5 Changes between 0.9.4 and 0.9.5 [xx XXX 1999]
7 *) Add OIDs for idea and blowfish in CBC mode. This will allow both
8 to be used in PKCS#5 v2.0 and S/MIME. Also add checking to
9 some routines that use cipher OIDs: some ciphers do not have OIDs
10 defined and so they cannot be used for S/MIME and PKCS#5 v2.0 for
14 *) Simplify the trust setting structure and code. Now we just have
15 two sequences of OIDs for trusted and rejected settings. These will
16 typically have values the same as the extended key usage extension
17 and any application specific purposes.
19 The trust checking code now has a default behaviour: it will just
20 check for an object with the same NID as the passed id. Functions can
21 be provided to override either the default behaviour or the behaviour
22 for a given id. SSL client, server and email already have functions
23 in place for compatability: they check the NID and also return "trusted"
24 if the certificate is self signed.
27 *) Add d2i,i2d bio/fp functions for PrivateKey: these convert the
28 traditional format into an EVP_PKEY structure.
31 *) Add a password callback function PEM_cb() which either prompts for
32 a password if usr_data is NULL or otherwise assumes it is a null
33 terminated password. Allow passwords to be passed on command line
34 environment or config files in a few more utilities.
37 *) Add a bunch of DER and PEM functions to handle PKCS#8 format private
38 keys. Add some short names for PKCS#8 PBE algorithms and allow them
39 to be specified on the command line for the pkcs8 and pkcs12 utilities.
43 *) Support for ASN1 "NULL" type. This could be handled before by using
44 ASN1_TYPE but there wasn't any function that would try to read a NULL
45 and produce an error if it couldn't. For compatibility we also have
46 ASN1_NULL_new() and ASN1_NULL_free() functions but these are faked and
47 don't allocate anything because they don't need to.
50 *) Initial support for MacOS is now provided. Examine INSTALL.MacOS
52 [Andy Polyakov, Roy Woods <roy@centicsystems.ca>]
54 *) Rebuild of the memory allocation routines used by OpenSSL code and
55 possibly others as well. The purpose is to make an interface that
56 provide hooks so anyone can build a separate set of allocation and
57 deallocation routines to be used by OpenSSL, for example if memory
58 pool implementations, or something else. The same is provided for
59 memory debugging code. OpenSSL already comes with code that finds
60 memory leaks, but this gives people a chance to debug other memory
63 With these changes, a new set of functions and macros have appeared:
65 CRYPTO_set_mem_debug_functions() [F]
66 CRYPTO_get_mem_debug_functions() [F]
67 CRYPTO_dbg_set_options() [F]
68 CRYPTO_dbg_get_options() [F]
69 CRYPTO_melloc_debug_init() [M]
71 The memory debug functions are NULL by default, unless the library
72 is compiled with CRYPTO_MDEBUG or friends is defined. If someone
73 wants to debug memory anyway, CRYPTO_malloc_debug_init() or
74 CRYPTO_set_mem_debug_functions() must be used.
76 Also, things like CRYPTO_set_mem_functions will always give the
77 expected result (the new set of functions is used for allocation
78 and deallocation) at all times, regardless of platform and compiler
81 To finish it up, some functions that were never use in any other
82 way than through macros have a new API and new semantic:
88 All macros of value have retained their old syntax.
91 *) Some S/MIME fixes. The OID for SMIMECapabilities was wrong, the
92 ordering of SMIMECapabilities wasn't in "strength order" and there
93 was a missing NULL in the AlgorithmIdentifier for the SHA1 signature
97 *) Some ASN1 types with illegal zero length encoding (INTEGER,
98 ENUMERATED and OBJECT IDENTIFIER) choked the ASN1 routines.
99 [Frans Heymans <fheymans@isaserver.be>, modified by Steve Henson]
101 *) Merge in my S/MIME library for OpenSSL. This provides a simple
102 S/MIME API on top of the PKCS#7 code, a MIME parser (with enough
103 functionality to handle multipart/signed properly) and a utility
104 called 'smime' to call all this stuff. This is based on code I
105 originally wrote for Celo who have kindly allowed it to be
109 *) Add variants des_set_key_checked and des_set_key_unchecked of
110 des_set_key (aka des_key_sched). Global variable des_check_key
111 decides which of these is called by des_set_key; this way
112 des_check_key behaves as it always did, but applications and
113 the library itself, which was buggy for des_check_key == 1,
114 have a cleaner way to pick the version they need.
117 *) New function PKCS12_newpass() which changes the password of a
121 *) Modify X509_TRUST and X509_PURPOSE so it also uses a static and
122 dynamic mix. In both cases the ids can be used as an index into the
123 table. Also modified the X509_TRUST_add() and X509_PURPOSE_add()
124 functions so they accept a list of the field values and the
125 application doesn't need to directly manipulate the X509_TRUST
129 *) Modify the ASN1_STRING_TABLE stuff so it also uses bsearch and doesn't
133 *) Modify the way the V3 extension code looks up extensions. This now
134 works in a similar way to the object code: we have some "standard"
135 extensions in a static table which is searched with OBJ_bsearch()
136 and the application can add dynamic ones if needed. The file
137 crypto/x509v3/ext_dat.h now has the info: this file needs to be
138 updated whenever a new extension is added to the core code and kept
139 in ext_nid order. There is a simple program 'tabtest.c' which checks
140 this. New extensions are not added too often so this file can readily
141 be maintained manually.
143 There are two big advantages in doing things this way. The extensions
144 can be looked up immediately and no longer need to be "added" using
145 X509V3_add_standard_extensions(): this function now does nothing.
146 [Side note: I get *lots* of email saying the extension code doesn't
147 work because people forget to call this function]
148 Also no dynamic allocation is done unless new extensions are added:
149 so if we don't add custom extensions there is no need to call
150 X509V3_EXT_cleanup().
153 *) Modify enc utility's salting as follows: make salting the default. Add a
154 magic header, so unsalted files fail gracefully instead of just decrypting
155 to garbage. This is because not salting is a big security hole, so people
156 should be discouraged from doing it.
159 *) Fixes and enhancements to the 'x509' utility. It allowed a message
160 digest to be passed on the command line but it only used this
161 parameter when signing a certificate. Modified so all relevant
162 operations are affected by the digest parameter including the
163 -fingerprint and -x509toreq options. Also -x509toreq choked if a
164 DSA key was used because it didn't fix the digest.
167 *) Initial certificate chain verify code. Currently tests the untrusted
168 certificates for consistency with the verify purpose (which is set
169 when the X509_STORE_CTX structure is set up) and checks the pathlength.
171 There is a NO_CHAIN_VERIFY compilation option to keep the old behaviour:
172 this is because it will reject chains with invalid extensions whereas
173 every previous version of OpenSSL and SSLeay made no checks at all.
175 Trust code: checks the root CA for the relevant trust settings. Trust
176 settings have an initial value consistent with the verify purpose: e.g.
177 if the verify purpose is for SSL client use it expects the CA to be
178 trusted for SSL client use. However the default value can be changed to
179 permit custom trust settings: one example of this would be to only trust
180 certificates from a specific "secure" set of CAs.
182 Also added X509_STORE_CTX_new() and X509_STORE_CTX_free() functions
183 which should be used for version portability: especially since the
184 verify structure is likely to change more often now.
186 SSL integration. Add purpose and trust to SSL_CTX and SSL and functions
187 to set them. If not set then assume SSL clients will verify SSL servers
190 Two new options to the verify program: -untrusted allows a set of
191 untrusted certificates to be passed in and -purpose which sets the
192 intended purpose of the certificate. If a purpose is set then the
193 new chain verify code is used to check extension consistency.
196 *) Support for the authority information access extension.
199 *) Modify RSA and DSA PEM read routines to transparently handle
200 PKCS#8 format private keys. New *_PUBKEY_* functions that handle
201 public keys in a format compatible with certificate
202 SubjectPublicKeyInfo structures. Unfortunately there were already
203 functions called *_PublicKey_* which used various odd formats so
204 these are retained for compatibility: however the DSA variants were
205 never in a public release so they have been deleted. Changed dsa/rsa
206 utilities to handle the new format: note no releases ever handled public
207 keys so we should be OK.
209 The primary motivation for this change is to avoid the same fiasco
210 that dogs private keys: there are several incompatible private key
211 formats some of which are standard and some OpenSSL specific and
212 require various evil hacks to allow partial transparent handling and
213 even then it doesn't work with DER formats. Given the option anything
214 other than PKCS#8 should be dumped: but the other formats have to
215 stay in the name of compatibility.
217 With public keys and the benefit of hindsight one standard format
218 is used which works with EVP_PKEY, RSA or DSA structures: though
219 it clearly returns an error if you try to read the wrong kind of key.
221 Added a -pubkey option to the 'x509' utility to output the public key.
222 Also rename the EVP_PKEY_get_*() to EVP_PKEY_rget_*() and add
223 EVP_PKEY_rset_*() functions that do the same as the EVP_PKEY_assign_*()
224 except they up the reference count of the added key (they don't "swallow"
228 *) Fixes to crypto/x509/by_file.c the code to read in certificates and
229 CRLs would fail if the file contained no certificates or no CRLs:
230 added a new function to read in both types and return the number
231 read: this means that if none are read it will be an error. The
232 DER versions of the certificate and CRL reader would always fail
233 because it isn't possible to mix certificates and CRLs in DER format
234 without choking one or the other routine. Changed this to just read
235 a certificate: this is the best we can do. Also modified the code
236 in apps/verify.c to take notice of return codes: it was previously
237 attempting to read in certificates from NULL pointers and ignoring
238 any errors: this is one reason why the cert and CRL reader seemed
239 to work. It doesn't check return codes from the default certificate
240 routines: these may well fail if the certificates aren't installed.
243 *) Code to support otherName option in GeneralName.
246 *) First update to verify code. Change the verify utility
247 so it warns if it is passed a self signed certificate:
248 for consistency with the normal behaviour. X509_verify
249 has been modified to it will now verify a self signed
250 certificate if *exactly* the same certificate appears
251 in the store: it was previously impossible to trust a
252 single self signed certificate. This means that:
253 openssl verify ss.pem
254 now gives a warning about a self signed certificate but
255 openssl verify -CAfile ss.pem ss.pem
259 *) For servers, store verify_result in SSL_SESSION data structure
260 (and add it to external session representation).
261 This is needed when client certificate verifications fails,
262 but an application-provided verification callback (set by
263 SSL_CTX_set_cert_verify_callback) allows accepting the session
264 anyway (i.e. leaves x509_store_ctx->error != X509_V_OK
265 but returns 1): When the session is reused, we have to set
266 ssl->verify_result to the appropriate error code to avoid
268 [Bodo Moeller, problem pointed out by Lutz Jaenicke]
270 *) Fix a bug in the new PKCS#7 code: it didn't consider the
271 case in PKCS7_dataInit() where the signed PKCS7 structure
272 didn't contain any existing data because it was being created.
273 [Po-Cheng Chen <pocheng@nst.com.tw>, slightly modified by Steve Henson]
275 *) Add a salt to the key derivation routines in enc.c. This
276 forms the first 8 bytes of the encrypted file. Also add a
277 -S option to allow a salt to be input on the command line.
