_sign/verify_hash signature
to that of a sign/verify_hash driver entry point.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Wrap sign/verify_hash software implementation into
psa_sign/verify_hash_internal() functions whose
signature is that of a sign/verify_hash driver
entry point.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Change the psa_crypto use of the CHACHA20 cipher to also use the new
MBEDTLS_PSA_BUILTIN_KE_TYPE_CHACHA20.
Signed-off-by: David Brown <david.brown@linaro.org>
Change a few conditionals in the psa library to be based on the
MBEDTLS_PSA_BUILTIN_KEY_TYPE_DES instead of the WANT macros. Future
additions of HW acceleration will need to be mindful of these
definitions if any of this code is needed in those instances.
Signed-off-by: David Brown <david.brown@linaro.org>
When converting definitions to use the new PSA defines, one erroneously
was conditionalized on the WANT macro instead of on the BUILTIN macro.
Signed-off-by: David Brown <david.brown@linaro.org>
There are a few instances of MBEDTLS_*_C (specifically for DES) in
psa_crypto.c. Change to the PSA_WANT_KEY_TYPE_DES macros to reflect the
new PSA crypto config.
Signed-off-by: David Brown <david.brown@linaro.org>
This file will always be used with the PSA configurations, so use the
MBEDTLS_PSA_BUILTIN... definitions for the symmetric cyphers.
Signed-off-by: David Brown <david.brown@linaro.org>
Use PSA_EXPORT_KEY_OUTPUT_SIZE macro to compute the
size of the buffer to contain the generated key
instead of computing it alongside the key type and
size validation.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
When generating transparent keys, we need to be able
to compute the size of the key buffer whether the
key is generated by the Mbed TLS library or by an
accelerator. Thus, change the RSA/ECP
MBEDTLS_PSA_BUILTIN_... compilation guards with
their PSA_WANT_... counterparts.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Change psa_generate_key_internal() signature to
that of a PSA driver generate_key entry point.
That way, this function can be called by the
driver wrapper when a software fallback is
necessary.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Preparatory commit to eventually change
psa_generate_key_internal() signature to that of
a PSA driver generate_key entry point.
To be able to change the signature, the buffer to
store the generated key has to be allocated before
the call to psa_generate_key_internal().
This commit moves the allocation and clean-up in
case of error of the buffer to store the generated
key from psa_generate_key_internal() to
psa_generate_key().
This has the nice benefit of factorizing the key
buffer allocation and clean-up.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Preparatory commit to eventually change
psa_generate_key_internal() signature to that of
a PSA driver generate_key entry point.
To be able to change the signature, the buffer to
store the key has to be allocated before the call
to psa_generate_key_internal() thus its size has
to be calculed beforehand as well.
This is the purpose of this commit: to move the
computation of the key size in bytes out of
psa_generate_key_internal().
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Small improvements to psa_generate_key_internal()
implementation:
. declare only once the status local variable and
initialize it to PSA_ERROR_CORRUPTION_DETECTED
to improve robustness against FI attacks.
. remove an unnecessary assignment.
. use type local variable instead of its global
variable equivalent.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
This brings them in line with PSA Crypto API 1.0.0
PSA_ALG_AEAD_WITH_DEFAULT_TAG_LENGTH -> PSA_ALG_AEAD_WITH_DEFAULT_LENGTH_TAG
PSA_ALG_AEAD_WITH_TAG_LENGTH -> PSA_ALG_AEAD_WITH_SHORTENED_TAG
Signed-off-by: Bence Szépkúti <bence.szepkuti@arm.com>
In psa_generate_derived_key_internal() an error case was returning
directly rather than jumping to the exit label, which meant that an
allocated buffer would not be free'd.
Found via coverity.
Signed-off-by: Paul Elliott <paul.elliott@arm.com>
Attempting to create an ECC key with a curve specification that is not
valid can plausibly fail with PSA_ERROR_INVALID_ARGUMENT ("this is not
a curve specification at all") or PSA_ERROR_NOT_SUPPORTED ("this may
be a curve specification, but not one I support"). The choice of error
is somewhat subjective.
Before this commit, due to happenstance in the implementation, an
attempt to use a curve that is declared in the PSA API but not
implemented in Mbed TLS returned PSA_ERROR_INVALID_ARGUMENT, whereas
an attempt to use a curve that Mbed TLS supports but for which support
was disabled at compile-time returned PSA_ERROR_NOT_SUPPORTED. This
inconsistency made it difficult to write negative tests that could
work whether the curve is implemented via Mbed TLS code or via a
driver.
After this commit, any attempt to use parameters that are not
recognized fails with NOT_SUPPORTED, whether a curve with the
specified size might plausibly exist or not, because "might plausibly
exist" is not something Mbed TLS can determine.
