5 Changes between 1.0.2 and 1.1.0 [xx XXX xxxx]
7 *) Experimental support for a new, fast, unbiased prime candidate generator,
8 bn_probable_prime_dh_coprime(). Not currently used by any prime generator.
9 [Felix Laurie von Massenbach <felix@erbridge.co.uk>]
11 *) New output format NSS in the sess_id command line tool. This allows
12 exporting the session id and the master key in NSS keylog format.
13 [Martin Kaiser <martin@kaiser.cx>]
15 *) Harmonize version and its documentation. -f flag is used to display
17 [mancha <mancha1@zoho.com>]
19 *) Fix eckey_priv_encode so it immediately returns an error upon a failure
21 [mancha <mancha1@zoho.com>]
23 *) Fix some double frees. These are not thought to be exploitable.
24 [mancha <mancha1@zoho.com>]
26 *) A missing bounds check in the handling of the TLS heartbeat extension
27 can be used to reveal up to 64k of memory to a connected client or
30 Thanks for Neel Mehta of Google Security for discovering this bug and to
31 Adam Langley <agl@chromium.org> and Bodo Moeller <bmoeller@acm.org> for
32 preparing the fix (CVE-2014-0160)
33 [Adam Langley, Bodo Moeller]
35 *) Fix for the attack described in the paper "Recovering OpenSSL
36 ECDSA Nonces Using the FLUSH+RELOAD Cache Side-channel Attack"
37 by Yuval Yarom and Naomi Benger. Details can be obtained from:
38 http://eprint.iacr.org/2014/140
40 Thanks to Yuval Yarom and Naomi Benger for discovering this
41 flaw and to Yuval Yarom for supplying a fix (CVE-2014-0076)
42 [Yuval Yarom and Naomi Benger]
44 *) Use algorithm specific chains in SSL_CTX_use_certificate_chain_file():
45 this fixes a limitation in previous versions of OpenSSL.
48 *) Experimental encrypt-then-mac support.
50 Experimental support for encrypt then mac from
51 draft-gutmann-tls-encrypt-then-mac-02.txt
53 To enable it set the appropriate extension number (0x42 for the test
54 server) using e.g. -DTLSEXT_TYPE_encrypt_then_mac=0x42
56 For non-compliant peers (i.e. just about everything) this should have no
59 WARNING: EXPERIMENTAL, SUBJECT TO CHANGE.
63 *) Add EVP support for key wrapping algorithms, to avoid problems with
64 existing code the flag EVP_CIPHER_CTX_WRAP_ALLOW has to be set in
65 the EVP_CIPHER_CTX or an error is returned. Add AES and DES3 wrap
66 algorithms and include tests cases.
69 *) Extend CMS code to support RSA-PSS signatures and RSA-OAEP for
73 *) Extended RSA OAEP support via EVP_PKEY API. Options to specify digest,
74 MGF1 digest and OAEP label.
77 *) Make openssl verify return errors.
78 [Chris Palmer <palmer@google.com> and Ben Laurie]
80 *) New function ASN1_TIME_diff to calculate the difference between two
81 ASN1_TIME structures or one structure and the current time.
84 *) Update fips_test_suite to support multiple command line options. New
85 test to induce all self test errors in sequence and check expected
89 *) Add FIPS_{rsa,dsa,ecdsa}_{sign,verify} functions which digest and
90 sign or verify all in one operation.
93 *) Add fips_algvs: a multicall fips utility incorporating all the algorithm
94 test programs and fips_test_suite. Includes functionality to parse
95 the minimal script output of fipsalgest.pl directly.
98 *) Add authorisation parameter to FIPS_module_mode_set().
101 *) Add FIPS selftest for ECDH algorithm using P-224 and B-233 curves.
104 *) Use separate DRBG fields for internal and external flags. New function
105 FIPS_drbg_health_check() to perform on demand health checking. Add
106 generation tests to fips_test_suite with reduced health check interval to
107 demonstrate periodic health checking. Add "nodh" option to
108 fips_test_suite to skip very slow DH test.
111 *) New function FIPS_get_cipherbynid() to lookup FIPS supported ciphers
115 *) More extensive health check for DRBG checking many more failure modes.
116 New function FIPS_selftest_drbg_all() to handle every possible DRBG
117 combination: call this in fips_test_suite.
120 *) Add support for Dual EC DRBG from SP800-90. Update DRBG algorithm test
121 and POST to handle Dual EC cases.
124 *) Add support for canonical generation of DSA parameter 'g'. See
127 *) Add support for HMAC DRBG from SP800-90. Update DRBG algorithm test and
128 POST to handle HMAC cases.
131 *) Add functions FIPS_module_version() and FIPS_module_version_text()
132 to return numerical and string versions of the FIPS module number.
135 *) Rename FIPS_mode_set and FIPS_mode to FIPS_module_mode_set and
136 FIPS_module_mode. FIPS_mode and FIPS_mode_set will be implemented
137 outside the validated module in the FIPS capable OpenSSL.
140 *) Minor change to DRBG entropy callback semantics. In some cases
141 there is no multiple of the block length between min_len and
142 max_len. Allow the callback to return more than max_len bytes
143 of entropy but discard any extra: it is the callback's responsibility
144 to ensure that the extra data discarded does not impact the
145 requested amount of entropy.
148 *) Add PRNG security strength checks to RSA, DSA and ECDSA using
149 information in FIPS186-3, SP800-57 and SP800-131A.
152 *) CCM support via EVP. Interface is very similar to GCM case except we
153 must supply all data in one chunk (i.e. no update, final) and the
154 message length must be supplied if AAD is used. Add algorithm test
158 *) Initial version of POST overhaul. Add POST callback to allow the status
159 of POST to be monitored and/or failures induced. Modify fips_test_suite
160 to use callback. Always run all selftests even if one fails.
163 *) XTS support including algorithm test driver in the fips_gcmtest program.
164 Note: this does increase the maximum key length from 32 to 64 bytes but
165 there should be no binary compatibility issues as existing applications
166 will never use XTS mode.
169 *) Extensive reorganisation of FIPS PRNG behaviour. Remove all dependencies
170 to OpenSSL RAND code and replace with a tiny FIPS RAND API which also
171 performs algorithm blocking for unapproved PRNG types. Also do not
172 set PRNG type in FIPS_mode_set(): leave this to the application.
173 Add default OpenSSL DRBG handling: sets up FIPS PRNG and seeds with
174 the standard OpenSSL PRNG: set additional data to a date time vector.
177 *) Rename old X9.31 PRNG functions of the form FIPS_rand* to FIPS_x931*.
178 This shouldn't present any incompatibility problems because applications
179 shouldn't be using these directly and any that are will need to rethink
180 anyway as the X9.31 PRNG is now deprecated by FIPS 140-2
183 *) Extensive self tests and health checking required by SP800-90 DRBG.
184 Remove strength parameter from FIPS_drbg_instantiate and always
185 instantiate at maximum supported strength.
188 *) Add ECDH code to fips module and fips_ecdhvs for primitives only testing.
191 *) New algorithm test program fips_dhvs to handle DH primitives only testing.
194 *) New function DH_compute_key_padded() to compute a DH key and pad with
195 leading zeroes if needed: this complies with SP800-56A et al.
198 *) Initial implementation of SP800-90 DRBGs for Hash and CTR. Not used by
199 anything, incomplete, subject to change and largely untested at present.
202 *) Modify fipscanisteronly build option to only build the necessary object
203 files by filtering FIPS_EX_OBJ through a perl script in crypto/Makefile.
206 *) Add experimental option FIPSSYMS to give all symbols in
207 fipscanister.o and FIPS or fips prefix. This will avoid
208 conflicts with future versions of OpenSSL. Add perl script
209 util/fipsas.pl to preprocess assembly language source files
210 and rename any affected symbols.
213 *) Add selftest checks and algorithm block of non-fips algorithms in
214 FIPS mode. Remove DES2 from selftests.
217 *) Add ECDSA code to fips module. Add tiny fips_ecdsa_check to just
218 return internal method without any ENGINE dependencies. Add new
219 tiny fips sign and verify functions.
222 *) New build option no-ec2m to disable characteristic 2 code.
225 *) New build option "fipscanisteronly". This only builds fipscanister.o
226 and (currently) associated fips utilities. Uses the file Makefile.fips
227 instead of Makefile.org as the prototype.
230 *) Add some FIPS mode restrictions to GCM. Add internal IV generator.
231 Update fips_gcmtest to use IV generator.
234 *) Initial, experimental EVP support for AES-GCM. AAD can be input by
235 setting output buffer to NULL. The *Final function must be
236 called although it will not retrieve any additional data. The tag
237 can be set or retrieved with a ctrl. The IV length is by default 12
238 bytes (96 bits) but can be set to an alternative value. If the IV
239 length exceeds the maximum IV length (currently 16 bytes) it cannot be
243 *) New flag in ciphers: EVP_CIPH_FLAG_CUSTOM_CIPHER. This means the
244 underlying do_cipher function handles all cipher semantics itself
245 including padding and finalisation. This is useful if (for example)
246 an ENGINE cipher handles block padding itself. The behaviour of
247 do_cipher is subtly changed if this flag is set: the return value
248 is the number of characters written to the output buffer (zero is
249 no longer an error code) or a negative error code. Also if the
250 input buffer is NULL and length 0 finalisation should be performed.
253 *) If a candidate issuer certificate is already part of the constructed
254 path ignore it: new debug notification X509_V_ERR_PATH_LOOP for this case.
257 *) Improve forward-security support: add functions
259 void SSL_CTX_set_not_resumable_session_callback(SSL_CTX *ctx, int (*cb)(SSL *ssl, int is_forward_secure))
260 void SSL_set_not_resumable_session_callback(SSL *ssl, int (*cb)(SSL *ssl, int is_forward_secure))
262 for use by SSL/TLS servers; the callback function will be called whenever a
263 new session is created, and gets to decide whether the session may be
264 cached to make it resumable (return 0) or not (return 1). (As by the
265 SSL/TLS protocol specifications, the session_id sent by the server will be
266 empty to indicate that the session is not resumable; also, the server will
267 not generate RFC 4507 (RFC 5077) session tickets.)
269 A simple reasonable callback implementation is to return is_forward_secure.
270 This parameter will be set to 1 or 0 depending on the ciphersuite selected
271 by the SSL/TLS server library, indicating whether it can provide forward
273 [Emilia Käsper <emilia.kasper@esat.kuleuven.be> (Google)]
275 *) New -verify_name option in command line utilities to set verification
279 *) Initial CMAC implementation. WARNING: EXPERIMENTAL, API MAY CHANGE.
280 Add CMAC pkey methods.
283 *) Experimental renegotiation in s_server -www mode. If the client
284 browses /reneg connection is renegotiated. If /renegcert it is
285 renegotiated requesting a certificate.
288 *) Add an "external" session cache for debugging purposes to s_server. This
289 should help trace issues which normally are only apparent in deployed
290 multi-process servers.
293 *) Extensive audit of libcrypto with DEBUG_UNUSED. Fix many cases where
294 return value is ignored. NB. The functions RAND_add(), RAND_seed(),
295 BIO_set_cipher() and some obscure PEM functions were changed so they
296 can now return an error. The RAND changes required a change to the
297 RAND_METHOD structure.
300 *) New macro __owur for "OpenSSL Warn Unused Result". This makes use of
301 a gcc attribute to warn if the result of a function is ignored. This
302 is enable if DEBUG_UNUSED is set. Add to several functions in evp.h
303 whose return value is often ignored.
306 Changes between 1.0.1j and 1.0.2 [xx XXX xxxx]
308 *) Accelerated NIST P-256 elliptic curve implementation for x86_64
309 (other platforms pending).
310 [Shay Gueron & Vlad Krasnov (Intel Corp), Andy Polyakov]
312 *) Add support for the SignedCertificateTimestampList certificate and
313 OCSP response extensions from RFC6962.
316 *) Fix ec_GFp_simple_points_make_affine (thus, EC_POINTs_mul etc.)
317 for corner cases. (Certain input points at infinity could lead to
318 bogus results, with non-infinity inputs mapped to infinity too.)
321 *) Initial support for PowerISA 2.0.7, first implemented in POWER8.
322 This covers AES, SHA256/512 and GHASH. "Initial" means that most
323 common cases are optimized and there still is room for further
324 improvements. Vector Permutation AES for Altivec is also added.
327 *) Add support for little-endian ppc64 Linux target.
328 [Marcelo Cerri (IBM)]
330 *) Initial support for AMRv8 ISA crypto extensions. This covers AES,
331 SHA1, SHA256 and GHASH. "Initial" means that most common cases
332 are optimized and there still is room for further improvements.
333 Both 32- and 64-bit modes are supported.
334 [Andy Polyakov, Ard Biesheuvel (Linaro)]
336 *) Improved ARMv7 NEON support.
339 *) Support for SPARC Architecture 2011 crypto extensions, first
340 implemented in SPARC T4. This covers AES, DES, Camellia, SHA1,
341 SHA256/512, MD5, GHASH and modular exponentiation.
342 [Andy Polyakov, David Miller]
344 *) Accelerated modular exponentiation for Intel processors, a.k.a.
346 [Shay Gueron & Vlad Krasnov (Intel Corp)]
348 *) Support for new and upcoming Intel processors, including AVX2,
349 BMI and SHA ISA extensions. This includes additional "stitched"
350 implementations, AESNI-SHA256 and GCM, and multi-buffer support
353 This work was sponsored by Intel Corp.
356 *) Support for DTLS 1.2. This adds two sets of DTLS methods: DTLS_*_method()
357 supports both DTLS 1.2 and 1.0 and should use whatever version the peer
358 supports and DTLSv1_2_*_method() which supports DTLS 1.2 only.
361 *) Use algorithm specific chains in SSL_CTX_use_certificate_chain_file():
362 this fixes a limiation in previous versions of OpenSSL.
365 *) Extended RSA OAEP support via EVP_PKEY API. Options to specify digest,
366 MGF1 digest and OAEP label.
369 *) Add EVP support for key wrapping algorithms, to avoid problems with
370 existing code the flag EVP_CIPHER_CTX_WRAP_ALLOW has to be set in
371 the EVP_CIPHER_CTX or an error is returned. Add AES and DES3 wrap
372 algorithms and include tests cases.
375 *) Add functions to allocate and set the fields of an ECDSA_METHOD
377 [Douglas E. Engert, Steve Henson]
379 *) New functions OPENSSL_gmtime_diff and ASN1_TIME_diff to find the
380 difference in days and seconds between two tm or ASN1_TIME structures.
383 *) Add -rev test option to s_server to just reverse order of characters
384 received by client and send back to server. Also prints an abbreviated
385 summary of the connection parameters.
388 *) New option -brief for s_client and s_server to print out a brief summary
389 of connection parameters.
392 *) Add callbacks for arbitrary TLS extensions.
393 [Trevor Perrin <trevp@trevp.net> and Ben Laurie]
395 *) New option -crl_download in several openssl utilities to download CRLs
396 from CRLDP extension in certificates.
399 *) New options -CRL and -CRLform for s_client and s_server for CRLs.
402 *) New function X509_CRL_diff to generate a delta CRL from the difference
403 of two full CRLs. Add support to "crl" utility.
406 *) New functions to set lookup_crls function and to retrieve
407 X509_STORE from X509_STORE_CTX.
410 *) Print out deprecated issuer and subject unique ID fields in
414 *) Extend OCSP I/O functions so they can be used for simple general purpose
415 HTTP as well as OCSP. New wrapper function which can be used to download
416 CRLs using the OCSP API.
419 *) Delegate command line handling in s_client/s_server to SSL_CONF APIs.
422 *) SSL_CONF* functions. These provide a common framework for application
423 configuration using configuration files or command lines.
426 *) SSL/TLS tracing code. This parses out SSL/TLS records using the
427 message callback and prints the results. Needs compile time option
428 "enable-ssl-trace". New options to s_client and s_server to enable
432 *) New ctrl and macro to retrieve supported points extensions.
433 Print out extension in s_server and s_client.
436 *) New functions to retrieve certificate signature and signature
440 *) Add functions to retrieve and manipulate the raw cipherlist sent by a
444 *) New Suite B modes for TLS code. These use and enforce the requirements
445 of RFC6460: restrict ciphersuites, only permit Suite B algorithms and
446 only use Suite B curves. The Suite B modes can be set by using the
447 strings "SUITEB128", "SUITEB192" or "SUITEB128ONLY" for the cipherstring.
450 *) New chain verification flags for Suite B levels of security. Check
451 algorithms are acceptable when flags are set in X509_verify_cert.
454 *) Make tls1_check_chain return a set of flags indicating checks passed
455 by a certificate chain. Add additional tests to handle client
456 certificates: checks for matching certificate type and issuer name
460 *) If an attempt is made to use a signature algorithm not in the peer
461 preference list abort the handshake. If client has no suitable
462 signature algorithms in response to a certificate request do not
466 *) If server EC tmp key is not in client preference list abort handshake.
469 *) Add support for certificate stores in CERT structure. This makes it
470 possible to have different stores per SSL structure or one store in
471 the parent SSL_CTX. Include distint stores for certificate chain
472 verification and chain building. New ctrl SSL_CTRL_BUILD_CERT_CHAIN
473 to build and store a certificate chain in CERT structure: returing
474 an error if the chain cannot be built: this will allow applications
475 to test if a chain is correctly configured.
477 Note: if the CERT based stores are not set then the parent SSL_CTX
478 store is used to retain compatibility with existing behaviour.
482 *) New function ssl_set_client_disabled to set a ciphersuite disabled
483 mask based on the current session, check mask when sending client
484 hello and checking the requested ciphersuite.
487 *) New ctrls to retrieve and set certificate types in a certificate
488 request message. Print out received values in s_client. If certificate
489 types is not set with custom values set sensible values based on
490 supported signature algorithms.
493 *) Support for distinct client and server supported signature algorithms.
496 *) Add certificate callback. If set this is called whenever a certificate
497 is required by client or server. An application can decide which
498 certificate chain to present based on arbitrary criteria: for example
499 supported signature algorithms. Add very simple example to s_server.
500 This fixes many of the problems and restrictions of the existing client
501 certificate callback: for example you can now clear an existing
502 certificate and specify the whole chain.
505 *) Add new "valid_flags" field to CERT_PKEY structure which determines what
506 the certificate can be used for (if anything). Set valid_flags field
507 in new tls1_check_chain function. Simplify ssl_set_cert_masks which used
508 to have similar checks in it.
510 Add new "cert_flags" field to CERT structure and include a "strict mode".
511 This enforces some TLS certificate requirements (such as only permitting
512 certificate signature algorithms contained in the supported algorithms
513 extension) which some implementations ignore: this option should be used
514 with caution as it could cause interoperability issues.
517 *) Update and tidy signature algorithm extension processing. Work out
518 shared signature algorithms based on preferences and peer algorithms
519 and print them out in s_client and s_server. Abort handshake if no
520 shared signature algorithms.