280 *) New function X509_cmp(). Oddly enough there wasn't a function
281 to compare two certificates. We do this by working out the SHA1
282 hash and comparing that. X509_cmp() will be needed by the trust
286 *) Correctly increment the reference count in the SSL_SESSION pointer
287 returned from SSL_get_session().
288 [Geoff Thorpe <geoff@eu.c2.net>]
290 *) Fix for 'req': it was adding a null to request attributes.
291 Also change the X509_LOOKUP and X509_INFO code to handle
292 certificate auxiliary information.
295 *) Add support for 40 and 64 bit RC2 and RC4 algorithms: document
299 *) Add the possibility to add extra information to the memory leak
300 detecting output, to form tracebacks, showing from where each
301 allocation was originated. Also updated sid code to be multi-
305 *) Add options -text and -noout to pkcs7 utility and delete the
306 encryption options which never did anything. Update docs.
309 *) Add options to some of the utilities to allow the pass phrase
310 to be included on either the command line (not recommended on
311 OSes like Unix) or read from the environment. Update the
312 manpages and fix a few bugs.
315 *) Add a few manpages for some of the openssl commands.
318 *) Fix the -revoke option in ca. It was freeing up memory twice,
319 leaking and not finding already revoked certificates.
322 *) Extensive changes to support certificate auxiliary information.
323 This involves the use of X509_CERT_AUX structure and X509_AUX
324 functions. An X509_AUX function such as PEM_read_X509_AUX()
325 can still read in a certificate file in the usual way but it
326 will also read in any additional "auxiliary information". By
327 doing things this way a fair degree of compatibility can be
328 retained: existing certificates can have this information added
329 using the new 'x509' options.
331 Current auxiliary information includes an "alias" and some trust
332 settings. The trust settings will ultimately be used in enhanced
333 certificate chain verification routines: currently a certificate
334 can only be trusted if it is self signed and then it is trusted
338 *) Fix assembler for Alpha (tested only on DEC OSF not Linux or *BSD). The
339 problem was that one of the replacement routines had not been working since
340 SSLeay releases. For now the offending routine has been replaced with
341 non-optimised assembler. Even so, this now gives around 95% performance
342 improvement for 1024 bit RSA signs.
345 *) Hack to fix PKCS#7 decryption when used with some unorthodox RC2
346 handling. Most clients have the effective key size in bits equal to
347 the key length in bits: so a 40 bit RC2 key uses a 40 bit (5 byte) key.
348 A few however don't do this and instead use the size of the decrypted key
349 to determine the RC2 key length and the AlgorithmIdentifier to determine
350 the effective key length. In this case the effective key lenth can still
351 be 40 bits but the key length can be 168 bits for example. This is fixed
352 by manually forcing an RC2 key into the EVP_PKEY structure because the
353 EVP code can't currently handle unusual RC2 key sizes: it always assumes
354 the key length and effective key length are equal.
357 *) Add a bunch of functions that should simplify the creation of
358 X509_NAME structures. Now you should be able to do:
359 X509_NAME_add_entry_by_txt(nm, "CN", MBSTRING_ASC, "Steve", -1, -1, 0);
360 and have it automatically work out the correct field type and fill in
361 the structures. The more adventurous can try:
362 X509_NAME_add_entry_by_txt(nm, field, MBSTRING_UTF8, str, -1, -1, 0);
363 and it will (hopefully) work out the correct multibyte encoding.
366 *) Change the 'req' utility to use the new field handling and multibyte
367 copy routines. Before the DN field creation was handled in an ad hoc
368 way in req, ca, and x509 which was rather broken and didn't support
369 BMPStrings or UTF8Strings. Since some software doesn't implement
370 BMPStrings or UTF8Strings yet, they can be enabled using the config file
371 using the dirstring_type option. See the new comment in the default
372 openssl.cnf for more info.
375 *) Make crypto/rand/md_rand.c more robust:
376 - Assure unique random numbers after fork().
377 - Make sure that concurrent threads access the global counter and
378 md serializably so that we never lose entropy in them
379 or use exactly the same state in multiple threads.
380 Access to the large state is not always serializable because
381 the additional locking could be a performance killer, and
382 md should be large enough anyway.
385 *) New file apps/app_rand.c with commonly needed functionality
386 for handling the random seed file.
388 Use the random seed file in some applications that previously did not:
390 dsaparam -genkey (which also ignored its '-rand' option),
394 Except on systems with /dev/urandom, it is crucial to have a random
395 seed file at least for key creation, DSA signing, and for DH exchanges;
396 for RSA signatures we could do without one.
398 gendh and gendsa (unlike genrsa) used to read only the first byte
399 of each file listed in the '-rand' option. The function as previously
400 found in genrsa is now in app_rand.c and is used by all programs
401 that support '-rand'.
404 *) In RAND_write_file, use mode 0600 for creating files;
405 don't just chmod when it may be too late.
408 *) Report an error from X509_STORE_load_locations
409 when X509_LOOKUP_load_file or X509_LOOKUP_add_dir failed.
412 *) New function ASN1_mbstring_copy() this copies a string in either
413 ASCII, Unicode, Universal (4 bytes per character) or UTF8 format
414 into an ASN1_STRING type. A mask of permissible types is passed
415 and it chooses the "minimal" type to use or an error if not type
419 *) Add function equivalents to the various macros in asn1.h. The old
420 macros are retained with an M_ prefix. Code inside the library can
421 use the M_ macros. External code (including the openssl utility)
422 should *NOT* in order to be "shared library friendly".
425 *) Add various functions that can check a certificate's extensions
426 to see if it usable for various purposes such as SSL client,
427 server or S/MIME and CAs of these types. This is currently
428 VERY EXPERIMENTAL but will ultimately be used for certificate chain
429 verification. Also added a -purpose flag to x509 utility to
430 print out all the purposes.
433 *) Add a CRYPTO_EX_DATA to X509 certificate structure and associated
437 *) New X509V3_{X509,CRL,REVOKED}_get_d2i() functions. These will search
438 for, obtain and decode and extension and obtain its critical flag.
439 This allows all the necessary extension code to be handled in a
440 single function call.
443 *) RC4 tune-up featuring 30-40% performance improvement on most RISC
444 platforms. See crypto/rc4/rc4_enc.c for further details.
447 *) New -noout option to asn1parse. This causes no output to be produced
448 its main use is when combined with -strparse and -out to extract data
449 from a file (which may not be in ASN.1 format).
452 *) Fix for pkcs12 program. It was hashing an invalid certificate pointer
453 when producing the local key id.
454 [Richard Levitte <levitte@stacken.kth.se>]
456 *) New option -dhparam in s_server. This allows a DH parameter file to be
457 stated explicitly. If it is not stated then it tries the first server
458 certificate file. The previous behaviour hard coded the filename
462 *) Add -pubin and -pubout options to the rsa and dsa commands. These allow
463 a public key to be input or output. For example:
464 openssl rsa -in key.pem -pubout -out pubkey.pem
465 Also added necessary DSA public key functions to handle this.
468 *) Fix so PKCS7_dataVerify() doesn't crash if no certificates are contained
469 in the message. This was handled by allowing
470 X509_find_by_issuer_and_serial() to tolerate a NULL passed to it.
471 [Steve Henson, reported by Sampo Kellomaki <sampo@mail.neuronio.pt>]
473 *) Fix for bug in d2i_ASN1_bytes(): other ASN1 functions add an extra null
474 to the end of the strings whereas this didn't. This would cause problems
475 if strings read with d2i_ASN1_bytes() were later modified.
476 [Steve Henson, reported by Arne Ansper <arne@ats.cyber.ee>]
478 *) Fix for base64 decode bug. When a base64 bio reads only one line of
479 data and it contains EOF it will end up returning an error. This is
480 caused by input 46 bytes long. The cause is due to the way base64
481 BIOs find the start of base64 encoded data. They do this by trying a
482 trial decode on each line until they find one that works. When they
483 do a flag is set and it starts again knowing it can pass all the
484 data directly through the decoder. Unfortunately it doesn't reset
485 the context it uses. This means that if EOF is reached an attempt
486 is made to pass two EOFs through the context and this causes the
487 resulting error. This can also cause other problems as well. As is
488 usual with these problems it takes *ages* to find and the fix is
489 trivial: move one line.
490 [Steve Henson, reported by ian@uns.ns.ac.yu (Ivan Nejgebauer) ]
492 *) Ugly workaround to get s_client and s_server working under Windows. The
493 old code wouldn't work because it needed to select() on sockets and the
494 tty (for keypresses and to see if data could be written). Win32 only
495 supports select() on sockets so we select() with a 1s timeout on the
496 sockets and then see if any characters are waiting to be read, if none
497 are present then we retry, we also assume we can always write data to
498 the tty. This isn't nice because the code then blocks until we've
499 received a complete line of data and it is effectively polling the
500 keyboard at 1s intervals: however it's quite a bit better than not
501 working at all :-) A dedicated Windows application might handle this
502 with an event loop for example.
505 *) Enhance RSA_METHOD structure. Now there are two extra methods, rsa_sign
506 and rsa_verify. When the RSA_FLAGS_SIGN_VER option is set these functions
507 will be called when RSA_sign() and RSA_verify() are used. This is useful
508 if rsa_pub_dec() and rsa_priv_enc() equivalents are not available.
509 For this to work properly RSA_public_decrypt() and RSA_private_encrypt()
510 should *not* be used: RSA_sign() and RSA_verify() must be used instead.
511 This necessitated the support of an extra signature type NID_md5_sha1
512 for SSL signatures and modifications to the SSL library to use it instead
513 of calling RSA_public_decrypt() and RSA_private_encrypt().
516 *) Add new -verify -CAfile and -CApath options to the crl program, these
517 will lookup a CRL issuers certificate and verify the signature in a
518 similar way to the verify program. Tidy up the crl program so it
519 no longer acesses structures directly. Make the ASN1 CRL parsing a bit
520 less strict. It will now permit CRL extensions even if it is not
521 a V2 CRL: this will allow it to tolerate some broken CRLs.
524 *) Initialize all non-automatic variables each time one of the openssl
525 sub-programs is started (this is necessary as they may be started
526 multiple times from the "OpenSSL>" prompt).
527 [Lennart Bang, Bodo Moeller]
529 *) Preliminary compilation option RSA_NULL which disables RSA crypto without
530 removing all other RSA functionality (this is what NO_RSA does). This
531 is so (for example) those in the US can disable those operations covered
532 by the RSA patent while allowing storage and parsing of RSA keys and RSA
536 *) Non-copying interface to BIO pairs.
537 (still largely untested)
540 *) New function ANS1_tag2str() to convert an ASN1 tag to a descriptive
541 ASCII string. This was handled independently in various places before.
544 *) New functions UTF8_getc() and UTF8_putc() that parse and generate
545 UTF8 strings a character at a time.
548 *) Use client_version from client hello to select the protocol
549 (s23_srvr.c) and for RSA client key exchange verification
550 (s3_srvr.c), as required by the SSL 3.0/TLS 1.0 specifications.
