5 Changes between 0.9.4 and 0.9.5 [xx XXX 2000]
7 *) Change the ASN1 types so they are typedefs by default. Before
8 almost all types were #define'd to ASN1_STRING which was causing
9 STACK_OF() problems: you couldn't declare STACK_OF(ASN1_UTF8STRING)
13 *) Change names of new functions to the new get1/get0 naming
14 convention: After 'get1', the caller owns a reference count
15 and has to call ..._free; 'get0' returns a pointer to some
16 data structure without incrementing reference counters.
17 (Some of the existing 'get' functions increment a reference
19 Similarly, 'set1' and 'add1' functions increase reference
20 counters or duplicate objects.
23 *) Allow for the possibility of temp RSA key generation failure:
24 the code used to assume it always worked and crashed on failure.
27 *) Fix potential buffer overrun problem in BIO_printf().
28 [Ulf Möller, using public domain code by Patrick Powell; problem
29 pointed out by David Sacerdote <das33@cornell.edu>]
31 *) Support EGD <http://www.lothar.com/tech/crypto/>. New functions
32 RAND_egd() and RAND_status(). In the command line application,
33 the EGD socket can be specified like a seed file using RANDFILE
37 *) Allow the string CERTIFICATE to be tolerated in PKCS#7 structures.
38 Some CAs (e.g. Verisign) distribute certificates in this form.
41 *) Remove the SSL_ALLOW_ADH compile option and set the default cipher
42 list to exclude them. This means that no special compilation option
43 is needed to use anonymous DH: it just needs to be included in the
47 *) Change the EVP_MD_CTX_type macro so its meaning consistent with
48 EVP_MD_type. The old functionality is available in a new macro called
49 EVP_MD_md(). Change code that uses it and update docs.
52 *) ..._ctrl functions now have corresponding ..._callback_ctrl functions
53 where the 'void *' argument is replaced by a function pointer argument.
54 Previously 'void *' was abused to point to functions, which works on
55 many platforms, but is not correct. As these functions are usually
56 called by macros defined in OpenSSL header files, most source code
57 should work without changes.
60 *) <openssl/opensslconf.h> (which is created by Configure) now contains
61 sections with information on -D... compiler switches used for
62 compiling the library so that applications can see them. To enable
63 one of these sections, a pre-processor symbol OPENSSL_..._DEFINES
64 must be defined. E.g.,
65 #define OPENSSL_ALGORITHM_DEFINES
66 #include <openssl/opensslconf.h>
67 defines all pertinent NO_<algo> symbols, such as NO_IDEA, NO_RSA, etc.
68 [Richard Levitte, Ulf and Bodo Möller]
70 *) Bugfix: Tolerate fragmentation and interleaving in the SSL 3/TLS
74 *) Change the 'other' type in certificate aux info to a STACK_OF
75 X509_ALGOR. Although not an AlgorithmIdentifier as such it has
76 the required ASN1 format: arbitrary types determined by an OID.
79 *) Add some PEM_write_X509_REQ_NEW() functions and a command line
80 argument to 'req'. This is not because the function is newer or
81 better than others it just uses the work 'NEW' in the certificate
82 request header lines. Some software needs this.
85 *) Reorganise password command line arguments: now passwords can be
86 obtained from various sources. Delete the PEM_cb function and make
87 it the default behaviour: i.e. if the callback is NULL and the
88 usrdata argument is not NULL interpret it as a null terminated pass
89 phrase. If usrdata and the callback are NULL then the pass phrase
90 is prompted for as usual.
93 *) Add support for the Compaq Atalla crypto accelerator. If it is installed,
94 the support is automatically enabled. The resulting binaries will
95 autodetect the card and use it if present.
96 [Ben Laurie and Compaq Inc.]
98 *) Work around for Netscape hang bug. This sends certificate request
99 and server done in one record. Since this is perfectly legal in the
100 SSL/TLS protocol it isn't a "bug" option and is on by default. See
101 the bugs/SSLv3 entry for more info.
104 *) HP-UX tune-up: new unified configs, HP C compiler bug workaround.
107 *) Add -rand argument to smime and pkcs12 applications and read/write
111 *) New 'passwd' tool for crypt(3) and apr1 password hashes.
114 *) Add command line password options to the remaining applications.
117 *) Bug fix for BN_div_recp() for numerators with an even number of
121 *) More tests in bntest.c, and changed test_bn output.
124 *) ./config recognizes MacOS X now.
127 *) Bug fix for BN_div() when the first words of num and divsor are
128 equal (it gave wrong results if (rem=(n1-q*d0)&BN_MASK2) < d0).
131 *) Add support for various broken PKCS#8 formats, and command line
132 options to produce them.
135 *) New functions BN_CTX_start(), BN_CTX_get() and BT_CTX_end() to
136 get temporary BIGNUMs from a BN_CTX.
139 *) Correct return values in BN_mod_exp_mont() and BN_mod_exp2_mont()
143 *) Change the SSLeay_add_all_*() functions to OpenSSL_add_all_*() and
144 include a #define from the old name to the new. The original intent
145 was that statically linked binaries could for example just call
146 SSLeay_add_all_ciphers() to just add ciphers to the table and not
147 link with digests. This never worked becayse SSLeay_add_all_digests()
148 and SSLeay_add_all_ciphers() were in the same source file so calling
149 one would link with the other. They are now in separate source files.
152 *) Add a new -notext option to 'ca' and a -pubkey option to 'spkac'.
155 *) Use a less unusual form of the Miller-Rabin primality test (it used
156 a binary algorithm for exponentiation integrated into the Miller-Rabin
157 loop, our standard modexp algorithms are faster).
160 *) Support for the EBCDIC character set completed.
161 [Martin Kraemer <Martin.Kraemer@Mch.SNI.De>]
163 *) Source code cleanups: use const where appropriate, eliminate casts,
164 use void * instead of char * in lhash.
167 *) Bugfix: ssl3_send_server_key_exchange was not restartable
168 (the state was not changed to SSL3_ST_SW_KEY_EXCH_B, and because of
169 this the server could overwrite ephemeral keys that the client
173 *) Turn DSA_is_prime into a macro that calls BN_is_prime,
174 using 50 iterations of the Rabin-Miller test.
176 DSA_generate_parameters now uses BN_is_prime_fasttest (with 50
177 iterations of the Rabin-Miller test as required by the appendix
178 to FIPS PUB 186[-1]) instead of DSA_is_prime.
179 As BN_is_prime_fasttest includes trial division, DSA parameter
180 generation becomes much faster.
182 This implies a change for the callback functions in DSA_is_prime
183 and DSA_generate_parameters: The callback function is called once
184 for each positive witness in the Rabin-Miller test, not just
185 occasionally in the inner loop; and the parameters to the
186 callback function now provide an iteration count for the outer
187 loop rather than for the current invocation of the inner loop.
188 DSA_generate_parameters additionally can call the callback
189 function with an 'iteration count' of -1, meaning that a
190 candidate has passed the trial division test (when q is generated
191 from an application-provided seed, trial division is skipped).
194 *) New function BN_is_prime_fasttest that optionally does trial
195 division before starting the Rabin-Miller test and has
196 an additional BN_CTX * argument (whereas BN_is_prime always
197 has to allocate at least one BN_CTX).
198 'callback(1, -1, cb_arg)' is called when a number has passed the
199 trial division stage.
202 *) Fix for bug in CRL encoding. The validity dates weren't being handled
206 *) New -pkcs12 option to CA.pl script to write out a PKCS#12 file.
209 *) New function BN_pseudo_rand().
212 *) Clean up BN_mod_mul_montgomery(): replace the broken (and unreadable)
213 bignum version of BN_from_montgomery() with the working code from
214 SSLeay 0.9.0 (the word based version is faster anyway), and clean up
218 *) Avoid a race condition in s2_clnt.c (function get_server_hello) that
219 made it impossible to use the same SSL_SESSION data structure in
220 SSL2 clients in multiple threads.
223 *) The return value of RAND_load_file() no longer counts bytes obtained
224 by stat(). RAND_load_file(..., -1) is new and uses the complete file
225 to seed the PRNG (previously an explicit byte count was required).
226 [Ulf Möller, Bodo Möller]
228 *) Clean up CRYPTO_EX_DATA functions, some of these didn't have prototypes
229 used (char *) instead of (void *) and had casts all over the place.
232 *) Make BN_generate_prime() return NULL on error if ret!=NULL.
235 *) Retain source code compatibility for BN_prime_checks macro:
236 BN_is_prime(..., BN_prime_checks, ...) now uses
237 BN_prime_checks_for_size to determine the appropriate number of
238 Rabin-Miller iterations.
241 *) Diffie-Hellman uses "safe" primes: DH_check() return code renamed to
242 DH_CHECK_P_NOT_SAFE_PRIME.
243 (Check if this is true? OpenPGP calls them "strong".)
246 *) Merge the functionality of "dh" and "gendh" programs into a new program
247 "dhparam". The old programs are retained for now but will handle DH keys
248 (instead of parameters) in future.
251 *) Make the ciphers, s_server and s_client programs check the return values
252 when a new cipher list is set.
255 *) Enhance the SSL/TLS cipher mechanism to correctly handle the TLS 56bit
256 ciphers. Before when the 56bit ciphers were enabled the sorting was
259 The syntax for the cipher sorting has been extended to support sorting by
260 cipher-strength (using the strength_bits hard coded in the tables).
261 The new command is "@STRENGTH" (see also doc/apps/ciphers.pod).
263 Fix a bug in the cipher-command parser: when supplying a cipher command
264 string with an "undefined" symbol (neither command nor alphanumeric
265 [A-Za-z0-9], ssl_set_cipher_list used to hang in an endless loop. Now
268 Due to the strength-sorting extension, the code of the
269 ssl_create_cipher_list() function was completely rearranged. I hope that
270 the readability was also increased :-)
271 [Lutz Jaenicke <Lutz.Jaenicke@aet.TU-Cottbus.DE>]
273 *) Minor change to 'x509' utility. The -CAcreateserial option now uses 1
274 for the first serial number and places 2 in the serial number file. This
275 avoids problems when the root CA is created with serial number zero and
276 the first user certificate has the same issuer name and serial number
280 *) Fixes to X509_ATTRIBUTE utilities, change the 'req' program so it uses
281 the new code. Add documentation for this stuff.
284 *) Changes to X509_ATTRIBUTE utilities. These have been renamed from
285 X509_*() to X509at_*() on the grounds that they don't handle X509
286 structures and behave in an analagous way to the X509v3 functions:
287 they shouldn't be called directly but wrapper functions should be used
290 So we also now have some wrapper functions that call the X509at functions
291 when passed certificate requests. (TO DO: similar things can be done with
292 PKCS#7 signed and unsigned attributes, PKCS#12 attributes and a few other
293 things. Some of these need some d2i or i2d and print functionality
294 because they handle more complex structures.)
297 *) Add missing #ifndefs that caused missing symbols when building libssl
298 as a shared library without RSA. Use #ifndef NO_SSL2 instead of
300 [Kris Kennaway <kris@hub.freebsd.org>, modified by Ulf Möller]
302 *) Precautions against using the PRNG uninitialized: RAND_bytes() now
303 has a return value which indicates the quality of the random data
304 (1 = ok, 0 = not seeded). Also an error is recorded on the thread's
305 error queue. New function RAND_pseudo_bytes() generates output that is
306 guaranteed to be unique but not unpredictable. RAND_add is like
307 RAND_seed, but takes an extra argument for an entropy estimate
308 (RAND_seed always assumes full entropy).
311 *) Do more iterations of Rabin-Miller probable prime test (specifically,
312 3 for 1024-bit primes, 6 for 512-bit primes, 12 for 256-bit primes
313 instead of only 2 for all lengths; see BN_prime_checks_for_size definition
314 in crypto/bn/bn_prime.c for the complete table). This guarantees a
315 false-positive rate of at most 2^-80 for random input.
318 *) Rewrite ssl3_read_n (ssl/s3_pkt.c) avoiding a couple of bugs.
321 *) New function X509_CTX_rget_chain() (renamed to X509_CTX_get1_chain
322 in the 0.9.5 release), this returns the chain
323 from an X509_CTX structure with a dup of the stack and all
324 the X509 reference counts upped: so the stack will exist
325 after X509_CTX_cleanup() has been called. Modify pkcs12.c
328 Also make SSL_SESSION_print() print out the verify return
332 *) Add manpage for the pkcs12 command. Also change the default
333 behaviour so MAC iteration counts are used unless the new
334 -nomaciter option is used. This improves file security and
335 only older versions of MSIE (4.0 for example) need it.
338 *) Honor the no-xxx Configure options when creating .DEF files.
341 *) Add PKCS#10 attributes to field table: challengePassword,
342 unstructuredName and unstructuredAddress. These are taken from
343 draft PKCS#9 v2.0 but are compatible with v1.2 provided no
344 international characters are used.
346 More changes to X509_ATTRIBUTE code: allow the setting of types
347 based on strings. Remove the 'loc' parameter when adding
348 attributes because these will be a SET OF encoding which is sorted
352 *) Initial changes to the 'req' utility to allow request generation
353 automation. This will allow an application to just generate a template
354 file containing all the field values and have req construct the
357 Initial support for X509_ATTRIBUTE handling. Stacks of these are
358 used all over the place including certificate requests and PKCS#7
359 structures. They are currently handled manually where necessary with
360 some primitive wrappers for PKCS#7. The new functions behave in a
361 manner analogous to the X509 extension functions: they allow
362 attributes to be looked up by NID and added.