To keep returning INVALID_ARGUMENT when importing an ECC key with an
explicit "bits" attribute that is inconsistent with the size of the
key material, this commit changes the way mbedtls_ecc_group_of_psa()
works: it now works on a size in bits rather than bytes, with an extra
flag indicating whether the bit-size must be exact or not.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Move key buffer allocation from psa_import_key_into_slot()
function up to the two functions calling it.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Rename ECP key import function before to move
it to psa_crypto_ecp.c to adapt to the naming
of exported functions in psa_crypto_ecp.c.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Rename psa_import_rsa_key to mbedtls_psa_rsa_import_key to
align its name with the naming conventions of exported
functions in psa_crypto_rsa.c.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Change psa_export_(public_)key_internal signature to
the signature of an export_(public_)key driver entry
point.
This is a preparatory commit to be able to call the
software implementations as a driver.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Disentangle public export and export code to be
able to move the call to export and public
export operations to the driver wrapper.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Align naming of export internal functions with the way
other psa_crypto.c internal functions are named, ending
with _internal.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Rename psa_export_rsa/ecp_key to
mbedtls_psa_rsa/ecp_export_key before to move them to
RSA/ECP specific PSA crypto C files.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Rename psa_load_rsa/ecp_representation to
mbedtls_psa_rsa/ecp_load_representation
before to move them in their RSA/ECP
specific PSA crypto C modules.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Change psa_driver_wrapper_export_public_key() signature
to the signature of an export_public_key driver entry point.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
In the course of the development of the PSA unified
driver interface, the validate_key entry point for
opaque drivers has been removed and replaced by an
import_key entry point. This commit takes into account
this change of specification.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
In key slots containing the description of a key of a
dynamically registered Secure Element (SE), store the
key slot number in a key context as defined in the
PSA driver interface for opaque drivers.
That way transparent key data and slot numbers are
, in a key slot, both stored in a dynamically allocated
buffer. The `data` union in structures of type
psa_key_slot_t to distinguish between the storage of
transparent key data and slot numbers is consequently
not necessary anymore and thus removed.
This alignement of some part of the code dedicated to
dynamically registered SE with the PSA driver interface
specification is done to ease the support of both
dynamically registered and statically defined secure
elements.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Rename the enum constants TLS12_PRF_xxx, which are declared in a
public header but not intended for use in application code, to start
with MBEDTLS_PSA_.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.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>
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>
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>
The previous revision of guards around SHA_224 and SHA_384 were not
correct. This set of changes ensures the proper code is available
when SHA_224 and SHA_384 are enabled for use.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
Removed unecessary checks on the ALG_SHA_224 and ALG_SHA_384 since
those are handled in config_psa.h by ensuring the correct _C is
included. Reformatted config_psa.h to be alphabetical and made the
assignments line up correctly for consistency. Fixed the guards
for ALG_SHA_224 and ALG_SHA_384 to be correct in the crypto library
source.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
This set of changes converts all the previous MBEDTLS_*_C guards in
the crypto library to use the new MBEDTLS_PSA_BUILTIN_ guards for
hash algorithms.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
The return values of the functions are updated in the documetation.
All possible return values are added including nested functions' return
values. The values which cannot be returned are removed.
Signed-off-by: gabor-mezei-arm <gabor.mezei@arm.com>
If the file is read correctly, but it contains data that isn't valid,
the crypto storage code returns PSA_ERROR_DATA_INVALID.
The PSA_ERROR_DATA_CORRUPT and PSA_ERROR_STORAGE_FAILURE error codes are
replaced with PSA_ERROR_DATA_INVALID, except in the ITS subsystem.
Signed-off-by: gabor-mezei-arm <gabor.mezei@arm.com>
Support using HMAC_DRBG instead of CTR_DRBG in the PSA subsystem.
Use HMAC_DRBG if CTR_DRBG is available. Choose between SHA-256 and
SHA-512 based on availability.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Implement support for MBEDTLS_PSA_CRYPTO_EXTERNAL_RNG.
For test purposes, write an implementation that uses libc rand().
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Create wrapper functions around calls to CTR_DRBG and around calls to
entropy+DRBG. This is in preparation for allowing alternative DRBG
implementations that use the Mbed TLS entropy module, or complete RNG
implementations that bypass the entropy module as well.
This is purely a refactoring commit. No behavior change.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Rename functions to get a key slot:
. to make their naming more consistent
. to emphasize that those functions set a lock on the
key slot they return to protect it from being wiped
out and re-used while some part of the library
is accessing it.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Move the return of the identifier of a created key from
psa_start_key_creation() to psa_finish_key_creation().
That way in case of creation error, it is less likely to
return the identifier that was temporarily assigned to
the key while trying to create it.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Decrement the slot access count in psa_finish_key_creation()
when the finalization succeeds instead of in functions calling
psa_finish_key_creation(). That way the decrementation cannot
be forgotten and it reduces the code size.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Special handling of volatile key identifiers is not
needed eventually, they can be handled just as
key identifier in the vendor range.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Moved new check_crypto_config.h file from include/psa to library
directory and the file is now included from *.c instead of the
crypto_config.h file. Fixed guards in PSA crypto library based
on review comments for new PSA crypto config features.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
There was one lingering MBEDTLS_MD_C that needed to be removed since
it is no longer needed.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
Revised the placement of various new MBEDTLS_PSA_BUILTIN_xxx
guards based on review comments. Corrected guards in psa
test driver to use _ACCEL version instead of _BUILTIN version.