553 *) Add various utility functions to handle SPKACs, these were previously
554 handled by poking round in the structure internals. Added new function
555 NETSCAPE_SPKI_print() to print out SPKAC and a new utility 'spkac' to
556 print, verify and generate SPKACs. Based on an original idea from
557 Massimiliano Pala <madwolf@comune.modena.it> but extensively modified.
560 *) RIPEMD160 is operational on all platforms and is back in 'make test'.
563 *) Allow the config file extension section to be overwritten on the
564 command line. Based on an original idea from Massimiliano Pala
565 <madwolf@comune.modena.it>. The new option is called -extensions
566 and can be applied to ca, req and x509. Also -reqexts to override
567 the request extensions in req and -crlexts to override the crl extensions
571 *) Add new feature to the SPKAC handling in ca. Now you can include
572 the same field multiple times by preceding it by "XXXX." for example:
575 this is the same syntax as used in the req config file.
578 *) Allow certificate extensions to be added to certificate requests. These
579 are specified in a 'req_extensions' option of the req section of the
580 config file. They can be printed out with the -text option to req but
581 are otherwise ignored at present.
584 *) Fix a horrible bug in enc_read() in crypto/evp/bio_enc.c: if the first
585 data read consists of only the final block it would not decrypted because
586 EVP_CipherUpdate() would correctly report zero bytes had been decrypted.
587 A misplaced 'break' also meant the decrypted final block might not be
588 copied until the next read.
591 *) Initial support for DH_METHOD. Again based on RSA_METHOD. Also added
592 a few extra parameters to the DH structure: these will be useful if
593 for example we want the value of 'q' or implement X9.42 DH.
596 *) Initial support for DSA_METHOD. This is based on the RSA_METHOD and
597 provides hooks that allow the default DSA functions or functions on a
598 "per key" basis to be replaced. This allows hardware acceleration and
599 hardware key storage to be handled without major modification to the
600 library. Also added low level modexp hooks and CRYPTO_EX structure and
601 associated functions.
604 *) Add a new flag to memory BIOs, BIO_FLAG_MEM_RDONLY. This marks the BIO
605 as "read only": it can't be written to and the buffer it points to will
606 not be freed. Reading from a read only BIO is much more efficient than
607 a normal memory BIO. This was added because there are several times when
608 an area of memory needs to be read from a BIO. The previous method was
609 to create a memory BIO and write the data to it, this results in two
610 copies of the data and an O(n^2) reading algorithm. There is a new
611 function BIO_new_mem_buf() which creates a read only memory BIO from
612 an area of memory. Also modified the PKCS#7 routines to use read only
616 *) Bugfix: ssl23_get_client_hello did not work properly when called in
617 state SSL23_ST_SR_CLNT_HELLO_B, i.e. when the first 7 bytes of
618 a SSLv2-compatible client hello for SSLv3 or TLSv1 could be read,
619 but a retry condition occured while trying to read the rest.
622 *) The PKCS7_ENC_CONTENT_new() function was setting the content type as
623 NID_pkcs7_encrypted by default: this was wrong since this should almost
624 always be NID_pkcs7_data. Also modified the PKCS7_set_type() to handle
625 the encrypted data type: this is a more sensible place to put it and it
626 allows the PKCS#12 code to be tidied up that duplicated this
630 *) Changed obj_dat.pl script so it takes its input and output files on
631 the command line. This should avoid shell escape redirection problems
635 *) Initial support for certificate extension requests, these are included
636 in things like Xenroll certificate requests. Included functions to allow
637 extensions to be obtained and added.
640 *) -crlf option to s_client and s_server for sending newlines as
641 CRLF (as required by many protocols).
644 Changes between 0.9.3a and 0.9.4 [09 Aug 1999]
646 *) Install libRSAglue.a when OpenSSL is built with RSAref.
647 [Ralf S. Engelschall]
649 *) A few more ``#ifndef NO_FP_API / #endif'' pairs for consistency.
650 [Andrija Antonijevic <TheAntony2@bigfoot.com>]
652 *) Fix -startdate and -enddate (which was missing) arguments to 'ca'
656 *) New function DSA_dup_DH, which duplicates DSA parameters/keys as
657 DH parameters/keys (q is lost during that conversion, but the resulting
658 DH parameters contain its length).
660 For 1024-bit p, DSA_generate_parameters followed by DSA_dup_DH is
661 much faster than DH_generate_parameters (which creates parameters
662 where p = 2*q + 1), and also the smaller q makes DH computations
663 much more efficient (160-bit exponentiation instead of 1024-bit
664 exponentiation); so this provides a convenient way to support DHE
665 ciphersuites in SSL/TLS servers (see ssl/ssltest.c). It is of
666 utter importance to use
667 SSL_CTX_set_options(s_ctx, SSL_OP_SINGLE_DH_USE);
669 SSL_set_options(s_ctx, SSL_OP_SINGLE_DH_USE);
670 when such DH parameters are used, because otherwise small subgroup
671 attacks may become possible!
674 *) Avoid memory leak in i2d_DHparams.
677 *) Allow the -k option to be used more than once in the enc program:
678 this allows the same encrypted message to be read by multiple recipients.
681 *) New function OBJ_obj2txt(buf, buf_len, a, no_name), this converts
682 an ASN1_OBJECT to a text string. If the "no_name" parameter is set then
683 it will always use the numerical form of the OID, even if it has a short
687 *) Added an extra RSA flag: RSA_FLAG_EXT_PKEY. Previously the rsa_mod_exp
688 method only got called if p,q,dmp1,dmq1,iqmp components were present,
689 otherwise bn_mod_exp was called. In the case of hardware keys for example
690 no private key components need be present and it might store extra data
691 in the RSA structure, which cannot be accessed from bn_mod_exp.
692 By setting RSA_FLAG_EXT_PKEY rsa_mod_exp will always be called for
693 private key operations.
696 *) Added support for SPARC Linux.
699 *) pem_password_cb function type incompatibly changed from
700 typedef int pem_password_cb(char *buf, int size, int rwflag);
702 ....(char *buf, int size, int rwflag, void *userdata);
703 so that applications can pass data to their callbacks:
704 The PEM[_ASN1]_{read,write}... functions and macros now take an
705 additional void * argument, which is just handed through whenever
706 the password callback is called.
707 [Damien Miller <dmiller@ilogic.com.au>; tiny changes by Bodo Moeller]
709 New function SSL_CTX_set_default_passwd_cb_userdata.
711 Compatibility note: As many C implementations push function arguments
712 onto the stack in reverse order, the new library version is likely to
713 interoperate with programs that have been compiled with the old
714 pem_password_cb definition (PEM_whatever takes some data that
715 happens to be on the stack as its last argument, and the callback
716 just ignores this garbage); but there is no guarantee whatsoever that
719 *) The -DPLATFORM="\"$(PLATFORM)\"" definition and the similar -DCFLAGS=...
720 (both in crypto/Makefile.ssl for use by crypto/cversion.c) caused
721 problems not only on Windows, but also on some Unix platforms.
722 To avoid problematic command lines, these definitions are now in an
723 auto-generated file crypto/buildinf.h (created by crypto/Makefile.ssl
724 for standard "make" builds, by util/mk1mf.pl for "mk1mf" builds).
727 *) MIPS III/IV assembler module is reimplemented.
730 *) More DES library cleanups: remove references to srand/rand and
731 delete an unused file.
734 *) Add support for the the free Netwide assembler (NASM) under Win32,
735 since not many people have MASM (ml) and it can be hard to obtain.
736 This is currently experimental but it seems to work OK and pass all
737 the tests. Check out INSTALL.W32 for info.
740 *) Fix memory leaks in s3_clnt.c: All non-anonymous SSL3/TLS1 connections
741 without temporary keys kept an extra copy of the server key,
742 and connections with temporary keys did not free everything in case
746 *) New function RSA_check_key and new openssl rsa option -check
747 for verifying the consistency of RSA keys.
748 [Ulf Moeller, Bodo Moeller]
750 *) Various changes to make Win32 compile work:
751 1. Casts to avoid "loss of data" warnings in p5_crpt2.c
752 2. Change unsigned int to int in b_dump.c to avoid "signed/unsigned
753 comparison" warnings.
754 3. Add sk_<TYPE>_sort to DEF file generator and do make update.
757 *) Add a debugging option to PKCS#5 v2 key generation function: when
758 you #define DEBUG_PKCS5V2 passwords, salts, iteration counts and
759 derived keys are printed to stderr.
762 *) Copy the flags in ASN1_STRING_dup().
763 [Roman E. Pavlov <pre@mo.msk.ru>]
765 *) The x509 application mishandled signing requests containing DSA
766 keys when the signing key was also DSA and the parameters didn't match.
768 It was supposed to omit the parameters when they matched the signing key:
769 the verifying software was then supposed to automatically use the CA's
770 parameters if they were absent from the end user certificate.
772 Omitting parameters is no longer recommended. The test was also
773 the wrong way round! This was probably due to unusual behaviour in
774 EVP_cmp_parameters() which returns 1 if the parameters match.
775 This meant that parameters were omitted when they *didn't* match and
776 the certificate was useless. Certificates signed with 'ca' didn't have
778 [Steve Henson, reported by Doug Erickson <Doug.Erickson@Part.NET>]
780 *) Memory leak checking (-DCRYPTO_MDEBUG) had some problems.
781 The interface is as follows:
783 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ON) aka MemCheck_start(),
784 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_OFF) aka MemCheck_stop();
785 "off" is now the default.
786 The library internally uses
787 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_DISABLE) aka MemCheck_off(),
788 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ENABLE) aka MemCheck_on()
789 to disable memory-checking temporarily.
791 Some inconsistent states that previously were possible (and were
792 even the default) are now avoided.
794 -DCRYPTO_MDEBUG_TIME is new and additionally stores the current time
795 with each memory chunk allocated; this is occasionally more helpful
796 than just having a counter.
798 -DCRYPTO_MDEBUG_THREAD is also new and adds the thread ID.
800 -DCRYPTO_MDEBUG_ALL enables all of the above, plus any future
804 *) Introduce "mode" for SSL structures (with defaults in SSL_CTX),
805 which largely parallels "options", but is for changing API behaviour,
806 whereas "options" are about protocol behaviour.
807 Initial "mode" flags are:
809 SSL_MODE_ENABLE_PARTIAL_WRITE Allow SSL_write to report success when
810 a single record has been written.
811 SSL_MODE_ACCEPT_MOVING_WRITE_BUFFER Don't insist that SSL_write
812 retries use the same buffer location.
813 (But all of the contents must be
817 *) Bugfix: SSL_set_mode ignored its parameter, only SSL_CTX_set_mode
820 *) Fix problems with no-hmac etc.
821 [Ulf Möller, pointed out by Brian Wellington <bwelling@tislabs.com>]
823 *) New functions RSA_get_default_method(), RSA_set_method() and
824 RSA_get_method(). These allows replacement of RSA_METHODs without having
825 to mess around with the internals of an RSA structure.
828 *) Fix memory leaks in DSA_do_sign and DSA_is_prime.
