5 Changes between 0.9.4 and 0.9.5 [xx XXX 2000]
7 *) New 'passwd' tool, currently only -fcrypt is implemented.
10 *) Add command line password options to the remaining applications.
13 *) Bug fix for BN_div_recp() for numerators with an even number of
17 *) More tests in bntest.c, and changed test_bn output.
20 *) ./config recognizes MacOS X now.
23 *) Bug fix for BN_div() when the first words of num and divsor are
24 equal (it gave wrong results if (rem=(n1-q*d0)&BN_MASK2) < d0).
27 *) Add support for various broken PKCS#8 formats, and command line
28 options to produce them.
31 *) New functions BN_CTX_start(), BN_CTX_get() and BT_CTX_end() to
32 get temporary BIGNUMs from a BN_CTX.
35 *) Correct return values in BN_mod_exp_mont() and BN_mod_exp2_mont()
39 *) Change the SSLeay_add_all_*() functions to OpenSSL_add_all_*() and
40 include a #define from the old name to the new. The original intent
41 was that statically linked binaries could for example just call
42 SSLeay_add_all_ciphers() to just add ciphers to the table and not
43 link with digests. This never worked becayse SSLeay_add_all_digests()
44 and SSLeay_add_all_ciphers() were in the same source file so calling
45 one would link with the other. They are now in separate source files.
48 *) Add a new -notext option to 'ca' and a -pubkey option to 'spkac'.
51 *) Use a less unusual form of the Miller-Rabin primality test (it used
52 a binary algorithm for exponentiation integrated into the Miller-Rabin
53 loop, our standard modexp algorithms are faster).
56 *) Support for the EBCDIC character set completed.
57 [Martin Kraemer <Martin.Kraemer@Mch.SNI.De>]
59 *) Source code cleanups: use const where appropriate, eliminate casts,
60 use void * instead of char * in lhash.
63 *) Bugfix: ssl3_send_server_key_exchange was not restartable
64 (the state was not changed to SSL3_ST_SW_KEY_EXCH_B, and because of
65 this the server could overwrite ephemeral keys that the client
69 *) Turn DSA_is_prime into a macro that calls BN_is_prime,
70 using 50 iterations of the Rabin-Miller test.
72 DSA_generate_parameters now uses BN_is_prime_fasttest (with 50
73 iterations of the Rabin-Miller test as required by the appendix
74 to FIPS PUB 186[-1]) instead of DSA_is_prime.
75 As BN_is_prime_fasttest includes trial division, DSA parameter
76 generation becomes much faster.
78 This implies a change for the callback functions in DSA_is_prime
79 and DSA_generate_parameters: The callback function is called once
80 for each positive witness in the Rabin-Miller test, not just
81 occasionally in the inner loop; and the parameters to the
82 callback function now provide an iteration count for the outer
83 loop rather than for the current invocation of the inner loop.
84 DSA_generate_parameters additionally can call the callback
85 function with an 'iteration count' of -1, meaning that a
86 candidate has passed the trial division test (when q is generated
87 from an application-provided seed, trial division is skipped).
90 *) New function BN_is_prime_fasttest that optionally does trial
91 division before starting the Rabin-Miller test and has
92 an additional BN_CTX * argument (whereas BN_is_prime always
93 has to allocate at least one BN_CTX).
94 'callback(1, -1, cb_arg)' is called when a number has passed the
98 *) Fix for bug in CRL encoding. The validity dates weren't being handled
102 *) New -pkcs12 option to CA.pl script to write out a PKCS#12 file.
105 *) New function BN_pseudo_rand().
108 *) Clean up BN_mod_mul_montgomery(): replace the broken (and unreadable)
109 bignum version of BN_from_montgomery() with the working code from
110 SSLeay 0.9.0 (the word based version is faster anyway), and clean up
114 *) Avoid a race condition in s2_clnt.c (function get_server_hello) that
115 made it impossible to use the same SSL_SESSION data structure in
116 SSL2 clients in multiple threads.
119 *) The return value of RAND_load_file() no longer counts bytes obtained
120 by stat(). RAND_load_file(..., -1) is new and uses the complete file
121 to seed the PRNG (previously an explicit byte count was required).
122 [Ulf Möller, Bodo Möller]
124 *) Clean up CRYPTO_EX_DATA functions, some of these didn't have prototypes
125 used (char *) instead of (void *) and had casts all over the place.
128 *) Make BN_generate_prime() return NULL on error if ret!=NULL.
131 *) Retain source code compatibility for BN_prime_checks macro:
132 BN_is_prime(..., BN_prime_checks, ...) now uses
133 BN_prime_checks_for_size to determine the appropriate number of
134 Rabin-Miller iterations.
137 *) Diffie-Hellman uses "safe" primes: DH_check() return code renamed to
138 DH_CHECK_P_NOT_SAFE_PRIME.
139 (Check if this is true? OpenPGP calls them "strong".)
142 *) Merge the functionality of "dh" and "gendh" programs into a new program
143 "dhparam". The old programs are retained for now but will handle DH keys
144 (instead of parameters) in future.
147 *) Make the ciphers, s_server and s_client programs check the return values
148 when a new cipher list is set.
151 *) Enhance the SSL/TLS cipher mechanism to correctly handle the TLS 56bit
152 ciphers. Before when the 56bit ciphers were enabled the sorting was
155 The syntax for the cipher sorting has been extended to support sorting by
156 cipher-strength (using the strength_bits hard coded in the tables).
157 The new command is "@STRENGTH" (see also doc/apps/ciphers.pod).
159 Fix a bug in the cipher-command parser: when supplying a cipher command
160 string with an "undefined" symbol (neither command nor alphanumeric
161 [A-Za-z0-9], ssl_set_cipher_list used to hang in an endless loop. Now
164 Due to the strength-sorting extension, the code of the
165 ssl_create_cipher_list() function was completely rearranged. I hope that
166 the readability was also increased :-)
167 [Lutz Jaenicke <Lutz.Jaenicke@aet.TU-Cottbus.DE>]
169 *) Minor change to 'x509' utility. The -CAcreateserial option now uses 1
170 for the first serial number and places 2 in the serial number file. This
171 avoids problems when the root CA is created with serial number zero and
172 the first user certificate has the same issuer name and serial number
176 *) Fixes to X509_ATTRIBUTE utilities, change the 'req' program so it uses
177 the new code. Add documentation for this stuff.
180 *) Changes to X509_ATTRIBUTE utilities. These have been renamed from
181 X509_*() to X509at_*() on the grounds that they don't handle X509
182 structures and behave in an analagous way to the X509v3 functions:
183 they shouldn't be called directly but wrapper functions should be used
186 So we also now have some wrapper functions that call the X509at functions
187 when passed certificate requests. (TO DO: similar things can be done with
188 PKCS#7 signed and unsigned attributes, PKCS#12 attributes and a few other
189 things. Some of these need some d2i or i2d and print functionality
190 because they handle more complex structures.)
193 *) Add missing #ifndefs that caused missing symbols when building libssl
194 as a shared library without RSA. Use #ifndef NO_SSL2 instead of
196 [Kris Kennaway <kris@hub.freebsd.org>, modified by Ulf Möller]
198 *) Precautions against using the PRNG uninitialized: RAND_bytes() now
199 has a return value which indicates the quality of the random data
200 (1 = ok, 0 = not seeded). Also an error is recorded on the thread's
201 error queue. New function RAND_pseudo_bytes() generates output that is
202 guaranteed to be unique but not unpredictable.
205 *) Do more iterations of Rabin-Miller probable prime test (specifically,
206 3 for 1024-bit primes, 6 for 512-bit primes, 12 for 256-bit primes
207 instead of only 2 for all lengths; see BN_prime_checks_for_size definition
208 in crypto/bn/bn_prime.c for the complete table). This guarantees a
209 false-positive rate of at most 2^-80 for random input.
212 *) Rewrite ssl3_read_n (ssl/s3_pkt.c) avoiding a couple of bugs.
215 *) New function X509_CTX_rget_chain(), this returns the chain
216 from an X509_CTX structure with a dup of the stack and all
217 the X509 reference counts upped: so the stack will exist
218 after X509_CTX_cleanup() has been called. Modify pkcs12.c
221 Also make SSL_SESSION_print() print out the verify return
225 *) Add manpage for the pkcs12 command. Also change the default
226 behaviour so MAC iteration counts are used unless the new
227 -nomaciter option is used. This improves file security and
228 only older versions of MSIE (4.0 for example) need it.
231 *) Honor the no-xxx Configure options when creating .DEF files.
234 *) Add PKCS#10 attributes to field table: challengePassword,
235 unstructuredName and unstructuredAddress. These are taken from
236 draft PKCS#9 v2.0 but are compatible with v1.2 provided no
237 international characters are used.
239 More changes to X509_ATTRIBUTE code: allow the setting of types
240 based on strings. Remove the 'loc' parameter when adding
241 attributes because these will be a SET OF encoding which is sorted
245 *) Initial changes to the 'req' utility to allow request generation
246 automation. This will allow an application to just generate a template
247 file containing all the field values and have req construct the
250 Initial support for X509_ATTRIBUTE handling. Stacks of these are
251 used all over the place including certificate requests and PKCS#7
252 structures. They are currently handled manually where necessary with
253 some primitive wrappers for PKCS#7. The new functions behave in a
254 manner analogous to the X509 extension functions: they allow
255 attributes to be looked up by NID and added.
257 Later something similar to the X509V3 code would be desirable to
258 automatically handle the encoding, decoding and printing of the
259 more complex types. The string types like challengePassword can
260 be handled by the string table functions.
262 Also modified the multi byte string table handling. Now there is
263 a 'global mask' which masks out certain types. The table itself
264 can use the flag STABLE_NO_MASK to ignore the mask setting: this
265 is useful when for example there is only one permissible type
266 (as in countryName) and using the mask might result in no valid
270 *) Clean up 'Finished' handling, and add functions SSL_get_finished and
271 SSL_get_peer_finished to allow applications to obtain the latest
272 Finished messages sent to the peer or expected from the peer,
273 respectively. (SSL_get_peer_finished is usually the Finished message
274 actually received from the peer, otherwise the protocol will be aborted.)
276 As the Finished message are message digests of the complete handshake
277 (with a total of 192 bits for TLS 1.0 and more for SSL 3.0), they can
278 be used for external authentication procedures when the authentication
279 provided by SSL/TLS is not desired or is not enough.
282 *) Enhanced support for Alpha Linux is added. Now ./config checks if
283 the host supports BWX extension and if Compaq C is present on the
284 $PATH. Just exploiting of the BWX extension results in 20-30%
285 performance kick for some algorithms, e.g. DES and RC4 to mention
286 a couple. Compaq C in turn generates ~20% faster code for MD5 and
290 *) Add support for MS "fast SGC". This is arguably a violation of the
291 SSL3/TLS protocol. Netscape SGC does two handshakes: the first with
292 weak crypto and after checking the certificate is SGC a second one
293 with strong crypto. MS SGC stops the first handshake after receiving
294 the server certificate message and sends a second client hello. Since
295 a server will typically do all the time consuming operations before
296 expecting any further messages from the client (server key exchange
297 is the most expensive) there is little difference between the two.
299 To get OpenSSL to support MS SGC we have to permit a second client
300 hello message after we have sent server done. In addition we have to
301 reset the MAC if we do get this second client hello and include the
305 *) Add a function 'd2i_AutoPrivateKey()' this will automatically decide
306 if a DER encoded private key is RSA or DSA traditional format. Changed
307 d2i_PrivateKey_bio() to use it. This is only needed for the "traditional"
308 format DER encoded private key. Newer code should use PKCS#8 format which
309 has the key type encoded in the ASN1 structure. Added DER private key
310 support to pkcs8 application.
313 *) SSL 3/TLS 1 servers now don't request certificates when an anonymous
314 ciphersuites has been selected (as required by the SSL 3/TLS 1
315 specifications). Exception: When SSL_VERIFY_FAIL_IF_NO_PEER_CERT
316 is set, we interpret this as a request to violate the specification
317 (the worst that can happen is a handshake failure, and 'correct'
318 behaviour would result in a handshake failure anyway).
321 *) In SSL_CTX_add_session, take into account that there might be multiple
322 SSL_SESSION structures with the same session ID (e.g. when two threads
323 concurrently obtain them from an external cache).
324 The internal cache can handle only one SSL_SESSION with a given ID,
325 so if there's a conflict, we now throw out the old one to achieve
329 *) Add OIDs for idea and blowfish in CBC mode. This will allow both
330 to be used in PKCS#5 v2.0 and S/MIME. Also add checking to
331 some routines that use cipher OIDs: some ciphers do not have OIDs
332 defined and so they cannot be used for S/MIME and PKCS#5 v2.0 for
336 *) Simplify the trust setting structure and code. Now we just have
337 two sequences of OIDs for trusted and rejected settings. These will
338 typically have values the same as the extended key usage extension
339 and any application specific purposes.