Updated check_config_psa.h to include additional dependency checks
for more algorithms. Renamed some of the new tests to be a little
more clear on the purpose.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
For consistency across the code base, prefer
persistent over permanent to qualify a key
stored in persistent storage.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
When psa_close/destroy/purge_key is called, do not
reset a key slot containing the description
of a persistent key if it is currently accessed.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Add key slot access counter to be able to
state if a key slot containing the description
of a permanent key can be reset or reset
and re-used.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
The identifier of keys created/registred should be in
the application range.
This is by spec for key creation.
This may change for registered key.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
The lifetime of key attributes now encodes whether a key is
volatile/persistent or not AND its location.
Fix PSA code where the fact that the lifetime encodes
the key location was not taken into account properly.
Fix the impacted tests and add two non regression tests.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Volatile key identifiers in the vendor range are
reserved to volatile keys thus don't allow them
for persistent keys when creating a key.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Move all the PSA crypto APIs using key handles
to use key identifiers but psa_key_open() and
psa_key_close(). This is done without modifying
any test as key handles and key identifiers are
now the same.
Update the library modules using PSA crypto APIs
to get rid of key handles.
Programs and unit tests are updated to not use
key handles in subsequent commits, not in this
one.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Define psa_key_handle_t to be equal to
mbedtls_svc_key_id_t. Make the handle of a persistent
key be equal to its key identifier. For volatile keys,
make the key handle equal to the volatile key
identifier of the created volatile key.
The unit tests are modified just to make them compile
not to make them run successfully. They are fixed in
the subsequent commits.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Volatile key identifiers are introduced in
PSA Crypto API v1.0.0. They are returned by the APIs
when importing or generating or deriving a volatile key.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Updated macros in config_psa.h that used ECC_xxx to use KEY_TYPE_ECC_xxx
per comments from review. Implemented a check_config_psa.h to help with
dependency checking of features enabled in config_psa.h. Added
check_config_psa.h to visual studio project.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
In the original attempt to add RSA support to PSA crypto config was too
generic. This set of changes adds support for the following RSA features:
PSA_WANT_ALG_RSA_PKCS1V15_CRYPT, PSA_WANT_ALG_RSA_PKCS1V15_SIGN,
PSA_WANT_ALG_RSA_OAEP, PSA_WANT_ALG_RSA_PSS, PSA_WANT_KEY_TYPE_RSA_KEY_PAIR,
and PSA_WANT_KEY_TYPE_RSA_PUBLIC_KEY.
There were also some updates to ensure the proper inclusion of PSA crypto
library code when certain features are enabled. These updates were made to
address warnings and errors in builds from the new tests for these
features being added for PSA crypto configuration.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
The KEY_TYPE_ECC_KEY_PAIR and KEY_TYPE_ECC_PUBLIC_KEY were previously
being guarded by MBEDTLS_ECP_C in the PSA crypto library code. This change
moves it to the new MBEDTLS_PSA_BUILTIN_xxx and separates KEY_PAIR
and PUBLIC_KEY as needed. Tests have also been added to validate the new
settings.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
This phase adds in support for the following features being
added to the list of features that can be configured in the
include/psa/crypto_config.h header file using the PSA_WANT_ALG_xxx
macros: ECDH, HMAC, HKDF, and RSA. These changes include changes to
the PSA crypto library to use the appropriate new guards that
will allow the feature to be compiled in or out either using
new PSA_WANT_ALG_xxx or the previous MBEDTLS_xxx macros.
For HKDF and HMAC, most of the PSA library code did not have a
specific matching MBEDTLS_xxx macro for that feature, but was instead
using the generic dependent MBEDTLS_MD_C macro. The ECDH and RSA
features more closely aligned with a direct replacement with a similar
macro.
The new tests for RSA, HMAC, and HKDF would normally unset additional
dependent macros, but when attempting to implement that level of
testing it required removal of too many core features like MD_C, PK_C,
ECP_C and other low level features. This may point to additional phases of
work to complete the transition of these features to the new model.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
With key usage based on key identifiers and not
key handles (openless APIs), volatile keys will
also have a key identifier. Thus, isolate key
identifier validation from key persistence
validation to clarify that key identifiers
are not specific to persistent keys.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
* zero key buffer on failure
* readability improvements
* psa_finish_key_creation adjustment after removing import_key_into_slot
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
In the PSA crypto library, the code for verification of ECDSA is the same for
both MBEDTLS_PSA_BUILTIN_ALG_ECDSA and
MBEDTLS_PSA_BUILTIN_ALG_DETERMINISTIC_ECDSA. So, the guards should allow for
either one to enable the code blocks. The original implementation only had
the check for ECDSA. In order to make this work, config_psa.h was updated
to ensure when MBEDTLS_CRYPTO_CONFIG is disabled, the setting for DETERMINISTIC
is only updated if MBEDTLS_ECDSA_C is also enabled.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
Now that there's a validate_key entry point for drivers, it becomes
much more important to separate the import action (where a key needs
to be validated) from the load action (where a key has been
previously validated, and thus re-validating it would be a waste of
time).
