5 Changes between 0.9.4 and 0.9.5 [xx XXX 1999]
7 *) Simplify the trust setting structure and code. Now we just have
8 two sequences of OIDs for trusted and rejected settings. These will
9 typically have values the same as the extended key usage extension
10 and any application specific purposes.
12 The trust checking code now has a default behaviour: it will just
13 check for an object with the same NID as the passed id. Functions can
14 be provided to override either the default behaviour or the behaviour
15 for a given id. SSL client, server and email already have functions
16 in place for compatability: they check the NID and also return "trusted"
17 if the certificate is self signed.
20 *) Add d2i,i2d bio/fp functions for PrivateKey: these convert the
21 traditional format into an EVP_PKEY structure.
24 *) Add a password callback function PEM_cb() which either prompts for
25 a password if usr_data is NULL or otherwise assumes it is a null
26 terminated password. Allow passwords to be passed on command line
27 environment or config files in a few more utilities.
30 *) Add a bunch of DER and PEM functions to handle PKCS#8 format private
31 keys. Add some short names for PKCS#8 PBE algorithms and allow them
32 to be specified on the command line for the pkcs8 and pkcs12 utilities.
36 *) Support for ASN1 "NULL" type. This could be handled before by using
37 ASN1_TYPE but there wasn't any function that would try to read a NULL
38 and produce an error if it couldn't. For compatibility we also have
39 ASN1_NULL_new() and ASN1_NULL_free() functions but these are faked and
40 don't allocate anything because they don't need to.
43 *) Initial support for MacOS is now provided. Examine INSTALL.MacOS
45 [Andy Polyakov, Roy Woods <roy@centicsystems.ca>]
47 *) Rebuild of the memory allocation routines used by OpenSSL code and
48 possibly others as well. The purpose is to make an interface that
49 provide hooks so anyone can build a separate set of allocation and
50 deallocation routines to be used by OpenSSL, for example if memory
51 pool implementations, or something else. The same is provided for
52 memory debugging code. OpenSSL already comes with code that finds
53 memory leaks, but this gives people a chance to debug other memory
56 With these changes, a new set of functions and macros have appeared:
58 CRYPTO_set_mem_debug_functions() [F]
59 CRYPTO_get_mem_debug_functions() [F]
60 CRYPTO_dbg_set_options() [F]
61 CRYPTO_dbg_get_options() [F]
62 CRYPTO_melloc_debug_init() [M]
64 The memory debug functions are NULL by default, unless the library
65 is compiled with CRYPTO_MDEBUG or friends is defined. If someone
66 wants to debug memory anyway, CRYPTO_malloc_debug_init() or
67 CRYPTO_set_mem_debug_functions() must be used.
69 Also, things like CRYPTO_set_mem_functions will always give the
70 expected result (the new set of functions is used for allocation
71 and deallocation) at all times, regardless of platform and compiler
74 To finish it up, some functions that were never use in any other
75 way than through macros have a new API and new semantic:
81 All macros of value have retained their old syntax.
84 *) Some S/MIME fixes. The OID for SMIMECapabilities was wrong, the
85 ordering of SMIMECapabilities wasn't in "strength order" and there
86 was a missing NULL in the AlgorithmIdentifier for the SHA1 signature
90 *) Some ASN1 types with illegal zero length encoding (INTEGER,
91 ENUMERATED and OBJECT IDENTIFIER) choked the ASN1 routines.
92 [Frans Heymans <fheymans@isaserver.be>, modified by Steve Henson]
94 *) Merge in my S/MIME library for OpenSSL. This provides a simple
95 S/MIME API on top of the PKCS#7 code, a MIME parser (with enough
96 functionality to handle multipart/signed properly) and a utility
97 called 'smime' to call all this stuff. This is based on code I
98 originally wrote for Celo who have kindly allowed it to be
102 *) Add variants des_set_key_checked and des_set_key_unchecked of
103 des_set_key (aka des_key_sched). Global variable des_check_key
104 decides which of these is called by des_set_key; this way
105 des_check_key behaves as it always did, but applications and
106 the library itself, which was buggy for des_check_key == 1,
107 have a cleaner way to pick the version they need.
110 *) New function PKCS12_newpass() which changes the password of a
114 *) Modify X509_TRUST and X509_PURPOSE so it also uses a static and
115 dynamic mix. In both cases the ids can be used as an index into the
116 table. Also modified the X509_TRUST_add() and X509_PURPOSE_add()
117 functions so they accept a list of the field values and the
118 application doesn't need to directly manipulate the X509_TRUST
122 *) Modify the ASN1_STRING_TABLE stuff so it also uses bsearch and doesn't
126 *) Modify the way the V3 extension code looks up extensions. This now
127 works in a similar way to the object code: we have some "standard"
128 extensions in a static table which is searched with OBJ_bsearch()
129 and the application can add dynamic ones if needed. The file
130 crypto/x509v3/ext_dat.h now has the info: this file needs to be
131 updated whenever a new extension is added to the core code and kept
132 in ext_nid order. There is a simple program 'tabtest.c' which checks
133 this. New extensions are not added too often so this file can readily
134 be maintained manually.
136 There are two big advantages in doing things this way. The extensions
137 can be looked up immediately and no longer need to be "added" using
138 X509V3_add_standard_extensions(): this function now does nothing.
139 [Side note: I get *lots* of email saying the extension code doesn't
140 work because people forget to call this function]
141 Also no dynamic allocation is done unless new extensions are added:
142 so if we don't add custom extensions there is no need to call
143 X509V3_EXT_cleanup().
146 *) Modify enc utility's salting as follows: make salting the default. Add a
147 magic header, so unsalted files fail gracefully instead of just decrypting
148 to garbage. This is because not salting is a big security hole, so people
149 should be discouraged from doing it.