341 The trust checking code now has a default behaviour: it will just
342 check for an object with the same NID as the passed id. Functions can
343 be provided to override either the default behaviour or the behaviour
344 for a given id. SSL client, server and email already have functions
345 in place for compatibility: they check the NID and also return "trusted"
346 if the certificate is self signed.
349 *) Add d2i,i2d bio/fp functions for PrivateKey: these convert the
350 traditional format into an EVP_PKEY structure.
353 *) Add a password callback function PEM_cb() which either prompts for
354 a password if usr_data is NULL or otherwise assumes it is a null
355 terminated password. Allow passwords to be passed on command line
356 environment or config files in a few more utilities.
359 *) Add a bunch of DER and PEM functions to handle PKCS#8 format private
360 keys. Add some short names for PKCS#8 PBE algorithms and allow them
361 to be specified on the command line for the pkcs8 and pkcs12 utilities.
362 Update documentation.
365 *) Support for ASN1 "NULL" type. This could be handled before by using
366 ASN1_TYPE but there wasn't any function that would try to read a NULL
367 and produce an error if it couldn't. For compatibility we also have
368 ASN1_NULL_new() and ASN1_NULL_free() functions but these are faked and
369 don't allocate anything because they don't need to.
372 *) Initial support for MacOS is now provided. Examine INSTALL.MacOS
374 [Andy Polyakov, Roy Woods <roy@centicsystems.ca>]
376 *) Rebuild of the memory allocation routines used by OpenSSL code and
377 possibly others as well. The purpose is to make an interface that
378 provide hooks so anyone can build a separate set of allocation and
379 deallocation routines to be used by OpenSSL, for example memory
380 pool implementations, or something else, which was previously hard
381 since Malloc(), Realloc() and Free() were defined as macros having
382 the values malloc, realloc and free, respectively (except for Win32
383 compilations). The same is provided for memory debugging code.
384 OpenSSL already comes with functionality to find memory leaks, but
385 this gives people a chance to debug other memory problems.
387 With these changes, a new set of functions and macros have appeared:
389 CRYPTO_set_mem_debug_functions() [F]
390 CRYPTO_get_mem_debug_functions() [F]
391 CRYPTO_dbg_set_options() [F]
392 CRYPTO_dbg_get_options() [F]
393 CRYPTO_malloc_debug_init() [M]
395 The memory debug functions are NULL by default, unless the library
396 is compiled with CRYPTO_MDEBUG or friends is defined. If someone
397 wants to debug memory anyway, CRYPTO_malloc_debug_init() (which
398 gives the standard debugging functions that come with OpenSSL) or
399 CRYPTO_set_mem_debug_functions() (tells OpenSSL to use functions
400 provided by the library user) must be used. When the standard
401 debugging functions are used, CRYPTO_dbg_set_options can be used to
402 request additional information:
403 CRYPTO_dbg_set_options(V_CYRPTO_MDEBUG_xxx) corresponds to setting
404 the CRYPTO_MDEBUG_xxx macro when compiling the library.
406 Also, things like CRYPTO_set_mem_functions will always give the
407 expected result (the new set of functions is used for allocation
408 and deallocation) at all times, regardless of platform and compiler
411 To finish it up, some functions that were never use in any other
412 way than through macros have a new API and new semantic:
418 All macros of value have retained their old syntax.
419 [Richard Levitte and Bodo Moeller]
421 *) Some S/MIME fixes. The OID for SMIMECapabilities was wrong, the
422 ordering of SMIMECapabilities wasn't in "strength order" and there
423 was a missing NULL in the AlgorithmIdentifier for the SHA1 signature
427 *) Some ASN1 types with illegal zero length encoding (INTEGER,
428 ENUMERATED and OBJECT IDENTIFIER) choked the ASN1 routines.
429 [Frans Heymans <fheymans@isaserver.be>, modified by Steve Henson]
431 *) Merge in my S/MIME library for OpenSSL. This provides a simple
432 S/MIME API on top of the PKCS#7 code, a MIME parser (with enough
433 functionality to handle multipart/signed properly) and a utility
434 called 'smime' to call all this stuff. This is based on code I
435 originally wrote for Celo who have kindly allowed it to be
439 *) Add variants des_set_key_checked and des_set_key_unchecked of
440 des_set_key (aka des_key_sched). Global variable des_check_key
441 decides which of these is called by des_set_key; this way
442 des_check_key behaves as it always did, but applications and
443 the library itself, which was buggy for des_check_key == 1,
444 have a cleaner way to pick the version they need.
447 *) New function PKCS12_newpass() which changes the password of a
451 *) Modify X509_TRUST and X509_PURPOSE so it also uses a static and
452 dynamic mix. In both cases the ids can be used as an index into the
453 table. Also modified the X509_TRUST_add() and X509_PURPOSE_add()
454 functions so they accept a list of the field values and the
455 application doesn't need to directly manipulate the X509_TRUST
459 *) Modify the ASN1_STRING_TABLE stuff so it also uses bsearch and doesn't
463 *) Modify the way the V3 extension code looks up extensions. This now
464 works in a similar way to the object code: we have some "standard"
465 extensions in a static table which is searched with OBJ_bsearch()
466 and the application can add dynamic ones if needed. The file
467 crypto/x509v3/ext_dat.h now has the info: this file needs to be
468 updated whenever a new extension is added to the core code and kept
469 in ext_nid order. There is a simple program 'tabtest.c' which checks
470 this. New extensions are not added too often so this file can readily
471 be maintained manually.
473 There are two big advantages in doing things this way. The extensions
474 can be looked up immediately and no longer need to be "added" using
475 X509V3_add_standard_extensions(): this function now does nothing.
476 [Side note: I get *lots* of email saying the extension code doesn't
477 work because people forget to call this function]
478 Also no dynamic allocation is done unless new extensions are added:
479 so if we don't add custom extensions there is no need to call
480 X509V3_EXT_cleanup().
483 *) Modify enc utility's salting as follows: make salting the default. Add a
484 magic header, so unsalted files fail gracefully instead of just decrypting
485 to garbage. This is because not salting is a big security hole, so people
486 should be discouraged from doing it.
489 *) Fixes and enhancements to the 'x509' utility. It allowed a message
490 digest to be passed on the command line but it only used this
491 parameter when signing a certificate. Modified so all relevant
492 operations are affected by the digest parameter including the
493 -fingerprint and -x509toreq options. Also -x509toreq choked if a
494 DSA key was used because it didn't fix the digest.
497 *) Initial certificate chain verify code. Currently tests the untrusted
498 certificates for consistency with the verify purpose (which is set
499 when the X509_STORE_CTX structure is set up) and checks the pathlength.
501 There is a NO_CHAIN_VERIFY compilation option to keep the old behaviour:
502 this is because it will reject chains with invalid extensions whereas
503 every previous version of OpenSSL and SSLeay made no checks at all.
505 Trust code: checks the root CA for the relevant trust settings. Trust
506 settings have an initial value consistent with the verify purpose: e.g.
507 if the verify purpose is for SSL client use it expects the CA to be
508 trusted for SSL client use. However the default value can be changed to
509 permit custom trust settings: one example of this would be to only trust
510 certificates from a specific "secure" set of CAs.
512 Also added X509_STORE_CTX_new() and X509_STORE_CTX_free() functions
513 which should be used for version portability: especially since the
514 verify structure is likely to change more often now.
516 SSL integration. Add purpose and trust to SSL_CTX and SSL and functions
517 to set them. If not set then assume SSL clients will verify SSL servers
520 Two new options to the verify program: -untrusted allows a set of
521 untrusted certificates to be passed in and -purpose which sets the
522 intended purpose of the certificate. If a purpose is set then the
523 new chain verify code is used to check extension consistency.
526 *) Support for the authority information access extension.
529 *) Modify RSA and DSA PEM read routines to transparently handle
530 PKCS#8 format private keys. New *_PUBKEY_* functions that handle
531 public keys in a format compatible with certificate
532 SubjectPublicKeyInfo structures. Unfortunately there were already
533 functions called *_PublicKey_* which used various odd formats so
534 these are retained for compatibility: however the DSA variants were
535 never in a public release so they have been deleted. Changed dsa/rsa
536 utilities to handle the new format: note no releases ever handled public
537 keys so we should be OK.
539 The primary motivation for this change is to avoid the same fiasco
540 that dogs private keys: there are several incompatible private key
541 formats some of which are standard and some OpenSSL specific and
542 require various evil hacks to allow partial transparent handling and
543 even then it doesn't work with DER formats. Given the option anything
544 other than PKCS#8 should be dumped: but the other formats have to
545 stay in the name of compatibility.
547 With public keys and the benefit of hindsight one standard format
548 is used which works with EVP_PKEY, RSA or DSA structures: though
549 it clearly returns an error if you try to read the wrong kind of key.
551 Added a -pubkey option to the 'x509' utility to output the public key.
552 Also rename the EVP_PKEY_get_*() to EVP_PKEY_rget_*() and add
553 EVP_PKEY_rset_*() functions that do the same as the EVP_PKEY_assign_*()
554 except they up the reference count of the added key (they don't "swallow"
558 *) Fixes to crypto/x509/by_file.c the code to read in certificates and
559 CRLs would fail if the file contained no certificates or no CRLs:
560 added a new function to read in both types and return the number
561 read: this means that if none are read it will be an error. The
562 DER versions of the certificate and CRL reader would always fail
563 because it isn't possible to mix certificates and CRLs in DER format
564 without choking one or the other routine. Changed this to just read
565 a certificate: this is the best we can do. Also modified the code
566 in apps/verify.c to take notice of return codes: it was previously
567 attempting to read in certificates from NULL pointers and ignoring
568 any errors: this is one reason why the cert and CRL reader seemed
569 to work. It doesn't check return codes from the default certificate
570 routines: these may well fail if the certificates aren't installed.
573 *) Code to support otherName option in GeneralName.
576 *) First update to verify code. Change the verify utility
577 so it warns if it is passed a self signed certificate:
578 for consistency with the normal behaviour. X509_verify
579 has been modified to it will now verify a self signed
580 certificate if *exactly* the same certificate appears
581 in the store: it was previously impossible to trust a
582 single self signed certificate. This means that:
583 openssl verify ss.pem
584 now gives a warning about a self signed certificate but
585 openssl verify -CAfile ss.pem ss.pem
589 *) For servers, store verify_result in SSL_SESSION data structure
590 (and add it to external session representation).
591 This is needed when client certificate verifications fails,
592 but an application-provided verification callback (set by
593 SSL_CTX_set_cert_verify_callback) allows accepting the session
594 anyway (i.e. leaves x509_store_ctx->error != X509_V_OK
595 but returns 1): When the session is reused, we have to set
596 ssl->verify_result to the appropriate error code to avoid
598 [Bodo Moeller, problem pointed out by Lutz Jaenicke]
600 *) Fix a bug in the new PKCS#7 code: it didn't consider the
601 case in PKCS7_dataInit() where the signed PKCS7 structure
602 didn't contain any existing data because it was being created.
603 [Po-Cheng Chen <pocheng@nst.com.tw>, slightly modified by Steve Henson]
605 *) Add a salt to the key derivation routines in enc.c. This
606 forms the first 8 bytes of the encrypted file. Also add a
607 -S option to allow a salt to be input on the command line.
610 *) New function X509_cmp(). Oddly enough there wasn't a function
611 to compare two certificates. We do this by working out the SHA1
612 hash and comparing that. X509_cmp() will be needed by the trust
616 *) SSL_get1_session() is like SSL_get_session(), but increments
617 the reference count in the SSL_SESSION returned.
618 [Geoff Thorpe <geoff@eu.c2.net>]
620 *) Fix for 'req': it was adding a null to request attributes.
621 Also change the X509_LOOKUP and X509_INFO code to handle
622 certificate auxiliary information.
625 *) Add support for 40 and 64 bit RC2 and RC4 algorithms: document
629 *) Add the possibility to add extra information to the memory leak
630 detecting output, to form tracebacks, showing from where each
631 allocation was originated: CRYPTO_push_info("constant string") adds
632 the string plus current file name and line number to a per-thread
633 stack, CRYPTO_pop_info() does the obvious, CRYPTO_remove_all_info()
634 is like calling CYRPTO_pop_info() until the stack is empty.
635 Also updated memory leak detection code to be multi-thread-safe.
638 *) Add options -text and -noout to pkcs7 utility and delete the
639 encryption options which never did anything. Update docs.
642 *) Add options to some of the utilities to allow the pass phrase
643 to be included on either the command line (not recommended on
644 OSes like Unix) or read from the environment. Update the
645 manpages and fix a few bugs.
648 *) Add a few manpages for some of the openssl commands.
651 *) Fix the -revoke option in ca. It was freeing up memory twice,
652 leaking and not finding already revoked certificates.
655 *) Extensive changes to support certificate auxiliary information.