This also exposes why not storing the 'bits' attribute persistently
was a bad idea. The only reason there's a rather large function to
detect bit size is because loading from persistent storage requires
it.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
There's no need for calling export-and-import when the key is
guaranteed to have been stored in export representation.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
The psa crypto library was generically using PSA_WANT_ALG_xxx, but
should have been using the correct MBEDTLS_PSA_BUILTIN_ALG_xxx
definition since that code is the builtin version. There were also
a couple of spots that needed to ensure the code block was enabled
for either ECDSA or DETERMINISTIC_ECDSA, not just one of them.
Fixed all the new ALG_ECDSA_DETERMINISTIC names to be
ALG_DETERMINISTIC_ECDSA instead.
Fixed test to use correct definitions.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
* Stores bits in psa_persistent_key_storage_format.
* psa_load_persistent_key_into_slot still imports plaintext keys which
ensures that the bits value gets set.
* Updates key specification to match new implementation.
* Expands persistent store and load tests with to check for bits
attribute.
* Removes bits storage from psa_se_key_data_storage_t.
Signed-off-by: Torstein Nesse <torstein.nesse@silabs.com>
* #3742 After input of a key as SECRET in the derivation, allow the
derivation result to be used as key.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
* #3741 Allow key agreement inside derivation with a key that's allowed
for the relevant agreement.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
In psa_generate_key_internal() for ECC keys, remove the check that the
bit-size according to Mbed TLS is equal to the requested bit-size.
This check was necessary back when the PSA API encoded curves and key
sizes independently, in order to reject combinations such as SECP256R1
with a 512-bit size. Since the curve encoding changed to specifying a
curve family and a size separately, the Mbed TLS curve id (grp_id) and
the curve data (curve_info) are now determined from the size, and
checking that (curve_info->bit_size == bits) is now only a redundant
sanity check.
This check is actually buggy, because PSA Crypto and Mbed TLS don't
have exactly the same notion of key size. PSA thinks Curve25519 is
255-bit and secp224k1 is 225-bit, but Mbed TLS thinks they're 256-bit
and 224-bit respectively. Removing the check allows key generation to
work for these curves.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
... as opposed to PSA_ERROR_BAD_STATE.
The spec on psa_cipher_finish() states that PSA_ERROR_INVALID_ARGUMENT
should be returned when:
"The total input size passed to this operation is not valid for this
particular algorithm. For example, the algorithm is a based on block
cipher and requires a whole number of blocks, but the total input size
is not a multiple of the block size."
Currently, there is a distinction between encryption and decryption
on whether INVALID_ARGUMENT or BAD_STATE is returned, but this is not
a part of the spec.
This fix ensures that PSA_ERROR_INVALID_ARGUMENT is returned
consistently on invalid cipher input sizes.
Signed-off-by: Fredrik Strupe <fredrik.strupe@silabs.com>
Initial changes to PSA crypto core to support configuration
of ECDSA algorithm using PSA crypto configuration mechanism.
Guards using MBEDTLS_ECDSA_C and MBEDTLS_ECDSA_DETERMINISTIC have
been changed to be based off PSA_WANT_ALG_ECDSA and
PSA_WANT_ALG_ECDSA_DETERMINISTIC. Added new tests to all.sh to
confirm new settings are working properly. Current code does not
pass the tests since built in signature verification is not in place.
Signed-off-by: John Durkop <john.durkop@fermatsoftware.com>
* Reworked the cipher context once again to be more robustly defined
* Removed redundant memset
* Unified behaviour on failure between driver and software in cipher_finish
* Cipher test driver setup function now also returns early when its status
is overridden, like the other test driver functions
* Removed redundant test cases
* Added bad-order checking to verify the driver doesn't get called where
the spec says it won't.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
As pointed out by Ronald. The key slot is populated using
get_key_from_slot, and after calling the driver the slot is
validated to not contain an external key, so calling
get_transparent_key is superfluous.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
Once an operation has been 'accepted' by a driver, the remainder is bound
to the same driver, since driver-specific context structs cannot be shared.
This provides a pretty good gate mechanism for the fallback logic, too.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
This is a more natural place for handling the drivers belonging to the
'previous' SE driver spec. It makes for a cleaner psa_crypto.c, and
potentially an easier overview of how to migrate from the old SE driver
interface to the 'opaque accelerator' interface.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
As a result, the copyright of contributors other than Arm is now
acknowledged, and the years of publishing are no longer tracked in the
source files.
Also remove the now-redundant lines declaring that the files are part of
MbedTLS.