364 Later something similar to the X509V3 code would be desirable to
365 automatically handle the encoding, decoding and printing of the
366 more complex types. The string types like challengePassword can
367 be handled by the string table functions.
369 Also modified the multi byte string table handling. Now there is
370 a 'global mask' which masks out certain types. The table itself
371 can use the flag STABLE_NO_MASK to ignore the mask setting: this
372 is useful when for example there is only one permissible type
373 (as in countryName) and using the mask might result in no valid
377 *) Clean up 'Finished' handling, and add functions SSL_get_finished and
378 SSL_get_peer_finished to allow applications to obtain the latest
379 Finished messages sent to the peer or expected from the peer,
380 respectively. (SSL_get_peer_finished is usually the Finished message
381 actually received from the peer, otherwise the protocol will be aborted.)
383 As the Finished message are message digests of the complete handshake
384 (with a total of 192 bits for TLS 1.0 and more for SSL 3.0), they can
385 be used for external authentication procedures when the authentication
386 provided by SSL/TLS is not desired or is not enough.
389 *) Enhanced support for Alpha Linux is added. Now ./config checks if
390 the host supports BWX extension and if Compaq C is present on the
391 $PATH. Just exploiting of the BWX extension results in 20-30%
392 performance kick for some algorithms, e.g. DES and RC4 to mention
393 a couple. Compaq C in turn generates ~20% faster code for MD5 and
397 *) Add support for MS "fast SGC". This is arguably a violation of the
398 SSL3/TLS protocol. Netscape SGC does two handshakes: the first with
399 weak crypto and after checking the certificate is SGC a second one
400 with strong crypto. MS SGC stops the first handshake after receiving
401 the server certificate message and sends a second client hello. Since
402 a server will typically do all the time consuming operations before
403 expecting any further messages from the client (server key exchange
404 is the most expensive) there is little difference between the two.
406 To get OpenSSL to support MS SGC we have to permit a second client
407 hello message after we have sent server done. In addition we have to
408 reset the MAC if we do get this second client hello.
411 *) Add a function 'd2i_AutoPrivateKey()' this will automatically decide
412 if a DER encoded private key is RSA or DSA traditional format. Changed
413 d2i_PrivateKey_bio() to use it. This is only needed for the "traditional"
414 format DER encoded private key. Newer code should use PKCS#8 format which
415 has the key type encoded in the ASN1 structure. Added DER private key
416 support to pkcs8 application.
419 *) SSL 3/TLS 1 servers now don't request certificates when an anonymous
420 ciphersuites has been selected (as required by the SSL 3/TLS 1
421 specifications). Exception: When SSL_VERIFY_FAIL_IF_NO_PEER_CERT
422 is set, we interpret this as a request to violate the specification
423 (the worst that can happen is a handshake failure, and 'correct'
424 behaviour would result in a handshake failure anyway).
427 *) In SSL_CTX_add_session, take into account that there might be multiple
428 SSL_SESSION structures with the same session ID (e.g. when two threads
429 concurrently obtain them from an external cache).
430 The internal cache can handle only one SSL_SESSION with a given ID,
431 so if there's a conflict, we now throw out the old one to achieve
435 *) Add OIDs for idea and blowfish in CBC mode. This will allow both
436 to be used in PKCS#5 v2.0 and S/MIME. Also add checking to
437 some routines that use cipher OIDs: some ciphers do not have OIDs
438 defined and so they cannot be used for S/MIME and PKCS#5 v2.0 for
442 *) Simplify the trust setting structure and code. Now we just have
443 two sequences of OIDs for trusted and rejected settings. These will
444 typically have values the same as the extended key usage extension
445 and any application specific purposes.
447 The trust checking code now has a default behaviour: it will just
448 check for an object with the same NID as the passed id. Functions can
449 be provided to override either the default behaviour or the behaviour
450 for a given id. SSL client, server and email already have functions
451 in place for compatibility: they check the NID and also return "trusted"
452 if the certificate is self signed.
455 *) Add d2i,i2d bio/fp functions for PrivateKey: these convert the
456 traditional format into an EVP_PKEY structure.
459 *) Add a password callback function PEM_cb() which either prompts for
460 a password if usr_data is NULL or otherwise assumes it is a null
461 terminated password. Allow passwords to be passed on command line
462 environment or config files in a few more utilities.
465 *) Add a bunch of DER and PEM functions to handle PKCS#8 format private
466 keys. Add some short names for PKCS#8 PBE algorithms and allow them
467 to be specified on the command line for the pkcs8 and pkcs12 utilities.
468 Update documentation.
471 *) Support for ASN1 "NULL" type. This could be handled before by using
472 ASN1_TYPE but there wasn't any function that would try to read a NULL
473 and produce an error if it couldn't. For compatibility we also have
474 ASN1_NULL_new() and ASN1_NULL_free() functions but these are faked and
475 don't allocate anything because they don't need to.
478 *) Initial support for MacOS is now provided. Examine INSTALL.MacOS
480 [Andy Polyakov, Roy Woods <roy@centicsystems.ca>]
482 *) Rebuild of the memory allocation routines used by OpenSSL code and
483 possibly others as well. The purpose is to make an interface that
484 provide hooks so anyone can build a separate set of allocation and
485 deallocation routines to be used by OpenSSL, for example memory
486 pool implementations, or something else, which was previously hard
487 since Malloc(), Realloc() and Free() were defined as macros having
488 the values malloc, realloc and free, respectively (except for Win32
489 compilations). The same is provided for memory debugging code.
490 OpenSSL already comes with functionality to find memory leaks, but
491 this gives people a chance to debug other memory problems.
493 With these changes, a new set of functions and macros have appeared:
495 CRYPTO_set_mem_debug_functions() [F]
496 CRYPTO_get_mem_debug_functions() [F]
497 CRYPTO_dbg_set_options() [F]
498 CRYPTO_dbg_get_options() [F]
499 CRYPTO_malloc_debug_init() [M]
501 The memory debug functions are NULL by default, unless the library
502 is compiled with CRYPTO_MDEBUG or friends is defined. If someone
503 wants to debug memory anyway, CRYPTO_malloc_debug_init() (which
504 gives the standard debugging functions that come with OpenSSL) or
505 CRYPTO_set_mem_debug_functions() (tells OpenSSL to use functions
506 provided by the library user) must be used. When the standard
507 debugging functions are used, CRYPTO_dbg_set_options can be used to
508 request additional information:
509 CRYPTO_dbg_set_options(V_CYRPTO_MDEBUG_xxx) corresponds to setting
510 the CRYPTO_MDEBUG_xxx macro when compiling the library.
512 Also, things like CRYPTO_set_mem_functions will always give the
513 expected result (the new set of functions is used for allocation
514 and deallocation) at all times, regardless of platform and compiler
517 To finish it up, some functions that were never use in any other
518 way than through macros have a new API and new semantic:
524 All macros of value have retained their old syntax.
525 [Richard Levitte and Bodo Moeller]
527 *) Some S/MIME fixes. The OID for SMIMECapabilities was wrong, the
528 ordering of SMIMECapabilities wasn't in "strength order" and there
529 was a missing NULL in the AlgorithmIdentifier for the SHA1 signature
533 *) Some ASN1 types with illegal zero length encoding (INTEGER,
534 ENUMERATED and OBJECT IDENTIFIER) choked the ASN1 routines.
535 [Frans Heymans <fheymans@isaserver.be>, modified by Steve Henson]
537 *) Merge in my S/MIME library for OpenSSL. This provides a simple
538 S/MIME API on top of the PKCS#7 code, a MIME parser (with enough
539 functionality to handle multipart/signed properly) and a utility
540 called 'smime' to call all this stuff. This is based on code I
541 originally wrote for Celo who have kindly allowed it to be
545 *) Add variants des_set_key_checked and des_set_key_unchecked of
546 des_set_key (aka des_key_sched). Global variable des_check_key
547 decides which of these is called by des_set_key; this way
548 des_check_key behaves as it always did, but applications and
549 the library itself, which was buggy for des_check_key == 1,
550 have a cleaner way to pick the version they need.
553 *) New function PKCS12_newpass() which changes the password of a
557 *) Modify X509_TRUST and X509_PURPOSE so it also uses a static and
558 dynamic mix. In both cases the ids can be used as an index into the
559 table. Also modified the X509_TRUST_add() and X509_PURPOSE_add()
560 functions so they accept a list of the field values and the
561 application doesn't need to directly manipulate the X509_TRUST
565 *) Modify the ASN1_STRING_TABLE stuff so it also uses bsearch and doesn't
569 *) Modify the way the V3 extension code looks up extensions. This now
570 works in a similar way to the object code: we have some "standard"
571 extensions in a static table which is searched with OBJ_bsearch()
572 and the application can add dynamic ones if needed. The file
573 crypto/x509v3/ext_dat.h now has the info: this file needs to be
574 updated whenever a new extension is added to the core code and kept
575 in ext_nid order. There is a simple program 'tabtest.c' which checks
576 this. New extensions are not added too often so this file can readily
577 be maintained manually.
579 There are two big advantages in doing things this way. The extensions
580 can be looked up immediately and no longer need to be "added" using
581 X509V3_add_standard_extensions(): this function now does nothing.
582 [Side note: I get *lots* of email saying the extension code doesn't
583 work because people forget to call this function]
584 Also no dynamic allocation is done unless new extensions are added:
585 so if we don't add custom extensions there is no need to call
586 X509V3_EXT_cleanup().
589 *) Modify enc utility's salting as follows: make salting the default. Add a
590 magic header, so unsalted files fail gracefully instead of just decrypting
591 to garbage. This is because not salting is a big security hole, so people
592 should be discouraged from doing it.
595 *) Fixes and enhancements to the 'x509' utility. It allowed a message
596 digest to be passed on the command line but it only used this
597 parameter when signing a certificate. Modified so all relevant
598 operations are affected by the digest parameter including the
599 -fingerprint and -x509toreq options. Also -x509toreq choked if a
600 DSA key was used because it didn't fix the digest.
603 *) Initial certificate chain verify code. Currently tests the untrusted
604 certificates for consistency with the verify purpose (which is set
605 when the X509_STORE_CTX structure is set up) and checks the pathlength.
607 There is a NO_CHAIN_VERIFY compilation option to keep the old behaviour:
608 this is because it will reject chains with invalid extensions whereas
609 every previous version of OpenSSL and SSLeay made no checks at all.
611 Trust code: checks the root CA for the relevant trust settings. Trust
612 settings have an initial value consistent with the verify purpose: e.g.
613 if the verify purpose is for SSL client use it expects the CA to be
614 trusted for SSL client use. However the default value can be changed to
615 permit custom trust settings: one example of this would be to only trust
616 certificates from a specific "secure" set of CAs.
618 Also added X509_STORE_CTX_new() and X509_STORE_CTX_free() functions
619 which should be used for version portability: especially since the
620 verify structure is likely to change more often now.
622 SSL integration. Add purpose and trust to SSL_CTX and SSL and functions
623 to set them. If not set then assume SSL clients will verify SSL servers
626 Two new options to the verify program: -untrusted allows a set of
627 untrusted certificates to be passed in and -purpose which sets the
628 intended purpose of the certificate. If a purpose is set then the
629 new chain verify code is used to check extension consistency.
632 *) Support for the authority information access extension.
635 *) Modify RSA and DSA PEM read routines to transparently handle
636 PKCS#8 format private keys. New *_PUBKEY_* functions that handle
637 public keys in a format compatible with certificate
638 SubjectPublicKeyInfo structures. Unfortunately there were already
639 functions called *_PublicKey_* which used various odd formats so
640 these are retained for compatibility: however the DSA variants were
641 never in a public release so they have been deleted. Changed dsa/rsa
642 utilities to handle the new format: note no releases ever handled public
643 keys so we should be OK.
645 The primary motivation for this change is to avoid the same fiasco
646 that dogs private keys: there are several incompatible private key
647 formats some of which are standard and some OpenSSL specific and
648 require various evil hacks to allow partial transparent handling and
649 even then it doesn't work with DER formats. Given the option anything
650 other than PKCS#8 should be dumped: but the other formats have to
651 stay in the name of compatibility.
653 With public keys and the benefit of hindsight one standard format
654 is used which works with EVP_PKEY, RSA or DSA structures: though
655 it clearly returns an error if you try to read the wrong kind of key.
657 Added a -pubkey option to the 'x509' utility to output the public key.
658 Also rename the EVP_PKEY_get_*() to EVP_PKEY_rget_*()
659 (renamed to EVP_PKEY_get1_*() in the OpenSSL 0.9.5 release) and add
660 EVP_PKEY_rset_*() functions (renamed to EVP_PKEY_set1_*())
661 that do the same as the EVP_PKEY_assign_*() except they up the
662 reference count of the added key (they don't "swallow" the
666 *) Fixes to crypto/x509/by_file.c the code to read in certificates and
667 CRLs would fail if the file contained no certificates or no CRLs:
668 added a new function to read in both types and return the number
669 read: this means that if none are read it will be an error. The
670 DER versions of the certificate and CRL reader would always fail
671 because it isn't possible to mix certificates and CRLs in DER format
672 without choking one or the other routine. Changed this to just read
673 a certificate: this is the best we can do. Also modified the code
674 in apps/verify.c to take notice of return codes: it was previously
675 attempting to read in certificates from NULL pointers and ignoring
676 any errors: this is one reason why the cert and CRL reader seemed
677 to work. It doesn't check return codes from the default certificate
678 routines: these may well fail if the certificates aren't installed.