656 This involves the use of X509_CERT_AUX structure and X509_AUX
657 functions. An X509_AUX function such as PEM_read_X509_AUX()
658 can still read in a certificate file in the usual way but it
659 will also read in any additional "auxiliary information". By
660 doing things this way a fair degree of compatibility can be
661 retained: existing certificates can have this information added
662 using the new 'x509' options.
664 Current auxiliary information includes an "alias" and some trust
665 settings. The trust settings will ultimately be used in enhanced
666 certificate chain verification routines: currently a certificate
667 can only be trusted if it is self signed and then it is trusted
671 *) Fix assembler for Alpha (tested only on DEC OSF not Linux or *BSD).
672 The problem was that one of the replacement routines had not been working
673 since SSLeay releases. For now the offending routine has been replaced
674 with non-optimised assembler. Even so, this now gives around 95%
675 performance improvement for 1024 bit RSA signs.
678 *) Hack to fix PKCS#7 decryption when used with some unorthodox RC2
679 handling. Most clients have the effective key size in bits equal to
680 the key length in bits: so a 40 bit RC2 key uses a 40 bit (5 byte) key.
681 A few however don't do this and instead use the size of the decrypted key
682 to determine the RC2 key length and the AlgorithmIdentifier to determine
683 the effective key length. In this case the effective key length can still
684 be 40 bits but the key length can be 168 bits for example. This is fixed
685 by manually forcing an RC2 key into the EVP_PKEY structure because the
686 EVP code can't currently handle unusual RC2 key sizes: it always assumes
687 the key length and effective key length are equal.
690 *) Add a bunch of functions that should simplify the creation of
691 X509_NAME structures. Now you should be able to do:
692 X509_NAME_add_entry_by_txt(nm, "CN", MBSTRING_ASC, "Steve", -1, -1, 0);
693 and have it automatically work out the correct field type and fill in
694 the structures. The more adventurous can try:
695 X509_NAME_add_entry_by_txt(nm, field, MBSTRING_UTF8, str, -1, -1, 0);
696 and it will (hopefully) work out the correct multibyte encoding.
699 *) Change the 'req' utility to use the new field handling and multibyte
700 copy routines. Before the DN field creation was handled in an ad hoc
701 way in req, ca, and x509 which was rather broken and didn't support
702 BMPStrings or UTF8Strings. Since some software doesn't implement
703 BMPStrings or UTF8Strings yet, they can be enabled using the config file
704 using the dirstring_type option. See the new comment in the default
705 openssl.cnf for more info.
708 *) Make crypto/rand/md_rand.c more robust:
709 - Assure unique random numbers after fork().
710 - Make sure that concurrent threads access the global counter and
711 md serializably so that we never lose entropy in them
712 or use exactly the same state in multiple threads.
713 Access to the large state is not always serializable because
714 the additional locking could be a performance killer, and
715 md should be large enough anyway.
718 *) New file apps/app_rand.c with commonly needed functionality
719 for handling the random seed file.
721 Use the random seed file in some applications that previously did not:
723 dsaparam -genkey (which also ignored its '-rand' option),
727 Except on systems with /dev/urandom, it is crucial to have a random
728 seed file at least for key creation, DSA signing, and for DH exchanges;
729 for RSA signatures we could do without one.
731 gendh and gendsa (unlike genrsa) used to read only the first byte
732 of each file listed in the '-rand' option. The function as previously
733 found in genrsa is now in app_rand.c and is used by all programs
734 that support '-rand'.
737 *) In RAND_write_file, use mode 0600 for creating files;
738 don't just chmod when it may be too late.
741 *) Report an error from X509_STORE_load_locations
742 when X509_LOOKUP_load_file or X509_LOOKUP_add_dir failed.
745 *) New function ASN1_mbstring_copy() this copies a string in either
746 ASCII, Unicode, Universal (4 bytes per character) or UTF8 format
747 into an ASN1_STRING type. A mask of permissible types is passed
748 and it chooses the "minimal" type to use or an error if not type
752 *) Add function equivalents to the various macros in asn1.h. The old
753 macros are retained with an M_ prefix. Code inside the library can
754 use the M_ macros. External code (including the openssl utility)
755 should *NOT* in order to be "shared library friendly".
758 *) Add various functions that can check a certificate's extensions
759 to see if it usable for various purposes such as SSL client,
760 server or S/MIME and CAs of these types. This is currently
761 VERY EXPERIMENTAL but will ultimately be used for certificate chain
762 verification. Also added a -purpose flag to x509 utility to
763 print out all the purposes.
766 *) Add a CRYPTO_EX_DATA to X509 certificate structure and associated
770 *) New X509V3_{X509,CRL,REVOKED}_get_d2i() functions. These will search
771 for, obtain and decode and extension and obtain its critical flag.
772 This allows all the necessary extension code to be handled in a
773 single function call.
776 *) RC4 tune-up featuring 30-40% performance improvement on most RISC
777 platforms. See crypto/rc4/rc4_enc.c for further details.
780 *) New -noout option to asn1parse. This causes no output to be produced
781 its main use is when combined with -strparse and -out to extract data
782 from a file (which may not be in ASN.1 format).
785 *) Fix for pkcs12 program. It was hashing an invalid certificate pointer
786 when producing the local key id.
787 [Richard Levitte <levitte@stacken.kth.se>]
789 *) New option -dhparam in s_server. This allows a DH parameter file to be
790 stated explicitly. If it is not stated then it tries the first server
791 certificate file. The previous behaviour hard coded the filename
795 *) Add -pubin and -pubout options to the rsa and dsa commands. These allow
796 a public key to be input or output. For example:
797 openssl rsa -in key.pem -pubout -out pubkey.pem
798 Also added necessary DSA public key functions to handle this.
801 *) Fix so PKCS7_dataVerify() doesn't crash if no certificates are contained
802 in the message. This was handled by allowing
803 X509_find_by_issuer_and_serial() to tolerate a NULL passed to it.
804 [Steve Henson, reported by Sampo Kellomaki <sampo@mail.neuronio.pt>]
806 *) Fix for bug in d2i_ASN1_bytes(): other ASN1 functions add an extra null
807 to the end of the strings whereas this didn't. This would cause problems
808 if strings read with d2i_ASN1_bytes() were later modified.
809 [Steve Henson, reported by Arne Ansper <arne@ats.cyber.ee>]
811 *) Fix for base64 decode bug. When a base64 bio reads only one line of
812 data and it contains EOF it will end up returning an error. This is
813 caused by input 46 bytes long. The cause is due to the way base64
814 BIOs find the start of base64 encoded data. They do this by trying a
815 trial decode on each line until they find one that works. When they
816 do a flag is set and it starts again knowing it can pass all the
817 data directly through the decoder. Unfortunately it doesn't reset
818 the context it uses. This means that if EOF is reached an attempt
819 is made to pass two EOFs through the context and this causes the
820 resulting error. This can also cause other problems as well. As is
821 usual with these problems it takes *ages* to find and the fix is
822 trivial: move one line.
823 [Steve Henson, reported by ian@uns.ns.ac.yu (Ivan Nejgebauer) ]
825 *) Ugly workaround to get s_client and s_server working under Windows. The
826 old code wouldn't work because it needed to select() on sockets and the
827 tty (for keypresses and to see if data could be written). Win32 only
828 supports select() on sockets so we select() with a 1s timeout on the
829 sockets and then see if any characters are waiting to be read, if none
830 are present then we retry, we also assume we can always write data to
831 the tty. This isn't nice because the code then blocks until we've
832 received a complete line of data and it is effectively polling the
833 keyboard at 1s intervals: however it's quite a bit better than not
834 working at all :-) A dedicated Windows application might handle this
835 with an event loop for example.
838 *) Enhance RSA_METHOD structure. Now there are two extra methods, rsa_sign
839 and rsa_verify. When the RSA_FLAGS_SIGN_VER option is set these functions
840 will be called when RSA_sign() and RSA_verify() are used. This is useful
841 if rsa_pub_dec() and rsa_priv_enc() equivalents are not available.
842 For this to work properly RSA_public_decrypt() and RSA_private_encrypt()
843 should *not* be used: RSA_sign() and RSA_verify() must be used instead.
844 This necessitated the support of an extra signature type NID_md5_sha1
845 for SSL signatures and modifications to the SSL library to use it instead
846 of calling RSA_public_decrypt() and RSA_private_encrypt().
849 *) Add new -verify -CAfile and -CApath options to the crl program, these
850 will lookup a CRL issuers certificate and verify the signature in a
851 similar way to the verify program. Tidy up the crl program so it
852 no longer accesses structures directly. Make the ASN1 CRL parsing a bit
853 less strict. It will now permit CRL extensions even if it is not
854 a V2 CRL: this will allow it to tolerate some broken CRLs.
857 *) Initialize all non-automatic variables each time one of the openssl
858 sub-programs is started (this is necessary as they may be started
859 multiple times from the "OpenSSL>" prompt).
860 [Lennart Bang, Bodo Moeller]
862 *) Preliminary compilation option RSA_NULL which disables RSA crypto without
863 removing all other RSA functionality (this is what NO_RSA does). This
864 is so (for example) those in the US can disable those operations covered
865 by the RSA patent while allowing storage and parsing of RSA keys and RSA
869 *) Non-copying interface to BIO pairs.
870 (still largely untested)
873 *) New function ANS1_tag2str() to convert an ASN1 tag to a descriptive
874 ASCII string. This was handled independently in various places before.
877 *) New functions UTF8_getc() and UTF8_putc() that parse and generate
878 UTF8 strings a character at a time.
881 *) Use client_version from client hello to select the protocol
882 (s23_srvr.c) and for RSA client key exchange verification
883 (s3_srvr.c), as required by the SSL 3.0/TLS 1.0 specifications.
886 *) Add various utility functions to handle SPKACs, these were previously
887 handled by poking round in the structure internals. Added new function
888 NETSCAPE_SPKI_print() to print out SPKAC and a new utility 'spkac' to
889 print, verify and generate SPKACs. Based on an original idea from
890 Massimiliano Pala <madwolf@comune.modena.it> but extensively modified.
893 *) RIPEMD160 is operational on all platforms and is back in 'make test'.
896 *) Allow the config file extension section to be overwritten on the
897 command line. Based on an original idea from Massimiliano Pala
898 <madwolf@comune.modena.it>. The new option is called -extensions
899 and can be applied to ca, req and x509. Also -reqexts to override
900 the request extensions in req and -crlexts to override the crl extensions
904 *) Add new feature to the SPKAC handling in ca. Now you can include
905 the same field multiple times by preceding it by "XXXX." for example:
908 this is the same syntax as used in the req config file.
911 *) Allow certificate extensions to be added to certificate requests. These
912 are specified in a 'req_extensions' option of the req section of the
913 config file. They can be printed out with the -text option to req but
914 are otherwise ignored at present.
917 *) Fix a horrible bug in enc_read() in crypto/evp/bio_enc.c: if the first
918 data read consists of only the final block it would not decrypted because
919 EVP_CipherUpdate() would correctly report zero bytes had been decrypted.
920 A misplaced 'break' also meant the decrypted final block might not be
921 copied until the next read.
924 *) Initial support for DH_METHOD. Again based on RSA_METHOD. Also added
925 a few extra parameters to the DH structure: these will be useful if
926 for example we want the value of 'q' or implement X9.42 DH.
929 *) Initial support for DSA_METHOD. This is based on the RSA_METHOD and
930 provides hooks that allow the default DSA functions or functions on a
931 "per key" basis to be replaced. This allows hardware acceleration and
932 hardware key storage to be handled without major modification to the
933 library. Also added low level modexp hooks and CRYPTO_EX structure and
934 associated functions.
937 *) Add a new flag to memory BIOs, BIO_FLAG_MEM_RDONLY. This marks the BIO
938 as "read only": it can't be written to and the buffer it points to will
939 not be freed. Reading from a read only BIO is much more efficient than
940 a normal memory BIO. This was added because there are several times when
941 an area of memory needs to be read from a BIO. The previous method was
942 to create a memory BIO and write the data to it, this results in two
943 copies of the data and an O(n^2) reading algorithm. There is a new
944 function BIO_new_mem_buf() which creates a read only memory BIO from
945 an area of memory. Also modified the PKCS#7 routines to use read only
949 *) Bugfix: ssl23_get_client_hello did not work properly when called in
950 state SSL23_ST_SR_CLNT_HELLO_B, i.e. when the first 7 bytes of
951 a SSLv2-compatible client hello for SSLv3 or TLSv1 could be read,
952 but a retry condition occured while trying to read the rest.
955 *) The PKCS7_ENC_CONTENT_new() function was setting the content type as
956 NID_pkcs7_encrypted by default: this was wrong since this should almost
957 always be NID_pkcs7_data. Also modified the PKCS7_set_type() to handle
958 the encrypted data type: this is a more sensible place to put it and it
959 allows the PKCS#12 code to be tidied up that duplicated this
963 *) Changed obj_dat.pl script so it takes its input and output files on
964 the command line. This should avoid shell escape redirection problems
968 *) Initial support for certificate extension requests, these are included
969 in things like Xenroll certificate requests. Included functions to allow
970 extensions to be obtained and added.
