diff options
author | Adam Harrison <adamdharrison@gmail.com> | 2023-07-06 06:37:41 -0400 |
---|---|---|
committer | Adam Harrison <adamdharrison@gmail.com> | 2023-07-06 06:37:41 -0400 |
commit | 9db10386430479067795bec66bb26343ff176ded (patch) | |
tree | 5ad0cf95abde7cf03afaf8f70af8549d46b09a46 /lib/mbedtls-2.27.0/ChangeLog | |
parent | 57092d80cb07fa1a84873769fa92165426196054 (diff) | |
download | lite-xl-plugin-manager-9db10386430479067795bec66bb26343ff176ded.tar.gz lite-xl-plugin-manager-9db10386430479067795bec66bb26343ff176ded.zip |
Removed old mbedtls, replacing with submodule.
Diffstat (limited to 'lib/mbedtls-2.27.0/ChangeLog')
-rw-r--r-- | lib/mbedtls-2.27.0/ChangeLog | 4392 |
1 files changed, 0 insertions, 4392 deletions
diff --git a/lib/mbedtls-2.27.0/ChangeLog b/lib/mbedtls-2.27.0/ChangeLog deleted file mode 100644 index 89572ca..0000000 --- a/lib/mbedtls-2.27.0/ChangeLog +++ /dev/null @@ -1,4392 +0,0 @@ -mbed TLS ChangeLog (Sorted per branch, date) - -= mbed TLS 2.27.0 branch released 2021-07-07 - -API changes - * Update AEAD output size macros to bring them in line with the PSA Crypto - API version 1.0 spec. This version of the spec parameterizes them on the - key type used, as well as the key bit-size in the case of - PSA_AEAD_TAG_LENGTH. - The old versions of these macros were renamed and deprecated as follows: - - PSA_AEAD_TAG_LENGTH -> PSA_AEAD_TAG_LENGTH_1_ARG - - PSA_AEAD_ENCRYPT_OUTPUT_SIZE -> PSA_AEAD_ENCRYPT_OUTPUT_SIZE_2_ARG - - PSA_AEAD_DECRYPT_OUTPUT_SIZE -> PSA_AEAD_DECRYPT_OUTPUT_SIZE_2_ARG - - PSA_AEAD_UPDATE_OUTPUT_SIZE -> PSA_AEAD_UPDATE_OUTPUT_SIZE_2_ARG - - PSA_AEAD_FINISH_OUTPUT_SIZE -> PSA_AEAD_FINISH_OUTPUT_SIZE_1_ARG - - PSA_AEAD_VERIFY_OUTPUT_SIZE -> PSA_AEAD_VERIFY_OUTPUT_SIZE_1_ARG - * Implement one-shot cipher functions, psa_cipher_encrypt and - psa_cipher_decrypt, according to the PSA Crypto API 1.0.0 - specification. - -Requirement changes - * The library now uses the %zu format specifier with the printf() family of - functions, so requires a toolchain that supports it. This change does not - affect the maintained LTS branches, so when contributing changes please - bear this in mind and do not add them to backported code. - -Features - * Add mbedtls_rsa_rsassa_pss_sign_ext() function allowing to generate a - signature with a specific salt length. This function allows to validate - test cases provided in the NIST's CAVP test suite. Contributed by Cédric - Meuter in PR #3183. - * Added support for built-in driver keys through the PSA opaque crypto - driver interface. Refer to the documentation of - MBEDTLS_PSA_CRYPTO_BUILTIN_KEYS for more information. - * Implement psa_sign_message() and psa_verify_message(). - * The new function mbedtls_mpi_random() generates a random value in a - given range uniformly. - * Implement psa_mac_compute() and psa_mac_verify() as defined in the - PSA Cryptograpy API 1.0.0 specification. - * MBEDTLS_ECP_MAX_BITS is now determined automatically from the configured - curves and no longer needs to be configured explicitly to save RAM. - -Security - * Fix a bias in the generation of finite-field Diffie-Hellman-Merkle (DHM) - private keys and of blinding values for DHM and elliptic curves (ECP) - computations. Reported by FlorianF89 in #4245. - * Fix a potential side channel vulnerability in ECDSA ephemeral key generation. - An adversary who is capable of very precise timing measurements could - learn partial information about the leading bits of the nonce used for the - signature, allowing the recovery of the private key after observing a - large number of signature operations. This completes a partial fix in - Mbed TLS 2.20.0. - * It was possible to configure MBEDTLS_ECP_MAX_BITS to a value that is - too small, leading to buffer overflows in ECC operations. Fail the build - in such a case. - * An adversary with access to precise enough information about memory - accesses (typically, an untrusted operating system attacking a secure - enclave) could recover an RSA private key after observing the victim - performing a single private-key operation. Found and reported by - Zili KOU, Wenjian HE, Sharad Sinha, and Wei ZHANG. - * An adversary with access to precise enough timing information (typically, a - co-located process) could recover a Curve25519 or Curve448 static ECDH key - after inputting a chosen public key and observing the victim performing the - corresponding private-key operation. Found and reported by Leila Batina, - Lukas Chmielewski, Björn Haase, Niels Samwel and Peter Schwabe. - -Bugfix - * Add printf function attributes to mbedtls_debug_print_msg to ensure we - get printf format specifier warnings. - * Fix premature fopen() call in mbedtls_entropy_write_seed_file which may - lead to seed file corruption in the case where the path to the seed file is - equal to MBEDTLS_PLATFORM_STD_NV_SEED_FILE. Contributed by Victor - Krasnoshchok in #3616. - * PSA functions other than psa_open_key now return PSA_ERROR_INVALID_HANDLE - rather than PSA_ERROR_DOES_NOT_EXIST for an invalid handle, bringing them - in line with version 1.0.0 of the specification. Fix #4162. - * PSA functions creating a key now return PSA_ERROR_INVALID_ARGUMENT rather - than PSA_ERROR_INVALID_HANDLE when the identifier specified for the key - to create is not valid, bringing them in line with version 1.0.0 of the - specification. Fix #4271. - * Fix some cases in the bignum module where the library constructed an - unintended representation of the value 0 which was not processed - correctly by some bignum operations. This could happen when - mbedtls_mpi_read_string() was called on "-0", or when - mbedtls_mpi_mul_mpi() and mbedtls_mpi_mul_int() was called with one of - the arguments being negative and the other being 0. Fixes #4643. - * Fix a bug in ECDSA that would cause it to fail when the hash is all-bits - zero. Fixes #1792 - * Fix a compilation error when MBEDTLS_ECP_RANDOMIZE_MXZ_ALT is - defined. Fixes #4217. - * Fix an incorrect error code when parsing a PKCS#8 private key. - * In a TLS client, enforce the Diffie-Hellman minimum parameter size - set with mbedtls_ssl_conf_dhm_min_bitlen() precisely. Before, the - minimum size was rounded down to the nearest multiple of 8. - * In library/net_sockets.c, _POSIX_C_SOURCE and _XOPEN_SOURCE are - defined to specific values. If the code is used in a context - where these are already defined, this can result in a compilation - error. Instead, assume that if they are defined, the values will - be adequate to build Mbed TLS. - * The cipher suite TLS-RSA-WITH-CAMELLIA-256-GCM-SHA384 was not available - when SHA-1 was disabled and was offered when SHA-1 was enabled but SHA-384 - was disabled. Fix the dependency. Fixes #4472. - * Do not offer SHA384 cipher suites when SHA-384 is disabled. Fixes #4499. - * With MBEDTLS_PSA_CRYPTO_C disabled, some functions were getting built - nonetheless, resulting in undefined reference errors when building a - shared library. Reported by Guillermo Garcia M. in #4411. - * Fix test suite code on platforms where int32_t is not int, such as - Arm Cortex-M. Fixes #4530. - * Fix some issues affecting MBEDTLS_ARIA_ALT implementations: a misplaced - directive in a header and a missing initialization in the self-test. - * Fix a missing initialization in the Camellia self-test, affecting - MBEDTLS_CAMELLIA_ALT implementations. - * Restore the ability to configure PSA via Mbed TLS options to support RSA - key pair operations but exclude RSA key generation. When MBEDTLS_GENPRIME - is not defined PSA will no longer attempt to use mbedtls_rsa_gen_key(). - Fixes #4512. - * Fix a regression introduced in 2.24.0 which broke (D)TLS CBC ciphersuites - (when the encrypt-then-MAC extension is not in use) with some ALT - implementations of the underlying hash (SHA-1, SHA-256, SHA-384), causing - the affected side to wrongly reject valid messages. Fixes #4118. - * Remove outdated check-config.h check that prevented implementing the - timing module on Mbed OS. Fixes #4633. - * Fix PSA_ALG_TLS12_PRF and PSA_ALG_TLS12_PSK_TO_MS being too permissive - about missing inputs. - * Fix mbedtls_net_poll() and mbedtls_net_recv_timeout() often failing with - MBEDTLS_ERR_NET_POLL_FAILED on Windows. Fixes #4465. - * Fix a resource leak in a test suite with an alternative AES - implementation. Fixes #4176. - * Fix a crash in mbedtls_mpi_debug_mpi on a bignum having 0 limbs. This - could notably be triggered by setting the TLS debug level to 3 or above - and using a Montgomery curve for the key exchange. Reported by lhuang04 - in #4578. Fixes #4608. - * psa_verify_hash() was relying on implementation-specific behavior of - mbedtls_rsa_rsassa_pss_verify() and was causing failures in some _ALT - implementations. This reliance is now removed. Fixes #3990. - * Disallow inputs of length different from the corresponding hash when - signing or verifying with PSA_ALG_RSA_PSS (The PSA Crypto API mandates - that PSA_ALG_RSA_PSS uses the same hash throughout the algorithm.) - * Fix a null pointer dereference when mbedtls_mpi_exp_mod() was called with - A=0 represented with 0 limbs. Up to and including Mbed TLS 2.26, this bug - could not be triggered by code that constructed A with one of the - mbedtls_mpi_read_xxx functions (including in particular TLS code) since - those always built an mpi object with at least one limb. - Credit to OSS-Fuzz. Fixes #4641. - * Fix mbedtls_mpi_gcd(G,A,B) when the value of B is zero. This had no - effect on Mbed TLS's internal use of mbedtls_mpi_gcd(), but may affect - applications that call mbedtls_mpi_gcd() directly. Fixes #4642. - * The PSA API no longer allows the creation or destruction of keys with a - read-only lifetime. The persistence level PSA_KEY_PERSISTENCE_READ_ONLY - can now only be used as intended, for keys that cannot be modified through - normal use of the API. - * When MBEDTLS_PSA_CRYPTO_SPM is enabled, crypto_spe.h was not included - in all the right places. Include it from crypto_platform.h, which is - the natural place. Fixes #4649. - * mbedtls_pk_sign() and mbedtls_pk_verify() and their extended and - restartable variants now always honor the specified hash length if - nonzero. Before, for RSA, hash_len was ignored in favor of the length of - the specified hash algorithm. - * Fix which alert is sent in some cases to conform to the - applicable RFC: on an invalid Finished message value, an - invalid max_fragment_length extension, or an - unsupported extension used by the server. - * Correct (change from 12 to 13 bytes) the value of the macro describing the - maximum nonce length returned by psa_aead_generate_nonce(). - -Changes - * Add extra printf compiler warning flags to builds. - * Fix memsan build false positive in x509_crt.c with Clang 11 - * Fix the setting of the read timeout in the DTLS sample programs. - * Remove the AES sample application programs/aes/aescrypt2 which shows - bad cryptographic practice. Fix #1906. - * Alternative implementations of CMAC may now opt to not support 3DES as a - CMAC block cipher, and still pass the CMAC self test. - * Remove configs/config-psa-crypto.h, which was identical to the default - configuration except for having some extra cryptographic mechanisms - enabled and for unintended differences. This configuration was primarily - intended to demonstrate the PSA API, and lost most of its usefulness when - MBEDTLS_PSA_CRYPTO_C became enabled by default. - * When building the test suites with GNU make, invoke python3 or python, not - python2, which is no longer supported upstream. - * When using session cache based session resumption on the server, - double-check that custom session cache implementations return - sessions which are consistent with the negotiated ciphersuite - and compression method. - * Fix build failure on MinGW toolchain when __USE_MING_ANSI_STDIO is on. - When that flag is on, standard GNU C printf format specifiers - should be used. - * Reduce the default value of MBEDTLS_ECP_WINDOW_SIZE. This reduces RAM usage - during ECC operations at a negligible performance cost. - * mbedtls_mpi_read_binary(), mbedtls_mpi_read_binary_le() and - mbedtls_mpi_read_string() now construct an mbedtls_mpi object with 0 limbs - when their input has length 0. Note that this is an implementation detail - and can change at any time, so this change should be transparent, but it - may result in mbedtls_mpi_write_binary() or mbedtls_mpi_write_string() - now writing an empty string where it previously wrote one or more - zero digits when operating from values constructed with an mpi_read - function and some mpi operations. - * Implicitly add PSA_KEY_USAGE_SIGN_MESSAGE key usage policy flag when - PSA_KEY_USAGE_SIGN_HASH flag is set and PSA_KEY_USAGE_VERIFY_MESSAGE flag - when PSA_KEY_USAGE_VERIFY_HASH flag is set. This usage flag extension - is also applied when loading a key from storage. - -= mbed TLS 2.26.0 branch released 2021-03-08 - -API changes - * Renamed the PSA Crypto API output buffer size macros to bring them in line - with version 1.0.0 of the specification. - * The API glue function mbedtls_ecc_group_of_psa() now takes the curve size - in bits rather than bytes, with an additional flag to indicate if the - size may have been rounded up to a whole number of bytes. - * Renamed the PSA Crypto API AEAD tag length macros to bring them in line - with version 1.0.0 of the specification. - -Default behavior changes - * In mbedtls_rsa_context objects, the ver field was formerly documented - as always 0. It is now reserved for internal purposes and may take - different values. - -New deprecations - * PSA_KEY_EXPORT_MAX_SIZE, PSA_HASH_SIZE, PSA_MAC_FINAL_SIZE, - PSA_BLOCK_CIPHER_BLOCK_SIZE, PSA_MAX_BLOCK_CIPHER_BLOCK_SIZE and - PSA_ALG_TLS12_PSK_TO_MS_MAX_PSK_LEN have been renamed, and the old names - deprecated. - * PSA_ALG_AEAD_WITH_DEFAULT_TAG_LENGTH and PSA_ALG_AEAD_WITH_TAG_LENGTH - have been renamed, and the old names deprecated. - -Features - * The PSA crypto subsystem can now use HMAC_DRBG instead of CTR_DRBG. - CTR_DRBG is used by default if it is available, but you can override - this choice by setting MBEDTLS_PSA_HMAC_DRBG_MD_TYPE at compile time. - Fix #3354. - * Automatic fallback to a software implementation of ECP when - MBEDTLS_ECP_xxx_ALT accelerator hooks are in use can now be turned off - through setting the new configuration flag MBEDTLS_ECP_NO_FALLBACK. - * The PSA crypto subsystem can now be configured to use less static RAM by - tweaking the setting for the maximum amount of keys simultaneously in RAM. - MBEDTLS_PSA_KEY_SLOT_COUNT sets the maximum number of volatile keys that - can exist simultaneously. It has a sensible default if not overridden. - * Partial implementation of the PSA crypto driver interface: Mbed TLS can - now use an external random generator instead of the library's own - entropy collection and DRBG code. Enable MBEDTLS_PSA_CRYPTO_EXTERNAL_RNG - and see the documentation of mbedtls_psa_external_get_random() for details. - * Applications using both mbedtls_xxx and psa_xxx functions (for example, - applications using TLS and MBEDTLS_USE_PSA_CRYPTO) can now use the PSA - random generator with mbedtls_xxx functions. See the documentation of - mbedtls_psa_get_random() for details. - * In the PSA API, the policy for a MAC or AEAD algorithm can specify a - minimum MAC or tag length thanks to the new wildcards - PSA_ALG_AT_LEAST_THIS_LENGTH_MAC and - PSA_ALG_AEAD_WITH_AT_LEAST_THIS_LENGTH_TAG. - -Security - * Fix a security reduction in CTR_DRBG when the initial seeding obtained a - nonce from entropy. Applications were affected if they called - mbedtls_ctr_drbg_set_nonce_len(), if they called - mbedtls_ctr_drbg_set_entropy_len() with a size that was 3/2 times the key - length, or when the entropy module uses SHA-256 and CTR_DRBG uses AES-256. - In such cases, a random nonce was necessary to achieve the advertised - security strength, but the code incorrectly used a constant instead of - entropy from the nonce. - Found by John Stroebel in #3819 and fixed in #3973. - * Fix a buffer overflow in mbedtls_mpi_sub_abs() when calculating - |A| - |B| where |B| is larger than |A| and has more limbs (so the - function should return MBEDTLS_ERR_MPI_NEGATIVE_VALUE). Only - applications calling mbedtls_mpi_sub_abs() directly are affected: - all calls inside the library were safe since this function is - only called with |A| >= |B|. Reported by Guido Vranken in #4042. - * Fix an errorneous estimation for an internal buffer in - mbedtls_pk_write_key_pem(). If MBEDTLS_MPI_MAX_SIZE is set to an odd - value the function might fail to write a private RSA keys of the largest - supported size. - Found by Daniel Otte, reported in #4093 and fixed in #4094. - * Fix a stack buffer overflow with mbedtls_net_poll() and - mbedtls_net_recv_timeout() when given a file descriptor that is - beyond FD_SETSIZE. Reported by FigBug in #4169. - * Guard against strong local side channel attack against base64 tables by - making access aceess to them use constant flow code. - -Bugfix - * Fix use-after-scope error in programs/ssl/ssl_client2.c and ssl_server2.c - * Fix memory leak that occured when calling psa_close_key() on a - wrapped key with MBEDTLS_PSA_CRYPTO_SE_C defined. - * Fix an incorrect error code if an RSA private operation glitched. - * Fix a memory leak in an error case in psa_generate_derived_key_internal(). - * Fix a resource leak in CTR_DRBG and HMAC_DRBG when MBEDTLS_THREADING_C - is enabled, on platforms where initializing a mutex allocates resources. - This was a regression introduced in the previous release. Reported in - #4017, #4045 and #4071. - * Ensure that calling mbedtls_rsa_free() or mbedtls_entropy_free() - twice is safe. This happens for RSA when some Mbed TLS library functions - fail. Such a double-free was not safe when MBEDTLS_THREADING_C was - enabled on platforms where freeing a mutex twice is not safe. - * Fix a resource leak in a bad-arguments case of mbedtls_rsa_gen_key() - when MBEDTLS_THREADING_C is enabled on platforms where initializing - a mutex allocates resources. - * Fixes a bug where, if the library was configured to include support for - both the old SE interface and the new PSA driver interface, external keys were - not loaded from storage. This was fixed by #3996. - * This change makes 'mbedtls_x509write_crt_set_basic_constraints' - consistent with RFC 5280 4.2.1.9 which says: "Conforming CAs MUST - include this extension in all CA certificates that contain public keys - used to validate digital signatures on certificates and MUST mark the - extension as critical in such certificates." Previous to this change, - the extension was always marked as non-critical. This was fixed by - #3698. - -Changes - * A new library C file psa_crypto_client.c has been created to contain - the PSA code needed by a PSA crypto client when the PSA crypto - implementation is not included into the library. - * On recent enough versions of FreeBSD and DragonFlyBSD, the entropy module - now uses the getrandom syscall instead of reading from /dev/urandom. - -= mbed TLS 2.25.0 branch released 2020-12-11 - -API changes - * The numerical values of the PSA Crypto API macros have been updated to - conform to version 1.0.0 of the specification. - * PSA_ALG_STREAM_CIPHER replaces PSA_ALG_CHACHA20 and PSA_ALG_ARC4. - The underlying stream cipher is determined by the key type - (PSA_KEY_TYPE_CHACHA20 or PSA_KEY_TYPE_ARC4). - * The functions mbedtls_cipher_auth_encrypt() and - mbedtls_cipher_auth_decrypt() no longer accept NIST_KW contexts, - as they have no way to check if the output buffer is large enough. - Please use mbedtls_cipher_auth_encrypt_ext() and - mbedtls_cipher_auth_decrypt_ext() instead. Credit to OSS-Fuzz and - Cryptofuzz. Fixes #3665. - -Requirement changes - * Update the minimum required CMake version to 2.8.12. This silences a - warning on CMake 3.19.0. #3801 - -New deprecations - * PSA_ALG_CHACHA20 and PSA_ALG_ARC4 have been deprecated. - Use PSA_ALG_STREAM_CIPHER instead. - * The functions mbedtls_cipher_auth_encrypt() and - mbedtls_cipher_auth_decrypt() are deprecated in favour of the new - functions mbedtls_cipher_auth_encrypt_ext() and - mbedtls_cipher_auth_decrypt_ext(). Please note that with AEAD ciphers, - these new functions always append the tag to the ciphertext, and include - the tag in the ciphertext length. - -Features - * Partial implementation of the new PSA Crypto accelerator APIs. (Symmetric - ciphers, asymmetric signing/verification and key generation, validate_key - entry point, and export_public_key interface.) - * Add support for ECB to the PSA cipher API. - * In PSA, allow using a key declared with a base key agreement algorithm - in combined key agreement and derivation operations, as long as the key - agreement algorithm in use matches the algorithm the key was declared with. - This is currently non-standard behaviour, but expected to make it into a - future revision of the PSA Crypto standard. - * Add MBEDTLS_TARGET_PREFIX CMake variable, which is prefixed to the mbedtls, - mbedcrypto, mbedx509 and apidoc CMake target names. This can be used by - external CMake projects that include this one to avoid CMake target name - clashes. The default value of this variable is "", so default target names - are unchanged. - * Add support for DTLS-SRTP as defined in RFC 5764. Contributed by Johan - Pascal, improved by Ron Eldor. - * In the PSA API, it is no longer necessary to open persistent keys: - operations now accept the key identifier. The type psa_key_handle_t is now - identical to psa_key_id_t instead of being platform-defined. This bridges - the last major gap to compliance with the PSA Cryptography specification - version 1.0.0. Opening persistent keys is still supported for backward - compatibility, but will be deprecated and later removed in future - releases. - * PSA_AEAD_NONCE_LENGTH, PSA_AEAD_NONCE_MAX_SIZE, PSA_CIPHER_IV_LENGTH and - PSA_CIPHER_IV_MAX_SIZE macros have been added as defined in version - 1.0.0 of the PSA Crypto API specification. - -Security - * The functions mbedtls_cipher_auth_encrypt() and - mbedtls_cipher_auth_decrypt() would write past the minimum documented - size of the output buffer when used with NIST_KW. As a result, code using - those functions as documented with NIST_KW could have a buffer overwrite - of up to 15 bytes, with consequences ranging up to arbitrary code - execution depending on the location of the output buffer. - * Limit the size of calculations performed by mbedtls_mpi_exp_mod to - MBEDTLS_MPI_MAX_SIZE to prevent a potential denial of service when - generating Diffie-Hellman key pairs. Credit to OSS-Fuzz. - * A failure of the random generator was ignored in mbedtls_mpi_fill_random(), - which is how most uses of randomization in asymmetric cryptography - (including key generation, intermediate value randomization and blinding) - are implemented. This could cause failures or the silent use of non-random - values. A random generator can fail if it needs reseeding and cannot not - obtain entropy, or due to an internal failure (which, for Mbed TLS's own - CTR_DRBG or HMAC_DRBG, can only happen due to a misconfiguration). - * Fix a compliance issue whereby we were not checking the tag on the - algorithm parameters (only the size) when comparing the signature in the - description part of the cert to the real signature. This meant that a - NULL algorithm parameters entry would look identical to an array of REAL - (size zero) to the library and thus the certificate would be considered - valid. However, if the parameters do not match in *any* way then the - certificate should be considered invalid, and indeed OpenSSL marks these - certs as invalid when mbedtls did not. - Many thanks to guidovranken who found this issue via differential fuzzing - and reported it in #3629. - * Zeroising of local buffers and variables which are used for calculations - in mbedtls_pkcs5_pbkdf2_hmac(), mbedtls_internal_sha*_process(), - mbedtls_internal_md*_process() and mbedtls_internal_ripemd160_process() - functions to erase sensitive data from memory. Reported by - Johan Malmgren and Johan Uppman Bruce from Sectra. - -Bugfix - * Fix an invalid (but non-zero) return code from mbedtls_pk_parse_subpubkey() - when the input has trailing garbage. Fixes #2512. - * Fix build failure in configurations where MBEDTLS_USE_PSA_CRYPTO is - enabled but ECDSA is disabled. Contributed by jdurkop. Fixes #3294. - * Include the psa_constant_names generated source code in the source tree - instead of generating it at build time. Fixes #3524. - * Fix rsa_prepare_blinding() to retry when the blinding value is not - invertible (mod N), instead of returning MBEDTLS_ERR_RSA_RNG_FAILED. This - addresses a regression but is rare in practice (approx. 