523 *) Add new functions to allow customised supported signature algorithms
524 for SSL and SSL_CTX structures. Add options to s_client and s_server
528 *) New function SSL_certs_clear() to delete all references to certificates
529 from an SSL structure. Before this once a certificate had been added
530 it couldn't be removed.
533 *) Integrate hostname, email address and IP address checking with certificate
534 verification. New verify options supporting checking in opensl utility.
537 *) Fixes and wildcard matching support to hostname and email checking
538 functions. Add manual page.
539 [Florian Weimer (Red Hat Product Security Team)]
541 *) New functions to check a hostname email or IP address against a
542 certificate. Add options x509 utility to print results of checks against
546 *) Fix OCSP checking.
547 [Rob Stradling <rob.stradling@comodo.com> and Ben Laurie]
549 *) Initial experimental support for explicitly trusted non-root CAs.
550 OpenSSL still tries to build a complete chain to a root but if an
551 intermediate CA has a trust setting included that is used. The first
552 setting is used: whether to trust (e.g., -addtrust option to the x509
556 *) Add -trusted_first option which attempts to find certificates in the
557 trusted store even if an untrusted chain is also supplied.
560 *) MIPS assembly pack updates: support for MIPS32r2 and SmartMIPS ASE,
561 platform support for Linux and Android.
564 *) Support for linux-x32, ILP32 environment in x86_64 framework.
567 *) Experimental multi-implementation support for FIPS capable OpenSSL.
568 When in FIPS mode the approved implementations are used as normal,
569 when not in FIPS mode the internal unapproved versions are used instead.
570 This means that the FIPS capable OpenSSL isn't forced to use the
571 (often lower performance) FIPS implementations outside FIPS mode.
574 *) Transparently support X9.42 DH parameters when calling
575 PEM_read_bio_DHparameters. This means existing applications can handle
576 the new parameter format automatically.
579 *) Initial experimental support for X9.42 DH parameter format: mainly
580 to support use of 'q' parameter for RFC5114 parameters.
583 *) Add DH parameters from RFC5114 including test data to dhtest.
586 *) Support for automatic EC temporary key parameter selection. If enabled
587 the most preferred EC parameters are automatically used instead of
588 hardcoded fixed parameters. Now a server just has to call:
589 SSL_CTX_set_ecdh_auto(ctx, 1) and the server will automatically
590 support ECDH and use the most appropriate parameters.
593 *) Enhance and tidy EC curve and point format TLS extension code. Use
594 static structures instead of allocation if default values are used.
595 New ctrls to set curves we wish to support and to retrieve shared curves.
596 Print out shared curves in s_server. New options to s_server and s_client
597 to set list of supported curves.
600 *) New ctrls to retrieve supported signature algorithms and
601 supported curve values as an array of NIDs. Extend openssl utility
602 to print out received values.
605 *) Add new APIs EC_curve_nist2nid and EC_curve_nid2nist which convert
606 between NIDs and the more common NIST names such as "P-256". Enhance
607 ecparam utility and ECC method to recognise the NIST names for curves.
610 *) Enhance SSL/TLS certificate chain handling to support different
611 chains for each certificate instead of one chain in the parent SSL_CTX.
614 *) Support for fixed DH ciphersuite client authentication: where both
615 server and client use DH certificates with common parameters.
618 *) Support for fixed DH ciphersuites: those requiring DH server
622 *) New function i2d_re_X509_tbs for re-encoding the TBS portion of
624 Note: Related 1.0.2-beta specific macros X509_get_cert_info,
625 X509_CINF_set_modified, X509_CINF_get_issuer, X509_CINF_get_extensions and
626 X509_CINF_get_signature were reverted post internal team review.
628 Changes between 1.0.1i and 1.0.1j [15 Oct 2014]
632 A flaw in the DTLS SRTP extension parsing code allows an attacker, who
633 sends a carefully crafted handshake message, to cause OpenSSL to fail
634 to free up to 64k of memory causing a memory leak. This could be
635 exploited in a Denial Of Service attack. This issue affects OpenSSL
636 1.0.1 server implementations for both SSL/TLS and DTLS regardless of
637 whether SRTP is used or configured. Implementations of OpenSSL that
638 have been compiled with OPENSSL_NO_SRTP defined are not affected.
640 The fix was developed by the OpenSSL team.
644 *) Session Ticket Memory Leak.
646 When an OpenSSL SSL/TLS/DTLS server receives a session ticket the
647 integrity of that ticket is first verified. In the event of a session
648 ticket integrity check failing, OpenSSL will fail to free memory
649 causing a memory leak. By sending a large number of invalid session
650 tickets an attacker could exploit this issue in a Denial Of Service
655 *) Build option no-ssl3 is incomplete.
657 When OpenSSL is configured with "no-ssl3" as a build option, servers
658 could accept and complete a SSL 3.0 handshake, and clients could be
659 configured to send them.
661 [Akamai and the OpenSSL team]
663 *) Add support for TLS_FALLBACK_SCSV.
664 Client applications doing fallback retries should call
665 SSL_set_mode(s, SSL_MODE_SEND_FALLBACK_SCSV).
667 [Adam Langley, Bodo Moeller]
669 *) Add additional DigestInfo checks.
671 Reencode DigestInto in DER and check against the original when
672 verifying RSA signature: this will reject any improperly encoded
673 DigestInfo structures.
675 Note: this is a precautionary measure and no attacks are currently known.
679 Changes between 1.0.1g and 1.0.1h [5 Jun 2014]
681 *) Fix for SSL/TLS MITM flaw. An attacker using a carefully crafted
682 handshake can force the use of weak keying material in OpenSSL
683 SSL/TLS clients and servers.
685 Thanks to KIKUCHI Masashi (Lepidum Co. Ltd.) for discovering and
686 researching this issue. (CVE-2014-0224)
687 [KIKUCHI Masashi, Steve Henson]
689 *) Fix DTLS recursion flaw. By sending an invalid DTLS handshake to an
690 OpenSSL DTLS client the code can be made to recurse eventually crashing
693 Thanks to Imre Rad (Search-Lab Ltd.) for discovering this issue.
695 [Imre Rad, Steve Henson]
697 *) Fix DTLS invalid fragment vulnerability. A buffer overrun attack can
698 be triggered by sending invalid DTLS fragments to an OpenSSL DTLS
699 client or server. This is potentially exploitable to run arbitrary
700 code on a vulnerable client or server.
702 Thanks to Jüri Aedla for reporting this issue. (CVE-2014-0195)
703 [Jüri Aedla, Steve Henson]
705 *) Fix bug in TLS code where clients enable anonymous ECDH ciphersuites
706 are subject to a denial of service attack.
708 Thanks to Felix Gröbert and Ivan Fratric at Google for discovering
709 this issue. (CVE-2014-3470)
710 [Felix Gröbert, Ivan Fratric, Steve Henson]
712 *) Harmonize version and its documentation. -f flag is used to display
714 [mancha <mancha1@zoho.com>]
716 *) Fix eckey_priv_encode so it immediately returns an error upon a failure
718 [mancha <mancha1@zoho.com>]
720 *) Fix some double frees. These are not thought to be exploitable.
721 [mancha <mancha1@zoho.com>]
723 Changes between 1.0.1f and 1.0.1g [7 Apr 2014]
725 *) A missing bounds check in the handling of the TLS heartbeat extension
726 can be used to reveal up to 64k of memory to a connected client or
729 Thanks for Neel Mehta of Google Security for discovering this bug and to
730 Adam Langley <agl@chromium.org> and Bodo Moeller <bmoeller@acm.org> for
731 preparing the fix (CVE-2014-0160)
732 [Adam Langley, Bodo Moeller]
734 *) Fix for the attack described in the paper "Recovering OpenSSL
735 ECDSA Nonces Using the FLUSH+RELOAD Cache Side-channel Attack"
736 by Yuval Yarom and Naomi Benger. Details can be obtained from:
737 http://eprint.iacr.org/2014/140
739 Thanks to Yuval Yarom and Naomi Benger for discovering this
740 flaw and to Yuval Yarom for supplying a fix (CVE-2014-0076)
741 [Yuval Yarom and Naomi Benger]
743 *) TLS pad extension: draft-agl-tls-padding-03
745 Workaround for the "TLS hang bug" (see FAQ and PR#2771): if the
746 TLS client Hello record length value would otherwise be > 255 and
747 less that 512 pad with a dummy extension containing zeroes so it
748 is at least 512 bytes long.
750 [Adam Langley, Steve Henson]
752 Changes between 1.0.1e and 1.0.1f [6 Jan 2014]
754 *) Fix for TLS record tampering bug. A carefully crafted invalid
755 handshake could crash OpenSSL with a NULL pointer exception.
756 Thanks to Anton Johansson for reporting this issues.
759 *) Keep original DTLS digest and encryption contexts in retransmission
760 structures so we can use the previous session parameters if they need
761 to be resent. (CVE-2013-6450)
764 *) Add option SSL_OP_SAFARI_ECDHE_ECDSA_BUG (part of SSL_OP_ALL) which
765 avoids preferring ECDHE-ECDSA ciphers when the client appears to be
766 Safari on OS X. Safari on OS X 10.8..10.8.3 advertises support for
767 several ECDHE-ECDSA ciphers, but fails to negotiate them. The bug
768 is fixed in OS X 10.8.4, but Apple have ruled out both hot fixing
769 10.8..10.8.3 and forcing users to upgrade to 10.8.4 or newer.
770 [Rob Stradling, Adam Langley]
772 Changes between 1.0.1d and 1.0.1e [11 Feb 2013]
774 *) Correct fix for CVE-2013-0169. The original didn't work on AES-NI
775 supporting platforms or when small records were transferred.
776 [Andy Polyakov, Steve Henson]
778 Changes between 1.0.1c and 1.0.1d [5 Feb 2013]
780 *) Make the decoding of SSLv3, TLS and DTLS CBC records constant time.
782 This addresses the flaw in CBC record processing discovered by
783 Nadhem Alfardan and Kenny Paterson. Details of this attack can be found
784 at: http://www.isg.rhul.ac.uk/tls/
786 Thanks go to Nadhem Alfardan and Kenny Paterson of the Information
787 Security Group at Royal Holloway, University of London
788 (www.isg.rhul.ac.uk) for discovering this flaw and Adam Langley and
789 Emilia Käsper for the initial patch.
791 [Emilia Käsper, Adam Langley, Ben Laurie, Andy Polyakov, Steve Henson]
793 *) Fix flaw in AESNI handling of TLS 1.2 and 1.1 records for CBC mode
794 ciphersuites which can be exploited in a denial of service attack.
795 Thanks go to and to Adam Langley <agl@chromium.org> for discovering
796 and detecting this bug and to Wolfgang Ettlinger
797 <wolfgang.ettlinger@gmail.com> for independently discovering this issue.
801 *) Return an error when checking OCSP signatures when key is NULL.
802 This fixes a DoS attack. (CVE-2013-0166)
805 *) Make openssl verify return errors.
806 [Chris Palmer <palmer@google.com> and Ben Laurie]
808 *) Call OCSP Stapling callback after ciphersuite has been chosen, so
809 the right response is stapled. Also change SSL_get_certificate()
810 so it returns the certificate actually sent.
811 See http://rt.openssl.org/Ticket/Display.html?id=2836.
812 [Rob Stradling <rob.stradling@comodo.com>]
814 *) Fix possible deadlock when decoding public keys.
817 *) Don't use TLS 1.0 record version number in initial client hello
821 Changes between 1.0.1b and 1.0.1c [10 May 2012]
823 *) Sanity check record length before skipping explicit IV in TLS
824 1.2, 1.1 and DTLS to fix DoS attack.
826 Thanks to Codenomicon for discovering this issue using Fuzz-o-Matic
827 fuzzing as a service testing platform.
831 *) Initialise tkeylen properly when encrypting CMS messages.
832 Thanks to Solar Designer of Openwall for reporting this issue.
835 *) In FIPS mode don't try to use composite ciphers as they are not
839 Changes between 1.0.1a and 1.0.1b [26 Apr 2012]
841 *) OpenSSL 1.0.0 sets SSL_OP_ALL to 0x80000FFFL and OpenSSL 1.0.1 and
842 1.0.1a set SSL_OP_NO_TLSv1_1 to 0x00000400L which would unfortunately
843 mean any application compiled against OpenSSL 1.0.0 headers setting
844 SSL_OP_ALL would also set SSL_OP_NO_TLSv1_1, unintentionally disablng
845 TLS 1.1 also. Fix this by changing the value of SSL_OP_NO_TLSv1_1 to
846 0x10000000L Any application which was previously compiled against
847 OpenSSL 1.0.1 or 1.0.1a headers and which cares about SSL_OP_NO_TLSv1_1
848 will need to be recompiled as a result. Letting be results in
849 inability to disable specifically TLS 1.1 and in client context,
850 in unlike event, limit maximum offered version to TLS 1.0 [see below].
853 *) In order to ensure interoperabilty SSL_OP_NO_protocolX does not
854 disable just protocol X, but all protocols above X *if* there are
855 protocols *below* X still enabled. In more practical terms it means
856 that if application wants to disable TLS1.0 in favor of TLS1.1 and
857 above, it's not sufficient to pass SSL_OP_NO_TLSv1, one has to pass
858 SSL_OP_NO_TLSv1|SSL_OP_NO_SSLv3|SSL_OP_NO_SSLv2. This applies to
862 Changes between 1.0.1 and 1.0.1a [19 Apr 2012]
864 *) Check for potentially exploitable overflows in asn1_d2i_read_bio
865 BUF_mem_grow and BUF_mem_grow_clean. Refuse attempts to shrink buffer
866 in CRYPTO_realloc_clean.
868 Thanks to Tavis Ormandy, Google Security Team, for discovering this
869 issue and to Adam Langley <agl@chromium.org> for fixing it.
871 [Adam Langley (Google), Tavis Ormandy, Google Security Team]
873 *) Don't allow TLS 1.2 SHA-256 ciphersuites in TLS 1.0, 1.1 connections.
876 *) Workarounds for some broken servers that "hang" if a client hello
877 record length exceeds 255 bytes.
879 1. Do not use record version number > TLS 1.0 in initial client
880 hello: some (but not all) hanging servers will now work.
881 2. If we set OPENSSL_MAX_TLS1_2_CIPHER_LENGTH this will truncate
882 the number of ciphers sent in the client hello. This should be
883 set to an even number, such as 50, for example by passing:
884 -DOPENSSL_MAX_TLS1_2_CIPHER_LENGTH=50 to config or Configure.
885 Most broken servers should now work.
886 3. If all else fails setting OPENSSL_NO_TLS1_2_CLIENT will disable
887 TLS 1.2 client support entirely.
890 *) Fix SEGV in Vector Permutation AES module observed in OpenSSH.
893 Changes between 1.0.0h and 1.0.1 [14 Mar 2012]
895 *) Add compatibility with old MDC2 signatures which use an ASN1 OCTET
896 STRING form instead of a DigestInfo.
899 *) The format used for MDC2 RSA signatures is inconsistent between EVP
900 and the RSA_sign/RSA_verify functions. This was made more apparent when
901 OpenSSL used RSA_sign/RSA_verify for some RSA signatures in particular
902 those which went through EVP_PKEY_METHOD in 1.0.0 and later. Detect
903 the correct format in RSA_verify so both forms transparently work.
906 *) Some servers which support TLS 1.0 can choke if we initially indicate
907 support for TLS 1.2 and later renegotiate using TLS 1.0 in the RSA
908 encrypted premaster secret. As a workaround use the maximum permitted
909 client version in client hello, this should keep such servers happy
910 and still work with previous versions of OpenSSL.
913 *) Add support for TLS/DTLS heartbeats.
914 [Robin Seggelmann <seggelmann@fh-muenster.de>]
916 *) Add support for SCTP.
917 [Robin Seggelmann <seggelmann@fh-muenster.de>]
919 *) Improved PRNG seeding for VOS.
920 [Paul Green <Paul.Green@stratus.com>]
922 *) Extensive assembler packs updates, most notably:
924 - x86[_64]: AES-NI, PCLMULQDQ, RDRAND support;
925 - x86[_64]: SSSE3 support (SHA1, vector-permutation AES);
926 - x86_64: bit-sliced AES implementation;
927 - ARM: NEON support, contemporary platforms optimizations;
928 - s390x: z196 support;
929 - *: GHASH and GF(2^m) multiplication implementations;
933 *) Make TLS-SRP code conformant with RFC 5054 API cleanup
934 (removal of unnecessary code)
935 [Peter Sylvester <peter.sylvester@edelweb.fr>]
937 *) Add TLS key material exporter from RFC 5705.
940 *) Add DTLS-SRTP negotiation from RFC 5764.
943 *) Add Next Protocol Negotiation,
944 http://tools.ietf.org/html/draft-agl-tls-nextprotoneg-00. Can be
945 disabled with a no-npn flag to config or Configure. Code donated
947 [Adam Langley <agl@google.com> and Ben Laurie]
949 *) Add optional 64-bit optimized implementations of elliptic curves NIST-P224,
950 NIST-P256, NIST-P521, with constant-time single point multiplication on
951 typical inputs. Compiler support for the nonstandard type __uint128_t is
952 required to use this (present in gcc 4.4 and later, for 64-bit builds).
953 Code made available under Apache License version 2.0.
955 Specify "enable-ec_nistp_64_gcc_128" on the Configure (or config) command
956 line to include this in your build of OpenSSL, and run "make depend" (or
957 "make update"). This enables the following EC_METHODs:
959 EC_GFp_nistp224_method()
960 EC_GFp_nistp256_method()
961 EC_GFp_nistp521_method()
963 EC_GROUP_new_by_curve_name() will automatically use these (while
964 EC_GROUP_new_curve_GFp() currently prefers the more flexible
966 [Emilia Käsper, Adam Langley, Bodo Moeller (Google)]
968 *) Use type ossl_ssize_t instad of ssize_t which isn't available on
969 all platforms. Move ssize_t definition from e_os.h to the public
970 header file e_os2.h as it now appears in public header file cms.h
973 *) New -sigopt option to the ca, req and x509 utilities. Additional
974 signature parameters can be passed using this option and in
978 *) Add RSA PSS signing function. This will generate and set the
979 appropriate AlgorithmIdentifiers for PSS based on those in the
980 corresponding EVP_MD_CTX structure. No application support yet.
983 *) Support for companion algorithm specific ASN1 signing routines.
984 New function ASN1_item_sign_ctx() signs a pre-initialised
985 EVP_MD_CTX structure and sets AlgorithmIdentifiers based on
986 the appropriate parameters.
989 *) Add new algorithm specific ASN1 verification initialisation function
990 to EVP_PKEY_ASN1_METHOD: this is not in EVP_PKEY_METHOD since the ASN1
991 handling will be the same no matter what EVP_PKEY_METHOD is used.
992 Add a PSS handler to support verification of PSS signatures: checked
993 against a number of sample certificates.