829 Also really enable memory leak checks in openssl.c and in some
831 [Chad C. Mulligan, Bodo Moeller]
833 *) Fix a bug in d2i_ASN1_INTEGER() and i2d_ASN1_INTEGER() which can mess
834 up the length of negative integers. This has now been simplified to just
835 store the length when it is first determined and use it later, rather
836 than trying to keep track of where data is copied and updating it to
838 [Steve Henson, reported by Brien Wheeler
839 <bwheeler@authentica-security.com>]
841 *) Add a new function PKCS7_signatureVerify. This allows the verification
842 of a PKCS#7 signature but with the signing certificate passed to the
843 function itself. This contrasts with PKCS7_dataVerify which assumes the
844 certificate is present in the PKCS#7 structure. This isn't always the
845 case: certificates can be omitted from a PKCS#7 structure and be
846 distributed by "out of band" means (such as a certificate database).
849 *) Complete the PEM_* macros with DECLARE_PEM versions to replace the
850 function prototypes in pem.h, also change util/mkdef.pl to add the
851 necessary function names.
854 *) mk1mf.pl (used by Windows builds) did not properly read the
855 options set by Configure in the top level Makefile, and Configure
856 was not even able to write more than one option correctly.
857 Fixed, now "no-idea no-rc5 -DCRYPTO_MDEBUG" etc. works as intended.
860 *) New functions CONF_load_bio() and CONF_load_fp() to allow a config
861 file to be loaded from a BIO or FILE pointer. The BIO version will
862 for example allow memory BIOs to contain config info.
865 *) New function "CRYPTO_num_locks" that returns CRYPTO_NUM_LOCKS.
866 Whoever hopes to achieve shared-library compatibility across versions
867 must use this, not the compile-time macro.
868 (Exercise 0.9.4: Which is the minimum library version required by
870 Note: All this applies only to multi-threaded programs, others don't
874 *) Add missing case to s3_clnt.c state machine -- one of the new SSL tests
875 through a BIO pair triggered the default case, i.e.
876 SSLerr(...,SSL_R_UNKNOWN_STATE).
879 *) New "BIO pair" concept (crypto/bio/bss_bio.c) so that applications
880 can use the SSL library even if none of the specific BIOs is
884 *) Fix a bug in i2d_DSAPublicKey() which meant it returned the wrong value
885 for the encoded length.
886 [Jeon KyoungHo <khjeon@sds.samsung.co.kr>]
888 *) Add initial documentation of the X509V3 functions.
891 *) Add a new pair of functions PEM_write_PKCS8PrivateKey() and
892 PEM_write_bio_PKCS8PrivateKey() that are equivalent to
893 PEM_write_PrivateKey() and PEM_write_bio_PrivateKey() but use the more
894 secure PKCS#8 private key format with a high iteration count.
897 *) Fix determination of Perl interpreter: A perl or perl5
898 _directory_ in $PATH was also accepted as the interpreter.
899 [Ralf S. Engelschall]
901 *) Fix demos/sign/sign.c: well there wasn't anything strictly speaking
902 wrong with it but it was very old and did things like calling
903 PEM_ASN1_read() directly and used MD5 for the hash not to mention some
907 *) Fix demos/selfsign.c: it used obsolete and deleted functions, changed
908 to use the new extension code.
911 *) Implement the PEM_read/PEM_write functions in crypto/pem/pem_all.c
912 with macros. This should make it easier to change their form, add extra
913 arguments etc. Fix a few PEM prototypes which didn't have cipher as a
917 *) Add to configuration table a new entry that can specify an alternative
918 name for unistd.h (for pre-POSIX systems); we need this for NeXTstep,
919 according to Mark Crispin <MRC@Panda.COM>.
923 *) DES CBC did not update the IV. Weird.
926 des_cbc_encrypt does not update the IV, but des_ncbc_encrypt does.
927 Changing the behaviour of the former might break existing programs --
928 where IV updating is needed, des_ncbc_encrypt can be used.
931 *) When bntest is run from "make test" it drives bc to check its
932 calculations, as well as internally checking them. If an internal check
933 fails, it needs to cause bc to give a non-zero result or make test carries
934 on without noticing the failure. Fixed.
937 *) DES library cleanups.
940 *) Add support for PKCS#5 v2.0 PBE algorithms. This will permit PKCS#8 to be
941 used with any cipher unlike PKCS#5 v1.5 which can at most handle 64 bit
942 ciphers. NOTE: although the key derivation function has been verified
943 against some published test vectors it has not been extensively tested
944 yet. Added a -v2 "cipher" option to pkcs8 application to allow the use
948 *) Instead of "mkdir -p", which is not fully portable, use new
949 Perl script "util/mkdir-p.pl".
952 *) Rewrite the way password based encryption (PBE) is handled. It used to
953 assume that the ASN1 AlgorithmIdentifier parameter was a PBEParameter
954 structure. This was true for the PKCS#5 v1.5 and PKCS#12 PBE algorithms
955 but doesn't apply to PKCS#5 v2.0 where it can be something else. Now
956 the 'parameter' field of the AlgorithmIdentifier is passed to the
957 underlying key generation function so it must do its own ASN1 parsing.
958 This has also changed the EVP_PBE_CipherInit() function which now has a
959 'parameter' argument instead of literal salt and iteration count values
960 and the function EVP_PBE_ALGOR_CipherInit() has been deleted.
963 *) Support for PKCS#5 v1.5 compatible password based encryption algorithms
964 and PKCS#8 functionality. New 'pkcs8' application linked to openssl.
965 Needed to change the PEM_STRING_EVP_PKEY value which was just "PRIVATE
966 KEY" because this clashed with PKCS#8 unencrypted string. Since this
967 value was just used as a "magic string" and not used directly its
968 value doesn't matter.
971 *) Introduce some semblance of const correctness to BN. Shame C doesn't
975 *) "linux-sparc64" configuration (ultrapenguin).
976 [Ray Miller <ray.miller@oucs.ox.ac.uk>]
977 "linux-sparc" configuration.
978 [Christian Forster <fo@hawo.stw.uni-erlangen.de>]
980 *) config now generates no-xxx options for missing ciphers.
983 *) Support the EBCDIC character set (work in progress).
984 File ebcdic.c not yet included because it has a different license.
985 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>]
987 *) Support BS2000/OSD-POSIX.
988 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>]
990 *) Make callbacks for key generation use void * instead of char *.
993 *) Make S/MIME samples compile (not yet tested).
996 *) Additional typesafe stacks.
999 *) New configuration variants "bsdi-elf-gcc" (BSD/OS 4.x).
1003 Changes between 0.9.3 and 0.9.3a [29 May 1999]
1005 *) New configuration variant "sco5-gcc".
1007 *) Updated some demos.
1008 [Sean O Riordain, Wade Scholine]
1010 *) Add missing BIO_free at exit of pkcs12 application.
1013 *) Fix memory leak in conf.c.
1016 *) Updates for Win32 to assembler version of MD5.
1019 *) Set #! path to perl in apps/der_chop to where we found it
1020 instead of using a fixed path.
1023 *) SHA library changes for irix64-mips4-cc.
1026 *) Improvements for VMS support.
1030 Changes between 0.9.2b and 0.9.3 [24 May 1999]
1032 *) Bignum library bug fix. IRIX 6 passes "make test" now!
1033 This also avoids the problems with SC4.2 and unpatched SC5.
1034 [Andy Polyakov <appro@fy.chalmers.se>]
1036 *) New functions sk_num, sk_value and sk_set to replace the previous macros.
1037 These are required because of the typesafe stack would otherwise break
1038 existing code. If old code used a structure member which used to be STACK
1039 and is now STACK_OF (for example cert in a PKCS7_SIGNED structure) with
1040 sk_num or sk_value it would produce an error because the num, data members
1041 are not present in STACK_OF. Now it just produces a warning. sk_set
1042 replaces the old method of assigning a value to sk_value
1043 (e.g. sk_value(x, i) = y) which the library used in a few cases. Any code
1044 that does this will no longer work (and should use sk_set instead) but
1045 this could be regarded as a "questionable" behaviour anyway.
1048 *) Fix most of the other PKCS#7 bugs. The "experimental" code can now
1049 correctly handle encrypted S/MIME data.
1052 *) Change type of various DES function arguments from des_cblock
1053 (which means, in function argument declarations, pointer to char)
1054 to des_cblock * (meaning pointer to array with 8 char elements),
1055 which allows the compiler to do more typechecking; it was like
1056 that back in SSLeay, but with lots of ugly casts.
1058 Introduce new type const_des_cblock.
1061 *) Reorganise the PKCS#7 library and get rid of some of the more obvious
1062 problems: find RecipientInfo structure that matches recipient certificate
1063 and initialise the ASN1 structures properly based on passed cipher.
1066 *) Belatedly make the BN tests actually check the results.
1069 *) Fix the encoding and decoding of negative ASN1 INTEGERS and conversion
1070 to and from BNs: it was completely broken. New compilation option
1071 NEG_PUBKEY_BUG to allow for some broken certificates that encode public
1072 key elements as negative integers.
1075 *) Reorganize and speed up MD5.
1076 [Andy Polyakov <appro@fy.chalmers.se>]
1079 [Richard Levitte <richard@levitte.org>]
1081 *) New option -out to asn1parse to allow the parsed structure to be
1082 output to a file. This is most useful when combined with the -strparse
1083 option to examine the output of things like OCTET STRINGS.
1086 *) Make SSL library a little more fool-proof by not requiring any longer
1087 that SSL_set_{accept,connect}_state be called before
1088 SSL_{accept,connect} may be used (SSL_set_..._state is omitted
1089 in many applications because usually everything *appeared* to work as
1090 intended anyway -- now it really works as intended).
1093 *) Move openssl.cnf out of lib/.
1096 *) Fix various things to let OpenSSL even pass ``egcc -pipe -O2 -Wall
1097 -Wshadow -Wpointer-arith -Wcast-align -Wmissing-prototypes
1098 -Wmissing-declarations -Wnested-externs -Winline'' with EGCS 1.1.2+
1099 [Ralf S. Engelschall]
1101 *) Various fixes to the EVP and PKCS#7 code. It may now be able to
1102 handle PKCS#7 enveloped data properly.
1103 [Sebastian Akerman <sak@parallelconsulting.com>, modified by Steve]
1105 *) Create a duplicate of the SSL_CTX's CERT in SSL_new instead of
1106 copying pointers. The cert_st handling is changed by this in
1107 various ways (and thus what used to be known as ctx->default_cert
1108 is now called ctx->cert, since we don't resort to s->ctx->[default_]cert
1109 any longer when s->cert does not give us what we need).
1110 ssl_cert_instantiate becomes obsolete by this change.
1111 As soon as we've got the new code right (possibly it already is?),
1112 we have solved a couple of bugs of the earlier code where s->cert
1113 was used as if it could not have been shared with other SSL structures.
1115 Note that using the SSL API in certain dirty ways now will result
1116 in different behaviour than observed with earlier library versions:
1117 Changing settings for an SSL_CTX *ctx after having done s = SSL_new(ctx)
1118 does not influence s as it used to.
1120 In order to clean up things more thoroughly, inside SSL_SESSION
1121 we don't use CERT any longer, but a new structure SESS_CERT
1122 that holds per-session data (if available); currently, this is
1123 the peer's certificate chain and, for clients, the server's certificate
1124 and temporary key. CERT holds only those values that can have
1125 meaningful defaults in an SSL_CTX.