973 *) -crlf option to s_client and s_server for sending newlines as
974 CRLF (as required by many protocols).
977 Changes between 0.9.3a and 0.9.4 [09 Aug 1999]
979 *) Install libRSAglue.a when OpenSSL is built with RSAref.
980 [Ralf S. Engelschall]
982 *) A few more ``#ifndef NO_FP_API / #endif'' pairs for consistency.
983 [Andrija Antonijevic <TheAntony2@bigfoot.com>]
985 *) Fix -startdate and -enddate (which was missing) arguments to 'ca'
989 *) New function DSA_dup_DH, which duplicates DSA parameters/keys as
990 DH parameters/keys (q is lost during that conversion, but the resulting
991 DH parameters contain its length).
993 For 1024-bit p, DSA_generate_parameters followed by DSA_dup_DH is
994 much faster than DH_generate_parameters (which creates parameters
995 where p = 2*q + 1), and also the smaller q makes DH computations
996 much more efficient (160-bit exponentiation instead of 1024-bit
997 exponentiation); so this provides a convenient way to support DHE
998 ciphersuites in SSL/TLS servers (see ssl/ssltest.c). It is of
999 utter importance to use
1000 SSL_CTX_set_options(s_ctx, SSL_OP_SINGLE_DH_USE);
1002 SSL_set_options(s_ctx, SSL_OP_SINGLE_DH_USE);
1003 when such DH parameters are used, because otherwise small subgroup
1004 attacks may become possible!
1007 *) Avoid memory leak in i2d_DHparams.
1010 *) Allow the -k option to be used more than once in the enc program:
1011 this allows the same encrypted message to be read by multiple recipients.
1014 *) New function OBJ_obj2txt(buf, buf_len, a, no_name), this converts
1015 an ASN1_OBJECT to a text string. If the "no_name" parameter is set then
1016 it will always use the numerical form of the OID, even if it has a short
1020 *) Added an extra RSA flag: RSA_FLAG_EXT_PKEY. Previously the rsa_mod_exp
1021 method only got called if p,q,dmp1,dmq1,iqmp components were present,
1022 otherwise bn_mod_exp was called. In the case of hardware keys for example
1023 no private key components need be present and it might store extra data
1024 in the RSA structure, which cannot be accessed from bn_mod_exp.
1025 By setting RSA_FLAG_EXT_PKEY rsa_mod_exp will always be called for
1026 private key operations.
1029 *) Added support for SPARC Linux.
1032 *) pem_password_cb function type incompatibly changed from
1033 typedef int pem_password_cb(char *buf, int size, int rwflag);
1035 ....(char *buf, int size, int rwflag, void *userdata);
1036 so that applications can pass data to their callbacks:
1037 The PEM[_ASN1]_{read,write}... functions and macros now take an
1038 additional void * argument, which is just handed through whenever
1039 the password callback is called.
1040 [Damien Miller <dmiller@ilogic.com.au>; tiny changes by Bodo Moeller]
1042 New function SSL_CTX_set_default_passwd_cb_userdata.
1044 Compatibility note: As many C implementations push function arguments
1045 onto the stack in reverse order, the new library version is likely to
1046 interoperate with programs that have been compiled with the old
1047 pem_password_cb definition (PEM_whatever takes some data that
1048 happens to be on the stack as its last argument, and the callback
1049 just ignores this garbage); but there is no guarantee whatsoever that
1052 *) The -DPLATFORM="\"$(PLATFORM)\"" definition and the similar -DCFLAGS=...
1053 (both in crypto/Makefile.ssl for use by crypto/cversion.c) caused
1054 problems not only on Windows, but also on some Unix platforms.
1055 To avoid problematic command lines, these definitions are now in an
1056 auto-generated file crypto/buildinf.h (created by crypto/Makefile.ssl
1057 for standard "make" builds, by util/mk1mf.pl for "mk1mf" builds).
1060 *) MIPS III/IV assembler module is reimplemented.
1063 *) More DES library cleanups: remove references to srand/rand and
1064 delete an unused file.
1067 *) Add support for the the free Netwide assembler (NASM) under Win32,
1068 since not many people have MASM (ml) and it can be hard to obtain.
1069 This is currently experimental but it seems to work OK and pass all
1070 the tests. Check out INSTALL.W32 for info.
1073 *) Fix memory leaks in s3_clnt.c: All non-anonymous SSL3/TLS1 connections
1074 without temporary keys kept an extra copy of the server key,
1075 and connections with temporary keys did not free everything in case
1079 *) New function RSA_check_key and new openssl rsa option -check
1080 for verifying the consistency of RSA keys.
1081 [Ulf Moeller, Bodo Moeller]
1083 *) Various changes to make Win32 compile work:
1084 1. Casts to avoid "loss of data" warnings in p5_crpt2.c
1085 2. Change unsigned int to int in b_dump.c to avoid "signed/unsigned
1086 comparison" warnings.
1087 3. Add sk_<TYPE>_sort to DEF file generator and do make update.
1090 *) Add a debugging option to PKCS#5 v2 key generation function: when
1091 you #define DEBUG_PKCS5V2 passwords, salts, iteration counts and
1092 derived keys are printed to stderr.
1095 *) Copy the flags in ASN1_STRING_dup().
1096 [Roman E. Pavlov <pre@mo.msk.ru>]
1098 *) The x509 application mishandled signing requests containing DSA
1099 keys when the signing key was also DSA and the parameters didn't match.
1101 It was supposed to omit the parameters when they matched the signing key:
1102 the verifying software was then supposed to automatically use the CA's
1103 parameters if they were absent from the end user certificate.
1105 Omitting parameters is no longer recommended. The test was also
1106 the wrong way round! This was probably due to unusual behaviour in
1107 EVP_cmp_parameters() which returns 1 if the parameters match.
1108 This meant that parameters were omitted when they *didn't* match and
1109 the certificate was useless. Certificates signed with 'ca' didn't have
1111 [Steve Henson, reported by Doug Erickson <Doug.Erickson@Part.NET>]
1113 *) Memory leak checking (-DCRYPTO_MDEBUG) had some problems.
1114 The interface is as follows:
1115 Applications can use
1116 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ON) aka MemCheck_start(),
1117 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_OFF) aka MemCheck_stop();
1118 "off" is now the default.
1119 The library internally uses
1120 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_DISABLE) aka MemCheck_off(),
1121 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ENABLE) aka MemCheck_on()
1122 to disable memory-checking temporarily.
1124 Some inconsistent states that previously were possible (and were
1125 even the default) are now avoided.
1127 -DCRYPTO_MDEBUG_TIME is new and additionally stores the current time
1128 with each memory chunk allocated; this is occasionally more helpful
1129 than just having a counter.
1131 -DCRYPTO_MDEBUG_THREAD is also new and adds the thread ID.
1133 -DCRYPTO_MDEBUG_ALL enables all of the above, plus any future
1137 *) Introduce "mode" for SSL structures (with defaults in SSL_CTX),
1138 which largely parallels "options", but is for changing API behaviour,
1139 whereas "options" are about protocol behaviour.
1140 Initial "mode" flags are:
1142 SSL_MODE_ENABLE_PARTIAL_WRITE Allow SSL_write to report success when
1143 a single record has been written.
1144 SSL_MODE_ACCEPT_MOVING_WRITE_BUFFER Don't insist that SSL_write
1145 retries use the same buffer location.
1146 (But all of the contents must be
1150 *) Bugfix: SSL_set_mode ignored its parameter, only SSL_CTX_set_mode
1153 *) Fix problems with no-hmac etc.
1154 [Ulf Möller, pointed out by Brian Wellington <bwelling@tislabs.com>]
1156 *) New functions RSA_get_default_method(), RSA_set_method() and
1157 RSA_get_method(). These allows replacement of RSA_METHODs without having
1158 to mess around with the internals of an RSA structure.
1161 *) Fix memory leaks in DSA_do_sign and DSA_is_prime.
1162 Also really enable memory leak checks in openssl.c and in some
1164 [Chad C. Mulligan, Bodo Moeller]
1166 *) Fix a bug in d2i_ASN1_INTEGER() and i2d_ASN1_INTEGER() which can mess
1167 up the length of negative integers. This has now been simplified to just
1168 store the length when it is first determined and use it later, rather
1169 than trying to keep track of where data is copied and updating it to
1171 [Steve Henson, reported by Brien Wheeler
1172 <bwheeler@authentica-security.com>]
1174 *) Add a new function PKCS7_signatureVerify. This allows the verification
1175 of a PKCS#7 signature but with the signing certificate passed to the
1176 function itself. This contrasts with PKCS7_dataVerify which assumes the
1177 certificate is present in the PKCS#7 structure. This isn't always the
1178 case: certificates can be omitted from a PKCS#7 structure and be
1179 distributed by "out of band" means (such as a certificate database).
1182 *) Complete the PEM_* macros with DECLARE_PEM versions to replace the
1183 function prototypes in pem.h, also change util/mkdef.pl to add the
1184 necessary function names.
1187 *) mk1mf.pl (used by Windows builds) did not properly read the
1188 options set by Configure in the top level Makefile, and Configure
1189 was not even able to write more than one option correctly.
1190 Fixed, now "no-idea no-rc5 -DCRYPTO_MDEBUG" etc. works as intended.
1193 *) New functions CONF_load_bio() and CONF_load_fp() to allow a config
1194 file to be loaded from a BIO or FILE pointer. The BIO version will
1195 for example allow memory BIOs to contain config info.
1198 *) New function "CRYPTO_num_locks" that returns CRYPTO_NUM_LOCKS.
1199 Whoever hopes to achieve shared-library compatibility across versions
1200 must use this, not the compile-time macro.
1201 (Exercise 0.9.4: Which is the minimum library version required by
1203 Note: All this applies only to multi-threaded programs, others don't
1207 *) Add missing case to s3_clnt.c state machine -- one of the new SSL tests
1208 through a BIO pair triggered the default case, i.e.
1209 SSLerr(...,SSL_R_UNKNOWN_STATE).
1212 *) New "BIO pair" concept (crypto/bio/bss_bio.c) so that applications
1213 can use the SSL library even if none of the specific BIOs is
1217 *) Fix a bug in i2d_DSAPublicKey() which meant it returned the wrong value
1218 for the encoded length.
1219 [Jeon KyoungHo <khjeon@sds.samsung.co.kr>]
1221 *) Add initial documentation of the X509V3 functions.
1224 *) Add a new pair of functions PEM_write_PKCS8PrivateKey() and
1225 PEM_write_bio_PKCS8PrivateKey() that are equivalent to
1226 PEM_write_PrivateKey() and PEM_write_bio_PrivateKey() but use the more
1227 secure PKCS#8 private key format with a high iteration count.
1230 *) Fix determination of Perl interpreter: A perl or perl5
1231 _directory_ in $PATH was also accepted as the interpreter.
1232 [Ralf S. Engelschall]
1234 *) Fix demos/sign/sign.c: well there wasn't anything strictly speaking
1235 wrong with it but it was very old and did things like calling
1236 PEM_ASN1_read() directly and used MD5 for the hash not to mention some
1240 *) Fix demos/selfsign.c: it used obsolete and deleted functions, changed
1241 to use the new extension code.
1244 *) Implement the PEM_read/PEM_write functions in crypto/pem/pem_all.c
1245 with macros. This should make it easier to change their form, add extra
1246 arguments etc. Fix a few PEM prototypes which didn't have cipher as a
1250 *) Add to configuration table a new entry that can specify an alternative
1251 name for unistd.h (for pre-POSIX systems); we need this for NeXTstep,
1252 according to Mark Crispin <MRC@Panda.COM>.
1256 *) DES CBC did not update the IV. Weird.
1259 des_cbc_encrypt does not update the IV, but des_ncbc_encrypt does.
1260 Changing the behaviour of the former might break existing programs --
1261 where IV updating is needed, des_ncbc_encrypt can be used.
1264 *) When bntest is run from "make test" it drives bc to check its
1265 calculations, as well as internally checking them. If an internal check
1266 fails, it needs to cause bc to give a non-zero result or make test carries
1267 on without noticing the failure. Fixed.
1270 *) DES library cleanups.
1273 *) Add support for PKCS#5 v2.0 PBE algorithms. This will permit PKCS#8 to be
1274 used with any cipher unlike PKCS#5 v1.5 which can at most handle 64 bit
1275 ciphers. NOTE: although the key derivation function has been verified
1276 against some published test vectors it has not been extensively tested
1277 yet. Added a -v2 "cipher" option to pkcs8 application to allow the use
1281 *) Instead of "mkdir -p", which is not fully portable, use new
1282 Perl script "util/mkdir-p.pl".
1285 *) Rewrite the way password based encryption (PBE) is handled. It used to
1286 assume that the ASN1 AlgorithmIdentifier parameter was a PBEParameter
1287 structure. This was true for the PKCS#5 v1.5 and PKCS#12 PBE algorithms
1288 but doesn't apply to PKCS#5 v2.0 where it can be something else. Now
1289 the 'parameter' field of the AlgorithmIdentifier is passed to the
1290 underlying key generation function so it must do its own ASN1 parsing.