This commit was generated using the following script:
# ========================
#!/bin/sh
# Find files
find '(' -path './.git' -o -path './3rdparty' ')' -prune -o -type f -print | xargs sed -bi '
# Replace copyright attribution line
s/Copyright.*Arm.*/Copyright The Mbed TLS Contributors/I
# Remove redundant declaration and the preceding line
$!N
/This file is part of Mbed TLS/Id
P
D
'
# ========================
Signed-off-by: Bence Szépkúti <bence.szepkuti@arm.com>
Since it is being dereferenced by free on exit it should be inited to NULL.
Also added a small test that would trigger the issue.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
* return is treated as a function call
* space between opening and closing parentheses
* remove whiteline between assignment and checking of same variable
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
* No need to check for NULL before free'ing
* No need to reset variables that weren't touched
* Set output buffer to zero if key output fails
* Document internal functions and rearrange order of input arguments to
better match other functions.
* Clean up Montgomery fix to be less verbose code
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
PSA Crypto was checking the byte length of a to-be-imported public ECP key
against the expected length for Weierstrass keys, forgetting that
Curve25519/Curve448 exists.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
Avoids stack-allocating a key slot during ECDH, and mock-attaching a
key to a key slot during key import.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
* No null-check before calling free
* Close memory leak
* No need for double check of privkey validity
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
* Allocate internal representation contexts on the heap (i.e. don't change
where they're being allocated)
* Unify load_xxx_representation in terms of allocation and init behaviour
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
* Updated wording
* Split out buffer allocation to a convenience function
* Moved variable declarations to beginning of their code block
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
Now that both ECP and RSA keys are represented in export representation,
they can be treated more uniformly.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
Change to on-demand loading of the internal representation when required
in order to call an mbed TLS cryptography API.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
Change to on-demand loading of the internal representation when required
in order to call an mbed TLS cryptography API.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
In preparation for the implementation of the accelerator APIs. This is
ramping up to the goal of only storing the export representation in the
key slot, and not keeping the crypto implementation-specific representations
around.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
PSA_ALG_ECB_NO_PADDING came in to the PSA Crypto API spec v1.0.0, but
was not implemented yet in the mbed TLS implementation.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
Changed PSA core (and PKWrite) from reaching into MPI to using the proper
ecp function to fetch a private key.
Added changelog.
Signed-off-by: Steven Cooreman <steven.cooreman@silabs.com>
Rename PSA_ECC_CURVE_xxx to PSA_ECC_FAMILY_xxx, also rename
PSA_KEY_TYPE_GET_CURVE to PSA_KEY_TYPE_ECC_GET_FAMILY and rename
psa_ecc_curve_t to psa_ecc_family_t. Old defines are provided in
include/crypto_compat.h for backward compatibility.
Signed-off-by: Paul Elliott <paul.elliott@arm.com>
In library source files, include "common.h", which takes care of
including "mbedtls/config.h" (or the alternative MBEDTLS_CONFIG_FILE)
and other things that are used throughout the library.
FROM=$'#if !defined(MBEDTLS_CONFIG_FILE)\n#include "mbedtls/config.h"\n#else\n#include MBEDTLS_CONFIG_FILE\n#endif' perl -i -0777 -pe 's~\Q$ENV{FROM}~#include "common.h"~' library/*.c 3rdparty/*/library/*.c scripts/data_files/error.fmt scripts/data_files/version_features.fmt
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Also normalize the first line of the copyright headers.
This commit was generated using the following script:
# ========================
#!/bin/sh
# Find scripts
find -path './.git' -prune -o '(' -name '*.c' -o -name '*.cpp' -o -name '*.fmt' -o -name '*.h' ')' -print | xargs sed -i '
# Normalize the first line of the copyright headers (no text on the first line of a block comment)
/^\/\*.*Copyright.*Arm/I {
i\
/*
s/^\// /
}
/Copyright.*Arm/I {
# Print copyright declaration
p
# Read the two lines immediately following the copyright declaration
N
N
# Insert Apache header if it is missing
/SPDX/! i\
* SPDX-License-Identifier: Apache-2.0\
*\
* Licensed under the Apache License, Version 2.0 (the "License"); you may\
* not use this file except in compliance with the License.\
* You may obtain a copy of the License at\
*\
* http://www.apache.org/licenses/LICENSE-2.0\
*\
* Unless required by applicable law or agreed to in writing, software\
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT\
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\
* See the License for the specific language governing permissions and\
* limitations under the License.
# Clear copyright declaration from buffer
D
}
'
# ========================
Signed-off-by: Bence Szépkúti <bence.szepkuti@arm.com>
On dual world platforms, we want to run the PK module (pk.c) on the NS
side so TLS can use PSA APIs via the PK interface. PK currently has a
hard dependency on mbedtls_ecc_group_to_psa() which is declared in
crypto_extra.h, but only defined in psa_crypto.c, which is only built
for the S side.