996 *) Add signature printing for PSS. Add PSS OIDs.
997 [Steve Henson, Martin Kaiser <lists@kaiser.cx>]
999 *) Add algorithm specific signature printing. An individual ASN1 method
1000 can now print out signatures instead of the standard hex dump.
1002 More complex signatures (e.g. PSS) can print out more meaningful
1003 information. Include DSA version that prints out the signature
1007 *) Password based recipient info support for CMS library: implementing
1011 *) Split password based encryption into PBES2 and PBKDF2 functions. This
1012 neatly separates the code into cipher and PBE sections and is required
1013 for some algorithms that split PBES2 into separate pieces (such as
1014 password based CMS).
1017 *) Session-handling fixes:
1018 - Fix handling of connections that are resuming with a session ID,
1019 but also support Session Tickets.
1020 - Fix a bug that suppressed issuing of a new ticket if the client
1021 presented a ticket with an expired session.
1022 - Try to set the ticket lifetime hint to something reasonable.
1023 - Make tickets shorter by excluding irrelevant information.
1024 - On the client side, don't ignore renewed tickets.
1025 [Adam Langley, Bodo Moeller (Google)]
1027 *) Fix PSK session representation.
1030 *) Add RC4-MD5 and AESNI-SHA1 "stitched" implementations.
1032 This work was sponsored by Intel.
1035 *) Add GCM support to TLS library. Some custom code is needed to split
1036 the IV between the fixed (from PRF) and explicit (from TLS record)
1037 portions. This adds all GCM ciphersuites supported by RFC5288 and
1038 RFC5289. Generalise some AES* cipherstrings to include GCM and
1039 add a special AESGCM string for GCM only.
1042 *) Expand range of ctrls for AES GCM. Permit setting invocation
1043 field on decrypt and retrieval of invocation field only on encrypt.
1046 *) Add HMAC ECC ciphersuites from RFC5289. Include SHA384 PRF support.
1047 As required by RFC5289 these ciphersuites cannot be used if for
1048 versions of TLS earlier than 1.2.
1051 *) For FIPS capable OpenSSL interpret a NULL default public key method
1052 as unset and return the appropriate default but do *not* set the default.
1053 This means we can return the appropriate method in applications that
1054 switch between FIPS and non-FIPS modes.
1057 *) Redirect HMAC and CMAC operations to FIPS module in FIPS mode. If an
1058 ENGINE is used then we cannot handle that in the FIPS module so we
1059 keep original code iff non-FIPS operations are allowed.
1062 *) Add -attime option to openssl utilities.
1063 [Peter Eckersley <pde@eff.org>, Ben Laurie and Steve Henson]
1065 *) Redirect DSA and DH operations to FIPS module in FIPS mode.
1068 *) Redirect ECDSA and ECDH operations to FIPS module in FIPS mode. Also use
1069 FIPS EC methods unconditionally for now.
1072 *) New build option no-ec2m to disable characteristic 2 code.
1075 *) Backport libcrypto audit of return value checking from 1.1.0-dev; not
1076 all cases can be covered as some introduce binary incompatibilities.
1079 *) Redirect RSA operations to FIPS module including keygen,
1080 encrypt, decrypt, sign and verify. Block use of non FIPS RSA methods.
1083 *) Add similar low level API blocking to ciphers.
1086 *) Low level digest APIs are not approved in FIPS mode: any attempt
1087 to use these will cause a fatal error. Applications that *really* want
1088 to use them can use the private_* version instead.
1091 *) Redirect cipher operations to FIPS module for FIPS builds.
1094 *) Redirect digest operations to FIPS module for FIPS builds.
1097 *) Update build system to add "fips" flag which will link in fipscanister.o
1098 for static and shared library builds embedding a signature if needed.
1101 *) Output TLS supported curves in preference order instead of numerical
1102 order. This is currently hardcoded for the highest order curves first.
1103 This should be configurable so applications can judge speed vs strength.
1106 *) Add TLS v1.2 server support for client authentication.
1109 *) Add support for FIPS mode in ssl library: disable SSLv3, non-FIPS ciphers
1113 *) Functions FIPS_mode_set() and FIPS_mode() which call the underlying
1114 FIPS modules versions.
1117 *) Add TLS v1.2 client side support for client authentication. Keep cache
1118 of handshake records longer as we don't know the hash algorithm to use
1119 until after the certificate request message is received.
1122 *) Initial TLS v1.2 client support. Add a default signature algorithms
1123 extension including all the algorithms we support. Parse new signature
1124 format in client key exchange. Relax some ECC signing restrictions for
1125 TLS v1.2 as indicated in RFC5246.
1128 *) Add server support for TLS v1.2 signature algorithms extension. Switch
1129 to new signature format when needed using client digest preference.
1130 All server ciphersuites should now work correctly in TLS v1.2. No client
1131 support yet and no support for client certificates.
1134 *) Initial TLS v1.2 support. Add new SHA256 digest to ssl code, switch
1135 to SHA256 for PRF when using TLS v1.2 and later. Add new SHA256 based
1136 ciphersuites. At present only RSA key exchange ciphersuites work with
1137 TLS v1.2. Add new option for TLS v1.2 replacing the old and obsolete
1138 SSL_OP_PKCS1_CHECK flags with SSL_OP_NO_TLSv1_2. New TLSv1.2 methods
1139 and version checking.
1142 *) New option OPENSSL_NO_SSL_INTERN. If an application can be compiled
1143 with this defined it will not be affected by any changes to ssl internal
1144 structures. Add several utility functions to allow openssl application
1145 to work with OPENSSL_NO_SSL_INTERN defined.
1149 [Tom Wu <tjw@cs.stanford.edu> and Ben Laurie]
1151 *) Add functions to copy EVP_PKEY_METHOD and retrieve flags and id.
1154 *) Permit abbreviated handshakes when renegotiating using the function
1155 SSL_renegotiate_abbreviated().
1156 [Robin Seggelmann <seggelmann@fh-muenster.de>]
1158 *) Add call to ENGINE_register_all_complete() to
1159 ENGINE_load_builtin_engines(), so some implementations get used
1160 automatically instead of needing explicit application support.
1163 *) Add support for TLS key exporter as described in RFC5705.
1164 [Robin Seggelmann <seggelmann@fh-muenster.de>, Steve Henson]
1166 *) Initial TLSv1.1 support. Since TLSv1.1 is very similar to TLS v1.0 only
1167 a few changes are required:
1169 Add SSL_OP_NO_TLSv1_1 flag.
1170 Add TLSv1_1 methods.
1171 Update version checking logic to handle version 1.1.
1172 Add explicit IV handling (ported from DTLS code).
1173 Add command line options to s_client/s_server.
1176 Changes between 1.0.0j and 1.0.0k [5 Feb 2013]
1178 *) Make the decoding of SSLv3, TLS and DTLS CBC records constant time.
1180 This addresses the flaw in CBC record processing discovered by
1181 Nadhem Alfardan and Kenny Paterson. Details of this attack can be found
1182 at: http://www.isg.rhul.ac.uk/tls/
1184 Thanks go to Nadhem Alfardan and Kenny Paterson of the Information
1185 Security Group at Royal Holloway, University of London
1186 (www.isg.rhul.ac.uk) for discovering this flaw and Adam Langley and
1187 Emilia Käsper for the initial patch.
1189 [Emilia Käsper, Adam Langley, Ben Laurie, Andy Polyakov, Steve Henson]
1191 *) Return an error when checking OCSP signatures when key is NULL.
1192 This fixes a DoS attack. (CVE-2013-0166)
1195 *) Call OCSP Stapling callback after ciphersuite has been chosen, so
1196 the right response is stapled. Also change SSL_get_certificate()
1197 so it returns the certificate actually sent.
1198 See http://rt.openssl.org/Ticket/Display.html?id=2836.
1199 (This is a backport)
1200 [Rob Stradling <rob.stradling@comodo.com>]
1202 *) Fix possible deadlock when decoding public keys.
1205 Changes between 1.0.0i and 1.0.0j [10 May 2012]
1207 [NB: OpenSSL 1.0.0i and later 1.0.0 patch levels were released after
1210 *) Sanity check record length before skipping explicit IV in DTLS
1213 Thanks to Codenomicon for discovering this issue using Fuzz-o-Matic
1214 fuzzing as a service testing platform.
1218 *) Initialise tkeylen properly when encrypting CMS messages.
1219 Thanks to Solar Designer of Openwall for reporting this issue.
1222 Changes between 1.0.0h and 1.0.0i [19 Apr 2012]
1224 *) Check for potentially exploitable overflows in asn1_d2i_read_bio
1225 BUF_mem_grow and BUF_mem_grow_clean. Refuse attempts to shrink buffer
1226 in CRYPTO_realloc_clean.
1228 Thanks to Tavis Ormandy, Google Security Team, for discovering this
1229 issue and to Adam Langley <agl@chromium.org> for fixing it.
1231 [Adam Langley (Google), Tavis Ormandy, Google Security Team]
1233 Changes between 1.0.0g and 1.0.0h [12 Mar 2012]
1235 *) Fix MMA (Bleichenbacher's attack on PKCS #1 v1.5 RSA padding) weakness
1236 in CMS and PKCS7 code. When RSA decryption fails use a random key for
1237 content decryption and always return the same error. Note: this attack
1238 needs on average 2^20 messages so it only affects automated senders. The
1239 old behaviour can be reenabled in the CMS code by setting the
1240 CMS_DEBUG_DECRYPT flag: this is useful for debugging and testing where
1241 an MMA defence is not necessary.
1242 Thanks to Ivan Nestlerode <inestlerode@us.ibm.com> for discovering
1243 this issue. (CVE-2012-0884)
1246 *) Fix CVE-2011-4619: make sure we really are receiving a
1247 client hello before rejecting multiple SGC restarts. Thanks to
1248 Ivan Nestlerode <inestlerode@us.ibm.com> for discovering this bug.
1251 Changes between 1.0.0f and 1.0.0g [18 Jan 2012]
1253 *) Fix for DTLS DoS issue introduced by fix for CVE-2011-4109.
1254 Thanks to Antonio Martin, Enterprise Secure Access Research and
1255 Development, Cisco Systems, Inc. for discovering this bug and
1256 preparing a fix. (CVE-2012-0050)
1259 Changes between 1.0.0e and 1.0.0f [4 Jan 2012]
1261 *) Nadhem Alfardan and Kenny Paterson have discovered an extension
1262 of the Vaudenay padding oracle attack on CBC mode encryption
1263 which enables an efficient plaintext recovery attack against
1264 the OpenSSL implementation of DTLS. Their attack exploits timing
1265 differences arising during decryption processing. A research
1266 paper describing this attack can be found at:
1267 http://www.isg.rhul.ac.uk/~kp/dtls.pdf
1268 Thanks go to Nadhem Alfardan and Kenny Paterson of the Information
1269 Security Group at Royal Holloway, University of London
1270 (www.isg.rhul.ac.uk) for discovering this flaw and to Robin Seggelmann
1271 <seggelmann@fh-muenster.de> and Michael Tuexen <tuexen@fh-muenster.de>
1272 for preparing the fix. (CVE-2011-4108)
1273 [Robin Seggelmann, Michael Tuexen]
1275 *) Clear bytes used for block padding of SSL 3.0 records.
1277 [Adam Langley (Google)]
1279 *) Only allow one SGC handshake restart for SSL/TLS. Thanks to George
1280 Kadianakis <desnacked@gmail.com> for discovering this issue and
1281 Adam Langley for preparing the fix. (CVE-2011-4619)
1282 [Adam Langley (Google)]
1284 *) Check parameters are not NULL in GOST ENGINE. (CVE-2012-0027)
1285 [Andrey Kulikov <amdeich@gmail.com>]
1287 *) Prevent malformed RFC3779 data triggering an assertion failure.
1288 Thanks to Andrew Chi, BBN Technologies, for discovering the flaw
1289 and Rob Austein <sra@hactrn.net> for fixing it. (CVE-2011-4577)
1290 [Rob Austein <sra@hactrn.net>]
1292 *) Improved PRNG seeding for VOS.
1293 [Paul Green <Paul.Green@stratus.com>]
1295 *) Fix ssl_ciph.c set-up race.
1296 [Adam Langley (Google)]
1298 *) Fix spurious failures in ecdsatest.c.
1299 [Emilia Käsper (Google)]
1301 *) Fix the BIO_f_buffer() implementation (which was mixing different
1302 interpretations of the '..._len' fields).
1303 [Adam Langley (Google)]
1305 *) Fix handling of BN_BLINDING: now BN_BLINDING_invert_ex (rather than
1306 BN_BLINDING_invert_ex) calls BN_BLINDING_update, ensuring that concurrent
1307 threads won't reuse the same blinding coefficients.
1309 This also avoids the need to obtain the CRYPTO_LOCK_RSA_BLINDING
1310 lock to call BN_BLINDING_invert_ex, and avoids one use of
1311 BN_BLINDING_update for each BN_BLINDING structure (previously,
1312 the last update always remained unused).
1313 [Emilia Käsper (Google)]
1315 *) In ssl3_clear, preserve s3->init_extra along with s3->rbuf.
1316 [Bob Buckholz (Google)]
1318 Changes between 1.0.0d and 1.0.0e [6 Sep 2011]
1320 *) Fix bug where CRLs with nextUpdate in the past are sometimes accepted
1321 by initialising X509_STORE_CTX properly. (CVE-2011-3207)
1322 [Kaspar Brand <ossl@velox.ch>]
1324 *) Fix SSL memory handling for (EC)DH ciphersuites, in particular
1325 for multi-threaded use of ECDH. (CVE-2011-3210)
1326 [Adam Langley (Google)]
1328 *) Fix x509_name_ex_d2i memory leak on bad inputs.
1331 *) Remove hard coded ecdsaWithSHA1 signature tests in ssl code and check
1332 signature public key algorithm by using OID xref utilities instead.
1333 Before this you could only use some ECC ciphersuites with SHA1 only.
1336 *) Add protection against ECDSA timing attacks as mentioned in the paper
1337 by Billy Bob Brumley and Nicola Tuveri, see:
1339 http://eprint.iacr.org/2011/232.pdf
1341 [Billy Bob Brumley and Nicola Tuveri]
1343 Changes between 1.0.0c and 1.0.0d [8 Feb 2011]
1345 *) Fix parsing of OCSP stapling ClientHello extension. CVE-2011-0014
1346 [Neel Mehta, Adam Langley, Bodo Moeller (Google)]
1348 *) Fix bug in string printing code: if *any* escaping is enabled we must
1349 escape the escape character (backslash) or the resulting string is
1353 Changes between 1.0.0b and 1.0.0c [2 Dec 2010]
1355 *) Disable code workaround for ancient and obsolete Netscape browsers
1356 and servers: an attacker can use it in a ciphersuite downgrade attack.
1357 Thanks to Martin Rex for discovering this bug. CVE-2010-4180
1360 *) Fixed J-PAKE implementation error, originally discovered by
1361 Sebastien Martini, further info and confirmation from Stefan
1362 Arentz and Feng Hao. Note that this fix is a security fix. CVE-2010-4252
1365 Changes between 1.0.0a and 1.0.0b [16 Nov 2010]
1367 *) Fix extension code to avoid race conditions which can result in a buffer
1368 overrun vulnerability: resumed sessions must not be modified as they can
1369 be shared by multiple threads. CVE-2010-3864
1372 *) Fix WIN32 build system to correctly link an ENGINE directory into
1376 Changes between 1.0.0 and 1.0.0a [01 Jun 2010]
1378 *) Check return value of int_rsa_verify in pkey_rsa_verifyrecover
1380 [Steve Henson, Peter-Michael Hager <hager@dortmund.net>]
1382 Changes between 0.9.8n and 1.0.0 [29 Mar 2010]
1384 *) Add "missing" function EVP_CIPHER_CTX_copy(). This copies a cipher
1385 context. The operation can be customised via the ctrl mechanism in
1386 case ENGINEs want to include additional functionality.
1389 *) Tolerate yet another broken PKCS#8 key format: private key value negative.
1392 *) Add new -subject_hash_old and -issuer_hash_old options to x509 utility to
1393 output hashes compatible with older versions of OpenSSL.
1394 [Willy Weisz <weisz@vcpc.univie.ac.at>]
1396 *) Fix compression algorithm handling: if resuming a session use the
1397 compression algorithm of the resumed session instead of determining
1398 it from client hello again. Don't allow server to change algorithm.
1401 *) Add load_crls() function to apps tidying load_certs() too. Add option
1402 to verify utility to allow additional CRLs to be included.
1405 *) Update OCSP request code to permit adding custom headers to the request:
1406 some responders need this.
1409 *) The function EVP_PKEY_sign() returns <=0 on error: check return code
1411 [Julia Lawall <julia@diku.dk>]
1413 *) Update verify callback code in apps/s_cb.c and apps/verify.c, it
1414 needlessly dereferenced structures, used obsolete functions and
1415 didn't handle all updated verify codes correctly.
1418 *) Disable MD2 in the default configuration.
1421 *) In BIO_pop() and BIO_push() use the ctrl argument (which was NULL) to
1422 indicate the initial BIO being pushed or popped. This makes it possible
1423 to determine whether the BIO is the one explicitly called or as a result
1424 of the ctrl being passed down the chain. Fix BIO_pop() and SSL BIOs so
1425 it handles reference counts correctly and doesn't zero out the I/O bio
1426 when it is not being explicitly popped. WARNING: applications which
1427 included workarounds for the old buggy behaviour will need to be modified
1428 or they could free up already freed BIOs.
1431 *) Extend the uni2asc/asc2uni => OPENSSL_uni2asc/OPENSSL_asc2uni
1432 renaming to all platforms (within the 0.9.8 branch, this was
1433 done conditionally on Netware platforms to avoid a name clash).
1434 [Guenter <lists@gknw.net>]
1436 *) Add ECDHE and PSK support to DTLS.
1437 [Michael Tuexen <tuexen@fh-muenster.de>]
1439 *) Add CHECKED_STACK_OF macro to safestack.h, otherwise safestack can't
1443 *) Add "missing" function EVP_MD_flags() (without this the only way to
1444 retrieve a digest flags is by accessing the structure directly. Update
1445 EVP_MD_do_all*() and EVP_CIPHER_do_all*() to include the name a digest
1446 or cipher is registered as in the "from" argument. Print out all
1447 registered digests in the dgst usage message instead of manually
1448 attempting to work them out.
1451 *) If no SSLv2 ciphers are used don't use an SSLv2 compatible client hello:
1452 this allows the use of compression and extensions. Change default cipher
1453 string to remove SSLv2 ciphersuites. This effectively avoids ancient SSLv2
1454 by default unless an application cipher string requests it.
1457 *) Alter match criteria in PKCS12_parse(). It used to try to use local
1458 key ids to find matching certificates and keys but some PKCS#12 files
1459 don't follow the (somewhat unwritten) rules and this strategy fails.