681 *) Code to support otherName option in GeneralName.
684 *) First update to verify code. Change the verify utility
685 so it warns if it is passed a self signed certificate:
686 for consistency with the normal behaviour. X509_verify
687 has been modified to it will now verify a self signed
688 certificate if *exactly* the same certificate appears
689 in the store: it was previously impossible to trust a
690 single self signed certificate. This means that:
691 openssl verify ss.pem
692 now gives a warning about a self signed certificate but
693 openssl verify -CAfile ss.pem ss.pem
697 *) For servers, store verify_result in SSL_SESSION data structure
698 (and add it to external session representation).
699 This is needed when client certificate verifications fails,
700 but an application-provided verification callback (set by
701 SSL_CTX_set_cert_verify_callback) allows accepting the session
702 anyway (i.e. leaves x509_store_ctx->error != X509_V_OK
703 but returns 1): When the session is reused, we have to set
704 ssl->verify_result to the appropriate error code to avoid
706 [Bodo Moeller, problem pointed out by Lutz Jaenicke]
708 *) Fix a bug in the new PKCS#7 code: it didn't consider the
709 case in PKCS7_dataInit() where the signed PKCS7 structure
710 didn't contain any existing data because it was being created.
711 [Po-Cheng Chen <pocheng@nst.com.tw>, slightly modified by Steve Henson]
713 *) Add a salt to the key derivation routines in enc.c. This
714 forms the first 8 bytes of the encrypted file. Also add a
715 -S option to allow a salt to be input on the command line.
718 *) New function X509_cmp(). Oddly enough there wasn't a function
719 to compare two certificates. We do this by working out the SHA1
720 hash and comparing that. X509_cmp() will be needed by the trust
724 *) SSL_get1_session() is like SSL_get_session(), but increments
725 the reference count in the SSL_SESSION returned.
726 [Geoff Thorpe <geoff@eu.c2.net>]
728 *) Fix for 'req': it was adding a null to request attributes.
729 Also change the X509_LOOKUP and X509_INFO code to handle
730 certificate auxiliary information.
733 *) Add support for 40 and 64 bit RC2 and RC4 algorithms: document
737 *) Add the possibility to add extra information to the memory leak
738 detecting output, to form tracebacks, showing from where each
739 allocation was originated: CRYPTO_push_info("constant string") adds
740 the string plus current file name and line number to a per-thread
741 stack, CRYPTO_pop_info() does the obvious, CRYPTO_remove_all_info()
742 is like calling CYRPTO_pop_info() until the stack is empty.
743 Also updated memory leak detection code to be multi-thread-safe.
746 *) Add options -text and -noout to pkcs7 utility and delete the
747 encryption options which never did anything. Update docs.
750 *) Add options to some of the utilities to allow the pass phrase
751 to be included on either the command line (not recommended on
752 OSes like Unix) or read from the environment. Update the
753 manpages and fix a few bugs.
756 *) Add a few manpages for some of the openssl commands.
759 *) Fix the -revoke option in ca. It was freeing up memory twice,
760 leaking and not finding already revoked certificates.
763 *) Extensive changes to support certificate auxiliary information.
764 This involves the use of X509_CERT_AUX structure and X509_AUX
765 functions. An X509_AUX function such as PEM_read_X509_AUX()
766 can still read in a certificate file in the usual way but it
767 will also read in any additional "auxiliary information". By
768 doing things this way a fair degree of compatibility can be
769 retained: existing certificates can have this information added
770 using the new 'x509' options.
772 Current auxiliary information includes an "alias" and some trust
773 settings. The trust settings will ultimately be used in enhanced
774 certificate chain verification routines: currently a certificate
775 can only be trusted if it is self signed and then it is trusted
779 *) Fix assembler for Alpha (tested only on DEC OSF not Linux or *BSD).
780 The problem was that one of the replacement routines had not been working
781 since SSLeay releases. For now the offending routine has been replaced
782 with non-optimised assembler. Even so, this now gives around 95%
783 performance improvement for 1024 bit RSA signs.
786 *) Hack to fix PKCS#7 decryption when used with some unorthodox RC2
787 handling. Most clients have the effective key size in bits equal to
788 the key length in bits: so a 40 bit RC2 key uses a 40 bit (5 byte) key.
789 A few however don't do this and instead use the size of the decrypted key
790 to determine the RC2 key length and the AlgorithmIdentifier to determine
791 the effective key length. In this case the effective key length can still
792 be 40 bits but the key length can be 168 bits for example. This is fixed
793 by manually forcing an RC2 key into the EVP_PKEY structure because the
794 EVP code can't currently handle unusual RC2 key sizes: it always assumes
795 the key length and effective key length are equal.
798 *) Add a bunch of functions that should simplify the creation of
799 X509_NAME structures. Now you should be able to do:
800 X509_NAME_add_entry_by_txt(nm, "CN", MBSTRING_ASC, "Steve", -1, -1, 0);
801 and have it automatically work out the correct field type and fill in
802 the structures. The more adventurous can try:
803 X509_NAME_add_entry_by_txt(nm, field, MBSTRING_UTF8, str, -1, -1, 0);
804 and it will (hopefully) work out the correct multibyte encoding.
807 *) Change the 'req' utility to use the new field handling and multibyte
808 copy routines. Before the DN field creation was handled in an ad hoc
809 way in req, ca, and x509 which was rather broken and didn't support
810 BMPStrings or UTF8Strings. Since some software doesn't implement
811 BMPStrings or UTF8Strings yet, they can be enabled using the config file
812 using the dirstring_type option. See the new comment in the default
813 openssl.cnf for more info.
816 *) Make crypto/rand/md_rand.c more robust:
817 - Assure unique random numbers after fork().
818 - Make sure that concurrent threads access the global counter and
819 md serializably so that we never lose entropy in them
820 or use exactly the same state in multiple threads.
821 Access to the large state is not always serializable because
822 the additional locking could be a performance killer, and
823 md should be large enough anyway.
826 *) New file apps/app_rand.c with commonly needed functionality
827 for handling the random seed file.
829 Use the random seed file in some applications that previously did not:
831 dsaparam -genkey (which also ignored its '-rand' option),
835 Except on systems with /dev/urandom, it is crucial to have a random
836 seed file at least for key creation, DSA signing, and for DH exchanges;
837 for RSA signatures we could do without one.
839 gendh and gendsa (unlike genrsa) used to read only the first byte
840 of each file listed in the '-rand' option. The function as previously
841 found in genrsa is now in app_rand.c and is used by all programs
842 that support '-rand'.
845 *) In RAND_write_file, use mode 0600 for creating files;
846 don't just chmod when it may be too late.
849 *) Report an error from X509_STORE_load_locations
850 when X509_LOOKUP_load_file or X509_LOOKUP_add_dir failed.
853 *) New function ASN1_mbstring_copy() this copies a string in either
854 ASCII, Unicode, Universal (4 bytes per character) or UTF8 format
855 into an ASN1_STRING type. A mask of permissible types is passed
856 and it chooses the "minimal" type to use or an error if not type
860 *) Add function equivalents to the various macros in asn1.h. The old
861 macros are retained with an M_ prefix. Code inside the library can
862 use the M_ macros. External code (including the openssl utility)
863 should *NOT* in order to be "shared library friendly".
866 *) Add various functions that can check a certificate's extensions
867 to see if it usable for various purposes such as SSL client,
868 server or S/MIME and CAs of these types. This is currently
869 VERY EXPERIMENTAL but will ultimately be used for certificate chain
870 verification. Also added a -purpose flag to x509 utility to
871 print out all the purposes.
874 *) Add a CRYPTO_EX_DATA to X509 certificate structure and associated
878 *) New X509V3_{X509,CRL,REVOKED}_get_d2i() functions. These will search
879 for, obtain and decode and extension and obtain its critical flag.
880 This allows all the necessary extension code to be handled in a
881 single function call.
884 *) RC4 tune-up featuring 30-40% performance improvement on most RISC
885 platforms. See crypto/rc4/rc4_enc.c for further details.
888 *) New -noout option to asn1parse. This causes no output to be produced
889 its main use is when combined with -strparse and -out to extract data
890 from a file (which may not be in ASN.1 format).
893 *) Fix for pkcs12 program. It was hashing an invalid certificate pointer
894 when producing the local key id.
895 [Richard Levitte <levitte@stacken.kth.se>]
897 *) New option -dhparam in s_server. This allows a DH parameter file to be
898 stated explicitly. If it is not stated then it tries the first server
899 certificate file. The previous behaviour hard coded the filename
903 *) Add -pubin and -pubout options to the rsa and dsa commands. These allow
904 a public key to be input or output. For example:
905 openssl rsa -in key.pem -pubout -out pubkey.pem
906 Also added necessary DSA public key functions to handle this.
909 *) Fix so PKCS7_dataVerify() doesn't crash if no certificates are contained
910 in the message. This was handled by allowing
911 X509_find_by_issuer_and_serial() to tolerate a NULL passed to it.
912 [Steve Henson, reported by Sampo Kellomaki <sampo@mail.neuronio.pt>]
914 *) Fix for bug in d2i_ASN1_bytes(): other ASN1 functions add an extra null
915 to the end of the strings whereas this didn't. This would cause problems
916 if strings read with d2i_ASN1_bytes() were later modified.
917 [Steve Henson, reported by Arne Ansper <arne@ats.cyber.ee>]
919 *) Fix for base64 decode bug. When a base64 bio reads only one line of
920 data and it contains EOF it will end up returning an error. This is
921 caused by input 46 bytes long. The cause is due to the way base64
922 BIOs find the start of base64 encoded data. They do this by trying a
923 trial decode on each line until they find one that works. When they
924 do a flag is set and it starts again knowing it can pass all the
925 data directly through the decoder. Unfortunately it doesn't reset
926 the context it uses. This means that if EOF is reached an attempt
927 is made to pass two EOFs through the context and this causes the
928 resulting error. This can also cause other problems as well. As is
929 usual with these problems it takes *ages* to find and the fix is
930 trivial: move one line.
931 [Steve Henson, reported by ian@uns.ns.ac.yu (Ivan Nejgebauer) ]
933 *) Ugly workaround to get s_client and s_server working under Windows. The
934 old code wouldn't work because it needed to select() on sockets and the
935 tty (for keypresses and to see if data could be written). Win32 only
936 supports select() on sockets so we select() with a 1s timeout on the
937 sockets and then see if any characters are waiting to be read, if none
938 are present then we retry, we also assume we can always write data to
939 the tty. This isn't nice because the code then blocks until we've
940 received a complete line of data and it is effectively polling the
941 keyboard at 1s intervals: however it's quite a bit better than not
942 working at all :-) A dedicated Windows application might handle this
943 with an event loop for example.
946 *) Enhance RSA_METHOD structure. Now there are two extra methods, rsa_sign
947 and rsa_verify. When the RSA_FLAGS_SIGN_VER option is set these functions
948 will be called when RSA_sign() and RSA_verify() are used. This is useful
949 if rsa_pub_dec() and rsa_priv_enc() equivalents are not available.
950 For this to work properly RSA_public_decrypt() and RSA_private_encrypt()
951 should *not* be used: RSA_sign() and RSA_verify() must be used instead.
952 This necessitated the support of an extra signature type NID_md5_sha1
953 for SSL signatures and modifications to the SSL library to use it instead
954 of calling RSA_public_decrypt() and RSA_private_encrypt().
957 *) Add new -verify -CAfile and -CApath options to the crl program, these
958 will lookup a CRL issuers certificate and verify the signature in a
959 similar way to the verify program. Tidy up the crl program so it
960 no longer accesses structures directly. Make the ASN1 CRL parsing a bit
961 less strict. It will now permit CRL extensions even if it is not
962 a V2 CRL: this will allow it to tolerate some broken CRLs.
965 *) Initialize all non-automatic variables each time one of the openssl
966 sub-programs is started (this is necessary as they may be started
967 multiple times from the "OpenSSL>" prompt).
968 [Lennart Bang, Bodo Moeller]
970 *) Preliminary compilation option RSA_NULL which disables RSA crypto without
971 removing all other RSA functionality (this is what NO_RSA does). This
972 is so (for example) those in the US can disable those operations covered
973 by the RSA patent while allowing storage and parsing of RSA keys and RSA
977 *) Non-copying interface to BIO pairs.
978 (still largely untested)
981 *) New function ANS1_tag2str() to convert an ASN1 tag to a descriptive
982 ASCII string. This was handled independently in various places before.
985 *) New functions UTF8_getc() and UTF8_putc() that parse and generate
986 UTF8 strings a character at a time.
989 *) Use client_version from client hello to select the protocol
990 (s23_srvr.c) and for RSA client key exchange verification
991 (s3_srvr.c), as required by the SSL 3.0/TLS 1.0 specifications.
994 *) Add various utility functions to handle SPKACs, these were previously
995 handled by poking round in the structure internals. Added new function
996 NETSCAPE_SPKI_print() to print out SPKAC and a new utility 'spkac' to
997 print, verify and generate SPKACs. Based on an original idea from
998 Massimiliano Pala <madwolf@comune.modena.it> but extensively modified.
1001 *) RIPEMD160 is operational on all platforms and is back in 'make test'.