152 *) Fixes and enhancements to the 'x509' utility. It allowed a message
153 digest to be passed on the command line but it only used this
154 parameter when signing a certificate. Modified so all relevant
155 operations are affected by the digest parameter including the
156 -fingerprint and -x509toreq options. Also -x509toreq choked if a
157 DSA key was used because it didn't fix the digest.
160 *) Initial certificate chain verify code. Currently tests the untrusted
161 certificates for consistency with the verify purpose (which is set
162 when the X509_STORE_CTX structure is set up) and checks the pathlength.
164 There is a NO_CHAIN_VERIFY compilation option to keep the old behaviour:
165 this is because it will reject chains with invalid extensions whereas
166 every previous version of OpenSSL and SSLeay made no checks at all.
168 Trust code: checks the root CA for the relevant trust settings. Trust
169 settings have an initial value consistent with the verify purpose: e.g.
170 if the verify purpose is for SSL client use it expects the CA to be
171 trusted for SSL client use. However the default value can be changed to
172 permit custom trust settings: one example of this would be to only trust
173 certificates from a specific "secure" set of CAs.
175 Also added X509_STORE_CTX_new() and X509_STORE_CTX_free() functions
176 which should be used for version portability: especially since the
177 verify structure is likely to change more often now.
179 SSL integration. Add purpose and trust to SSL_CTX and SSL and functions
180 to set them. If not set then assume SSL clients will verify SSL servers
183 Two new options to the verify program: -untrusted allows a set of
184 untrusted certificates to be passed in and -purpose which sets the
185 intended purpose of the certificate. If a purpose is set then the
186 new chain verify code is used to check extension consistency.
189 *) Support for the authority information access extension.
192 *) Modify RSA and DSA PEM read routines to transparently handle
193 PKCS#8 format private keys. New *_PUBKEY_* functions that handle
194 public keys in a format compatible with certificate
195 SubjectPublicKeyInfo structures. Unfortunately there were already
196 functions called *_PublicKey_* which used various odd formats so
197 these are retained for compatibility: however the DSA variants were
198 never in a public release so they have been deleted. Changed dsa/rsa
199 utilities to handle the new format: note no releases ever handled public
200 keys so we should be OK.
202 The primary motivation for this change is to avoid the same fiasco
203 that dogs private keys: there are several incompatible private key
204 formats some of which are standard and some OpenSSL specific and
205 require various evil hacks to allow partial transparent handling and
206 even then it doesn't work with DER formats. Given the option anything
207 other than PKCS#8 should be dumped: but the other formats have to
208 stay in the name of compatibility.
210 With public keys and the benefit of hindsight one standard format
211 is used which works with EVP_PKEY, RSA or DSA structures: though
212 it clearly returns an error if you try to read the wrong kind of key.
214 Added a -pubkey option to the 'x509' utility to output the public key.
215 Also rename the EVP_PKEY_get_*() to EVP_PKEY_rget_*() and add
216 EVP_PKEY_rset_*() functions that do the same as the EVP_PKEY_assign_*()
217 except they up the reference count of the added key (they don't "swallow"
221 *) Fixes to crypto/x509/by_file.c the code to read in certificates and
222 CRLs would fail if the file contained no certificates or no CRLs:
223 added a new function to read in both types and return the number
224 read: this means that if none are read it will be an error. The
225 DER versions of the certificate and CRL reader would always fail
226 because it isn't possible to mix certificates and CRLs in DER format
227 without choking one or the other routine. Changed this to just read
228 a certificate: this is the best we can do. Also modified the code
229 in apps/verify.c to take notice of return codes: it was previously
230 attempting to read in certificates from NULL pointers and ignoring
231 any errors: this is one reason why the cert and CRL reader seemed
232 to work. It doesn't check return codes from the default certificate
233 routines: these may well fail if the certificates aren't installed.
236 *) Code to support otherName option in GeneralName.
239 *) First update to verify code. Change the verify utility
240 so it warns if it is passed a self signed certificate:
241 for consistency with the normal behaviour. X509_verify
242 has been modified to it will now verify a self signed
243 certificate if *exactly* the same certificate appears
244 in the store: it was previously impossible to trust a
245 single self signed certificate. This means that:
246 openssl verify ss.pem
247 now gives a warning about a self signed certificate but
248 openssl verify -CAfile ss.pem ss.pem
252 *) For servers, store verify_result in SSL_SESSION data structure
253 (and add it to external session representation).
254 This is needed when client certificate verifications fails,
255 but an application-provided verification callback (set by
256 SSL_CTX_set_cert_verify_callback) allows accepting the session
257 anyway (i.e. leaves x509_store_ctx->error != X509_V_OK
258 but returns 1): When the session is reused, we have to set
259 ssl->verify_result to the appropriate error code to avoid
261 [Bodo Moeller, problem pointed out by Lutz Jaenicke]
263 *) Fix a bug in the new PKCS#7 code: it didn't consider the
264 case in PKCS7_dataInit() where the signed PKCS7 structure
265 didn't contain any existing data because it was being created.
266 [Po-Cheng Chen <pocheng@nst.com.tw>, slightly modified by Steve Henson]
268 *) Add a salt to the key derivation routines in enc.c. This
269 forms the first 8 bytes of the encrypted file. Also add a
270 -S option to allow a salt to be input on the command line.
273 *) New function X509_cmp(). Oddly enough there wasn't a function
274 to compare two certificates. We do this by working out the SHA1
275 hash and comparing that. X509_cmp() will be needed by the trust
279 *) Correctly increment the reference count in the SSL_SESSION pointer
280 returned from SSL_get_session().