1 in 2/sqrt(N)). - Found by Synopsys Coverity, fix contributed by Peter Kolbus (Garmin). - Fixes #3647. - * Use socklen_t on Android and other POSIX-compliant system - * Fix the build when the macro _GNU_SOURCE is defined to a non-empty value. - Fix #3432. - * Consistently return PSA_ERROR_INVALID_ARGUMENT on invalid cipher input - sizes (instead of PSA_ERROR_BAD_STATE in some cases) to make the - psa_cipher_* functions compliant with the PSA Crypto API specification. - * mbedtls_ecp_curve_list() now lists Curve25519 and Curve448 under the names - "x25519" and "x448". These curves support ECDH but not ECDSA. If you need - only the curves that support ECDSA, filter the list with - mbedtls_ecdsa_can_do(). - * Fix psa_generate_key() returning an error when asked to generate - an ECC key pair on Curve25519 or secp244k1. - * Fix psa_key_derivation_output_key() to allow the output of a combined key - agreement and subsequent key derivation operation to be used as a key - inside of the PSA Crypto core. - * Fix handling of EOF against 0xff bytes and on platforms with unsigned - chars. Fixes a build failure on platforms where char is unsigned. Fixes - #3794. - * Fix an off-by-one error in the additional data length check for - CCM, which allowed encryption with a non-standard length field. - Fixes #3719. - * Correct the default IV size for mbedtls_cipher_info_t structures using - MBEDTLS_MODE_ECB to 0, since ECB mode ciphers don't use IVs. - * Make arc4random_buf available on NetBSD and OpenBSD when _POSIX_C_SOURCE is - defined. Fix contributed in #3571. - * Fix conditions for including string.h in error.c. Fixes #3866. - * psa_set_key_id() now also sets the lifetime to persistent for keys located - in a secure element. - * Attempting to create a volatile key with a non-zero key identifier now - fails. Previously the key identifier was just ignored when creating a - volatile key. - * Attempting to create or register a key with a key identifier in the vendor - range now fails. - * Fix build failures on GCC 11. Fixes #3782. - * Add missing arguments of debug message in mbedtls_ssl_decrypt_buf. - * Fix a memory leak in mbedtls_mpi_sub_abs() when the result was negative - (an error condition) and the second operand was aliased to the result. - * Fix a case in elliptic curve arithmetic where an out-of-memory condition - could go undetected, resulting in an incorrect result. - * In CTR_DRBG and HMAC_DRBG, don't reset the reseed interval in seed(). - Fixes #2927. - * In PEM writing functions, fill the trailing part of the buffer with null - bytes. This guarantees that the corresponding parsing function can read - the buffer back, which was the case for mbedtls_x509write_{crt,csr}_pem - until this property was inadvertently broken in Mbed TLS 2.19.0. - Fixes #3682. - * Fix a build failure that occurred with the MBEDTLS_AES_SETKEY_DEC_ALT - option on. In this configuration key management methods that are required - for MBEDTLS_CIPHER_MODE_XTS were excluded from the build and made it fail. - Fixes #3818. Reported by John Stroebel. - -Changes - * Reduce stack usage significantly during sliding window exponentiation. - Reported in #3591 and fix contributed in #3592 by Daniel Otte. - * The PSA persistent storage format is updated to always store the key bits - attribute. No automatic upgrade path is provided. Previously stored keys - must be erased, or manually upgraded based on the key storage format - specification (docs/architecture/mbed-crypto-storage-specification.md). - Fixes #3740. - * Remove the zeroization of a pointer variable in AES rounds. It was valid - but spurious and misleading since it looked like a mistaken attempt to - zeroize the pointed-to buffer. Reported by Antonio de la Piedra, CEA - Leti, France. - -= mbed TLS 2.24.0 branch released 2020-09-01 - -API changes - * In the PSA API, rename the types of elliptic curve and Diffie-Hellman - group families to psa_ecc_family_t and psa_dh_family_t, in line with the - PSA Crypto API specification version 1.0.0. - Rename associated macros as well: - PSA_ECC_CURVE_xxx renamed to PSA_ECC_FAMILY_xxx - PSA_DH_GROUP_xxx renamed to PSA_DH_FAMILY_xxx - PSA_KEY_TYPE_GET_CURVE renamed to to PSA_KEY_TYPE_ECC_GET_FAMILY - PSA_KEY_TYPE_GET_GROUP renamed to PSA_KEY_TYPE_DH_GET_FAMILY - -Default behavior changes - * Stop storing persistent information about externally stored keys created - through PSA Crypto with a volatile lifetime. Reported in #3288 and - contributed by Steven Cooreman in #3382. - -Features - * The new function mbedtls_ecp_write_key() exports private ECC keys back to - a byte buffer. It is the inverse of the existing mbedtls_ecp_read_key(). - * Support building on e2k (Elbrus) architecture: correctly enable - -Wformat-signedness, and fix the code that causes signed-one-bit-field - and sign-compare warnings. Contributed by makise-homura (Igor Molchanov) - <akemi_homura@kurisa.ch>. - -Security - * Fix a vulnerability in the verification of X.509 certificates when - matching the expected common name (the cn argument of - mbedtls_x509_crt_verify()) with the actual certificate name: when the - subjecAltName extension is present, the expected name was compared to any - name in that extension regardless of its type. This means that an - attacker could for example impersonate a 4-bytes or 16-byte domain by - getting a certificate for the corresponding IPv4 or IPv6 (this would - require the attacker to control that IP address, though). Similar attacks - using other subjectAltName name types might be possible. Found and - reported by kFYatek in #3498. - * When checking X.509 CRLs, a certificate was only considered as revoked if - its revocationDate was in the past according to the local clock if - available. In particular, on builds without MBEDTLS_HAVE_TIME_DATE, - certificates were never considered as revoked. On builds with - MBEDTLS_HAVE_TIME_DATE, an attacker able to control the local clock (for - example, an untrusted OS attacking a secure enclave) could prevent - revocation of certificates via CRLs. Fixed by no longer checking the - revocationDate field, in accordance with RFC 5280. Reported by - yuemonangong in #3340. Reported independently and fixed by - Raoul Strackx and Jethro Beekman in #3433. - * In (D)TLS record decryption, when using a CBC ciphersuites without the - Encrypt-then-Mac extension, use constant code flow memory access patterns - to extract and check the MAC. This is an improvement to the existing - countermeasure against Lucky 13 attacks. The previous countermeasure was - effective against network-based attackers, but less so against local - attackers. The new countermeasure defends against local attackers, even - if they have access to fine-grained measurements. In particular, this - fixes a local Lucky 13 cache attack found and reported by Tuba Yavuz, - Farhaan Fowze, Ken (Yihan) Bai, Grant Hernandez, and Kevin Butler - (University of Florida) and Dave Tian (Purdue University). - * Fix side channel in RSA private key operations and static (finite-field) - Diffie-Hellman. An adversary with precise enough timing and memory access - information (typically an untrusted operating system attacking a secure - enclave) could bypass an existing counter-measure (base blinding) and - potentially fully recover the private key. - * Fix a 1-byte buffer overread in mbedtls_x509_crl_parse_der(). - Credit to OSS-Fuzz for detecting the problem and to Philippe Antoine - for pinpointing the problematic code. - * Zeroising of plaintext buffers in mbedtls_ssl_read() to erase unused - application data from memory. Reported in #689 by - Johan Uppman Bruce of Sectra. - -Bugfix - * Library files installed after a CMake build no longer have execute - permission. - * Use local labels in mbedtls_padlock_has_support() to fix an invalid symbol - redefinition if the function is inlined. - Reported in #3451 and fix contributed in #3452 by okhowang. - * Fix the endianness of Curve25519 keys imported/exported through the PSA - APIs. psa_import_key and psa_export_key will now correctly expect/output - Montgomery keys in little-endian as defined by RFC7748. Contributed by - Steven Cooreman in #3425. - * Fix build errors when the only enabled elliptic curves are Montgomery - curves. Raised by signpainter in #941 and by Taiki-San in #1412. This - also fixes missing declarations reported by Steven Cooreman in #1147. - * Fix self-test failure when the only enabled short Weierstrass elliptic - curve is secp192k1. Fixes #2017. - * PSA key import will now correctly import a Curve25519/Curve448 public key - instead of erroring out. Contributed by Steven Cooreman in #3492. - * Use arc4random_buf on NetBSD instead of rand implementation with cyclical - lower bits. Fix contributed in #3540. - * Fix a memory leak in mbedtls_md_setup() when using HMAC under low memory - conditions. Reported and fix suggested by Guido Vranken in #3486. - * Fix bug in redirection of unit test outputs on platforms where stdout is - defined as a macro. First reported in #2311 and fix contributed in #3528. - -Changes - * Only pass -Wformat-signedness to versions of GCC that support it. Reported - in #3478 and fix contributed in #3479 by okhowang. - * Reduce the stack consumption of mbedtls_x509write_csr_der() which - previously could lead to stack overflow on constrained devices. - Contributed by Doru Gucea and Simon Leet in #3464. - * Undefine the ASSERT macro before defining it locally, in case it is defined - in a platform header. Contributed by Abdelatif Guettouche in #3557. - * Update copyright notices to use Linux Foundation guidance. 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. This also - eliminates the need for the lines declaring the files to be part of - MbedTLS. Fixes #3457. - * Add the command line parameter key_pwd to the ssl_client2 and ssl_server2 - example applications which allows to provide a password for the key file - specified through the existing key_file argument. This allows the use of - these applications with password-protected key files. Analogously but for - ssl_server2 only, add the command line parameter key_pwd2 which allows to - set a password for the key file provided through the existing key_file2 - argument. - -= mbed TLS 2.23.0 branch released 2020-07-01 - -Default behavior changes - * In the experimental PSA secure element interface, change the encoding of - key lifetimes to encode a persistence level and the location. Although C - prototypes do not effectively change, code calling - psa_register_se_driver() must be modified to pass the driver's location - instead of the keys' lifetime. If the library is upgraded on an existing - device, keys created with the old lifetime value will not be readable or - removable through Mbed TLS after the upgrade. - -Features - * New functions in the error module return constant strings for - high- and low-level error codes, complementing mbedtls_strerror() - which constructs a string for any error code, including compound - ones, but requires a writable buffer. Contributed by Gaurav Aggarwal - in #3176. - * The new utility programs/ssl/ssl_context_info prints a human-readable - dump of an SSL context saved with mbedtls_ssl_context_save(). - * Add support for midipix, a POSIX layer for Microsoft Windows. - * Add new mbedtls_x509_crt_parse_der_with_ext_cb() routine which allows - parsing unsupported certificate extensions via user provided callback. - Contributed by Nicola Di Lieto <nicola.dilieto@gmail.com> in #3243 as - a solution to #3241. - * Pass the "certificate policies" extension to the callback supplied to - mbedtls_x509_crt_parse_der_with_ext_cb() if it contains unsupported - policies (#3419). - * Added support to entropy_poll for the kern.arandom syscall supported on - some BSD systems. Contributed by Nia Alarie in #3423. - * Add support for Windows 2000 in net_sockets. Contributed by opatomic. #3239 - -Security - * Fix a side channel vulnerability in modular exponentiation that could - reveal an RSA private key used in a secure enclave. Noticed by Sangho Lee, - Ming-Wei Shih, Prasun Gera, Taesoo Kim and Hyesoon Kim (Georgia Institute - of Technology); and Marcus Peinado (Microsoft Research). Reported by Raoul - Strackx (Fortanix) in #3394. - * Fix side channel in mbedtls_ecp_check_pub_priv() and - mbedtls_pk_parse_key() / mbedtls_pk_parse_keyfile() (when loading a - private key that didn't include the uncompressed public key), as well as - mbedtls_ecp_mul() / mbedtls_ecp_mul_restartable() when called with a NULL - f_rng argument. An attacker with access to precise enough timing and - memory access information (typically an untrusted operating system - attacking a secure enclave) could fully recover the ECC private key. - Found and reported by Alejandro Cabrera Aldaya and Billy Brumley. - * Fix issue in Lucky 13 counter-measure that could make it ineffective when - hardware accelerators were used (using one of the MBEDTLS_SHAxxx_ALT - macros). This would cause the original Lucky 13 attack to be possible in - those configurations, allowing an active network attacker to recover - plaintext after repeated timing measurements under some conditions. - Reported and fix suggested by Luc Perneel in #3246. - -Bugfix - * Fix the Visual Studio Release x64 build configuration for mbedtls itself. - Completes a previous fix in Mbed TLS 2.19 that only fixed the build for - the example programs. Reported in #1430 and fix contributed by irwir. - * Fix undefined behavior in X.509 certificate parsing if the - pathLenConstraint basic constraint value is equal to INT_MAX. - The actual effect with almost every compiler is the intended - behavior, so this is unlikely to be exploitable anywhere. #3192 - * Fix issue with a detected HW accelerated record error not being exposed - due to shadowed variable. Contributed by Sander Visser in #3310. - * Avoid NULL pointer dereferencing if mbedtls_ssl_free() is called with a - NULL pointer argument. Contributed by Sander Visser in #3312. - * Fix potential linker errors on dual world platforms by inlining - mbedtls_gcc_group_to_psa(). This allows the pk.c module to link separately - from psa_crypto.c. Fixes #3300. - * Remove dead code in X.509 certificate parsing. Contributed by irwir in - #2855. - * Include asn1.h in error.c. Fixes #3328 reported by David Hu. - * Fix potential memory leaks in ecp_randomize_jac() and ecp_randomize_mxz() - when PRNG function fails. Contributed by Jonas Lejeune in #3318. - * Remove unused macros from MSVC projects. Reported in #3297 and fix - submitted in #3333 by irwir. - * Add additional bounds checks in ssl_write_client_hello() preventing - output buffer overflow if the configuration declared a buffer that was - too small. - * Set _POSIX_C_SOURCE to at least 200112L in C99 code. Reported in #3420 and - fix submitted in #3421 by Nia Alarie. - * Fix building library/net_sockets.c and the ssl_mail_client program on - NetBSD. Contributed by Nia Alarie in #3422. - * Fix false positive uninitialised variable reported by cpp-check. - Contributed by Sander Visser in #3311. - * Update iv and len context pointers manually when reallocating buffers - using the MBEDTLS_SSL_VARIABLE_BUFFER_LENGTH feature. This caused issues - when receiving a connection with CID, when these fields were shifted - in ssl_parse_record_header(). - -Changes - * Fix warnings about signedness issues in format strings. The build is now - clean of -Wformat-signedness warnings. Contributed by Kenneth Soerensen - in #3153. - * Fix minor performance issue in operations on Curve25519 caused by using a - suboptimal modular reduction in one place. Found and fix contributed by - Aurelien Jarno in #3209. - * Combine identical cases in switch statements in md.c. Contributed - by irwir in #3208. - * Simplify a bounds check in ssl_write_certificate_request(). Contributed - by irwir in #3150. - * Unify the example programs termination to call mbedtls_exit() instead of - using a return command. This has been done to enable customization of the - behavior in bare metal environments. - * Fix mbedtls_x509_dn_gets to escape non-ASCII characters as "?". - Contributed by Koh M. Nakagawa in #3326. - * Use FindPython3 when cmake version >= 3.15.0 - * Abort the ClientHello writing function as soon as some extension doesn't - fit into the record buffer. Previously, such extensions were silently - dropped. As a consequence, the TLS handshake now fails when the output - buffer is not large enough to hold the ClientHello. - * The unit tests now rely on header files in tests/include/test and source - files in tests/src. When building with make or cmake, the files in - tests/src are compiled and the resulting object linked into each test - executable. - * The ECP module, enabled by `MBEDTLS_ECP_C`, now depends on - `MBEDTLS_CTR_DRBG_C` or `MBEDTLS_HMAC_DRBG_C` for some side-channel - coutermeasures. If side channels are not a concern, this dependency can - be avoided by enabling the new option `MBEDTLS_ECP_NO_INTERNAL_RNG`. - * Align MSVC error flag with GCC and Clang. Contributed by Carlos Gomes - Martinho. #3147 - * Remove superfluous assignment in mbedtls_ssl_parse_certificate(). Reported - in #3182 and fix submitted by irwir. #3217 - * Fix typo in XTS tests. Reported and fix submitted by Kxuan. #3319 - -= mbed TLS 2.22.0 branch released 2020-04-14 - -New deprecations - * Deprecate MBEDTLS_SSL_HW_RECORD_ACCEL that enables function hooks in the - SSL module for hardware acceleration of individual records. - * Deprecate mbedtls_ssl_get_max_frag_len() in favour of - mbedtls_ssl_get_output_max_frag_len() and - mbedtls_ssl_get_input_max_frag_len() to be more precise about which max - fragment length is desired. - -Security - * Fix issue in DTLS handling of new associations with the same parameters - (RFC 6347 section 4.2.8): an attacker able to send forged UDP packets to - the server could cause it to drop established associations with - legitimate clients, resulting in a Denial of Service. This could only - happen when MBEDTLS_SSL_DTLS_CLIENT_PORT_REUSE was enabled in config.h - (which it is by default). - * Fix side channel in ECC code that allowed an adversary with access to - precise enough timing and memory access information (typically an - untrusted operating system attacking a secure enclave) to fully recover - an ECDSA private key. Found and reported by Alejandro Cabrera Aldaya, - Billy Brumley and Cesar Pereida Garcia. CVE-2020-10932 - * Fix a potentially remotely exploitable buffer overread in a - DTLS client when parsing the Hello Verify Request message. - -Features - * The new build option MBEDTLS_SSL_VARIABLE_BUFFER_LENGTH automatically - resizes the I/O buffers before and after handshakes, reducing the memory - consumption during application data transfer. - -Bugfix - * Fix compilation failure when both MBEDTLS_SSL_PROTO_DTLS and - MBEDTLS_SSL_HW_RECORD_ACCEL are enabled. - * Remove a spurious check in ssl_parse_client_psk_identity that triggered - a warning with some compilers. Fix contributed by irwir in #2856. - * Fix a function name in a debug message. Contributed by Ercan Ozturk in - #3013. - -Changes - * Mbed Crypto is no longer a Git submodule. The crypto part of the library - is back directly in the present repository. - * Split mbedtls_ssl_get_max_frag_len() into - mbedtls_ssl_get_output_max_frag_len() and - mbedtls_ssl_get_input_max_frag_len() to ensure that a sufficient input - buffer is allocated by the server (if MBEDTLS_SSL_VARIABLE_BUFFER_LENGTH - is defined), regardless of what MFL was configured for it. - -= mbed TLS 2.21.0 branch released 2020-02-20 - -New deprecations - * Deprecate MBEDTLS_SSL_SRV_SUPPORT_SSLV2_CLIENT_HELLO that enables parsing - SSLv2 ClientHello messages. - * Deprecate MBEDTLS_SSL_PROTO_SSL3 that enables support for SSLv3. - * Deprecate for MBEDTLS_PKCS11_C, the wrapper around the pkcs11-helper - library which allows TLS authentication to use keys stored in a - PKCS#11 token such as a smartcard. - -Security - * Fix potential memory overread when performing an ECDSA signature - operation. The overread only happens with cryptographically low - probability (of the order of 2^-n where n is the bitsize of the curve) - unless the RNG is broken, and could result in information disclosure or - denial of service (application crash or extra resource consumption). - Found by Auke Zeilstra and Peter Schwabe, using static analysis. - * To avoid a side channel vulnerability when parsing an RSA private key, - read all the CRT parameters from the DER structure rather than - reconstructing them. Found by Alejandro Cabrera Aldaya and Billy Bob - Brumley. Reported and fix contributed by Jack Lloyd. - ARMmbed/mbed-crypto#352 - -Features - * The new build option MBEDTLS_SHA512_NO_SHA384 allows building SHA-512 - support without SHA-384. - -API changes - * Change the encoding of key types and curves in the PSA API. The new - values are aligned with the upcoming release of the PSA Crypto API - specification version 1.0.0. The main change which may break some - existing code is that elliptic curve key types no longer encode the - exact curve: a psa_ecc_curve_t or psa_key_type_t value only encodes - a curve family and the key size determines the exact curve (for example, - PSA_ECC_CURVE_SECP_R1 with 256 bits is P256R1). ARMmbed/mbed-crypto#330 - -Bugfix - * Fix an unchecked call to mbedtls_md() in the x509write module. - * Fix build failure with MBEDTLS_ZLIB_SUPPORT enabled. Reported by - Jack Lloyd in #2859. Fix submitted by jiblime in #2963. - * Fix some false-positive uninitialized variable warnings in X.509. Fix - contributed by apple-ihack-geek in #2663. - * Fix a possible error code mangling in psa_mac_verify_finish() when - a cryptographic accelerator fails. ARMmbed/mbed-crypto#345 - * Fix a bug in mbedtls_pk_parse_key() that would cause it to accept some - RSA keys that would later be rejected by functions expecting private - keys. Found by Catena cyber using oss-fuzz (issue 20467). - * Fix a bug in mbedtls_pk_parse_key() that would cause it to - accept some RSA keys with invalid values by silently fixing those values. - -= mbed TLS 2.20.0 branch released 2020-01-15 - -Default behavior changes - * The initial seeding of a CTR_DRBG instance makes a second call to the - entropy function to obtain entropy for a nonce if the entropy size is less - than 3/2 times the key size. In case you want to disable the extra call to - grab entropy, you can call mbedtls_ctr_drbg_set_nonce_len() to force the - nonce length to 0. - -Security - * Enforce that mbedtls_entropy_func() gathers a total of - MBEDTLS_ENTROPY_BLOCK_SIZE bytes or more from strong sources. In the - default configuration, on a platform with a single entropy source, the - entropy module formerly only grabbed 32 bytes, which is good enough for - security if the source is genuinely strong, but less than the expected 64 - bytes (size of the entropy accumulator). - * Zeroize local variables in mbedtls_internal_aes_encrypt() and - mbedtls_internal_aes_decrypt() before exiting the function. The value of - these variables can be used to recover the last round key. To follow best - practice and to limit the impact of buffer overread vulnerabilities (like - Heartbleed) we need to zeroize them before exiting the function. - Issue reported by Tuba Yavuz, Farhaan Fowze, Ken (Yihang) Bai, - Grant Hernandez, and Kevin Butler (University of Florida) and - Dave Tian (Purdue University). - * Fix side channel vulnerability in ECDSA. Our bignum implementation is not - constant time/constant trace, so side channel attacks can retrieve the - blinded value, factor it (as it is smaller than RSA keys and not guaranteed - to have only large prime factors), and then, by brute force, recover the - key. Reported by Alejandro Cabrera Aldaya and Billy Brumley. - * Fix side channel vulnerability in ECDSA key generation. Obtaining precise - timings on the comparison in the key generation enabled the attacker to - learn leading bits of the ephemeral key used during ECDSA signatures and to - recover the private key. Reported by Jeremy Dubeuf. - * Catch failure of AES functions in mbedtls_ctr_drbg_random(). Uncaught - failures could happen with alternative implementations of AES. Bug - reported and fix proposed by Johan Uppman Bruce and Christoffer Lauri, - Sectra. - -Features - * Key derivation inputs in the PSA API can now either come from a key object - or from a buffer regardless of the step type. - * The CTR_DRBG module can grab a nonce from the entropy source during the - initial seeding. The default nonce length is chosen based on the key size - to achieve the security strength defined by NIST SP 800-90A. You can - change it with mbedtls_ctr_drbg_set_nonce_len(). - * Add ENUMERATED tag support to the ASN.1 module. Contributed by - msopiha-linaro in ARMmbed/mbed-crypto#307. - -API changes - * In the PSA API, forbid zero-length keys. To pass a zero-length input to a - key derivation function, use a buffer instead (this is now always - possible). - * Rename psa_asymmetric_sign() to psa_sign_hash() and - psa_asymmetric_verify() to psa_verify_hash(). - -Bugfix - * Fix an incorrect size in a debugging message. Reported and fix - submitted by irwir. Fixes #2717. - * Fix an unused variable warning when compiling without DTLS. - Reported and fix submitted by irwir. Fixes #2800. - * Remove a useless assignment. Reported and fix submitted by irwir. - Fixes #2801. - * Fix a buffer overflow in the PSA HMAC code when using a long key with an - unsupported algorithm. Fixes ARMmbed/mbed-crypto#254. - * Fix mbedtls_asn1_get_int to support any number of leading zeros. Credit - to OSS-Fuzz for finding a bug in an intermediate version of the fix. - * Fix mbedtls_asn1_get_bitstring_null to correctly parse bitstrings of at - most 2 bytes. - * mbedtls_ctr_drbg_set_entropy_len() and - mbedtls_hmac_drbg_set_entropy_len() now work if you call them before - mbedtls_ctr_drbg_seed() or mbedtls_hmac_drbg_seed(). - -Changes - * Remove the technical possibility to define custom mbedtls_md_info - structures, which was exposed only in an internal header. - * psa_close_key(0) and psa_destroy_key(0) now succeed (doing nothing, as - before). - * Variables containing error codes are now initialized to an error code - rather than success, so that coding mistakes or memory corruption tends to - cause functions to return this error code rather than a success. There are - no known instances where this changes the behavior of the library: this is - merely a robustness improvement. ARMmbed/mbed-crypto#323 - * Remove a useless call to mbedtls_ecp_group_free(). Contributed by - Alexander Krizhanovsky in ARMmbed/mbed-crypto#210. - * Speed up PBKDF2 by caching the digest calculation. Contributed by Jack - Lloyd and Fortanix Inc in ARMmbed/mbed-crypto#277. - * Small performance improvement of mbedtls_mpi_div_mpi(). Contributed by - Alexander Krizhanovsky in ARMmbed/mbed-crypto#308. - -= mbed TLS 2.19.1 branch released 2019-09-16 - -Features - * Declare include headers as PUBLIC to propagate to CMake project consumers - Contributed by Zachary J. Fields in PR #2949. - * Add nss_keylog to ssl_client2 and ssl_server2, enabling easier analysis of - TLS sessions with tools like Wireshark. - -API Changes - * Make client_random and server_random const in - mbedtls_ssl_export_keys_ext_t, so that the key exporter is discouraged - from modifying the client/server hello. - -Bugfix - * Fix some false-positive uninitialized variable warnings in crypto. Fix - contributed by apple-ihack-geek in #2663. - -= mbed TLS 2.19.0 branch released 2019-09-06 - -Security - * Fix a missing error detection in ECJPAKE. This could have caused a - predictable shared secret if a hardware accelerator failed and the other - side of the key exchange had a similar bug. - * When writing a private EC key, use a constant size for the private - value, as specified in RFC 5915. Previously, the value was written - as an ASN.1 INTEGER, which caused the size of the key to leak - about 1 bit of information on average and could cause the value to be - 1 byte too large for the output buffer. - * The deterministic ECDSA calculation reused the scheme's HMAC-DRBG to - implement blinding. Because of this for the same key and message the same - blinding value was generated. This reduced the effectiveness of the - countermeasure and leaked information about the private key through side - channels. Reported by Jack Lloyd. - -Features - * Add new API functions mbedtls_ssl_session_save() and - mbedtls_ssl_session_load() to allow serializing a session, for example to - store it in non-volatile storage, and later using it for TLS session - resumption. - * Add a new API function mbedtls_ssl_check_record() to allow checking that - an incoming record is valid, authentic and has not been seen before. This - feature can be used alongside Connection ID and SSL context serialisation. - The feature is enabled at compile-time by MBEDTLS_SSL_RECORD_CHECKING - option. - * New implementation of X25519 (ECDH using Curve25519) from Project Everest - (https://project-everest.github.io/). It can be enabled at compile time - with MBEDTLS_ECDH_VARIANT_EVEREST_ENABLED. This implementation is formally - verified and significantly faster, but is only supported on x86 platforms - (32-bit and 64-bit) using GCC, Clang or Visual Studio. Contributed by - Christoph Wintersteiger from Microsoft Research. - * Add mbedtls_net_close(), enabling the building of forking servers where - the parent process closes the client socket and continue accepting, and - the child process closes the listening socket and handles the client - socket. Contributed by Robert Larsen in #2803. - -API Changes - * Add DER-encoded test CRTs to library/certs.c, allowing - the example programs ssl_server2 and ssl_client2 to be run - if MBEDTLS_FS_IO and MBEDTLS_PEM_PARSE_C are unset. Fixes #2254. - * The HAVEGE state type now uses uint32_t elements instead of int. - * The functions mbedtls_ecp_curve_list() and mbedtls_ecp_grp_id_list() now - list all curves for which at least one of ECDH or ECDSA is supported, not - just curves for which both are supported. Call mbedtls_ecdsa_can_do() or - mbedtls_ecdh_can_do() on each result to check whether each algorithm is - supported. - * The new function mbedtls_ecdsa_sign_det_ext() is similar to - mbedtls_ecdsa_sign_det() but allows passing an external RNG for the - purpose of blinding. - -New deprecations - * Deprecate mbedtls_ecdsa_sign_det() in favor of a functions that can take an - RNG function as an input. - * Calling mbedtls_ecdsa_write_signature() with NULL as the f_rng argument - is now deprecated. - -Bugfix - * Fix missing bounds checks in X.509 parsing functions that could - lead to successful parsing of ill-formed X.509 CRTs. Fixes #2437. - * Fix multiple X.509 functions previously returning ASN.1 low-level error - codes to always wrap these codes into X.509 high level error codes before - returning. Fixes #2431. - * Fix to allow building test suites with any warning that detects unused - functions. Fixes #1628. - * Fix typo in net_would_block(). Fixes #528 reported by github-monoculture. - * Remove redundant include file in timing.c. Fixes #2640 reported by irwir. - * Fix build failure when building with mingw on Windows by including - stdarg.h where needed. Fixes #2656. - * Fix Visual Studio Release x64 build configuration by inheriting - PlatformToolset from the project configuration. Fixes #1430 reported by - irwir. - * Enable Suite B with subset of ECP curves. Make sure the code compiles even - if some curves are not defined. Fixes #1591 reported by dbedev. - * Fix misuse of signed arithmetic in the HAVEGE module. #2598 - * Avoid use of statically sized stack buffers for certificate writing. - This previously limited the maximum size of DER encoded certificates - in mbedtls_x509write_crt_der() to 2Kb. Reported by soccerGB in #2631. - * Fix partial zeroing in x509_get_other_name. Found and fixed by ekse, #2716. - * Update test