1004 *) Allow the config file extension section to be overwritten on the
1005 command line. Based on an original idea from Massimiliano Pala
1006 <madwolf@comune.modena.it>. The new option is called -extensions
1007 and can be applied to ca, req and x509. Also -reqexts to override
1008 the request extensions in req and -crlexts to override the crl extensions
1012 *) Add new feature to the SPKAC handling in ca. Now you can include
1013 the same field multiple times by preceding it by "XXXX." for example:
1016 this is the same syntax as used in the req config file.
1019 *) Allow certificate extensions to be added to certificate requests. These
1020 are specified in a 'req_extensions' option of the req section of the
1021 config file. They can be printed out with the -text option to req but
1022 are otherwise ignored at present.
1025 *) Fix a horrible bug in enc_read() in crypto/evp/bio_enc.c: if the first
1026 data read consists of only the final block it would not decrypted because
1027 EVP_CipherUpdate() would correctly report zero bytes had been decrypted.
1028 A misplaced 'break' also meant the decrypted final block might not be
1029 copied until the next read.
1032 *) Initial support for DH_METHOD. Again based on RSA_METHOD. Also added
1033 a few extra parameters to the DH structure: these will be useful if
1034 for example we want the value of 'q' or implement X9.42 DH.
1037 *) Initial support for DSA_METHOD. This is based on the RSA_METHOD and
1038 provides hooks that allow the default DSA functions or functions on a
1039 "per key" basis to be replaced. This allows hardware acceleration and
1040 hardware key storage to be handled without major modification to the
1041 library. Also added low level modexp hooks and CRYPTO_EX structure and
1042 associated functions.
1045 *) Add a new flag to memory BIOs, BIO_FLAG_MEM_RDONLY. This marks the BIO
1046 as "read only": it can't be written to and the buffer it points to will
1047 not be freed. Reading from a read only BIO is much more efficient than
1048 a normal memory BIO. This was added because there are several times when
1049 an area of memory needs to be read from a BIO. The previous method was
1050 to create a memory BIO and write the data to it, this results in two
1051 copies of the data and an O(n^2) reading algorithm. There is a new
1052 function BIO_new_mem_buf() which creates a read only memory BIO from
1053 an area of memory. Also modified the PKCS#7 routines to use read only
1057 *) Bugfix: ssl23_get_client_hello did not work properly when called in
1058 state SSL23_ST_SR_CLNT_HELLO_B, i.e. when the first 7 bytes of
1059 a SSLv2-compatible client hello for SSLv3 or TLSv1 could be read,
1060 but a retry condition occured while trying to read the rest.
1063 *) The PKCS7_ENC_CONTENT_new() function was setting the content type as
1064 NID_pkcs7_encrypted by default: this was wrong since this should almost
1065 always be NID_pkcs7_data. Also modified the PKCS7_set_type() to handle
1066 the encrypted data type: this is a more sensible place to put it and it
1067 allows the PKCS#12 code to be tidied up that duplicated this
1071 *) Changed obj_dat.pl script so it takes its input and output files on
1072 the command line. This should avoid shell escape redirection problems
1076 *) Initial support for certificate extension requests, these are included
1077 in things like Xenroll certificate requests. Included functions to allow
1078 extensions to be obtained and added.
1081 *) -crlf option to s_client and s_server for sending newlines as
1082 CRLF (as required by many protocols).
1085 Changes between 0.9.3a and 0.9.4 [09 Aug 1999]
1087 *) Install libRSAglue.a when OpenSSL is built with RSAref.
1088 [Ralf S. Engelschall]
1090 *) A few more ``#ifndef NO_FP_API / #endif'' pairs for consistency.
1091 [Andrija Antonijevic <TheAntony2@bigfoot.com>]
1093 *) Fix -startdate and -enddate (which was missing) arguments to 'ca'
1097 *) New function DSA_dup_DH, which duplicates DSA parameters/keys as
1098 DH parameters/keys (q is lost during that conversion, but the resulting
1099 DH parameters contain its length).
1101 For 1024-bit p, DSA_generate_parameters followed by DSA_dup_DH is
1102 much faster than DH_generate_parameters (which creates parameters
1103 where p = 2*q + 1), and also the smaller q makes DH computations
1104 much more efficient (160-bit exponentiation instead of 1024-bit
1105 exponentiation); so this provides a convenient way to support DHE
1106 ciphersuites in SSL/TLS servers (see ssl/ssltest.c). It is of
1107 utter importance to use
1108 SSL_CTX_set_options(s_ctx, SSL_OP_SINGLE_DH_USE);
1110 SSL_set_options(s_ctx, SSL_OP_SINGLE_DH_USE);
1111 when such DH parameters are used, because otherwise small subgroup
1112 attacks may become possible!
1115 *) Avoid memory leak in i2d_DHparams.
1118 *) Allow the -k option to be used more than once in the enc program:
1119 this allows the same encrypted message to be read by multiple recipients.
1122 *) New function OBJ_obj2txt(buf, buf_len, a, no_name), this converts
1123 an ASN1_OBJECT to a text string. If the "no_name" parameter is set then
1124 it will always use the numerical form of the OID, even if it has a short
1128 *) Added an extra RSA flag: RSA_FLAG_EXT_PKEY. Previously the rsa_mod_exp
1129 method only got called if p,q,dmp1,dmq1,iqmp components were present,
1130 otherwise bn_mod_exp was called. In the case of hardware keys for example
1131 no private key components need be present and it might store extra data
1132 in the RSA structure, which cannot be accessed from bn_mod_exp.
1133 By setting RSA_FLAG_EXT_PKEY rsa_mod_exp will always be called for
1134 private key operations.
1137 *) Added support for SPARC Linux.
1140 *) pem_password_cb function type incompatibly changed from
1141 typedef int pem_password_cb(char *buf, int size, int rwflag);
1143 ....(char *buf, int size, int rwflag, void *userdata);
1144 so that applications can pass data to their callbacks:
1145 The PEM[_ASN1]_{read,write}... functions and macros now take an
1146 additional void * argument, which is just handed through whenever
1147 the password callback is called.
1148 [Damien Miller <dmiller@ilogic.com.au>; tiny changes by Bodo Moeller]
1150 New function SSL_CTX_set_default_passwd_cb_userdata.
1152 Compatibility note: As many C implementations push function arguments
1153 onto the stack in reverse order, the new library version is likely to
1154 interoperate with programs that have been compiled with the old
1155 pem_password_cb definition (PEM_whatever takes some data that
1156 happens to be on the stack as its last argument, and the callback
1157 just ignores this garbage); but there is no guarantee whatsoever that
1160 *) The -DPLATFORM="\"$(PLATFORM)\"" definition and the similar -DCFLAGS=...
1161 (both in crypto/Makefile.ssl for use by crypto/cversion.c) caused
1162 problems not only on Windows, but also on some Unix platforms.
1163 To avoid problematic command lines, these definitions are now in an
1164 auto-generated file crypto/buildinf.h (created by crypto/Makefile.ssl
1165 for standard "make" builds, by util/mk1mf.pl for "mk1mf" builds).
1168 *) MIPS III/IV assembler module is reimplemented.
1171 *) More DES library cleanups: remove references to srand/rand and
1172 delete an unused file.
1175 *) Add support for the the free Netwide assembler (NASM) under Win32,
1176 since not many people have MASM (ml) and it can be hard to obtain.
1177 This is currently experimental but it seems to work OK and pass all
1178 the tests. Check out INSTALL.W32 for info.
1181 *) Fix memory leaks in s3_clnt.c: All non-anonymous SSL3/TLS1 connections
1182 without temporary keys kept an extra copy of the server key,
1183 and connections with temporary keys did not free everything in case
1187 *) New function RSA_check_key and new openssl rsa option -check
1188 for verifying the consistency of RSA keys.
1189 [Ulf Moeller, Bodo Moeller]
1191 *) Various changes to make Win32 compile work:
1192 1. Casts to avoid "loss of data" warnings in p5_crpt2.c
1193 2. Change unsigned int to int in b_dump.c to avoid "signed/unsigned
1194 comparison" warnings.
1195 3. Add sk_<TYPE>_sort to DEF file generator and do make update.
1198 *) Add a debugging option to PKCS#5 v2 key generation function: when
1199 you #define DEBUG_PKCS5V2 passwords, salts, iteration counts and
1200 derived keys are printed to stderr.
1203 *) Copy the flags in ASN1_STRING_dup().
1204 [Roman E. Pavlov <pre@mo.msk.ru>]
1206 *) The x509 application mishandled signing requests containing DSA
1207 keys when the signing key was also DSA and the parameters didn't match.
1209 It was supposed to omit the parameters when they matched the signing key:
1210 the verifying software was then supposed to automatically use the CA's
1211 parameters if they were absent from the end user certificate.
1213 Omitting parameters is no longer recommended. The test was also
1214 the wrong way round! This was probably due to unusual behaviour in
1215 EVP_cmp_parameters() which returns 1 if the parameters match.
1216 This meant that parameters were omitted when they *didn't* match and
1217 the certificate was useless. Certificates signed with 'ca' didn't have
1219 [Steve Henson, reported by Doug Erickson <Doug.Erickson@Part.NET>]
1221 *) Memory leak checking (-DCRYPTO_MDEBUG) had some problems.
1222 The interface is as follows:
1223 Applications can use
1224 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ON) aka MemCheck_start(),
1225 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_OFF) aka MemCheck_stop();
1226 "off" is now the default.
1227 The library internally uses
1228 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_DISABLE) aka MemCheck_off(),
1229 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ENABLE) aka MemCheck_on()
1230 to disable memory-checking temporarily.
1232 Some inconsistent states that previously were possible (and were
1233 even the default) are now avoided.
1235 -DCRYPTO_MDEBUG_TIME is new and additionally stores the current time
1236 with each memory chunk allocated; this is occasionally more helpful
1237 than just having a counter.
1239 -DCRYPTO_MDEBUG_THREAD is also new and adds the thread ID.
1241 -DCRYPTO_MDEBUG_ALL enables all of the above, plus any future
1245 *) Introduce "mode" for SSL structures (with defaults in SSL_CTX),
1246 which largely parallels "options", but is for changing API behaviour,
1247 whereas "options" are about protocol behaviour.
1248 Initial "mode" flags are:
1250 SSL_MODE_ENABLE_PARTIAL_WRITE Allow SSL_write to report success when
1251 a single record has been written.
1252 SSL_MODE_ACCEPT_MOVING_WRITE_BUFFER Don't insist that SSL_write
1253 retries use the same buffer location.
1254 (But all of the contents must be
1258 *) Bugfix: SSL_set_mode ignored its parameter, only SSL_CTX_set_mode
1261 *) Fix problems with no-hmac etc.
1262 [Ulf Möller, pointed out by Brian Wellington <bwelling@tislabs.com>]
1264 *) New functions RSA_get_default_method(), RSA_set_method() and
1265 RSA_get_method(). These allows replacement of RSA_METHODs without having
1266 to mess around with the internals of an RSA structure.
1269 *) Fix memory leaks in DSA_do_sign and DSA_is_prime.
1270 Also really enable memory leak checks in openssl.c and in some
1272 [Chad C. Mulligan, Bodo Moeller]
1274 *) Fix a bug in d2i_ASN1_INTEGER() and i2d_ASN1_INTEGER() which can mess
1275 up the length of negative integers. This has now been simplified to just
1276 store the length when it is first determined and use it later, rather
1277 than trying to keep track of where data is copied and updating it to
1279 [Steve Henson, reported by Brien Wheeler
1280 <bwheeler@authentica-security.com>]
1282 *) Add a new function PKCS7_signatureVerify. This allows the verification
1283 of a PKCS#7 signature but with the signing certificate passed to the
1284 function itself. This contrasts with PKCS7_dataVerify which assumes the
1285 certificate is present in the PKCS#7 structure. This isn't always the
1286 case: certificates can be omitted from a PKCS#7 structure and be
1287 distributed by "out of band" means (such as a certificate database).
1290 *) Complete the PEM_* macros with DECLARE_PEM versions to replace the
1291 function prototypes in pem.h, also change util/mkdef.pl to add the
1292 necessary function names.
1295 *) mk1mf.pl (used by Windows builds) did not properly read the
1296 options set by Configure in the top level Makefile, and Configure
1297 was not even able to write more than one option correctly.
1298 Fixed, now "no-idea no-rc5 -DCRYPTO_MDEBUG" etc. works as intended.
1301 *) New functions CONF_load_bio() and CONF_load_fp() to allow a config
1302 file to be loaded from a BIO or FILE pointer. The BIO version will
1303 for example allow memory BIOs to contain config info.
1306 *) New function "CRYPTO_num_locks" that returns CRYPTO_NUM_LOCKS.
1307 Whoever hopes to achieve shared-library compatibility across versions
1308 must use this, not the compile-time macro.
1309 (Exercise 0.9.4: Which is the minimum library version required by
1311 Note: All this applies only to multi-threaded programs, others don't
1315 *) Add missing case to s3_clnt.c state machine -- one of the new SSL tests
1316 through a BIO pair triggered the default case, i.e.
1317 SSLerr(...,SSL_R_UNKNOWN_STATE).
1320 *) New "BIO pair" concept (crypto/bio/bss_bio.c) so that applications
1321 can use the SSL library even if none of the specific BIOs is
1325 *) Fix a bug in i2d_DSAPublicKey() which meant it returned the wrong value
1326 for the encoded length.
1327 [Jeon KyoungHo <khjeon@sds.samsung.co.kr>]
1329 *) Add initial documentation of the X509V3 functions.