Without this change, dual world platforms get error messages like the
following.
[Error] @0,0: L6218E: Undefined symbol mbedtls_ecc_group_to_psa (referred from BUILD/LPC55S69_NS/ARM/mbed-os/features/mbedtls/mbed-crypto/src/pk.o)
Make mbedtls_ecc_group_to_psa() inline within crypto_extra.h so that it
is available to both NS and S world code.
Fixes#3300
Signed-off-by: Darryl Green <darryl.green@arm.com>
Signed-off-by: Jaeden Amero <jaeden.amero@arm.com>
Clear bits in mbedtls_ecc_group_to_psa() to avoid static analyzers and
possibly compilers from warning that bits may be used uninitialized in
certain code paths.
For example, if mbedtls_ecc_group_to_psa() were to be inlined in
crypto_extra.h, the following compiler warning is likely.
In file included from ../include/psa/crypto.h:3774:0,
from ../include/mbedtls/pk.h:49,
from pk.c:29:
pk.c: In function 'mbedtls_pk_wrap_as_opaque':
../include/psa/crypto_struct.h:460:33: error: 'bits' may be used uninitialized in this function [-Werror=maybe-uninitialized]
attributes->core.bits = (psa_key_bits_t) bits;
^~~~~~~~~~~~~~~~~~~~~
pk.c:608:12: note: 'bits' was declared here
size_t bits;
^~~~
Signed-off-by: Jaeden Amero <jaeden.amero@arm.com>
If psa_key_agreement_ecdh fails, there may be output that leaks
sensitive information in the output buffer. Zeroize it.
If this is due to an underlying failure in the ECDH implementation, it
is currently not an issue since both the traditional Mbed TLS/Crypto
implementation and Everest only write to the output buffer once every
intermediate step has succeeded, but zeroizing is more robust. If this
is because the recently added key size check fails, a leak could be a
serious issue.
All key types now have an encoding on 32 bits where the bottom 16 bits
are zero. Change to using 16 bits only.
Keep 32 bits for key types in storage, but move the significant
half-word from the top to the bottom.
Likewise, change EC curve and DH group families from 32 bits out of
which the top 8 and bottom 16 bits are zero, to 8 bits only.
Reorder psa_core_key_attributes_t to avoid padding.
Remove the values of curve encodings that are based on the TLS registry
and include the curve size, keeping only the new encoding that merely
encodes a curve family in 8 bits.
Keep the old constant names as aliases for the new values and
deprecate the old names.
Define constants for ECC curve families and DH group families. These
constants have 0x0000 in the lower 16 bits of the key type.
Support these constants in the implementation and in the PSA metadata
tests.
Switch the slot management and secure element driver HAL tests to the
new curve encodings. This requires SE driver code to become slightly
more clever when figuring out the bit-size of an imported EC key since
it now needs to take the data size into account.
Switch some documentation to the new encodings.
Remove the macro PSA_ECC_CURVE_BITS which can no longer be implemented.
Change the representation of psa_ecc_curve_t and psa_dh_group_t from
the IETF 16-bit encoding to a custom 24-bit encoding where the upper 8
bits represent a curve family and the lower 16 bits are the key size
in bits. Families are based on naming and mathematical similarity,
with sufficiently precise families that no two curves in a family have
the same bit size (for example SECP-R1 and SECP-R2 are two different
families).
As a consequence, the lower 16 bits of a key type value are always
either the key size or 0.
Don't rely on the bit size encoded in the PSA curve identifier, in
preparation for removing that.
For some inputs, the error code on EC key creation changes from
PSA_ERROR_INVALID_ARGUMENT to PSA_ERROR_NOT_SUPPORTED or vice versa.
There will be further such changes in subsequent commits.
If psa_mac_finish_internal fails (which can only happen due to bad
parameters or hardware problem), the error code was converted to
PSA_ERROR_INVALID_SIGNATURE if the uninitialized stack variable
actual_mac happened to contain the expected MAC. This is a minor bug
but it may be possible to leverage it as part of a longer attack path
in some scenarios.
Reported externally. Found by static analysis.
Rename some macros and functions related to signature which are
changing as part of the addition of psa_sign_message and
psa_verify_message.
perl -i -pe '%t = (
PSA_KEY_USAGE_SIGN => PSA_KEY_USAGE_SIGN_HASH,
PSA_KEY_USAGE_VERIFY => PSA_KEY_USAGE_VERIFY_HASH,
PSA_ASYMMETRIC_SIGNATURE_MAX_SIZE => PSA_SIGNATURE_MAX_SIZE,
PSA_ASYMMETRIC_SIGN_OUTPUT_SIZE => PSA_SIGN_OUTPUT_SIZE,
psa_asymmetric_sign => psa_sign_hash,
psa_asymmetric_verify => psa_verify_hash,
); s/\b(@{[join("|", keys %t)]})\b/$t{$1}/ge' $(git ls-files . ':!:**/crypto_compat.h')
When registering a key in a secure element, go through the transaction
mechanism. This makes the code simpler, at the expense of a few extra
storage operations. Given that registering a key is typically very
rare over the lifetime of a device, this is an acceptable loss.