281 [Geoff Thorpe <geoff@eu.c2.net>]
283 *) Fix for 'req': it was adding a null to request attributes.
284 Also change the X509_LOOKUP and X509_INFO code to handle
285 certificate auxiliary information.
288 *) Add support for 40 and 64 bit RC2 and RC4 algorithms: document
292 *) Add the possibility to add extra information to the memory leak
293 detecting output, to form tracebacks, showing from where each
294 allocation was originated. Also updated sid code to be multi-
298 *) Add options -text and -noout to pkcs7 utility and delete the
299 encryption options which never did anything. Update docs.
302 *) Add options to some of the utilities to allow the pass phrase
303 to be included on either the command line (not recommended on
304 OSes like Unix) or read from the environment. Update the
305 manpages and fix a few bugs.
308 *) Add a few manpages for some of the openssl commands.
311 *) Fix the -revoke option in ca. It was freeing up memory twice,
312 leaking and not finding already revoked certificates.
315 *) Extensive changes to support certificate auxiliary information.
316 This involves the use of X509_CERT_AUX structure and X509_AUX
317 functions. An X509_AUX function such as PEM_read_X509_AUX()
318 can still read in a certificate file in the usual way but it
319 will also read in any additional "auxiliary information". By
320 doing things this way a fair degree of compatibility can be
321 retained: existing certificates can have this information added
322 using the new 'x509' options.
324 Current auxiliary information includes an "alias" and some trust
325 settings. The trust settings will ultimately be used in enhanced
326 certificate chain verification routines: currently a certificate
327 can only be trusted if it is self signed and then it is trusted
331 *) Fix assembler for Alpha (tested only on DEC OSF not Linux or *BSD). The
332 problem was that one of the replacement routines had not been working since
333 SSLeay releases. For now the offending routine has been replaced with
334 non-optimised assembler. Even so, this now gives around 95% performance
335 improvement for 1024 bit RSA signs.
338 *) Hack to fix PKCS#7 decryption when used with some unorthodox RC2
339 handling. Most clients have the effective key size in bits equal to
340 the key length in bits: so a 40 bit RC2 key uses a 40 bit (5 byte) key.
341 A few however don't do this and instead use the size of the decrypted key
342 to determine the RC2 key length and the AlgorithmIdentifier to determine
343 the effective key length. In this case the effective key lenth can still
344 be 40 bits but the key length can be 168 bits for example. This is fixed
345 by manually forcing an RC2 key into the EVP_PKEY structure because the
346 EVP code can't currently handle unusual RC2 key sizes: it always assumes
347 the key length and effective key length are equal.
350 *) Add a bunch of functions that should simplify the creation of
351 X509_NAME structures. Now you should be able to do:
352 X509_NAME_add_entry_by_txt(nm, "CN", MBSTRING_ASC, "Steve", -1, -1, 0);
353 and have it automatically work out the correct field type and fill in
354 the structures. The more adventurous can try:
355 X509_NAME_add_entry_by_txt(nm, field, MBSTRING_UTF8, str, -1, -1, 0);
356 and it will (hopefully) work out the correct multibyte encoding.
359 *) Change the 'req' utility to use the new field handling and multibyte
360 copy routines. Before the DN field creation was handled in an ad hoc
361 way in req, ca, and x509 which was rather broken and didn't support
362 BMPStrings or UTF8Strings. Since some software doesn't implement
363 BMPStrings or UTF8Strings yet, they can be enabled using the config file
364 using the dirstring_type option. See the new comment in the default
365 openssl.cnf for more info.
368 *) Make crypto/rand/md_rand.c more robust:
369 - Assure unique random numbers after fork().
370 - Make sure that concurrent threads access the global counter and
371 md serializably so that we never lose entropy in them
372 or use exactly the same state in multiple threads.
373 Access to the large state is not always serializable because
374 the additional locking could be a performance killer, and
375 md should be large enough anyway.
378 *) New file apps/app_rand.c with commonly needed functionality
379 for handling the random seed file.
381 Use the random seed file in some applications that previously did not:
383 dsaparam -genkey (which also ignored its '-rand' option),
387 Except on systems with /dev/urandom, it is crucial to have a random
388 seed file at least for key creation, DSA signing, and for DH exchanges;
389 for RSA signatures we could do without one.
391 gendh and gendsa (unlike genrsa) used to read only the first byte
392 of each file listed in the '-rand' option. The function as previously
393 found in genrsa is now in app_rand.c and is used by all programs
394 that support '-rand'.
397 *) In RAND_write_file, use mode 0600 for creating files;
398 don't just chmod when it may be too late.
401 *) Report an error from X509_STORE_load_locations
402 when X509_LOOKUP_load_file or X509_LOOKUP_add_dir failed.
405 *) New function ASN1_mbstring_copy() this copies a string in either
406 ASCII, Unicode, Universal (4 bytes per character) or UTF8 format
407 into an ASN1_STRING type. A mask of permissible types is passed
408 and it chooses the "minimal" type to use or an error if not type
412 *) Add function equivalents to the various macros in asn1.h. The old
413 macros are retained with an M_ prefix. Code inside the library can
414 use the M_ macros. External code (including the openssl utility)
415 should *NOT* in order to be "shared library friendly".
418 *) Add various functions that can check a certificate's extensions
419 to see if it usable for various purposes such as SSL client,
420 server or S/MIME and CAs of these types. This is currently
421 VERY EXPERIMENTAL but will ultimately be used for certificate chain
422 verification. Also added a -purpose flag to x509 utility to
423 print out all the purposes.