1332 *) Add a new pair of functions PEM_write_PKCS8PrivateKey() and
1333 PEM_write_bio_PKCS8PrivateKey() that are equivalent to
1334 PEM_write_PrivateKey() and PEM_write_bio_PrivateKey() but use the more
1335 secure PKCS#8 private key format with a high iteration count.
1338 *) Fix determination of Perl interpreter: A perl or perl5
1339 _directory_ in $PATH was also accepted as the interpreter.
1340 [Ralf S. Engelschall]
1342 *) Fix demos/sign/sign.c: well there wasn't anything strictly speaking
1343 wrong with it but it was very old and did things like calling
1344 PEM_ASN1_read() directly and used MD5 for the hash not to mention some
1348 *) Fix demos/selfsign.c: it used obsolete and deleted functions, changed
1349 to use the new extension code.
1352 *) Implement the PEM_read/PEM_write functions in crypto/pem/pem_all.c
1353 with macros. This should make it easier to change their form, add extra
1354 arguments etc. Fix a few PEM prototypes which didn't have cipher as a
1358 *) Add to configuration table a new entry that can specify an alternative
1359 name for unistd.h (for pre-POSIX systems); we need this for NeXTstep,
1360 according to Mark Crispin <MRC@Panda.COM>.
1364 *) DES CBC did not update the IV. Weird.
1367 des_cbc_encrypt does not update the IV, but des_ncbc_encrypt does.
1368 Changing the behaviour of the former might break existing programs --
1369 where IV updating is needed, des_ncbc_encrypt can be used.
1372 *) When bntest is run from "make test" it drives bc to check its
1373 calculations, as well as internally checking them. If an internal check
1374 fails, it needs to cause bc to give a non-zero result or make test carries
1375 on without noticing the failure. Fixed.
1378 *) DES library cleanups.
1381 *) Add support for PKCS#5 v2.0 PBE algorithms. This will permit PKCS#8 to be
1382 used with any cipher unlike PKCS#5 v1.5 which can at most handle 64 bit
1383 ciphers. NOTE: although the key derivation function has been verified
1384 against some published test vectors it has not been extensively tested
1385 yet. Added a -v2 "cipher" option to pkcs8 application to allow the use
1389 *) Instead of "mkdir -p", which is not fully portable, use new
1390 Perl script "util/mkdir-p.pl".
1393 *) Rewrite the way password based encryption (PBE) is handled. It used to
1394 assume that the ASN1 AlgorithmIdentifier parameter was a PBEParameter
1395 structure. This was true for the PKCS#5 v1.5 and PKCS#12 PBE algorithms
1396 but doesn't apply to PKCS#5 v2.0 where it can be something else. Now
1397 the 'parameter' field of the AlgorithmIdentifier is passed to the
1398 underlying key generation function so it must do its own ASN1 parsing.
1399 This has also changed the EVP_PBE_CipherInit() function which now has a
1400 'parameter' argument instead of literal salt and iteration count values
1401 and the function EVP_PBE_ALGOR_CipherInit() has been deleted.
1404 *) Support for PKCS#5 v1.5 compatible password based encryption algorithms
1405 and PKCS#8 functionality. New 'pkcs8' application linked to openssl.
1406 Needed to change the PEM_STRING_EVP_PKEY value which was just "PRIVATE
1407 KEY" because this clashed with PKCS#8 unencrypted string. Since this
1408 value was just used as a "magic string" and not used directly its
1409 value doesn't matter.
1412 *) Introduce some semblance of const correctness to BN. Shame C doesn't
1416 *) "linux-sparc64" configuration (ultrapenguin).
1417 [Ray Miller <ray.miller@oucs.ox.ac.uk>]
1418 "linux-sparc" configuration.
1419 [Christian Forster <fo@hawo.stw.uni-erlangen.de>]
1421 *) config now generates no-xxx options for missing ciphers.
1424 *) Support the EBCDIC character set (work in progress).
1425 File ebcdic.c not yet included because it has a different license.
1426 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>]
1428 *) Support BS2000/OSD-POSIX.
1429 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>]
1431 *) Make callbacks for key generation use void * instead of char *.
1434 *) Make S/MIME samples compile (not yet tested).
1437 *) Additional typesafe stacks.
1440 *) New configuration variants "bsdi-elf-gcc" (BSD/OS 4.x).
1444 Changes between 0.9.3 and 0.9.3a [29 May 1999]
1446 *) New configuration variant "sco5-gcc".
1448 *) Updated some demos.
1449 [Sean O Riordain, Wade Scholine]
1451 *) Add missing BIO_free at exit of pkcs12 application.
1454 *) Fix memory leak in conf.c.
1457 *) Updates for Win32 to assembler version of MD5.
1460 *) Set #! path to perl in apps/der_chop to where we found it
1461 instead of using a fixed path.
1464 *) SHA library changes for irix64-mips4-cc.
1467 *) Improvements for VMS support.
1471 Changes between 0.9.2b and 0.9.3 [24 May 1999]
1473 *) Bignum library bug fix. IRIX 6 passes "make test" now!
1474 This also avoids the problems with SC4.2 and unpatched SC5.
1475 [Andy Polyakov <appro@fy.chalmers.se>]
1477 *) New functions sk_num, sk_value and sk_set to replace the previous macros.
1478 These are required because of the typesafe stack would otherwise break
1479 existing code. If old code used a structure member which used to be STACK
1480 and is now STACK_OF (for example cert in a PKCS7_SIGNED structure) with
1481 sk_num or sk_value it would produce an error because the num, data members
1482 are not present in STACK_OF. Now it just produces a warning. sk_set
1483 replaces the old method of assigning a value to sk_value
1484 (e.g. sk_value(x, i) = y) which the library used in a few cases. Any code
1485 that does this will no longer work (and should use sk_set instead) but
1486 this could be regarded as a "questionable" behaviour anyway.
1489 *) Fix most of the other PKCS#7 bugs. The "experimental" code can now
1490 correctly handle encrypted S/MIME data.
1493 *) Change type of various DES function arguments from des_cblock
1494 (which means, in function argument declarations, pointer to char)
1495 to des_cblock * (meaning pointer to array with 8 char elements),
1496 which allows the compiler to do more typechecking; it was like
1497 that back in SSLeay, but with lots of ugly casts.
1499 Introduce new type const_des_cblock.
1502 *) Reorganise the PKCS#7 library and get rid of some of the more obvious
1503 problems: find RecipientInfo structure that matches recipient certificate
1504 and initialise the ASN1 structures properly based on passed cipher.
1507 *) Belatedly make the BN tests actually check the results.
1510 *) Fix the encoding and decoding of negative ASN1 INTEGERS and conversion
1511 to and from BNs: it was completely broken. New compilation option
1512 NEG_PUBKEY_BUG to allow for some broken certificates that encode public
1513 key elements as negative integers.
1516 *) Reorganize and speed up MD5.
1517 [Andy Polyakov <appro@fy.chalmers.se>]
1520 [Richard Levitte <richard@levitte.org>]
1522 *) New option -out to asn1parse to allow the parsed structure to be
1523 output to a file. This is most useful when combined with the -strparse
1524 option to examine the output of things like OCTET STRINGS.
1527 *) Make SSL library a little more fool-proof by not requiring any longer
1528 that SSL_set_{accept,connect}_state be called before
1529 SSL_{accept,connect} may be used (SSL_set_..._state is omitted
1530 in many applications because usually everything *appeared* to work as
1531 intended anyway -- now it really works as intended).
1534 *) Move openssl.cnf out of lib/.
1537 *) Fix various things to let OpenSSL even pass ``egcc -pipe -O2 -Wall
1538 -Wshadow -Wpointer-arith -Wcast-align -Wmissing-prototypes
1539 -Wmissing-declarations -Wnested-externs -Winline'' with EGCS 1.1.2+
1540 [Ralf S. Engelschall]
1542 *) Various fixes to the EVP and PKCS#7 code. It may now be able to
1543 handle PKCS#7 enveloped data properly.
1544 [Sebastian Akerman <sak@parallelconsulting.com>, modified by Steve]
1546 *) Create a duplicate of the SSL_CTX's CERT in SSL_new instead of
1547 copying pointers. The cert_st handling is changed by this in
1548 various ways (and thus what used to be known as ctx->default_cert
1549 is now called ctx->cert, since we don't resort to s->ctx->[default_]cert
1550 any longer when s->cert does not give us what we need).
1551 ssl_cert_instantiate becomes obsolete by this change.
1552 As soon as we've got the new code right (possibly it already is?),
1553 we have solved a couple of bugs of the earlier code where s->cert
1554 was used as if it could not have been shared with other SSL structures.
1556 Note that using the SSL API in certain dirty ways now will result
1557 in different behaviour than observed with earlier library versions:
1558 Changing settings for an SSL_CTX *ctx after having done s = SSL_new(ctx)
1559 does not influence s as it used to.
1561 In order to clean up things more thoroughly, inside SSL_SESSION
1562 we don't use CERT any longer, but a new structure SESS_CERT
1563 that holds per-session data (if available); currently, this is
1564 the peer's certificate chain and, for clients, the server's certificate
1565 and temporary key. CERT holds only those values that can have
1566 meaningful defaults in an SSL_CTX.
1569 *) New function X509V3_EXT_i2d() to create an X509_EXTENSION structure
1570 from the internal representation. Various PKCS#7 fixes: remove some
1571 evil casts and set the enc_dig_alg field properly based on the signing
1575 *) Allow PKCS#12 password to be set from the command line or the
1576 environment. Let 'ca' get its config file name from the environment
1577 variables "OPENSSL_CONF" or "SSLEAY_CONF" (for consistency with 'req'
1581 *) Allow certificate policies extension to use an IA5STRING for the
1582 organization field. This is contrary to the PKIX definition but
1583 VeriSign uses it and IE5 only recognises this form. Document 'x509'
1587 *) Add PEDANTIC compiler flag to allow compilation with gcc -pedantic,
1588 without disallowing inline assembler and the like for non-pedantic builds.
1591 *) Support Borland C++ builder.
1592 [Janez Jere <jj@void.si>, modified by Ulf Möller]
1597 *) SHA-1 cleanups and performance enhancements.
1598 [Andy Polyakov <appro@fy.chalmers.se>]
1600 *) Sparc v8plus assembler for the bignum library.
1601 [Andy Polyakov <appro@fy.chalmers.se>]
1603 *) Accept any -xxx and +xxx compiler options in Configure.
1606 *) Update HPUX configuration.
1609 *) Add missing sk_<type>_unshift() function to safestack.h
1610 [Ralf S. Engelschall]
1612 *) New function SSL_CTX_use_certificate_chain_file that sets the
1613 "extra_cert"s in addition to the certificate. (This makes sense
1614 only for "PEM" format files, as chains as a whole are not
1618 *) Support verify_depth from the SSL API.
1619 x509_vfy.c had what can be considered an off-by-one-error:
1620 Its depth (which was not part of the external interface)
1621 was actually counting the number of certificates in a chain;
1622 now it really counts the depth.
1625 *) Bugfix in crypto/x509/x509_cmp.c: The SSLerr macro was used
1626 instead of X509err, which often resulted in confusing error
1627 messages since the error codes are not globally unique
1628 (e.g. an alleged error in ssl3_accept when a certificate
1629 didn't match the private key).
1631 *) New function SSL_CTX_set_session_id_context that allows to set a default
1632 value (so that you don't need SSL_set_session_id_context for each
1633 connection using the SSL_CTX).
1636 *) OAEP decoding bug fix.
1639 *) Support INSTALL_PREFIX for package builders, as proposed by
1643 *) New Configure options "threads" and "no-threads". For systems
1644 where the proper compiler options are known (currently Solaris
1645 and Linux), "threads" is the default.
1648 *) New script util/mklink.pl as a faster substitute for util/mklink.sh.
1651 *) Install various scripts to $(OPENSSLDIR)/misc, not to
1652 $(INSTALLTOP)/bin -- they shouldn't clutter directories
1653 such as /usr/local/bin.
1656 *) "make linux-shared" to build shared libraries.
1657 [Niels Poppe <niels@netbox.org>]
1659 *) New Configure option no-<cipher> (rsa, idea, rc5, ...).
1662 *) Add the PKCS#12 API documentation to openssl.txt. Preliminary support for
1663 extension adding in x509 utility.
1666 *) Remove NOPROTO sections and error code comments.
1669 *) Partial rewrite of the DEF file generator to now parse the ANSI
1673 *) New Configure options --prefix=DIR and --openssldir=DIR.
1676 *) Complete rewrite of the error code script(s). It is all now handled
1677 by one script at the top level which handles error code gathering,
1678 header rewriting and C source file generation. It should be much better
1679 than the old method: it now uses a modified version of Ulf's parser to
1680 read the ANSI prototypes in all header files (thus the old K&R definitions
1681 aren't needed for error creation any more) and do a better job of
1682 translating function codes into names. The old 'ASN1 error code imbedded
1683 in a comment' is no longer necessary and it doesn't use .err files which
1684 have now been deleted. Also the error code call doesn't have to appear all
1685 on one line (which resulted in some large lines...).
1688 *) Change #include filenames from <foo.h> to <openssl/foo.h>.
1691 *) Change behaviour of ssl2_read when facing length-0 packets: Don't return
1692 0 (which usually indicates a closed connection), but continue reading.
1695 *) Fix some race conditions.
1698 *) Add support for CRL distribution points extension. Add Certificate
1699 Policies and CRL distribution points documentation.
1702 *) Move the autogenerated header file parts to crypto/opensslconf.h.