certificates that were about to expire. Reported by - Bernhard M. Wiedemann in #2357. - * Fix the build on ARMv5TE in ARM mode to not use assembly instructions - that are only available in Thumb mode. Fix contributed by Aurelien Jarno - in #2169. - * Fix propagation of restart contexts in restartable EC operations. - This could previously lead to segmentation faults in builds using an - address-sanitizer and enabling but not using MBEDTLS_ECP_RESTARTABLE. - * Fix memory leak in in mpi_miller_rabin(). Contributed by - Jens Wiklander <jens.wiklander@linaro.org> in #2363 - * Improve code clarity in x509_crt module, removing false-positive - uninitialized variable warnings on some recent toolchains (GCC8, etc). - Discovered and fixed by Andy Gross (Linaro), #2392. - * Fix bug in endianness conversion in bignum module. This lead to - functionally incorrect code on bigendian systems which don't have - __BYTE_ORDER__ defined. Reported by Brendan Shanks. Fixes #2622. - -Changes - * Replace multiple uses of MD2 by SHA-256 in X.509 test suite. Fixes #821. - * Make it easier to define MBEDTLS_PARAM_FAILED as assert (which config.h - suggests). #2671 - * Make `make clean` clean all programs always. Fixes #1862. - * Add a Dockerfile and helper scripts (all-in-docker.sh, basic-in-docker.sh, - docker-env.sh) to simplify running test suites on a Linux host. Contributed - by Peter Kolbus (Garmin). - * Add `reproducible` option to `ssl_client2` and `ssl_server2` to enable - test runs without variability. Contributed by Philippe Antoine (Catena - cyber) in #2681. - * Extended .gitignore to ignore Visual Studio artifacts. Fixed by ConfusedSushi. - * Adds fuzz targets, especially for continuous fuzzing with OSS-Fuzz. - Contributed by Philippe Antoine (Catena cyber). - * Remove the crypto part of the library from Mbed TLS. The crypto - code and tests are now only available via Mbed Crypto, which - Mbed TLS references as a Git submodule. - -= mbed TLS 2.18.1 branch released 2019-07-12 - -Bugfix - * Fix build failure when building with mingw on Windows by including - stdarg.h where needed. Fixes #2656. - -Changes - * Enable building of Mbed TLS as a CMake subproject. Suggested and fixed by - Ashley Duncan in #2609. - -= mbed TLS 2.18.0 branch released 2019-06-11 - -Features - * Add the Any Policy certificate policy oid, as defined in - rfc 5280 section 4.2.1.4. - * It is now possible to use NIST key wrap mode via the mbedtls_cipher API. - Contributed by Jack Lloyd and Fortanix Inc. - * Add the Wi-SUN Field Area Network (FAN) device extended key usage. - * Add the oid certificate policy x509 extension. - * It is now possible to perform RSA PKCS v1.5 signatures with RIPEMD-160 digest. - Contributed by Jack Lloyd and Fortanix Inc. - * Extend the MBEDTLS_SSL_EXPORT_KEYS to export the handshake randbytes, - and the used tls-prf. - * Add public API for tls-prf function, according to requested enum. - * Add support for parsing otherName entries in the Subject Alternative Name - X.509 certificate extension, specifically type hardware module name, - as defined in RFC 4108 section 5. - * Add support for parsing certificate policies extension, as defined in - RFC 5280 section 4.2.1.4. Currently, only the "Any Policy" policy is - supported. - * List all SAN types in the subject_alt_names field of the certificate. - Resolves #459. - * Add support for draft-05 of the Connection ID extension, as specified - in https://tools.ietf.org/html/draft-ietf-tls-dtls-connection-id-05. - The Connection ID extension allows to keep DTLS connections beyond the - lifetime of the underlying transport by adding a connection identifier - to the DTLS record header. This identifier can be used to associated an - incoming record with the correct connection data even after the peer has - changed its IP or port. The feature is enabled at compile-time by setting - MBEDTLS_SSL_DTLS_CONNECTION_ID (disabled by default), and at run-time - through the new APIs mbedtls_ssl_conf_cid() and mbedtls_ssl_set_cid(). - - -API Changes - * Extend the MBEDTLS_SSL_EXPORT_KEYS to export the handshake randbytes, - and the used tls-prf. - * Add public API for tls-prf function, according to requested enum. - -Bugfix - * Fix private key DER output in the key_app_writer example. File contents - were shifted by one byte, creating an invalid ASN.1 tag. Fixed by - Christian Walther in #2239. - * Fix potential memory leak in X.509 self test. Found and fixed by - Junhwan Park, #2106. - * Reduce stack usage of hkdf tests. Fixes #2195. - * Fix 1-byte buffer overflow in mbedtls_mpi_write_string() when - used with negative inputs. Found by Guido Vranken in #2404. Credit to - OSS-Fuzz. - * Fix bugs in the AEAD test suite which would be exposed by ciphers which - either used both encrypt and decrypt key schedules, or which perform padding. - GCM and CCM were not affected. Fixed by Jack Lloyd. - * Fix incorrect default port number in ssl_mail_client example's usage. - Found and fixed by irwir. #2337 - * Add psa_util.h to test/cpp_dummy_build to fix build_default_make_gcc_and_cxx. - Fixed by Peter Kolbus (Garmin). #2579 - * Add missing parentheses around parameters in the definition of the - public macro MBEDTLS_X509_ID_FLAG. This could lead to invalid evaluation - in case operators binding less strongly than subtraction were used - for the parameter. - * Add a check for MBEDTLS_X509_CRL_PARSE_C in ssl_server2, guarding the crl - sni entry parameter. Reported by inestlerode in #560. - * Set the next sequence of the subject_alt_name to NULL when deleting - sequence on failure. Found and fix suggested by Philippe Antoine. - Credit to OSS-Fuzz. - -Changes - * Server's RSA certificate in certs.c was SHA-1 signed. In the default - mbedTLS configuration only SHA-2 signed certificates are accepted. - This certificate is used in the demo server programs, which lead the - client programs to fail at the peer's certificate verification - due to an unacceptable hash signature. The certificate has been - updated to one that is SHA-256 signed. Fix contributed by - Illya Gerasymchuk. - * Return from various debugging routines immediately if the - provided SSL context is unset. - * Remove dead code from bignum.c in the default configuration. - Found by Coverity, reported and fixed by Peter Kolbus (Garmin). Fixes #2309. - * Add test for minimal value of MBEDTLS_MPI_WINDOW_SIZE to all.sh. - Contributed by Peter Kolbus (Garmin). - * Change wording in the `mbedtls_ssl_conf_max_frag_len()`'s documentation to - improve clarity. Fixes #2258. - -= mbed TLS 2.17.0 branch released 2019-03-19 - -Features - * Add a new X.509 API call `mbedtls_x509_parse_der_nocopy()` - which allows copy-less parsing of DER encoded X.509 CRTs, - at the cost of additional lifetime constraints on the input - buffer, but at the benefit of reduced RAM consumption. - * Add a new function mbedtls_asn1_write_named_bitstring() to write ASN.1 - named bitstring in DER as required by RFC 5280 Appendix B. - * Add MBEDTLS_REMOVE_3DES_CIPHERSUITES to allow removing 3DES ciphersuites - from the default list (enabled by default). See - https://sweet32.info/SWEET32_CCS16.pdf. - -API Changes - * Add a new X.509 API call `mbedtls_x509_parse_der_nocopy()`. - See the Features section for more information. - * Allow to opt in to the removal the API mbedtls_ssl_get_peer_cert() - for the benefit of saving RAM, by disabling the new compile-time - option MBEDTLS_SSL_KEEP_PEER_CERTIFICATE (enabled by default for - API stability). Disabling this option makes mbedtls_ssl_get_peer_cert() - always return NULL, and removes the peer_cert field from the - mbedtls_ssl_session structure which otherwise stores the peer's - certificate. - -Security - * Make mbedtls_ecdh_get_params return an error if the second key - belongs to a different group from the first. Before, if an application - passed keys that belonged to different group, the first key's data was - interpreted according to the second group, which could lead to either - an error or a meaningless output from mbedtls_ecdh_get_params. In the - latter case, this could expose at most 5 bits of the private key. - -Bugfix - * Fix a compilation issue with mbedtls_ecp_restart_ctx not being defined - when MBEDTLS_ECP_ALT is defined. Reported by jwhui. Fixes #2242. - * Run the AD too long test only if MBEDTLS_CCM_ALT is not defined. - Raised as a comment in #1996. - * Reduce the stack consumption of mbedtls_mpi_fill_random() which could - previously lead to a stack overflow on constrained targets. - * Add `MBEDTLS_SELF_TEST` for the mbedtls_self_test functions - in the header files, which missed the precompilation check. #971 - * Fix returning the value 1 when mbedtls_ecdsa_genkey failed. - * Remove a duplicate #include in a sample program. Fixed by Masashi Honma #2326. - * Remove the mbedtls namespacing from the header file, to fix a "file not found" - build error. Fixed by Haijun Gu #2319. - * Fix signed-to-unsigned integer conversion warning - in X.509 module. Fixes #2212. - * Reduce stack usage of `mpi_write_hlp()` by eliminating recursion. - Fixes #2190. - * Fix false failure in all.sh when backup files exist in include/mbedtls - (e.g. config.h.bak). Fixed by Peter Kolbus (Garmin) #2407. - * Ensure that unused bits are zero when writing ASN.1 bitstrings when using - mbedtls_asn1_write_bitstring(). - * Fix issue when writing the named bitstrings in KeyUsage and NsCertType - extensions in CSRs and CRTs that caused these bitstrings to not be encoded - correctly as trailing zeroes were not accounted for as unused bits in the - leading content octet. Fixes #1610. - -Changes - * Reduce RAM consumption during session renegotiation by not storing - the peer CRT chain and session ticket twice. - * Include configuration file in all header files that use configuration, - instead of relying on other header files that they include. - Inserted as an enhancement for #1371 - * Add support for alternative CSR headers, as used by Microsoft and defined - in RFC 7468. Found by Michael Ernst. Fixes #767. - * Correct many misspellings. Fixed by MisterDA #2371. - * Provide an abstraction of vsnprintf to allow alternative implementations - for platforms that don't provide it. Based on contributions by Joris Aerts - and Nathaniel Wesley Filardo. - * Fix clobber list in MIPS assembly for large integer multiplication. - Previously, this could lead to functionally incorrect assembly being - produced by some optimizing compilers, showing up as failures in - e.g. RSA or ECC signature operations. Reported in #1722, fix suggested - by Aurelien Jarno and submitted by Jeffrey Martin. - * Reduce the complexity of the timing tests. They were assuming more than the - underlying OS actually guarantees. - * Fix configuration queries in ssl-opt.h. #2030 - * Ensure that ssl-opt.h can be run in OS X. #2029 - * Re-enable certain interoperability tests in ssl-opt.sh which had previously - been disabled for lack of a sufficiently recent version of GnuTLS on the CI. - * Ciphersuites based on 3DES now have the lowest priority by default when - they are enabled. - -= mbed TLS 2.16.0 branch released 2018-12-21 - -Features - * Add a new config.h option of MBEDTLS_CHECK_PARAMS that enables validation - of parameters in the API. This allows detection of obvious misuses of the - API, such as passing NULL pointers. The API of existing functions hasn't - changed, but requirements on parameters have been made more explicit in - the documentation. See the corresponding API documentation for each - function to see for which parameter values it is defined. This feature is - disabled by default. See its API documentation in config.h for additional - steps you have to take when enabling it. - -API Changes - * The following functions in the random generator modules have been - deprecated and replaced as shown below. The new functions change - the return type from void to int to allow returning error codes when - using MBEDTLS_<MODULE>_ALT for the underlying AES or message digest - primitive. Fixes #1798. - mbedtls_ctr_drbg_update() -> mbedtls_ctr_drbg_update_ret() - mbedtls_hmac_drbg_update() -> mbedtls_hmac_drbg_update_ret() - * Extend ECDH interface to enable alternative implementations. - * Deprecate error codes of the form MBEDTLS_ERR_xxx_INVALID_KEY_LENGTH for - ARIA, CAMELLIA and Blowfish. These error codes will be replaced by - the more generic per-module error codes MBEDTLS_ERR_xxx_BAD_INPUT_DATA. - * Additional parameter validation checks have been added for the following - modules - AES, ARIA, Blowfish, CAMELLIA, CCM, GCM, DHM, ECP, ECDSA, ECDH, - ECJPAKE, SHA, Chacha20 and Poly1305, cipher, pk, RSA, and MPI. - Where modules have had parameter validation added, existing parameter - checks may have changed. Some modules, such as Chacha20 had existing - parameter validation whereas other modules had little. This has now been - changed so that the same level of validation is present in all modules, and - that it is now optional with the MBEDTLS_CHECK_PARAMS flag which by default - is off. That means that checks which were previously present by default - will no longer be. - -New deprecations - * Deprecate mbedtls_ctr_drbg_update and mbedtls_hmac_drbg_update - in favor of functions that can return an error code. - -Bugfix - * Fix for Clang, which was reporting a warning for the bignum.c inline - assembly for AMD64 targets creating string literals greater than those - permitted by the ISO C99 standard. Found by Aaron Jones. Fixes #482. - * Fix runtime error in `mbedtls_platform_entropy_poll()` when run - through qemu user emulation. Reported and fix suggested by randombit - in #1212. Fixes #1212. - * Fix an unsafe bounds check when restoring an SSL session from a ticket. - This could lead to a buffer overflow, but only in case ticket authentication - was broken. Reported and fix suggested by Guido Vranken in #659. - * Add explicit integer to enumeration type casts to example program - programs/pkey/gen_key which previously led to compilation failure - on some toolchains. Reported by phoenixmcallister. Fixes #2170. - * Fix double initialization of ECC hardware that made some accelerators - hang. - * Clarify documentation of mbedtls_ssl_set_own_cert() regarding the absence - of check for certificate/key matching. Reported by Attila Molnar, #507. - - = mbed TLS 2.15.1 branch released 2018-11-30 - - Changes - * Update the Mbed Crypto submodule to version 0.1.0b2. - - = mbed TLS 2.15.0 branch released 2018-11-23 - - Features - * Add an experimental build option, USE_CRYPTO_SUBMODULE, to enable use of - Mbed Crypto as the source of the cryptography implementation. - * Add an experimental configuration option, MBEDTLS_PSA_CRYPTO_C, to enable - the PSA Crypto API from Mbed Crypto when additionally used with the - USE_CRYPTO_SUBMODULE build option. - - Changes - * Add unit tests for AES-GCM when called through mbedtls_cipher_auth_xxx() - from the cipher abstraction layer. Fixes #2198. - -= mbed TLS 2.14.1 branch released 2018-11-30 - -Security - * Fix timing variations and memory access variations in RSA PKCS#1 v1.5 - decryption that could lead to a Bleichenbacher-style padding oracle - attack. In TLS, this affects servers that accept ciphersuites based on - RSA decryption (i.e. ciphersuites whose name contains RSA but not - (EC)DH(E)). Discovered by Eyal Ronen (Weizmann Institute), Robert Gillham - (University of Adelaide), Daniel Genkin (University of Michigan), - Adi Shamir (Weizmann Institute), David Wong (NCC Group), and Yuval Yarom - (University of Adelaide, Data61). The attack is described in more detail - in the paper available here: http://cat.eyalro.net/cat.pdf CVE-2018-19608 - * In mbedtls_mpi_write_binary(), don't leak the exact size of the number - via branching and memory access patterns. An attacker who could submit - a plaintext for RSA PKCS#1 v1.5 decryption but only observe the timing - of the decryption and not its result could nonetheless decrypt RSA - plaintexts and forge RSA signatures. Other asymmetric algorithms may - have been similarly vulnerable. Reported by Eyal Ronen, Robert Gillham, - Daniel Genkin, Adi Shamir, David Wong and Yuval Yarom. - * Wipe sensitive buffers on the stack in the CTR_DRBG and HMAC_DRBG - modules. - -API Changes - * The new functions mbedtls_ctr_drbg_update_ret() and - mbedtls_hmac_drbg_update_ret() are similar to mbedtls_ctr_drbg_update() - and mbedtls_hmac_drbg_update() respectively, but the new functions - report errors whereas the old functions return void. We recommend that - applications use the new functions. - -= mbed TLS 2.14.0 branch released 2018-11-19 - -Security - * Fix overly strict DN comparison when looking for CRLs belonging to a - particular CA. This previously led to ignoring CRLs when the CRL's issuer - name and the CA's subject name differed in their string encoding (e.g., - one using PrintableString and the other UTF8String) or in the choice of - upper and lower case. Reported by Henrik Andersson of Bosch GmbH in issue - #1784. - * Fix a flawed bounds check in server PSK hint parsing. In case the - incoming message buffer was placed within the first 64KiB of address - space and a PSK-(EC)DHE ciphersuite was used, this allowed an attacker - to trigger a memory access up to 64KiB beyond the incoming message buffer, - potentially leading to an application crash or information disclosure. - * Fix mbedtls_mpi_is_prime() to use more rounds of probabilistic testing. The - previous settings for the number of rounds made it practical for an - adversary to construct non-primes that would be erroneously accepted as - primes with high probability. This does not have an impact on the - security of TLS, but can matter in other contexts with numbers chosen - potentially by an adversary that should be prime and can be validated. - For example, the number of rounds was enough to securely generate RSA key - pairs or Diffie-Hellman parameters, but was insufficient to validate - Diffie-Hellman parameters properly. - See "Prime and Prejudice" by by Martin R. Albrecht and Jake Massimo and - Kenneth G. Paterson and Juraj Somorovsky. - -Features - * Add support for temporarily suspending expensive ECC computations after - some configurable amount of operations. This is intended to be used in - constrained, single-threaded systems where ECC is time consuming and can - block other operations until they complete. This is disabled by default, - but can be enabled by MBEDTLS_ECP_RESTARTABLE at compile time and - configured by mbedtls_ecp_set_max_ops() at runtime. It applies to the new - xxx_restartable functions in ECP, ECDSA, PK and X.509 (CRL not supported - yet), and to existing functions in ECDH and SSL (currently only - implemented client-side, for ECDHE-ECDSA ciphersuites in TLS 1.2, - including client authentication). - * Add support for Arm CPU DSP extensions to accelerate asymmetric key - operations. On CPUs where the extensions are available, they can accelerate - MPI multiplications used in ECC and RSA cryptography. Contributed by - Aurelien Jarno. - * Extend RSASSA-PSS signature to allow a smaller salt size. Previously, PSS - signature always used a salt with the same length as the hash, and returned - an error if this was not possible. Now the salt size may be up to two bytes - shorter. This allows the library to support all hash and signature sizes - that comply with FIPS 186-4, including SHA-512 with a 1024-bit key. - * Add support for 128-bit keys in CTR_DRBG. Note that using keys shorter - than 256 bits limits the security of generated material to 128 bits. - -API Changes - * Add a common error code of `MBEDTLS_ERR_PLATFORM_FEATURE_UNSUPPORTED` for - a feature that is not supported by underlying alternative - implementations implementing cryptographic primitives. This is useful for - hardware accelerators that don't implement all options or features. - -New deprecations - * All module specific errors following the form - MBEDTLS_ERR_XXX_FEATURE_UNAVAILABLE that indicate a feature is not - supported are deprecated and are now replaced by the new equivalent - platform error. - * All module specific generic hardware acceleration errors following the - form MBEDTLS_ERR_XXX_HW_ACCEL_FAILED that are deprecated and are replaced - by the equivalent plaform error. - * Deprecate the function mbedtls_mpi_is_prime() in favor of - mbedtls_mpi_is_prime_ext() which allows specifying the number of - Miller-Rabin rounds. - -Bugfix - * Fix wrong order of freeing in programs/ssl/ssl_server2 example - application leading to a memory leak in case both - MBEDTLS_MEMORY_BUFFER_ALLOC_C and MBEDTLS_MEMORY_BACKTRACE are set. - Fixes #2069. - * Fix a bug in the update function for SSL ticket keys which previously - invalidated keys of a lifetime of less than a 1s. Fixes #1968. - * Fix failure in hmac_drbg in the benchmark sample application, when - MBEDTLS_THREADING_C is defined. Found by TrinityTonic, #1095 - * Fix a bug in the record decryption routine ssl_decrypt_buf() - which lead to accepting properly authenticated but improperly - padded records in case of CBC ciphersuites using Encrypt-then-MAC. - * Fix memory leak and freeing without initialization in the example - program programs/x509/cert_write. Fixes #1422. - * Ignore IV in mbedtls_cipher_set_iv() when the cipher mode is - MBEDTLS_MODE_ECB. Found by ezdevelop. Fixes #1091. - * Zeroize memory used for buffering or reassembling handshake messages - after use. - * Use `mbedtls_platform_zeroize()` instead of `memset()` for zeroization - of sensitive data in the example programs aescrypt2 and crypt_and_hash. - * Change the default string format used for various X.509 DN attributes to - UTF8String. Previously, the use of the PrintableString format led to - wildcards and non-ASCII characters being unusable in some DN attributes. - Reported by raprepo in #1860 and by kevinpt in #468. Fix contributed by - Thomas-Dee. - * Fix compilation failure for configurations which use compile time - replacements of standard calloc/free functions through the macros - MBEDTLS_PLATFORM_CALLOC_MACRO and MBEDTLS_PLATFORM_FREE_MACRO. - Reported by ole-de and ddhome2006. Fixes #882, #1642 and #1706. - -Changes - * Removed support for Yotta as a build tool. - * Add tests for session resumption in DTLS. - * Close a test gap in (D)TLS between the client side and the server side: - test the handling of large packets and small packets on the client side - in the same way as on the server side. - * Change the dtls_client and dtls_server samples to work by default over - IPv6 and optionally by a build option over IPv4. - * Change the use of Windows threading to use Microsoft Visual C++ runtime - calls, rather than Win32 API calls directly. This is necessary to avoid - conflict with C runtime usage. Found and fixed by irwir. - * Remember the string format of X.509 DN attributes when replicating - X.509 DNs. Previously, DN attributes were always written in their default - string format (mostly PrintableString), which could lead to CRTs being - created which used PrintableStrings in the issuer field even though the - signing CA used UTF8Strings in its subject field; while X.509 compliant, - such CRTs were rejected in some applications, e.g. some versions of - Firefox, curl and GnuTLS. Reported in #1033 by Moschn. Fix contributed by - Thomas-Dee. - * Improve documentation of mbedtls_ssl_get_verify_result(). - Fixes #517 reported by github-monoculture. - * Add MBEDTLS_MPI_GEN_PRIME_FLAG_LOW_ERR flag to mbedtls_mpi_gen_prime() and - use it to reduce error probability in RSA key generation to levels mandated - by FIPS-186-4. - -= mbed TLS 2.13.1 branch released 2018-09-06 - -API Changes - * Extend the platform module with an abstraction mbedtls_platform_gmtime_r() - whose implementation should behave as a thread-safe version of gmtime(). - This allows users to configure such an implementation at compile time when - the target system cannot be deduced automatically, by setting the option - MBEDTLS_PLATFORM_GMTIME_R_ALT. At this stage Mbed TLS is only able to - automatically select implementations for Windows and POSIX C libraries. - -Bugfix - * Fix build failures on platforms where only gmtime() is available but - neither gmtime_r() nor gmtime_s() are present. Fixes #1907. - -= mbed TLS 2.13.0 branch released 2018-08-31 - -Security - * Fix an issue in the X.509 module which could lead to a buffer overread - during certificate extensions parsing. In case of receiving malformed - input (extensions length field equal to 0), an illegal read of one byte - beyond the input buffer is made. Found and analyzed by Nathan Crandall. - -Features - * Add support for fragmentation of outgoing DTLS handshake messages. This - is controlled by the maximum fragment length as set locally or negotiated - with the peer, as well as by a new per-connection MTU option, set using - mbedtls_ssl_set_mtu(). - * Add support for auto-adjustment of MTU to a safe value during the - handshake when flights do not get through (RFC 6347, section 4.1.1.1, - last paragraph). - * Add support for packing multiple records within a single datagram, - enabled by default. - * Add support for buffering out-of-order handshake messages in DTLS. - The maximum amount of RAM used for this can be controlled by the - compile-time constant MBEDTLS_SSL_DTLS_MAX_BUFFERING defined - in mbedtls/config.h. - -API Changes - * Add function mbedtls_ssl_set_datagram_packing() to configure - the use of datagram packing (enabled by default). - -Bugfix - * Fix a potential memory leak in mbedtls_ssl_setup() function. An allocation - failure in the function could lead to other buffers being leaked. - * Fixes an issue with MBEDTLS_CHACHAPOLY_C which would not compile if - MBEDTLS_ARC4_C and MBEDTLS_CIPHER_NULL_CIPHER weren't also defined. #1890 - * Fix a memory leak in ecp_mul_comb() if ecp_precompute_comb() fails. - Fix contributed by Espressif Systems. - * Add ecc extensions only if an ecc based ciphersuite is used. - This improves compliance to RFC 4492, and as a result, solves - interoperability issues with BouncyCastle. Raised by milenamil in #1157. - * Replace printf with mbedtls_printf in the ARIA module. Found by - TrinityTonic in #1908. - * Fix potential use-after-free in mbedtls_ssl_get_max_frag_len() - and mbedtls_ssl_get_record_expansion() after a session reset. Fixes #1941. - * Fix a bug that caused SSL/TLS clients to incorrectly abort the handshake - with TLS versions 1.1 and earlier when the server requested authentication - without providing a list of CAs. This was due to an overly strict bounds - check in parsing the CertificateRequest message, - introduced in Mbed TLS 2.12.0. Fixes #1954. - * Fix a miscalculation of the maximum record expansion in - mbedtls_ssl_get_record_expansion() in case of ChachaPoly ciphersuites, - or CBC ciphersuites in (D)TLS versions 1.1 or higher. Fixes #1913, #1914. - * Fix undefined shifts with negative values in certificates parsing - (found by Catena cyber using oss-fuzz) - * Fix memory leak and free without initialization in pk_encrypt - and pk_decrypt example programs. Reported by Brace Stout. Fixes #1128. - * Remove redundant else statement. Raised by irwir. Fixes #1776. - -Changes - * Copy headers preserving timestamps when doing a "make install". - Contributed by xueruini. - * Allow the forward declaration of public structs. Contributed by Dawid - Drozd. Fixes #1215 raised by randombit. - * Improve compatibility with some alternative CCM implementations by using - CCM test vectors from RAM. - * Add support for buffering of out-of-order handshake messages. - * Add warnings to the documentation of the HKDF module to reduce the risk - of misusing the mbedtls_hkdf_extract() and mbedtls_hkdf_expand() - functions. Fixes #1775. Reported by Brian J. Murray. - -= mbed TLS 2.12.0 branch released 2018-07-25 - -Security - * Fix a vulnerability in TLS ciphersuites based on CBC and using SHA-384, - in (D)TLS 1.0 to 1.2, that allowed an active network attacker to - partially recover the plaintext of messages under some conditions by - exploiting timing measurements. With DTLS, the attacker could perform - this recovery by sending many messages in the same connection. With TLS - or if mbedtls_ssl_conf_dtls_badmac_limit() was used, the attack only - worked if the same secret (for example a HTTP Cookie) has been repeatedly - sent over connections manipulated by the