Drivers must now have a p_validate_slot_number method, otherwise
registering a key is not possible. This reduces the risk that due to a
mistake during the integration of a device, an application might claim
a slot in a way that is not supported by the driver.
If none of the inputs to a key derivation is a
PSA_KEY_DERIVATION_INPUT_SECRET passed with
psa_key_derivation_input_key(), forbid
psa_key_derivation_output_key(). It usually doesn't make sense to
derive a key object if the secret isn't itself a proper key.
Allow a direct input as the SECRET input step in a key derivation, in
addition to allowing DERIVE keys. This makes it easier for
applications to run a key derivation where the "secret" input is
obtained from somewhere else. This makes it possible for the "secret"
input to be empty (keys cannot be empty), which some protocols do (for
example the IV derivation in EAP-TLS).
Conversely, allow a RAW_DATA key as the INFO/LABEL/SALT/SEED input to a key
derivation, in addition to allowing direct inputs. This doesn't
improve security, but removes a step when a personalization parameter
is stored in the key store, and allows this personalization parameter
to remain opaque.
Add test cases that explore step/key-type-and-keyhood combinations.
The signature must have exactly the same length as the key, it can't
be longer. Fix#258
If the signature doesn't have the correct size, that's an invalid
signature, not a problem with an output buffer size. Fix the error code.
Add test cases.
In psa_asymmetric_sign, immediately reject an empty signature buffer.
This can never be right.
Add test cases (one RSA and one ECDSA).
Change the SE HAL mock tests not to use an empty signature buffer.
Zero-length keys are rejected at creation time, so we don't need any
special handling internally.
When exporting a key, we do need to take care of the case where the
output buffer is empty, but this is easy: an empty output buffer is
never valid.
At the end of `psa_hmac_setup_internal()`, the ipad is cleared.
However, the size that was given to clear was `key_len` which is larger
than the size of `ipad`.
In psa_import_key, the key bits value was uninitialized before
calling the secure element driver import function. There is a
potential issue if the driver returns PSA_SUCCESS without setting
the key bits. This shouldn't happen, but shouldn't be discounted
either, so we initialize the key bits to an invalid issue.
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
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
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.
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.
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".
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.
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.
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).
Add a slot_number field to psa_key_attributes_t and getter/setter
functions. Since slot numbers can have the value 0, indicate the
presence of the field via a separate flag.
In psa_get_key_attributes(), report the slot number if the key is in a
secure element.
When creating a key, for now, applications cannot choose a slot
number. A subsequent commit will add this capability in the secure
element HAL.
Add infrastructure for internal, external and dual-use flags, with a
compile-time check (if static_assert is available) to ensure that the
same numerical value doesn't get declared for two different purposes
in crypto_struct.h (external or dual-use) and
psa_crypto_core.h (internal).
mbedtls_ctr_drbg_random can only return up to
MBEDTLS_CTR_DRBG_MAX_REQUEST (normally 1024) bytes at a time. So if
more than that is requested, call mbedtls_ctr_drbg_random in a loop.
When psa_generate_random fails, psa_generate_key_internal frees the
key buffer but a the pointer to the now-freed buffer in the slot. Then
psa_generate_key calls psa_fail_key_creation which sees the pointer
and calls free() again.
This bug was introduced by ff5f0e7221
"Implement atomic-creation psa_{generate,generator_import}_key" which
changed how psa_generate_key() cleans up on errors. I went through the
code and could not find a similar bug in cleanup on an error during
key creation.
Fix#207
Conflict resolution:
* `scripts/config.pl`:
Take the exclusion of `MBEDTLS_PSA_CRYPTO_SE_C` from the API branch.
Take the removal of `MBEDTLS_PSA_CRYPTO_STORAGE_ITS_C` (obsolete) from
the development branch.
* `tests/scripts/all.sh`:
Multiple instances of factoring a sequence of `config.pl` calls into
a mere `config.pl baremetal` in the development branch, and a change in
the composition of `baremetal` in the API branch. In each case, take the
version from development.
* `tests/suites/test_suite_psa_crypto_slot_management.function`:
A function became non-static in development and disappeared in the API
branch. Keep the version from the API branch. Functions need to be
non-static if they're defined but unused in some configurations,
which is not the case for any function in this file at the moment.
* `tests/suites/test_suite_psa_crypto.function`:
Consecutive changes in the two branches, reconciled.
The flag to mark key slots as allocated was introduced to mark slots
that are claimed and in use, but do not have key material yet, at a
time when creating a key used several API functions: allocate a slot,
then progressively set its metadata, and finally create the key
material. Now that all of these steps are combined into a single
API function call, the notion of allocated-but-not-filled slot is no
longer relevant. So remove the corresponding flag.