426 *) Add a CRYPTO_EX_DATA to X509 certificate structure and associated
430 *) New X509V3_{X509,CRL,REVOKED}_get_d2i() functions. These will search
431 for, obtain and decode and extension and obtain its critical flag.
432 This allows all the necessary extension code to be handled in a
433 single function call.
436 *) RC4 tune-up featuring 30-40% performance improvement on most RISC
437 platforms. See crypto/rc4/rc4_enc.c for further details.
440 *) New -noout option to asn1parse. This causes no output to be produced
441 its main use is when combined with -strparse and -out to extract data
442 from a file (which may not be in ASN.1 format).
445 *) Fix for pkcs12 program. It was hashing an invalid certificate pointer
446 when producing the local key id.
447 [Richard Levitte <levitte@stacken.kth.se>]
449 *) New option -dhparam in s_server. This allows a DH parameter file to be
450 stated explicitly. If it is not stated then it tries the first server
451 certificate file. The previous behaviour hard coded the filename
455 *) Add -pubin and -pubout options to the rsa and dsa commands. These allow
456 a public key to be input or output. For example:
457 openssl rsa -in key.pem -pubout -out pubkey.pem
458 Also added necessary DSA public key functions to handle this.
461 *) Fix so PKCS7_dataVerify() doesn't crash if no certificates are contained
462 in the message. This was handled by allowing
463 X509_find_by_issuer_and_serial() to tolerate a NULL passed to it.
464 [Steve Henson, reported by Sampo Kellomaki <sampo@mail.neuronio.pt>]
466 *) Fix for bug in d2i_ASN1_bytes(): other ASN1 functions add an extra null
467 to the end of the strings whereas this didn't. This would cause problems
468 if strings read with d2i_ASN1_bytes() were later modified.
469 [Steve Henson, reported by Arne Ansper <arne@ats.cyber.ee>]
471 *) Fix for base64 decode bug. When a base64 bio reads only one line of
472 data and it contains EOF it will end up returning an error. This is
473 caused by input 46 bytes long. The cause is due to the way base64
474 BIOs find the start of base64 encoded data. They do this by trying a
475 trial decode on each line until they find one that works. When they
476 do a flag is set and it starts again knowing it can pass all the
477 data directly through the decoder. Unfortunately it doesn't reset
478 the context it uses. This means that if EOF is reached an attempt
479 is made to pass two EOFs through the context and this causes the
480 resulting error. This can also cause other problems as well. As is
481 usual with these problems it takes *ages* to find and the fix is
482 trivial: move one line.
483 [Steve Henson, reported by ian@uns.ns.ac.yu (Ivan Nejgebauer) ]
485 *) Ugly workaround to get s_client and s_server working under Windows. The
486 old code wouldn't work because it needed to select() on sockets and the
487 tty (for keypresses and to see if data could be written). Win32 only
488 supports select() on sockets so we select() with a 1s timeout on the
489 sockets and then see if any characters are waiting to be read, if none
490 are present then we retry, we also assume we can always write data to
491 the tty. This isn't nice because the code then blocks until we've
492 received a complete line of data and it is effectively polling the
493 keyboard at 1s intervals: however it's quite a bit better than not
494 working at all :-) A dedicated Windows application might handle this
495 with an event loop for example.
498 *) Enhance RSA_METHOD structure. Now there are two extra methods, rsa_sign
499 and rsa_verify. When the RSA_FLAGS_SIGN_VER option is set these functions
500 will be called when RSA_sign() and RSA_verify() are used. This is useful
501 if rsa_pub_dec() and rsa_priv_enc() equivalents are not available.
502 For this to work properly RSA_public_decrypt() and RSA_private_encrypt()
503 should *not* be used: RSA_sign() and RSA_verify() must be used instead.
504 This necessitated the support of an extra signature type NID_md5_sha1
505 for SSL signatures and modifications to the SSL library to use it instead
506 of calling RSA_public_decrypt() and RSA_private_encrypt().
509 *) Add new -verify -CAfile and -CApath options to the crl program, these
510 will lookup a CRL issuers certificate and verify the signature in a
511 similar way to the verify program. Tidy up the crl program so it
512 no longer acesses structures directly. Make the ASN1 CRL parsing a bit
513 less strict. It will now permit CRL extensions even if it is not
514 a V2 CRL: this will allow it to tolerate some broken CRLs.
517 *) Initialize all non-automatic variables each time one of the openssl
518 sub-programs is started (this is necessary as they may be started
519 multiple times from the "OpenSSL>" prompt).
520 [Lennart Bang, Bodo Moeller]
522 *) Preliminary compilation option RSA_NULL which disables RSA crypto without
523 removing all other RSA functionality (this is what NO_RSA does). This
524 is so (for example) those in the US can disable those operations covered
525 by the RSA patent while allowing storage and parsing of RSA keys and RSA
529 *) Non-copying interface to BIO pairs.
530 (still largely untested)
533 *) New function ANS1_tag2str() to convert an ASN1 tag to a descriptive
534 ASCII string. This was handled independently in various places before.
537 *) New functions UTF8_getc() and UTF8_putc() that parse and generate
538 UTF8 strings a character at a time.
541 *) Use client_version from client hello to select the protocol
542 (s23_srvr.c) and for RSA client key exchange verification
543 (s3_srvr.c), as required by the SSL 3.0/TLS 1.0 specifications.
546 *) Add various utility functions to handle SPKACs, these were previously
547 handled by poking round in the structure internals. Added new function
548 NETSCAPE_SPKI_print() to print out SPKAC and a new utility 'spkac' to
549 print, verify and generate SPKACs. Based on an original idea from
550 Massimiliano Pala <madwolf@comune.modena.it> but extensively modified.
553 *) RIPEMD160 is operational on all platforms and is back in 'make test'.