attacker. Connections using GCM - or CCM instead of CBC, using hash sizes other than SHA-384, or using - Encrypt-then-Mac (RFC 7366) were not affected. The vulnerability was - caused by a miscalculation (for SHA-384) in a countermeasure to the - original Lucky 13 attack. Found by Kenny Paterson, Eyal Ronen and Adi - Shamir. - * Fix a vulnerability in TLS ciphersuites based on CBC, in (D)TLS 1.0 to - 1.2, that allowed a local attacker, able to execute code on the local - machine as well as manipulate network packets, to partially recover the - plaintext of messages under some conditions by using a cache attack - targeting an internal MD/SHA buffer. With TLS or if - mbedtls_ssl_conf_dtls_badmac_limit() was used, the attack only worked if - the same secret (for example a HTTP Cookie) has been repeatedly sent over - connections manipulated by the attacker. Connections using GCM or CCM - instead of CBC or using Encrypt-then-Mac (RFC 7366) were not affected. - Found by Kenny Paterson, Eyal Ronen and Adi Shamir. - * Add a counter-measure against a vulnerability in TLS ciphersuites based - on CBC, in (D)TLS 1.0 to 1.2, that allowed a local attacker, able to - execute code on the local machine as well as manipulate network packets, - to partially recover the plaintext of messages under some conditions (see - previous entry) by using a cache attack targeting the SSL input record - buffer. Connections using GCM or CCM instead of CBC or using - Encrypt-then-Mac (RFC 7366) were not affected. Found by Kenny Paterson, - Eyal Ronen and Adi Shamir. - -Features - * Add new crypto primitives from RFC 7539: stream cipher Chacha20, one-time - authenticator Poly1305 and AEAD construct Chacha20-Poly1305. Contributed - by Daniel King. - * Add support for CHACHA20-POLY1305 ciphersuites from RFC 7905. - * Add platform support for the Haiku OS. (https://www.haiku-os.org). - Contributed by Augustin Cavalier. - * Make the receive and transmit buffers independent sizes, for situations - where the outgoing buffer can be fixed at a smaller size than the incoming - buffer, which can save some RAM. If buffer lengths are kept equal, there - is no functional difference. Contributed by Angus Gratton, and also - independently contributed again by Paul Sokolovsky. - * Add support for key wrapping modes based on AES as defined by - NIST SP 800-38F algorithms KW and KWP and by RFC 3394 and RFC 5649. - -Bugfix - * Fix the key_app_writer example which was writing a leading zero byte which - was creating an invalid ASN.1 tag. Found by Aryeh R. Fixes #1257. - * Fix compilation error on C++, because of a variable named new. - Found and fixed by Hirotaka Niisato in #1783. - * Fix "no symbols" warning issued by ranlib when building on Mac OS X. Fix - contributed by tabascoeye. - * Clarify documentation for mbedtls_ssl_write() to include 0 as a valid - return value. Found by @davidwu2000. #839 - * Fix a memory leak in mbedtls_x509_csr_parse(), found by catenacyber, - Philippe Antoine. Fixes #1623. - * Remove unused headers included in x509.c. Found by Chris Hanson and fixed - by Brendan Shanks. Part of a fix for #992. - * Fix compilation error when MBEDTLS_ARC4_C is disabled and - MBEDTLS_CIPHER_NULL_CIPHER is enabled. Found by TrinityTonic in #1719. - * Added length checks to some TLS parsing functions. Found and fixed by - Philippe Antoine from Catena cyber. #1663. - * Fix the inline assembly for the MPI multiply helper function for i386 and - i386 with SSE2. Found by László Langó. Fixes #1550 - * Fix namespacing in header files. Remove the `mbedtls` namespacing in - the `#include` in the header files. Resolves #857 - * Fix compiler warning of 'use before initialisation' in - mbedtls_pk_parse_key(). Found by Martin Boye Petersen and fixed by Dawid - Drozd. #1098 - * Fix decryption for zero length messages (which contain all padding) when a - CBC based ciphersuite is used together with Encrypt-then-MAC. Previously, - such a message was wrongly reported as an invalid record and therefore lead - to the connection being terminated. Seen most often with OpenSSL using - TLS 1.0. Reported by @kFYatek and by Conor Murphy on the forum. Fix - contributed by Espressif Systems. Fixes #1632 - * Fix ssl_client2 example to send application data with 0-length content - when the request_size argument is set to 0 as stated in the documentation. - Fixes #1833. - * Correct the documentation for `mbedtls_ssl_get_session()`. This API has - deep copy of the session, and the peer certificate is not lost. Fixes #926. - * Fix build using -std=c99. Fixed by Nick Wilson. - -Changes - * Fail when receiving a TLS alert message with an invalid length, or invalid - zero-length messages when using TLS 1.2. Contributed by Espressif Systems. - * Change the default behaviour of mbedtls_hkdf_extract() to return an error - when calling with a NULL salt and non-zero salt_len. Contributed by - Brian J Murray - * Change the shebang line in Perl scripts to look up perl in the PATH. - Contributed by fbrosson. - * Allow overriding the time on Windows via the platform-time abstraction. - Fixed by Nick Wilson. - * Use gmtime_r/gmtime_s for thread-safety. Fixed by Nick Wilson. - -= mbed TLS 2.11.0 branch released 2018-06-18 - -Features - * Add additional block mode, OFB (Output Feedback), to the AES module and - cipher abstraction module. - * Implement the HMAC-based extract-and-expand key derivation function - (HKDF) per RFC 5869. Contributed by Thomas Fossati. - * Add support for the CCM* block cipher mode as defined in IEEE Std. 802.15.4. - * Add support for the XTS block cipher mode with AES (AES-XTS). - Contributed by Aorimn in pull request #414. - * In TLS servers, support offloading private key operations to an external - cryptoprocessor. Private key operations can be asynchronous to allow - non-blocking operation of the TLS server stack. - -Bugfix - * Fix the cert_write example to handle certificates signed with elliptic - curves as well as RSA. Fixes #777 found by dbedev. - * Fix for redefinition of _WIN32_WINNT to avoid overriding a definition - used by user applications. Found and fixed by Fabio Alessandrelli. - * Fix compilation warnings with IAR toolchain, on 32 bit platform. - Reported by rahmanih in #683 - * Fix braces in mbedtls_memory_buffer_alloc_status(). Found by sbranden, #552. - -Changes - * Changed CMake defaults for IAR to treat all compiler warnings as errors. - * Changed the Clang parameters used in the CMake build files to work for - versions later than 3.6. Versions of Clang earlier than this may no longer - work. Fixes #1072 - -= mbed TLS 2.10.0 branch released 2018-06-06 - -Features - * Add support for ARIA cipher (RFC 5794) and associated TLS ciphersuites - (RFC 6209). Disabled by default, see MBEDTLS_ARIA_C in config.h - -API Changes - * Extend the platform module with a util component that contains - functionality shared by multiple Mbed TLS modules. At this stage - platform_util.h (and its associated platform_util.c) only contain - mbedtls_platform_zeroize(), which is a critical function from a security - point of view. mbedtls_platform_zeroize() needs to be regularly tested - against compilers to ensure that calls to it are not removed from the - output binary as part of redundant code elimination optimizations. - Therefore, mbedtls_platform_zeroize() is moved to the platform module to - facilitate testing and maintenance. - -Bugfix - * Fix an issue with MicroBlaze support in bn_mul.h which was causing the - build to fail. Found by zv-io. Fixes #1651. - -Changes - * Support TLS testing in out-of-source builds using cmake. Fixes #1193. - * Fix redundant declaration of mbedtls_ssl_list_ciphersuites. Raised by - TrinityTonic. #1359. - -= mbed TLS 2.9.0 branch released 2018-04-30 - -Security - * Fix an issue in the X.509 module which could lead to a buffer overread - during certificate validation. Additionally, the issue could also lead to - unnecessary callback checks being made or to some validation checks to be - omitted. The overread could be triggered remotely, while the other issues - would require a non DER-compliant certificate to be correctly signed by a - trusted CA, or a trusted CA with a non DER-compliant certificate. Found by - luocm. Fixes #825. - * Fix the buffer length assertion in the ssl_parse_certificate_request() - function which led to an arbitrary overread of the message buffer. The - overreads could be caused by receiving a malformed message at the point - where an optional signature algorithms list is expected when the signature - algorithms section is too short. In builds with debug output, the overread - data is output with the debug data. - * Fix a client-side bug in the validation of the server's ciphersuite choice - which could potentially lead to the client accepting a ciphersuite it didn't - offer or a ciphersuite that cannot be used with the TLS or DTLS version - chosen by the server. This could lead to corruption of internal data - structures for some configurations. - -Features - * Add an option, MBEDTLS_AES_FEWER_TABLES, to dynamically compute smaller AES - tables during runtime, thereby reducing the RAM/ROM footprint by ~6KiB. - Suggested and contributed by jkivilin in pull request #394. - * Add initial support for Curve448 (RFC 7748). Only mbedtls_ecp_mul() and - ECDH primitive functions (mbedtls_ecdh_gen_public(), - mbedtls_ecdh_compute_shared()) are supported for now. Contributed by - Nicholas Wilson in pull request #348. - -API Changes - * Extend the public API with the function of mbedtls_net_poll() to allow user - applications to wait for a network context to become ready before reading - or writing. - * Add function mbedtls_ssl_check_pending() to the public API to allow - a check for whether more more data is pending to be processed in the - internal message buffers. - This function is necessary to determine when it is safe to idle on the - underlying transport in case event-driven IO is used. - -Bugfix - * Fix a spurious uninitialized variable warning in cmac.c. Fix independently - contributed by Brian J Murray and David Brown. - * Add missing dependencies in test suites that led to build failures - in configurations that omit certain hashes or public-key algorithms. - Fixes #1040. - * Fix C89 incompatibility in benchmark.c. Contributed by Brendan Shanks. - #1353 - * Add missing dependencies for MBEDTLS_HAVE_TIME_DATE and - MBEDTLS_VERSION_FEATURES in some test suites. Contributed by - Deomid Ryabkov. Fixes #1299, #1475. - * Fix the Makefile build process for building shared libraries on Mac OS X. - Fixed by mnacamura. - * Fix parsing of PKCS#8 encoded Elliptic Curve keys. Previously Mbed TLS was - unable to parse keys which had only the optional parameters field of the - ECPrivateKey structure. Found by Jethro Beekman, fixed in #1379. - * Return the plaintext data more quickly on unpadded CBC decryption, as - stated in the mbedtls_cipher_update() documentation. Contributed by - Andy Leiserson. - * Fix overriding and ignoring return values when parsing and writing to - a file in pk_sign program. Found by kevlut in #1142. - * Restrict usage of error code MBEDTLS_ERR_SSL_WANT_READ to situations - where data needs to be fetched from the underlying transport in order - to make progress. Previously, this error code was also occasionally - returned when unexpected messages were being discarded, ignoring that - further messages could potentially already be pending to be processed - in the internal buffers; these cases led to deadlocks when event-driven - I/O was used. Found and reported by Hubert Mis in #772. - * Fix buffer length assertions in the ssl_parse_certificate_request() - function which leads to a potential one byte overread of the message - buffer. - * Fix invalid buffer sizes passed to zlib during record compression and - decompression. - * Fix the soversion of libmbedcrypto to match the soversion of the - maintained 2.7 branch. The soversion was increased in Mbed TLS - version 2.7.1 to reflect breaking changes in that release, but the - increment was missed in 2.8.0 and later releases outside of the 2.7 branch. - -Changes - * Remove some redundant code in bignum.c. Contributed by Alexey Skalozub. - * Support cmake builds where Mbed TLS is a subproject. Fix contributed - independently by Matthieu Volat and Arne Schwabe. - * Improve testing in configurations that omit certain hashes or - public-key algorithms. Includes contributions by Gert van Dijk. - * Improve negative testing of X.509 parsing. - * Do not define global mutexes around readdir() and gmtime() in - configurations where the feature is disabled. Found and fixed by Gergely - Budai. - * Harden the function mbedtls_ssl_config_free() against misuse, so that it - doesn't leak memory if the user doesn't use mbedtls_ssl_conf_psk() and - instead incorrectly manipulates the configuration structure directly. - Found and fix submitted by junyeonLEE in #1220. - * Provide an empty implementation of mbedtls_pkcs5_pbes2() when - MBEDTLS_ASN1_PARSE_C is not enabled. This allows the use of PBKDF2 - without PBES2. Fixed by Marcos Del Sol Vives. - * Add the order of the base point as N in the mbedtls_ecp_group structure - for Curve25519 (other curves had it already). Contributed by Nicholas - Wilson #481 - * Improve the documentation of mbedtls_net_accept(). Contributed by Ivan - Krylov. - * Improve the documentation of mbedtls_ssl_write(). Suggested by - Paul Sokolovsky in #1356. - * Add an option in the Makefile to support ar utilities where the operation - letter must not be prefixed by '-', such as LLVM. Found and fixed by - Alex Hixon. - * Allow configuring the shared library extension by setting the DLEXT - environment variable when using the project makefiles. - * Optimize unnecessary zeroing in mbedtls_mpi_copy. Based on a contribution - by Alexey Skalozub in #405. - * In the SSL module, when f_send, f_recv or f_recv_timeout report - transmitting more than the required length, return an error. Raised by - Sam O'Connor in #1245. - * Improve robustness of mbedtls_ssl_derive_keys against the use of - HMAC functions with non-HMAC ciphersuites. Independently contributed - by Jiayuan Chen in #1377. Fixes #1437. - * Improve security of RSA key generation by including criteria from - FIPS 186-4. Contributed by Jethro Beekman. #1380 - * Declare functions in header files even when an alternative implementation - of the corresponding module is activated by defining the corresponding - MBEDTLS_XXX_ALT macro. This means that alternative implementations do - not need to copy the declarations, and ensures that they will have the - same API. - * Add platform setup and teardown calls in test suites. - -= mbed TLS 2.8.0 branch released 2018-03-16 - -Default behavior changes - * The truncated HMAC extension now conforms to RFC 6066. This means - that when both sides of a TLS connection negotiate the truncated - HMAC extension, Mbed TLS can now interoperate with other - compliant implementations, but this breaks interoperability with - prior versions of Mbed TLS. To restore the old behavior, enable - the (deprecated) option MBEDTLS_SSL_TRUNCATED_HMAC_COMPAT in - config.h. Found by Andreas Walz (ivESK, Offenburg University of - Applied Sciences). - -Security - * Fix implementation of the truncated HMAC extension. The previous - implementation allowed an offline 2^80 brute force attack on the - HMAC key of a single, uninterrupted connection (with no - resumption of the session). - * Verify results of RSA private key operations to defend - against Bellcore glitch attack. - * Fix a buffer overread in ssl_parse_server_key_exchange() that could cause - a crash on invalid input. - * Fix a buffer overread in ssl_parse_server_psk_hint() that could cause a - crash on invalid input. - * Fix CRL parsing to reject CRLs containing unsupported critical - extensions. Found by Falko Strenzke and Evangelos Karatsiolis. - -Features - * Extend PKCS#8 interface by introducing support for the entire SHA - algorithms family when encrypting private keys using PKCS#5 v2.0. - This allows reading encrypted PEM files produced by software that - uses PBKDF2-SHA2, such as OpenSSL 1.1. Submitted by Antonio Quartulli, - OpenVPN Inc. Fixes #1339 - * Add support for public keys encoded in PKCS#1 format. #1122 - -New deprecations - * Deprecate support for record compression (configuration option - MBEDTLS_ZLIB_SUPPORT). - -Bugfix - * Fix the name of a DHE parameter that was accidentally changed in 2.7.0. - Fixes #1358. - * Fix test_suite_pk to work on 64-bit ILP32 systems. #849 - * Fix mbedtls_x509_crt_profile_suiteb, which used to reject all certificates - with flag MBEDTLS_X509_BADCERT_BAD_PK even when the key type was correct. - In the context of SSL, this resulted in handshake failure. Reported by - daniel in the Mbed TLS forum. #1351 - * Fix Windows x64 builds with the included mbedTLS.sln file. #1347 - * Fix setting version TLSv1 as minimal version, even if TLS 1 - is not enabled. Set MBEDTLS_SSL_MIN_MAJOR_VERSION - and MBEDTLS_SSL_MIN_MINOR_VERSION instead of - MBEDTLS_SSL_MAJOR_VERSION_3 and MBEDTLS_SSL_MINOR_VERSION_1. #664 - * Fix compilation error on Mingw32 when _TRUNCATE is defined. Use _TRUNCATE - only if __MINGW32__ not defined. Fix suggested by Thomas Glanzmann and - Nick Wilson on issue #355 - * In test_suite_pk, pass valid parameters when testing for hash length - overflow. #1179 - * Fix memory allocation corner cases in memory_buffer_alloc.c module. Found - by Guido Vranken. #639 - * Log correct number of ciphersuites used in Client Hello message. #918 - * Fix X509 CRT parsing that would potentially accept an invalid tag when - parsing the subject alternative names. - * Fix a possible arithmetic overflow in ssl_parse_server_key_exchange() - that could cause a key exchange to fail on valid data. - * Fix a possible arithmetic overflow in ssl_parse_server_psk_hint() that - could cause a key exchange to fail on valid data. - * Don't define mbedtls_aes_decrypt and mbedtls_aes_encrypt under - MBEDTLS_DEPRECATED_REMOVED. #1388 - * Fix a 1-byte heap buffer overflow (read-only) during private key parsing. - Found through fuzz testing. - -Changes - * Fix tag lengths and value ranges in the documentation of CCM encryption. - Contributed by Mathieu Briand. - * Fix typo in a comment ctr_drbg.c. Contributed by Paul Sokolovsky. - * Remove support for the library reference configuration for picocoin. - * MD functions deprecated in 2.7.0 are no longer inline, to provide - a migration path for those depending on the library's ABI. - * Clarify the documentation of mbedtls_ssl_setup. - * Use (void) when defining functions with no parameters. Contributed by - Joris Aerts. #678 - -= mbed TLS 2.7.0 branch released 2018-02-03 - -Security - * Fix a heap corruption issue in the implementation of the truncated HMAC - extension. When the truncated HMAC extension is enabled and CBC is used, - sending a malicious application packet could be used to selectively corrupt - 6 bytes on the peer's heap, which could potentially lead to crash or remote - code execution. The issue could be triggered remotely from either side in - both TLS and DTLS. CVE-2018-0488 - * Fix a buffer overflow in RSA-PSS verification when the hash was too large - for the key size, which could potentially lead to crash or remote code - execution. Found by Seth Terashima, Qualcomm Product Security Initiative, - Qualcomm Technologies Inc. CVE-2018-0487 - * Fix buffer overflow in RSA-PSS verification when the unmasked data is all - zeros. - * Fix an unsafe bounds check in ssl_parse_client_psk_identity() when adding - 64 KiB to the address of the SSL buffer and causing a wrap around. - * Fix a potential heap buffer overflow in mbedtls_ssl_write(). When the (by - default enabled) maximum fragment length extension is disabled in the - config and the application data buffer passed to mbedtls_ssl_write - is larger than the internal message buffer (16384 bytes by default), the - latter overflows. The exploitability of this issue depends on whether the - application layer can be forced into sending such large packets. The issue - was independently reported by Tim Nordell via e-mail and by Florin Petriuc - and sjorsdewit on GitHub. Fix proposed by Florin Petriuc in #1022. - Fixes #707. - * Add a provision to prevent compiler optimizations breaking the time - constancy of mbedtls_ssl_safer_memcmp(). - * Ensure that buffers are cleared after use if they contain sensitive data. - Changes were introduced in multiple places in the library. - * Set PEM buffer to zero before freeing it, to avoid decoded private keys - being leaked to memory after release. - * Fix dhm_check_range() failing to detect trivial subgroups and potentially - leaking 1 bit of the private key. Reported by prashantkspatil. - * Make mbedtls_mpi_read_binary() constant-time with respect to the input - data. Previously, trailing zero bytes were detected and omitted for the - sake of saving memory, but potentially leading to slight timing - differences. Reported by Marco Macchetti, Kudelski Group. - * Wipe stack buffer temporarily holding EC private exponent - after keypair generation. - * Fix a potential heap buffer over-read in ALPN extension parsing - (server-side). Could result in application crash, but only if an ALPN - name larger than 16 bytes had been configured on the server. - * Change default choice of DHE parameters from untrustworthy RFC 5114 - to RFC 3526 containing parameters generated in a nothing-up-my-sleeve - manner. - -Features - * Allow comments in test data files. - * The selftest program can execute a subset of the tests based on command - line arguments. - * New unit tests for timing. Improve the self-test to be more robust - when run on a heavily-loaded machine. - * Add alternative implementation support for CCM and CMAC (MBEDTLS_CCM_ALT, - MBEDTLS_CMAC_ALT). Submitted by Steven Cooreman, Silicon Labs. - * Add support for alternative implementations of GCM, selected by the - configuration flag MBEDTLS_GCM_ALT. - * Add support for alternative implementations for ECDSA, controlled by new - configuration flags MBEDTLS_ECDSA_SIGN_ALT, MBEDTLS_ECDSA_VERIFY_ALT and - MBEDTLS_ECDSDA_GENKEY_AT in config.h. - The following functions from the ECDSA module can be replaced - with alternative implementation: - mbedtls_ecdsa_sign(), mbedtls_ecdsa_verify() and mbedtls_ecdsa_genkey(). - * Add support for alternative implementation of ECDH, controlled by the - new configuration flags MBEDTLS_ECDH_COMPUTE_SHARED_ALT and - MBEDTLS_ECDH_GEN_PUBLIC_ALT in config.h. - The following functions from the ECDH module can be replaced - with an alternative implementation: - mbedtls_ecdh_gen_public() and mbedtls_ecdh_compute_shared(). - * Add support for alternative implementation of ECJPAKE, controlled by - the new configuration flag MBEDTLS_ECJPAKE_ALT. - * Add mechanism to provide alternative implementation of the DHM module. - -API Changes - * Extend RSA interface by multiple functions allowing structure- - independent setup and export of RSA contexts. Most notably, - mbedtls_rsa_import() and mbedtls_rsa_complete() are introduced for setting - up RSA contexts from partial key material and having them completed to the - needs of the implementation automatically. This allows to setup private RSA - contexts from keys consisting of N,D,E only, even if P,Q are needed for the - purpose or CRT and/or blinding. - * The configuration option MBEDTLS_RSA_ALT can be used to define alternative - implementations of the RSA interface declared in rsa.h. - * The following functions in the message digest modules (MD2, MD4, MD5, - SHA1, SHA256, SHA512) have been deprecated and replaced as shown below. - The new functions change the return type from void to int to allow - returning error codes when using MBEDTLS_<MODULE>_ALT. - mbedtls_<MODULE>_starts() -> mbedtls_<MODULE>_starts_ret() - mbedtls_<MODULE>_update() -> mbedtls_<MODULE>_update_ret() - mbedtls_<MODULE>_finish() -> mbedtls_<MODULE>_finish_ret() - mbedtls_<MODULE>_process() -> mbedtls_internal_<MODULE>_process() - -New deprecations - * Deprecate usage of RSA primitives with non-matching key-type - (e.g. signing with a public key). - * Direct manipulation of structure fields of RSA contexts is deprecated. - Users are advised to use the extended RSA API instead. - * Deprecate usage of message digest functions that return void - (mbedtls_<MODULE>_starts, mbedtls_<MODULE>_update, - mbedtls_<MODULE>_finish and mbedtls_<MODULE>_process where <MODULE> is - any of MD2, MD4, MD5, SHA1, SHA256, SHA512) in favor of functions - that can return an error code. - * Deprecate untrustworthy DHE parameters from RFC 5114. Superseded by - parameters from RFC 3526 or the newly added parameters from RFC 7919. - * Deprecate hex string DHE constants MBEDTLS_DHM_RFC3526_MODP_2048_P etc. - Supserseded by binary encoded constants MBEDTLS_DHM_RFC3526_MODP_2048_P_BIN - etc. - * Deprecate mbedtls_ssl_conf_dh_param() for setting default DHE parameters - from hex strings. Superseded by mbedtls_ssl_conf_dh_param_bin() - accepting DHM parameters in binary form, matching the new constants. - -Bugfix - * Fix ssl_parse_record_header() to silently discard invalid DTLS records - as recommended in RFC 6347 Section 4.1.2.7. - * Fix memory leak in mbedtls_ssl_set_hostname() when called multiple times. - Found by projectgus and Jethro Beekman, #836. - * Fix usage help in ssl_server2 example. Found and fixed by Bei Lin. - * Parse signature algorithm extension when renegotiating. Previously, - renegotiated handshakes would only accept signatures using SHA-1 - regardless of the peer's preferences, or fail if SHA-1 was disabled. - * Fix leap year calculation in x509_date_is_valid() to ensure that invalid - dates on leap years with 100 and 400 intervals are handled correctly. Found - by Nicholas Wilson. #694 - * Fix some invalid RSA-PSS signatures with keys of size 8N+1 that were - accepted. Generating these signatures required the private key. - * Fix out-of-memory problem when parsing 4096-bit PKCS8-encrypted RSA keys. - Found independently by Florian in the mbed TLS forum and by Mishamax. - #878, #1019. - * Fix variable used before assignment compilation warnings with IAR - toolchain. Found by gkerrien38. - * Fix unchecked return codes from AES, DES and 3DES functions in - pem_aes_decrypt(), pem_des_decrypt() and pem_des3_decrypt() respectively. - If a call to one of the functions of the cryptographic primitive modules - failed, the error may not be noticed by the function - mbedtls_pem_read_buffer() causing it to return invalid values. Found by - Guido Vranken. #756 - * Include configuration file in md.h, to fix compilation warnings. - Reported by aaronmdjones in #1001 - * Correct extraction of signature-type from PK instance in X.509 CRT and CSR - writing routines that prevented these functions to work with alternative - RSA implementations. Raised by J.B. in the Mbed TLS forum. Fixes #1011. - * Don't print X.509 version tag for v1 CRT's, and omit extensions for - non-v3 CRT's. - * Fix bugs in RSA test suite under MBEDTLS_NO_PLATFORM_ENTROPY. #1023 #1024 - * Fix net_would_block() to avoid modification by errno through fcntl() call. - Found by nkolban. Fixes #845. - * Fix handling of handshake messages in mbedtls_ssl_read() in case - MBEDTLS_SSL_RENEGOTIATION is disabled. Found by erja-gp. - * Add a check for invalid private parameters in mbedtls_ecdsa_sign(). - Reported by Yolan Romailler. - * Fix word size check in in pk.c to not depend on MBEDTLS_HAVE_INT64. - * Fix incorrect unit in benchmark output. #850 - * Add size-checks for record and handshake message content, securing - fragile yet non-exploitable code-paths. - * Fix crash when calling mbedtls_ssl_cache_free() twice. Found by - MilenkoMitrovic, #1104 - * Fix mbedtls_timing_alarm(0) on Unix and MinGW. - * Fix use of uninitialized memory in mbedtls_timing_get_timer() when reset=1. - * Fix possible memory leaks in mbedtls_gcm_self_test(). - * Added missing return code checks in mbedtls_aes_self_test(). - * Fix issues in RSA key generation program programs/x509/rsa_genkey and the - RSA test suite where the failure of CTR DRBG initialization lead to - freeing an RSA context and several MPI's without proper initialization - beforehand. - * Fix error message in programs/pkey/gen_key.c. Found and fixed by Chris Xue. - * Fix programs/pkey/dh_server.c so that