mbedtls_md_setup() allocates a hash-specific context and then, if
requested, an extra HMAC context. If the second allocation failed, the
hash context was not freed.
Fix this by ensuring that the mbedtls_md_context_t object is always in
a consistent state, in particular, that the md_info field is always
set. For robustness, ensure that the object is in a consistent state
even on errors (other than BAD_INPUT_DATA if the object was not in a
consistent state on entry).
Fix#3486
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.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>
The client previously reproted the offered ciphersuites through
their numerical identifier only, while the server reported them
through their name.
This commit modifies the debug output on client and server to
both use the format `ID (NAME)` for the ciphersuites.
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>
Run some self-test both for a short Weierstrass curve and for a
Montgomery curve, if the build-time configuration includes a curve of
both types. Run both because there are significant differences in the
implementation.
The test data is suitable for Curve25519.
Signed-off-by: Gilles Peskine <gilles.peskine@arm.com>
The constants used in the test worked with every supported curve
except secp192k1. For secp192k1, the "N-1" exponent was too large.
Signed-off-by: Gilles Peskine <gilles.peskine@arm.com>
For some curves (semi-coincidentally, short Weierstrass curves), the
ECP module calculates some group parameters dynamically. Build the
code to calculate the parameters only if a relevant curve is enabled.
This fixes an unused function warning when building with only
Montgomery curves.
Signed-off-by: Gilles Peskine <gilles.peskine@arm.com>
Replace the now-redundant internal curve type macros ECP_xxx by the
macros MBEDTLS_ECP__xxx_ENABLED which are declared in ecp.h.
Signed-off-by: Gilles Peskine <gilles.peskine@arm.com>
Document that mbedtls_ecp_muladd and mbedtls_ecp_muladd_restartable
are only implemented on short Weierstrass curves.
Exclude these functions at build time if no short Weierstrass curve
is included in the build. Before, these functions failed to compile in
such a configuration.
Signed-off-by: Gilles Peskine <gilles.peskine@arm.com>
Address remaining PR comments for #2118
- Add ChangeLog.d/x509write_csr_heap_alloc.txt.
- Fix parameter alignment per Gille's recommendation.
- Update comments to more explicitly describe the manipulation of buf.
- Replace use of `MBEDTLS_MPI_MAX_SIZE` as `sig` buffer size for
call to `x509write_csr_der_internal()` with more intuitive
`MBEDTLS_PK_SIGNATURE_MAX_SIZE`.
- Update `mbedtls_x509write_csr_der()` to return
`MBEDTLS_ERR_X509_ALLOC_FAILED` on mbedtls_calloc error.
Signed-off-by: Simon Leet <simon.leet@microsoft.com>
Using a stack-buffer with a size > 2K could easily produce a stack
overflow for an embedded device which has a limited stack size.
This commit dynamically allocates the large CSR buffer.
This commit avoids using a temporary buffer for storing the OIDs.
A single buffer is used:
a) OIDs are written backwards starting with the end of the buffer;
b) OIDs are memmove'd to the beginning of the buffer;
c) signature over this OIDs is computed and written backwards from the
end of the buffer;
d) the two memory regions are compacted.
Signed-off-by: Doru Gucea <doru-cristian.gucea@nxp.com>