1460 Now just gather all certificates together and the first private key
1461 then look for the first certificate that matches the key.
1464 *) Support use of registered digest and cipher names for dgst and cipher
1465 commands instead of having to add each one as a special case. So now
1472 openssl dgst -sha256 foo
1474 and this works for ENGINE based algorithms too.
1478 *) Update Gost ENGINE to support parameter files.
1479 [Victor B. Wagner <vitus@cryptocom.ru>]
1481 *) Support GeneralizedTime in ca utility.
1482 [Oliver Martin <oliver@volatilevoid.net>, Steve Henson]
1484 *) Enhance the hash format used for certificate directory links. The new
1485 form uses the canonical encoding (meaning equivalent names will work
1486 even if they aren't identical) and uses SHA1 instead of MD5. This form
1487 is incompatible with the older format and as a result c_rehash should
1488 be used to rebuild symbolic links.
1491 *) Make PKCS#8 the default write format for private keys, replacing the
1492 traditional format. This form is standardised, more secure and doesn't
1493 include an implicit MD5 dependency.
1496 *) Add a $gcc_devteam_warn option to Configure. The idea is that any code
1497 committed to OpenSSL should pass this lot as a minimum.
1500 *) Add session ticket override functionality for use by EAP-FAST.
1501 [Jouni Malinen <j@w1.fi>]
1503 *) Modify HMAC functions to return a value. Since these can be implemented
1504 in an ENGINE errors can occur.
1507 *) Type-checked OBJ_bsearch_ex.
1510 *) Type-checked OBJ_bsearch. Also some constification necessitated
1511 by type-checking. Still to come: TXT_DB, bsearch(?),
1512 OBJ_bsearch_ex, qsort, CRYPTO_EX_DATA, ASN1_VALUE, ASN1_STRING,
1516 *) New function OPENSSL_gmtime_adj() to add a specific number of days and
1517 seconds to a tm structure directly, instead of going through OS
1518 specific date routines. This avoids any issues with OS routines such
1519 as the year 2038 bug. New *_adj() functions for ASN1 time structures
1520 and X509_time_adj_ex() to cover the extended range. The existing
1521 X509_time_adj() is still usable and will no longer have any date issues.
1524 *) Delta CRL support. New use deltas option which will attempt to locate
1525 and search any appropriate delta CRLs available.
1527 This work was sponsored by Google.
1530 *) Support for CRLs partitioned by reason code. Reorganise CRL processing
1531 code and add additional score elements. Validate alternate CRL paths
1532 as part of the CRL checking and indicate a new error "CRL path validation
1533 error" in this case. Applications wanting additional details can use
1534 the verify callback and check the new "parent" field. If this is not
1535 NULL CRL path validation is taking place. Existing applications wont
1536 see this because it requires extended CRL support which is off by
1539 This work was sponsored by Google.
1542 *) Support for freshest CRL extension.
1544 This work was sponsored by Google.
1547 *) Initial indirect CRL support. Currently only supported in the CRLs
1548 passed directly and not via lookup. Process certificate issuer
1549 CRL entry extension and lookup CRL entries by bother issuer name
1550 and serial number. Check and process CRL issuer entry in IDP extension.
1552 This work was sponsored by Google.
1555 *) Add support for distinct certificate and CRL paths. The CRL issuer
1556 certificate is validated separately in this case. Only enabled if
1557 an extended CRL support flag is set: this flag will enable additional
1558 CRL functionality in future.
1560 This work was sponsored by Google.
1563 *) Add support for policy mappings extension.
1565 This work was sponsored by Google.
1568 *) Fixes to pathlength constraint, self issued certificate handling,
1569 policy processing to align with RFC3280 and PKITS tests.
1571 This work was sponsored by Google.
1574 *) Support for name constraints certificate extension. DN, email, DNS
1575 and URI types are currently supported.
1577 This work was sponsored by Google.
1580 *) To cater for systems that provide a pointer-based thread ID rather
1581 than numeric, deprecate the current numeric thread ID mechanism and
1582 replace it with a structure and associated callback type. This
1583 mechanism allows a numeric "hash" to be extracted from a thread ID in
1584 either case, and on platforms where pointers are larger than 'long',
1585 mixing is done to help ensure the numeric 'hash' is usable even if it
1586 can't be guaranteed unique. The default mechanism is to use "&errno"
1587 as a pointer-based thread ID to distinguish between threads.
1589 Applications that want to provide their own thread IDs should now use
1590 CRYPTO_THREADID_set_callback() to register a callback that will call
1591 either CRYPTO_THREADID_set_numeric() or CRYPTO_THREADID_set_pointer().
1593 Note that ERR_remove_state() is now deprecated, because it is tied
1594 to the assumption that thread IDs are numeric. ERR_remove_state(0)
1595 to free the current thread's error state should be replaced by
1596 ERR_remove_thread_state(NULL).
1598 (This new approach replaces the functions CRYPTO_set_idptr_callback(),
1599 CRYPTO_get_idptr_callback(), and CRYPTO_thread_idptr() that existed in
1600 OpenSSL 0.9.9-dev between June 2006 and August 2008. Also, if an
1601 application was previously providing a numeric thread callback that
1602 was inappropriate for distinguishing threads, then uniqueness might
1603 have been obtained with &errno that happened immediately in the
1604 intermediate development versions of OpenSSL; this is no longer the
1605 case, the numeric thread callback will now override the automatic use
1607 [Geoff Thorpe, with help from Bodo Moeller]
1609 *) Initial support for different CRL issuing certificates. This covers a
1610 simple case where the self issued certificates in the chain exist and
1611 the real CRL issuer is higher in the existing chain.
1613 This work was sponsored by Google.
1616 *) Removed effectively defunct crypto/store from the build.
1619 *) Revamp of STACK to provide stronger type-checking. Still to come:
1620 TXT_DB, bsearch(?), OBJ_bsearch, qsort, CRYPTO_EX_DATA, ASN1_VALUE,
1621 ASN1_STRING, CONF_VALUE.
1624 *) Add a new SSL_MODE_RELEASE_BUFFERS mode flag to release unused buffer
1625 RAM on SSL connections. This option can save about 34k per idle SSL.
1628 *) Revamp of LHASH to provide stronger type-checking. Still to come:
1629 STACK, TXT_DB, bsearch, qsort.
1632 *) Initial support for Cryptographic Message Syntax (aka CMS) based
1633 on RFC3850, RFC3851 and RFC3852. New cms directory and cms utility,
1634 support for data, signedData, compressedData, digestedData and
1635 encryptedData, envelopedData types included. Scripts to check against
1636 RFC4134 examples draft and interop and consistency checks of many
1637 content types and variants.
1640 *) Add options to enc utility to support use of zlib compression BIO.
1643 *) Extend mk1mf to support importing of options and assembly language
1644 files from Configure script, currently only included in VC-WIN32.
1645 The assembly language rules can now optionally generate the source
1646 files from the associated perl scripts.
1649 *) Implement remaining functionality needed to support GOST ciphersuites.
1650 Interop testing has been performed using CryptoPro implementations.
1651 [Victor B. Wagner <vitus@cryptocom.ru>]
1653 *) s390x assembler pack.
1656 *) ARMv4 assembler pack. ARMv4 refers to v4 and later ISA, not CPU
1660 *) Implement Opaque PRF Input TLS extension as specified in
1661 draft-rescorla-tls-opaque-prf-input-00.txt. Since this is not an
1662 official specification yet and no extension type assignment by
1663 IANA exists, this extension (for now) will have to be explicitly
1664 enabled when building OpenSSL by providing the extension number
1665 to use. For example, specify an option
1667 -DTLSEXT_TYPE_opaque_prf_input=0x9527
1669 to the "config" or "Configure" script to enable the extension,
1670 assuming extension number 0x9527 (which is a completely arbitrary
1671 and unofficial assignment based on the MD5 hash of the Internet
1672 Draft). Note that by doing so, you potentially lose
1673 interoperability with other TLS implementations since these might
1674 be using the same extension number for other purposes.
1676 SSL_set_tlsext_opaque_prf_input(ssl, src, len) is used to set the
1677 opaque PRF input value to use in the handshake. This will create
1678 an interal copy of the length-'len' string at 'src', and will
1679 return non-zero for success.
1681 To get more control and flexibility, provide a callback function
1684 SSL_CTX_set_tlsext_opaque_prf_input_callback(ctx, cb)
1685 SSL_CTX_set_tlsext_opaque_prf_input_callback_arg(ctx, arg)
1689 int (*cb)(SSL *, void *peerinput, size_t len, void *arg);
1692 Callback function 'cb' will be called in handshakes, and is
1693 expected to use SSL_set_tlsext_opaque_prf_input() as appropriate.
1694 Argument 'arg' is for application purposes (the value as given to
1695 SSL_CTX_set_tlsext_opaque_prf_input_callback_arg() will directly
1696 be provided to the callback function). The callback function
1697 has to return non-zero to report success: usually 1 to use opaque
1698 PRF input just if possible, or 2 to enforce use of the opaque PRF
1699 input. In the latter case, the library will abort the handshake
1700 if opaque PRF input is not successfully negotiated.
1702 Arguments 'peerinput' and 'len' given to the callback function
1703 will always be NULL and 0 in the case of a client. A server will
1704 see the client's opaque PRF input through these variables if
1705 available (NULL and 0 otherwise). Note that if the server
1706 provides an opaque PRF input, the length must be the same as the
1707 length of the client's opaque PRF input.
1709 Note that the callback function will only be called when creating
1710 a new session (session resumption can resume whatever was
1711 previously negotiated), and will not be called in SSL 2.0
1712 handshakes; thus, SSL_CTX_set_options(ctx, SSL_OP_NO_SSLv2) or
1713 SSL_set_options(ssl, SSL_OP_NO_SSLv2) is especially recommended
1714 for applications that need to enforce opaque PRF input.
1718 *) Update ssl code to support digests other than SHA1+MD5 for handshake
1721 [Victor B. Wagner <vitus@cryptocom.ru>]
1723 *) Add RFC4507 support to OpenSSL. This includes the corrections in
1724 RFC4507bis. The encrypted ticket format is an encrypted encoded
1725 SSL_SESSION structure, that way new session features are automatically
1728 If a client application caches session in an SSL_SESSION structure
1729 support is transparent because tickets are now stored in the encoded
1732 The SSL_CTX structure automatically generates keys for ticket
1733 protection in servers so again support should be possible
1734 with no application modification.
1736 If a client or server wishes to disable RFC4507 support then the option
1737 SSL_OP_NO_TICKET can be set.
1739 Add a TLS extension debugging callback to allow the contents of any client
1740 or server extensions to be examined.
1742 This work was sponsored by Google.
1745 *) Final changes to avoid use of pointer pointer casts in OpenSSL.
1746 OpenSSL should now compile cleanly on gcc 4.2
1747 [Peter Hartley <pdh@utter.chaos.org.uk>, Steve Henson]
1749 *) Update SSL library to use new EVP_PKEY MAC API. Include generic MAC
1750 support including streaming MAC support: this is required for GOST
1751 ciphersuite support.
1752 [Victor B. Wagner <vitus@cryptocom.ru>, Steve Henson]
1754 *) Add option -stream to use PKCS#7 streaming in smime utility. New
1755 function i2d_PKCS7_bio_stream() and PEM_write_PKCS7_bio_stream()
1756 to output in BER and PEM format.
1759 *) Experimental support for use of HMAC via EVP_PKEY interface. This
1760 allows HMAC to be handled via the EVP_DigestSign*() interface. The
1761 EVP_PKEY "key" in this case is the HMAC key, potentially allowing
1762 ENGINE support for HMAC keys which are unextractable. New -mac and
1763 -macopt options to dgst utility.
1766 *) New option -sigopt to dgst utility. Update dgst to use
1767 EVP_Digest{Sign,Verify}*. These two changes make it possible to use
1768 alternative signing paramaters such as X9.31 or PSS in the dgst
1772 *) Change ssl_cipher_apply_rule(), the internal function that does
1773 the work each time a ciphersuite string requests enabling
1774 ("foo+bar"), moving ("+foo+bar"), disabling ("-foo+bar", or
1775 removing ("!foo+bar") a class of ciphersuites: Now it maintains
1776 the order of disabled ciphersuites such that those ciphersuites
1777 that most recently went from enabled to disabled not only stay
1778 in order with respect to each other, but also have higher priority
1779 than other disabled ciphersuites the next time ciphersuites are
1782 This means that you can now say, e.g., "PSK:-PSK:HIGH" to enable
1783 the same ciphersuites as with "HIGH" alone, but in a specific
1784 order where the PSK ciphersuites come first (since they are the
1785 most recently disabled ciphersuites when "HIGH" is parsed).
1787 Also, change ssl_create_cipher_list() (using this new
1788 funcionality) such that between otherwise identical
1789 cihpersuites, ephemeral ECDH is preferred over ephemeral DH in
1793 *) Change ssl_create_cipher_list() so that it automatically
1794 arranges the ciphersuites in reasonable order before starting
1795 to process the rule string. Thus, the definition for "DEFAULT"
1796 (SSL_DEFAULT_CIPHER_LIST) now is just "ALL:!aNULL:!eNULL", but
1797 remains equivalent to "AES:ALL:!aNULL:!eNULL:+aECDH:+kRSA:+RC4:@STRENGTH".
1798 This makes it much easier to arrive at a reasonable default order
1799 in applications for which anonymous ciphers are OK (meaning
1800 that you can't actually use DEFAULT).
1801 [Bodo Moeller; suggested by Victor Duchovni]
1803 *) Split the SSL/TLS algorithm mask (as used for ciphersuite string
1804 processing) into multiple integers instead of setting
1805 "SSL_MKEY_MASK" bits, "SSL_AUTH_MASK" bits, "SSL_ENC_MASK",
1806 "SSL_MAC_MASK", and "SSL_SSL_MASK" bits all in a single integer.
1807 (These masks as well as the individual bit definitions are hidden
1808 away into the non-exported interface ssl/ssl_locl.h, so this
1809 change to the definition of the SSL_CIPHER structure shouldn't
1810 affect applications.) This give us more bits for each of these
1811 categories, so there is no longer a need to coagulate AES128 and
1812 AES256 into a single algorithm bit, and to coagulate Camellia128
1813 and Camellia256 into a single algorithm bit, which has led to all
1816 Thus, among other things, the kludge introduced in 0.9.7m and
1817 0.9.8e for masking out AES256 independently of AES128 or masking
1818 out Camellia256 independently of AES256 is not needed here in 0.9.9.
1820 With the change, we also introduce new ciphersuite aliases that
1821 so far were missing: "AES128", "AES256", "CAMELLIA128", and
1825 *) Add support for dsa-with-SHA224 and dsa-with-SHA256.
1826 Use the leftmost N bytes of the signature input if the input is
1827 larger than the prime q (with N being the size in bytes of q).
1830 *) Very *very* experimental PKCS#7 streaming encoder support. Nothing uses
1831 it yet and it is largely untested.
1834 *) Add support for the ecdsa-with-SHA224/256/384/512 signature types.
1837 *) Initial incomplete changes to avoid need for function casts in OpenSSL
1838 some compilers (gcc 4.2 and later) reject their use. Safestack is
1839 reimplemented. Update ASN1 to avoid use of legacy functions.
1842 *) Win32/64 targets are linked with Winsock2.
1845 *) Add an X509_CRL_METHOD structure to allow CRL processing to be redirected
1846 to external functions. This can be used to increase CRL handling
1847 efficiency especially when CRLs are very large by (for example) storing
1848 the CRL revoked certificates in a database.
1851 *) Overhaul of by_dir code. Add support for dynamic loading of CRLs so
1852 new CRLs added to a directory can be used. New command line option
1853 -verify_return_error to s_client and s_server. This causes real errors
1854 to be returned by the verify callback instead of carrying on no matter
1855 what. This reflects the way a "real world" verify callback would behave.
1858 *) GOST engine, supporting several GOST algorithms and public key formats.
1859 Kindly donated by Cryptocom.
1862 *) Partial support for Issuing Distribution Point CRL extension. CRLs
1863 partitioned by DP are handled but no indirect CRL or reason partitioning
1864 (yet). Complete overhaul of CRL handling: now the most suitable CRL is
1865 selected via a scoring technique which handles IDP and AKID in CRLs.
1868 *) New X509_STORE_CTX callbacks lookup_crls() and lookup_certs() which
1869 will ultimately be used for all verify operations: this will remove the
1870 X509_STORE dependency on certificate verification and allow alternative
1871 lookup methods. X509_STORE based implementations of these two callbacks.
1874 *) Allow multiple CRLs to exist in an X509_STORE with matching issuer names.
1875 Modify get_crl() to find a valid (unexpired) CRL if possible.
1878 *) New function X509_CRL_match() to check if two CRLs are identical. Normally
1879 this would be called X509_CRL_cmp() but that name is already used by
1880 a function that just compares CRL issuer names. Cache several CRL
1881 extensions in X509_CRL structure and cache CRLDP in X509.
1884 *) Store a "canonical" representation of X509_NAME structure (ASN1 Name)
1885 this maps equivalent X509_NAME structures into a consistent structure.
1886 Name comparison can then be performed rapidly using memcmp().
1889 *) Non-blocking OCSP request processing. Add -timeout option to ocsp
1893 *) Allow digests to supply their own micalg string for S/MIME type using
1894 the ctrl EVP_MD_CTRL_MICALG.
1897 *) During PKCS7 signing pass the PKCS7 SignerInfo structure to the
1898 EVP_PKEY_METHOD before and after signing via the EVP_PKEY_CTRL_PKCS7_SIGN
1899 ctrl. It can then customise the structure before and/or after signing
1903 *) New function OBJ_add_sigid() to allow application defined signature OIDs
1904 to be added to OpenSSLs internal tables. New function OBJ_sigid_free()
1905 to free up any added signature OIDs.
1908 *) New functions EVP_CIPHER_do_all(), EVP_CIPHER_do_all_sorted(),
1909 EVP_MD_do_all() and EVP_MD_do_all_sorted() to enumerate internal
1910 digest and cipher tables. New options added to openssl utility:
1911 list-message-digest-algorithms and list-cipher-algorithms.
1914 *) Change the array representation of binary polynomials: the list
1915 of degrees of non-zero coefficients is now terminated with -1.
1916 Previously it was terminated with 0, which was also part of the
1917 value; thus, the array representation was not applicable to
1918 polynomials where t^0 has coefficient zero. This change makes
1919 the array representation useful in a more general context.
1922 *) Various modifications and fixes to SSL/TLS cipher string
1923 handling. For ECC, the code now distinguishes between fixed ECDH
1924 with RSA certificates on the one hand and with ECDSA certificates
1925 on the other hand, since these are separate ciphersuites. The
1926 unused code for Fortezza ciphersuites has been removed.