it actually works with dh_client.c. - Found and fixed by Martijn de Milliano. - * Fix an issue in the cipher decryption with the mode - MBEDTLS_PADDING_ONE_AND_ZEROS that sometimes accepted invalid padding. - Note, this padding mode is not used by the TLS protocol. Found and fixed by - Micha Kraus. - * Fix the entropy.c module to not call mbedtls_sha256_starts() or - mbedtls_sha512_starts() in the mbedtls_entropy_init() function. - * Fix the entropy.c module to ensure that mbedtls_sha256_init() or - mbedtls_sha512_init() is called before operating on the relevant context - structure. Do not assume that zeroizing a context is a correct way to - reset it. Found independently by ccli8 on Github. - * In mbedtls_entropy_free(), properly free the message digest context. - * Fix status handshake status message in programs/ssl/dtls_client.c. Found - and fixed by muddog. - -Changes - * Extend cert_write example program by options to set the certificate version - and the message digest. Further, allow enabling/disabling of authority - identifier, subject identifier and basic constraints extensions. - * Only check for necessary RSA structure fields in `mbedtls_rsa_private`. In - particular, don't require P,Q if neither CRT nor blinding are - used. Reported and fix proposed independently by satur9nine and sliai - on GitHub. - * Only run AES-192 self-test if AES-192 is available. Fixes #963. - * Tighten the RSA PKCS#1 v1.5 signature verification code and remove the - undeclared dependency of the RSA module on the ASN.1 module. - * Update all internal usage of deprecated message digest functions to the - new ones with return codes. In particular, this modifies the - mbedtls_md_info_t structure. Propagate errors from these functions - everywhere except some locations in the ssl_tls.c module. - * Improve CTR_DRBG error handling by propagating underlying AES errors. - * Add MBEDTLS_ERR_XXX_HW_ACCEL_FAILED error codes for all cryptography - modules where the software implementation can be replaced by a hardware - implementation. - * Add explicit warnings for the use of MD2, MD4, MD5, SHA-1, DES and ARC4 - throughout the library. - -= mbed TLS 2.6.0 branch released 2017-08-10 - -Security - * Fix authentication bypass in SSL/TLS: when authmode is set to optional, - mbedtls_ssl_get_verify_result() would incorrectly return 0 when the peer's - X.509 certificate chain had more than MBEDTLS_X509_MAX_INTERMEDIATE_CA - (default: 8) intermediates, even when it was not trusted. This could be - triggered remotely from either side. (With authmode set to 'required' - (the default), the handshake was correctly aborted). - * Reliably wipe sensitive data after use in the AES example applications - programs/aes/aescrypt2 and programs/aes/crypt_and_hash. - Found by Laurent Simon. - -Features - * Add the functions mbedtls_platform_setup() and mbedtls_platform_teardown() - and the context struct mbedtls_platform_context to perform - platform-specific setup and teardown operations. The macro - MBEDTLS_PLATFORM_SETUP_TEARDOWN_ALT allows the functions to be overridden - by the user in a platform_alt.h file. These new functions are required in - some embedded environments to provide a means of initialising underlying - cryptographic acceleration hardware. - -API Changes - * Reverted API/ABI breaking changes introduced in mbed TLS 2.5.1, to make the - API consistent with mbed TLS 2.5.0. Specifically removed the inline - qualifier from the functions mbedtls_aes_decrypt, mbedtls_aes_encrypt, - mbedtls_ssl_ciphersuite_uses_ec and mbedtls_ssl_ciphersuite_uses_psk. Found - by James Cowgill. #978 - * Certificate verification functions now set flags to -1 in case the full - chain was not verified due to an internal error (including in the verify - callback) or chain length limitations. - * With authmode set to optional, the TLS handshake is now aborted if the - verification of the peer's certificate failed due to an overlong chain or - a fatal error in the verify callback. - -Bugfix - * Add a check if iv_len is zero in GCM, and return an error if it is zero. - Reported by roberto. #716 - * Replace preprocessor condition from #if defined(MBEDTLS_THREADING_PTHREAD) - to #if defined(MBEDTLS_THREADING_C) as the library cannot assume they will - always be implemented by pthread support. #696 - * Fix a resource leak on Windows platforms in mbedtls_x509_crt_parse_path(), - in the case of an error. Found by redplait. #590 - * Add MBEDTLS_MPI_CHK to check for error value of mbedtls_mpi_fill_random. - Reported and fix suggested by guidovranken. #740 - * Fix conditional preprocessor directives in bignum.h to enable 64-bit - compilation when using ARM Compiler 6. - * Fix a potential integer overflow in the version verification for DER - encoded X.509 CRLs. The overflow could enable maliciously constructed CRLs - to bypass the version verification check. Found by Peng Li/Yueh-Hsun Lin, - KNOX Security, Samsung Research America - * Fix potential integer overflow in the version verification for DER - encoded X.509 CSRs. The overflow could enable maliciously constructed CSRs - to bypass the version verification check. Found by Peng Li/Yueh-Hsun Lin, - KNOX Security, Samsung Research America - * Fix a potential integer overflow in the version verification for DER - encoded X.509 certificates. The overflow could enable maliciously - constructed certificates to bypass the certificate verification check. - * Fix a call to the libc function time() to call the platform abstraction - function mbedtls_time() instead. Found by wairua. #666 - * Avoid shadowing of time and index functions through mbed TLS function - arguments. Found by inestlerode. #557. - -Changes - * Added config.h option MBEDTLS_NO_UDBL_DIVISION, to prevent the use of - 64-bit division. This is useful on embedded platforms where 64-bit division - created a dependency on external libraries. #708 - * Removed mutexes from ECP hardware accelerator code. Now all hardware - accelerator code in the library leaves concurrency handling to the - platform. Reported by Steven Cooreman. #863 - * Define the macro MBEDTLS_AES_ROM_TABLES in the configuration file - config-no-entropy.h to reduce the RAM footprint. - * Added a test script that can be hooked into git that verifies commits - before they are pushed. - * Improve documentation of PKCS1 decryption functions. - -= mbed TLS 2.5.1 released 2017-06-21 - -Security - * Fixed unlimited overread of heap-based buffer in mbedtls_ssl_read(). - The issue could only happen client-side with renegotiation enabled. - Could result in DoS (application crash) or information leak - (if the application layer sent data read from mbedtls_ssl_read() - back to the server or to a third party). Can be triggered remotely. - * Removed SHA-1 and RIPEMD-160 from the default hash algorithms for - certificate verification. SHA-1 can be turned back on with a compile-time - option if needed. - * Fixed offset in FALLBACK_SCSV parsing that caused TLS server to fail to - detect it sometimes. Reported by Hugo Leisink. #810 - * Tighten parsing of RSA PKCS#1 v1.5 signatures, to avoid a - potential Bleichenbacher/BERserk-style attack. - -Bugfix - * Remove size zero arrays from ECJPAKE test suite. Size zero arrays are not - valid C and they prevented the test from compiling in Visual Studio 2015 - and with GCC using the -Wpedantic compilation option. - * Fix insufficient support for signature-hash-algorithm extension, - resulting in compatibility problems with Chrome. Found by hfloyrd. #823 - * Fix behaviour that hid the original cause of fatal alerts in some cases - when sending the alert failed. The fix makes sure not to hide the error - that triggered the alert. - * Fix SSLv3 renegotiation behaviour and stop processing data received from - peer after sending a fatal alert to refuse a renegotiation attempt. - Previous behaviour was to keep processing data even after the alert has - been sent. - * Accept empty trusted CA chain in authentication mode - MBEDTLS_SSL_VERIFY_OPTIONAL. Found by Jethro Beekman. #864 - * Fix implementation of mbedtls_ssl_parse_certificate() to not annihilate - fatal errors in authentication mode MBEDTLS_SSL_VERIFY_OPTIONAL and to - reflect bad EC curves within verification result. - * Fix bug that caused the modular inversion function to accept the invalid - modulus 1 and therefore to hang. Found by blaufish. #641. - * Fix incorrect sign computation in modular exponentiation when the base is - a negative MPI. Previously the result was always negative. Found by Guido - Vranken. - * Fix a numerical underflow leading to stack overflow in mpi_read_file() - that was triggered uppon reading an empty line. Found by Guido Vranken. - -Changes - * Send fatal alerts in more cases. The previous behaviour was to skip - sending the fatal alert and just drop the connection. - * Clarify ECDSA documentation and improve the sample code to avoid - misunderstanding and potentially dangerous use of the API. Pointed out - by Jean-Philippe Aumasson. - -= mbed TLS 2.5.0 branch released 2017-05-17 - -Security - * Wipe stack buffers in RSA private key operations - (rsa_rsaes_pkcs1_v15_decrypt(), rsa_rsaes_oaep_decrypt). Found by Laurent - Simon. - * Add exponent blinding to RSA private operations as a countermeasure - against side-channel attacks like the cache attack described in - https://arxiv.org/abs/1702.08719v2. - Found and fix proposed by Michael Schwarz, Samuel Weiser, Daniel Gruss, - Clémentine Maurice and Stefan Mangard. - -Features - * Add hardware acceleration support for the Elliptic Curve Point module. - This involved exposing parts of the internal interface to enable - replacing the core functions and adding and alternative, module level - replacement support for enabling the extension of the interface. - * Add a new configuration option to 'mbedtls_ssl_config' to enable - suppressing the CA list in Certificate Request messages. The default - behaviour has not changed, namely every configured CAs name is included. - -API Changes - * The following functions in the AES module have been deprecated and replaced - by the functions shown below. The new functions change the return type from - void to int to allow returning error codes when using MBEDTLS_AES_ALT, - MBEDTLS_AES_DECRYPT_ALT or MBEDTLS_AES_ENCRYPT_ALT. - mbedtls_aes_decrypt() -> mbedtls_internal_aes_decrypt() - mbedtls_aes_encrypt() -> mbedtls_internal_aes_encrypt() - -Bugfix - * Remove macros from compat-1.3.h that correspond to deleted items from most - recent versions of the library. Found by Kyle Keen. - * Fixed issue in the Threading module that prevented mutexes from - initialising. Found by sznaider. #667 #843 - * Add checks in the PK module for the RSA functions on 64-bit systems. - The PK and RSA modules use different types for passing hash length and - without these checks the type cast could lead to data loss. Found by Guido - Vranken. - -= mbed TLS 2.4.2 branch released 2017-03-08 - -Security - * Add checks to prevent signature forgeries for very large messages while - using RSA through the PK module in 64-bit systems. The issue was caused by - some data loss when casting a size_t to an unsigned int value in the - functions rsa_verify_wrap(), rsa_sign_wrap(), rsa_alt_sign_wrap() and - mbedtls_pk_sign(). Found by Jean-Philippe Aumasson. - * Fixed potential livelock during the parsing of a CRL in PEM format in - mbedtls_x509_crl_parse(). A string containing a CRL followed by trailing - characters after the footer could result in the execution of an infinite - loop. The issue can be triggered remotely. Found by Greg Zaverucha, - Microsoft. - * Removed MD5 from the allowed hash algorithms for CertificateRequest and - CertificateVerify messages, to prevent SLOTH attacks against TLS 1.2. - Introduced by interoperability fix for #513. - * Fixed a bug that caused freeing a buffer that was allocated on the stack, - when verifying the validity of a key on secp224k1. This could be - triggered remotely for example with a maliciously constructed certificate - and potentially could lead to remote code execution on some platforms. - Reported independently by rongsaws and Aleksandar Nikolic, Cisco Talos - team. #569 CVE-2017-2784 - -Bugfix - * Fix output certificate verification flags set by x509_crt_verify_top() when - traversing a chain of trusted CA. The issue would cause both flags, - MBEDTLS_X509_BADCERT_NOT_TRUSTED and MBEDTLS_X509_BADCERT_EXPIRED, to be - set when the verification conditions are not met regardless of the cause. - Found by Harm Verhagen and inestlerode. #665 #561 - * Fix the redefinition of macro ssl_set_bio to an undefined symbol - mbedtls_ssl_set_bio_timeout in compat-1.3.h, by removing it. - Found by omlib-lin. #673 - * Fix unused variable/function compilation warnings in pem.c, x509_crt.c and - x509_csr.c that are reported when building mbed TLS with a config.h that - does not define MBEDTLS_PEM_PARSE_C. Found by omnium21. #562 - * Fix incorrect renegotiation condition in ssl_check_ctr_renegotiate() that - would compare 64 bits of the record counter instead of 48 bits as indicated - in RFC 6347 Section 4.3.1. This could cause the execution of the - renegotiation routines at unexpected times when the protocol is DTLS. Found - by wariua. #687 - * Fixed multiple buffer overreads in mbedtls_pem_read_buffer() when parsing - the input string in PEM format to extract the different components. Found - by Eyal Itkin. - * Fixed potential arithmetic overflow in mbedtls_ctr_drbg_reseed() that could - cause buffer bound checks to be bypassed. Found by Eyal Itkin. - * Fixed potential arithmetic overflows in mbedtls_cipher_update() that could - cause buffer bound checks to be bypassed. Found by Eyal Itkin. - * Fixed potential arithmetic overflow in mbedtls_md2_update() that could - cause buffer bound checks to be bypassed. Found by Eyal Itkin. - * Fixed potential arithmetic overflow in mbedtls_base64_decode() that could - cause buffer bound checks to be bypassed. Found by Eyal Itkin. - * Fixed heap overreads in mbedtls_x509_get_time(). Found by Peng - Li/Yueh-Hsun Lin, KNOX Security, Samsung Research America. - * Fix potential memory leak in mbedtls_x509_crl_parse(). The leak was caused - by missing calls to mbedtls_pem_free() in cases when a - MBEDTLS_ERR_PEM_NO_HEADER_FOOTER_PRESENT error was encountered. Found and - fix proposed by Guido Vranken. #722 - * Fixed the templates used to generate project and solution files for Visual - Studio 2015 as well as the files themselves, to remove a build warning - generated in Visual Studio 2015. Reported by Steve Valliere. #742 - * Fix a resource leak in ssl_cookie, when using MBEDTLS_THREADING_C. - Raised and fix suggested by Alan Gillingham in the mbed TLS forum. #771 - * Fix 1 byte buffer overflow in mbedtls_mpi_write_string() when the MPI - number to write in hexadecimal is negative and requires an odd number of - digits. Found and fixed by Guido Vranken. - * Fix unlisted DES configuration dependency in some pkparse test cases. Found - by inestlerode. #555 - -= mbed TLS 2.4.1 branch released 2016-12-13 - -Changes - * Update to CMAC test data, taken from - NIST Special Publication 800-38B - - Recommendation for Block Cipher Modes of Operation: The CMAC Mode for - Authentication – October 2016 - -= mbed TLS 2.4.0 branch released 2016-10-17 - -Security - * Removed the MBEDTLS_SSL_AEAD_RANDOM_IV option, because it was not compliant - with RFC-5116 and could lead to session key recovery in very long TLS - sessions. "Nonce-Disrespecting Adversaries Practical Forgery Attacks on GCM in - TLS" - H. Bock, A. Zauner, S. Devlin, J. Somorovsky, P. Jovanovic. - https://eprint.iacr.org/2016/475.pdf - * Fixed potential stack corruption in mbedtls_x509write_crt_der() and - mbedtls_x509write_csr_der() when the signature is copied to the buffer - without checking whether there is enough space in the destination. The - issue cannot be triggered remotely. Found by Jethro Beekman. - -Features - * Added support for CMAC for AES and 3DES and AES-CMAC-PRF-128, as defined by - NIST SP 800-38B, RFC-4493 and RFC-4615. - * Added hardware entropy selftest to verify that the hardware entropy source - is functioning correctly. - * Added a script to print build environment info for diagnostic use in test - scripts, which is also now called by all.sh. - * Added the macro MBEDTLS_X509_MAX_FILE_PATH_LEN that enables the user to - configure the maximum length of a file path that can be buffered when - calling mbedtls_x509_crt_parse_path(). - * Added a configuration file config-no-entropy.h that configures the subset of - library features that do not require an entropy source. - * Added the macro MBEDTLS_ENTROPY_MIN_HARDWARE in config.h. This allows users - to configure the minimum number of bytes for entropy sources using the - mbedtls_hardware_poll() function. - -Bugfix - * Fix for platform time abstraction to avoid dependency issues where a build - may need time but not the standard C library abstraction, and added - configuration consistency checks to check_config.h - * Fix dependency issue in Makefile to allow parallel builds. - * Fix incorrect handling of block lengths in crypt_and_hash.c sample program, - when GCM is used. Found by udf2457. #441 - * Fix for key exchanges based on ECDH-RSA or ECDH-ECDSA which weren't - enabled unless others were also present. Found by David Fernandez. #428 - * Fix for out-of-tree builds using CMake. Found by jwurzer, and fix based on - a contribution from Tobias Tangemann. #541 - * Fixed cert_app.c sample program for debug output and for use when no root - certificates are provided. - * Fix conditional statement that would cause a 1 byte overread in - mbedtls_asn1_get_int(). Found and fixed by Guido Vranken. #599 - * Fixed pthread implementation to avoid unintended double initialisations - and double frees. Found by Niklas Amnebratt. - * Fixed the sample applications gen_key.c, cert_req.c and cert_write.c for - builds where the configuration MBEDTLS_PEM_WRITE_C is not defined. Found - by inestlerode. #559. - * Fix mbedtls_x509_get_sig() to update the ASN1 type in the mbedtls_x509_buf - data structure until after error checks are successful. Found by - subramanyam-c. #622 - * Fix documentation and implementation missmatch for function arguments of - mbedtls_gcm_finish(). Found by cmiatpaar. #602 - * Guarantee that P>Q at RSA key generation. Found by inestlerode. #558 - * Fix potential byte overread when verifying malformed SERVER_HELLO in - ssl_parse_hello_verify_request() for DTLS. Found by Guido Vranken. - * Fix check for validity of date when parsing in mbedtls_x509_get_time(). - Found by subramanyam-c. #626 - * Fix compatibility issue with Internet Explorer client authentication, - where the limited hash choices prevented the client from sending its - certificate. Found by teumas. #513 - * Fix compilation without MBEDTLS_SELF_TEST enabled. - -Changes - * Extended test coverage of special cases, and added new timing test suite. - * Removed self-tests from the basic-built-test.sh script, and added all - missing self-tests to the test suites, to ensure self-tests are only - executed once. - * Added support for 3 and 4 byte lengths to mbedtls_asn1_write_len(). - * Added support for a Yotta specific configuration file - - through the symbol YOTTA_CFG_MBEDTLS_TARGET_CONFIG_FILE. - * Added optimization for code space for X.509/OID based on configured - features. Contributed by Aviv Palivoda. - * Renamed source file library/net.c to library/net_sockets.c to avoid - naming collision in projects which also have files with the common name - net.c. For consistency, the corresponding header file, net.h, is marked as - deprecated, and its contents moved to net_sockets.h. - * Changed the strategy for X.509 certificate parsing and validation, to no - longer disregard certificates with unrecognised fields. - -= mbed TLS 2.3.0 branch released 2016-06-28 - -Security - * Fix missing padding length check in mbedtls_rsa_rsaes_pkcs1_v15_decrypt - required by PKCS1 v2.2 - * Fix potential integer overflow to buffer overflow in - mbedtls_rsa_rsaes_pkcs1_v15_encrypt and mbedtls_rsa_rsaes_oaep_encrypt - (not triggerable remotely in (D)TLS). - * Fix a potential integer underflow to buffer overread in - mbedtls_rsa_rsaes_oaep_decrypt. It is not triggerable remotely in - SSL/TLS. - -Features - * Support for platform abstraction of the standard C library time() - function. - -Bugfix - * Fix bug in mbedtls_mpi_add_mpi() that caused wrong results when the three - arguments where the same (in-place doubling). Found and fixed by Janos - Follath. #309 - * Fix potential build failures related to the 'apidoc' target, introduced - in the previous patch release. Found by Robert Scheck. #390 #391 - * Fix issue in Makefile that prevented building using armar. #386 - * Fix memory leak that occurred only when ECJPAKE was enabled and ECDHE and - ECDSA was disabled in config.h . The leak didn't occur by default. - * Fix an issue that caused valid certificates to be rejected whenever an - expired or not yet valid certificate was parsed before a valid certificate - in the trusted certificate list. - * Fix bug in mbedtls_x509_crt_parse that caused trailing extra data in the - buffer after DER certificates to be included in the raw representation. - * Fix issue that caused a hang when generating RSA keys of odd bitlength - * Fix bug in mbedtls_rsa_rsaes_pkcs1_v15_encrypt that made null pointer - dereference possible. - * Fix issue that caused a crash if invalid curves were passed to - mbedtls_ssl_conf_curves. #373 - * Fix issue in ssl_fork_server which was preventing it from functioning. #429 - * Fix memory leaks in test framework - * Fix test in ssl-opt.sh that does not run properly with valgrind - * Fix unchecked calls to mmbedtls_md_setup(). Fix by Brian Murray. #502 - -Changes - * On ARM platforms, when compiling with -O0 with GCC, Clang or armcc5, - don't use the optimized assembly for bignum multiplication. This removes - the need to pass -fomit-frame-pointer to avoid a build error with -O0. - * Disabled SSLv3 in the default configuration. - * Optimized mbedtls_mpi_zeroize() for MPI integer size. (Fix by Alexey - Skalozub). - * Fix non-compliance server extension handling. Extensions for SSLv3 are now - ignored, as required by RFC6101. - -= mbed TLS 2.2.1 released 2016-01-05 - -Security - * Fix potential double free when mbedtls_asn1_store_named_data() fails to - allocate memory. Only used for certificate generation, not triggerable - remotely in SSL/TLS. Found by Rafał Przywara. #367 - * Disable MD5 handshake signatures in TLS 1.2 by default to prevent the - SLOTH attack on TLS 1.2 server authentication (other attacks from the - SLOTH paper do not apply to any version of mbed TLS or PolarSSL). - https://www.mitls.org/pages/attacks/SLOTH - -Bugfix - * Fix over-restrictive length limit in GCM. Found by Andreas-N. #362 - * Fix bug in certificate validation that caused valid chains to be rejected - when the first intermediate certificate has pathLenConstraint=0. Found by - Nicholas Wilson. Introduced in mbed TLS 2.2.0. #280 - * Removed potential leak in mbedtls_rsa_rsassa_pkcs1_v15_sign(), found by - JayaraghavendranK. #372 - * Fix suboptimal handling of unexpected records that caused interop issues - with some peers over unreliable links. Avoid dropping an entire DTLS - datagram if a single record in a datagram is unexpected, instead only - drop the record and look at subsequent records (if any are present) in - the same datagram. Found by jeannotlapin. #345 - -= mbed TLS 2.2.0 released 2015-11-04 - -Security - * Fix potential double free if mbedtls_ssl_conf_psk() is called more than - once and some allocation fails. Cannot be forced remotely. Found by Guido - Vranken, Intelworks. - * Fix potential heap corruption on Windows when - mbedtls_x509_crt_parse_path() is passed a path longer than 2GB. Cannot be - triggered remotely. Found by Guido Vranken, Intelworks. - * Fix potential buffer overflow in some asn1_write_xxx() functions. - Cannot be triggered remotely unless you create X.509 certificates based - on untrusted input or write keys of untrusted origin. Found by Guido - Vranken, Intelworks. - * The X509 max_pathlen constraint was not enforced on intermediate - certificates. Found by Nicholas Wilson, fix and tests provided by - Janos Follath. #280 and #319 - -Features - * Experimental support for EC J-PAKE as defined in Thread 1.0.0. - Disabled by default as the specification might still change. - * Added a key extraction callback to accees the master secret and key - block. (Potential uses include EAP-TLS and Thread.) - -Bugfix - * Self-signed certificates were not excluded from pathlen counting, - resulting in some valid X.509 being incorrectly rejected. Found and fix - provided by Janos Follath. #319 - * Fix build error with configurations where ECDHE-PSK is the only key - exchange. Found and fix provided by Chris Hammond. #270 - * Fix build error with configurations where RSA, RSA-PSK, ECDH-RSA or - ECHD-ECDSA if the only key exchange. Multiple reports. #310 - * Fixed a bug causing some handshakes to fail due to some non-fatal alerts - not being properly ignored. Found by mancha and Kasom Koht-arsa, #308 - * mbedtls_x509_crt_verify(_with_profile)() now also checks the key type and - size/curve against the profile. Before that, there was no way to set a - minimum key size for end-entity certificates with RSA keys. Found by - Matthew Page of Scannex Electronics Ltd. - * Fix failures in MPI on Sparc(64) due to use of bad assembly code. - Found by Kurt Danielson. #292 - * Fix typo in name of the extKeyUsage OID. Found by inestlerode, #314 - * Fix bug in ASN.1 encoding of booleans that caused generated CA - certificates to be rejected by some applications, including OS X - Keychain. Found and fixed by Jonathan Leroy, Inikup. - -Changes - * Improved performance of mbedtls_ecp_muladd() when one of the scalars is 1 - or -1. - -= mbed TLS 2.1.2 released 2015-10-06 - -Security - * Added fix for CVE-2015-5291 to prevent heap corruption due to buffer - overflow of the hostname or session ticket. Found by Guido Vranken, - Intelworks. - * Fix potential double-free if mbedtls_ssl_set_hs_psk() is called more than - once in the same handhake and mbedtls_ssl_conf_psk() was used. - Found and patch provided by Guido Vranken, Intelworks. Cannot be forced - remotely. - * Fix stack buffer overflow in pkcs12 decryption (used by - mbedtls_pk_parse_key(file)() when the password is > 129 bytes. - Found by Guido Vranken, Intelworks. Not triggerable remotely. - * Fix potential buffer overflow in mbedtls_mpi_read_string(). - Found by Guido Vranken, Intelworks. Not exploitable remotely in the context - of TLS, but might be in other uses. On 32 bit machines, requires reading a - string of close to or larger than 1GB to exploit; on 64 bit machines, would - require reading a string of close to or larger than 2^62 bytes. - * Fix potential random memory allocation in mbedtls_pem_read_buffer() - on crafted PEM input data. Found and fix provided by Guido Vranken, - Intelworks. Not triggerable remotely in TLS. Triggerable remotely if you - accept PEM data from an untrusted source. - * Fix possible heap buffer overflow in base64_encoded() when the input - buffer is 512MB or larger on 32-bit platforms. Found by Guido Vranken, - Intelworks. Not trigerrable remotely in TLS. - * Fix potential double-free if mbedtls_conf_psk() is called repeatedly on - the same mbedtls_ssl_config object and memory allocation fails. Found by - Guido Vranken, Intelworks. Cannot be forced remotely. - * Fix potential heap buffer overflow in servers that perform client - authentication against a crafted CA cert. Cannot be triggered remotely - unless you allow third parties to pick trust CAs for client auth. - Found by Guido Vranken, Intelworks. - -Bugfix - * Fix compile error in net.c with musl libc. Found and patch provided by - zhasha (#278). - * Fix macroization of 'inline' keyword when building as C++. (#279) - -Changes - * Added checking of hostname length in mbedtls_ssl_set_hostname() to ensure - domain names are compliant with RFC 1035. - * Fixed paths for check_config.h in example config files. (Found by bachp) - (#291) - -= mbed TLS 2.1.1 released 2015-09-17 - -Security - * Add countermeasure against Lenstra's RSA-CRT attack for PKCS#1 v1.5 - signatures. (Found by