556 *) Allow the config file extension section to be overwritten on the
557 command line. Based on an original idea from Massimiliano Pala
558 <madwolf@comune.modena.it>. The new option is called -extensions
559 and can be applied to ca, req and x509. Also -reqexts to override
560 the request extensions in req and -crlexts to override the crl extensions
564 *) Add new feature to the SPKAC handling in ca. Now you can include
565 the same field multiple times by preceding it by "XXXX." for example:
568 this is the same syntax as used in the req config file.
571 *) Allow certificate extensions to be added to certificate requests. These
572 are specified in a 'req_extensions' option of the req section of the
573 config file. They can be printed out with the -text option to req but
574 are otherwise ignored at present.
577 *) Fix a horrible bug in enc_read() in crypto/evp/bio_enc.c: if the first
578 data read consists of only the final block it would not decrypted because
579 EVP_CipherUpdate() would correctly report zero bytes had been decrypted.
580 A misplaced 'break' also meant the decrypted final block might not be
581 copied until the next read.
584 *) Initial support for DH_METHOD. Again based on RSA_METHOD. Also added
585 a few extra parameters to the DH structure: these will be useful if
586 for example we want the value of 'q' or implement X9.42 DH.
589 *) Initial support for DSA_METHOD. This is based on the RSA_METHOD and
590 provides hooks that allow the default DSA functions or functions on a
591 "per key" basis to be replaced. This allows hardware acceleration and
592 hardware key storage to be handled without major modification to the
593 library. Also added low level modexp hooks and CRYPTO_EX structure and
594 associated functions.
597 *) Add a new flag to memory BIOs, BIO_FLAG_MEM_RDONLY. This marks the BIO
598 as "read only": it can't be written to and the buffer it points to will
599 not be freed. Reading from a read only BIO is much more efficient than
600 a normal memory BIO. This was added because there are several times when
601 an area of memory needs to be read from a BIO. The previous method was
602 to create a memory BIO and write the data to it, this results in two
603 copies of the data and an O(n^2) reading algorithm. There is a new
604 function BIO_new_mem_buf() which creates a read only memory BIO from
605 an area of memory. Also modified the PKCS#7 routines to use read only
609 *) Bugfix: ssl23_get_client_hello did not work properly when called in
610 state SSL23_ST_SR_CLNT_HELLO_B, i.e. when the first 7 bytes of
611 a SSLv2-compatible client hello for SSLv3 or TLSv1 could be read,
612 but a retry condition occured while trying to read the rest.
615 *) The PKCS7_ENC_CONTENT_new() function was setting the content type as
616 NID_pkcs7_encrypted by default: this was wrong since this should almost
617 always be NID_pkcs7_data. Also modified the PKCS7_set_type() to handle
618 the encrypted data type: this is a more sensible place to put it and it
619 allows the PKCS#12 code to be tidied up that duplicated this
623 *) Changed obj_dat.pl script so it takes its input and output files on
624 the command line. This should avoid shell escape redirection problems
628 *) Initial support for certificate extension requests, these are included
629 in things like Xenroll certificate requests. Included functions to allow
630 extensions to be obtained and added.
633 *) -crlf option to s_client and s_server for sending newlines as
634 CRLF (as required by many protocols).
637 Changes between 0.9.3a and 0.9.4 [09 Aug 1999]
639 *) Install libRSAglue.a when OpenSSL is built with RSAref.
640 [Ralf S. Engelschall]
642 *) A few more ``#ifndef NO_FP_API / #endif'' pairs for consistency.
643 [Andrija Antonijevic <TheAntony2@bigfoot.com>]
645 *) Fix -startdate and -enddate (which was missing) arguments to 'ca'
649 *) New function DSA_dup_DH, which duplicates DSA parameters/keys as
650 DH parameters/keys (q is lost during that conversion, but the resulting
651 DH parameters contain its length).
653 For 1024-bit p, DSA_generate_parameters followed by DSA_dup_DH is
654 much faster than DH_generate_parameters (which creates parameters
655 where p = 2*q + 1), and also the smaller q makes DH computations
656 much more efficient (160-bit exponentiation instead of 1024-bit
657 exponentiation); so this provides a convenient way to support DHE
658 ciphersuites in SSL/TLS servers (see ssl/ssltest.c). It is of
659 utter importance to use
660 SSL_CTX_set_options(s_ctx, SSL_OP_SINGLE_DH_USE);
662 SSL_set_options(s_ctx, SSL_OP_SINGLE_DH_USE);
663 when such DH parameters are used, because otherwise small subgroup
664 attacks may become possible!
667 *) Avoid memory leak in i2d_DHparams.
670 *) Allow the -k option to be used more than once in the enc program:
671 this allows the same encrypted message to be read by multiple recipients.
674 *) New function OBJ_obj2txt(buf, buf_len, a, no_name), this converts
675 an ASN1_OBJECT to a text string. If the "no_name" parameter is set then
676 it will always use the numerical form of the OID, even if it has a short
680 *) Added an extra RSA flag: RSA_FLAG_EXT_PKEY. Previously the rsa_mod_exp
681 method only got called if p,q,dmp1,dmq1,iqmp components were present,
682 otherwise bn_mod_exp was called. In the case of hardware keys for example
683 no private key components need be present and it might store extra data
684 in the RSA structure, which cannot be accessed from bn_mod_exp.
685 By setting RSA_FLAG_EXT_PKEY rsa_mod_exp will always be called for
686 private key operations.
689 *) Added support for SPARC Linux.