1928 For consistency with EDH, ephemeral ECDH is now called "EECDH"
1929 (not "ECDHE"). For consistency with the code for DH
1930 certificates, use of ECDH certificates is now considered ECDH
1931 authentication, not RSA or ECDSA authentication (the latter is
1932 merely the CA's signing algorithm and not actively used in the
1935 The temporary ciphersuite alias "ECCdraft" is no longer
1936 available, and ECC ciphersuites are no longer excluded from "ALL"
1937 and "DEFAULT". The following aliases now exist for RFC 4492
1938 ciphersuites, most of these by analogy with the DH case:
1940 kECDHr - ECDH cert, signed with RSA
1941 kECDHe - ECDH cert, signed with ECDSA
1942 kECDH - ECDH cert (signed with either RSA or ECDSA)
1943 kEECDH - ephemeral ECDH
1944 ECDH - ECDH cert or ephemeral ECDH
1950 AECDH - anonymous ECDH
1951 EECDH - non-anonymous ephemeral ECDH (equivalent to "kEECDH:-AECDH")
1955 *) Add additional S/MIME capabilities for AES and GOST ciphers if supported.
1956 Use correct micalg parameters depending on digest(s) in signed message.
1959 *) Add engine support for EVP_PKEY_ASN1_METHOD. Add functions to process
1960 an ENGINE asn1 method. Support ENGINE lookups in the ASN1 code.
1963 *) Initial engine support for EVP_PKEY_METHOD. New functions to permit
1964 an engine to register a method. Add ENGINE lookups for methods and
1965 functional reference processing.
1968 *) New functions EVP_Digest{Sign,Verify)*. These are enchance versions of
1969 EVP_{Sign,Verify}* which allow an application to customise the signature
1973 *) New -resign option to smime utility. This adds one or more signers
1974 to an existing PKCS#7 signedData structure. Also -md option to use an
1975 alternative message digest algorithm for signing.
1978 *) Tidy up PKCS#7 routines and add new functions to make it easier to
1979 create PKCS7 structures containing multiple signers. Update smime
1980 application to support multiple signers.
1983 *) New -macalg option to pkcs12 utility to allow setting of an alternative
1987 *) Initial support for PKCS#5 v2.0 PRFs other than default SHA1 HMAC.
1988 Reorganize PBE internals to lookup from a static table using NIDs,
1989 add support for HMAC PBE OID translation. Add a EVP_CIPHER ctrl:
1990 EVP_CTRL_PBE_PRF_NID this allows a cipher to specify an alternative
1991 PRF which will be automatically used with PBES2.
1994 *) Replace the algorithm specific calls to generate keys in "req" with the
1998 *) Update PKCS#7 enveloped data routines to use new API. This is now
1999 supported by any public key method supporting the encrypt operation. A
2000 ctrl is added to allow the public key algorithm to examine or modify
2001 the PKCS#7 RecipientInfo structure if it needs to: for RSA this is
2005 *) Add a ctrl to asn1 method to allow a public key algorithm to express
2006 a default digest type to use. In most cases this will be SHA1 but some
2007 algorithms (such as GOST) need to specify an alternative digest. The
2008 return value indicates how strong the preference is 1 means optional and
2009 2 is mandatory (that is it is the only supported type). Modify
2010 ASN1_item_sign() to accept a NULL digest argument to indicate it should
2011 use the default md. Update openssl utilities to use the default digest
2012 type for signing if it is not explicitly indicated.
2015 *) Use OID cross reference table in ASN1_sign() and ASN1_verify(). New
2016 EVP_MD flag EVP_MD_FLAG_PKEY_METHOD_SIGNATURE. This uses the relevant
2017 signing method from the key type. This effectively removes the link
2018 between digests and public key types.
2021 *) Add an OID cross reference table and utility functions. Its purpose is to
2022 translate between signature OIDs such as SHA1WithrsaEncryption and SHA1,
2023 rsaEncryption. This will allow some of the algorithm specific hackery
2024 needed to use the correct OID to be removed.
2027 *) Remove algorithm specific dependencies when setting PKCS7_SIGNER_INFO
2028 structures for PKCS7_sign(). They are now set up by the relevant public
2032 *) Add provisional EC pkey method with support for ECDSA and ECDH.
2035 *) Add support for key derivation (agreement) in the API, DH method and
2039 *) Add DSA pkey method and DH pkey methods, extend DH ASN1 method to support
2040 public and private key formats. As a side effect these add additional
2041 command line functionality not previously available: DSA signatures can be
2042 generated and verified using pkeyutl and DH key support and generation in
2047 [Oliver Tappe <zooey@hirschkaefer.de>]
2049 *) New make target "install_html_docs" installs HTML renditions of the
2051 [Oliver Tappe <zooey@hirschkaefer.de>]
2053 *) New utility "genpkey" this is analogous to "genrsa" etc except it can
2054 generate keys for any algorithm. Extend and update EVP_PKEY_METHOD to
2055 support key and parameter generation and add initial key generation
2056 functionality for RSA.
2059 *) Add functions for main EVP_PKEY_method operations. The undocumented
2060 functions EVP_PKEY_{encrypt,decrypt} have been renamed to
2061 EVP_PKEY_{encrypt,decrypt}_old.
2064 *) Initial definitions for EVP_PKEY_METHOD. This will be a high level public
2065 key API, doesn't do much yet.
2068 *) New function EVP_PKEY_asn1_get0_info() to retrieve information about
2069 public key algorithms. New option to openssl utility:
2070 "list-public-key-algorithms" to print out info.
2073 *) Implement the Supported Elliptic Curves Extension for
2074 ECC ciphersuites from draft-ietf-tls-ecc-12.txt.
2077 *) Don't free up OIDs in OBJ_cleanup() if they are in use by EVP_MD or
2078 EVP_CIPHER structures to avoid later problems in EVP_cleanup().
2081 *) New utilities pkey and pkeyparam. These are similar to algorithm specific
2082 utilities such as rsa, dsa, dsaparam etc except they process any key
2086 *) Transfer public key printing routines to EVP_PKEY_ASN1_METHOD. New
2087 functions EVP_PKEY_print_public(), EVP_PKEY_print_private(),
2088 EVP_PKEY_print_param() to print public key data from an EVP_PKEY
2092 *) Initial support for pluggable public key ASN1.
2093 De-spaghettify the public key ASN1 handling. Move public and private
2094 key ASN1 handling to a new EVP_PKEY_ASN1_METHOD structure. Relocate
2095 algorithm specific handling to a single module within the relevant
2096 algorithm directory. Add functions to allow (near) opaque processing
2097 of public and private key structures.
2100 *) Implement the Supported Point Formats Extension for
2101 ECC ciphersuites from draft-ietf-tls-ecc-12.txt.
2104 *) Add initial support for RFC 4279 PSK TLS ciphersuites. Add members
2105 for the psk identity [hint] and the psk callback functions to the
2106 SSL_SESSION, SSL and SSL_CTX structure.
2109 PSK-RC4-SHA, PSK-3DES-EDE-CBC-SHA, PSK-AES128-CBC-SHA,
2113 SSL_CTX_use_psk_identity_hint
2114 SSL_get_psk_identity_hint
2115 SSL_get_psk_identity
2116 SSL_use_psk_identity_hint
2118 [Mika Kousa and Pasi Eronen of Nokia Corporation]
2120 *) Add RFC 3161 compliant time stamp request creation, response generation
2121 and response verification functionality.
2122 [Zoltán Glózik <zglozik@opentsa.org>, The OpenTSA Project]
2124 *) Add initial support for TLS extensions, specifically for the server_name
2125 extension so far. The SSL_SESSION, SSL_CTX, and SSL data structures now
2126 have new members for a host name. The SSL data structure has an
2127 additional member SSL_CTX *initial_ctx so that new sessions can be
2128 stored in that context to allow for session resumption, even after the
2129 SSL has been switched to a new SSL_CTX in reaction to a client's
2130 server_name extension.
2132 New functions (subject to change):
2134 SSL_get_servername()
2135 SSL_get_servername_type()
2138 New CTRL codes and macros (subject to change):
2140 SSL_CTRL_SET_TLSEXT_SERVERNAME_CB
2141 - SSL_CTX_set_tlsext_servername_callback()
2142 SSL_CTRL_SET_TLSEXT_SERVERNAME_ARG
2143 - SSL_CTX_set_tlsext_servername_arg()
2144 SSL_CTRL_SET_TLSEXT_HOSTNAME - SSL_set_tlsext_host_name()
2146 openssl s_client has a new '-servername ...' option.
2148 openssl s_server has new options '-servername_host ...', '-cert2 ...',
2149 '-key2 ...', '-servername_fatal' (subject to change). This allows
2150 testing the HostName extension for a specific single host name ('-cert'
2151 and '-key' remain fallbacks for handshakes without HostName
2152 negotiation). If the unrecognized_name alert has to be sent, this by
2153 default is a warning; it becomes fatal with the '-servername_fatal'
2156 [Peter Sylvester, Remy Allais, Christophe Renou]
2158 *) Whirlpool hash implementation is added.
2161 *) BIGNUM code on 64-bit SPARCv9 targets is switched from bn(64,64) to
2162 bn(64,32). Because of instruction set limitations it doesn't have
2163 any negative impact on performance. This was done mostly in order
2164 to make it possible to share assembler modules, such as bn_mul_mont
2165 implementations, between 32- and 64-bit builds without hassle.
2168 *) Move code previously exiled into file crypto/ec/ec2_smpt.c
2169 to ec2_smpl.c, and no longer require the OPENSSL_EC_BIN_PT_COMP
2173 *) New candidate for BIGNUM assembler implementation, bn_mul_mont,
2174 dedicated Montgomery multiplication procedure, is introduced.
2175 BN_MONT_CTX is modified to allow bn_mul_mont to reach for higher
2176 "64-bit" performance on certain 32-bit targets.
2179 *) New option SSL_OP_NO_COMP to disable use of compression selectively
2180 in SSL structures. New SSL ctrl to set maximum send fragment size.
2181 Save memory by seeting the I/O buffer sizes dynamically instead of
2182 using the maximum available value.
2185 *) New option -V for 'openssl ciphers'. This prints the ciphersuite code
2186 in addition to the text details.
2189 *) Very, very preliminary EXPERIMENTAL support for printing of general
2190 ASN1 structures. This currently produces rather ugly output and doesn't
2191 handle several customised structures at all.
2194 *) Integrated support for PVK file format and some related formats such
2195 as MS PUBLICKEYBLOB and PRIVATEKEYBLOB. Command line switches to support
2196 these in the 'rsa' and 'dsa' utilities.
2199 *) Support for PKCS#1 RSAPublicKey format on rsa utility command line.
2202 *) Remove the ancient ASN1_METHOD code. This was only ever used in one
2203 place for the (very old) "NETSCAPE" format certificates which are now
2204 handled using new ASN1 code equivalents.
2207 *) Let the TLSv1_method() etc. functions return a 'const' SSL_METHOD
2208 pointer and make the SSL_METHOD parameter in SSL_CTX_new,
2209 SSL_CTX_set_ssl_version and SSL_set_ssl_method 'const'.
2212 *) Modify CRL distribution points extension code to print out previously
2213 unsupported fields. Enhance extension setting code to allow setting of
2217 *) Add print and set support for Issuing Distribution Point CRL extension.
2220 *) Change 'Configure' script to enable Camellia by default.
2223 Changes between 0.9.8x and 0.9.8y [5 Feb 2013]
2225 *) Make the decoding of SSLv3, TLS and DTLS CBC records constant time.
2227 This addresses the flaw in CBC record processing discovered by
2228 Nadhem Alfardan and Kenny Paterson. Details of this attack can be found
2229 at: http://www.isg.rhul.ac.uk/tls/
2231 Thanks go to Nadhem Alfardan and Kenny Paterson of the Information
2232 Security Group at Royal Holloway, University of London
2233 (www.isg.rhul.ac.uk) for discovering this flaw and Adam Langley and
2234 Emilia Käsper for the initial patch.
2236 [Emilia Käsper, Adam Langley, Ben Laurie, Andy Polyakov, Steve Henson]
2238 *) Return an error when checking OCSP signatures when key is NULL.
2239 This fixes a DoS attack. (CVE-2013-0166)
2242 *) Call OCSP Stapling callback after ciphersuite has been chosen, so
2243 the right response is stapled. Also change SSL_get_certificate()
2244 so it returns the certificate actually sent.
2245 See http://rt.openssl.org/Ticket/Display.html?id=2836.
2246 (This is a backport)
2247 [Rob Stradling <rob.stradling@comodo.com>]
2249 *) Fix possible deadlock when decoding public keys.
2252 Changes between 0.9.8w and 0.9.8x [10 May 2012]
2254 *) Sanity check record length before skipping explicit IV in DTLS
2257 Thanks to Codenomicon for discovering this issue using Fuzz-o-Matic
2258 fuzzing as a service testing platform.
2262 *) Initialise tkeylen properly when encrypting CMS messages.
2263 Thanks to Solar Designer of Openwall for reporting this issue.
2266 Changes between 0.9.8v and 0.9.8w [23 Apr 2012]
2268 *) The fix for CVE-2012-2110 did not take into account that the
2269 'len' argument to BUF_MEM_grow and BUF_MEM_grow_clean is an
2270 int in OpenSSL 0.9.8, making it still vulnerable. Fix by
2271 rejecting negative len parameter. (CVE-2012-2131)
2272 [Tomas Hoger <thoger@redhat.com>]
2274 Changes between 0.9.8u and 0.9.8v [19 Apr 2012]
2276 *) Check for potentially exploitable overflows in asn1_d2i_read_bio
2277 BUF_mem_grow and BUF_mem_grow_clean. Refuse attempts to shrink buffer
2278 in CRYPTO_realloc_clean.
2280 Thanks to Tavis Ormandy, Google Security Team, for discovering this
2281 issue and to Adam Langley <agl@chromium.org> for fixing it.
2283 [Adam Langley (Google), Tavis Ormandy, Google Security Team]
2285 Changes between 0.9.8t and 0.9.8u [12 Mar 2012]
2287 *) Fix MMA (Bleichenbacher's attack on PKCS #1 v1.5 RSA padding) weakness
2288 in CMS and PKCS7 code. When RSA decryption fails use a random key for
2289 content decryption and always return the same error. Note: this attack
2290 needs on average 2^20 messages so it only affects automated senders. The
2291 old behaviour can be reenabled in the CMS code by setting the
2292 CMS_DEBUG_DECRYPT flag: this is useful for debugging and testing where
2293 an MMA defence is not necessary.
2294 Thanks to Ivan Nestlerode <inestlerode@us.ibm.com> for discovering
2295 this issue. (CVE-2012-0884)
2298 *) Fix CVE-2011-4619: make sure we really are receiving a
2299 client hello before rejecting multiple SGC restarts. Thanks to
2300 Ivan Nestlerode <inestlerode@us.ibm.com> for discovering this bug.
2303 Changes between 0.9.8s and 0.9.8t [18 Jan 2012]
2305 *) Fix for DTLS DoS issue introduced by fix for CVE-2011-4109.
2306 Thanks to Antonio Martin, Enterprise Secure Access Research and
2307 Development, Cisco Systems, Inc. for discovering this bug and
2308 preparing a fix. (CVE-2012-0050)
2311 Changes between 0.9.8r and 0.9.8s [4 Jan 2012]
2313 *) Nadhem Alfardan and Kenny Paterson have discovered an extension
2314 of the Vaudenay padding oracle attack on CBC mode encryption
2315 which enables an efficient plaintext recovery attack against
2316 the OpenSSL implementation of DTLS. Their attack exploits timing
2317 differences arising during decryption processing. A research
2318 paper describing this attack can be found at:
2319 http://www.isg.rhul.ac.uk/~kp/dtls.pdf
2320 Thanks go to Nadhem Alfardan and Kenny Paterson of the Information
2321 Security Group at Royal Holloway, University of London
2322 (www.isg.rhul.ac.uk) for discovering this flaw and to Robin Seggelmann
2323 <seggelmann@fh-muenster.de> and Michael Tuexen <tuexen@fh-muenster.de>
2324 for preparing the fix. (CVE-2011-4108)
2325 [Robin Seggelmann, Michael Tuexen]
2327 *) Stop policy check failure freeing same buffer twice. (CVE-2011-4109)
2328 [Ben Laurie, Kasper <ekasper@google.com>]
2330 *) Clear bytes used for block padding of SSL 3.0 records.
2332 [Adam Langley (Google)]
2334 *) Only allow one SGC handshake restart for SSL/TLS. Thanks to George
2335 Kadianakis <desnacked@gmail.com> for discovering this issue and
2336 Adam Langley for preparing the fix. (CVE-2011-4619)
2337 [Adam Langley (Google)]
2339 *) Prevent malformed RFC3779 data triggering an assertion failure.
2340 Thanks to Andrew Chi, BBN Technologies, for discovering the flaw
2341 and Rob Austein <sra@hactrn.net> for fixing it. (CVE-2011-4577)
2342 [Rob Austein <sra@hactrn.net>]
2344 *) Fix ssl_ciph.c set-up race.
2345 [Adam Langley (Google)]
2347 *) Fix spurious failures in ecdsatest.c.
2348 [Emilia Käsper (Google)]
2350 *) Fix the BIO_f_buffer() implementation (which was mixing different
2351 interpretations of the '..._len' fields).
2352 [Adam Langley (Google)]
2354 *) Fix handling of BN_BLINDING: now BN_BLINDING_invert_ex (rather than
2355 BN_BLINDING_invert_ex) calls BN_BLINDING_update, ensuring that concurrent
2356 threads won't reuse the same blinding coefficients.
2358 This also avoids the need to obtain the CRYPTO_LOCK_RSA_BLINDING
2359 lock to call BN_BLINDING_invert_ex, and avoids one use of
2360 BN_BLINDING_update for each BN_BLINDING structure (previously,
2361 the last update always remained unused).
2362 [Emilia Käsper (Google)]
2364 *) Fix SSL memory handling for (EC)DH ciphersuites, in particular
2365 for multi-threaded use of ECDH.
2366 [Adam Langley (Google)]
2368 *) Fix x509_name_ex_d2i memory leak on bad inputs.
2371 *) Add protection against ECDSA timing attacks as mentioned in the paper
2372 by Billy Bob Brumley and Nicola Tuveri, see:
2374 http://eprint.iacr.org/2011/232.pdf
2376 [Billy Bob Brumley and Nicola Tuveri]
2378 Changes between 0.9.8q and 0.9.8r [8 Feb 2011]
2380 *) Fix parsing of OCSP stapling ClientHello extension. CVE-2011-0014
2381 [Neel Mehta, Adam Langley, Bodo Moeller (Google)]
2383 *) Fix bug in string printing code: if *any* escaping is enabled we must
2384 escape the escape character (backslash) or the resulting string is
2388 Changes between 0.9.8p and 0.9.8q [2 Dec 2010]
2390 *) Disable code workaround for ancient and obsolete Netscape browsers
2391 and servers: an attacker can use it in a ciphersuite downgrade attack.