Florian Weimer, Red Hat.) - https://securityblog.redhat.com/2015/09/02/factoring-rsa-keys-with-tls-perfect-forward-secrecy/ - * Fix possible client-side NULL pointer dereference (read) when the client - tries to continue the handshake after it failed (a misuse of the API). - (Found and patch provided by Fabian Foerg, Gotham Digital Science using - afl-fuzz.) - -Bugfix - * Fix warning when using a 64bit platform. (found by embedthis) (#275) - * Fix off-by-one error in parsing Supported Point Format extension that - caused some handshakes to fail. - -Changes - * Made X509 profile pointer const in mbedtls_ssl_conf_cert_profile() to allow - use of mbedtls_x509_crt_profile_next. (found by NWilson) - * When a client initiates a reconnect from the same port as a live - connection, if cookie verification is available - (MBEDTLS_SSL_DTLS_HELLO_VERIFY defined in config.h, and usable cookie - callbacks set with mbedtls_ssl_conf_dtls_cookies()), this will be - detected and mbedtls_ssl_read() will return - MBEDTLS_ERR_SSL_CLIENT_RECONNECT - it is then possible to start a new - handshake with the same context. (See RFC 6347 section 4.2.8.) - -= mbed TLS 2.1.0 released 2015-09-04 - -Features - * Added support for yotta as a build system. - * Primary open source license changed to Apache 2.0 license. - -Bugfix - * Fix segfault in the benchmark program when benchmarking DHM. - * Fix build error with CMake and pre-4.5 versions of GCC (found by Hugo - Leisink). - * Fix bug when parsing a ServerHello without extensions (found by David - Sears). - * Fix bug in CMake lists that caused libmbedcrypto.a not to be installed - (found by Benoit Lecocq). - * Fix bug in Makefile that caused libmbedcrypto and libmbedx509 not to be - installed (found by Rawi666). - * Fix compile error with armcc 5 with --gnu option. - * Fix bug in Makefile that caused programs not to be installed correctly - (found by robotanarchy) (#232). - * Fix bug in Makefile that prevented from installing without building the - tests (found by robotanarchy) (#232). - * Fix missing -static-libgcc when building shared libraries for Windows - with make. - * Fix link error when building shared libraries for Windows with make. - * Fix error when loading libmbedtls.so. - * Fix bug in mbedtls_ssl_conf_default() that caused the default preset to - be always used (found by dcb314) (#235) - * Fix bug in mbedtls_rsa_public() and mbedtls_rsa_private() that could - result trying to unlock an unlocked mutex on invalid input (found by - Fredrik Axelsson) (#257) - * Fix -Wshadow warnings (found by hnrkp) (#240) - * Fix memory corruption on client with overlong PSK identity, around - SSL_MAX_CONTENT_LEN or higher - not triggerrable remotely (found by - Aleksandrs Saveljevs) (#238) - * Fix unused function warning when using MBEDTLS_MDx_ALT or - MBEDTLS_SHAxxx_ALT (found by Henrik) (#239) - * Fix memory corruption in pkey programs (found by yankuncheng) (#210) - -Changes - * The PEM parser now accepts a trailing space at end of lines (#226). - * It is now possible to #include a user-provided configuration file at the - end of the default config.h by defining MBEDTLS_USER_CONFIG_FILE on the - compiler's command line. - * When verifying a certificate chain, if an intermediate certificate is - trusted, no later cert is checked. (suggested by hannes-landeholm) - (#220). - * Prepend a "thread identifier" to debug messages (issue pointed out by - Hugo Leisink) (#210). - * Add mbedtls_ssl_get_max_frag_len() to query the current maximum fragment - length. - -= mbed TLS 2.0.0 released 2015-07-13 - -Features - * Support for DTLS 1.0 and 1.2 (RFC 6347). - * Ability to override core functions from MDx, SHAx, AES and DES modules - with custom implementation (eg hardware accelerated), complementing the - ability to override the whole module. - * New server-side implementation of session tickets that rotate keys to - preserve forward secrecy, and allows sharing across multiple contexts. - * Added a concept of X.509 cerificate verification profile that controls - which algorithms and key sizes (curves for ECDSA) are acceptable. - * Expanded configurability of security parameters in the SSL module with - mbedtls_ssl_conf_dhm_min_bitlen() and mbedtls_ssl_conf_sig_hashes(). - * Introduced a concept of presets for SSL security-relevant configuration - parameters. - -API Changes - * The library has been split into libmbedcrypto, libmbedx509, libmbedtls. - You now need to link to all of them if you use TLS for example. - * All public identifiers moved to the mbedtls_* or MBEDTLS_* namespace. - Some names have been further changed to make them more consistent. - Migration helpers scripts/rename.pl and include/mbedtls/compat-1.3.h are - provided. Full list of renamings in scripts/data_files/rename-1.3-2.0.txt - * Renamings of fields inside structures, not covered by the previous list: - mbedtls_cipher_info_t.key_length -> key_bitlen - mbedtls_cipher_context_t.key_length -> key_bitlen - mbedtls_ecp_curve_info.size -> bit_size - * Headers are now found in the 'mbedtls' directory (previously 'polarssl'). - * The following _init() functions that could return errors have - been split into an _init() that returns void and another function that - should generally be the first function called on this context after init: - mbedtls_ssl_init() -> mbedtls_ssl_setup() - mbedtls_ccm_init() -> mbedtls_ccm_setkey() - mbedtls_gcm_init() -> mbedtls_gcm_setkey() - mbedtls_hmac_drbg_init() -> mbedtls_hmac_drbg_seed(_buf)() - mbedtls_ctr_drbg_init() -> mbedtls_ctr_drbg_seed() - Note that for mbedtls_ssl_setup(), you need to be done setting up the - ssl_config structure before calling it. - * Most ssl_set_xxx() functions (all except ssl_set_bio(), ssl_set_hostname(), - ssl_set_session() and ssl_set_client_transport_id(), plus - ssl_legacy_renegotiation()) have been renamed to mbedtls_ssl_conf_xxx() - (see rename.pl and compat-1.3.h above) and their first argument's type - changed from ssl_context to ssl_config. - * ssl_set_bio() changed signature (contexts merged, order switched, one - additional callback for read-with-timeout). - * The following functions have been introduced and must be used in callback - implementations (SNI, PSK) instead of their *conf counterparts: - mbedtls_ssl_set_hs_own_cert() - mbedtls_ssl_set_hs_ca_chain() - mbedtls_ssl_set_hs_psk() - * mbedtls_ssl_conf_ca_chain() lost its last argument (peer_cn), now set - using mbedtls_ssl_set_hostname(). - * mbedtls_ssl_conf_session_cache() changed prototype (only one context - pointer, parameters reordered). - * On server, mbedtls_ssl_conf_session_tickets_cb() must now be used in - place of mbedtls_ssl_conf_session_tickets() to enable session tickets. - * The SSL debug callback gained two new arguments (file name, line number). - * Debug modes were removed. - * mbedtls_ssl_conf_truncated_hmac() now returns void. - * mbedtls_memory_buffer_alloc_init() now returns void. - * X.509 verification flags are now an uint32_t. Affect the signature of: - mbedtls_ssl_get_verify_result() - mbedtls_x509_ctr_verify_info() - mbedtls_x509_crt_verify() (flags, f_vrfy -> needs to be updated) - mbedtls_ssl_conf_verify() (f_vrfy -> needs to be updated) - * The following functions changed prototype to avoid an in-out length - parameter: - mbedtls_base64_encode() - mbedtls_base64_decode() - mbedtls_mpi_write_string() - mbedtls_dhm_calc_secret() - * In the NET module, all "int" and "int *" arguments for file descriptors - changed type to "mbedtls_net_context *". - * net_accept() gained new arguments for the size of the client_ip buffer. - * In the threading layer, mbedtls_mutex_init() and mbedtls_mutex_free() now - return void. - * ecdsa_write_signature() gained an additional md_alg argument and - ecdsa_write_signature_det() was deprecated. - * pk_sign() no longer accepts md_alg == POLARSSL_MD_NONE with ECDSA. - * Last argument of x509_crt_check_key_usage() and - mbedtls_x509write_crt_set_key_usage() changed from int to unsigned. - * test_ca_list (from certs.h) is renamed to test_cas_pem and is only - available if POLARSSL_PEM_PARSE_C is defined (it never worked without). - * Test certificates in certs.c are no longer guaranteed to be nul-terminated - strings; use the new *_len variables instead of strlen(). - * Functions mbedtls_x509_xxx_parse(), mbedtls_pk_parse_key(), - mbedtls_pk_parse_public_key() and mbedtls_dhm_parse_dhm() now expect the - length parameter to include the terminating null byte for PEM input. - * Signature of mpi_mul_mpi() changed to make the last argument unsigned - * calloc() is now used instead of malloc() everywhere. API of platform - layer and the memory_buffer_alloc module changed accordingly. - (Thanks to Mansour Moufid for helping with the replacement.) - * Change SSL_DISABLE_RENEGOTIATION config.h flag to SSL_RENEGOTIATION - (support for renegotiation now needs explicit enabling in config.h). - * Split MBEDTLS_HAVE_TIME into MBEDTLS_HAVE_TIME and MBEDTLS_HAVE_TIME_DATE - in config.h - * net_connect() and net_bind() have a new 'proto' argument to choose - between TCP and UDP, using the macros NET_PROTO_TCP or NET_PROTO_UDP. - Their 'port' argument type is changed to a string. - * Some constness fixes - -Removals - * Removed mbedtls_ecp_group_read_string(). Only named groups are supported. - * Removed mbedtls_ecp_sub() and mbedtls_ecp_add(), use - mbedtls_ecp_muladd(). - * Removed individual mdX_hmac, shaX_hmac, mdX_file and shaX_file functions - (use generic functions from md.h) - * Removed mbedtls_timing_msleep(). Use mbedtls_net_usleep() or a custom - waiting function. - * Removed test DHM parameters from the test certs module. - * Removed the PBKDF2 module (use PKCS5). - * Removed POLARSSL_ERROR_STRERROR_BC (use mbedtls_strerror()). - * Removed compat-1.2.h (helper for migrating from 1.2 to 1.3). - * Removed openssl.h (very partial OpenSSL compatibility layer). - * Configuration options POLARSSL_HAVE_LONGLONG was removed (now always on). - * Configuration options POLARSSL_HAVE_INT8 and POLARSSL_HAVE_INT16 have - been removed (compiler is required to support 32-bit operations). - * Configuration option POLARSSL_HAVE_IPV6 was removed (always enabled). - * Removed test program o_p_test, the script compat.sh does more. - * Removed test program ssl_test, superseded by ssl-opt.sh. - * Removed helper script active-config.pl - -New deprecations - * md_init_ctx() is deprecated in favour of md_setup(), that adds a third - argument (allowing memory savings if HMAC is not used) - -Semi-API changes (technically public, morally private) - * Renamed a few headers to include _internal in the name. Those headers are - not supposed to be included by users. - * Changed md_info_t into an opaque structure (use md_get_xxx() accessors). - * Changed pk_info_t into an opaque structure. - * Changed cipher_base_t into an opaque structure. - * Removed sig_oid2 and rename sig_oid1 to sig_oid in x509_crt and x509_crl. - * x509_crt.key_usage changed from unsigned char to unsigned int. - * Removed r and s from ecdsa_context - * Removed mode from des_context and des3_context - -Default behavior changes - * The default minimum TLS version is now TLS 1.0. - * RC4 is now blacklisted by default in the SSL/TLS layer, and excluded from the - default ciphersuite list returned by ssl_list_ciphersuites() - * Support for receiving SSLv2 ClientHello is now disabled by default at - compile time. - * The default authmode for SSL/TLS clients is now REQUIRED. - * Support for RSA_ALT contexts in the PK layer is now optional. Since is is - enabled in the default configuration, this is only noticeable if using a - custom config.h - * Default DHM parameters server-side upgraded from 1024 to 2048 bits. - * A minimum RSA key size of 2048 bits is now enforced during ceritificate - chain verification. - * Negotiation of truncated HMAC is now disabled by default on server too. - * The following functions are now case-sensitive: - mbedtls_cipher_info_from_string() - mbedtls_ecp_curve_info_from_name() - mbedtls_md_info_from_string() - mbedtls_ssl_ciphersuite_from_string() - mbedtls_version_check_feature() - -Requirement changes - * The minimum MSVC version required is now 2010 (better C99 support). - * The NET layer now unconditionnaly relies on getaddrinfo() and select(). - * Compiler is required to support C99 types such as long long and uint32_t. - -API changes from the 1.4 preview branch - * ssl_set_bio_timeout() was removed, split into mbedtls_ssl_set_bio() with - new prototype, and mbedtls_ssl_set_read_timeout(). - * The following functions now return void: - mbedtls_ssl_conf_transport() - mbedtls_ssl_conf_max_version() - mbedtls_ssl_conf_min_version() - * DTLS no longer hard-depends on TIMING_C, but uses a callback interface - instead, see mbedtls_ssl_set_timer_cb(), with the Timing module providing - an example implementation, see mbedtls_timing_delay_context and - mbedtls_timing_set/get_delay(). - * With UDP sockets, it is no longer necessary to call net_bind() again - after a successful net_accept(). - -Changes - * mbedtls_ctr_drbg_random() and mbedtls_hmac_drbg_random() are now - thread-safe if MBEDTLS_THREADING_C is enabled. - * Reduced ROM fooprint of SHA-256 and added an option to reduce it even - more (at the expense of performance) MBEDTLS_SHA256_SMALLER. - -= mbed TLS 1.3 branch - -Security - * With authmode set to SSL_VERIFY_OPTIONAL, verification of keyUsage and - extendedKeyUsage on the leaf certificate was lost (results not accessible - via ssl_get_verify_results()). - * Add countermeasure against "Lucky 13 strikes back" cache-based attack, - https://dl.acm.org/citation.cfm?id=2714625 - -Features - * Improve ECC performance by using more efficient doubling formulas - (contributed by Peter Dettman). - * Add x509_crt_verify_info() to display certificate verification results. - * Add support for reading DH parameters with privateValueLength included - (contributed by Daniel Kahn Gillmor). - * Add support for bit strings in X.509 names (request by Fredrik Axelsson). - * Add support for id-at-uniqueIdentifier in X.509 names. - * Add support for overriding snprintf() (except on Windows) and exit() in - the platform layer. - * Add an option to use macros instead of function pointers in the platform - layer (helps get rid of unwanted references). - * Improved Makefiles for Windows targets by fixing library targets and making - cross-compilation easier (thanks to Alon Bar-Lev). - * The benchmark program also prints heap usage for public-key primitives - if POLARSSL_MEMORY_BUFFER_ALLOC_C and POLARSSL_MEMORY_DEBUG are defined. - * New script ecc-heap.sh helps measuring the impact of ECC parameters on - speed and RAM (heap only for now) usage. - * New script memory.sh helps measuring the ROM and RAM requirements of two - reduced configurations (PSK-CCM and NSA suite B). - * Add config flag POLARSSL_DEPRECATED_WARNING (off by default) to produce - warnings on use of deprecated functions (with GCC and Clang only). - * Add config flag POLARSSL_DEPRECATED_REMOVED (off by default) to produce - errors on use of deprecated functions. - -Bugfix - * Fix compile errors with PLATFORM_NO_STD_FUNCTIONS. - * Fix compile error with PLATFORM_EXIT_ALT (thanks to Rafał Przywara). - * Fix bug in entropy.c when THREADING_C is also enabled that caused - entropy_free() to crash (thanks to Rafał Przywara). - * Fix memory leak when gcm_setkey() and ccm_setkey() are used more than - once on the same context. - * Fix bug in ssl_mail_client when password is longer that username (found - by Bruno Pape). - * Fix undefined behaviour (memcmp( NULL, NULL, 0 );) in X.509 modules - (detected by Clang's 3.6 UBSan). - * mpi_size() and mpi_msb() would segfault when called on an mpi that is - initialized but not set (found by pravic). - * Fix detection of support for getrandom() on Linux (reported by syzzer) by - doing it at runtime (using uname) rather that compile time. - * Fix handling of symlinks by "make install" (found by Gaël PORTAY). - * Fix potential NULL pointer dereference (not trigerrable remotely) when - ssl_write() is called before the handshake is finished (introduced in - 1.3.10) (first reported by Martin Blumenstingl). - * Fix bug in pk_parse_key() that caused some valid private EC keys to be - rejected. - * Fix bug in Via Padlock support (found by Nikos Mavrogiannopoulos). - * Fix thread safety bug in RSA operations (found by Fredrik Axelsson). - * Fix hardclock() (only used in the benchmarking program) with some - versions of mingw64 (found by kxjhlele). - * Fix warnings from mingw64 in timing.c (found by kxjklele). - * Fix potential unintended sign extension in asn1_get_len() on 64-bit - platforms. - * Fix potential memory leak in ssl_set_psk() (found by Mansour Moufid). - * Fix compile error when POLARSSL_SSL_DISABLE_RENEGOTATION and - POLARSSL_SSL_SSESSION_TICKETS where both enabled in config.h (introduced - in 1.3.10). - * Add missing extern "C" guard in aesni.h (reported by amir zamani). - * Add missing dependency on SHA-256 in some x509 programs (reported by - Gergely Budai). - * Fix bug related to ssl_set_curves(): the client didn't check that the - curve picked by the server was actually allowed. - -Changes - * Remove bias in mpi_gen_prime (contributed by Pascal Junod). - * Remove potential sources of timing variations (some contributed by Pascal - Junod). - * Options POLARSSL_HAVE_INT8 and POLARSSL_HAVE_INT16 are deprecated. - * Enabling POLARSSL_NET_C without POLARSSL_HAVE_IPV6 is deprecated. - * compat-1.2.h and openssl.h are deprecated. - * Adjusting/overriding CFLAGS and LDFLAGS with the make build system is now - more flexible (warning: OFLAGS is not used any more) (see the README) - (contributed by Alon Bar-Lev). - * ssl_set_own_cert() no longer calls pk_check_pair() since the - performance impact was bad for some users (this was introduced in 1.3.10). - * Move from SHA-1 to SHA-256 in example programs using signatures - (suggested by Thorsten Mühlfelder). - * Remove some unneeded inclusions of header files from the standard library - "minimize" others (eg use stddef.h if only size_t is needed). - * Change #include lines in test files to use double quotes instead of angle - brackets for uniformity with the rest of the code. - * Remove dependency on sscanf() in X.509 parsing modules. - -= mbed TLS 1.3.10 released 2015-02-09 -Security - * NULL pointer dereference in the buffer-based allocator when the buffer is - full and polarssl_free() is called (found by Mark Hasemeyer) - (only possible if POLARSSL_MEMORY_BUFFER_ALLOC_C is enabled, which it is - not by default). - * Fix remotely-triggerable uninitialised pointer dereference caused by - crafted X.509 certificate (TLS server is not affected if it doesn't ask for a - client certificate) (found using Codenomicon Defensics). - * Fix remotely-triggerable memory leak caused by crafted X.509 certificates - (TLS server is not affected if it doesn't ask for a client certificate) - (found using Codenomicon Defensics). - * Fix potential stack overflow while parsing crafted X.509 certificates - (TLS server is not affected if it doesn't ask for a client certificate) - (found using Codenomicon Defensics). - * Fix timing difference that could theoretically lead to a - Bleichenbacher-style attack in the RSA and RSA-PSK key exchanges - (reported by Sebastian Schinzel). - -Features - * Add support for FALLBACK_SCSV (draft-ietf-tls-downgrade-scsv). - * Add support for Extended Master Secret (draft-ietf-tls-session-hash). - * Add support for Encrypt-then-MAC (RFC 7366). - * Add function pk_check_pair() to test if public and private keys match. - * Add x509_crl_parse_der(). - * Add compile-time option POLARSSL_X509_MAX_INTERMEDIATE_CA to limit the - length of an X.509 verification chain. - * Support for renegotiation can now be disabled at compile-time - * Support for 1/n-1 record splitting, a countermeasure against BEAST. - * Certificate selection based on signature hash, preferring SHA-1 over SHA-2 - for pre-1.2 clients when multiple certificates are available. - * Add support for getrandom() syscall on recent Linux kernels with Glibc or - a compatible enough libc (eg uClibc). - * Add ssl_set_arc4_support() to make it easier to disable RC4 at runtime - while using the default ciphersuite list. - * Added new error codes and debug messages about selection of - ciphersuite/certificate. - -Bugfix - * Stack buffer overflow if ctr_drbg_update() is called with too large - add_len (found by Jean-Philippe Aumasson) (not triggerable remotely). - * Possible buffer overflow of length at most POLARSSL_MEMORY_ALIGN_MULTIPLE - if memory_buffer_alloc_init() was called with buf not aligned and len not - a multiple of POLARSSL_MEMORY_ALIGN_MULTIPLE (not triggerable remotely). - * User set CFLAGS were ignored by Cmake with gcc (introduced in 1.3.9, found - by Julian Ospald). - * Fix potential undefined behaviour in Camellia. - * Fix potential failure in ECDSA signatures when POLARSSL_ECP_MAX_BITS is a - multiple of 8 (found by Gergely Budai). - * Fix unchecked return code in x509_crt_parse_path() on Windows (found by - Peter Vaskovic). - * Fix assembly selection for MIPS64 (thanks to James Cowgill). - * ssl_get_verify_result() now works even if the handshake was aborted due - to a failed verification (found by Fredrik Axelsson). - * Skip writing and parsing signature_algorithm extension if none of the - key exchanges enabled needs certificates. This fixes a possible interop - issue with some servers when a zero-length extension was sent. (Reported - by Peter Dettman.) - * On a 0-length input, base64_encode() did not correctly set output length - (found by Hendrik van den Boogaard). - -Changes - * Use deterministic nonces for AEAD ciphers in TLS by default (possible to - switch back to random with POLARSSL_SSL_AEAD_RANDOM_IV in config.h). - * Blind RSA private operations even when POLARSSL_RSA_NO_CRT is defined. - * ssl_set_own_cert() now returns an error on key-certificate mismatch. - * Forbid repeated extensions in X.509 certificates. - * debug_print_buf() now prints a text view in addition to hexadecimal. - * A specific error is now returned when there are ciphersuites in common - but none of them is usable due to external factors such as no certificate - with a suitable (extended)KeyUsage or curve or no PSK set. - * It is now possible to disable negotiation of truncated HMAC server-side - at runtime with ssl_set_truncated_hmac(). - * Example programs for SSL client and server now disable SSLv3 by default. - * Example programs for SSL client and server now disable RC4 by default. - * Use platform.h in all test suites and programs. - -= PolarSSL 1.3.9 released 2014-10-20 -Security - * Lowest common hash was selected from signature_algorithms extension in - TLS 1.2 (found by Darren Bane) (introduced in 1.3.8). - * Remotely-triggerable memory leak when parsing some X.509 certificates - (server is not affected if it doesn't ask for a client certificate) - (found using Codenomicon Defensics). - * Remotely-triggerable memory leak when parsing crafted ClientHello - (not affected if ECC support was compiled out) (found using Codenomicon - Defensics). - -Bugfix - * Support escaping of commas in x509_string_to_names() - * Fix compile error in ssl_pthread_server (found by Julian Ospald). - * Fix net_accept() regarding non-blocking sockets (found by Luca Pesce). - * Don't print uninitialised buffer in ssl_mail_client (found by Marc Abel). - * Fix warnings from Clang's scan-build (contributed by Alfred Klomp). - * Fix compile error in timing.c when POLARSSL_NET_C and POLARSSL_SELFTEST - are defined but not POLARSSL_HAVE_TIME (found by Stephane Di Vito). - * Remove non-existent file from VS projects (found by Peter Vaskovic). - * ssl_read() could return non-application data records on server while - renegotation was pending, and on client when a HelloRequest was received. - * Server-initiated renegotiation would fail with non-blocking I/O if the - write callback returned WANT_WRITE when requesting renegotiation. - * ssl_close_notify() could send more than one message in some circumstances - with non-blocking I/O. - * Fix compiler warnings on iOS (found by Sander Niemeijer). - * x509_crt_parse() did not increase total_failed on PEM error - * Fix compile error with armcc in mpi_is_prime() - * Fix potential bad read in parsing ServerHello (found by Adrien - Vialletelle). - -Changes - * Ciphersuites using SHA-256 or SHA-384 now require TLS 1.x (there is no - standard defining how to use SHA-2 with SSL 3.0). - * Ciphersuites using RSA-PSK key exchange new require TLS 1.x (the spec is - ambiguous on how to encode some packets with SSL 3.0). - * Made buffer size in pk_write_(pub)key_pem() more dynamic, eg smaller if - RSA is disabled, larger if POLARSSL_MPI_MAX_SIZE is larger. - * ssl_read() now returns POLARSSL_ERR_NET_WANT_READ rather than - POLARSSL_ERR_SSL_UNEXPECTED_MESSAGE on harmless alerts. - * POLARSSL_MPI_MAX_SIZE now defaults to 1024 in order to allow 8192 bits - RSA keys. - * Accept spaces at end of line or end of buffer in base64_decode(). - * X.509 certificates with more than one AttributeTypeAndValue per - RelativeDistinguishedName are not accepted any more. - -= PolarSSL 1.3.8 released 2014-07-11 -Security - * Fix length checking for AEAD ciphersuites (found by Codenomicon). - It was possible to crash the server (and client) using crafted messages - when a GCM suite was chosen. - -Features - * Add CCM module and cipher mode to Cipher Layer - * Support for CCM and CCM_8 ciphersuites - * Support for parsing and verifying RSASSA-PSS signatures in the X.509 - modules (certificates, CRLs and CSRs). - * Blowfish in the cipher layer now supports variable length keys. - * Add example config.h for PSK with CCM, optimized for low RAM usage. - * Optimize for RAM usage in example config.h for NSA Suite B profile. - * Add POLARSSL_REMOVE_ARC4_CIPHERSUITES to allow removing RC4 ciphersuites - from the default list (inactive by default). - * Add server-side enforcement of sent renegotiation requests - (ssl_set_renegotiation_enforced()) - * Add SSL_CIPHERSUITES config.h flag to allow specifying a list of - ciphersuites to use and save some memory if the list is small. - -Changes - * Add LINK_WITH_PTHREAD option in CMake for explicit linking that is - required on some platforms (e.g. OpenBSD) - * Migrate zeroizing of data to polarssl_zeroize() instead of memset() - against unwanted compiler optimizations - * md_list() now returns hashes strongest first - * Selection of hash for signing ServerKeyExchange in TLS 1.2 now picks - strongest offered by client. - * All public contexts have _init() and _free() functions now for simpler - usage pattern - -Bugfix - * Fix in debug_print_msg() - * Enforce alignment in the buffer allocator even if buffer is not aligned - * Remove less-than-zero checks on unsigned numbers - * Stricter check on SSL ClientHello internal sizes compared to actual packet - size (found by TrustInSoft) - * Fix WSAStartup() return value check (found by Peter Vaskovic) - * Other minor issues (found by Peter Vaskovic) - * Fix symlink command for cross compiling with CMake (found by Andre - Heinecke) - * Fix DER output of gen_key app (found by Gergely Budai) - * Very small records were incorrectly rejected when truncated HMAC was in - use with some ciphersuites and versions (RC4 in all versions, CBC with - versions < TLS 1.1). - * Very large records using more than 224 bytes of padding were incorrectly - rejected with CBC-based ciphersuites and TLS >= 1.1 - * Very large records using less padding could cause a buffer overread of up - to 32 bytes with CBC-based ciphersuites and TLS >= 1.1 - * Restore ability to use a v1 cert as a CA if trusted locally. (This had - been removed in 1.3.6.) - * Restore ability to locally trust a self-signed cert that is not a proper - CA for use as an end entity certificate. (This had been removed in - 1.3.6.) - * Fix preprocessor checks for bn_mul PPC asm (found by Barry K. Nathan). - * Use \n\t rather than semicolons for bn_mul asm, since some assemblers - interpret semicolons as comment delimiters (found by Barry K. Nathan). - * Fix off-by-one error in parsing Supported Point Format extension that - caused some handshakes to fail. - * Fix possible miscomputation of the premaster secret with DHE-PSK key - exchange that caused some handshakes to fail with other