692 *) pem_password_cb function type incompatibly changed from
693 typedef int pem_password_cb(char *buf, int size, int rwflag);
695 ....(char *buf, int size, int rwflag, void *userdata);
696 so that applications can pass data to their callbacks:
697 The PEM[_ASN1]_{read,write}... functions and macros now take an
698 additional void * argument, which is just handed through whenever
699 the password callback is called.
700 [Damien Miller <dmiller@ilogic.com.au>; tiny changes by Bodo Moeller]
702 New function SSL_CTX_set_default_passwd_cb_userdata.
704 Compatibility note: As many C implementations push function arguments
705 onto the stack in reverse order, the new library version is likely to
706 interoperate with programs that have been compiled with the old
707 pem_password_cb definition (PEM_whatever takes some data that
708 happens to be on the stack as its last argument, and the callback
709 just ignores this garbage); but there is no guarantee whatsoever that
712 *) The -DPLATFORM="\"$(PLATFORM)\"" definition and the similar -DCFLAGS=...
713 (both in crypto/Makefile.ssl for use by crypto/cversion.c) caused
714 problems not only on Windows, but also on some Unix platforms.
715 To avoid problematic command lines, these definitions are now in an
716 auto-generated file crypto/buildinf.h (created by crypto/Makefile.ssl
717 for standard "make" builds, by util/mk1mf.pl for "mk1mf" builds).
720 *) MIPS III/IV assembler module is reimplemented.
723 *) More DES library cleanups: remove references to srand/rand and
724 delete an unused file.
727 *) Add support for the the free Netwide assembler (NASM) under Win32,
728 since not many people have MASM (ml) and it can be hard to obtain.
729 This is currently experimental but it seems to work OK and pass all
730 the tests. Check out INSTALL.W32 for info.
733 *) Fix memory leaks in s3_clnt.c: All non-anonymous SSL3/TLS1 connections
734 without temporary keys kept an extra copy of the server key,
735 and connections with temporary keys did not free everything in case
739 *) New function RSA_check_key and new openssl rsa option -check
740 for verifying the consistency of RSA keys.
741 [Ulf Moeller, Bodo Moeller]
743 *) Various changes to make Win32 compile work:
744 1. Casts to avoid "loss of data" warnings in p5_crpt2.c
745 2. Change unsigned int to int in b_dump.c to avoid "signed/unsigned
746 comparison" warnings.
747 3. Add sk_<TYPE>_sort to DEF file generator and do make update.
750 *) Add a debugging option to PKCS#5 v2 key generation function: when
751 you #define DEBUG_PKCS5V2 passwords, salts, iteration counts and
752 derived keys are printed to stderr.
755 *) Copy the flags in ASN1_STRING_dup().
756 [Roman E. Pavlov <pre@mo.msk.ru>]
758 *) The x509 application mishandled signing requests containing DSA
759 keys when the signing key was also DSA and the parameters didn't match.
761 It was supposed to omit the parameters when they matched the signing key:
762 the verifying software was then supposed to automatically use the CA's
763 parameters if they were absent from the end user certificate.
765 Omitting parameters is no longer recommended. The test was also
766 the wrong way round! This was probably due to unusual behaviour in
767 EVP_cmp_parameters() which returns 1 if the parameters match.
768 This meant that parameters were omitted when they *didn't* match and
769 the certificate was useless. Certificates signed with 'ca' didn't have
771 [Steve Henson, reported by Doug Erickson <Doug.Erickson@Part.NET>]
773 *) Memory leak checking (-DCRYPTO_MDEBUG) had some problems.
774 The interface is as follows:
776 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ON) aka MemCheck_start(),
777 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_OFF) aka MemCheck_stop();
778 "off" is now the default.
779 The library internally uses
780 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_DISABLE) aka MemCheck_off(),
781 CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ENABLE) aka MemCheck_on()
782 to disable memory-checking temporarily.
784 Some inconsistent states that previously were possible (and were
785 even the default) are now avoided.
787 -DCRYPTO_MDEBUG_TIME is new and additionally stores the current time
788 with each memory chunk allocated; this is occasionally more helpful
789 than just having a counter.
791 -DCRYPTO_MDEBUG_THREAD is also new and adds the thread ID.
793 -DCRYPTO_MDEBUG_ALL enables all of the above, plus any future
797 *) Introduce "mode" for SSL structures (with defaults in SSL_CTX),
798 which largely parallels "options", but is for changing API behaviour,
799 whereas "options" are about protocol behaviour.
800 Initial "mode" flags are:
802 SSL_MODE_ENABLE_PARTIAL_WRITE Allow SSL_write to report success when
803 a single record has been written.
804 SSL_MODE_ACCEPT_MOVING_WRITE_BUFFER Don't insist that SSL_write
805 retries use the same buffer location.
806 (But all of the contents must be
810 *) Bugfix: SSL_set_mode ignored its parameter, only SSL_CTX_set_mode
813 *) Fix problems with no-hmac etc.
814 [Ulf Möller, pointed out by Brian Wellington <bwelling@tislabs.com>]
816 *) New functions RSA_get_default_method(), RSA_set_method() and
817 RSA_get_method(). These allows replacement of RSA_METHODs without having
818 to mess around with the internals of an RSA structure.
821 *) Fix memory leaks in DSA_do_sign and DSA_is_prime.
822 Also really enable memory leak checks in openssl.c and in some
824 [Chad C. Mulligan, Bodo Moeller]
826 *) Fix a bug in d2i_ASN1_INTEGER() and i2d_ASN1_INTEGER() which can mess
827 up the length of negative integers. This has now been simplified to just
828 store the length when it is first determined and use it later, rather
829 than trying to keep track of where data is copied and updating it to
831 [Steve Henson, reported by Brien Wheeler
832 <bwheeler@authentica-security.com>]
834 *) Add a new function PKCS7_signatureVerify. This allows the verification
835 of a PKCS#7 signature but with the signing certificate passed to the
836 function itself. This contrasts with PKCS7_dataVerify which assumes the
837 certificate is present in the PKCS#7 structure. This isn't always the
838 case: certificates can be omitted from a PKCS#7 structure and be
839 distributed by "out of band" means (such as a certificate database).