2392 Thanks to Martin Rex for discovering this bug. CVE-2010-4180
2395 *) Fixed J-PAKE implementation error, originally discovered by
2396 Sebastien Martini, further info and confirmation from Stefan
2397 Arentz and Feng Hao. Note that this fix is a security fix. CVE-2010-4252
2400 Changes between 0.9.8o and 0.9.8p [16 Nov 2010]
2402 *) Fix extension code to avoid race conditions which can result in a buffer
2403 overrun vulnerability: resumed sessions must not be modified as they can
2404 be shared by multiple threads. CVE-2010-3864
2407 *) Fix for double free bug in ssl/s3_clnt.c CVE-2010-2939
2410 *) Don't reencode certificate when calculating signature: cache and use
2411 the original encoding instead. This makes signature verification of
2412 some broken encodings work correctly.
2415 *) ec2_GF2m_simple_mul bugfix: compute correct result if the output EC_POINT
2416 is also one of the inputs.
2417 [Emilia Käsper <emilia.kasper@esat.kuleuven.be> (Google)]
2419 *) Don't repeatedly append PBE algorithms to table if they already exist.
2420 Sort table on each new add. This effectively makes the table read only
2421 after all algorithms are added and subsequent calls to PKCS12_pbe_add
2425 Changes between 0.9.8n and 0.9.8o [01 Jun 2010]
2427 [NB: OpenSSL 0.9.8o and later 0.9.8 patch levels were released after
2430 *) Correct a typo in the CMS ASN1 module which can result in invalid memory
2431 access or freeing data twice (CVE-2010-0742)
2432 [Steve Henson, Ronald Moesbergen <intercommit@gmail.com>]
2434 *) Add SHA2 algorithms to SSL_library_init(). SHA2 is becoming far more
2435 common in certificates and some applications which only call
2436 SSL_library_init and not OpenSSL_add_all_algorithms() will fail.
2440 Reduce copying into .apps and .test in makevms.com
2441 Don't try to use blank CA certificate in CA.com
2442 Allow use of C files from original directories in maketests.com
2443 [Steven M. Schweda" <sms@antinode.info>]
2445 Changes between 0.9.8m and 0.9.8n [24 Mar 2010]
2447 *) When rejecting SSL/TLS records due to an incorrect version number, never
2448 update s->server with a new major version number. As of
2449 - OpenSSL 0.9.8m if 'short' is a 16-bit type,
2450 - OpenSSL 0.9.8f if 'short' is longer than 16 bits,
2451 the previous behavior could result in a read attempt at NULL when
2452 receiving specific incorrect SSL/TLS records once record payload
2453 protection is active. (CVE-2010-0740)
2454 [Bodo Moeller, Adam Langley <agl@chromium.org>]
2456 *) Fix for CVE-2010-0433 where some kerberos enabled versions of OpenSSL
2457 could be crashed if the relevant tables were not present (e.g. chrooted).
2458 [Tomas Hoger <thoger@redhat.com>]
2460 Changes between 0.9.8l and 0.9.8m [25 Feb 2010]
2462 *) Always check bn_wexpend() return values for failure. (CVE-2009-3245)
2463 [Martin Olsson, Neel Mehta]
2465 *) Fix X509_STORE locking: Every 'objs' access requires a lock (to
2466 accommodate for stack sorting, always a write lock!).
2469 *) On some versions of WIN32 Heap32Next is very slow. This can cause
2470 excessive delays in the RAND_poll(): over a minute. As a workaround
2471 include a time check in the inner Heap32Next loop too.
2474 *) The code that handled flushing of data in SSL/TLS originally used the
2475 BIO_CTRL_INFO ctrl to see if any data was pending first. This caused
2476 the problem outlined in PR#1949. The fix suggested there however can
2477 trigger problems with buggy BIO_CTRL_WPENDING (e.g. some versions
2478 of Apache). So instead simplify the code to flush unconditionally.
2479 This should be fine since flushing with no data to flush is a no op.
2482 *) Handle TLS versions 2.0 and later properly and correctly use the
2483 highest version of TLS/SSL supported. Although TLS >= 2.0 is some way
2484 off ancient servers have a habit of sticking around for a while...
2487 *) Modify compression code so it frees up structures without using the
2488 ex_data callbacks. This works around a problem where some applications
2489 call CRYPTO_cleanup_all_ex_data() before application exit (e.g. when
2490 restarting) then use compression (e.g. SSL with compression) later.
2491 This results in significant per-connection memory leaks and
2492 has caused some security issues including CVE-2008-1678 and
2496 *) Constify crypto/cast (i.e., <openssl/cast.h>): a CAST_KEY doesn't
2497 change when encrypting or decrypting.
2500 *) Add option SSL_OP_LEGACY_SERVER_CONNECT which will allow clients to
2501 connect and renegotiate with servers which do not support RI.
2502 Until RI is more widely deployed this option is enabled by default.
2505 *) Add "missing" ssl ctrls to clear options and mode.
2508 *) If client attempts to renegotiate and doesn't support RI respond with
2509 a no_renegotiation alert as required by RFC5746. Some renegotiating
2510 TLS clients will continue a connection gracefully when they receive
2511 the alert. Unfortunately OpenSSL mishandled this alert and would hang
2512 waiting for a server hello which it will never receive. Now we treat a
2513 received no_renegotiation alert as a fatal error. This is because
2514 applications requesting a renegotiation might well expect it to succeed
2515 and would have no code in place to handle the server denying it so the
2516 only safe thing to do is to terminate the connection.
2519 *) Add ctrl macro SSL_get_secure_renegotiation_support() which returns 1 if
2520 peer supports secure renegotiation and 0 otherwise. Print out peer
2521 renegotiation support in s_client/s_server.
2524 *) Replace the highly broken and deprecated SPKAC certification method with
2525 the updated NID creation version. This should correctly handle UTF8.
2528 *) Implement RFC5746. Re-enable renegotiation but require the extension
2529 as needed. Unfortunately, SSL3_FLAGS_ALLOW_UNSAFE_LEGACY_RENEGOTIATION
2530 turns out to be a bad idea. It has been replaced by
2531 SSL_OP_ALLOW_UNSAFE_LEGACY_RENEGOTIATION which can be set with
2532 SSL_CTX_set_options(). This is really not recommended unless you
2533 know what you are doing.
2534 [Eric Rescorla <ekr@networkresonance.com>, Ben Laurie, Steve Henson]
2536 *) Fixes to stateless session resumption handling. Use initial_ctx when
2537 issuing and attempting to decrypt tickets in case it has changed during
2538 servername handling. Use a non-zero length session ID when attempting
2539 stateless session resumption: this makes it possible to determine if
2540 a resumption has occurred immediately after receiving server hello
2541 (several places in OpenSSL subtly assume this) instead of later in
2545 *) The functions ENGINE_ctrl(), OPENSSL_isservice(),
2546 CMS_get1_RecipientRequest() and RAND_bytes() can return <=0 on error
2547 fixes for a few places where the return code is not checked
2549 [Julia Lawall <julia@diku.dk>]
2551 *) Add --strict-warnings option to Configure script to include devteam
2552 warnings in other configurations.
2555 *) Add support for --libdir option and LIBDIR variable in makefiles. This
2556 makes it possible to install openssl libraries in locations which
2557 have names other than "lib", for example "/usr/lib64" which some
2559 [Steve Henson, based on patch from Jeremy Utley]
2561 *) Don't allow the use of leading 0x80 in OIDs. This is a violation of
2562 X690 8.9.12 and can produce some misleading textual output of OIDs.
2563 [Steve Henson, reported by Dan Kaminsky]
2565 *) Delete MD2 from algorithm tables. This follows the recommendation in
2566 several standards that it is not used in new applications due to
2567 several cryptographic weaknesses. For binary compatibility reasons
2568 the MD2 API is still compiled in by default.
2571 *) Add compression id to {d2i,i2d}_SSL_SESSION so it is correctly saved
2575 *) Rename uni2asc and asc2uni functions to OPENSSL_uni2asc and
2576 OPENSSL_asc2uni conditionally on Netware platforms to avoid a name
2578 [Guenter <lists@gknw.net>]
2580 *) Fix the server certificate chain building code to use X509_verify_cert(),
2581 it used to have an ad-hoc builder which was unable to cope with anything
2582 other than a simple chain.
2583 [David Woodhouse <dwmw2@infradead.org>, Steve Henson]
2585 *) Don't check self signed certificate signatures in X509_verify_cert()
2586 by default (a flag can override this): it just wastes time without
2587 adding any security. As a useful side effect self signed root CAs
2588 with non-FIPS digests are now usable in FIPS mode.
2591 *) In dtls1_process_out_of_seq_message() the check if the current message
2592 is already buffered was missing. For every new message was memory
2593 allocated, allowing an attacker to perform an denial of service attack
2594 with sending out of seq handshake messages until there is no memory
2595 left. Additionally every future messege was buffered, even if the
2596 sequence number made no sense and would be part of another handshake.
2597 So only messages with sequence numbers less than 10 in advance will be
2598 buffered. (CVE-2009-1378)
2599 [Robin Seggelmann, discovered by Daniel Mentz]
2601 *) Records are buffered if they arrive with a future epoch to be
2602 processed after finishing the corresponding handshake. There is
2603 currently no limitation to this buffer allowing an attacker to perform
2604 a DOS attack with sending records with future epochs until there is no
2605 memory left. This patch adds the pqueue_size() function to determine
2606 the size of a buffer and limits the record buffer to 100 entries.
2608 [Robin Seggelmann, discovered by Daniel Mentz]
2610 *) Keep a copy of frag->msg_header.frag_len so it can be used after the
2611 parent structure is freed. (CVE-2009-1379)
2614 *) Handle non-blocking I/O properly in SSL_shutdown() call.
2615 [Darryl Miles <darryl-mailinglists@netbauds.net>]
2618 [Ilya O. <vrghost@gmail.com>]
2620 Changes between 0.9.8k and 0.9.8l [5 Nov 2009]
2622 *) Disable renegotiation completely - this fixes a severe security
2623 problem (CVE-2009-3555) at the cost of breaking all
2624 renegotiation. Renegotiation can be re-enabled by setting
2625 SSL3_FLAGS_ALLOW_UNSAFE_LEGACY_RENEGOTIATION in s3->flags at
2626 run-time. This is really not recommended unless you know what
2630 Changes between 0.9.8j and 0.9.8k [25 Mar 2009]
2632 *) Don't set val to NULL when freeing up structures, it is freed up by
2633 underlying code. If sizeof(void *) > sizeof(long) this can result in
2634 zeroing past the valid field. (CVE-2009-0789)
2635 [Paolo Ganci <Paolo.Ganci@AdNovum.CH>]
2637 *) Fix bug where return value of CMS_SignerInfo_verify_content() was not
2638 checked correctly. This would allow some invalid signed attributes to
2639 appear to verify correctly. (CVE-2009-0591)
2640 [Ivan Nestlerode <inestlerode@us.ibm.com>]
2642 *) Reject UniversalString and BMPString types with invalid lengths. This
2643 prevents a crash in ASN1_STRING_print_ex() which assumes the strings have
2644 a legal length. (CVE-2009-0590)
2647 *) Set S/MIME signing as the default purpose rather than setting it
2648 unconditionally. This allows applications to override it at the store
2652 *) Permit restricted recursion of ASN1 strings. This is needed in practice
2653 to handle some structures.
2656 *) Improve efficiency of mem_gets: don't search whole buffer each time
2658 [Jeremy Shapiro <jnshapir@us.ibm.com>]
2660 *) New -hex option for openssl rand.
2663 *) Print out UTF8String and NumericString when parsing ASN1.
2666 *) Support NumericString type for name components.
2669 *) Allow CC in the environment to override the automatically chosen
2670 compiler. Note that nothing is done to ensure flags work with the
2674 Changes between 0.9.8i and 0.9.8j [07 Jan 2009]
2676 *) Properly check EVP_VerifyFinal() and similar return values
2678 [Ben Laurie, Bodo Moeller, Google Security Team]
2680 *) Enable TLS extensions by default.
2683 *) Allow the CHIL engine to be loaded, whether the application is
2684 multithreaded or not. (This does not release the developer from the
2685 obligation to set up the dynamic locking callbacks.)
2686 [Sander Temme <sander@temme.net>]
2688 *) Use correct exit code if there is an error in dgst command.
2689 [Steve Henson; problem pointed out by Roland Dirlewanger]
2691 *) Tweak Configure so that you need to say "experimental-jpake" to enable
2692 JPAKE, and need to use -DOPENSSL_EXPERIMENTAL_JPAKE in applications.
2695 *) Add experimental JPAKE support, including demo authentication in
2696 s_client and s_server.
2699 *) Set the comparison function in v3_addr_canonize().
2700 [Rob Austein <sra@hactrn.net>]
2702 *) Add support for XMPP STARTTLS in s_client.
2703 [Philip Paeps <philip@freebsd.org>]
2705 *) Change the server-side SSL_OP_NETSCAPE_REUSE_CIPHER_CHANGE_BUG behavior
2706 to ensure that even with this option, only ciphersuites in the
2707 server's preference list will be accepted. (Note that the option
2708 applies only when resuming a session, so the earlier behavior was
2709 just about the algorithm choice for symmetric cryptography.)
2712 Changes between 0.9.8h and 0.9.8i [15 Sep 2008]
2714 *) Fix NULL pointer dereference if a DTLS server received
2715 ChangeCipherSpec as first record (CVE-2009-1386).
2718 *) Fix a state transition in s3_srvr.c and d1_srvr.c
2719 (was using SSL3_ST_CW_CLNT_HELLO_B, should be ..._ST_SW_SRVR_...).
2722 *) The fix in 0.9.8c that supposedly got rid of unsafe
2723 double-checked locking was incomplete for RSA blinding,
2724 addressing just one layer of what turns out to have been
2725 doubly unsafe triple-checked locking.
2727 So now fix this for real by retiring the MONT_HELPER macro
2728 in crypto/rsa/rsa_eay.c.
2730 [Bodo Moeller; problem pointed out by Marius Schilder]
2732 *) Various precautionary measures:
2734 - Avoid size_t integer overflow in HASH_UPDATE (md32_common.h).
2736 - Avoid a buffer overflow in d2i_SSL_SESSION() (ssl_asn1.c).
2737 (NB: This would require knowledge of the secret session ticket key
2738 to exploit, in which case you'd be SOL either way.)
2740 - Change bn_nist.c so that it will properly handle input BIGNUMs
2741 outside the expected range.
2743 - Enforce the 'num' check in BN_div() (bn_div.c) for non-BN_DEBUG
2746 [Neel Mehta, Bodo Moeller]
2748 *) Allow engines to be "soft loaded" - i.e. optionally don't die if
2749 the load fails. Useful for distros.
2750 [Ben Laurie and the FreeBSD team]
2752 *) Add support for Local Machine Keyset attribute in PKCS#12 files.
2755 *) Fix BN_GF2m_mod_arr() top-bit cleanup code.
2758 *) Expand ENGINE to support engine supplied SSL client certificate functions.
2760 This work was sponsored by Logica.
2763 *) Add CryptoAPI ENGINE to support use of RSA and DSA keys held in Windows
2764 keystores. Support for SSL/TLS client authentication too.
2765 Not compiled unless enable-capieng specified to Configure.
2767 This work was sponsored by Logica.
2770 *) Fix bug in X509_ATTRIBUTE creation: dont set attribute using
2771 ASN1_TYPE_set1 if MBSTRING flag set. This bug would crash certain
2772 attribute creation routines such as certifcate requests and PKCS#12
2776 Changes between 0.9.8g and 0.9.8h [28 May 2008]
2778 *) Fix flaw if 'Server Key exchange message' is omitted from a TLS
2779 handshake which could lead to a cilent crash as found using the
2780 Codenomicon TLS test suite (CVE-2008-1672)
2781 [Steve Henson, Mark Cox]
2783 *) Fix double free in TLS server name extensions which could lead to
2784 a remote crash found by Codenomicon TLS test suite (CVE-2008-0891)
2787 *) Clear error queue in SSL_CTX_use_certificate_chain_file()
2789 Clear the error queue to ensure that error entries left from
2790 older function calls do not interfere with the correct operation.
2791 [Lutz Jaenicke, Erik de Castro Lopo]
2793 *) Remove root CA certificates of commercial CAs:
2795 The OpenSSL project does not recommend any specific CA and does not
2796 have any policy with respect to including or excluding any CA.
2797 Therefore it does not make any sense to ship an arbitrary selection
2798 of root CA certificates with the OpenSSL software.
2801 *) RSA OAEP patches to fix two separate invalid memory reads.
2802 The first one involves inputs when 'lzero' is greater than
2803 'SHA_DIGEST_LENGTH' (it would read about SHA_DIGEST_LENGTH bytes
2804 before the beginning of from). The second one involves inputs where
2805 the 'db' section contains nothing but zeroes (there is a one-byte
2806 invalid read after the end of 'db').
2807 [Ivan Nestlerode <inestlerode@us.ibm.com>]
2809 *) Partial backport from 0.9.9-dev:
2811 Introduce bn_mul_mont (dedicated Montgomery multiplication
2812 procedure) as a candidate for BIGNUM assembler implementation.
2813 While 0.9.9-dev uses assembler for various architectures, only
2814 x86_64 is available by default here in the 0.9.8 branch, and
2815 32-bit x86 is available through a compile-time setting.
2817 To try the 32-bit x86 assembler implementation, use Configure
2818 option "enable-montasm" (which exists only for this backport).
2820 As "enable-montasm" for 32-bit x86 disclaims code stability
2821 anyway, in this constellation we activate additional code
2822 backported from 0.9.9-dev for further performance improvements,
2823 namely BN_from_montgomery_word. (To enable this otherwise,
2824 e.g. x86_64, try "-DMONT_FROM_WORD___NON_DEFAULT_0_9_8_BUILD".)
2826 [Andy Polyakov (backport partially by Bodo Moeller)]
2828 *) Add TLS session ticket callback. This allows an application to set
2829 TLS ticket cipher and HMAC keys rather than relying on hardcoded fixed
2830 values. This is useful for key rollover for example where several key
2831 sets may exist with different names.
2834 *) Reverse ENGINE-internal logic for caching default ENGINE handles.
2835 This was broken until now in 0.9.8 releases, such that the only way
2836 a registered ENGINE could be used (assuming it initialises
2837 successfully on the host) was to explicitly set it as the default
2838 for the relevant algorithms. This is in contradiction with 0.9.7
2839 behaviour and the documentation. With this fix, when an ENGINE is
2840 registered into a given algorithm's table of implementations, the
2841 'uptodate' flag is reset so that auto-discovery will be used next
2842 time a new context for that algorithm attempts to select an
2844 [Ian Lister (tweaked by Geoff Thorpe)]
2846 *) Backport of CMS code to OpenSSL 0.9.8. This differs from the 0.9.9
2847 implemention in the following ways:
2849 Lack of EVP_PKEY_ASN1_METHOD means algorithm parameters have to be
2852 Lack of BER streaming support means one pass streaming processing is
2853 only supported if data is detached: setting the streaming flag is
2854 ignored for embedded content.