1128 *) New function X509V3_EXT_i2d() to create an X509_EXTENSION structure
1129 from the internal representation. Various PKCS#7 fixes: remove some
1130 evil casts and set the enc_dig_alg field properly based on the signing
1134 *) Allow PKCS#12 password to be set from the command line or the
1135 environment. Let 'ca' get its config file name from the environment
1136 variables "OPENSSL_CONF" or "SSLEAY_CONF" (for consistency with 'req'
1140 *) Allow certificate policies extension to use an IA5STRING for the
1141 organization field. This is contrary to the PKIX definition but
1142 VeriSign uses it and IE5 only recognises this form. Document 'x509'
1146 *) Add PEDANTIC compiler flag to allow compilation with gcc -pedantic,
1147 without disallowing inline assembler and the like for non-pedantic builds.
1150 *) Support Borland C++ builder.
1151 [Janez Jere <jj@void.si>, modified by Ulf Möller]
1156 *) SHA-1 cleanups and performance enhancements.
1157 [Andy Polyakov <appro@fy.chalmers.se>]
1159 *) Sparc v8plus assembler for the bignum library.
1160 [Andy Polyakov <appro@fy.chalmers.se>]
1162 *) Accept any -xxx and +xxx compiler options in Configure.
1165 *) Update HPUX configuration.
1168 *) Add missing sk_<type>_unshift() function to safestack.h
1169 [Ralf S. Engelschall]
1171 *) New function SSL_CTX_use_certificate_chain_file that sets the
1172 "extra_cert"s in addition to the certificate. (This makes sense
1173 only for "PEM" format files, as chains as a whole are not
1177 *) Support verify_depth from the SSL API.
1178 x509_vfy.c had what can be considered an off-by-one-error:
1179 Its depth (which was not part of the external interface)
1180 was actually counting the number of certificates in a chain;
1181 now it really counts the depth.
1184 *) Bugfix in crypto/x509/x509_cmp.c: The SSLerr macro was used
1185 instead of X509err, which often resulted in confusing error
1186 messages since the error codes are not globally unique
1187 (e.g. an alleged error in ssl3_accept when a certificate
1188 didn't match the private key).
1190 *) New function SSL_CTX_set_session_id_context that allows to set a default
1191 value (so that you don't need SSL_set_session_id_context for each
1192 connection using the SSL_CTX).
1195 *) OAEP decoding bug fix.
1198 *) Support INSTALL_PREFIX for package builders, as proposed by
1202 *) New Configure options "threads" and "no-threads". For systems
1203 where the proper compiler options are known (currently Solaris
1204 and Linux), "threads" is the default.
1207 *) New script util/mklink.pl as a faster substitute for util/mklink.sh.
1210 *) Install various scripts to $(OPENSSLDIR)/misc, not to
1211 $(INSTALLTOP)/bin -- they shouldn't clutter directories
1212 such as /usr/local/bin.
1215 *) "make linux-shared" to build shared libraries.
1216 [Niels Poppe <niels@netbox.org>]
1218 *) New Configure option no-<cipher> (rsa, idea, rc5, ...).
1221 *) Add the PKCS#12 API documentation to openssl.txt. Preliminary support for
1222 extension adding in x509 utility.
1225 *) Remove NOPROTO sections and error code comments.
1228 *) Partial rewrite of the DEF file generator to now parse the ANSI
1232 *) New Configure options --prefix=DIR and --openssldir=DIR.
1235 *) Complete rewrite of the error code script(s). It is all now handled
1236 by one script at the top level which handles error code gathering,
1237 header rewriting and C source file generation. It should be much better
1238 than the old method: it now uses a modified version of Ulf's parser to
1239 read the ANSI prototypes in all header files (thus the old K&R definitions
1240 aren't needed for error creation any more) and do a better job of
1241 translating function codes into names. The old 'ASN1 error code imbedded
1242 in a comment' is no longer necessary and it doesn't use .err files which
1243 have now been deleted. Also the error code call doesn't have to appear all
1244 on one line (which resulted in some large lines...).
1247 *) Change #include filenames from <foo.h> to <openssl/foo.h>.
1250 *) Change behaviour of ssl2_read when facing length-0 packets: Don't return
1251 0 (which usually indicates a closed connection), but continue reading.
1254 *) Fix some race conditions.
1257 *) Add support for CRL distribution points extension. Add Certificate
1258 Policies and CRL distribution points documentation.
1261 *) Move the autogenerated header file parts to crypto/opensslconf.h.
1264 *) Fix new 56-bit DES export ciphersuites: they were using 7 bytes instead of
1265 8 of keying material. Merlin has also confirmed interop with this fix
1266 between OpenSSL and Baltimore C/SSL 2.0 and J/SSL 2.0.
1267 [Merlin Hughes <merlin@baltimore.ie>]
1269 *) Fix lots of warnings.
1270 [Richard Levitte <levitte@stacken.kth.se>]
1272 *) In add_cert_dir() in crypto/x509/by_dir.c, break out of the loop if
1273 the directory spec didn't end with a LIST_SEPARATOR_CHAR.
1274 [Richard Levitte <levitte@stacken.kth.se>]
1276 *) Fix problems with sizeof(long) == 8.
1277 [Andy Polyakov <appro@fy.chalmers.se>]
1279 *) Change functions to ANSI C.
1282 *) Fix typos in error codes.
1283 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>, Ulf Möller]
1285 *) Remove defunct assembler files from Configure.
1288 *) SPARC v8 assembler BIGNUM implementation.
1289 [Andy Polyakov <appro@fy.chalmers.se>]
1291 *) Support for Certificate Policies extension: both print and set.
1292 Various additions to support the r2i method this uses.
1295 *) A lot of constification, and fix a bug in X509_NAME_oneline() that could
1296 return a const string when you are expecting an allocated buffer.
1299 *) Add support for ASN1 types UTF8String and VISIBLESTRING, also the CHOICE
1300 types DirectoryString and DisplayText.
1303 *) Add code to allow r2i extensions to access the configuration database,
1304 add an LHASH database driver and add several ctx helper functions.
1307 *) Fix an evil bug in bn_expand2() which caused various BN functions to
1308 fail when they extended the size of a BIGNUM.
1311 *) Various utility functions to handle SXNet extension. Modify mkdef.pl to
1312 support typesafe stack.
1315 *) Fix typo in SSL_[gs]et_options().
1316 [Nils Frostberg <nils@medcom.se>]
1318 *) Delete various functions and files that belonged to the (now obsolete)
1319 old X509V3 handling code.
1322 *) New Configure option "rsaref".
1325 *) Don't auto-generate pem.h.
1328 *) Introduce type-safe ASN.1 SETs.
1331 *) Convert various additional casted stacks to type-safe STACK_OF() variants.
1332 [Ben Laurie, Ralf S. Engelschall, Steve Henson]
1334 *) Introduce type-safe STACKs. This will almost certainly break lots of code
1335 that links with OpenSSL (well at least cause lots of warnings), but fear
1336 not: the conversion is trivial, and it eliminates loads of evil casts. A
1337 few STACKed things have been converted already. Feel free to convert more.
1338 In the fullness of time, I'll do away with the STACK type altogether.
1341 *) Add `openssl ca -revoke <certfile>' facility which revokes a certificate
1342 specified in <certfile> by updating the entry in the index.txt file.
1343 This way one no longer has to edit the index.txt file manually for
1344 revoking a certificate. The -revoke option does the gory details now.
1345 [Massimiliano Pala <madwolf@openca.org>, Ralf S. Engelschall]
1347 *) Fix `openssl crl -noout -text' combination where `-noout' killed the
1348 `-text' option at all and this way the `-noout -text' combination was
1349 inconsistent in `openssl crl' with the friends in `openssl x509|rsa|dsa'.
1350 [Ralf S. Engelschall]
1352 *) Make sure a corresponding plain text error message exists for the
1353 X509_V_ERR_CERT_REVOKED/23 error number which can occur when a
1354 verify callback function determined that a certificate was revoked.
1355 [Ralf S. Engelschall]
1357 *) Bugfix: In test/testenc, don't test "openssl <cipher>" for
1358 ciphers that were excluded, e.g. by -DNO_IDEA. Also, test
1359 all available cipers including rc5, which was forgotten until now.
1360 In order to let the testing shell script know which algorithms
1361 are available, a new (up to now undocumented) command
1362 "openssl list-cipher-commands" is used.
1365 *) Bugfix: s_client occasionally would sleep in select() when
1366 it should have checked SSL_pending() first.
1369 *) New functions DSA_do_sign and DSA_do_verify to provide access to
1370 the raw DSA values prior to ASN.1 encoding.
1373 *) Tweaks to Configure
1374 [Niels Poppe <niels@netbox.org>]
1376 *) Add support for PKCS#5 v2.0 ASN1 PBES2 structures. No other support,
1380 *) New variables $(RANLIB) and $(PERL) in the Makefiles.
1383 *) New config option to avoid instructions that are illegal on the 80386.
1384 The default code is faster, but requires at least a 486.
1387 *) Got rid of old SSL2_CLIENT_VERSION (inconsistently used) and
1388 SSL2_SERVER_VERSION (not used at all) macros, which are now the
1389 same as SSL2_VERSION anyway.
1392 *) New "-showcerts" option for s_client.
1395 *) Still more PKCS#12 integration. Add pkcs12 application to openssl
1396 application. Various cleanups and fixes.
1399 *) More PKCS#12 integration. Add new pkcs12 directory with Makefile.ssl and
1400 modify error routines to work internally. Add error codes and PBE init
1401 to library startup routines.
1404 *) Further PKCS#12 integration. Added password based encryption, PKCS#8 and
1405 packing functions to asn1 and evp. Changed function names and error
1406 codes along the way.
1409 *) PKCS12 integration: and so it begins... First of several patches to
1410 slowly integrate PKCS#12 functionality into OpenSSL. Add PKCS#12
1411 objects to objects.h
1414 *) Add a new 'indent' option to some X509V3 extension code. Initial ASN1
1415 and display support for Thawte strong extranet extension.
1418 *) Add LinuxPPC support.
1419 [Jeff Dubrule <igor@pobox.org>]
1421 *) Get rid of redundant BN file bn_mulw.c, and rename bn_div64 to
1422 bn_div_words in alpha.s.
1423 [Hannes Reinecke <H.Reinecke@hw.ac.uk> and Ben Laurie]
1425 *) Make sure the RSA OAEP test is skipped under -DRSAref because
1426 OAEP isn't supported when OpenSSL is built with RSAref.
1427 [Ulf Moeller <ulf@fitug.de>]
1429 *) Move definitions of IS_SET/IS_SEQUENCE inside crypto/asn1/asn1.h
1430 so they no longer are missing under -DNOPROTO.
1431 [Soren S. Jorvang <soren@t.dk>]
1434 Changes between 0.9.1c and 0.9.2b [22 Mar 1999]
1436 *) Make SSL_get_peer_cert_chain() work in servers. Unfortunately, it still
1437 doesn't work when the session is reused. Coming soon!
1440 *) Fix a security hole, that allows sessions to be reused in the wrong
1441 context thus bypassing client cert protection! All software that uses
1442 client certs and session caches in multiple contexts NEEDS PATCHING to
1443 allow session reuse! A fuller solution is in the works.