1291 This has also changed the EVP_PBE_CipherInit() function which now has a
1292 'parameter' argument instead of literal salt and iteration count values
1293 and the function EVP_PBE_ALGOR_CipherInit() has been deleted.
1296 *) Support for PKCS#5 v1.5 compatible password based encryption algorithms
1297 and PKCS#8 functionality. New 'pkcs8' application linked to openssl.
1298 Needed to change the PEM_STRING_EVP_PKEY value which was just "PRIVATE
1299 KEY" because this clashed with PKCS#8 unencrypted string. Since this
1300 value was just used as a "magic string" and not used directly its
1301 value doesn't matter.
1304 *) Introduce some semblance of const correctness to BN. Shame C doesn't
1308 *) "linux-sparc64" configuration (ultrapenguin).
1309 [Ray Miller <ray.miller@oucs.ox.ac.uk>]
1310 "linux-sparc" configuration.
1311 [Christian Forster <fo@hawo.stw.uni-erlangen.de>]
1313 *) config now generates no-xxx options for missing ciphers.
1316 *) Support the EBCDIC character set (work in progress).
1317 File ebcdic.c not yet included because it has a different license.
1318 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>]
1320 *) Support BS2000/OSD-POSIX.
1321 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>]
1323 *) Make callbacks for key generation use void * instead of char *.
1326 *) Make S/MIME samples compile (not yet tested).
1329 *) Additional typesafe stacks.
1332 *) New configuration variants "bsdi-elf-gcc" (BSD/OS 4.x).
1336 Changes between 0.9.3 and 0.9.3a [29 May 1999]
1338 *) New configuration variant "sco5-gcc".
1340 *) Updated some demos.
1341 [Sean O Riordain, Wade Scholine]
1343 *) Add missing BIO_free at exit of pkcs12 application.
1346 *) Fix memory leak in conf.c.
1349 *) Updates for Win32 to assembler version of MD5.
1352 *) Set #! path to perl in apps/der_chop to where we found it
1353 instead of using a fixed path.
1356 *) SHA library changes for irix64-mips4-cc.
1359 *) Improvements for VMS support.
1363 Changes between 0.9.2b and 0.9.3 [24 May 1999]
1365 *) Bignum library bug fix. IRIX 6 passes "make test" now!
1366 This also avoids the problems with SC4.2 and unpatched SC5.
1367 [Andy Polyakov <appro@fy.chalmers.se>]
1369 *) New functions sk_num, sk_value and sk_set to replace the previous macros.
1370 These are required because of the typesafe stack would otherwise break
1371 existing code. If old code used a structure member which used to be STACK
1372 and is now STACK_OF (for example cert in a PKCS7_SIGNED structure) with
1373 sk_num or sk_value it would produce an error because the num, data members
1374 are not present in STACK_OF. Now it just produces a warning. sk_set
1375 replaces the old method of assigning a value to sk_value
1376 (e.g. sk_value(x, i) = y) which the library used in a few cases. Any code
1377 that does this will no longer work (and should use sk_set instead) but
1378 this could be regarded as a "questionable" behaviour anyway.
1381 *) Fix most of the other PKCS#7 bugs. The "experimental" code can now
1382 correctly handle encrypted S/MIME data.
1385 *) Change type of various DES function arguments from des_cblock
1386 (which means, in function argument declarations, pointer to char)
1387 to des_cblock * (meaning pointer to array with 8 char elements),
1388 which allows the compiler to do more typechecking; it was like
1389 that back in SSLeay, but with lots of ugly casts.
1391 Introduce new type const_des_cblock.
1394 *) Reorganise the PKCS#7 library and get rid of some of the more obvious
1395 problems: find RecipientInfo structure that matches recipient certificate
1396 and initialise the ASN1 structures properly based on passed cipher.
1399 *) Belatedly make the BN tests actually check the results.
1402 *) Fix the encoding and decoding of negative ASN1 INTEGERS and conversion
1403 to and from BNs: it was completely broken. New compilation option
1404 NEG_PUBKEY_BUG to allow for some broken certificates that encode public
1405 key elements as negative integers.
1408 *) Reorganize and speed up MD5.
1409 [Andy Polyakov <appro@fy.chalmers.se>]
1412 [Richard Levitte <richard@levitte.org>]
1414 *) New option -out to asn1parse to allow the parsed structure to be
1415 output to a file. This is most useful when combined with the -strparse
1416 option to examine the output of things like OCTET STRINGS.
1419 *) Make SSL library a little more fool-proof by not requiring any longer
1420 that SSL_set_{accept,connect}_state be called before
1421 SSL_{accept,connect} may be used (SSL_set_..._state is omitted
1422 in many applications because usually everything *appeared* to work as
1423 intended anyway -- now it really works as intended).
1426 *) Move openssl.cnf out of lib/.
1429 *) Fix various things to let OpenSSL even pass ``egcc -pipe -O2 -Wall
1430 -Wshadow -Wpointer-arith -Wcast-align -Wmissing-prototypes
1431 -Wmissing-declarations -Wnested-externs -Winline'' with EGCS 1.1.2+
1432 [Ralf S. Engelschall]
1434 *) Various fixes to the EVP and PKCS#7 code. It may now be able to
1435 handle PKCS#7 enveloped data properly.
1436 [Sebastian Akerman <sak@parallelconsulting.com>, modified by Steve]
1438 *) Create a duplicate of the SSL_CTX's CERT in SSL_new instead of
1439 copying pointers. The cert_st handling is changed by this in
1440 various ways (and thus what used to be known as ctx->default_cert
1441 is now called ctx->cert, since we don't resort to s->ctx->[default_]cert
1442 any longer when s->cert does not give us what we need).
1443 ssl_cert_instantiate becomes obsolete by this change.
1444 As soon as we've got the new code right (possibly it already is?),
1445 we have solved a couple of bugs of the earlier code where s->cert
1446 was used as if it could not have been shared with other SSL structures.
1448 Note that using the SSL API in certain dirty ways now will result
1449 in different behaviour than observed with earlier library versions:
1450 Changing settings for an SSL_CTX *ctx after having done s = SSL_new(ctx)
1451 does not influence s as it used to.
1453 In order to clean up things more thoroughly, inside SSL_SESSION
1454 we don't use CERT any longer, but a new structure SESS_CERT
1455 that holds per-session data (if available); currently, this is
1456 the peer's certificate chain and, for clients, the server's certificate
1457 and temporary key. CERT holds only those values that can have
1458 meaningful defaults in an SSL_CTX.
1461 *) New function X509V3_EXT_i2d() to create an X509_EXTENSION structure
1462 from the internal representation. Various PKCS#7 fixes: remove some
1463 evil casts and set the enc_dig_alg field properly based on the signing
1467 *) Allow PKCS#12 password to be set from the command line or the
1468 environment. Let 'ca' get its config file name from the environment
1469 variables "OPENSSL_CONF" or "SSLEAY_CONF" (for consistency with 'req'
1473 *) Allow certificate policies extension to use an IA5STRING for the
1474 organization field. This is contrary to the PKIX definition but
1475 VeriSign uses it and IE5 only recognises this form. Document 'x509'
1479 *) Add PEDANTIC compiler flag to allow compilation with gcc -pedantic,
1480 without disallowing inline assembler and the like for non-pedantic builds.
1483 *) Support Borland C++ builder.
1484 [Janez Jere <jj@void.si>, modified by Ulf Möller]
1489 *) SHA-1 cleanups and performance enhancements.
1490 [Andy Polyakov <appro@fy.chalmers.se>]
1492 *) Sparc v8plus assembler for the bignum library.
1493 [Andy Polyakov <appro@fy.chalmers.se>]
1495 *) Accept any -xxx and +xxx compiler options in Configure.
1498 *) Update HPUX configuration.
1501 *) Add missing sk_<type>_unshift() function to safestack.h
1502 [Ralf S. Engelschall]
1504 *) New function SSL_CTX_use_certificate_chain_file that sets the
1505 "extra_cert"s in addition to the certificate. (This makes sense
1506 only for "PEM" format files, as chains as a whole are not
1510 *) Support verify_depth from the SSL API.
1511 x509_vfy.c had what can be considered an off-by-one-error:
1512 Its depth (which was not part of the external interface)
1513 was actually counting the number of certificates in a chain;
1514 now it really counts the depth.
1517 *) Bugfix in crypto/x509/x509_cmp.c: The SSLerr macro was used
1518 instead of X509err, which often resulted in confusing error
1519 messages since the error codes are not globally unique
1520 (e.g. an alleged error in ssl3_accept when a certificate
1521 didn't match the private key).
1523 *) New function SSL_CTX_set_session_id_context that allows to set a default
1524 value (so that you don't need SSL_set_session_id_context for each
1525 connection using the SSL_CTX).
1528 *) OAEP decoding bug fix.
1531 *) Support INSTALL_PREFIX for package builders, as proposed by
1535 *) New Configure options "threads" and "no-threads". For systems
1536 where the proper compiler options are known (currently Solaris
1537 and Linux), "threads" is the default.
1540 *) New script util/mklink.pl as a faster substitute for util/mklink.sh.
1543 *) Install various scripts to $(OPENSSLDIR)/misc, not to
1544 $(INSTALLTOP)/bin -- they shouldn't clutter directories
1545 such as /usr/local/bin.
1548 *) "make linux-shared" to build shared libraries.
1549 [Niels Poppe <niels@netbox.org>]
1551 *) New Configure option no-<cipher> (rsa, idea, rc5, ...).
1554 *) Add the PKCS#12 API documentation to openssl.txt. Preliminary support for
1555 extension adding in x509 utility.
1558 *) Remove NOPROTO sections and error code comments.
1561 *) Partial rewrite of the DEF file generator to now parse the ANSI
1565 *) New Configure options --prefix=DIR and --openssldir=DIR.
1568 *) Complete rewrite of the error code script(s). It is all now handled
1569 by one script at the top level which handles error code gathering,
1570 header rewriting and C source file generation. It should be much better
1571 than the old method: it now uses a modified version of Ulf's parser to
1572 read the ANSI prototypes in all header files (thus the old K&R definitions
1573 aren't needed for error creation any more) and do a better job of
1574 translating function codes into names. The old 'ASN1 error code imbedded
1575 in a comment' is no longer necessary and it doesn't use .err files which
1576 have now been deleted. Also the error code call doesn't have to appear all
1577 on one line (which resulted in some large lines...).
1580 *) Change #include filenames from <foo.h> to <openssl/foo.h>.
1583 *) Change behaviour of ssl2_read when facing length-0 packets: Don't return
1584 0 (which usually indicates a closed connection), but continue reading.
1587 *) Fix some race conditions.
1590 *) Add support for CRL distribution points extension. Add Certificate
1591 Policies and CRL distribution points documentation.
1594 *) Move the autogenerated header file parts to crypto/opensslconf.h.
1597 *) Fix new 56-bit DES export ciphersuites: they were using 7 bytes instead of
1598 8 of keying material. Merlin has also confirmed interop with this fix
1599 between OpenSSL and Baltimore C/SSL 2.0 and J/SSL 2.0.
1600 [Merlin Hughes <merlin@baltimore.ie>]
1602 *) Fix lots of warnings.
1603 [Richard Levitte <levitte@stacken.kth.se>]
1605 *) In add_cert_dir() in crypto/x509/by_dir.c, break out of the loop if
1606 the directory spec didn't end with a LIST_SEPARATOR_CHAR.
1607 [Richard Levitte <levitte@stacken.kth.se>]
1609 *) Fix problems with sizeof(long) == 8.
1610 [Andy Polyakov <appro@fy.chalmers.se>]
1612 *) Change functions to ANSI C.
1615 *) Fix typos in error codes.
1616 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>, Ulf Möller]
1618 *) Remove defunct assembler files from Configure.
1621 *) SPARC v8 assembler BIGNUM implementation.
1622 [Andy Polyakov <appro@fy.chalmers.se>]
1624 *) Support for Certificate Policies extension: both print and set.
1625 Various additions to support the r2i method this uses.
1628 *) A lot of constification, and fix a bug in X509_NAME_oneline() that could
1629 return a const string when you are expecting an allocated buffer.
1632 *) Add support for ASN1 types UTF8String and VISIBLESTRING, also the CHOICE
1633 types DirectoryString and DisplayText.
1636 *) Add code to allow r2i extensions to access the configuration database,
1637 add an LHASH database driver and add several ctx helper functions.