1705 *) Fix new 56-bit DES export ciphersuites: they were using 7 bytes instead of
1706 8 of keying material. Merlin has also confirmed interop with this fix
1707 between OpenSSL and Baltimore C/SSL 2.0 and J/SSL 2.0.
1708 [Merlin Hughes <merlin@baltimore.ie>]
1710 *) Fix lots of warnings.
1711 [Richard Levitte <levitte@stacken.kth.se>]
1713 *) In add_cert_dir() in crypto/x509/by_dir.c, break out of the loop if
1714 the directory spec didn't end with a LIST_SEPARATOR_CHAR.
1715 [Richard Levitte <levitte@stacken.kth.se>]
1717 *) Fix problems with sizeof(long) == 8.
1718 [Andy Polyakov <appro@fy.chalmers.se>]
1720 *) Change functions to ANSI C.
1723 *) Fix typos in error codes.
1724 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>, Ulf Möller]
1726 *) Remove defunct assembler files from Configure.
1729 *) SPARC v8 assembler BIGNUM implementation.
1730 [Andy Polyakov <appro@fy.chalmers.se>]
1732 *) Support for Certificate Policies extension: both print and set.
1733 Various additions to support the r2i method this uses.
1736 *) A lot of constification, and fix a bug in X509_NAME_oneline() that could
1737 return a const string when you are expecting an allocated buffer.
1740 *) Add support for ASN1 types UTF8String and VISIBLESTRING, also the CHOICE
1741 types DirectoryString and DisplayText.
1744 *) Add code to allow r2i extensions to access the configuration database,
1745 add an LHASH database driver and add several ctx helper functions.
1748 *) Fix an evil bug in bn_expand2() which caused various BN functions to
1749 fail when they extended the size of a BIGNUM.
1752 *) Various utility functions to handle SXNet extension. Modify mkdef.pl to
1753 support typesafe stack.
1756 *) Fix typo in SSL_[gs]et_options().
1757 [Nils Frostberg <nils@medcom.se>]
1759 *) Delete various functions and files that belonged to the (now obsolete)
1760 old X509V3 handling code.
1763 *) New Configure option "rsaref".
1766 *) Don't auto-generate pem.h.
1769 *) Introduce type-safe ASN.1 SETs.
1772 *) Convert various additional casted stacks to type-safe STACK_OF() variants.
1773 [Ben Laurie, Ralf S. Engelschall, Steve Henson]
1775 *) Introduce type-safe STACKs. This will almost certainly break lots of code
1776 that links with OpenSSL (well at least cause lots of warnings), but fear
1777 not: the conversion is trivial, and it eliminates loads of evil casts. A
1778 few STACKed things have been converted already. Feel free to convert more.
1779 In the fullness of time, I'll do away with the STACK type altogether.
1782 *) Add `openssl ca -revoke <certfile>' facility which revokes a certificate
1783 specified in <certfile> by updating the entry in the index.txt file.
1784 This way one no longer has to edit the index.txt file manually for
1785 revoking a certificate. The -revoke option does the gory details now.
1786 [Massimiliano Pala <madwolf@openca.org>, Ralf S. Engelschall]
1788 *) Fix `openssl crl -noout -text' combination where `-noout' killed the
1789 `-text' option at all and this way the `-noout -text' combination was
1790 inconsistent in `openssl crl' with the friends in `openssl x509|rsa|dsa'.
1791 [Ralf S. Engelschall]
1793 *) Make sure a corresponding plain text error message exists for the
1794 X509_V_ERR_CERT_REVOKED/23 error number which can occur when a
1795 verify callback function determined that a certificate was revoked.
1796 [Ralf S. Engelschall]
1798 *) Bugfix: In test/testenc, don't test "openssl <cipher>" for
1799 ciphers that were excluded, e.g. by -DNO_IDEA. Also, test
1800 all available cipers including rc5, which was forgotten until now.
1801 In order to let the testing shell script know which algorithms
1802 are available, a new (up to now undocumented) command
1803 "openssl list-cipher-commands" is used.
1806 *) Bugfix: s_client occasionally would sleep in select() when
1807 it should have checked SSL_pending() first.
1810 *) New functions DSA_do_sign and DSA_do_verify to provide access to
1811 the raw DSA values prior to ASN.1 encoding.
1814 *) Tweaks to Configure
1815 [Niels Poppe <niels@netbox.org>]
1817 *) Add support for PKCS#5 v2.0 ASN1 PBES2 structures. No other support,
1821 *) New variables $(RANLIB) and $(PERL) in the Makefiles.
1824 *) New config option to avoid instructions that are illegal on the 80386.
1825 The default code is faster, but requires at least a 486.
1828 *) Got rid of old SSL2_CLIENT_VERSION (inconsistently used) and
1829 SSL2_SERVER_VERSION (not used at all) macros, which are now the
1830 same as SSL2_VERSION anyway.
1833 *) New "-showcerts" option for s_client.
1836 *) Still more PKCS#12 integration. Add pkcs12 application to openssl
1837 application. Various cleanups and fixes.
1840 *) More PKCS#12 integration. Add new pkcs12 directory with Makefile.ssl and
1841 modify error routines to work internally. Add error codes and PBE init
1842 to library startup routines.
1845 *) Further PKCS#12 integration. Added password based encryption, PKCS#8 and
1846 packing functions to asn1 and evp. Changed function names and error
1847 codes along the way.
1850 *) PKCS12 integration: and so it begins... First of several patches to
1851 slowly integrate PKCS#12 functionality into OpenSSL. Add PKCS#12
1852 objects to objects.h
1855 *) Add a new 'indent' option to some X509V3 extension code. Initial ASN1
1856 and display support for Thawte strong extranet extension.
1859 *) Add LinuxPPC support.
1860 [Jeff Dubrule <igor@pobox.org>]
1862 *) Get rid of redundant BN file bn_mulw.c, and rename bn_div64 to
1863 bn_div_words in alpha.s.
1864 [Hannes Reinecke <H.Reinecke@hw.ac.uk> and Ben Laurie]
1866 *) Make sure the RSA OAEP test is skipped under -DRSAref because
1867 OAEP isn't supported when OpenSSL is built with RSAref.
1868 [Ulf Moeller <ulf@fitug.de>]
1870 *) Move definitions of IS_SET/IS_SEQUENCE inside crypto/asn1/asn1.h
1871 so they no longer are missing under -DNOPROTO.
1872 [Soren S. Jorvang <soren@t.dk>]
1875 Changes between 0.9.1c and 0.9.2b [22 Mar 1999]
1877 *) Make SSL_get_peer_cert_chain() work in servers. Unfortunately, it still
1878 doesn't work when the session is reused. Coming soon!
1881 *) Fix a security hole, that allows sessions to be reused in the wrong
1882 context thus bypassing client cert protection! All software that uses
1883 client certs and session caches in multiple contexts NEEDS PATCHING to
1884 allow session reuse! A fuller solution is in the works.
1885 [Ben Laurie, problem pointed out by Holger Reif, Bodo Moeller (and ???)]
1887 *) Some more source tree cleanups (removed obsolete files
1888 crypto/bf/asm/bf586.pl, test/test.txt and crypto/sha/asm/f.s; changed
1889 permission on "config" script to be executable) and a fix for the INSTALL
1891 [Ulf Moeller <ulf@fitug.de>]
1893 *) Remove some legacy and erroneous uses of malloc, free instead of
1895 [Lennart Bang <lob@netstream.se>, with minor changes by Steve]
1897 *) Make rsa_oaep_test return non-zero on error.
1898 [Ulf Moeller <ulf@fitug.de>]
1900 *) Add support for native Solaris shared libraries. Configure
1901 solaris-sparc-sc4-pic, make, then run shlib/solaris-sc4.sh. It'd be nice
1902 if someone would make that last step automatic.
1903 [Matthias Loepfe <Matthias.Loepfe@AdNovum.CH>]
1905 *) ctx_size was not built with the right compiler during "make links". Fixed.
1908 *) Change the meaning of 'ALL' in the cipher list. It now means "everything
1909 except NULL ciphers". This means the default cipher list will no longer
1910 enable NULL ciphers. They need to be specifically enabled e.g. with
1911 the string "DEFAULT:eNULL".
1914 *) Fix to RSA private encryption routines: if p < q then it would
1915 occasionally produce an invalid result. This will only happen with
1916 externally generated keys because OpenSSL (and SSLeay) ensure p > q.
1919 *) Be less restrictive and allow also `perl util/perlpath.pl
1920 /path/to/bin/perl' in addition to `perl util/perlpath.pl /path/to/bin',
1921 because this way one can also use an interpreter named `perl5' (which is
1922 usually the name of Perl 5.xxx on platforms where an Perl 4.x is still
1923 installed as `perl').
1924 [Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
1926 *) Let util/clean-depend.pl work also with older Perl 5.00x versions.
1927 [Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
1929 *) Fix Makefile.org so CC,CFLAG etc are passed to 'make links' add
1930 advapi32.lib to Win32 build and change the pem test comparision
1931 to fc.exe (thanks to Ulrich Kroener <kroneru@yahoo.com> for the
1932 suggestion). Fix misplaced ASNI prototypes and declarations in evp.h
1933 and crypto/des/ede_cbcm_enc.c.
1936 *) DES quad checksum was broken on big-endian architectures. Fixed.
1939 *) Comment out two functions in bio.h that aren't implemented. Fix up the
1940 Win32 test batch file so it (might) work again. The Win32 test batch file
1941 is horrible: I feel ill....
1944 *) Move various #ifdefs around so NO_SYSLOG, NO_DIRENT etc are now selected
1945 in e_os.h. Audit of header files to check ANSI and non ANSI
1946 sections: 10 functions were absent from non ANSI section and not exported
1947 from Windows DLLs. Fixed up libeay.num for new functions.
1950 *) Make `openssl version' output lines consistent.
1951 [Ralf S. Engelschall]
1953 *) Fix Win32 symbol export lists for BIO functions: Added
1954 BIO_get_ex_new_index, BIO_get_ex_num, BIO_get_ex_data and BIO_set_ex_data
1955 to ms/libeay{16,32}.def.
1956 [Ralf S. Engelschall]
1958 *) Second round of fixing the OpenSSL perl/ stuff. It now at least compiled
1959 fine under Unix and passes some trivial tests I've now added. But the
1960 whole stuff is horribly incomplete, so a README.1ST with a disclaimer was
1961 added to make sure no one expects that this stuff really works in the
1962 OpenSSL 0.9.2 release. Additionally I've started to clean the XS sources
1963 up and fixed a few little bugs and inconsistencies in OpenSSL.{pm,xs} and
1965 [Ralf S. Engelschall]
1967 *) Fix the generation of two part addresses in perl.
1968 [Kenji Miyake <kenji@miyake.org>, integrated by Ben Laurie]
1970 *) Add config entry for Linux on MIPS.
1971 [John Tobey <jtobey@channel1.com>]
1973 *) Make links whenever Configure is run, unless we are on Windoze.
1976 *) Permit extensions to be added to CRLs using crl_section in openssl.cnf.
1977 Currently only issuerAltName and AuthorityKeyIdentifier make any sense
1981 *) Add a useful kludge to allow package maintainers to specify compiler and
1982 other platforms details on the command line without having to patch the
1983 Configure script everytime: One now can use ``perl Configure
1984 <id>:<details>'', i.e. platform ids are allowed to have details appended
1985 to them (seperated by colons). This is treated as there would be a static
1986 pre-configured entry in Configure's %table under key <id> with value
1987 <details> and ``perl Configure <id>'' is called. So, when you want to
1988 perform a quick test-compile under FreeBSD 3.1 with pgcc and without
1989 assembler stuff you can use ``perl Configure "FreeBSD-elf:pgcc:-O6:::"''
1990 now, which overrides the FreeBSD-elf entry on-the-fly.
1991 [Ralf S. Engelschall]
1993 *) Disable new TLS1 ciphersuites by default: they aren't official yet.
1996 *) Allow DSO flags like -fpic, -fPIC, -KPIC etc. to be specified
1997 on the `perl Configure ...' command line. This way one can compile
1998 OpenSSL libraries with Position Independent Code (PIC) which is needed
1999 for linking it into DSOs.
2000 [Ralf S. Engelschall]
2002 *) Remarkably, export ciphers were totally broken and no-one had noticed!
2006 *) Cleaned up the LICENSE document: The official contact for any license
2007 questions now is the OpenSSL core team under openssl-core@openssl.org.
2008 And add a paragraph about the dual-license situation to make sure people
2009 recognize that _BOTH_ the OpenSSL license _AND_ the SSLeay license apply
2010 to the OpenSSL toolkit.
2011 [Ralf S. Engelschall]
2013 *) General source tree makefile cleanups: Made `making xxx in yyy...'
2014 display consistent in the source tree and replaced `/bin/rm' by `rm'.
2015 Additonally cleaned up the `make links' target: Remove unnecessary
2016 semicolons, subsequent redundant removes, inline point.sh into mklink.sh
2017 to speed processing and no longer clutter the display with confusing
2018 stuff. Instead only the actually done links are displayed.
2019 [Ralf S. Engelschall]
2021 *) Permit null encryption ciphersuites, used for authentication only. It used
2022 to be necessary to set the preprocessor define SSL_ALLOW_ENULL to do this.