1444 [Ben Laurie, problem pointed out by Holger Reif, Bodo Moeller (and ???)]
1446 *) Some more source tree cleanups (removed obsolete files
1447 crypto/bf/asm/bf586.pl, test/test.txt and crypto/sha/asm/f.s; changed
1448 permission on "config" script to be executable) and a fix for the INSTALL
1450 [Ulf Moeller <ulf@fitug.de>]
1452 *) Remove some legacy and erroneous uses of malloc, free instead of
1454 [Lennart Bang <lob@netstream.se>, with minor changes by Steve]
1456 *) Make rsa_oaep_test return non-zero on error.
1457 [Ulf Moeller <ulf@fitug.de>]
1459 *) Add support for native Solaris shared libraries. Configure
1460 solaris-sparc-sc4-pic, make, then run shlib/solaris-sc4.sh. It'd be nice
1461 if someone would make that last step automatic.
1462 [Matthias Loepfe <Matthias.Loepfe@AdNovum.CH>]
1464 *) ctx_size was not built with the right compiler during "make links". Fixed.
1467 *) Change the meaning of 'ALL' in the cipher list. It now means "everything
1468 except NULL ciphers". This means the default cipher list will no longer
1469 enable NULL ciphers. They need to be specifically enabled e.g. with
1470 the string "DEFAULT:eNULL".
1473 *) Fix to RSA private encryption routines: if p < q then it would
1474 occasionally produce an invalid result. This will only happen with
1475 externally generated keys because OpenSSL (and SSLeay) ensure p > q.
1478 *) Be less restrictive and allow also `perl util/perlpath.pl
1479 /path/to/bin/perl' in addition to `perl util/perlpath.pl /path/to/bin',
1480 because this way one can also use an interpreter named `perl5' (which is
1481 usually the name of Perl 5.xxx on platforms where an Perl 4.x is still
1482 installed as `perl').
1483 [Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
1485 *) Let util/clean-depend.pl work also with older Perl 5.00x versions.
1486 [Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
1488 *) Fix Makefile.org so CC,CFLAG etc are passed to 'make links' add
1489 advapi32.lib to Win32 build and change the pem test comparision
1490 to fc.exe (thanks to Ulrich Kroener <kroneru@yahoo.com> for the
1491 suggestion). Fix misplaced ASNI prototypes and declarations in evp.h
1492 and crypto/des/ede_cbcm_enc.c.
1495 *) DES quad checksum was broken on big-endian architectures. Fixed.
1498 *) Comment out two functions in bio.h that aren't implemented. Fix up the
1499 Win32 test batch file so it (might) work again. The Win32 test batch file
1500 is horrible: I feel ill....
1503 *) Move various #ifdefs around so NO_SYSLOG, NO_DIRENT etc are now selected
1504 in e_os.h. Audit of header files to check ANSI and non ANSI
1505 sections: 10 functions were absent from non ANSI section and not exported
1506 from Windows DLLs. Fixed up libeay.num for new functions.
1509 *) Make `openssl version' output lines consistent.
1510 [Ralf S. Engelschall]
1512 *) Fix Win32 symbol export lists for BIO functions: Added
1513 BIO_get_ex_new_index, BIO_get_ex_num, BIO_get_ex_data and BIO_set_ex_data
1514 to ms/libeay{16,32}.def.
1515 [Ralf S. Engelschall]
1517 *) Second round of fixing the OpenSSL perl/ stuff. It now at least compiled
1518 fine under Unix and passes some trivial tests I've now added. But the
1519 whole stuff is horribly incomplete, so a README.1ST with a disclaimer was
1520 added to make sure no one expects that this stuff really works in the
1521 OpenSSL 0.9.2 release. Additionally I've started to clean the XS sources
1522 up and fixed a few little bugs and inconsistencies in OpenSSL.{pm,xs} and
1524 [Ralf S. Engelschall]
1526 *) Fix the generation of two part addresses in perl.
1527 [Kenji Miyake <kenji@miyake.org>, integrated by Ben Laurie]
1529 *) Add config entry for Linux on MIPS.
1530 [John Tobey <jtobey@channel1.com>]
1532 *) Make links whenever Configure is run, unless we are on Windoze.
1535 *) Permit extensions to be added to CRLs using crl_section in openssl.cnf.
1536 Currently only issuerAltName and AuthorityKeyIdentifier make any sense
1540 *) Add a useful kludge to allow package maintainers to specify compiler and
1541 other platforms details on the command line without having to patch the
1542 Configure script everytime: One now can use ``perl Configure
1543 <id>:<details>'', i.e. platform ids are allowed to have details appended
1544 to them (seperated by colons). This is treated as there would be a static
1545 pre-configured entry in Configure's %table under key <id> with value
1546 <details> and ``perl Configure <id>'' is called. So, when you want to
1547 perform a quick test-compile under FreeBSD 3.1 with pgcc and without
1548 assembler stuff you can use ``perl Configure "FreeBSD-elf:pgcc:-O6:::"''
1549 now, which overrides the FreeBSD-elf entry on-the-fly.
1550 [Ralf S. Engelschall]
1552 *) Disable new TLS1 ciphersuites by default: they aren't official yet.
1555 *) Allow DSO flags like -fpic, -fPIC, -KPIC etc. to be specified
1556 on the `perl Configure ...' command line. This way one can compile
1557 OpenSSL libraries with Position Independent Code (PIC) which is needed
1558 for linking it into DSOs.
1559 [Ralf S. Engelschall]
1561 *) Remarkably, export ciphers were totally broken and no-one had noticed!
1565 *) Cleaned up the LICENSE document: The official contact for any license
1566 questions now is the OpenSSL core team under openssl-core@openssl.org.
1567 And add a paragraph about the dual-license situation to make sure people
1568 recognize that _BOTH_ the OpenSSL license _AND_ the SSLeay license apply
1569 to the OpenSSL toolkit.
1570 [Ralf S. Engelschall]
1572 *) General source tree makefile cleanups: Made `making xxx in yyy...'
1573 display consistent in the source tree and replaced `/bin/rm' by `rm'.
1574 Additonally cleaned up the `make links' target: Remove unnecessary
1575 semicolons, subsequent redundant removes, inline point.sh into mklink.sh
1576 to speed processing and no longer clutter the display with confusing
1577 stuff. Instead only the actually done links are displayed.
1578 [Ralf S. Engelschall]
1580 *) Permit null encryption ciphersuites, used for authentication only. It used
1581 to be necessary to set the preprocessor define SSL_ALLOW_ENULL to do this.
1582 It is now necessary to set SSL_FORBID_ENULL to prevent the use of null
1586 *) Add a bunch of fixes to the PKCS#7 stuff. It used to sometimes reorder
1587 signed attributes when verifying signatures (this would break them),
1588 the detached data encoding was wrong and public keys obtained using
1589 X509_get_pubkey() weren't freed.
1592 *) Add text documentation for the BUFFER functions. Also added a work around
1593 to a Win95 console bug. This was triggered by the password read stuff: the
1594 last character typed gets carried over to the next fread(). If you were
1595 generating a new cert request using 'req' for example then the last
1596 character of the passphrase would be CR which would then enter the first
1600 *) Added the new `Includes OpenSSL Cryptography Software' button as
1601 doc/openssl_button.{gif,html} which is similar in style to the old SSLeay
1602 button and can be used by applications based on OpenSSL to show the
1603 relationship to the OpenSSL project.
1604 [Ralf S. Engelschall]
1606 *) Remove confusing variables in function signatures in files
1607 ssl/ssl_lib.c and ssl/ssl.h.
1608 [Lennart Bong <lob@kulthea.stacken.kth.se>]
1610 *) Don't install bss_file.c under PREFIX/include/
1611 [Lennart Bong <lob@kulthea.stacken.kth.se>]
1613 *) Get the Win32 compile working again. Modify mkdef.pl so it can handle
1614 functions that return function pointers and has support for NT specific
1615 stuff. Fix mk1mf.pl and VC-32.pl to support NT differences also. Various
1616 #ifdef WIN32 and WINNTs sprinkled about the place and some changes from
1617 unsigned to signed types: this was killing the Win32 compile.
1620 *) Add new certificate file to stack functions,
1621 SSL_add_dir_cert_subjects_to_stack() and
1622 SSL_add_file_cert_subjects_to_stack(). These largely supplant
1623 SSL_load_client_CA_file(), and can be used to add multiple certs easily
1624 to a stack (usually this is then handed to SSL_CTX_set_client_CA_list()).
1625 This means that Apache-SSL and similar packages don't have to mess around
1626 to add as many CAs as they want to the preferred list.
1629 *) Experiment with doxygen documentation. Currently only partially applied to
1631 See http://www.stack.nl/~dimitri/doxygen/index.html, and run doxygen with
1632 openssl.doxy as the configuration file.
1635 *) Get rid of remaining C++-style comments which strict C compilers hate.
1636 [Ralf S. Engelschall, pointed out by Carlos Amengual]
1638 *) Changed BN_RECURSION in bn_mont.c to BN_RECURSION_MONT so it is not
1639 compiled in by default: it has problems with large keys.
1642 *) Add a bunch of SSL_xxx() functions for configuring the temporary RSA and
1643 DH private keys and/or callback functions which directly correspond to
1644 their SSL_CTX_xxx() counterparts but work on a per-connection basis. This
1645 is needed for applications which have to configure certificates on a
1646 per-connection basis (e.g. Apache+mod_ssl) instead of a per-context basis
1648 For the RSA certificate situation is makes no difference, but
1649 for the DSA certificate situation this fixes the "no shared cipher"
1650 problem where the OpenSSL cipher selection procedure failed because the
1651 temporary keys were not overtaken from the context and the API provided
1652 no way to reconfigure them.
1653 The new functions now let applications reconfigure the stuff and they
1654 are in detail: SSL_need_tmp_RSA, SSL_set_tmp_rsa, SSL_set_tmp_dh,
1655 SSL_set_tmp_rsa_callback and SSL_set_tmp_dh_callback. Additionally a new
1656 non-public-API function ssl_cert_instantiate() is used as a helper
1657 function and also to reduce code redundancy inside ssl_rsa.c.
1658 [Ralf S. Engelschall]
1660 *) Move s_server -dcert and -dkey options out of the undocumented feature
1661 area because they are useful for the DSA situation and should be
1662 recognized by the users.
1663 [Ralf S. Engelschall]
1665 *) Fix the cipher decision scheme for export ciphers: the export bits are
1666 *not* within SSL_MKEY_MASK or SSL_AUTH_MASK, they are within
1667 SSL_EXP_MASK. So, the original variable has to be used instead of the
1668 already masked variable.
1669 [Richard Levitte <levitte@stacken.kth.se>]
1671 *) Fix 'port' variable from `int' to `unsigned int' in crypto/bio/b_sock.c
1672 [Richard Levitte <levitte@stacken.kth.se>]
1674 *) Change type of another md_len variable in pk7_doit.c:PKCS7_dataFinal()
1675 from `int' to `unsigned int' because it's a length and initialized by
1676 EVP_DigestFinal() which expects an `unsigned int *'.
1677 [Richard Levitte <levitte@stacken.kth.se>]
1679 *) Don't hard-code path to Perl interpreter on shebang line of Configure
1680 script. Instead use the usual Shell->Perl transition trick.