1640 *) Fix an evil bug in bn_expand2() which caused various BN functions to
1641 fail when they extended the size of a BIGNUM.
1644 *) Various utility functions to handle SXNet extension. Modify mkdef.pl to
1645 support typesafe stack.
1648 *) Fix typo in SSL_[gs]et_options().
1649 [Nils Frostberg <nils@medcom.se>]
1651 *) Delete various functions and files that belonged to the (now obsolete)
1652 old X509V3 handling code.
1655 *) New Configure option "rsaref".
1658 *) Don't auto-generate pem.h.
1661 *) Introduce type-safe ASN.1 SETs.
1664 *) Convert various additional casted stacks to type-safe STACK_OF() variants.
1665 [Ben Laurie, Ralf S. Engelschall, Steve Henson]
1667 *) Introduce type-safe STACKs. This will almost certainly break lots of code
1668 that links with OpenSSL (well at least cause lots of warnings), but fear
1669 not: the conversion is trivial, and it eliminates loads of evil casts. A
1670 few STACKed things have been converted already. Feel free to convert more.
1671 In the fullness of time, I'll do away with the STACK type altogether.
1674 *) Add `openssl ca -revoke <certfile>' facility which revokes a certificate
1675 specified in <certfile> by updating the entry in the index.txt file.
1676 This way one no longer has to edit the index.txt file manually for
1677 revoking a certificate. The -revoke option does the gory details now.
1678 [Massimiliano Pala <madwolf@openca.org>, Ralf S. Engelschall]
1680 *) Fix `openssl crl -noout -text' combination where `-noout' killed the
1681 `-text' option at all and this way the `-noout -text' combination was
1682 inconsistent in `openssl crl' with the friends in `openssl x509|rsa|dsa'.
1683 [Ralf S. Engelschall]
1685 *) Make sure a corresponding plain text error message exists for the
1686 X509_V_ERR_CERT_REVOKED/23 error number which can occur when a
1687 verify callback function determined that a certificate was revoked.
1688 [Ralf S. Engelschall]
1690 *) Bugfix: In test/testenc, don't test "openssl <cipher>" for
1691 ciphers that were excluded, e.g. by -DNO_IDEA. Also, test
1692 all available cipers including rc5, which was forgotten until now.
1693 In order to let the testing shell script know which algorithms
1694 are available, a new (up to now undocumented) command
1695 "openssl list-cipher-commands" is used.
1698 *) Bugfix: s_client occasionally would sleep in select() when
1699 it should have checked SSL_pending() first.
1702 *) New functions DSA_do_sign and DSA_do_verify to provide access to
1703 the raw DSA values prior to ASN.1 encoding.
1706 *) Tweaks to Configure
1707 [Niels Poppe <niels@netbox.org>]
1709 *) Add support for PKCS#5 v2.0 ASN1 PBES2 structures. No other support,
1713 *) New variables $(RANLIB) and $(PERL) in the Makefiles.
1716 *) New config option to avoid instructions that are illegal on the 80386.
1717 The default code is faster, but requires at least a 486.
1720 *) Got rid of old SSL2_CLIENT_VERSION (inconsistently used) and
1721 SSL2_SERVER_VERSION (not used at all) macros, which are now the
1722 same as SSL2_VERSION anyway.
1725 *) New "-showcerts" option for s_client.
1728 *) Still more PKCS#12 integration. Add pkcs12 application to openssl
1729 application. Various cleanups and fixes.
1732 *) More PKCS#12 integration. Add new pkcs12 directory with Makefile.ssl and
1733 modify error routines to work internally. Add error codes and PBE init
1734 to library startup routines.
1737 *) Further PKCS#12 integration. Added password based encryption, PKCS#8 and
1738 packing functions to asn1 and evp. Changed function names and error
1739 codes along the way.
1742 *) PKCS12 integration: and so it begins... First of several patches to
1743 slowly integrate PKCS#12 functionality into OpenSSL. Add PKCS#12
1744 objects to objects.h
1747 *) Add a new 'indent' option to some X509V3 extension code. Initial ASN1
1748 and display support for Thawte strong extranet extension.
1751 *) Add LinuxPPC support.
1752 [Jeff Dubrule <igor@pobox.org>]
1754 *) Get rid of redundant BN file bn_mulw.c, and rename bn_div64 to
1755 bn_div_words in alpha.s.
1756 [Hannes Reinecke <H.Reinecke@hw.ac.uk> and Ben Laurie]
1758 *) Make sure the RSA OAEP test is skipped under -DRSAref because
1759 OAEP isn't supported when OpenSSL is built with RSAref.
1760 [Ulf Moeller <ulf@fitug.de>]
1762 *) Move definitions of IS_SET/IS_SEQUENCE inside crypto/asn1/asn1.h
1763 so they no longer are missing under -DNOPROTO.
1764 [Soren S. Jorvang <soren@t.dk>]
1767 Changes between 0.9.1c and 0.9.2b [22 Mar 1999]
1769 *) Make SSL_get_peer_cert_chain() work in servers. Unfortunately, it still
1770 doesn't work when the session is reused. Coming soon!
1773 *) Fix a security hole, that allows sessions to be reused in the wrong
1774 context thus bypassing client cert protection! All software that uses
1775 client certs and session caches in multiple contexts NEEDS PATCHING to
1776 allow session reuse! A fuller solution is in the works.
1777 [Ben Laurie, problem pointed out by Holger Reif, Bodo Moeller (and ???)]
1779 *) Some more source tree cleanups (removed obsolete files
1780 crypto/bf/asm/bf586.pl, test/test.txt and crypto/sha/asm/f.s; changed
1781 permission on "config" script to be executable) and a fix for the INSTALL
1783 [Ulf Moeller <ulf@fitug.de>]
1785 *) Remove some legacy and erroneous uses of malloc, free instead of
1787 [Lennart Bang <lob@netstream.se>, with minor changes by Steve]
1789 *) Make rsa_oaep_test return non-zero on error.
1790 [Ulf Moeller <ulf@fitug.de>]
1792 *) Add support for native Solaris shared libraries. Configure
1793 solaris-sparc-sc4-pic, make, then run shlib/solaris-sc4.sh. It'd be nice
1794 if someone would make that last step automatic.
1795 [Matthias Loepfe <Matthias.Loepfe@AdNovum.CH>]
1797 *) ctx_size was not built with the right compiler during "make links". Fixed.
1800 *) Change the meaning of 'ALL' in the cipher list. It now means "everything
1801 except NULL ciphers". This means the default cipher list will no longer
1802 enable NULL ciphers. They need to be specifically enabled e.g. with
1803 the string "DEFAULT:eNULL".
1806 *) Fix to RSA private encryption routines: if p < q then it would
1807 occasionally produce an invalid result. This will only happen with
1808 externally generated keys because OpenSSL (and SSLeay) ensure p > q.
1811 *) Be less restrictive and allow also `perl util/perlpath.pl
1812 /path/to/bin/perl' in addition to `perl util/perlpath.pl /path/to/bin',
1813 because this way one can also use an interpreter named `perl5' (which is
1814 usually the name of Perl 5.xxx on platforms where an Perl 4.x is still
1815 installed as `perl').
1816 [Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
1818 *) Let util/clean-depend.pl work also with older Perl 5.00x versions.
1819 [Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
1821 *) Fix Makefile.org so CC,CFLAG etc are passed to 'make links' add
1822 advapi32.lib to Win32 build and change the pem test comparision
1823 to fc.exe (thanks to Ulrich Kroener <kroneru@yahoo.com> for the
1824 suggestion). Fix misplaced ASNI prototypes and declarations in evp.h
1825 and crypto/des/ede_cbcm_enc.c.
1828 *) DES quad checksum was broken on big-endian architectures. Fixed.
1831 *) Comment out two functions in bio.h that aren't implemented. Fix up the
1832 Win32 test batch file so it (might) work again. The Win32 test batch file
1833 is horrible: I feel ill....
1836 *) Move various #ifdefs around so NO_SYSLOG, NO_DIRENT etc are now selected
1837 in e_os.h. Audit of header files to check ANSI and non ANSI
1838 sections: 10 functions were absent from non ANSI section and not exported
1839 from Windows DLLs. Fixed up libeay.num for new functions.
1842 *) Make `openssl version' output lines consistent.
1843 [Ralf S. Engelschall]
1845 *) Fix Win32 symbol export lists for BIO functions: Added
1846 BIO_get_ex_new_index, BIO_get_ex_num, BIO_get_ex_data and BIO_set_ex_data
1847 to ms/libeay{16,32}.def.
1848 [Ralf S. Engelschall]
1850 *) Second round of fixing the OpenSSL perl/ stuff. It now at least compiled
1851 fine under Unix and passes some trivial tests I've now added. But the
1852 whole stuff is horribly incomplete, so a README.1ST with a disclaimer was
1853 added to make sure no one expects that this stuff really works in the
1854 OpenSSL 0.9.2 release. Additionally I've started to clean the XS sources
1855 up and fixed a few little bugs and inconsistencies in OpenSSL.{pm,xs} and
1857 [Ralf S. Engelschall]
1859 *) Fix the generation of two part addresses in perl.
1860 [Kenji Miyake <kenji@miyake.org>, integrated by Ben Laurie]
1862 *) Add config entry for Linux on MIPS.
1863 [John Tobey <jtobey@channel1.com>]
1865 *) Make links whenever Configure is run, unless we are on Windoze.
1868 *) Permit extensions to be added to CRLs using crl_section in openssl.cnf.
1869 Currently only issuerAltName and AuthorityKeyIdentifier make any sense
1873 *) Add a useful kludge to allow package maintainers to specify compiler and
1874 other platforms details on the command line without having to patch the
1875 Configure script everytime: One now can use ``perl Configure
1876 <id>:<details>'', i.e. platform ids are allowed to have details appended
1877 to them (seperated by colons). This is treated as there would be a static
1878 pre-configured entry in Configure's %table under key <id> with value
1879 <details> and ``perl Configure <id>'' is called. So, when you want to
1880 perform a quick test-compile under FreeBSD 3.1 with pgcc and without
1881 assembler stuff you can use ``perl Configure "FreeBSD-elf:pgcc:-O6:::"''
1882 now, which overrides the FreeBSD-elf entry on-the-fly.
1883 [Ralf S. Engelschall]
1885 *) Disable new TLS1 ciphersuites by default: they aren't official yet.
1888 *) Allow DSO flags like -fpic, -fPIC, -KPIC etc. to be specified
1889 on the `perl Configure ...' command line. This way one can compile
1890 OpenSSL libraries with Position Independent Code (PIC) which is needed
1891 for linking it into DSOs.
1892 [Ralf S. Engelschall]
1894 *) Remarkably, export ciphers were totally broken and no-one had noticed!
1898 *) Cleaned up the LICENSE document: The official contact for any license
1899 questions now is the OpenSSL core team under openssl-core@openssl.org.
1900 And add a paragraph about the dual-license situation to make sure people
1901 recognize that _BOTH_ the OpenSSL license _AND_ the SSLeay license apply
1902 to the OpenSSL toolkit.
1903 [Ralf S. Engelschall]
1905 *) General source tree makefile cleanups: Made `making xxx in yyy...'
1906 display consistent in the source tree and replaced `/bin/rm' by `rm'.
1907 Additonally cleaned up the `make links' target: Remove unnecessary
1908 semicolons, subsequent redundant removes, inline point.sh into mklink.sh
1909 to speed processing and no longer clutter the display with confusing
1910 stuff. Instead only the actually done links are displayed.
1911 [Ralf S. Engelschall]
1913 *) Permit null encryption ciphersuites, used for authentication only. It used
1914 to be necessary to set the preprocessor define SSL_ALLOW_ENULL to do this.
1915 It is now necessary to set SSL_FORBID_ENULL to prevent the use of null
1919 *) Add a bunch of fixes to the PKCS#7 stuff. It used to sometimes reorder
1920 signed attributes when verifying signatures (this would break them),
1921 the detached data encoding was wrong and public keys obtained using
1922 X509_get_pubkey() weren't freed.
1925 *) Add text documentation for the BUFFER functions. Also added a work around
1926 to a Win95 console bug. This was triggered by the password read stuff: the
1927 last character typed gets carried over to the next fread(). If you were
1928 generating a new cert request using 'req' for example then the last
1929 character of the passphrase would be CR which would then enter the first
1933 *) Added the new `Includes OpenSSL Cryptography Software' button as
1934 doc/openssl_button.{gif,html} which is similar in style to the old SSLeay
1935 button and can be used by applications based on OpenSSL to show the
1936 relationship to the OpenSSL project.
1937 [Ralf S. Engelschall]
1939 *) Remove confusing variables in function signatures in files
1940 ssl/ssl_lib.c and ssl/ssl.h.