2023 It is now necessary to set SSL_FORBID_ENULL to prevent the use of null
2027 *) Add a bunch of fixes to the PKCS#7 stuff. It used to sometimes reorder
2028 signed attributes when verifying signatures (this would break them),
2029 the detached data encoding was wrong and public keys obtained using
2030 X509_get_pubkey() weren't freed.
2033 *) Add text documentation for the BUFFER functions. Also added a work around
2034 to a Win95 console bug. This was triggered by the password read stuff: the
2035 last character typed gets carried over to the next fread(). If you were
2036 generating a new cert request using 'req' for example then the last
2037 character of the passphrase would be CR which would then enter the first
2041 *) Added the new `Includes OpenSSL Cryptography Software' button as
2042 doc/openssl_button.{gif,html} which is similar in style to the old SSLeay
2043 button and can be used by applications based on OpenSSL to show the
2044 relationship to the OpenSSL project.
2045 [Ralf S. Engelschall]
2047 *) Remove confusing variables in function signatures in files
2048 ssl/ssl_lib.c and ssl/ssl.h.
2049 [Lennart Bong <lob@kulthea.stacken.kth.se>]
2051 *) Don't install bss_file.c under PREFIX/include/
2052 [Lennart Bong <lob@kulthea.stacken.kth.se>]
2054 *) Get the Win32 compile working again. Modify mkdef.pl so it can handle
2055 functions that return function pointers and has support for NT specific
2056 stuff. Fix mk1mf.pl and VC-32.pl to support NT differences also. Various
2057 #ifdef WIN32 and WINNTs sprinkled about the place and some changes from
2058 unsigned to signed types: this was killing the Win32 compile.
2061 *) Add new certificate file to stack functions,
2062 SSL_add_dir_cert_subjects_to_stack() and
2063 SSL_add_file_cert_subjects_to_stack(). These largely supplant
2064 SSL_load_client_CA_file(), and can be used to add multiple certs easily
2065 to a stack (usually this is then handed to SSL_CTX_set_client_CA_list()).
2066 This means that Apache-SSL and similar packages don't have to mess around
2067 to add as many CAs as they want to the preferred list.
2070 *) Experiment with doxygen documentation. Currently only partially applied to
2072 See http://www.stack.nl/~dimitri/doxygen/index.html, and run doxygen with
2073 openssl.doxy as the configuration file.
2076 *) Get rid of remaining C++-style comments which strict C compilers hate.
2077 [Ralf S. Engelschall, pointed out by Carlos Amengual]
2079 *) Changed BN_RECURSION in bn_mont.c to BN_RECURSION_MONT so it is not
2080 compiled in by default: it has problems with large keys.
2083 *) Add a bunch of SSL_xxx() functions for configuring the temporary RSA and
2084 DH private keys and/or callback functions which directly correspond to
2085 their SSL_CTX_xxx() counterparts but work on a per-connection basis. This
2086 is needed for applications which have to configure certificates on a
2087 per-connection basis (e.g. Apache+mod_ssl) instead of a per-context basis
2089 For the RSA certificate situation is makes no difference, but
2090 for the DSA certificate situation this fixes the "no shared cipher"
2091 problem where the OpenSSL cipher selection procedure failed because the
2092 temporary keys were not overtaken from the context and the API provided
2093 no way to reconfigure them.
2094 The new functions now let applications reconfigure the stuff and they
2095 are in detail: SSL_need_tmp_RSA, SSL_set_tmp_rsa, SSL_set_tmp_dh,
2096 SSL_set_tmp_rsa_callback and SSL_set_tmp_dh_callback. Additionally a new
2097 non-public-API function ssl_cert_instantiate() is used as a helper
2098 function and also to reduce code redundancy inside ssl_rsa.c.
2099 [Ralf S. Engelschall]
2101 *) Move s_server -dcert and -dkey options out of the undocumented feature
2102 area because they are useful for the DSA situation and should be
2103 recognized by the users.
2104 [Ralf S. Engelschall]
2106 *) Fix the cipher decision scheme for export ciphers: the export bits are
2107 *not* within SSL_MKEY_MASK or SSL_AUTH_MASK, they are within
2108 SSL_EXP_MASK. So, the original variable has to be used instead of the
2109 already masked variable.
2110 [Richard Levitte <levitte@stacken.kth.se>]
2112 *) Fix 'port' variable from `int' to `unsigned int' in crypto/bio/b_sock.c
2113 [Richard Levitte <levitte@stacken.kth.se>]
2115 *) Change type of another md_len variable in pk7_doit.c:PKCS7_dataFinal()
2116 from `int' to `unsigned int' because it's a length and initialized by
2117 EVP_DigestFinal() which expects an `unsigned int *'.
2118 [Richard Levitte <levitte@stacken.kth.se>]
2120 *) Don't hard-code path to Perl interpreter on shebang line of Configure
2121 script. Instead use the usual Shell->Perl transition trick.
2122 [Ralf S. Engelschall]
2124 *) Make `openssl x509 -noout -modulus' functional also for DSA certificates
2125 (in addition to RSA certificates) to match the behaviour of `openssl dsa
2126 -noout -modulus' as it's already the case for `openssl rsa -noout
2127 -modulus'. For RSA the -modulus is the real "modulus" while for DSA
2128 currently the public key is printed (a decision which was already done by
2129 `openssl dsa -modulus' in the past) which serves a similar purpose.
2130 Additionally the NO_RSA no longer completely removes the whole -modulus
2131 option; it now only avoids using the RSA stuff. Same applies to NO_DSA
2133 [Ralf S. Engelschall]
2135 *) Add Arne Ansper's reliable BIO - this is an encrypted, block-digested
2136 BIO. See the source (crypto/evp/bio_ok.c) for more info.
2137 [Arne Ansper <arne@ats.cyber.ee>]
2139 *) Dump the old yucky req code that tried (and failed) to allow raw OIDs
2140 to be added. Now both 'req' and 'ca' can use new objects defined in the
2144 *) Add cool BIO that does syslog (or event log on NT).
2145 [Arne Ansper <arne@ats.cyber.ee>, integrated by Ben Laurie]
2147 *) Add support for new TLS ciphersuites, TLS_RSA_EXPORT56_WITH_RC4_56_MD5,
2148 TLS_RSA_EXPORT56_WITH_RC2_CBC_56_MD5 and
2149 TLS_RSA_EXPORT56_WITH_DES_CBC_SHA, as specified in "56-bit Export Cipher
2150 Suites For TLS", draft-ietf-tls-56-bit-ciphersuites-00.txt.
2153 *) Add preliminary config info for new extension code.
2156 *) Make RSA_NO_PADDING really use no padding.
2157 [Ulf Moeller <ulf@fitug.de>]
2159 *) Generate errors when private/public key check is done.
2162 *) Overhaul for 'crl' utility. New function X509_CRL_print. Partial support
2163 for some CRL extensions and new objects added.
2166 *) Really fix the ASN1 IMPLICIT bug this time... Partial support for private
2167 key usage extension and fuller support for authority key id.
2170 *) Add OAEP encryption for the OpenSSL crypto library. OAEP is the improved
2171 padding method for RSA, which is recommended for new applications in PKCS
2172 #1 v2.0 (RFC 2437, October 1998).
2173 OAEP (Optimal Asymmetric Encryption Padding) has better theoretical
2174 foundations than the ad-hoc padding used in PKCS #1 v1.5. It is secure
2175 against Bleichbacher's attack on RSA.
2176 [Ulf Moeller <ulf@fitug.de>, reformatted, corrected and integrated by
2179 *) Updates to the new SSL compression code
2180 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
2182 *) Fix so that the version number in the master secret, when passed
2183 via RSA, checks that if TLS was proposed, but we roll back to SSLv3
2184 (because the server will not accept higher), that the version number
2185 is 0x03,0x01, not 0x03,0x00
2186 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
2188 *) Run extensive memory leak checks on SSL apps. Fixed *lots* of memory
2189 leaks in ssl/ relating to new X509_get_pubkey() behaviour. Also fixes
2190 in apps/ and an unrelated leak in crypto/dsa/dsa_vrf.c
2193 *) Support for RAW extensions where an arbitrary extension can be
2194 created by including its DER encoding. See apps/openssl.cnf for
2198 *) Make sure latest Perl versions don't interpret some generated C array
2199 code as Perl array code in the crypto/err/err_genc.pl script.
2200 [Lars Weber <3weber@informatik.uni-hamburg.de>]
2202 *) Modify ms/do_ms.bat to not generate assembly language makefiles since
2203 not many people have the assembler. Various Win32 compilation fixes and
2204 update to the INSTALL.W32 file with (hopefully) more accurate Win32
2208 *) Modify configure script 'Configure' to automatically create crypto/date.h
2209 file under Win32 and also build pem.h from pem.org. New script
2210 util/mkfiles.pl to create the MINFO file on environments that can't do a
2211 'make files': perl util/mkfiles.pl >MINFO should work.
2214 *) Major rework of DES function declarations, in the pursuit of correctness
2215 and purity. As a result, many evil casts evaporated, and some weirdness,
2216 too. You may find this causes warnings in your code. Zapping your evil
2217 casts will probably fix them. Mostly.
2220 *) Fix for a typo in asn1.h. Bug fix to object creation script
2221 obj_dat.pl. It considered a zero in an object definition to mean
2222 "end of object": none of the objects in objects.h have any zeros
2223 so it wasn't spotted.
2224 [Steve Henson, reported by Erwann ABALEA <eabalea@certplus.com>]
2226 *) Add support for Triple DES Cipher Block Chaining with Output Feedback
2227 Masking (CBCM). In the absence of test vectors, the best I have been able
2228 to do is check that the decrypt undoes the encrypt, so far. Send me test
2229 vectors if you have them.
2232 *) Correct calculation of key length for export ciphers (too much space was
2233 allocated for null ciphers). This has not been tested!
2236 *) Modifications to the mkdef.pl for Win32 DEF file creation. The usage
2237 message is now correct (it understands "crypto" and "ssl" on its
2238 command line). There is also now an "update" option. This will update
2239 the util/ssleay.num and util/libeay.num files with any new functions.
2241 perl util/mkdef.pl crypto ssl update
2242 it will update them.
2245 *) Overhauled the Perl interface (perl/*):
2246 - ported BN stuff to OpenSSL's different BN library
2247 - made the perl/ source tree CVS-aware
2248 - renamed the package from SSLeay to OpenSSL (the files still contain
2249 their history because I've copied them in the repository)
2250 - removed obsolete files (the test scripts will be replaced
2251 by better Test::Harness variants in the future)
2252 [Ralf S. Engelschall]
2254 *) First cut for a very conservative source tree cleanup:
2255 1. merge various obsolete readme texts into doc/ssleay.txt
2256 where we collect the old documents and readme texts.
2257 2. remove the first part of files where I'm already sure that we no
2258 longer need them because of three reasons: either they are just temporary
2259 files which were left by Eric or they are preserved original files where
2260 I've verified that the diff is also available in the CVS via "cvs diff
2261 -rSSLeay_0_8_1b" or they were renamed (as it was definitely the case for
2262 the crypto/md/ stuff).
2263 [Ralf S. Engelschall]
2265 *) More extension code. Incomplete support for subject and issuer alt
2266 name, issuer and authority key id. Change the i2v function parameters
2267 and add an extra 'crl' parameter in the X509V3_CTX structure: guess
2268 what that's for :-) Fix to ASN1 macro which messed up
2269 IMPLICIT tag and add f_enum.c which adds a2i, i2a for ENUMERATED.
2272 *) Preliminary support for ENUMERATED type. This is largely copied from the
2276 *) Add new function, EVP_MD_CTX_copy() to replace frequent use of memcpy.
2277 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
2279 *) Make sure `make rehash' target really finds the `openssl' program.
2280 [Ralf S. Engelschall, Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
2282 *) Squeeze another 7% of speed out of MD5 assembler, at least on a P2. I'd
2283 like to hear about it if this slows down other processors.
2286 *) Add CygWin32 platform information to Configure script.
2287 [Alan Batie <batie@aahz.jf.intel.com>]
2289 *) Fixed ms/32all.bat script: `no_asm' -> `no-asm'
2290 [Rainer W. Gerling <gerling@mpg-gv.mpg.de>]
2292 *) New program nseq to manipulate netscape certificate sequences
2295 *) Modify crl2pkcs7 so it supports multiple -certfile arguments. Fix a
2299 *) Fixes to BN code. Previously the default was to define BN_RECURSION
2300 but the BN code had some problems that would cause failures when
2301 doing certificate verification and some other functions.
2302 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
2304 *) Add ASN1 and PEM code to support netscape certificate sequences.
2307 *) Add ASN1 and PEM code to support netscape certificate sequences.
2310 *) Add several PKIX and private extended key usage OIDs.
2313 *) Modify the 'ca' program to handle the new extension code. Modify
2314 openssl.cnf for new extension format, add comments.
2317 *) More X509 V3 changes. Fix typo in v3_bitstr.c. Add support to 'req'
2318 and add a sample to openssl.cnf so req -x509 now adds appropriate
2322 *) Continued X509 V3 changes. Add to other makefiles, integrate with the
2323 error code, add initial support to X509_print() and x509 application.
2326 *) Takes a deep breath and start addding X509 V3 extension support code. Add
2327 files in crypto/x509v3. Move original stuff to crypto/x509v3/old. All this
2328 stuff is currently isolated and isn't even compiled yet.
2331 *) Continuing patches for GeneralizedTime. Fix up certificate and CRL
2332 ASN1 to use ASN1_TIME and modify print routines to use ASN1_TIME_print.
2333 Removed the versions check from X509 routines when loading extensions:
2334 this allows certain broken certificates that don't set the version
2335 properly to be processed.