842 *) Complete the PEM_* macros with DECLARE_PEM versions to replace the
843 function prototypes in pem.h, also change util/mkdef.pl to add the
844 necessary function names.
847 *) mk1mf.pl (used by Windows builds) did not properly read the
848 options set by Configure in the top level Makefile, and Configure
849 was not even able to write more than one option correctly.
850 Fixed, now "no-idea no-rc5 -DCRYPTO_MDEBUG" etc. works as intended.
853 *) New functions CONF_load_bio() and CONF_load_fp() to allow a config
854 file to be loaded from a BIO or FILE pointer. The BIO version will
855 for example allow memory BIOs to contain config info.
858 *) New function "CRYPTO_num_locks" that returns CRYPTO_NUM_LOCKS.
859 Whoever hopes to achieve shared-library compatibility across versions
860 must use this, not the compile-time macro.
861 (Exercise 0.9.4: Which is the minimum library version required by
863 Note: All this applies only to multi-threaded programs, others don't
867 *) Add missing case to s3_clnt.c state machine -- one of the new SSL tests
868 through a BIO pair triggered the default case, i.e.
869 SSLerr(...,SSL_R_UNKNOWN_STATE).
872 *) New "BIO pair" concept (crypto/bio/bss_bio.c) so that applications
873 can use the SSL library even if none of the specific BIOs is
877 *) Fix a bug in i2d_DSAPublicKey() which meant it returned the wrong value
878 for the encoded length.
879 [Jeon KyoungHo <khjeon@sds.samsung.co.kr>]
881 *) Add initial documentation of the X509V3 functions.
884 *) Add a new pair of functions PEM_write_PKCS8PrivateKey() and
885 PEM_write_bio_PKCS8PrivateKey() that are equivalent to
886 PEM_write_PrivateKey() and PEM_write_bio_PrivateKey() but use the more
887 secure PKCS#8 private key format with a high iteration count.
890 *) Fix determination of Perl interpreter: A perl or perl5
891 _directory_ in $PATH was also accepted as the interpreter.
892 [Ralf S. Engelschall]
894 *) Fix demos/sign/sign.c: well there wasn't anything strictly speaking
895 wrong with it but it was very old and did things like calling
896 PEM_ASN1_read() directly and used MD5 for the hash not to mention some
900 *) Fix demos/selfsign.c: it used obsolete and deleted functions, changed
901 to use the new extension code.
904 *) Implement the PEM_read/PEM_write functions in crypto/pem/pem_all.c
905 with macros. This should make it easier to change their form, add extra
906 arguments etc. Fix a few PEM prototypes which didn't have cipher as a
910 *) Add to configuration table a new entry that can specify an alternative
911 name for unistd.h (for pre-POSIX systems); we need this for NeXTstep,
912 according to Mark Crispin <MRC@Panda.COM>.
916 *) DES CBC did not update the IV. Weird.
919 des_cbc_encrypt does not update the IV, but des_ncbc_encrypt does.
920 Changing the behaviour of the former might break existing programs --
921 where IV updating is needed, des_ncbc_encrypt can be used.
924 *) When bntest is run from "make test" it drives bc to check its
925 calculations, as well as internally checking them. If an internal check
926 fails, it needs to cause bc to give a non-zero result or make test carries
927 on without noticing the failure. Fixed.
930 *) DES library cleanups.
933 *) Add support for PKCS#5 v2.0 PBE algorithms. This will permit PKCS#8 to be
934 used with any cipher unlike PKCS#5 v1.5 which can at most handle 64 bit
935 ciphers. NOTE: although the key derivation function has been verified
936 against some published test vectors it has not been extensively tested
937 yet. Added a -v2 "cipher" option to pkcs8 application to allow the use
941 *) Instead of "mkdir -p", which is not fully portable, use new
942 Perl script "util/mkdir-p.pl".
945 *) Rewrite the way password based encryption (PBE) is handled. It used to
946 assume that the ASN1 AlgorithmIdentifier parameter was a PBEParameter
947 structure. This was true for the PKCS#5 v1.5 and PKCS#12 PBE algorithms
948 but doesn't apply to PKCS#5 v2.0 where it can be something else. Now
949 the 'parameter' field of the AlgorithmIdentifier is passed to the
950 underlying key generation function so it must do its own ASN1 parsing.
951 This has also changed the EVP_PBE_CipherInit() function which now has a
952 'parameter' argument instead of literal salt and iteration count values
953 and the function EVP_PBE_ALGOR_CipherInit() has been deleted.
956 *) Support for PKCS#5 v1.5 compatible password based encryption algorithms
957 and PKCS#8 functionality. New 'pkcs8' application linked to openssl.
958 Needed to change the PEM_STRING_EVP_PKEY value which was just "PRIVATE
959 KEY" because this clashed with PKCS#8 unencrypted string. Since this
960 value was just used as a "magic string" and not used directly its
961 value doesn't matter.
964 *) Introduce some semblance of const correctness to BN. Shame C doesn't
968 *) "linux-sparc64" configuration (ultrapenguin).
969 [Ray Miller <ray.miller@oucs.ox.ac.uk>]
970 "linux-sparc" configuration.
971 [Christian Forster <fo@hawo.stw.uni-erlangen.de>]
973 *) config now generates no-xxx options for missing ciphers.