1681 [Ralf S. Engelschall]
1683 *) Make `openssl x509 -noout -modulus' functional also for DSA certificates
1684 (in addition to RSA certificates) to match the behaviour of `openssl dsa
1685 -noout -modulus' as it's already the case for `openssl rsa -noout
1686 -modulus'. For RSA the -modulus is the real "modulus" while for DSA
1687 currently the public key is printed (a decision which was already done by
1688 `openssl dsa -modulus' in the past) which serves a similar purpose.
1689 Additionally the NO_RSA no longer completely removes the whole -modulus
1690 option; it now only avoids using the RSA stuff. Same applies to NO_DSA
1692 [Ralf S. Engelschall]
1694 *) Add Arne Ansper's reliable BIO - this is an encrypted, block-digested
1695 BIO. See the source (crypto/evp/bio_ok.c) for more info.
1696 [Arne Ansper <arne@ats.cyber.ee>]
1698 *) Dump the old yucky req code that tried (and failed) to allow raw OIDs
1699 to be added. Now both 'req' and 'ca' can use new objects defined in the
1703 *) Add cool BIO that does syslog (or event log on NT).
1704 [Arne Ansper <arne@ats.cyber.ee>, integrated by Ben Laurie]
1706 *) Add support for new TLS ciphersuites, TLS_RSA_EXPORT56_WITH_RC4_56_MD5,
1707 TLS_RSA_EXPORT56_WITH_RC2_CBC_56_MD5 and
1708 TLS_RSA_EXPORT56_WITH_DES_CBC_SHA, as specified in "56-bit Export Cipher
1709 Suites For TLS", draft-ietf-tls-56-bit-ciphersuites-00.txt.
1712 *) Add preliminary config info for new extension code.
1715 *) Make RSA_NO_PADDING really use no padding.
1716 [Ulf Moeller <ulf@fitug.de>]
1718 *) Generate errors when private/public key check is done.
1721 *) Overhaul for 'crl' utility. New function X509_CRL_print. Partial support
1722 for some CRL extensions and new objects added.
1725 *) Really fix the ASN1 IMPLICIT bug this time... Partial support for private
1726 key usage extension and fuller support for authority key id.
1729 *) Add OAEP encryption for the OpenSSL crypto library. OAEP is the improved
1730 padding method for RSA, which is recommended for new applications in PKCS
1731 #1 v2.0 (RFC 2437, October 1998).
1732 OAEP (Optimal Asymmetric Encryption Padding) has better theoretical
1733 foundations than the ad-hoc padding used in PKCS #1 v1.5. It is secure
1734 against Bleichbacher's attack on RSA.
1735 [Ulf Moeller <ulf@fitug.de>, reformatted, corrected and integrated by
1738 *) Updates to the new SSL compression code
1739 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
1741 *) Fix so that the version number in the master secret, when passed
1742 via RSA, checks that if TLS was proposed, but we roll back to SSLv3
1743 (because the server will not accept higher), that the version number
1744 is 0x03,0x01, not 0x03,0x00
1745 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
1747 *) Run extensive memory leak checks on SSL apps. Fixed *lots* of memory
1748 leaks in ssl/ relating to new X509_get_pubkey() behaviour. Also fixes
1749 in apps/ and an unrelated leak in crypto/dsa/dsa_vrf.c
1752 *) Support for RAW extensions where an arbitrary extension can be
1753 created by including its DER encoding. See apps/openssl.cnf for
1757 *) Make sure latest Perl versions don't interpret some generated C array
1758 code as Perl array code in the crypto/err/err_genc.pl script.
1759 [Lars Weber <3weber@informatik.uni-hamburg.de>]
1761 *) Modify ms/do_ms.bat to not generate assembly language makefiles since
1762 not many people have the assembler. Various Win32 compilation fixes and
1763 update to the INSTALL.W32 file with (hopefully) more accurate Win32
1767 *) Modify configure script 'Configure' to automatically create crypto/date.h
1768 file under Win32 and also build pem.h from pem.org. New script
1769 util/mkfiles.pl to create the MINFO file on environments that can't do a
1770 'make files': perl util/mkfiles.pl >MINFO should work.
1773 *) Major rework of DES function declarations, in the pursuit of correctness
1774 and purity. As a result, many evil casts evaporated, and some weirdness,
1775 too. You may find this causes warnings in your code. Zapping your evil
1776 casts will probably fix them. Mostly.
1779 *) Fix for a typo in asn1.h. Bug fix to object creation script
1780 obj_dat.pl. It considered a zero in an object definition to mean
1781 "end of object": none of the objects in objects.h have any zeros
1782 so it wasn't spotted.
1783 [Steve Henson, reported by Erwann ABALEA <eabalea@certplus.com>]
1785 *) Add support for Triple DES Cipher Block Chaining with Output Feedback
1786 Masking (CBCM). In the absence of test vectors, the best I have been able
1787 to do is check that the decrypt undoes the encrypt, so far. Send me test
1788 vectors if you have them.
1791 *) Correct calculation of key length for export ciphers (too much space was
1792 allocated for null ciphers). This has not been tested!
1795 *) Modifications to the mkdef.pl for Win32 DEF file creation. The usage
1796 message is now correct (it understands "crypto" and "ssl" on its
1797 command line). There is also now an "update" option. This will update
1798 the util/ssleay.num and util/libeay.num files with any new functions.
1800 perl util/mkdef.pl crypto ssl update
1801 it will update them.
1804 *) Overhauled the Perl interface (perl/*):
1805 - ported BN stuff to OpenSSL's different BN library
1806 - made the perl/ source tree CVS-aware
1807 - renamed the package from SSLeay to OpenSSL (the files still contain
1808 their history because I've copied them in the repository)
1809 - removed obsolete files (the test scripts will be replaced
1810 by better Test::Harness variants in the future)
1811 [Ralf S. Engelschall]
1813 *) First cut for a very conservative source tree cleanup:
1814 1. merge various obsolete readme texts into doc/ssleay.txt
1815 where we collect the old documents and readme texts.
1816 2. remove the first part of files where I'm already sure that we no
1817 longer need them because of three reasons: either they are just temporary
1818 files which were left by Eric or they are preserved original files where
1819 I've verified that the diff is also available in the CVS via "cvs diff
1820 -rSSLeay_0_8_1b" or they were renamed (as it was definitely the case for
1821 the crypto/md/ stuff).
1822 [Ralf S. Engelschall]
1824 *) More extension code. Incomplete support for subject and issuer alt
1825 name, issuer and authority key id. Change the i2v function parameters
1826 and add an extra 'crl' parameter in the X509V3_CTX structure: guess
1827 what that's for :-) Fix to ASN1 macro which messed up
1828 IMPLICIT tag and add f_enum.c which adds a2i, i2a for ENUMERATED.
1831 *) Preliminary support for ENUMERATED type. This is largely copied from the
1835 *) Add new function, EVP_MD_CTX_copy() to replace frequent use of memcpy.
1836 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
1838 *) Make sure `make rehash' target really finds the `openssl' program.
1839 [Ralf S. Engelschall, Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
1841 *) Squeeze another 7% of speed out of MD5 assembler, at least on a P2. I'd
1842 like to hear about it if this slows down other processors.
1845 *) Add CygWin32 platform information to Configure script.
1846 [Alan Batie <batie@aahz.jf.intel.com>]
1848 *) Fixed ms/32all.bat script: `no_asm' -> `no-asm'
1849 [Rainer W. Gerling <gerling@mpg-gv.mpg.de>]
1851 *) New program nseq to manipulate netscape certificate sequences
1854 *) Modify crl2pkcs7 so it supports multiple -certfile arguments. Fix a
1858 *) Fixes to BN code. Previously the default was to define BN_RECURSION
1859 but the BN code had some problems that would cause failures when
1860 doing certificate verification and some other functions.
1861 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
1863 *) Add ASN1 and PEM code to support netscape certificate sequences.
1866 *) Add ASN1 and PEM code to support netscape certificate sequences.
1869 *) Add several PKIX and private extended key usage OIDs.
1872 *) Modify the 'ca' program to handle the new extension code. Modify
1873 openssl.cnf for new extension format, add comments.
1876 *) More X509 V3 changes. Fix typo in v3_bitstr.c. Add support to 'req'
1877 and add a sample to openssl.cnf so req -x509 now adds appropriate
1881 *) Continued X509 V3 changes. Add to other makefiles, integrate with the
1882 error code, add initial support to X509_print() and x509 application.
1885 *) Takes a deep breath and start addding X509 V3 extension support code. Add
1886 files in crypto/x509v3. Move original stuff to crypto/x509v3/old. All this
1887 stuff is currently isolated and isn't even compiled yet.
1890 *) Continuing patches for GeneralizedTime. Fix up certificate and CRL
1891 ASN1 to use ASN1_TIME and modify print routines to use ASN1_TIME_print.
1892 Removed the versions check from X509 routines when loading extensions:
1893 this allows certain broken certificates that don't set the version
1894 properly to be processed.
1897 *) Deal with irritating shit to do with dependencies, in YAAHW (Yet Another
1898 Ad Hoc Way) - Makefile.ssls now all contain local dependencies, which
1899 can still be regenerated with "make depend".
1902 *) Spelling mistake in C version of CAST-128.
1903 [Ben Laurie, reported by Jeremy Hylton <jeremy@cnri.reston.va.us>]
1905 *) Changes to the error generation code. The perl script err-code.pl
1906 now reads in the old error codes and retains the old numbers, only
1907 adding new ones if necessary. It also only changes the .err files if new
1908 codes are added. The makefiles have been modified to only insert errors
1909 when needed (to avoid needlessly modifying header files). This is done
1910 by only inserting errors if the .err file is newer than the auto generated
1911 C file. To rebuild all the error codes from scratch (the old behaviour)
1912 either modify crypto/Makefile.ssl to pass the -regen flag to err_code.pl
1913 or delete all the .err files.
1916 *) CAST-128 was incorrectly implemented for short keys. The C version has
1917 been fixed, but is untested. The assembler versions are also fixed, but
1918 new assembler HAS NOT BEEN GENERATED FOR WIN32 - the Makefile needs fixing
1919 to regenerate it if needed.
1920 [Ben Laurie, reported (with fix for C version) by Jun-ichiro itojun
1921 Hagino <itojun@kame.net>]
1923 *) File was opened incorrectly in randfile.c.
1924 [Ulf Möller <ulf@fitug.de>]
1926 *) Beginning of support for GeneralizedTime. d2i, i2d, check and print
1927 functions. Also ASN1_TIME suite which is a CHOICE of UTCTime or
1928 GeneralizedTime. ASN1_TIME is the proper type used in certificates et
1929 al: it's just almost always a UTCTime. Note this patch adds new error
1930 codes so do a "make errors" if there are problems.
1933 *) Correct Linux 1 recognition in config.
1934 [Ulf Möller <ulf@fitug.de>]
1936 *) Remove pointless MD5 hash when using DSA keys in ca.
1937 [Anonymous <nobody@replay.com>]
1939 *) Generate an error if given an empty string as a cert directory. Also
1940 generate an error if handed NULL (previously returned 0 to indicate an
1941 error, but didn't set one).