implementations. - (Failure rate <= 1/255 with common DHM moduli.) - * Disable broken Sparc64 bn_mul assembly (found by Florian Obser). - * Fix base64_decode() to return and check length correctly (in case of - tight buffers) - * Fix mpi_write_string() to write "00" as hex output for empty MPI (found - by Hui Dong) - -= PolarSSL 1.3.7 released on 2014-05-02 -Features - * debug_set_log_mode() added to determine raw or full logging - * debug_set_threshold() added to ignore messages over threshold level - * version_check_feature() added to check for compile-time options at - run-time - -Changes - * POLARSSL_CONFIG_OPTIONS has been removed. All values are individually - checked and filled in the relevant module headers - * Debug module only outputs full lines instead of parts - * Better support for the different Attribute Types from IETF PKIX (RFC 5280) - * AES-NI now compiles with "old" assemblers too - * Ciphersuites based on RC4 now have the lowest priority by default - -Bugfix - * Only iterate over actual certificates in ssl_write_certificate_request() - (found by Matthew Page) - * Typos in platform.c and pkcs11.c (found by Daniel Phillips and Steffan - Karger) - * cert_write app should use subject of issuer certificate as issuer of cert - * Fix false reject in padding check in ssl_decrypt_buf() for CBC - ciphersuites, for full SSL frames of data. - * Improve interoperability by not writing extension length in ClientHello / - ServerHello when no extensions are present (found by Matthew Page) - * rsa_check_pubkey() now allows an E up to N - * On OpenBSD, use arc4random_buf() instead of rand() to prevent warnings - * mpi_fill_random() was creating numbers larger than requested on - big-endian platform when size was not an integer number of limbs - * Fix dependencies issues in X.509 test suite. - * Some parts of ssl_tls.c were compiled even when the module was disabled. - * Fix detection of DragonflyBSD in net.c (found by Markus Pfeiffer) - * Fix detection of Clang on some Apple platforms with CMake - (found by Barry K. Nathan) - -= PolarSSL 1.3.6 released on 2014-04-11 - -Features - * Support for the ALPN SSL extension - * Add option 'use_dev_random' to gen_key application - * Enable verification of the keyUsage extension for CA and leaf - certificates (POLARSSL_X509_CHECK_KEY_USAGE) - * Enable verification of the extendedKeyUsage extension - (POLARSSL_X509_CHECK_EXTENDED_KEY_USAGE) - -Changes - * x509_crt_info() now prints information about parsed extensions as well - * pk_verify() now returns a specific error code when the signature is valid - but shorter than the supplied length. - * Use UTC time to check certificate validity. - * Reject certificates with times not in UTC, per RFC 5280. - -Security - * Avoid potential timing leak in ecdsa_sign() by blinding modular division. - (Found by Watson Ladd.) - * The notAfter date of some certificates was no longer checked since 1.3.5. - This affects certificates in the user-supplied chain except the top - certificate. If the user-supplied chain contains only one certificates, - it is not affected (ie, its notAfter date is properly checked). - * Prevent potential NULL pointer dereference in ssl_read_record() (found by - TrustInSoft) - -Bugfix - * The length of various ClientKeyExchange messages was not properly checked. - * Some example server programs were not sending the close_notify alert. - * Potential memory leak in mpi_exp_mod() when error occurs during - calculation of RR. - * Fixed malloc/free default #define in platform.c (found by Gergely Budai). - * Fixed type which made POLARSSL_ENTROPY_FORCE_SHA256 uneffective (found by - Gergely Budai). - * Fix #include path in ecdsa.h which wasn't accepted by some compilers. - (found by Gergely Budai) - * Fix compile errors when POLARSSL_ERROR_STRERROR_BC is undefined (found by - Shuo Chen). - * oid_get_numeric_string() used to truncate the output without returning an - error if the output buffer was just 1 byte too small. - * dhm_parse_dhm() (hence dhm_parse_dhmfile()) did not set dhm->len. - * Calling pk_debug() on an RSA-alt key would segfault. - * pk_get_size() and pk_get_len() were off by a factor 8 for RSA-alt keys. - * Potential buffer overwrite in pem_write_buffer() because of low length - indication (found by Thijs Alkemade) - * EC curves constants, which should be only in ROM since 1.3.3, were also - stored in RAM due to missing 'const's (found by Gergely Budai). - -= PolarSSL 1.3.5 released on 2014-03-26 -Features - * HMAC-DRBG as a separate module - * Option to set the Curve preference order (disabled by default) - * Single Platform compatilibity layer (for memory / printf / fprintf) - * Ability to provide alternate timing implementation - * Ability to force the entropy module to use SHA-256 as its basis - (POLARSSL_ENTROPY_FORCE_SHA256) - * Testing script ssl-opt.sh added for testing 'live' ssl option - interoperability against OpenSSL and PolarSSL - * Support for reading EC keys that use SpecifiedECDomain in some cases. - * Entropy module now supports seed writing and reading - -Changes - * Deprecated the Memory layer - * entropy_add_source(), entropy_update_manual() and entropy_gather() - now thread-safe if POLARSSL_THREADING_C defined - * Improvements to the CMake build system, contributed by Julian Ospald. - * Work around a bug of the version of Clang shipped by Apple with Mavericks - that prevented bignum.c from compiling. (Reported by Rafael Baptista.) - * Revamped the compat.sh interoperatibility script to include support for - testing against GnuTLS - * Deprecated ssl_set_own_cert_rsa() and ssl_set_own_cert_rsa_alt() - * Improvements to tests/Makefile, contributed by Oden Eriksson. - -Security - * Forbid change of server certificate during renegotiation to prevent - "triple handshake" attack when authentication mode is 'optional' (the - attack was already impossible when authentication is required). - * Check notBefore timestamp of certificates and CRLs from the future. - * Forbid sequence number wrapping - * Fixed possible buffer overflow with overlong PSK - * Possible remotely-triggered out-of-bounds memory access fixed (found by - TrustInSoft) - -Bugfix - * ecp_gen_keypair() does more tries to prevent failure because of - statistics - * Fixed bug in RSA PKCS#1 v1.5 "reversed" operations - * Fixed testing with out-of-source builds using cmake - * Fixed version-major intolerance in server - * Fixed CMake symlinking on out-of-source builds - * Fixed dependency issues in test suite - * Programs rsa_sign_pss and rsa_verify_pss were not using PSS since 1.3.0 - * Bignum's MIPS-32 assembly was used on MIPS-64, causing chaos. (Found by - Alex Wilson.) - * ssl_cache was creating entries when max_entries=0 if TIMING_C was enabled. - * m_sleep() was sleeping twice too long on most Unix platforms. - * Fixed bug with session tickets and non-blocking I/O in the unlikely case - send() would return an EAGAIN error when sending the ticket. - * ssl_cache was leaking memory when reusing a timed out entry containing a - client certificate. - * ssl_srv was leaking memory when client presented a timed out ticket - containing a client certificate - * ssl_init() was leaving a dirty pointer in ssl_context if malloc of - out_ctr failed - * ssl_handshake_init() was leaving dirty pointers in subcontexts if malloc - of one of them failed - * Fix typo in rsa_copy() that impacted PKCS#1 v2 contexts - * x509_get_current_time() uses localtime_r() to prevent thread issues - -= PolarSSL 1.3.4 released on 2014-01-27 -Features - * Support for the Koblitz curves: secp192k1, secp224k1, secp256k1 - * Support for RIPEMD-160 - * Support for AES CFB8 mode - * Support for deterministic ECDSA (RFC 6979) - -Bugfix - * Potential memory leak in bignum_selftest() - * Replaced expired test certificate - * ssl_mail_client now terminates lines with CRLF, instead of LF - * net module handles timeouts on blocking sockets better (found by Tilman - Sauerbeck) - * Assembly format fixes in bn_mul.h - -Security - * Missing MPI_CHK calls added around unguarded mpi calls (found by - TrustInSoft) - -= PolarSSL 1.3.3 released on 2013-12-31 -Features - * EC key generation support in gen_key app - * Support for adhering to client ciphersuite order preference - (POLARSSL_SSL_SRV_RESPECT_CLIENT_PREFERENCE) - * Support for Curve25519 - * Support for ECDH-RSA and ECDH-ECDSA key exchanges and ciphersuites - * Support for IPv6 in the NET module - * AES-NI support for AES, AES-GCM and AES key scheduling - * SSL Pthread-based server example added (ssl_pthread_server) - -Changes - * gen_prime() speedup - * Speedup of ECP multiplication operation - * Relaxed some SHA2 ciphersuite's version requirements - * Dropped use of readdir_r() instead of readdir() with threading support - * More constant-time checks in the RSA module - * Split off curves from ecp.c into ecp_curves.c - * Curves are now stored fully in ROM - * Memory usage optimizations in ECP module - * Removed POLARSSL_THREADING_DUMMY - -Bugfix - * Fixed bug in mpi_set_bit() on platforms where t_uint is wider than int - * Fixed X.509 hostname comparison (with non-regular characters) - * SSL now gracefully handles missing RNG - * Missing defines / cases for RSA_PSK key exchange - * crypt_and_hash app checks MAC before final decryption - * Potential memory leak in ssl_ticket_keys_init() - * Memory leak in benchmark application - * Fixed x509_crt_parse_path() bug on Windows platforms - * Added missing MPI_CHK() around some statements in mpi_div_mpi() (found by - TrustInSoft) - * Fixed potential overflow in certificate size verification in - ssl_write_certificate() (found by TrustInSoft) - -Security - * Possible remotely-triggered out-of-bounds memory access fixed (found by - TrustInSoft) - -= PolarSSL 1.3.2 released on 2013-11-04 -Features - * PK tests added to test framework - * Added optional optimization for NIST MODP curves (POLARSSL_ECP_NIST_OPTIM) - * Support for Camellia-GCM mode and ciphersuites - -Changes - * Padding checks in cipher layer are now constant-time - * Value comparisons in SSL layer are now constant-time - * Support for serialNumber, postalAddress and postalCode in X509 names - * SSL Renegotiation was refactored - -Bugfix - * More stringent checks in cipher layer - * Server does not send out extensions not advertised by client - * Prevent possible alignment warnings on casting from char * to 'aligned *' - * Misc fixes and additions to dependency checks - * Const correctness - * cert_write with selfsign should use issuer_name as subject_name - * Fix ECDSA corner case: missing reduction mod N (found by DualTachyon) - * Defines to handle UEFI environment under MSVC - * Server-side initiated renegotiations send HelloRequest - -= PolarSSL 1.3.1 released on 2013-10-15 -Features - * Support for Brainpool curves and TLS ciphersuites (RFC 7027) - * Support for ECDHE-PSK key-exchange and ciphersuites - * Support for RSA-PSK key-exchange and ciphersuites - -Changes - * RSA blinding locks for a smaller amount of time - * TLS compression only allocates working buffer once - * Introduced POLARSSL_HAVE_READDIR_R for systems without it - * config.h is more script-friendly - -Bugfix - * Missing MSVC defines added - * Compile errors with POLARSSL_RSA_NO_CRT - * Header files with 'polarssl/' - * Const correctness - * Possible naming collision in dhm_context - * Better support for MSVC - * threading_set_alt() name - * Added missing x509write_crt_set_version() - -= PolarSSL 1.3.0 released on 2013-10-01 -Features - * Elliptic Curve Cryptography module added - * Elliptic Curve Diffie Hellman module added - * Ephemeral Elliptic Curve Diffie Hellman support for SSL/TLS - (ECDHE-based ciphersuites) - * Ephemeral Elliptic Curve Digital Signature Algorithm support for SSL/TLS - (ECDSA-based ciphersuites) - * Ability to specify allowed ciphersuites based on the protocol version. - * PSK and DHE-PSK based ciphersuites added - * Memory allocation abstraction layer added - * Buffer-based memory allocator added (no malloc() / free() / HEAP usage) - * Threading abstraction layer added (dummy / pthread / alternate) - * Public Key abstraction layer added - * Parsing Elliptic Curve keys - * Parsing Elliptic Curve certificates - * Support for max_fragment_length extension (RFC 6066) - * Support for truncated_hmac extension (RFC 6066) - * Support for zeros-and-length (ANSI X.923) padding, one-and-zeros - (ISO/IEC 7816-4) padding and zero padding in the cipher layer - * Support for session tickets (RFC 5077) - * Certificate Request (CSR) generation with extensions (key_usage, - ns_cert_type) - * X509 Certificate writing with extensions (basic_constraints, - issuer_key_identifier, etc) - * Optional blinding for RSA, DHM and EC - * Support for multiple active certificate / key pairs in SSL servers for - the same host (Not to be confused with SNI!) - -Changes - * Ability to enable / disable SSL v3 / TLS 1.0 / TLS 1.1 / TLS 1.2 - individually - * Introduced separate SSL Ciphersuites module that is based on - Cipher and MD information - * Internals for SSL module adapted to have separate IV pointer that is - dynamically set (Better support for hardware acceleration) - * Moved all OID functionality to a separate module. RSA function - prototypes for the RSA sign and verify functions changed as a result - * Split up the GCM module into a starts/update/finish cycle - * Client and server now filter sent and accepted ciphersuites on minimum - and maximum protocol version - * Ability to disable server_name extension (RFC 6066) - * Renamed error_strerror() to the less conflicting polarssl_strerror() - (Ability to keep old as well with POLARSSL_ERROR_STRERROR_BC) - * SHA2 renamed to SHA256, SHA4 renamed to SHA512 and functions accordingly - * All RSA operations require a random generator for blinding purposes - * X509 core refactored - * x509_crt_verify() now case insensitive for cn (RFC 6125 6.4) - * Also compiles / runs without time-based functions (!POLARSSL_HAVE_TIME) - * Support faulty X509 v1 certificates with extensions - (POLARSSL_X509_ALLOW_EXTENSIONS_NON_V3) - -Bugfix - * Fixed parse error in ssl_parse_certificate_request() - * zlib compression/decompression skipped on empty blocks - * Support for AIX header locations in net.c module - * Fixed file descriptor leaks - -Security - * RSA blinding on CRT operations to counter timing attacks - (found by Cyril Arnaud and Pierre-Alain Fouque) - - -= Version 1.2.14 released 2015-05-?? - -Security - * Fix potential invalid memory read in the server, that allows a client to - crash it remotely (found by Caj Larsson). - * Fix potential invalid memory read in certificate parsing, that allows a - client to crash the server remotely if client authentication is enabled - (found using Codenomicon Defensics). - * Add countermeasure against "Lucky 13 strikes back" cache-based attack, - https://dl.acm.org/citation.cfm?id=2714625 - -Bugfix - * Fix bug in Via Padlock support (found by Nikos Mavrogiannopoulos). - * Fix hardclock() (only used in the benchmarking program) with some - versions of mingw64 (found by kxjhlele). - * Fix warnings from mingw64 in timing.c (found by kxjklele). - * Fix potential unintended sign extension in asn1_get_len() on 64-bit - platforms (found with Coverity Scan). - -= Version 1.2.13 released 2015-02-16 -Note: Although PolarSSL has been renamed to mbed TLS, no changes reflecting - this will be made in the 1.2 branch at this point. - -Security - * Fix remotely-triggerable uninitialised pointer dereference caused by - crafted X.509 certificate (TLS server is not affected if it doesn't ask - for a client certificate) (found using Codenomicon Defensics). - * Fix remotely-triggerable memory leak caused by crafted X.509 certificates - (TLS server is not affected if it doesn't ask for a client certificate) - (found using Codenomicon Defensics). - * Fix potential stack overflow while parsing crafted X.509 certificates - (TLS server is not affected if it doesn't ask for a client certificate) - found using Codenomicon Defensics). - * Fix buffer overread of size 1 when parsing crafted X.509 certificates - (TLS server is not affected if it doesn't ask for a client certificate). - -Bugfix - * Fix potential undefined behaviour in Camellia. - * Fix memory leaks in PKCS#5 and PKCS#12. - * Stack buffer overflow if ctr_drbg_update() is called with too large - add_len (found by Jean-Philippe Aumasson) (not triggerable remotely). - * Fix bug in MPI/bignum on s390/s390x (reported by Dan Horák) (introduced - in 1.2.12). - * Fix unchecked return code in x509_crt_parse_path() on Windows (found by - Peter Vaskovic). - * Fix assembly selection for MIPS64 (thanks to James Cowgill). - * ssl_get_verify_result() now works even if the handshake was aborted due - to a failed verification (found by Fredrik Axelsson). - * Skip writing and parsing signature_algorithm extension if none of the - key exchanges enabled needs certificates. This fixes a possible interop - issue with some servers when a zero-length extension was sent. (Reported - by Peter Dettman.) - * On a 0-length input, base64_encode() did not correctly set output length - (found by Hendrik van den Boogaard). - -Changes - * Blind RSA private operations even when POLARSSL_RSA_NO_CRT is defined. - * Forbid repeated extensions in X.509 certificates. - * Add compile-time option POLARSSL_X509_MAX_INTERMEDIATE_CA to limit the - length of an X.509 verification chain (default = 8). -= Version 1.2.12 released 2014-10-24 - -Security - * Remotely-triggerable memory leak when parsing some X.509 certificates - (server is not affected if it doesn't ask for a client certificate). - (Found using Codenomicon Defensics.) - -Bugfix - * Fix potential bad read in parsing ServerHello (found by Adrien - Vialletelle). - * ssl_close_notify() could send more than one message in some circumstances - with non-blocking I/O. - * x509_crt_parse() did not increase total_failed on PEM error - * Fix compiler warnings on iOS (found by Sander Niemeijer). - * Don't print uninitialised buffer in ssl_mail_client (found by Marc Abel). - * Fix net_accept() regarding non-blocking sockets (found by Luca Pesce). - * ssl_read() could return non-application data records on server while - renegotation was pending, and on client when a HelloRequest was received. - * Fix warnings from Clang's scan-build (contributed by Alfred Klomp). - -Changes - * X.509 certificates with more than one AttributeTypeAndValue per - RelativeDistinguishedName are not accepted any more. - * ssl_read() now returns POLARSSL_ERR_NET_WANT_READ rather than - POLARSSL_ERR_SSL_UNEXPECTED_MESSAGE on harmless alerts. - * Accept spaces at end of line or end of buffer in base64_decode(). - -= Version 1.2.11 released 2014-07-11 -Features - * Entropy module now supports seed writing and reading - -Changes - * Introduced POLARSSL_HAVE_READDIR_R for systems without it - * Improvements to the CMake build system, contributed by Julian Ospald. - * Work around a bug of the version of Clang shipped by Apple with Mavericks - that prevented bignum.c from compiling. (Reported by Rafael Baptista.) - * Improvements to tests/Makefile, contributed by Oden Eriksson. - * Use UTC time to check certificate validity. - * Reject certificates with times not in UTC, per RFC 5280. - * Migrate zeroizing of data to polarssl_zeroize() instead of memset() - against unwanted compiler optimizations - -Security - * Forbid change of server certificate during renegotiation to prevent - "triple handshake" attack when authentication mode is optional (the - attack was already impossible when authentication is required). - * Check notBefore timestamp of certificates and CRLs from the future. - * Forbid sequence number wrapping - * Prevent potential NULL pointer dereference in ssl_read_record() (found by - TrustInSoft) - * Fix length checking for AEAD ciphersuites (found by Codenomicon). - It was possible to crash the server (and client) using crafted messages - when a GCM suite was chosen. - -Bugfix - * Fixed X.509 hostname comparison (with non-regular characters) - * SSL now gracefully handles missing RNG - * crypt_and_hash app checks MAC before final decryption - * Fixed x509_crt_parse_path() bug on Windows platforms - * Added missing MPI_CHK() around some statements in mpi_div_mpi() (found by - TrustInSoft) - * Fixed potential overflow in certificate size verification in - ssl_write_certificate() (found by TrustInSoft) - * Fix ASM format in bn_mul.h - * Potential memory leak in bignum_selftest() - * Replaced expired test certificate - * ssl_mail_client now terminates lines with CRLF, instead of LF - * Fix bug in RSA PKCS#1 v1.5 "reversed" operations - * Fixed testing with out-of-source builds using cmake - * Fixed version-major intolerance in server - * Fixed CMake symlinking on out-of-source builds - * Bignum's MIPS-32 assembly was used on MIPS-64, causing chaos. (Found by - Alex Wilson.) - * ssl_init() was leaving a dirty pointer in ssl_context if malloc of - out_ctr failed - * ssl_handshake_init() was leaving dirty pointers in subcontexts if malloc - of one of them failed - * x509_get_current_time() uses localtime_r() to prevent thread issues - * Some example server programs were not sending the close_notify alert. - * Potential memory leak in mpi_exp_mod() when error occurs during - calculation of RR. - * Improve interoperability by not writing extension length in ClientHello - when no extensions are present (found by Matthew Page) - * rsa_check_pubkey() now allows an E up to N - * On OpenBSD, use arc4random_buf() instead of rand() to prevent warnings - * mpi_fill_random() was creating numbers larger than requested on - big-endian platform when size was not an integer number of limbs - * Fix detection of DragonflyBSD in net.c (found by Markus Pfeiffer) - * Stricter check on SSL ClientHello internal sizes compared to actual packet - size (found by TrustInSoft) - * Fix preprocessor checks for bn_mul PPC asm (found by Barry K. Nathan). - * Use \n\t rather than semicolons for bn_mul asm, since some assemblers - interpret semicolons as comment delimiters (found by Barry K. Nathan). - * Disable broken Sparc64 bn_mul assembly (found by Florian Obser). - * Fix base64_decode() to return and check length correctly (in case of - tight buffers) - -= Version 1.2.10 released 2013-10-07 -Changes - * Changed RSA blinding to a slower but thread-safe version - -Bugfix - * Fixed memory leak in RSA as a result of introduction of blinding - * Fixed ssl_pkcs11_decrypt() prototype - * Fixed MSVC project files - -= Version 1.2.9 released 2013-10-01 -Changes - * x509_verify() now case insensitive for cn (RFC 6125 6.4) - -Bugfix - * Fixed potential memory leak when failing to resume a session - * Fixed potential file descriptor leaks (found by Remi Gacogne) - * Minor fixes - -Security - * Fixed potential heap buffer overflow on large hostname setting - * Fixed potential negative value misinterpretation in load_file() - * RSA blinding on CRT operations to counter timing attacks - (found by Cyril Arnaud and Pierre-Alain Fouque) - -= Version 1.2.8 released 2013-06-19 -Features - * Parsing of PKCS#8 encrypted private key files - * PKCS#12 PBE and derivation functions - * Centralized module option values in config.h to allow user-defined - settings without editing header files by using POLARSSL_CONFIG_OPTIONS - -Changes - * HAVEGE random generator disabled by default - * Internally split up x509parse_key() into a (PEM) handler function - and specific DER parser functions for the PKCS#1 and unencrypted - PKCS#8 private key formats - * Added mechanism to provide alternative implementations for all - symmetric cipher and hash algorithms (e.g. POLARSSL_AES_ALT in - config.h) - * PKCS#5 module added. Moved PBKDF2 functionality inside and deprecated - old PBKDF2 module - -Bugfix - * Secure renegotiation extension should only be sent in case client - supports secure renegotiation - * Fixed offset for cert_type list in ssl_parse_certificate_request() - * Fixed const correctness issues that have no impact on the ABI - * x509parse_crt() now better handles PEM error situations - * ssl_parse_certificate() now calls x509parse_crt_der() directly - instead of the x509parse_crt() wrapper that can also parse PEM - certificates - * x509parse_crtpath() is now reentrant and uses more portable stat() - * Fixed bignum.c and bn_mul.h to support Thumb2 and LLVM compiler - * Fixed values for 2-key Triple DES in cipher layer - * ssl_write_certificate_request() can handle empty ca_chain - -Security - * A possible DoS during the SSL Handshake, due to faulty parsing of - PEM-encoded certificates has been fixed (found by Jack Lloyd) - -= Version 1.2.7 released 2013-04-13 -Features - * Ability to specify allowed ciphersuites based on the protocol version. - -Changes - * Default Blowfish keysize is now 128-bits - * Test suites made smaller to accommodate Raspberry Pi - -Bugfix - * Fix for MPI assembly for ARM - * GCM adapted to support sizes > 2^29 - -= Version 1.2.6 released 2013-03-11 -Bugfix - * Fixed memory leak in ssl_free() and ssl_reset() for active session - * Corrected GCM counter incrementation to use only 32-bits instead of - 128-bits (found by Yawning Angel) - * Fixes for 64-bit compilation with MS Visual Studio - * Fixed net_bind() for specified IP addresses on little endian systems - * Fixed assembly code for ARM (Thumb and regular) for some compilers - -Changes - * Internally split up rsa_pkcs1_encrypt(), rsa_pkcs1_decrypt(), - rsa_pkcs1_sign() and rsa_pkcs1_verify() to separate PKCS#1 v1.5 and - PKCS#1 v2.1 functions - * Added support for custom labels when using rsa_rsaes_oaep_encrypt() - or rsa_rsaes_oaep_decrypt() - * Re-added handling for SSLv2 Client Hello when the define - POLARSSL_SSL_SRV_SUPPORT_SSLV2_CLIENT_HELLO is set - * The SSL session cache module (ssl_cache) now also retains peer_cert - information (not the entire chain) - -Security - * Removed further timing differences during SSL message decryption in - ssl_decrypt_buf() - * Removed timing differences due to bad padding from - rsa_rsaes_pkcs1_v15_decrypt() and rsa_pkcs1_decrypt() for PKCS#1 v1.5 - operations - -= Version 1.2.5 released 2013-02-02 -Changes - * Allow enabling of dummy error_strerror() to support some use-cases - * Debug messages about padding errors during SSL message decryption are - disabled by default and can be enabled with POLARSSL_SSL_DEBUG_ALL - * Sending of security-relevant alert messages that do not break - interoperability can be switched on/off with the flag - POLARSSL_SSL_ALL_ALERT_MESSAGES - -Security - * Removed timing differences during SSL message decryption in - ssl_decrypt_buf() due to badly formatted padding - -= Version 1.2.4 released 2013-01-25 -Changes - * More advanced SSL ciphersuite representation and moved to more dynamic - SSL core - * Added ssl_handshake_step() to allow single stepping the handshake process - -Bugfix - * Memory leak when using RSA_PKCS_V21 operations fixed - * Handle future version properly in ssl_write_certificate_request() - * Correctly handle CertificateRequest message in client for <= TLS 1.1 - without DN list - -= Version 1.2.3 released 2012-11-26 -Bugfix - * Server not always sending correct CertificateRequest message - -= Version 1.2.2 released 2012-11-24 -Changes - * Added p_hw_data to ssl_context for context specific hardware acceleration - data - * During verify trust-CA is only checked for expiration and CRL presence - -Bugfixes - * Fixed client authentication compatibility - * Fixed dependency on POLARSSL_SHA4_C in SSL modules - -= Version 1.2.1 released 2012-11-20 -Changes - * Depth that the certificate verify callback receives is now numbered - bottom-up (Peer cert depth is 0) - -Bugfixes - * Fixes for MSVC6 - * Moved mpi_inv_mod() outside POLARSSL_GENPRIME - * Allow R and A to point to same mpi in mpi_div_mpi (found by Manuel - Pégourié-Gonnard) - * Fixed possible segfault in mpi_shift_r() (found by Manuel - Pégourié-Gonnard) - * Added max length check for rsa_pkcs1_sign with PKCS#1 v2.1 - -= Version 1.2.0 released 2012-10-31 -Features - * Added support for NULL cipher (POLARSSL_CIPHER_NULL_CIPHER) and weak - ciphersuites (POLARSSL_ENABLE_WEAK_CIPHERSUITES). They are disabled by - default! - * Added support for wildcard certificates - * Added support for multi-domain certificates through the X509 Subject - Alternative Name extension - * Added preliminary ASN.1 