2856 CMS support is disabled by default and must be explicitly enabled
2857 with the enable-cms configuration option.
2860 *) Update the GMP engine glue to do direct copies between BIGNUM and
2861 mpz_t when openssl and GMP use the same limb size. Otherwise the
2862 existing "conversion via a text string export" trick is still used.
2863 [Paul Sheer <paulsheer@gmail.com>]
2865 *) Zlib compression BIO. This is a filter BIO which compressed and
2866 uncompresses any data passed through it.
2869 *) Add AES_wrap_key() and AES_unwrap_key() functions to implement
2870 RFC3394 compatible AES key wrapping.
2873 *) Add utility functions to handle ASN1 structures. ASN1_STRING_set0():
2874 sets string data without copying. X509_ALGOR_set0() and
2875 X509_ALGOR_get0(): set and retrieve X509_ALGOR (AlgorithmIdentifier)
2876 data. Attribute function X509at_get0_data_by_OBJ(): retrieves data
2877 from an X509_ATTRIBUTE structure optionally checking it occurs only
2878 once. ASN1_TYPE_set1(): set and ASN1_TYPE structure copying supplied
2882 *) Fix BN flag handling in RSA_eay_mod_exp() and BN_MONT_CTX_set()
2883 to get the expected BN_FLG_CONSTTIME behavior.
2884 [Bodo Moeller (Google)]
2888 - fixed wrong usage of ioctlsocket() when build for LIBC BSD sockets
2889 - fixed do_tests.pl to run the test suite with CLIB builds too (CLIB_OPT)
2890 - added some more tests to do_tests.pl
2891 - fixed RunningProcess usage so that it works with newer LIBC NDKs too
2892 - removed usage of BN_LLONG for CLIB builds to avoid runtime dependency
2893 - added new Configure targets netware-clib-bsdsock, netware-clib-gcc,
2894 netware-clib-bsdsock-gcc, netware-libc-bsdsock-gcc
2895 - various changes to netware.pl to enable gcc-cross builds on Win32
2897 - changed crypto/bio/b_sock.c to work with macro functions (CLIB BSD)
2898 - various changes to fix missing prototype warnings
2899 - fixed x86nasm.pl to create correct asm files for NASM COFF output
2900 - added AES, WHIRLPOOL and CPUID assembler code to build files
2901 - added missing AES assembler make rules to mk1mf.pl
2902 - fixed order of includes in apps/ocsp.c so that e_os.h settings apply
2903 [Guenter Knauf <eflash@gmx.net>]
2905 *) Implement certificate status request TLS extension defined in RFC3546.
2906 A client can set the appropriate parameters and receive the encoded
2907 OCSP response via a callback. A server can query the supplied parameters
2908 and set the encoded OCSP response in the callback. Add simplified examples
2909 to s_client and s_server.
2912 Changes between 0.9.8f and 0.9.8g [19 Oct 2007]
2914 *) Fix various bugs:
2915 + Binary incompatibility of ssl_ctx_st structure
2916 + DTLS interoperation with non-compliant servers
2917 + Don't call get_session_cb() without proposed session
2918 + Fix ia64 assembler code
2919 [Andy Polyakov, Steve Henson]
2921 Changes between 0.9.8e and 0.9.8f [11 Oct 2007]
2923 *) DTLS Handshake overhaul. There were longstanding issues with
2924 OpenSSL DTLS implementation, which were making it impossible for
2925 RFC 4347 compliant client to communicate with OpenSSL server.
2926 Unfortunately just fixing these incompatibilities would "cut off"
2927 pre-0.9.8f clients. To allow for hassle free upgrade post-0.9.8e
2928 server keeps tolerating non RFC compliant syntax. The opposite is
2929 not true, 0.9.8f client can not communicate with earlier server.
2930 This update even addresses CVE-2007-4995.
2933 *) Changes to avoid need for function casts in OpenSSL: some compilers
2934 (gcc 4.2 and later) reject their use.
2935 [Kurt Roeckx <kurt@roeckx.be>, Peter Hartley <pdh@utter.chaos.org.uk>,
2938 *) Add RFC4507 support to OpenSSL. This includes the corrections in
2939 RFC4507bis. The encrypted ticket format is an encrypted encoded
2940 SSL_SESSION structure, that way new session features are automatically
2943 If a client application caches session in an SSL_SESSION structure
2944 support is transparent because tickets are now stored in the encoded
2947 The SSL_CTX structure automatically generates keys for ticket
2948 protection in servers so again support should be possible
2949 with no application modification.
2951 If a client or server wishes to disable RFC4507 support then the option
2952 SSL_OP_NO_TICKET can be set.
2954 Add a TLS extension debugging callback to allow the contents of any client
2955 or server extensions to be examined.
2957 This work was sponsored by Google.
2960 *) Add initial support for TLS extensions, specifically for the server_name
2961 extension so far. The SSL_SESSION, SSL_CTX, and SSL data structures now
2962 have new members for a host name. The SSL data structure has an
2963 additional member SSL_CTX *initial_ctx so that new sessions can be
2964 stored in that context to allow for session resumption, even after the
2965 SSL has been switched to a new SSL_CTX in reaction to a client's
2966 server_name extension.
2968 New functions (subject to change):
2970 SSL_get_servername()
2971 SSL_get_servername_type()
2974 New CTRL codes and macros (subject to change):
2976 SSL_CTRL_SET_TLSEXT_SERVERNAME_CB
2977 - SSL_CTX_set_tlsext_servername_callback()
2978 SSL_CTRL_SET_TLSEXT_SERVERNAME_ARG
2979 - SSL_CTX_set_tlsext_servername_arg()
2980 SSL_CTRL_SET_TLSEXT_HOSTNAME - SSL_set_tlsext_host_name()
2982 openssl s_client has a new '-servername ...' option.
2984 openssl s_server has new options '-servername_host ...', '-cert2 ...',
2985 '-key2 ...', '-servername_fatal' (subject to change). This allows
2986 testing the HostName extension for a specific single host name ('-cert'
2987 and '-key' remain fallbacks for handshakes without HostName
2988 negotiation). If the unrecognized_name alert has to be sent, this by
2989 default is a warning; it becomes fatal with the '-servername_fatal'
2992 [Peter Sylvester, Remy Allais, Christophe Renou, Steve Henson]
2994 *) Add AES and SSE2 assembly language support to VC++ build.
2997 *) Mitigate attack on final subtraction in Montgomery reduction.
3000 *) Fix crypto/ec/ec_mult.c to work properly with scalars of value 0
3001 (which previously caused an internal error).
3004 *) Squeeze another 10% out of IGE mode when in != out.
3007 *) AES IGE mode speedup.
3008 [Dean Gaudet (Google)]
3010 *) Add the Korean symmetric 128-bit cipher SEED (see
3011 http://www.kisa.or.kr/kisa/seed/jsp/seed_eng.jsp) and
3012 add SEED ciphersuites from RFC 4162:
3014 TLS_RSA_WITH_SEED_CBC_SHA = "SEED-SHA"
3015 TLS_DHE_DSS_WITH_SEED_CBC_SHA = "DHE-DSS-SEED-SHA"
3016 TLS_DHE_RSA_WITH_SEED_CBC_SHA = "DHE-RSA-SEED-SHA"
3017 TLS_DH_anon_WITH_SEED_CBC_SHA = "ADH-SEED-SHA"
3019 To minimize changes between patchlevels in the OpenSSL 0.9.8
3020 series, SEED remains excluded from compilation unless OpenSSL
3021 is configured with 'enable-seed'.
3022 [KISA, Bodo Moeller]
3024 *) Mitigate branch prediction attacks, which can be practical if a
3025 single processor is shared, allowing a spy process to extract
3026 information. For detailed background information, see
3027 http://eprint.iacr.org/2007/039 (O. Aciicmez, S. Gueron,
3028 J.-P. Seifert, "New Branch Prediction Vulnerabilities in OpenSSL
3029 and Necessary Software Countermeasures"). The core of the change
3030 are new versions BN_div_no_branch() and
3031 BN_mod_inverse_no_branch() of BN_div() and BN_mod_inverse(),
3032 respectively, which are slower, but avoid the security-relevant
3033 conditional branches. These are automatically called by BN_div()
3034 and BN_mod_inverse() if the flag BN_FLG_CONSTTIME is set for one
3035 of the input BIGNUMs. Also, BN_is_bit_set() has been changed to
3036 remove a conditional branch.
3038 BN_FLG_CONSTTIME is the new name for the previous
3039 BN_FLG_EXP_CONSTTIME flag, since it now affects more than just
3040 modular exponentiation. (Since OpenSSL 0.9.7h, setting this flag
3041 in the exponent causes BN_mod_exp_mont() to use the alternative
3042 implementation in BN_mod_exp_mont_consttime().) The old name
3043 remains as a deprecated alias.
3045 Similary, RSA_FLAG_NO_EXP_CONSTTIME is replaced by a more general
3046 RSA_FLAG_NO_CONSTTIME flag since the RSA implementation now uses
3047 constant-time implementations for more than just exponentiation.
3048 Here too the old name is kept as a deprecated alias.
3050 BN_BLINDING_new() will now use BN_dup() for the modulus so that
3051 the BN_BLINDING structure gets an independent copy of the
3052 modulus. This means that the previous "BIGNUM *m" argument to
3053 BN_BLINDING_new() and to BN_BLINDING_create_param() now
3054 essentially becomes "const BIGNUM *m", although we can't actually
3055 change this in the header file before 0.9.9. It allows
3056 RSA_setup_blinding() to use BN_with_flags() on the modulus to
3057 enable BN_FLG_CONSTTIME.
3059 [Matthew D Wood (Intel Corp)]
3061 *) In the SSL/TLS server implementation, be strict about session ID
3062 context matching (which matters if an application uses a single
3063 external cache for different purposes). Previously,
3064 out-of-context reuse was forbidden only if SSL_VERIFY_PEER was
3065 set. This did ensure strict client verification, but meant that,
3066 with applications using a single external cache for quite
3067 different requirements, clients could circumvent ciphersuite
3068 restrictions for a given session ID context by starting a session
3069 in a different context.
3072 *) Include "!eNULL" in SSL_DEFAULT_CIPHER_LIST to make sure that
3073 a ciphersuite string such as "DEFAULT:RSA" cannot enable
3074 authentication-only ciphersuites.
3077 *) Update the SSL_get_shared_ciphers() fix CVE-2006-3738 which was
3078 not complete and could lead to a possible single byte overflow
3079 (CVE-2007-5135) [Ben Laurie]
3081 Changes between 0.9.8d and 0.9.8e [23 Feb 2007]
3083 *) Since AES128 and AES256 (and similarly Camellia128 and
3084 Camellia256) share a single mask bit in the logic of
3085 ssl/ssl_ciph.c, the code for masking out disabled ciphers needs a
3086 kludge to work properly if AES128 is available and AES256 isn't
3087 (or if Camellia128 is available and Camellia256 isn't).
3090 *) Fix the BIT STRING encoding generated by crypto/ec/ec_asn1.c
3091 (within i2d_ECPrivateKey, i2d_ECPKParameters, i2d_ECParameters):
3092 When a point or a seed is encoded in a BIT STRING, we need to
3093 prevent the removal of trailing zero bits to get the proper DER
3094 encoding. (By default, crypto/asn1/a_bitstr.c assumes the case
3095 of a NamedBitList, for which trailing 0 bits need to be removed.)
3098 *) Have SSL/TLS server implementation tolerate "mismatched" record
3099 protocol version while receiving ClientHello even if the
3100 ClientHello is fragmented. (The server can't insist on the
3101 particular protocol version it has chosen before the ServerHello
3102 message has informed the client about his choice.)
3105 *) Add RFC 3779 support.
3106 [Rob Austein for ARIN, Ben Laurie]
3108 *) Load error codes if they are not already present instead of using a
3109 static variable. This allows them to be cleanly unloaded and reloaded.
3110 Improve header file function name parsing.
3113 *) extend SMTP and IMAP protocol emulation in s_client to use EHLO
3114 or CAPABILITY handshake as required by RFCs.
3117 Changes between 0.9.8c and 0.9.8d [28 Sep 2006]
3119 *) Introduce limits to prevent malicious keys being able to
3120 cause a denial of service. (CVE-2006-2940)
3121 [Steve Henson, Bodo Moeller]
3123 *) Fix ASN.1 parsing of certain invalid structures that can result
3124 in a denial of service. (CVE-2006-2937) [Steve Henson]
3126 *) Fix buffer overflow in SSL_get_shared_ciphers() function.
3127 (CVE-2006-3738) [Tavis Ormandy and Will Drewry, Google Security Team]
3129 *) Fix SSL client code which could crash if connecting to a
3130 malicious SSLv2 server. (CVE-2006-4343)
3131 [Tavis Ormandy and Will Drewry, Google Security Team]
3133 *) Since 0.9.8b, ciphersuite strings naming explicit ciphersuites
3134 match only those. Before that, "AES256-SHA" would be interpreted
3135 as a pattern and match "AES128-SHA" too (since AES128-SHA got
3136 the same strength classification in 0.9.7h) as we currently only
3137 have a single AES bit in the ciphersuite description bitmap.
3138 That change, however, also applied to ciphersuite strings such as
3139 "RC4-MD5" that intentionally matched multiple ciphersuites --
3140 namely, SSL 2.0 ciphersuites in addition to the more common ones
3141 from SSL 3.0/TLS 1.0.
3143 So we change the selection algorithm again: Naming an explicit
3144 ciphersuite selects this one ciphersuite, and any other similar
3145 ciphersuite (same bitmap) from *other* protocol versions.
3146 Thus, "RC4-MD5" again will properly select both the SSL 2.0
3147 ciphersuite and the SSL 3.0/TLS 1.0 ciphersuite.
3149 Since SSL 2.0 does not have any ciphersuites for which the
3150 128/256 bit distinction would be relevant, this works for now.
3151 The proper fix will be to use different bits for AES128 and
3152 AES256, which would have avoided the problems from the beginning;
3153 however, bits are scarce, so we can only do this in a new release
3154 (not just a patchlevel) when we can change the SSL_CIPHER
3155 definition to split the single 'unsigned long mask' bitmap into
3156 multiple values to extend the available space.
3160 Changes between 0.9.8b and 0.9.8c [05 Sep 2006]
3162 *) Avoid PKCS #1 v1.5 signature attack discovered by Daniel Bleichenbacher
3163 (CVE-2006-4339) [Ben Laurie and Google Security Team]
3165 *) Add AES IGE and biIGE modes.
3168 *) Change the Unix randomness entropy gathering to use poll() when
3169 possible instead of select(), since the latter has some
3170 undesirable limitations.
3171 [Darryl Miles via Richard Levitte and Bodo Moeller]
3173 *) Disable "ECCdraft" ciphersuites more thoroughly. Now special
3174 treatment in ssl/ssl_ciph.s makes sure that these ciphersuites
3175 cannot be implicitly activated as part of, e.g., the "AES" alias.
3176 However, please upgrade to OpenSSL 0.9.9[-dev] for
3177 non-experimental use of the ECC ciphersuites to get TLS extension
3178 support, which is required for curve and point format negotiation
3179 to avoid potential handshake problems.
3182 *) Disable rogue ciphersuites:
3184 - SSLv2 0x08 0x00 0x80 ("RC4-64-MD5")
3185 - SSLv3/TLSv1 0x00 0x61 ("EXP1024-RC2-CBC-MD5")
3186 - SSLv3/TLSv1 0x00 0x60 ("EXP1024-RC4-MD5")
3188 The latter two were purportedly from
3189 draft-ietf-tls-56-bit-ciphersuites-0[01].txt, but do not really
3192 Also deactivate the remaining ciphersuites from
3193 draft-ietf-tls-56-bit-ciphersuites-01.txt. These are just as
3194 unofficial, and the ID has long expired.
3197 *) Fix RSA blinding Heisenbug (problems sometimes occured on
3198 dual-core machines) and other potential thread-safety issues.
3201 *) Add the symmetric cipher Camellia (128-bit, 192-bit, 256-bit key
3202 versions), which is now available for royalty-free use
3203 (see http://info.isl.ntt.co.jp/crypt/eng/info/chiteki.html).
3204 Also, add Camellia TLS ciphersuites from RFC 4132.
3206 To minimize changes between patchlevels in the OpenSSL 0.9.8
3207 series, Camellia remains excluded from compilation unless OpenSSL
3208 is configured with 'enable-camellia'.
3211 *) Disable the padding bug check when compression is in use. The padding
3212 bug check assumes the first packet is of even length, this is not
3213 necessarily true if compresssion is enabled and can result in false
3214 positives causing handshake failure. The actual bug test is ancient
3215 code so it is hoped that implementations will either have fixed it by
3216 now or any which still have the bug do not support compression.
3219 Changes between 0.9.8a and 0.9.8b [04 May 2006]
3221 *) When applying a cipher rule check to see if string match is an explicit
3222 cipher suite and only match that one cipher suite if it is.
3225 *) Link in manifests for VC++ if needed.
3226 [Austin Ziegler <halostatue@gmail.com>]
3228 *) Update support for ECC-based TLS ciphersuites according to
3229 draft-ietf-tls-ecc-12.txt with proposed changes (but without
3230 TLS extensions, which are supported starting with the 0.9.9
3231 branch, not in the OpenSSL 0.9.8 branch).
3234 *) New functions EVP_CIPHER_CTX_new() and EVP_CIPHER_CTX_free() to support
3235 opaque EVP_CIPHER_CTX handling.
3238 *) Fixes and enhancements to zlib compression code. We now only use
3239 "zlib1.dll" and use the default __cdecl calling convention on Win32
3240 to conform with the standards mentioned here:
3241 http://www.zlib.net/DLL_FAQ.txt
3242 Static zlib linking now works on Windows and the new --with-zlib-include
3243 --with-zlib-lib options to Configure can be used to supply the location
3244 of the headers and library. Gracefully handle case where zlib library
3248 *) Several fixes and enhancements to the OID generation code. The old code
3249 sometimes allowed invalid OIDs (1.X for X >= 40 for example), couldn't
3250 handle numbers larger than ULONG_MAX, truncated printing and had a
3251 non standard OBJ_obj2txt() behaviour.
3254 *) Add support for building of engines under engine/ as shared libraries
3255 under VC++ build system.
3258 *) Corrected the numerous bugs in the Win32 path splitter in DSO.
3259 Hopefully, we will not see any false combination of paths any more.
3262 Changes between 0.9.8 and 0.9.8a [11 Oct 2005]
3264 *) Remove the functionality of SSL_OP_MSIE_SSLV2_RSA_PADDING
3265 (part of SSL_OP_ALL). This option used to disable the
3266 countermeasure against man-in-the-middle protocol-version
3267 rollback in the SSL 2.0 server implementation, which is a bad
3268 idea. (CVE-2005-2969)
3270 [Bodo Moeller; problem pointed out by Yutaka Oiwa (Research Center
3271 for Information Security, National Institute of Advanced Industrial
3272 Science and Technology [AIST], Japan)]
3274 *) Add two function to clear and return the verify parameter flags.