1941 [Lennart Bong <lob@kulthea.stacken.kth.se>]
1943 *) Don't install bss_file.c under PREFIX/include/
1944 [Lennart Bong <lob@kulthea.stacken.kth.se>]
1946 *) Get the Win32 compile working again. Modify mkdef.pl so it can handle
1947 functions that return function pointers and has support for NT specific
1948 stuff. Fix mk1mf.pl and VC-32.pl to support NT differences also. Various
1949 #ifdef WIN32 and WINNTs sprinkled about the place and some changes from
1950 unsigned to signed types: this was killing the Win32 compile.
1953 *) Add new certificate file to stack functions,
1954 SSL_add_dir_cert_subjects_to_stack() and
1955 SSL_add_file_cert_subjects_to_stack(). These largely supplant
1956 SSL_load_client_CA_file(), and can be used to add multiple certs easily
1957 to a stack (usually this is then handed to SSL_CTX_set_client_CA_list()).
1958 This means that Apache-SSL and similar packages don't have to mess around
1959 to add as many CAs as they want to the preferred list.
1962 *) Experiment with doxygen documentation. Currently only partially applied to
1964 See http://www.stack.nl/~dimitri/doxygen/index.html, and run doxygen with
1965 openssl.doxy as the configuration file.
1968 *) Get rid of remaining C++-style comments which strict C compilers hate.
1969 [Ralf S. Engelschall, pointed out by Carlos Amengual]
1971 *) Changed BN_RECURSION in bn_mont.c to BN_RECURSION_MONT so it is not
1972 compiled in by default: it has problems with large keys.
1975 *) Add a bunch of SSL_xxx() functions for configuring the temporary RSA and
1976 DH private keys and/or callback functions which directly correspond to
1977 their SSL_CTX_xxx() counterparts but work on a per-connection basis. This
1978 is needed for applications which have to configure certificates on a
1979 per-connection basis (e.g. Apache+mod_ssl) instead of a per-context basis
1981 For the RSA certificate situation is makes no difference, but
1982 for the DSA certificate situation this fixes the "no shared cipher"
1983 problem where the OpenSSL cipher selection procedure failed because the
1984 temporary keys were not overtaken from the context and the API provided
1985 no way to reconfigure them.
1986 The new functions now let applications reconfigure the stuff and they
1987 are in detail: SSL_need_tmp_RSA, SSL_set_tmp_rsa, SSL_set_tmp_dh,
1988 SSL_set_tmp_rsa_callback and SSL_set_tmp_dh_callback. Additionally a new
1989 non-public-API function ssl_cert_instantiate() is used as a helper
1990 function and also to reduce code redundancy inside ssl_rsa.c.
1991 [Ralf S. Engelschall]
1993 *) Move s_server -dcert and -dkey options out of the undocumented feature
1994 area because they are useful for the DSA situation and should be
1995 recognized by the users.
1996 [Ralf S. Engelschall]
1998 *) Fix the cipher decision scheme for export ciphers: the export bits are
1999 *not* within SSL_MKEY_MASK or SSL_AUTH_MASK, they are within
2000 SSL_EXP_MASK. So, the original variable has to be used instead of the
2001 already masked variable.
2002 [Richard Levitte <levitte@stacken.kth.se>]
2004 *) Fix 'port' variable from `int' to `unsigned int' in crypto/bio/b_sock.c
2005 [Richard Levitte <levitte@stacken.kth.se>]
2007 *) Change type of another md_len variable in pk7_doit.c:PKCS7_dataFinal()
2008 from `int' to `unsigned int' because it's a length and initialized by
2009 EVP_DigestFinal() which expects an `unsigned int *'.
2010 [Richard Levitte <levitte@stacken.kth.se>]
2012 *) Don't hard-code path to Perl interpreter on shebang line of Configure
2013 script. Instead use the usual Shell->Perl transition trick.
2014 [Ralf S. Engelschall]
2016 *) Make `openssl x509 -noout -modulus' functional also for DSA certificates
2017 (in addition to RSA certificates) to match the behaviour of `openssl dsa
2018 -noout -modulus' as it's already the case for `openssl rsa -noout
2019 -modulus'. For RSA the -modulus is the real "modulus" while for DSA
2020 currently the public key is printed (a decision which was already done by
2021 `openssl dsa -modulus' in the past) which serves a similar purpose.
2022 Additionally the NO_RSA no longer completely removes the whole -modulus
2023 option; it now only avoids using the RSA stuff. Same applies to NO_DSA
2025 [Ralf S. Engelschall]
2027 *) Add Arne Ansper's reliable BIO - this is an encrypted, block-digested
2028 BIO. See the source (crypto/evp/bio_ok.c) for more info.
2029 [Arne Ansper <arne@ats.cyber.ee>]
2031 *) Dump the old yucky req code that tried (and failed) to allow raw OIDs
2032 to be added. Now both 'req' and 'ca' can use new objects defined in the
2036 *) Add cool BIO that does syslog (or event log on NT).
2037 [Arne Ansper <arne@ats.cyber.ee>, integrated by Ben Laurie]
2039 *) Add support for new TLS ciphersuites, TLS_RSA_EXPORT56_WITH_RC4_56_MD5,
2040 TLS_RSA_EXPORT56_WITH_RC2_CBC_56_MD5 and
2041 TLS_RSA_EXPORT56_WITH_DES_CBC_SHA, as specified in "56-bit Export Cipher
2042 Suites For TLS", draft-ietf-tls-56-bit-ciphersuites-00.txt.
2045 *) Add preliminary config info for new extension code.
2048 *) Make RSA_NO_PADDING really use no padding.
2049 [Ulf Moeller <ulf@fitug.de>]
2051 *) Generate errors when private/public key check is done.
2054 *) Overhaul for 'crl' utility. New function X509_CRL_print. Partial support
2055 for some CRL extensions and new objects added.
2058 *) Really fix the ASN1 IMPLICIT bug this time... Partial support for private
2059 key usage extension and fuller support for authority key id.
2062 *) Add OAEP encryption for the OpenSSL crypto library. OAEP is the improved
2063 padding method for RSA, which is recommended for new applications in PKCS
2064 #1 v2.0 (RFC 2437, October 1998).
2065 OAEP (Optimal Asymmetric Encryption Padding) has better theoretical
2066 foundations than the ad-hoc padding used in PKCS #1 v1.5. It is secure
2067 against Bleichbacher's attack on RSA.
2068 [Ulf Moeller <ulf@fitug.de>, reformatted, corrected and integrated by
2071 *) Updates to the new SSL compression code
2072 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
2074 *) Fix so that the version number in the master secret, when passed
2075 via RSA, checks that if TLS was proposed, but we roll back to SSLv3
2076 (because the server will not accept higher), that the version number
2077 is 0x03,0x01, not 0x03,0x00
2078 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
2080 *) Run extensive memory leak checks on SSL apps. Fixed *lots* of memory
2081 leaks in ssl/ relating to new X509_get_pubkey() behaviour. Also fixes
2082 in apps/ and an unrelated leak in crypto/dsa/dsa_vrf.c
2085 *) Support for RAW extensions where an arbitrary extension can be
2086 created by including its DER encoding. See apps/openssl.cnf for
2090 *) Make sure latest Perl versions don't interpret some generated C array
2091 code as Perl array code in the crypto/err/err_genc.pl script.
2092 [Lars Weber <3weber@informatik.uni-hamburg.de>]
2094 *) Modify ms/do_ms.bat to not generate assembly language makefiles since
2095 not many people have the assembler. Various Win32 compilation fixes and
2096 update to the INSTALL.W32 file with (hopefully) more accurate Win32
2100 *) Modify configure script 'Configure' to automatically create crypto/date.h
2101 file under Win32 and also build pem.h from pem.org. New script
2102 util/mkfiles.pl to create the MINFO file on environments that can't do a
2103 'make files': perl util/mkfiles.pl >MINFO should work.
2106 *) Major rework of DES function declarations, in the pursuit of correctness
2107 and purity. As a result, many evil casts evaporated, and some weirdness,
2108 too. You may find this causes warnings in your code. Zapping your evil
2109 casts will probably fix them. Mostly.
2112 *) Fix for a typo in asn1.h. Bug fix to object creation script
2113 obj_dat.pl. It considered a zero in an object definition to mean
2114 "end of object": none of the objects in objects.h have any zeros
2115 so it wasn't spotted.
2116 [Steve Henson, reported by Erwann ABALEA <eabalea@certplus.com>]
2118 *) Add support for Triple DES Cipher Block Chaining with Output Feedback
2119 Masking (CBCM). In the absence of test vectors, the best I have been able
2120 to do is check that the decrypt undoes the encrypt, so far. Send me test
2121 vectors if you have them.
2124 *) Correct calculation of key length for export ciphers (too much space was
2125 allocated for null ciphers). This has not been tested!
2128 *) Modifications to the mkdef.pl for Win32 DEF file creation. The usage
2129 message is now correct (it understands "crypto" and "ssl" on its
2130 command line). There is also now an "update" option. This will update
2131 the util/ssleay.num and util/libeay.num files with any new functions.
2133 perl util/mkdef.pl crypto ssl update
2134 it will update them.
2137 *) Overhauled the Perl interface (perl/*):
2138 - ported BN stuff to OpenSSL's different BN library
2139 - made the perl/ source tree CVS-aware
2140 - renamed the package from SSLeay to OpenSSL (the files still contain
2141 their history because I've copied them in the repository)
2142 - removed obsolete files (the test scripts will be replaced
2143 by better Test::Harness variants in the future)
2144 [Ralf S. Engelschall]
2146 *) First cut for a very conservative source tree cleanup:
2147 1. merge various obsolete readme texts into doc/ssleay.txt
2148 where we collect the old documents and readme texts.
2149 2. remove the first part of files where I'm already sure that we no
2150 longer need them because of three reasons: either they are just temporary
2151 files which were left by Eric or they are preserved original files where
2152 I've verified that the diff is also available in the CVS via "cvs diff
2153 -rSSLeay_0_8_1b" or they were renamed (as it was definitely the case for
2154 the crypto/md/ stuff).
2155 [Ralf S. Engelschall]
2157 *) More extension code. Incomplete support for subject and issuer alt
2158 name, issuer and authority key id. Change the i2v function parameters
2159 and add an extra 'crl' parameter in the X509V3_CTX structure: guess
2160 what that's for :-) Fix to ASN1 macro which messed up
2161 IMPLICIT tag and add f_enum.c which adds a2i, i2a for ENUMERATED.
2164 *) Preliminary support for ENUMERATED type. This is largely copied from the
2168 *) Add new function, EVP_MD_CTX_copy() to replace frequent use of memcpy.
2169 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
2171 *) Make sure `make rehash' target really finds the `openssl' program.
2172 [Ralf S. Engelschall, Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
2174 *) Squeeze another 7% of speed out of MD5 assembler, at least on a P2. I'd
2175 like to hear about it if this slows down other processors.
2178 *) Add CygWin32 platform information to Configure script.
2179 [Alan Batie <batie@aahz.jf.intel.com>]
2181 *) Fixed ms/32all.bat script: `no_asm' -> `no-asm'
2182 [Rainer W. Gerling <gerling@mpg-gv.mpg.de>]
2184 *) New program nseq to manipulate netscape certificate sequences
2187 *) Modify crl2pkcs7 so it supports multiple -certfile arguments. Fix a
2191 *) Fixes to BN code. Previously the default was to define BN_RECURSION
2192 but the BN code had some problems that would cause failures when
2193 doing certificate verification and some other functions.
2194 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
2196 *) Add ASN1 and PEM code to support netscape certificate sequences.
2199 *) Add ASN1 and PEM code to support netscape certificate sequences.
2202 *) Add several PKIX and private extended key usage OIDs.
2205 *) Modify the 'ca' program to handle the new extension code. Modify
2206 openssl.cnf for new extension format, add comments.
2209 *) More X509 V3 changes. Fix typo in v3_bitstr.c. Add support to 'req'
2210 and add a sample to openssl.cnf so req -x509 now adds appropriate
2214 *) Continued X509 V3 changes. Add to other makefiles, integrate with the
2215 error code, add initial support to X509_print() and x509 application.
2218 *) Takes a deep breath and start addding X509 V3 extension support code. Add
2219 files in crypto/x509v3. Move original stuff to crypto/x509v3/old. All this
2220 stuff is currently isolated and isn't even compiled yet.
2223 *) Continuing patches for GeneralizedTime. Fix up certificate and CRL
2224 ASN1 to use ASN1_TIME and modify print routines to use ASN1_TIME_print.
2225 Removed the versions check from X509 routines when loading extensions:
2226 this allows certain broken certificates that don't set the version
2227 properly to be processed.
2230 *) Deal with irritating shit to do with dependencies, in YAAHW (Yet Another
2231 Ad Hoc Way) - Makefile.ssls now all contain local dependencies, which
2232 can still be regenerated with "make depend".
2235 *) Spelling mistake in C version of CAST-128.