2338 *) Deal with irritating shit to do with dependencies, in YAAHW (Yet Another
2339 Ad Hoc Way) - Makefile.ssls now all contain local dependencies, which
2340 can still be regenerated with "make depend".
2343 *) Spelling mistake in C version of CAST-128.
2344 [Ben Laurie, reported by Jeremy Hylton <jeremy@cnri.reston.va.us>]
2346 *) Changes to the error generation code. The perl script err-code.pl
2347 now reads in the old error codes and retains the old numbers, only
2348 adding new ones if necessary. It also only changes the .err files if new
2349 codes are added. The makefiles have been modified to only insert errors
2350 when needed (to avoid needlessly modifying header files). This is done
2351 by only inserting errors if the .err file is newer than the auto generated
2352 C file. To rebuild all the error codes from scratch (the old behaviour)
2353 either modify crypto/Makefile.ssl to pass the -regen flag to err_code.pl
2354 or delete all the .err files.
2357 *) CAST-128 was incorrectly implemented for short keys. The C version has
2358 been fixed, but is untested. The assembler versions are also fixed, but
2359 new assembler HAS NOT BEEN GENERATED FOR WIN32 - the Makefile needs fixing
2360 to regenerate it if needed.
2361 [Ben Laurie, reported (with fix for C version) by Jun-ichiro itojun
2362 Hagino <itojun@kame.net>]
2364 *) File was opened incorrectly in randfile.c.
2365 [Ulf Möller <ulf@fitug.de>]
2367 *) Beginning of support for GeneralizedTime. d2i, i2d, check and print
2368 functions. Also ASN1_TIME suite which is a CHOICE of UTCTime or
2369 GeneralizedTime. ASN1_TIME is the proper type used in certificates et
2370 al: it's just almost always a UTCTime. Note this patch adds new error
2371 codes so do a "make errors" if there are problems.
2374 *) Correct Linux 1 recognition in config.
2375 [Ulf Möller <ulf@fitug.de>]
2377 *) Remove pointless MD5 hash when using DSA keys in ca.
2378 [Anonymous <nobody@replay.com>]
2380 *) Generate an error if given an empty string as a cert directory. Also
2381 generate an error if handed NULL (previously returned 0 to indicate an
2382 error, but didn't set one).
2383 [Ben Laurie, reported by Anonymous <nobody@replay.com>]
2385 *) Add prototypes to SSL methods. Make SSL_write's buffer const, at last.
2388 *) Fix the dummy function BN_ref_mod_exp() in rsaref.c to have the correct
2389 parameters. This was causing a warning which killed off the Win32 compile.
2392 *) Remove C++ style comments from crypto/bn/bn_local.h.
2393 [Neil Costigan <neil.costigan@celocom.com>]
2395 *) The function OBJ_txt2nid was broken. It was supposed to return a nid
2396 based on a text string, looking up short and long names and finally
2397 "dot" format. The "dot" format stuff didn't work. Added new function
2398 OBJ_txt2obj to do the same but return an ASN1_OBJECT and rewrote
2399 OBJ_txt2nid to use it. OBJ_txt2obj can also return objects even if the
2400 OID is not part of the table.
2403 *) Add prototypes to X509 lookup/verify methods, fixing a bug in
2404 X509_LOOKUP_by_alias().
2407 *) Sort openssl functions by name.
2410 *) Get the gendsa program working (hopefully) and add it to app list. Remove
2411 encryption from sample DSA keys (in case anyone is interested the password
2415 *) Make _all_ *_free functions accept a NULL pointer.
2416 [Frans Heymans <fheymans@isaserver.be>]
2418 *) If a DH key is generated in s3_srvr.c, don't blow it by trying to use
2420 [Anonymous <nobody@replay.com>]
2422 *) s_server should send the CAfile as acceptable CAs, not its own cert.
2423 [Bodo Moeller <3moeller@informatik.uni-hamburg.de>]
2425 *) Don't blow it for numeric -newkey arguments to apps/req.
2426 [Bodo Moeller <3moeller@informatik.uni-hamburg.de>]
2428 *) Temp key "for export" tests were wrong in s3_srvr.c.
2429 [Anonymous <nobody@replay.com>]
2431 *) Add prototype for temp key callback functions
2432 SSL_CTX_set_tmp_{rsa,dh}_callback().
2435 *) Make DH_free() tolerate being passed a NULL pointer (like RSA_free() and
2436 DSA_free()). Make X509_PUBKEY_set() check for errors in d2i_PublicKey().
2439 *) X509_name_add_entry() freed the wrong thing after an error.
2440 [Arne Ansper <arne@ats.cyber.ee>]
2442 *) rsa_eay.c would attempt to free a NULL context.
2443 [Arne Ansper <arne@ats.cyber.ee>]
2445 *) BIO_s_socket() had a broken should_retry() on Windoze.
2446 [Arne Ansper <arne@ats.cyber.ee>]
2448 *) BIO_f_buffer() didn't pass on BIO_CTRL_FLUSH.
2449 [Arne Ansper <arne@ats.cyber.ee>]
2451 *) Make sure the already existing X509_STORE->depth variable is initialized
2452 in X509_STORE_new(), but document the fact that this variable is still
2453 unused in the certificate verification process.
2454 [Ralf S. Engelschall]
2456 *) Fix the various library and apps files to free up pkeys obtained from
2457 X509_PUBKEY_get() et al. Also allow x509.c to handle netscape extensions.
2460 *) Fix reference counting in X509_PUBKEY_get(). This makes
2461 demos/maurice/example2.c work, amongst others, probably.
2462 [Steve Henson and Ben Laurie]
2464 *) First cut of a cleanup for apps/. First the `ssleay' program is now named
2465 `openssl' and second, the shortcut symlinks for the `openssl <command>'
2466 are no longer created. This way we have a single and consistent command
2467 line interface `openssl <command>', similar to `cvs <command>'.
2468 [Ralf S. Engelschall, Paul Sutton and Ben Laurie]
2470 *) ca.c: move test for DSA keys inside #ifndef NO_DSA. Make pubkey
2471 BIT STRING wrapper always have zero unused bits.
2474 *) Add CA.pl, perl version of CA.sh, add extended key usage OID.
2477 *) Make the top-level INSTALL documentation easier to understand.
2480 *) Makefiles updated to exit if an error occurs in a sub-directory
2481 make (including if user presses ^C) [Paul Sutton]
2483 *) Make Montgomery context stuff explicit in RSA data structure.
2486 *) Fix build order of pem and err to allow for generated pem.h.
2489 *) Fix renumbering bug in X509_NAME_delete_entry().
2492 *) Enhanced the err-ins.pl script so it makes the error library number
2493 global and can add a library name. This is needed for external ASN1 and
2494 other error libraries.
2497 *) Fixed sk_insert which never worked properly.
2500 *) Fix ASN1 macros so they can handle indefinite length construted
2501 EXPLICIT tags. Some non standard certificates use these: they can now
2505 *) Merged the various old/obsolete SSLeay documentation files (doc/xxx.doc)
2506 into a single doc/ssleay.txt bundle. This way the information is still
2507 preserved but no longer messes up this directory. Now it's new room for
2508 the new set of documenation files.
2509 [Ralf S. Engelschall]
2511 *) SETs were incorrectly DER encoded. This was a major pain, because they
2512 shared code with SEQUENCEs, which aren't coded the same. This means that
2513 almost everything to do with SETs or SEQUENCEs has either changed name or
2514 number of arguments.
2515 [Ben Laurie, based on a partial fix by GP Jayan <gp@nsj.co.jp>]
2517 *) Fix test data to work with the above.
2520 *) Fix the RSA header declarations that hid a bug I fixed in 0.9.0b but
2521 was already fixed by Eric for 0.9.1 it seems.
2522 [Ben Laurie - pointed out by Ulf Möller <ulf@fitug.de>]
2524 *) Autodetect FreeBSD3.
2527 *) Fix various bugs in Configure. This affects the following platforms:
2531 unixware-2.0-pentium
2535 *) Eliminate generated files from CVS. Reorder tests to regenerate files
2536 before they are needed.
2539 *) Generate Makefile.ssl from Makefile.org (to keep CVS happy).
2543 Changes between 0.9.1b and 0.9.1c [23-Dec-1998]
2545 *) Added OPENSSL_VERSION_NUMBER to crypto/crypto.h and
2546 changed SSLeay to OpenSSL in version strings.
2547 [Ralf S. Engelschall]
2549 *) Some fixups to the top-level documents.
2552 *) Fixed the nasty bug where rsaref.h was not found under compile-time
2553 because the symlink to include/ was missing.
2554 [Ralf S. Engelschall]
2556 *) Incorporated the popular no-RSA/DSA-only patches
2557 which allow to compile a RSA-free SSLeay.
2558 [Andrew Cooke / Interrader Ldt., Ralf S. Engelschall]
2560 *) Fixed nasty rehash problem under `make -f Makefile.ssl links'
2561 when "ssleay" is still not found.
2562 [Ralf S. Engelschall]
2564 *) Added more platforms to Configure: Cray T3E, HPUX 11,
2565 [Ralf S. Engelschall, Beckmann <beckman@acl.lanl.gov>]
2567 *) Updated the README file.
2568 [Ralf S. Engelschall]
2570 *) Added various .cvsignore files in the CVS repository subdirs
2571 to make a "cvs update" really silent.
2572 [Ralf S. Engelschall]
2574 *) Recompiled the error-definition header files and added
2575 missing symbols to the Win32 linker tables.
2576 [Ralf S. Engelschall]
2578 *) Cleaned up the top-level documents;
2579 o new files: CHANGES and LICENSE
2580 o merged VERSION, HISTORY* and README* files a CHANGES.SSLeay
2581 o merged COPYRIGHT into LICENSE
2582 o removed obsolete TODO file
2583 o renamed MICROSOFT to INSTALL.W32
2584 [Ralf S. Engelschall]
2586 *) Removed dummy files from the 0.9.1b source tree:
2587 crypto/asn1/x crypto/bio/cd crypto/bio/fg crypto/bio/grep crypto/bio/vi
2588 crypto/bn/asm/......add.c crypto/bn/asm/a.out crypto/dsa/f crypto/md5/f
2589 crypto/pem/gmon.out crypto/perlasm/f crypto/pkcs7/build crypto/rsa/f
2590 crypto/sha/asm/f crypto/threads/f ms/zzz ssl/f ssl/f.mak test/f
2591 util/f.mak util/pl/f util/pl/f.mak crypto/bf/bf_locl.old apps/f
2592 [Ralf S. Engelschall]
2594 *) Added various platform portability fixes.
2597 *) The Genesis of the OpenSSL rpject:
2598 We start with the latest (unreleased) SSLeay version 0.9.1b which Eric A.
2599 Young and Tim J. Hudson created while they were working for C2Net until
2601 [The OpenSSL Project]
2604 Changes between 0.9.0b and 0.9.1b [not released]
2606 *) Updated a few CA certificates under certs/
2609 *) Changed some BIGNUM api stuff.
2612 *) Various platform ports: OpenBSD, Ultrix, IRIX 64bit, NetBSD,
2613 DGUX x86, Linux Alpha, etc.
2616 *) New COMP library [crypto/comp/] for SSL Record Layer Compression:
2617 RLE (dummy implemented) and ZLIB (really implemented when ZLIB is
2621 *) Add -strparse option to asn1pars program which parses nested
2623 [Dr Stephen Henson <shenson@bigfoot.com>]
2625 *) Added "oid_file" to ssleay.cnf for "ca" and "req" programs.
2628 *) DSA fix for "ca" program.
2631 *) Added "-genkey" option to "dsaparam" program.
2634 *) Added RIPE MD160 (rmd160) message digest.
2637 *) Added -a (all) option to "ssleay version" command.
2640 *) Added PLATFORM define which is the id given to Configure.
2643 *) Added MemCheck_XXXX functions to crypto/mem.c for memory checking.
2646 *) Extended the ASN.1 parser routines.
2649 *) Extended BIO routines to support REUSEADDR, seek, tell, etc.
2652 *) Added a BN_CTX to the BN library.
2655 *) Fixed the weak key values in DES library
2658 *) Changed API in EVP library for cipher aliases.
2661 *) Added support for RC2/64bit cipher.
2664 *) Converted the lhash library to the crypto/mem.c functions.
2667 *) Added more recognized ASN.1 object ids.
2670 *) Added more RSA padding checks for SSL/TLS.
2673 *) Added BIO proxy/filter functionality.
2676 *) Added extra_certs to SSL_CTX which can be used
2677 send extra CA certificates to the client in the CA cert chain sending
2678 process. It can be configured with SSL_CTX_add_extra_chain_cert().
2681 *) Now Fortezza is denied in the authentication phase because
2682 this is key exchange mechanism is not supported by SSLeay at all.
2685 *) Additional PKCS1 checks.
2688 *) Support the string "TLSv1" for all TLS v1 ciphers.
2691 *) Added function SSL_get_ex_data_X509_STORE_CTX_idx() which gives the
2692 ex_data index of the SSL context in the X509_STORE_CTX ex_data.
2695 *) Fixed a few memory leaks.
2698 *) Fixed various code and comment typos.
2701 *) A minor bug in ssl/s3_clnt.c where there would always be 4 0
2702 bytes sent in the client random.
2703 [Edward Bishop <ebishop@spyglass.com>]
projects / openssl.git / blob