buffer writing support - * Added preliminary X509 Certificate Request writing support - * Added key_app_writer example application - * Added cert_req example application - * Added base Galois Counter Mode (GCM) for AES - * Added TLS 1.2 support (RFC 5246) - * Added GCM suites to TLS 1.2 (RFC 5288) - * Added commandline error code convertor (util/strerror) - * Added support for Hardware Acceleration hooking in SSL/TLS - * Added OpenSSL / PolarSSL compatibility script (tests/compat.sh) and - example application (programs/ssl/o_p_test) (requires OpenSSL) - * Added X509 CA Path support - * Added Thumb assembly optimizations - * Added DEFLATE compression support as per RFC3749 (requires zlib) - * Added blowfish algorithm (Generic and cipher layer) - * Added PKCS#5 PBKDF2 key derivation function - * Added Secure Renegotiation (RFC 5746) - * Added predefined DHM groups from RFC 5114 - * Added simple SSL session cache implementation - * Added ServerName extension parsing (SNI) at server side - * Added option to add minimum accepted SSL/TLS protocol version - -Changes - * Removed redundant POLARSSL_DEBUG_MSG define - * AES code only check for Padlock once - * Fixed const-correctness mpi_get_bit() - * Documentation for mpi_lsb() and mpi_msb() - * Moved out_msg to out_hdr + 32 to support hardware acceleration - * Changed certificate verify behaviour to comply with RFC 6125 section 6.3 - to not match CN if subjectAltName extension is present (Closes ticket #56) - * Cipher layer cipher_mode_t POLARSSL_MODE_CFB128 is renamed to - POLARSSL_MODE_CFB, to also handle different block size CFB modes. - * Removed handling for SSLv2 Client Hello (as per RFC 5246 recommendation) - * Revamped session resumption handling - * Generalized external private key implementation handling (like PKCS#11) - in SSL/TLS - * Revamped x509_verify() and the SSL f_vrfy callback implementations - * Moved from unsigned long to fixed width uint32_t types throughout code - * Renamed ciphersuites naming scheme to IANA reserved names - -Bugfix - * Fixed handling error in mpi_cmp_mpi() on longer B values (found by - Hui Dong) - * Fixed potential heap corruption in x509_name allocation - * Fixed single RSA test that failed on Big Endian systems (Closes ticket #54) - * mpi_exp_mod() now correctly handles negative base numbers (Closes ticket - #52) - * Handle encryption with private key and decryption with public key as per - RFC 2313 - * Handle empty certificate subject names - * Prevent reading over buffer boundaries on X509 certificate parsing - * mpi_add_abs() now correctly handles adding short numbers to long numbers - with carry rollover (found by Ruslan Yushchenko) - * Handle existence of OpenSSL Trust Extensions at end of X.509 DER blob - * Fixed MPI assembly for SPARC64 platform - -Security - * Fixed potential memory zeroization on miscrafted RSA key (found by Eloi - Vanderbeken) - -= Version 1.1.8 released on 2013-10-01 -Bugfix - * Fixed potential memory leak when failing to resume a session - * Fixed potential file descriptor leaks - -Security - * Potential buffer-overflow for ssl_read_record() (independently found by - both TrustInSoft and Paul Brodeur of Leviathan Security Group) - * Potential negative value misinterpretation in load_file() - * Potential heap buffer overflow on large hostname setting - -= Version 1.1.7 released on 2013-06-19 -Changes - * HAVEGE random generator disabled by default - -Bugfix - * x509parse_crt() now better handles PEM error situations - * ssl_parse_certificate() now calls x509parse_crt_der() directly - instead of the x509parse_crt() wrapper that can also parse PEM - certificates - * Fixed values for 2-key Triple DES in cipher layer - * ssl_write_certificate_request() can handle empty ca_chain - -Security - * A possible DoS during the SSL Handshake, due to faulty parsing of - PEM-encoded certificates has been fixed (found by Jack Lloyd) - -= Version 1.1.6 released on 2013-03-11 -Bugfix - * Fixed net_bind() for specified IP addresses on little endian systems - -Changes - * Allow enabling of dummy error_strerror() to support some use-cases - * Debug messages about padding errors during SSL message decryption are - disabled by default and can be enabled with POLARSSL_SSL_DEBUG_ALL - -Security - * Removed timing differences during SSL message decryption in - ssl_decrypt_buf() - * Removed timing differences due to bad padding from - rsa_rsaes_pkcs1_v15_decrypt() and rsa_pkcs1_decrypt() for PKCS#1 v1.5 - operations - -= Version 1.1.5 released on 2013-01-16 -Bugfix - * Fixed MPI assembly for SPARC64 platform - * Handle existence of OpenSSL Trust Extensions at end of X.509 DER blob - * mpi_add_abs() now correctly handles adding short numbers to long numbers - with carry rollover - * Moved mpi_inv_mod() outside POLARSSL_GENPRIME - * Prevent reading over buffer boundaries on X509 certificate parsing - * mpi_exp_mod() now correctly handles negative base numbers (Closes ticket - #52) - * Fixed possible segfault in mpi_shift_r() (found by Manuel - Pégourié-Gonnard) - * Allow R and A to point to same mpi in mpi_div_mpi (found by Manuel - Pégourié-Gonnard) - * Added max length check for rsa_pkcs1_sign with PKCS#1 v2.1 - * Memory leak when using RSA_PKCS_V21 operations fixed - * Handle encryption with private key and decryption with public key as per - RFC 2313 - * Fixes for MSVC6 - -Security - * Fixed potential memory zeroization on miscrafted RSA key (found by Eloi - Vanderbeken) - -= Version 1.1.4 released on 2012-05-31 -Bugfix - * Correctly handle empty SSL/TLS packets (Found by James Yonan) - * Fixed potential heap corruption in x509_name allocation - * Fixed single RSA test that failed on Big Endian systems (Closes ticket #54) - -= Version 1.1.3 released on 2012-04-29 -Bugfix - * Fixed random MPI generation to not generate more size than requested. - -= Version 1.1.2 released on 2012-04-26 -Bugfix - * Fixed handling error in mpi_cmp_mpi() on longer B values (found by - Hui Dong) - -Security - * Fixed potential memory corruption on miscrafted client messages (found by - Frama-C team at CEA LIST) - * Fixed generation of DHM parameters to correct length (found by Ruslan - Yushchenko) - -= Version 1.1.1 released on 2012-01-23 -Bugfix - * Check for failed malloc() in ssl_set_hostname() and x509_get_entries() - (Closes ticket #47, found by Hugo Leisink) - * Fixed issues with Intel compiler on 64-bit systems (Closes ticket #50) - * Fixed multiple compiler warnings for VS6 and armcc - * Fixed bug in CTR_CRBG selftest - -= Version 1.1.0 released on 2011-12-22 -Features - * Added ssl_session_reset() to allow better multi-connection pools of - SSL contexts without needing to set all non-connection-specific - data and pointers again. Adapted ssl_server to use this functionality. - * Added ssl_set_max_version() to allow clients to offer a lower maximum - supported version to a server to help buggy server implementations. - (Closes ticket #36) - * Added cipher_get_cipher_mode() and cipher_get_cipher_operation() - introspection functions (Closes ticket #40) - * Added CTR_DRBG based on AES-256-CTR (NIST SP 800-90) random generator - * Added a generic entropy accumulator that provides support for adding - custom entropy sources and added some generic and platform dependent - entropy sources - -Changes - * Documentation for AES and Camellia in modes CTR and CFB128 clarified. - * Fixed rsa_encrypt and rsa_decrypt examples to use public key for - encryption and private key for decryption. (Closes ticket #34) - * Inceased maximum size of ASN1 length reads to 32-bits. - * Added an EXPLICIT tag number parameter to x509_get_ext() - * Added a separate CRL entry extension parsing function - * Separated the ASN.1 parsing code from the X.509 specific parsing code. - So now there is a module that is controlled with POLARSSL_ASN1_PARSE_C. - * Changed the defined key-length of DES ciphers in cipher.h to include the - parity bits, to prevent mistakes in copying data. (Closes ticket #33) - * Loads of minimal changes to better support WINCE as a build target - (Credits go to Marco Lizza) - * Added POLARSSL_MPI_WINDOW_SIZE definition to allow easier time to memory - trade-off - * Introduced POLARSSL_MPI_MAX_SIZE and POLARSSL_MPI_MAX_BITS for MPI size - management (Closes ticket #44) - * Changed the used random function pointer to more flexible format. Renamed - havege_rand() to havege_random() to prevent mistakes. Lots of changes as - a consequence in library code and programs - * Moved all examples programs to use the new entropy and CTR_DRBG - * Added permissive certificate parsing to x509parse_crt() and - x509parse_crtfile(). With permissive parsing the parsing does not stop on - encountering a parse-error. Beware that the meaning of return values has - changed! - * All error codes are now negative. Even on mermory failures and IO errors. - -Bugfix - * Fixed faulty HMAC-MD2 implementation. Found by dibac. (Closes - ticket #37) - * Fixed a bug where the CRL parser expected an EXPLICIT ASN.1 tag - before version numbers - * Allowed X509 key usage parsing to accept 4 byte values instead of the - standard 1 byte version sometimes used by Microsoft. (Closes ticket #38) - * Fixed incorrect behaviour in case of RSASSA-PSS with a salt length - smaller than the hash length. (Closes ticket #41) - * If certificate serial is longer than 32 octets, serial number is now - appended with '....' after first 28 octets - * Improved build support for s390x and sparc64 in bignum.h - * Fixed MS Visual C++ name clash with int64 in sha4.h - * Corrected removal of leading "00:" in printing serial numbers in - certificates and CRLs - -= Version 1.0.0 released on 2011-07-27 -Features - * Expanded cipher layer with support for CFB128 and CTR mode - * Added rsa_encrypt and rsa_decrypt simple example programs. - -Changes - * The generic cipher and message digest layer now have normal error - codes instead of integers - -Bugfix - * Undid faulty bug fix in ssl_write() when flushing old data (Ticket - #18) - -= Version 0.99-pre5 released on 2011-05-26 -Features - * Added additional Cipher Block Modes to symmetric ciphers - (AES CTR, Camellia CTR, XTEA CBC) including the option to - enable and disable individual modes when needed - * Functions requiring File System functions can now be disabled - by undefining POLARSSL_FS_IO - * A error_strerror function() has been added to translate between - error codes and their description. - * Added mpi_get_bit() and mpi_set_bit() individual bit setter/getter - functions. - * Added ssl_mail_client and ssl_fork_server as example programs. - -Changes - * Major argument / variable rewrite. Introduced use of size_t - instead of int for buffer lengths and loop variables for - better unsigned / signed use. Renamed internal bigint types - t_int and t_dbl to t_uint and t_udbl in the process - * mpi_init() and mpi_free() now only accept a single MPI - argument and do not accept variable argument lists anymore. - * The error codes have been remapped and combining error codes - is now done with a PLUS instead of an OR as error codes - used are negative. - * Changed behaviour of net_read(), ssl_fetch_input() and ssl_recv(). - net_recv() now returns 0 on EOF instead of - POLARSSL_ERR_NET_CONN_RESET. ssl_fetch_input() returns - POLARSSL_ERR_SSL_CONN_EOF on an EOF from its f_recv() function. - ssl_read() returns 0 if a POLARSSL_ERR_SSL_CONN_EOF is received - after the handshake. - * Network functions now return POLARSSL_ERR_NET_WANT_READ or - POLARSSL_ERR_NET_WANT_WRITE instead of the ambiguous - POLARSSL_ERR_NET_TRY_AGAIN - -= Version 0.99-pre4 released on 2011-04-01 -Features - * Added support for PKCS#1 v2.1 encoding and thus support - for the RSAES-OAEP and RSASSA-PSS operations. - * Reading of Public Key files incorporated into default x509 - functionality as well. - * Added mpi_fill_random() for centralized filling of big numbers - with random data (Fixed ticket #10) - -Changes - * Debug print of MPI now removes leading zero octets and - displays actual bit size of the value. - * x509parse_key() (and as a consequence x509parse_keyfile()) - does not zeroize memory in advance anymore. Use rsa_init() - before parsing a key or keyfile! - -Bugfix - * Debug output of MPI's now the same independent of underlying - platform (32-bit / 64-bit) (Fixes ticket #19, found by Mads - Kiilerich and Mihai Militaru) - * Fixed bug in ssl_write() when flushing old data (Fixed ticket - #18, found by Nikolay Epifanov) - * Fixed proper handling of RSASSA-PSS verification with variable - length salt lengths - -= Version 0.99-pre3 released on 2011-02-28 -This release replaces version 0.99-pre2 which had possible copyright issues. -Features - * Parsing PEM private keys encrypted with DES and AES - are now supported as well (Fixes ticket #5) - * Added crl_app program to allow easy reading and - printing of X509 CRLs from file - -Changes - * Parsing of PEM files moved to separate module (Fixes - ticket #13). Also possible to remove PEM support for - systems only using DER encoding - -Bugfixes - * Corrected parsing of UTCTime dates before 1990 and - after 1950 - * Support more exotic OID's when parsing certificates - (found by Mads Kiilerich) - * Support more exotic name representations when parsing - certificates (found by Mads Kiilerich) - * Replaced the expired test certificates - * Do not bail out if no client certificate specified. Try - to negotiate anonymous connection (Fixes ticket #12, - found by Boris Krasnovskiy) - -Security fixes - * Fixed a possible Man-in-the-Middle attack on the - Diffie Hellman key exchange (thanks to Larry Highsmith, - Subreption LLC) - -= Version 0.99-pre1 released on 2011-01-30 -Features -Note: Most of these features have been donated by Fox-IT - * Added Doxygen source code documentation parts - * Added reading of DHM context from memory and file - * Improved X509 certificate parsing to include extended - certificate fields, including Key Usage - * Improved certificate verification and verification - against the available CRLs - * Detection for DES weak keys and parity bits added - * Improvements to support integration in other - applications: - + Added generic message digest and cipher wrapper - + Improved information about current capabilities, - status, objects and configuration - + Added verification callback on certificate chain - verification to allow external blacklisting - + Additional example programs to show usage - * Added support for PKCS#11 through the use of the - libpkcs11-helper library - -Changes - * x509parse_time_expired() checks time in addition to - the existing date check - * The ciphers member of ssl_context and the cipher member - of ssl_session have been renamed to ciphersuites and - ciphersuite respectively. This clarifies the difference - with the generic cipher layer and is better naming - altogether - -= Version 0.14.0 released on 2010-08-16 -Features - * Added support for SSL_EDH_RSA_AES_128_SHA and - SSL_EDH_RSA_CAMELLIA_128_SHA ciphersuites - * Added compile-time and run-time version information - * Expanded ssl_client2 arguments for more flexibility - * Added support for TLS v1.1 - -Changes - * Made Makefile cleaner - * Removed dependency on rand() in rsa_pkcs1_encrypt(). - Now using random fuction provided to function and - changed the prototype of rsa_pkcs1_encrypt(), - rsa_init() and rsa_gen_key(). - * Some SSL defines were renamed in order to avoid - future confusion - -Bug fixes - * Fixed CMake out of source build for tests (found by - kkert) - * rsa_check_private() now supports PKCS1v2 keys as well - * Fixed deadlock in rsa_pkcs1_encrypt() on failing random - generator - -= Version 0.13.1 released on 2010-03-24 -Bug fixes - * Fixed Makefile in library that was mistakenly merged - * Added missing const string fixes - -= Version 0.13.0 released on 2010-03-21 -Features - * Added option parsing for host and port selection to - ssl_client2 - * Added support for GeneralizedTime in X509 parsing - * Added cert_app program to allow easy reading and - printing of X509 certificates from file or SSL - connection. - -Changes - * Added const correctness for main code base - * X509 signature algorithm determination is now - in a function to allow easy future expansion - * Changed symmetric cipher functions to - identical interface (returning int result values) - * Changed ARC4 to use separate input/output buffer - * Added reset function for HMAC context as speed-up - for specific use-cases - -Bug fixes - * Fixed bug resulting in failure to send the last - certificate in the chain in ssl_write_certificate() and - ssl_write_certificate_request() (found by fatbob) - * Added small fixes for compiler warnings on a Mac - (found by Frank de Brabander) - * Fixed algorithmic bug in mpi_is_prime() (found by - Smbat Tonoyan) - -= Version 0.12.1 released on 2009-10-04 -Changes - * Coverage test definitions now support 'depends_on' - tagging system. - * Tests requiring specific hashing algorithms now honor - the defines. - -Bug fixes - * Changed typo in #ifdef in x509parse.c (found - by Eduardo) - -= Version 0.12.0 released on 2009-07-28 -Features - * Added CMake makefiles as alternative to regular Makefiles. - * Added preliminary Code Coverage tests for AES, ARC4, - Base64, MPI, SHA-family, MD-family, HMAC-SHA-family, - Camellia, DES, 3-DES, RSA PKCS#1, XTEA, Diffie-Hellman - and X509parse. - -Changes - * Error codes are not (necessarily) negative. Keep - this is mind when checking for errors. - * RSA_RAW renamed to SIG_RSA_RAW for consistency. - * Fixed typo in name of POLARSSL_ERR_RSA_OUTPUT_TOO_LARGE. - * Changed interface for AES and Camellia setkey functions - to indicate invalid key lengths. - -Bug fixes - * Fixed include location of endian.h on FreeBSD (found by - Gabriel) - * Fixed include location of endian.h and name clash on - Apples (found by Martin van Hensbergen) - * Fixed HMAC-MD2 by modifying md2_starts(), so that the - required HMAC ipad and opad variables are not cleared. - (found by code coverage tests) - * Prevented use of long long in bignum if - POLARSSL_HAVE_LONGLONG not defined (found by Giles - Bathgate). - * Fixed incorrect handling of negative strings in - mpi_read_string() (found by code coverage tests). - * Fixed segfault on handling empty rsa_context in - rsa_check_pubkey() and rsa_check_privkey() (found by - code coverage tests). - * Fixed incorrect handling of one single negative input - value in mpi_add_abs() (found by code coverage tests). - * Fixed incorrect handling of negative first input - value in mpi_sub_abs() (found by code coverage tests). - * Fixed incorrect handling of negative first input - value in mpi_mod_mpi() and mpi_mod_int(). Resulting - change also affects mpi_write_string() (found by code - coverage tests). - * Corrected is_prime() results for 0, 1 and 2 (found by - code coverage tests). - * Fixed Camellia and XTEA for 64-bit Windows systems. - -= Version 0.11.1 released on 2009-05-17 - * Fixed missing functionality for SHA-224, SHA-256, SHA384, - SHA-512 in rsa_pkcs1_sign() - -= Version 0.11.0 released on 2009-05-03 - * Fixed a bug in mpi_gcd() so that it also works when both - input numbers are even and added testcases to check - (found by Pierre Habouzit). - * Added support for SHA-224, SHA-256, SHA-384 and SHA-512 - one way hash functions with the PKCS#1 v1.5 signing and - verification. - * Fixed minor bug regarding mpi_gcd located within the - POLARSSL_GENPRIME block. - * Fixed minor memory leak in x509parse_crt() and added better - handling of 'full' certificate chains (found by Mathias - Olsson). - * Centralized file opening and reading for x509 files into - load_file() - * Made definition of net_htons() endian-clean for big endian - systems (Found by Gernot). - * Undefining POLARSSL_HAVE_ASM now also handles prevents asm in - padlock and timing code. - * Fixed an off-by-one buffer allocation in ssl_set_hostname() - responsible for crashes and unwanted behaviour. - * Added support for Certificate Revocation List (CRL) parsing. - * Added support for CRL revocation to x509parse_verify() and - SSL/TLS code. - * Fixed compatibility of XTEA and Camellia on a 64-bit system - (found by Felix von Leitner). - -= Version 0.10.0 released on 2009-01-12 - * Migrated XySSL to PolarSSL - * Added XTEA symmetric cipher - * Added Camellia symmetric cipher - * Added support for ciphersuites: SSL_RSA_CAMELLIA_128_SHA, - SSL_RSA_CAMELLIA_256_SHA and SSL_EDH_RSA_CAMELLIA_256_SHA - * Fixed dangerous bug that can cause a heap overflow in - rsa_pkcs1_decrypt (found by Christophe Devine) - -================================================================ -XySSL ChangeLog - -= Version 0.9 released on 2008-03-16 - - * Added support for ciphersuite: SSL_RSA_AES_128_SHA - * Enabled support for large files by default in aescrypt2.c - * Preliminary openssl wrapper contributed by David Barrett - * Fixed a bug in ssl_write() that caused the same payload to - be sent twice in non-blocking mode when send returns EAGAIN - * Fixed ssl_parse_client_hello(): session id and challenge must - not be swapped in the SSLv2 ClientHello (found by Greg Robson) - * Added user-defined callback debug function (Krystian Kolodziej) - * Before freeing a certificate, properly zero out all cert. data - * Fixed the "mode" parameter so that encryption/decryption are - not swapped on PadLock; also fixed compilation on older versions - of gcc (bug reported by David Barrett) - * Correctly handle the case in padlock_xcryptcbc() when input or - output data is non-aligned by falling back to the software - implementation, as VIA Nehemiah cannot handle non-aligned buffers - * Fixed a memory leak in x509parse_crt() which was reported by Greg - Robson-Garth; some x509write.c fixes by Pascal Vizeli, thanks to - Matthew Page who reported several bugs - * Fixed x509_get_ext() to accept some rare certificates which have - an INTEGER instead of a BOOLEAN for BasicConstraints::cA. - * Added support on the client side for the TLS "hostname" extension - (patch contributed by David Patino) - * Make x509parse_verify() return BADCERT_CN_MISMATCH when an empty - string is passed as the CN (bug reported by spoofy) - * Added an option to enable/disable the BN assembly code - * Updated rsa_check_privkey() to verify that (D*E) = 1 % (P-1)*(Q-1) - * Disabled obsolete hash functions by default (MD2, MD4); updated - selftest and benchmark to not test ciphers that have been disabled - * Updated x509parse_cert_info() to correctly display byte 0 of the - serial number, setup correct server port in the ssl client example - * Fixed a critical denial-of-service with X.509 cert. verification: - peer may cause xyssl to loop indefinitely by sending a certificate - for which the RSA signature check fails (bug reported by Benoit) - * Added test vectors for: AES-CBC, AES-CFB, DES-CBC and 3DES-CBC, - HMAC-MD5, HMAC-SHA1, HMAC-SHA-256, HMAC-SHA-384, and HMAC-SHA-512 - * Fixed HMAC-SHA-384 and HMAC-SHA-512 (thanks to Josh Sinykin) - * Modified ssl_parse_client_key_exchange() to protect against - Daniel Bleichenbacher attack on PKCS#1 v1.5 padding, as well - as the Klima-Pokorny-Rosa extension of Bleichenbacher's attack - * Updated rsa_gen_key() so that ctx->N is always nbits in size - * Fixed assembly PPC compilation errors on Mac OS X, thanks to - David Barrett and Dusan Semen - -= Version 0.8 released on 2007-10-20 - - * Modified the HMAC functions to handle keys larger - than 64 bytes, thanks to Stephane Desneux and gary ng - * Fixed ssl_read_record() to properly update the handshake - message digests, which fixes IE6/IE7 client authentication - * Cleaned up the XYSSL* #defines, suggested by Azriel Fasten - * Fixed net_recv(), thanks to Lorenz Schori and Egon Kocjan - * Added user-defined callbacks for handling I/O and sessions - * Added lots of debugging output in the SSL/TLS functions - * Added preliminary X.509 cert. writing by Pascal Vizeli - * Added preliminary support for the VIA PadLock routines - * Added AES-CFB mode of operation, contributed by chmike - * Added an SSL/TLS stress testing program (ssl_test.c) - * Updated the RSA PKCS#1 code to allow choosing between - RSA_PUBLIC and RSA_PRIVATE, as suggested by David Barrett - * Updated ssl_read() to skip 0-length records from OpenSSL - * Fixed the make install target to comply with *BSD make - * Fixed a bug in mpi_read_binary() on 64-bit platforms - * mpi_is_prime() speedups, thanks to Kevin McLaughlin - * Fixed a long standing memory leak in mpi_is_prime() - * Replaced realloc with malloc in mpi_grow(), and set - the sign of zero as positive in mpi_init() (reported - by Jonathan M. McCune) - -= Version 0.7 released on 2007-07-07 - - * Added support for the MicroBlaze soft-core processor - * Fixed a bug in ssl_tls.c which sometimes prevented SSL - connections from being established with non-blocking I/O - * Fixed a couple bugs in the VS6 and UNIX Makefiles - * Fixed the "PIC register ebx clobbered in asm" bug - * Added HMAC starts/update/finish support functions - * Added the SHA-224, SHA-384 and SHA-512 hash functions - * Fixed the net_set_*block routines, thanks to Andreas - * Added a few demonstration programs: md5sum, sha1sum, - dh_client, dh_server, rsa_genkey, rsa_sign, rsa_verify - * Added new bignum import and export helper functions - * Rewrote README.txt in program/ssl/ca to better explain - how to create a test PKI - -= Version 0.6 released on 2007-04-01 - - * Ciphers used in SSL/TLS can now be disabled at compile - time, to reduce the memory footprint on embedded systems - * Added multiply assembly code for the TriCore and modified - havege_struct for this processor, thanks to David Patiño - * Added multiply assembly code for 64-bit PowerPCs, - thanks to Peking University and the OSU Open Source Lab - * Added experimental support of Quantum Cryptography - * Added support for autoconf, contributed by Arnaud Cornet - * Fixed "long long" compilation issues on IA-64 and PPC64 - * Fixed a bug introduced in xyssl-0.5/timing.c: hardclock - was not being correctly defined on ARM and MIPS - -= Version 0.5 released on 2007-03-01 - - * Added multiply assembly code for SPARC and Alpha - * Added (beta) support for non-blocking I/O operations - * Implemented session resuming and client authentication - * Fixed some portability issues on WinCE, MINIX 3, Plan9 - (thanks to Benjamin Newman), HP-UX, FreeBSD and Solaris - * Improved the performance of the EDH key exchange - * Fixed a bug that caused valid packets with a payload - size of 16384 bytes to be rejected - -= Version 0.4 released on 2007-02-01 - - * Added support for Ephemeral Diffie-Hellman key exchange - * Added multiply asm code for SSE2, ARM, PPC, MIPS and M68K - * Various improvement to the modular exponentiation code - * Rewrote the headers to generate the API docs with doxygen - * Fixed a bug in ssl_encrypt_buf (incorrect padding was - generated) and in ssl_parse_client_hello (max. client - version was not properly set), thanks to Didier Rebeix - * Fixed another bug in ssl_parse_client_hello: clients with - cipherlists larger than 96 bytes were incorrectly rejected - * Fixed a couple memory leak in x509_read.c - -= Version 0.3 released on 2007-01-01 - - * Added server-side SSLv3 and TLSv1.0 support - * Multiple fixes to enhance the compatibility with g++, - thanks to Xosé Antón Otero Ferreira - * Fixed a bug in the CBC code, thanks to dowst; also, - the bignum code is no longer dependent on long long - * Updated rsa_pkcs1_sign to handle arbitrary large inputs - * Updated timing.c for improved compatibility with i386 - and 486 processors, thanks to Arnaud Cornet - -= Version 0.2 released on 2006-12-01 - - * Updated timing.c to support ARM and MIPS arch - * Updated the MPI code to support 8086 on MSVC 1.5 - * Added the copyright notice at the top of havege.h - * Fixed a bug in sha2_hmac, thanks to newsoft/Wenfang Zhang - * Fixed a bug reported by Adrian Rüegsegger in x509_read_key - * Fixed a bug reported by Torsten Lauter in ssl_read_record - * Fixed a bug in rsa_check_privkey that would wrongly cause - valid RSA keys to be dismissed (thanks to oldwolf) - * Fixed a bug in mpi_is_prime that caused some primes to fail - the Miller-Rabin primality test - - I'd also like to thank Younès Hafri for the CRUX linux port, - Khalil Petit who added XySSL into pkgsrc and Arnaud Cornet - who maintains the Debian package :-) - -= Version 0.1 released on 2006-11-01 |