Mbed TLS before 3.6.4 allows a use-after-free in certain situations of applications that are developed in accordance with the documentation. The function mbedtls_x509_string_to_names() takes a head argument that is documented as an output argument. The documentation does not suggest that the function will free that pointer; however, the function does call mbedtls_asn1_free_named_data_list() on that argument, which performs a deep free(). As a result, application code that uses this function (relying only on documented behavior) is likely to still hold pointers to the memory blocks that were freed, resulting in a high risk of use-after-free or double-free. In particular, the two sample programs x509/cert_write and x509/cert_req are affected (use-after-free if the san string contains more than one DN).
Mbed TLS before 3.6.4 has a NULL pointer dereference because mbedtls_asn1_store_named_data can trigger conflicting data with val.p of NULL but val.len greater than zero.
In MbedTLS 3.3.0 before 3.6.4, mbedtls_lms_verify may accept invalid signatures if hash computation fails and internal errors go unchecked, enabling LMS (Leighton-Micali Signature) forgery in a fault scenario. Specifically, unchecked return values in mbedtls_lms_verify allow an attacker (who can induce a hardware hash accelerator fault) to bypass LMS signature verification by reusing stale stack data, resulting in acceptance of an invalid signature. In mbedtls_lms_verify, the return values of the internal Merkle tree functions create_merkle_leaf_value and create_merkle_internal_value are not checked. These functions return an integer that indicates whether the call succeeded or not. If a failure occurs, the output buffer (Tc_candidate_root_node) may remain uninitialized, and the result of the signature verification is unpredictable. When the software implementation of SHA-256 is used, these functions will not fail. However, with hardware-accelerated hashing, an attacker could use fault injection against the accelerator to bypass verification.
In MbedTLS 3.3.0 before 3.6.4, mbedtls_lms_import_public_key does not check that the input buffer is at least 4 bytes before reading a 32-bit field, allowing a possible out-of-bounds read on truncated input. Specifically, an out-of-bounds read in mbedtls_lms_import_public_key allows context-dependent attackers to trigger a crash or limited adjacent-memory disclosure by supplying a truncated LMS (Leighton-Micali Signature) public-key buffer under four bytes. An LMS public key starts with a 4-byte type indicator. The function mbedtls_lms_import_public_key reads this type indicator before validating the size of its input.
Mbed TLS before 3.6.4 has a PEM parsing one-byte heap-based buffer underflow, in mbedtls_pem_read_buffer and two mbedtls_pk_parse functions, via untrusted PEM input.
Mbed TLS before 2.28.10 and 3.x before 3.6.3, on the client side, accepts servers that have trusted certificates for arbitrary hostnames unless the TLS client application calls mbedtls_ssl_set_hostname.
An issue was discovered in Mbed TLS 3.x before 3.6.1. With TLS 1.3, when a server enables optional authentication of the client, if the client-provided certificate does not have appropriate values in if keyUsage or extKeyUsage extensions, then the return value of mbedtls_ssl_get_verify_result() would incorrectly have the MBEDTLS_X509_BADCERT_KEY_USAGE and MBEDTLS_X509_BADCERT_KEY_USAGE bits clear. As a result, an attacker that had a certificate valid for uses other than TLS client authentication would nonetheless be able to use it for TLS client authentication. Only TLS 1.3 servers were affected, and only with optional authentication (with required authentication, the handshake would be aborted with a fatal alert).
An issue was discovered in Mbed TLS before 2.28.9 and 3.x before 3.6.1, in which the user-selected algorithm is not used. Unlike previously documented, enabling MBEDTLS_PSA_HMAC_DRBG_MD_TYPE does not cause the PSA subsystem to use HMAC_DRBG: it uses HMAC_DRBG only when MBEDTLS_PSA_CRYPTO_EXTERNAL_RNG and MBEDTLS_CTR_DRBG_C are disabled.
In Mbed TLS 3.3.0 through 3.5.2 before 3.6.0, a malicious client can cause information disclosure or a denial of service because of a stack buffer over-read (of less than 256 bytes) in a TLS 1.3 server via a TLS 3.1 ClientHello.
An issue was discovered in Mbed TLS 2.18.0 through 2.28.x before 2.28.8 and 3.x before 3.6.0, and Mbed Crypto. The PSA Crypto API mishandles shared memory.