We were successful in adding transparent memory-poisoning testing, so
simplify to the real design decision we made.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
The finally implemented functions were significantly different from the
initial design idea, so update the document accordingly.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Update the guards generated by the wrapper generation script to use
!MBEDTLS_PSA_ASSUME_EXCLUSIVE_BUFFERS and regenerate the PSA test
wrappers.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Replace MBEDTLS_PSA_COPY_CALLER_BUFFERS with inverse:
!MBEDTLS_PSA_ASSUME_EXCLUSIVE_BUFFERS. This ensures that buffer
protection is enabled by default without any change to the Mbed TLS
config file.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
This is the issue that tracks incomplete support for buffer overlap, so
we should refer to it when we discuss partial support whenever
MBEDTLS_PSA_ASSUME_EXCLUSIVE_BUFFERS is set.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
* Fix a grammatical typo
* Mention shared memory
* Mention overlap support in the security section
* Improve wording
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Mention that arbitrary overlap is now supported, except whenever
MBEDTLS_PSA_ASSUME_EXCLUSIVE_BUFFERS is set.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Specifically:
* Clarify that passing shared buffers is now secure by default (not
newly supported)
* Remove spurious hyphen
* Clarify that we do not guarantee copying, but rather guarantee
protection, allowing us to implement this differently in future if
required.
* Mention both protection of inputs from modification and outputs from
exposure of intermediate results.
* Invert the config option, from an enable-option to a disable-option.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Fix errors in merge conflict resolution - change
psa_generate_random_internal() to return directly rather than jumping to
an exit label and restore the variable psa_status_t status.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
The sign/verify functions happily use the wrong algorithm, but the
encrypt/decrypt functions error out if the padding mode specifies V21.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
If the output buffer is NULL, it either:
* Does not need wiping because it is zero-length.
* Has failed allocation of a copy.
* Has not yet been written to as a copy hasn't been allocated.
In any of these circumstances, we should not try to write the buffer,
so perform a NULL check before wiping it.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Add buffer protection to:
* psa_sign_hash_start(), which takes an input buffer for the hash.
* psa_sign_hash_complete(), which takes an output buffer for the
calculated signature.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Use of volatile is more an internal implementation detail (ensuring
const-time) than part of the contract (the caller doesn't care about
volatile as such).
Signed-off-by: Janos Follath <janos.follath@arm.com>
We need the pointer, A, to be volatile, to ensure the reads happen. bits
does not need to be volatile.
Signed-off-by: Janos Follath <janos.follath@arm.com>