3277 *) Keep cipherlists sorted in the source instead of sorting them at
3278 runtime, thus removing the need for a lock.
3281 *) Avoid some small subgroup attacks in Diffie-Hellman.
3282 [Nick Mathewson and Ben Laurie]
3284 *) Add functions for well-known primes.
3287 *) Extended Windows CE support.
3288 [Satoshi Nakamura and Andy Polyakov]
3290 *) Initialize SSL_METHOD structures at compile time instead of during
3291 runtime, thus removing the need for a lock.
3294 *) Make PKCS7_decrypt() work even if no certificate is supplied by
3295 attempting to decrypt each encrypted key in turn. Add support to
3299 Changes between 0.9.7h and 0.9.8 [05 Jul 2005]
3301 [NB: OpenSSL 0.9.7i and later 0.9.7 patch levels were released after
3304 *) Add libcrypto.pc and libssl.pc for those who feel they need them.
3307 *) Change CA.sh and CA.pl so they don't bundle the CSR and the private
3308 key into the same file any more.
3311 *) Add initial support for Win64, both IA64 and AMD64/x64 flavors.
3314 *) Add -utf8 command line and config file option to 'ca'.
3315 [Stefan <stf@udoma.org]
3317 *) Removed the macro des_crypt(), as it seems to conflict with some
3318 libraries. Use DES_crypt().
3321 *) Correct naming of the 'chil' and '4758cca' ENGINEs. This
3322 involves renaming the source and generated shared-libs for
3323 both. The engines will accept the corrected or legacy ids
3324 ('ncipher' and '4758_cca' respectively) when binding. NB,
3325 this only applies when building 'shared'.
3326 [Corinna Vinschen <vinschen@redhat.com> and Geoff Thorpe]
3328 *) Add attribute functions to EVP_PKEY structure. Modify
3329 PKCS12_create() to recognize a CSP name attribute and
3330 use it. Make -CSP option work again in pkcs12 utility.
3333 *) Add new functionality to the bn blinding code:
3334 - automatic re-creation of the BN_BLINDING parameters after
3335 a fixed number of uses (currently 32)
3336 - add new function for parameter creation
3337 - introduce flags to control the update behaviour of the
3338 BN_BLINDING parameters
3339 - hide BN_BLINDING structure
3340 Add a second BN_BLINDING slot to the RSA structure to improve
3341 performance when a single RSA object is shared among several
3345 *) Add support for DTLS.
3346 [Nagendra Modadugu <nagendra@cs.stanford.edu> and Ben Laurie]
3348 *) Add support for DER encoded private keys (SSL_FILETYPE_ASN1)
3349 to SSL_CTX_use_PrivateKey_file() and SSL_use_PrivateKey_file()
3352 *) Remove buggy and incomplete DH cert support from
3353 ssl/ssl_rsa.c and ssl/s3_both.c
3356 *) Use SHA-1 instead of MD5 as the default digest algorithm for
3357 the apps/openssl applications.
3360 *) Compile clean with "-Wall -Wmissing-prototypes
3361 -Wstrict-prototypes -Wmissing-declarations -Werror". Currently
3362 DEBUG_SAFESTACK must also be set.
3365 *) Change ./Configure so that certain algorithms can be disabled by default.
3366 The new counterpiece to "no-xxx" is "enable-xxx".
3368 The patented RC5 and MDC2 algorithms will now be disabled unless
3369 "enable-rc5" and "enable-mdc2", respectively, are specified.
3371 (IDEA remains enabled despite being patented. This is because IDEA
3372 is frequently required for interoperability, and there is no license
3373 fee for non-commercial use. As before, "no-idea" can be used to
3374 avoid this algorithm.)
3378 *) Add processing of proxy certificates (see RFC 3820). This work was
3379 sponsored by KTH (The Royal Institute of Technology in Stockholm) and
3380 EGEE (Enabling Grids for E-science in Europe).
3383 *) RC4 performance overhaul on modern architectures/implementations, such
3384 as Intel P4, IA-64 and AMD64.
3387 *) New utility extract-section.pl. This can be used specify an alternative
3388 section number in a pod file instead of having to treat each file as
3389 a separate case in Makefile. This can be done by adding two lines to the
3392 =for comment openssl_section:XXX
3394 The blank line is mandatory.
3398 *) New arguments -certform, -keyform and -pass for s_client and s_server
3399 to allow alternative format key and certificate files and passphrase
3403 *) New structure X509_VERIFY_PARAM which combines current verify parameters,
3404 update associated structures and add various utility functions.
3406 Add new policy related verify parameters, include policy checking in
3407 standard verify code. Enhance 'smime' application with extra parameters
3408 to support policy checking and print out.
3411 *) Add a new engine to support VIA PadLock ACE extensions in the VIA C3
3412 Nehemiah processors. These extensions support AES encryption in hardware
3413 as well as RNG (though RNG support is currently disabled).
3414 [Michal Ludvig <michal@logix.cz>, with help from Andy Polyakov]
3416 *) Deprecate BN_[get|set]_params() functions (they were ignored internally).
3419 *) New FIPS 180-2 algorithms, SHA-224/-256/-384/-512 are implemented.
3420 [Andy Polyakov and a number of other people]
3422 *) Improved PowerPC platform support. Most notably BIGNUM assembler
3423 implementation contributed by IBM.
3424 [Suresh Chari, Peter Waltenberg, Andy Polyakov]
3426 *) The new 'RSA_generate_key_ex' function now takes a BIGNUM for the public
3427 exponent rather than 'unsigned long'. There is a corresponding change to
3428 the new 'rsa_keygen' element of the RSA_METHOD structure.
3429 [Jelte Jansen, Geoff Thorpe]
3431 *) Functionality for creating the initial serial number file is now
3432 moved from CA.pl to the 'ca' utility with a new option -create_serial.
3434 (Before OpenSSL 0.9.7e, CA.pl used to initialize the serial
3435 number file to 1, which is bound to cause problems. To avoid
3436 the problems while respecting compatibility between different 0.9.7
3437 patchlevels, 0.9.7e employed 'openssl x509 -next_serial' in
3438 CA.pl for serial number initialization. With the new release 0.9.8,
3439 we can fix the problem directly in the 'ca' utility.)
3442 *) Reduced header interdepencies by declaring more opaque objects in
3443 ossl_typ.h. As a consequence, including some headers (eg. engine.h) will
3444 give fewer recursive includes, which could break lazy source code - so
3445 this change is covered by the OPENSSL_NO_DEPRECATED symbol. As always,
3446 developers should define this symbol when building and using openssl to
3447 ensure they track the recommended behaviour, interfaces, [etc], but
3448 backwards-compatible behaviour prevails when this isn't defined.
3451 *) New function X509_POLICY_NODE_print() which prints out policy nodes.
3454 *) Add new EVP function EVP_CIPHER_CTX_rand_key and associated functionality.
3455 This will generate a random key of the appropriate length based on the
3456 cipher context. The EVP_CIPHER can provide its own random key generation
3457 routine to support keys of a specific form. This is used in the des and
3458 3des routines to generate a key of the correct parity. Update S/MIME
3459 code to use new functions and hence generate correct parity DES keys.
3460 Add EVP_CHECK_DES_KEY #define to return an error if the key is not
3461 valid (weak or incorrect parity).
3464 *) Add a local set of CRLs that can be used by X509_verify_cert() as well
3465 as looking them up. This is useful when the verified structure may contain
3466 CRLs, for example PKCS#7 signedData. Modify PKCS7_verify() to use any CRLs
3467 present unless the new PKCS7_NO_CRL flag is asserted.
3470 *) Extend ASN1 oid configuration module. It now additionally accepts the
3473 shortName = some long name, 1.2.3.4
3476 *) Reimplemented the BN_CTX implementation. There is now no more static
3477 limitation on the number of variables it can handle nor the depth of the
3478 "stack" handling for BN_CTX_start()/BN_CTX_end() pairs. The stack
3479 information can now expand as required, and rather than having a single
3480 static array of bignums, BN_CTX now uses a linked-list of such arrays
3481 allowing it to expand on demand whilst maintaining the usefulness of
3482 BN_CTX's "bundling".
3485 *) Add a missing BN_CTX parameter to the 'rsa_mod_exp' callback in RSA_METHOD
3486 to allow all RSA operations to function using a single BN_CTX.
3489 *) Preliminary support for certificate policy evaluation and checking. This
3490 is initially intended to pass the tests outlined in "Conformance Testing
3491 of Relying Party Client Certificate Path Processing Logic" v1.07.
3494 *) bn_dup_expand() has been deprecated, it was introduced in 0.9.7 and
3495 remained unused and not that useful. A variety of other little bignum
3496 tweaks and fixes have also been made continuing on from the audit (see
3500 *) Constify all or almost all d2i, c2i, s2i and r2i functions, along with
3501 associated ASN1, EVP and SSL functions and old ASN1 macros.
3504 *) BN_zero() only needs to set 'top' and 'neg' to zero for correct results,
3505 and this should never fail. So the return value from the use of
3506 BN_set_word() (which can fail due to needless expansion) is now deprecated;
3507 if OPENSSL_NO_DEPRECATED is defined, BN_zero() is a void macro.
3510 *) BN_CTX_get() should return zero-valued bignums, providing the same
3511 initialised value as BN_new().
3512 [Geoff Thorpe, suggested by Ulf Möller]
3514 *) Support for inhibitAnyPolicy certificate extension.
3517 *) An audit of the BIGNUM code is underway, for which debugging code is
3518 enabled when BN_DEBUG is defined. This makes stricter enforcements on what
3519 is considered valid when processing BIGNUMs, and causes execution to
3520 assert() when a problem is discovered. If BN_DEBUG_RAND is defined,
3521 further steps are taken to deliberately pollute unused data in BIGNUM
3522 structures to try and expose faulty code further on. For now, openssl will
3523 (in its default mode of operation) continue to tolerate the inconsistent
3524 forms that it has tolerated in the past, but authors and packagers should
3525 consider trying openssl and their own applications when compiled with
3526 these debugging symbols defined. It will help highlight potential bugs in
3527 their own code, and will improve the test coverage for OpenSSL itself. At
3528 some point, these tighter rules will become openssl's default to improve
3529 maintainability, though the assert()s and other overheads will remain only
3530 in debugging configurations. See bn.h for more details.
3531 [Geoff Thorpe, Nils Larsch, Ulf Möller]
3533 *) BN_CTX_init() has been deprecated, as BN_CTX is an opaque structure
3534 that can only be obtained through BN_CTX_new() (which implicitly
3535 initialises it). The presence of this function only made it possible
3536 to overwrite an existing structure (and cause memory leaks).
3539 *) Because of the callback-based approach for implementing LHASH as a
3540 template type, lh_insert() adds opaque objects to hash-tables and
3541 lh_doall() or lh_doall_arg() are typically used with a destructor callback
3542 to clean up those corresponding objects before destroying the hash table
3543 (and losing the object pointers). So some over-zealous constifications in
3544 LHASH have been relaxed so that lh_insert() does not take (nor store) the
3545 objects as "const" and the lh_doall[_arg] callback wrappers are not
3546 prototyped to have "const" restrictions on the object pointers they are
3547 given (and so aren't required to cast them away any more).
3550 *) The tmdiff.h API was so ugly and minimal that our own timing utility
3551 (speed) prefers to use its own implementation. The two implementations
3552 haven't been consolidated as yet (volunteers?) but the tmdiff API has had
3553 its object type properly exposed (MS_TM) instead of casting to/from "char
3554 *". This may still change yet if someone realises MS_TM and "ms_time_***"
3555 aren't necessarily the greatest nomenclatures - but this is what was used
3556 internally to the implementation so I've used that for now.
3559 *) Ensure that deprecated functions do not get compiled when
3560 OPENSSL_NO_DEPRECATED is defined. Some "openssl" subcommands and a few of
3561 the self-tests were still using deprecated key-generation functions so
3562 these have been updated also.
3565 *) Reorganise PKCS#7 code to separate the digest location functionality
3566 into PKCS7_find_digest(), digest addition into PKCS7_bio_add_digest().
3567 New function PKCS7_set_digest() to set the digest type for PKCS#7
3568 digestedData type. Add additional code to correctly generate the
3569 digestedData type and add support for this type in PKCS7 initialization
3573 *) New function PKCS7_set0_type_other() this initializes a PKCS7
3574 structure of type "other".
3577 *) Fix prime generation loop in crypto/bn/bn_prime.pl by making
3578 sure the loop does correctly stop and breaking ("division by zero")
3579 modulus operations are not performed. The (pre-generated) prime
3580 table crypto/bn/bn_prime.h was already correct, but it could not be
3581 re-generated on some platforms because of the "division by zero"
3582 situation in the script.
3583 [Ralf S. Engelschall]
3585 *) Update support for ECC-based TLS ciphersuites according to
3586 draft-ietf-tls-ecc-03.txt: the KDF1 key derivation function with
3587 SHA-1 now is only used for "small" curves (where the
3588 representation of a field element takes up to 24 bytes); for
3589 larger curves, the field element resulting from ECDH is directly
3590 used as premaster secret.
3591 [Douglas Stebila (Sun Microsystems Laboratories)]
3593 *) Add code for kP+lQ timings to crypto/ec/ectest.c, and add SEC2
3594 curve secp160r1 to the tests.
3595 [Douglas Stebila (Sun Microsystems Laboratories)]
3597 *) Add the possibility to load symbols globally with DSO.
3598 [Götz Babin-Ebell <babin-ebell@trustcenter.de> via Richard Levitte]
3600 *) Add the functions ERR_set_mark() and ERR_pop_to_mark() for better
3601 control of the error stack.
3604 *) Add support for STORE in ENGINE.
3607 *) Add the STORE type. The intention is to provide a common interface
3608 to certificate and key stores, be they simple file-based stores, or
3609 HSM-type store, or LDAP stores, or...
3610 NOTE: The code is currently UNTESTED and isn't really used anywhere.
3613 *) Add a generic structure called OPENSSL_ITEM. This can be used to
3614 pass a list of arguments to any function as well as provide a way
3615 for a function to pass data back to the caller.
3618 *) Add the functions BUF_strndup() and BUF_memdup(). BUF_strndup()
3619 works like BUF_strdup() but can be used to duplicate a portion of
3620 a string. The copy gets NUL-terminated. BUF_memdup() duplicates
3624 *) Add the function sk_find_ex() which works like sk_find(), but will
3625 return an index to an element even if an exact match couldn't be
3626 found. The index is guaranteed to point at the element where the
3627 searched-for key would be inserted to preserve sorting order.
3630 *) Add the function OBJ_bsearch_ex() which works like OBJ_bsearch() but
3631 takes an extra flags argument for optional functionality. Currently,
3632 the following flags are defined:
3634 OBJ_BSEARCH_VALUE_ON_NOMATCH
3635 This one gets OBJ_bsearch_ex() to return a pointer to the first
3636 element where the comparing function returns a negative or zero
3639 OBJ_BSEARCH_FIRST_VALUE_ON_MATCH
3640 This one gets OBJ_bsearch_ex() to return a pointer to the first
3641 element where the comparing function returns zero. This is useful
3642 if there are more than one element where the comparing function
3646 *) Make it possible to create self-signed certificates with 'openssl ca'
3647 in such a way that the self-signed certificate becomes part of the
3648 CA database and uses the same mechanisms for serial number generation
3649 as all other certificate signing. The new flag '-selfsign' enables
3650 this functionality. Adapt CA.sh and CA.pl.in.
3653 *) Add functionality to check the public key of a certificate request
3654 against a given private. This is useful to check that a certificate
3655 request can be signed by that key (self-signing).
3658 *) Make it possible to have multiple active certificates with the same
3659 subject in the CA index file. This is done only if the keyword
3660 'unique_subject' is set to 'no' in the main CA section (default
3661 if 'CA_default') of the configuration file. The value is saved
3662 with the database itself in a separate index attribute file,
3663 named like the index file with '.attr' appended to the name.
3666 *) Generate muti valued AVAs using '+' notation in config files for
3670 *) Support for nameConstraints certificate extension.
3673 *) Support for policyConstraints certificate extension.
3676 *) Support for policyMappings certificate extension.
3679 *) Make sure the default DSA_METHOD implementation only uses its
3680 dsa_mod_exp() and/or bn_mod_exp() handlers if they are non-NULL,
3681 and change its own handlers to be NULL so as to remove unnecessary
3682 indirection. This lets alternative implementations fallback to the
3683 default implementation more easily.
3686 *) Support for directoryName in GeneralName related extensions
3690 *) Make it possible to link applications using Makefile.shared.
3691 Make that possible even when linking against static libraries!
3694 *) Support for single pass processing for S/MIME signing. This now
3695 means that S/MIME signing can be done from a pipe, in addition
3696 cleartext signing (multipart/signed type) is effectively streaming
3697 and the signed data does not need to be all held in memory.
3699 This is done with a new flag PKCS7_STREAM. When this flag is set
3700 PKCS7_sign() only initializes the PKCS7 structure and the actual signing
3701 is done after the data is output (and digests calculated) in
3702 SMIME_write_PKCS7().
3705 *) Add full support for -rpath/-R, both in shared libraries and
3706 applications, at least on the platforms where it's known how
3710 *) In crypto/ec/ec_mult.c, implement fast point multiplication with
3711 precomputation, based on wNAF splitting: EC_GROUP_precompute_mult()
3712 will now compute a table of multiples of the generator that
3713 makes subsequent invocations of EC_POINTs_mul() or EC_POINT_mul()
3714 faster (notably in the case of a single point multiplication,
3715 scalar * generator).
3716 [Nils Larsch, Bodo Moeller]
3718 *) IPv6 support for certificate extensions. The various extensions
3719 which use the IP:a.b.c.d can now take IPv6 addresses using the
3720 formats of RFC1884 2.2 . IPv6 addresses are now also displayed
3724 *) Added an ENGINE that implements RSA by performing private key
3725 exponentiations with the GMP library. The conversions to and from
3726 GMP's mpz_t format aren't optimised nor are any montgomery forms
3727 cached, and on x86 it appears OpenSSL's own performance has caught up.
3728 However there are likely to be other architectures where GMP could
3729 provide a boost. This ENGINE is not built in by default, but it can be
3730 specified at Configure time and should be accompanied by the necessary
3731 linker additions, eg;
3732 ./config -DOPENSSL_USE_GMP -lgmp
3735 *) "openssl engine" will not display ENGINE/DSO load failure errors when
3736 testing availability of engines with "-t" - the old behaviour is
3737 produced by increasing the feature's verbosity with "-tt".
projects / openssl.git / blob