A slot is occupied iff there is a key in it. (For a key in a secure
element, the key material is not present, but the slot contains the
key metadata.) This key must have a type which is nonzero, so use this
as an indicator that a slot is in use.
There is now a field for the key size in the key slot in memory. Use
it.
This makes psa_get_key_attributes() marginally faster at the expense
of memory that is available anyway in the current memory layout (16
bits for the size, 16 bits for flags). That's not the goal, though:
the goal is to simplify the code, in particular to make it more
uniform between transparent keys (whose size can be recomputed) and
keys in secure elements (whose size cannot be recomputed).
For keys in a secure element, the bit size is now saved by serializing
the type psa_key_bits_t (which is an alias for uint16_t) rather than
size_t.
Change the type of key slots in memory to use
psa_core_key_attributes_t rather than separate fields. The goal is to
simplify some parts of the code. This commit only does the mechanical
replacement, not the substitution.
The bit-field `allocate` is now a flag `PSA_KEY_SLOT_FLAG_ALLOCATED`
in the `flags` field.
Write accessor functions for flags.
Key slots now contain a bit size field which is currently unused.
Subsequent commits will make use of it.
65528 bits is more than any reasonable key until we start supporting
post-quantum cryptography.
This limit is chosen to allow bit-sizes to be stored in 16 bits, with
65535 left to indicate an invalid value. It's a whole number of bytes,
which facilitates some calculations, in particular allowing a key of
exactly PSA_CRYPTO_MAX_STORAGE_SIZE to be created but not one bit
more.
As a resource usage limit, this is arguably too large, but that's out
of scope of the current commit.
Test that key import, generation and derivation reject overly large
sizes.
Move the "core attributes" to a substructure of psa_key_attribute_t.
The motivation is to be able to use the new structure
psa_core_key_attributes_t internally.
For a key in a secure element, save the bit size alongside the slot
number.
This is a quick-and-dirty implementation where the storage format
depends on sizeof(size_t), which is fragile. This should be replaced
by a more robust implementation before going into production.
Add a parameter to the key import method of a secure element driver to
make it report the key size in bits. This is necessary (otherwise the
core has no idea what the bit-size is), and making import report it is
easier than adding a separate method (for other key creation methods,
this information is an input, not an output).
Nothing has been saved to disk yet, but there is stale data in
psa_crypto_transaction. This stale data should not be reused, but do
wipe it to reduce the risk of it mattering somehow in the future.
Introduce a new function psa_get_transparent_key which returns
NOT_SUPPORTED if the key is in a secure element. Use this function in
functions that don't support keys in a secure element.
After this commit, all functions that access a key slot directly via
psa_get_key_slot or psa_get_key_from_slot rather than via
psa_get_transparent_key have at least enough support for secure
elements not to crash or otherwise cause undefined behavior. Lesser
bad behavior such as wrong results or resource leakage is still
possible in error cases.
Pass information via a key attribute structure rather than as separate
parameters to psa_crypto_storage functions. This makes it easier to
maintain the code when the metadata of a key evolves.
This has negligible impact on code size (+4B with "gcc -Os" on x86_64).
Key creation and key destruction for a key in a secure element both
require updating three pieces of data: the key data in the secure
element, the key metadata in internal storage, and the SE driver's
persistent data. Perform these actions in a transaction so that
recovery is possible if the action is interrupted midway.
When creating a key with a lifetime that places it in a secure
element, retrieve the appropriate driver table entry.
This commit doesn't yet achieve behavior: so far the code only
retrieves the driver, it doesn't call the driver.
The psa_tls12_prf_set_seed() and psa_tls12_prf_set_label() functions did
not work on platforms where malloc(0) returns NULL.
It does not affect the TLS use case but these PRFs are used in other
protocols as well and might not be used the same way. For example EAP
uses the TLS PRF with an empty secret. (This would not trigger the bug,
but is a strong indication that it is not safe to assume that certain
inputs to this function are not zero length.)
The conditional block includes the memcpy() call as well to avoid
passing a NULL pointer as a parameter resulting in undefined behaviour.
The current tests are already using zero length label and seed, there is
no need to add new test for this bug.
Secure element support has its own source file, and in addition
requires many hooks in other files. This is a nontrivial amount of
code, so make it optional (but default on).
PSA_ERROR_BAD_STATE means that the function was called on a context in a
bad state.
This error is something that can't happen while only using the PSA API and
therefore a PSA_ERROR_CORRUPTION_DETECTED is a more appropriate error
code.
The macro initialiser might leave bytes in the union unspecified.
Zeroising it in setup makes sure that the behaviour is the same
independently of the initialisation method used.
The TLS 1.2 pseudorandom function does a lot of distinct HMAC operations
with the same key. To save the battery and CPU cycles spent on
calculating the paddings and hashing the inner padding, we keep the
hash context in the status right after the inner padding having been
hashed and clone it as needed.
Technically we could have reused the old one for the new API, but then
we had to set an extra field during setup. The new version works when
all the fields that haven't been set explicitely are zero-initialised.