2236 [Ben Laurie, reported by Jeremy Hylton <jeremy@cnri.reston.va.us>]
2238 *) Changes to the error generation code. The perl script err-code.pl
2239 now reads in the old error codes and retains the old numbers, only
2240 adding new ones if necessary. It also only changes the .err files if new
2241 codes are added. The makefiles have been modified to only insert errors
2242 when needed (to avoid needlessly modifying header files). This is done
2243 by only inserting errors if the .err file is newer than the auto generated
2244 C file. To rebuild all the error codes from scratch (the old behaviour)
2245 either modify crypto/Makefile.ssl to pass the -regen flag to err_code.pl
2246 or delete all the .err files.
2249 *) CAST-128 was incorrectly implemented for short keys. The C version has
2250 been fixed, but is untested. The assembler versions are also fixed, but
2251 new assembler HAS NOT BEEN GENERATED FOR WIN32 - the Makefile needs fixing
2252 to regenerate it if needed.
2253 [Ben Laurie, reported (with fix for C version) by Jun-ichiro itojun
2254 Hagino <itojun@kame.net>]
2256 *) File was opened incorrectly in randfile.c.
2257 [Ulf Möller <ulf@fitug.de>]
2259 *) Beginning of support for GeneralizedTime. d2i, i2d, check and print
2260 functions. Also ASN1_TIME suite which is a CHOICE of UTCTime or
2261 GeneralizedTime. ASN1_TIME is the proper type used in certificates et
2262 al: it's just almost always a UTCTime. Note this patch adds new error
2263 codes so do a "make errors" if there are problems.
2266 *) Correct Linux 1 recognition in config.
2267 [Ulf Möller <ulf@fitug.de>]
2269 *) Remove pointless MD5 hash when using DSA keys in ca.
2270 [Anonymous <nobody@replay.com>]
2272 *) Generate an error if given an empty string as a cert directory. Also
2273 generate an error if handed NULL (previously returned 0 to indicate an
2274 error, but didn't set one).
2275 [Ben Laurie, reported by Anonymous <nobody@replay.com>]
2277 *) Add prototypes to SSL methods. Make SSL_write's buffer const, at last.
2280 *) Fix the dummy function BN_ref_mod_exp() in rsaref.c to have the correct
2281 parameters. This was causing a warning which killed off the Win32 compile.
2284 *) Remove C++ style comments from crypto/bn/bn_local.h.
2285 [Neil Costigan <neil.costigan@celocom.com>]
2287 *) The function OBJ_txt2nid was broken. It was supposed to return a nid
2288 based on a text string, looking up short and long names and finally
2289 "dot" format. The "dot" format stuff didn't work. Added new function
2290 OBJ_txt2obj to do the same but return an ASN1_OBJECT and rewrote
2291 OBJ_txt2nid to use it. OBJ_txt2obj can also return objects even if the
2292 OID is not part of the table.
2295 *) Add prototypes to X509 lookup/verify methods, fixing a bug in
2296 X509_LOOKUP_by_alias().
2299 *) Sort openssl functions by name.
2302 *) Get the gendsa program working (hopefully) and add it to app list. Remove
2303 encryption from sample DSA keys (in case anyone is interested the password
2307 *) Make _all_ *_free functions accept a NULL pointer.
2308 [Frans Heymans <fheymans@isaserver.be>]
2310 *) If a DH key is generated in s3_srvr.c, don't blow it by trying to use
2312 [Anonymous <nobody@replay.com>]
2314 *) s_server should send the CAfile as acceptable CAs, not its own cert.
2315 [Bodo Moeller <3moeller@informatik.uni-hamburg.de>]
2317 *) Don't blow it for numeric -newkey arguments to apps/req.
2318 [Bodo Moeller <3moeller@informatik.uni-hamburg.de>]
2320 *) Temp key "for export" tests were wrong in s3_srvr.c.
2321 [Anonymous <nobody@replay.com>]
2323 *) Add prototype for temp key callback functions
2324 SSL_CTX_set_tmp_{rsa,dh}_callback().
2327 *) Make DH_free() tolerate being passed a NULL pointer (like RSA_free() and
2328 DSA_free()). Make X509_PUBKEY_set() check for errors in d2i_PublicKey().
2331 *) X509_name_add_entry() freed the wrong thing after an error.
2332 [Arne Ansper <arne@ats.cyber.ee>]
2334 *) rsa_eay.c would attempt to free a NULL context.
2335 [Arne Ansper <arne@ats.cyber.ee>]
2337 *) BIO_s_socket() had a broken should_retry() on Windoze.
2338 [Arne Ansper <arne@ats.cyber.ee>]
2340 *) BIO_f_buffer() didn't pass on BIO_CTRL_FLUSH.
2341 [Arne Ansper <arne@ats.cyber.ee>]
2343 *) Make sure the already existing X509_STORE->depth variable is initialized
2344 in X509_STORE_new(), but document the fact that this variable is still
2345 unused in the certificate verification process.
2346 [Ralf S. Engelschall]
2348 *) Fix the various library and apps files to free up pkeys obtained from
2349 X509_PUBKEY_get() et al. Also allow x509.c to handle netscape extensions.
2352 *) Fix reference counting in X509_PUBKEY_get(). This makes
2353 demos/maurice/example2.c work, amongst others, probably.
2354 [Steve Henson and Ben Laurie]
2356 *) First cut of a cleanup for apps/. First the `ssleay' program is now named
2357 `openssl' and second, the shortcut symlinks for the `openssl <command>'
2358 are no longer created. This way we have a single and consistent command
2359 line interface `openssl <command>', similar to `cvs <command>'.
2360 [Ralf S. Engelschall, Paul Sutton and Ben Laurie]
2362 *) ca.c: move test for DSA keys inside #ifndef NO_DSA. Make pubkey
2363 BIT STRING wrapper always have zero unused bits.
2366 *) Add CA.pl, perl version of CA.sh, add extended key usage OID.
2369 *) Make the top-level INSTALL documentation easier to understand.
2372 *) Makefiles updated to exit if an error occurs in a sub-directory
2373 make (including if user presses ^C) [Paul Sutton]
2375 *) Make Montgomery context stuff explicit in RSA data structure.
2378 *) Fix build order of pem and err to allow for generated pem.h.
2381 *) Fix renumbering bug in X509_NAME_delete_entry().
2384 *) Enhanced the err-ins.pl script so it makes the error library number
2385 global and can add a library name. This is needed for external ASN1 and
2386 other error libraries.
2389 *) Fixed sk_insert which never worked properly.
2392 *) Fix ASN1 macros so they can handle indefinite length construted
2393 EXPLICIT tags. Some non standard certificates use these: they can now
2397 *) Merged the various old/obsolete SSLeay documentation files (doc/xxx.doc)
2398 into a single doc/ssleay.txt bundle. This way the information is still
2399 preserved but no longer messes up this directory. Now it's new room for
2400 the new set of documenation files.
2401 [Ralf S. Engelschall]
2403 *) SETs were incorrectly DER encoded. This was a major pain, because they
2404 shared code with SEQUENCEs, which aren't coded the same. This means that
2405 almost everything to do with SETs or SEQUENCEs has either changed name or
2406 number of arguments.
2407 [Ben Laurie, based on a partial fix by GP Jayan <gp@nsj.co.jp>]
2409 *) Fix test data to work with the above.
2412 *) Fix the RSA header declarations that hid a bug I fixed in 0.9.0b but
2413 was already fixed by Eric for 0.9.1 it seems.
2414 [Ben Laurie - pointed out by Ulf Möller <ulf@fitug.de>]
2416 *) Autodetect FreeBSD3.
2419 *) Fix various bugs in Configure. This affects the following platforms:
2423 unixware-2.0-pentium
2427 *) Eliminate generated files from CVS. Reorder tests to regenerate files
2428 before they are needed.
2431 *) Generate Makefile.ssl from Makefile.org (to keep CVS happy).
2435 Changes between 0.9.1b and 0.9.1c [23-Dec-1998]
2437 *) Added OPENSSL_VERSION_NUMBER to crypto/crypto.h and
2438 changed SSLeay to OpenSSL in version strings.
2439 [Ralf S. Engelschall]
2441 *) Some fixups to the top-level documents.
2444 *) Fixed the nasty bug where rsaref.h was not found under compile-time
2445 because the symlink to include/ was missing.
2446 [Ralf S. Engelschall]
2448 *) Incorporated the popular no-RSA/DSA-only patches
2449 which allow to compile a RSA-free SSLeay.
2450 [Andrew Cooke / Interrader Ldt., Ralf S. Engelschall]
2452 *) Fixed nasty rehash problem under `make -f Makefile.ssl links'
2453 when "ssleay" is still not found.
2454 [Ralf S. Engelschall]
2456 *) Added more platforms to Configure: Cray T3E, HPUX 11,
2457 [Ralf S. Engelschall, Beckmann <beckman@acl.lanl.gov>]
2459 *) Updated the README file.
2460 [Ralf S. Engelschall]
2462 *) Added various .cvsignore files in the CVS repository subdirs
2463 to make a "cvs update" really silent.
2464 [Ralf S. Engelschall]
2466 *) Recompiled the error-definition header files and added
2467 missing symbols to the Win32 linker tables.
2468 [Ralf S. Engelschall]
2470 *) Cleaned up the top-level documents;
2471 o new files: CHANGES and LICENSE
2472 o merged VERSION, HISTORY* and README* files a CHANGES.SSLeay
2473 o merged COPYRIGHT into LICENSE
2474 o removed obsolete TODO file
2475 o renamed MICROSOFT to INSTALL.W32
2476 [Ralf S. Engelschall]
2478 *) Removed dummy files from the 0.9.1b source tree:
2479 crypto/asn1/x crypto/bio/cd crypto/bio/fg crypto/bio/grep crypto/bio/vi
2480 crypto/bn/asm/......add.c crypto/bn/asm/a.out crypto/dsa/f crypto/md5/f
2481 crypto/pem/gmon.out crypto/perlasm/f crypto/pkcs7/build crypto/rsa/f
2482 crypto/sha/asm/f crypto/threads/f ms/zzz ssl/f ssl/f.mak test/f
2483 util/f.mak util/pl/f util/pl/f.mak crypto/bf/bf_locl.old apps/f
2484 [Ralf S. Engelschall]
2486 *) Added various platform portability fixes.
2489 *) The Genesis of the OpenSSL rpject:
2490 We start with the latest (unreleased) SSLeay version 0.9.1b which Eric A.
2491 Young and Tim J. Hudson created while they were working for C2Net until
2493 [The OpenSSL Project]
2496 Changes between 0.9.0b and 0.9.1b [not released]
2498 *) Updated a few CA certificates under certs/
2501 *) Changed some BIGNUM api stuff.
2504 *) Various platform ports: OpenBSD, Ultrix, IRIX 64bit, NetBSD,
2505 DGUX x86, Linux Alpha, etc.
2508 *) New COMP library [crypto/comp/] for SSL Record Layer Compression:
2509 RLE (dummy implemented) and ZLIB (really implemented when ZLIB is
2513 *) Add -strparse option to asn1pars program which parses nested
2515 [Dr Stephen Henson <shenson@bigfoot.com>]
2517 *) Added "oid_file" to ssleay.cnf for "ca" and "req" programs.
2520 *) DSA fix for "ca" program.
2523 *) Added "-genkey" option to "dsaparam" program.
2526 *) Added RIPE MD160 (rmd160) message digest.
2529 *) Added -a (all) option to "ssleay version" command.
2532 *) Added PLATFORM define which is the id given to Configure.
2535 *) Added MemCheck_XXXX functions to crypto/mem.c for memory checking.
2538 *) Extended the ASN.1 parser routines.
2541 *) Extended BIO routines to support REUSEADDR, seek, tell, etc.
2544 *) Added a BN_CTX to the BN library.
2547 *) Fixed the weak key values in DES library
2550 *) Changed API in EVP library for cipher aliases.
2553 *) Added support for RC2/64bit cipher.
2556 *) Converted the lhash library to the crypto/mem.c functions.
2559 *) Added more recognized ASN.1 object ids.
2562 *) Added more RSA padding checks for SSL/TLS.
2565 *) Added BIO proxy/filter functionality.
2568 *) Added extra_certs to SSL_CTX which can be used
2569 send extra CA certificates to the client in the CA cert chain sending
2570 process. It can be configured with SSL_CTX_add_extra_chain_cert().
2573 *) Now Fortezza is denied in the authentication phase because
2574 this is key exchange mechanism is not supported by SSLeay at all.
2577 *) Additional PKCS1 checks.
2580 *) Support the string "TLSv1" for all TLS v1 ciphers.
2583 *) Added function SSL_get_ex_data_X509_STORE_CTX_idx() which gives the
2584 ex_data index of the SSL context in the X509_STORE_CTX ex_data.
2587 *) Fixed a few memory leaks.
2590 *) Fixed various code and comment typos.
2593 *) A minor bug in ssl/s3_clnt.c where there would always be 4 0
2594 bytes sent in the client random.
2595 [Edward Bishop <ebishop@spyglass.com>]