1942 [Ben Laurie, reported by Anonymous <nobody@replay.com>]
1944 *) Add prototypes to SSL methods. Make SSL_write's buffer const, at last.
1947 *) Fix the dummy function BN_ref_mod_exp() in rsaref.c to have the correct
1948 parameters. This was causing a warning which killed off the Win32 compile.
1951 *) Remove C++ style comments from crypto/bn/bn_local.h.
1952 [Neil Costigan <neil.costigan@celocom.com>]
1954 *) The function OBJ_txt2nid was broken. It was supposed to return a nid
1955 based on a text string, looking up short and long names and finally
1956 "dot" format. The "dot" format stuff didn't work. Added new function
1957 OBJ_txt2obj to do the same but return an ASN1_OBJECT and rewrote
1958 OBJ_txt2nid to use it. OBJ_txt2obj can also return objects even if the
1959 OID is not part of the table.
1962 *) Add prototypes to X509 lookup/verify methods, fixing a bug in
1963 X509_LOOKUP_by_alias().
1966 *) Sort openssl functions by name.
1969 *) Get the gendsa program working (hopefully) and add it to app list. Remove
1970 encryption from sample DSA keys (in case anyone is interested the password
1974 *) Make _all_ *_free functions accept a NULL pointer.
1975 [Frans Heymans <fheymans@isaserver.be>]
1977 *) If a DH key is generated in s3_srvr.c, don't blow it by trying to use
1979 [Anonymous <nobody@replay.com>]
1981 *) s_server should send the CAfile as acceptable CAs, not its own cert.
1982 [Bodo Moeller <3moeller@informatik.uni-hamburg.de>]
1984 *) Don't blow it for numeric -newkey arguments to apps/req.
1985 [Bodo Moeller <3moeller@informatik.uni-hamburg.de>]
1987 *) Temp key "for export" tests were wrong in s3_srvr.c.
1988 [Anonymous <nobody@replay.com>]
1990 *) Add prototype for temp key callback functions
1991 SSL_CTX_set_tmp_{rsa,dh}_callback().
1994 *) Make DH_free() tolerate being passed a NULL pointer (like RSA_free() and
1995 DSA_free()). Make X509_PUBKEY_set() check for errors in d2i_PublicKey().
1998 *) X509_name_add_entry() freed the wrong thing after an error.
1999 [Arne Ansper <arne@ats.cyber.ee>]
2001 *) rsa_eay.c would attempt to free a NULL context.
2002 [Arne Ansper <arne@ats.cyber.ee>]
2004 *) BIO_s_socket() had a broken should_retry() on Windoze.
2005 [Arne Ansper <arne@ats.cyber.ee>]
2007 *) BIO_f_buffer() didn't pass on BIO_CTRL_FLUSH.
2008 [Arne Ansper <arne@ats.cyber.ee>]
2010 *) Make sure the already existing X509_STORE->depth variable is initialized
2011 in X509_STORE_new(), but document the fact that this variable is still
2012 unused in the certificate verification process.
2013 [Ralf S. Engelschall]
2015 *) Fix the various library and apps files to free up pkeys obtained from
2016 X509_PUBKEY_get() et al. Also allow x509.c to handle netscape extensions.
2019 *) Fix reference counting in X509_PUBKEY_get(). This makes
2020 demos/maurice/example2.c work, amongst others, probably.
2021 [Steve Henson and Ben Laurie]
2023 *) First cut of a cleanup for apps/. First the `ssleay' program is now named
2024 `openssl' and second, the shortcut symlinks for the `openssl <command>'
2025 are no longer created. This way we have a single and consistent command
2026 line interface `openssl <command>', similar to `cvs <command>'.
2027 [Ralf S. Engelschall, Paul Sutton and Ben Laurie]
2029 *) ca.c: move test for DSA keys inside #ifndef NO_DSA. Make pubkey
2030 BIT STRING wrapper always have zero unused bits.
2033 *) Add CA.pl, perl version of CA.sh, add extended key usage OID.
2036 *) Make the top-level INSTALL documentation easier to understand.
2039 *) Makefiles updated to exit if an error occurs in a sub-directory
2040 make (including if user presses ^C) [Paul Sutton]
2042 *) Make Montgomery context stuff explicit in RSA data structure.
2045 *) Fix build order of pem and err to allow for generated pem.h.
2048 *) Fix renumbering bug in X509_NAME_delete_entry().
2051 *) Enhanced the err-ins.pl script so it makes the error library number
2052 global and can add a library name. This is needed for external ASN1 and
2053 other error libraries.
2056 *) Fixed sk_insert which never worked properly.
2059 *) Fix ASN1 macros so they can handle indefinite length construted
2060 EXPLICIT tags. Some non standard certificates use these: they can now
2064 *) Merged the various old/obsolete SSLeay documentation files (doc/xxx.doc)
2065 into a single doc/ssleay.txt bundle. This way the information is still
2066 preserved but no longer messes up this directory. Now it's new room for
2067 the new set of documenation files.
2068 [Ralf S. Engelschall]
2070 *) SETs were incorrectly DER encoded. This was a major pain, because they
2071 shared code with SEQUENCEs, which aren't coded the same. This means that
2072 almost everything to do with SETs or SEQUENCEs has either changed name or
2073 number of arguments.
2074 [Ben Laurie, based on a partial fix by GP Jayan <gp@nsj.co.jp>]
2076 *) Fix test data to work with the above.
2079 *) Fix the RSA header declarations that hid a bug I fixed in 0.9.0b but
2080 was already fixed by Eric for 0.9.1 it seems.
2081 [Ben Laurie - pointed out by Ulf Möller <ulf@fitug.de>]
2083 *) Autodetect FreeBSD3.
2086 *) Fix various bugs in Configure. This affects the following platforms:
2090 unixware-2.0-pentium
2094 *) Eliminate generated files from CVS. Reorder tests to regenerate files
2095 before they are needed.
2098 *) Generate Makefile.ssl from Makefile.org (to keep CVS happy).
2102 Changes between 0.9.1b and 0.9.1c [23-Dec-1998]
2104 *) Added OPENSSL_VERSION_NUMBER to crypto/crypto.h and
2105 changed SSLeay to OpenSSL in version strings.
2106 [Ralf S. Engelschall]
2108 *) Some fixups to the top-level documents.
2111 *) Fixed the nasty bug where rsaref.h was not found under compile-time
2112 because the symlink to include/ was missing.
2113 [Ralf S. Engelschall]
2115 *) Incorporated the popular no-RSA/DSA-only patches
2116 which allow to compile a RSA-free SSLeay.
2117 [Andrew Cooke / Interrader Ldt., Ralf S. Engelschall]
2119 *) Fixed nasty rehash problem under `make -f Makefile.ssl links'
2120 when "ssleay" is still not found.
2121 [Ralf S. Engelschall]
2123 *) Added more platforms to Configure: Cray T3E, HPUX 11,
2124 [Ralf S. Engelschall, Beckmann <beckman@acl.lanl.gov>]
2126 *) Updated the README file.
2127 [Ralf S. Engelschall]
2129 *) Added various .cvsignore files in the CVS repository subdirs
2130 to make a "cvs update" really silent.
2131 [Ralf S. Engelschall]
2133 *) Recompiled the error-definition header files and added
2134 missing symbols to the Win32 linker tables.
2135 [Ralf S. Engelschall]
2137 *) Cleaned up the top-level documents;
2138 o new files: CHANGES and LICENSE
2139 o merged VERSION, HISTORY* and README* files a CHANGES.SSLeay
2140 o merged COPYRIGHT into LICENSE
2141 o removed obsolete TODO file
2142 o renamed MICROSOFT to INSTALL.W32
2143 [Ralf S. Engelschall]
2145 *) Removed dummy files from the 0.9.1b source tree:
2146 crypto/asn1/x crypto/bio/cd crypto/bio/fg crypto/bio/grep crypto/bio/vi
2147 crypto/bn/asm/......add.c crypto/bn/asm/a.out crypto/dsa/f crypto/md5/f
2148 crypto/pem/gmon.out crypto/perlasm/f crypto/pkcs7/build crypto/rsa/f
2149 crypto/sha/asm/f crypto/threads/f ms/zzz ssl/f ssl/f.mak test/f
2150 util/f.mak util/pl/f util/pl/f.mak crypto/bf/bf_locl.old apps/f
2151 [Ralf S. Engelschall]
2153 *) Added various platform portability fixes.
2156 *) The Genesis of the OpenSSL rpject:
2157 We start with the latest (unreleased) SSLeay version 0.9.1b which Eric A.
2158 Young and Tim J. Hudson created while they were working for C2Net until
2160 [The OpenSSL Project]
2163 Changes between 0.9.0b and 0.9.1b [not released]
2165 *) Updated a few CA certificates under certs/
2168 *) Changed some BIGNUM api stuff.
2171 *) Various platform ports: OpenBSD, Ultrix, IRIX 64bit, NetBSD,
2172 DGUX x86, Linux Alpha, etc.
2175 *) New COMP library [crypto/comp/] for SSL Record Layer Compression:
2176 RLE (dummy implemented) and ZLIB (really implemented when ZLIB is
2180 *) Add -strparse option to asn1pars program which parses nested
2182 [Dr Stephen Henson <shenson@bigfoot.com>]
2184 *) Added "oid_file" to ssleay.cnf for "ca" and "req" programs.
2187 *) DSA fix for "ca" program.
2190 *) Added "-genkey" option to "dsaparam" program.
2193 *) Added RIPE MD160 (rmd160) message digest.
2196 *) Added -a (all) option to "ssleay version" command.
2199 *) Added PLATFORM define which is the id given to Configure.
2202 *) Added MemCheck_XXXX functions to crypto/mem.c for memory checking.
2205 *) Extended the ASN.1 parser routines.
2208 *) Extended BIO routines to support REUSEADDR, seek, tell, etc.
2211 *) Added a BN_CTX to the BN library.
2214 *) Fixed the weak key values in DES library
2217 *) Changed API in EVP library for cipher aliases.
2220 *) Added support for RC2/64bit cipher.
2223 *) Converted the lhash library to the crypto/mem.c functions.
2226 *) Added more recognized ASN.1 object ids.
2229 *) Added more RSA padding checks for SSL/TLS.
2232 *) Added BIO proxy/filter functionality.
2235 *) Added extra_certs to SSL_CTX which can be used
2236 send extra CA certificates to the client in the CA cert chain sending
2237 process. It can be configured with SSL_CTX_add_extra_chain_cert().
2240 *) Now Fortezza is denied in the authentication phase because
2241 this is key exchange mechanism is not supported by SSLeay at all.
2244 *) Additional PKCS1 checks.
2247 *) Support the string "TLSv1" for all TLS v1 ciphers.
2250 *) Added function SSL_get_ex_data_X509_STORE_CTX_idx() which gives the
2251 ex_data index of the SSL context in the X509_STORE_CTX ex_data.
2254 *) Fixed a few memory leaks.
2257 *) Fixed various code and comment typos.
2260 *) A minor bug in ssl/s3_clnt.c where there would always be 4 0
2261 bytes sent in the client random.
2262 [Edward Bishop <ebishop@spyglass.com>]