976 *) Support the EBCDIC character set (work in progress).
977 File ebcdic.c not yet included because it has a different license.
978 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>]
980 *) Support BS2000/OSD-POSIX.
981 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>]
983 *) Make callbacks for key generation use void * instead of char *.
986 *) Make S/MIME samples compile (not yet tested).
989 *) Additional typesafe stacks.
992 *) New configuration variants "bsdi-elf-gcc" (BSD/OS 4.x).
996 Changes between 0.9.3 and 0.9.3a [29 May 1999]
998 *) New configuration variant "sco5-gcc".
1000 *) Updated some demos.
1001 [Sean O Riordain, Wade Scholine]
1003 *) Add missing BIO_free at exit of pkcs12 application.
1006 *) Fix memory leak in conf.c.
1009 *) Updates for Win32 to assembler version of MD5.
1012 *) Set #! path to perl in apps/der_chop to where we found it
1013 instead of using a fixed path.
1016 *) SHA library changes for irix64-mips4-cc.
1019 *) Improvements for VMS support.
1023 Changes between 0.9.2b and 0.9.3 [24 May 1999]
1025 *) Bignum library bug fix. IRIX 6 passes "make test" now!
1026 This also avoids the problems with SC4.2 and unpatched SC5.
1027 [Andy Polyakov <appro@fy.chalmers.se>]
1029 *) New functions sk_num, sk_value and sk_set to replace the previous macros.
1030 These are required because of the typesafe stack would otherwise break
1031 existing code. If old code used a structure member which used to be STACK
1032 and is now STACK_OF (for example cert in a PKCS7_SIGNED structure) with
1033 sk_num or sk_value it would produce an error because the num, data members
1034 are not present in STACK_OF. Now it just produces a warning. sk_set
1035 replaces the old method of assigning a value to sk_value
1036 (e.g. sk_value(x, i) = y) which the library used in a few cases. Any code
1037 that does this will no longer work (and should use sk_set instead) but
1038 this could be regarded as a "questionable" behaviour anyway.
1041 *) Fix most of the other PKCS#7 bugs. The "experimental" code can now
1042 correctly handle encrypted S/MIME data.
1045 *) Change type of various DES function arguments from des_cblock
1046 (which means, in function argument declarations, pointer to char)
1047 to des_cblock * (meaning pointer to array with 8 char elements),
1048 which allows the compiler to do more typechecking; it was like
1049 that back in SSLeay, but with lots of ugly casts.
1051 Introduce new type const_des_cblock.
1054 *) Reorganise the PKCS#7 library and get rid of some of the more obvious
1055 problems: find RecipientInfo structure that matches recipient certificate
1056 and initialise the ASN1 structures properly based on passed cipher.
1059 *) Belatedly make the BN tests actually check the results.
1062 *) Fix the encoding and decoding of negative ASN1 INTEGERS and conversion
1063 to and from BNs: it was completely broken. New compilation option
1064 NEG_PUBKEY_BUG to allow for some broken certificates that encode public
1065 key elements as negative integers.
1068 *) Reorganize and speed up MD5.
1069 [Andy Polyakov <appro@fy.chalmers.se>]
1072 [Richard Levitte <richard@levitte.org>]
1074 *) New option -out to asn1parse to allow the parsed structure to be
1075 output to a file. This is most useful when combined with the -strparse
1076 option to examine the output of things like OCTET STRINGS.
1079 *) Make SSL library a little more fool-proof by not requiring any longer
1080 that SSL_set_{accept,connect}_state be called before
1081 SSL_{accept,connect} may be used (SSL_set_..._state is omitted
1082 in many applications because usually everything *appeared* to work as
1083 intended anyway -- now it really works as intended).
1086 *) Move openssl.cnf out of lib/.
1089 *) Fix various things to let OpenSSL even pass ``egcc -pipe -O2 -Wall
1090 -Wshadow -Wpointer-arith -Wcast-align -Wmissing-prototypes
1091 -Wmissing-declarations -Wnested-externs -Winline'' with EGCS 1.1.2+
1092 [Ralf S. Engelschall]
1094 *) Various fixes to the EVP and PKCS#7 code. It may now be able to
1095 handle PKCS#7 enveloped data properly.
1096 [Sebastian Akerman <sak@parallelconsulting.com>, modified by Steve]
1098 *) Create a duplicate of the SSL_CTX's CERT in SSL_new instead of
1099 copying pointers. The cert_st handling is changed by this in
1100 various ways (and thus what used to be known as ctx->default_cert
1101 is now called ctx->cert, since we don't resort to s->ctx->[default_]cert
1102 any longer when s->cert does not give us what we need).
1103 ssl_cert_instantiate becomes obsolete by this change.
1104 As soon as we've got the new code right (possibly it already is?),
1105 we have solved a couple of bugs of the earlier code where s->cert
1106 was used as if it could not have been shared with other SSL structures.
1108 Note that using the SSL API in certain dirty ways now will result
1109 in different behaviour than observed with earlier library versions:
1110 Changing settings for an SSL_CTX *ctx after having done s = SSL_new(ctx)
1111 does not influence s as it used to.
1113 In order to clean up things more thoroughly, inside SSL_SESSION
1114 we don't use CERT any longer, but a new structure SESS_CERT
1115 that holds per-session data (if available); currently, this is
1116 the peer's certificate chain and, for clients, the server's certificate
1117 and temporary key. CERT holds only those values that can have
1118 meaningful defaults in an SSL_CTX.
1121 *) New function X509V3_EXT_i2d() to create an X509_EXTENSION structure
1122 from the internal representation. Various PKCS#7 fixes: remove some
1123 evil casts and set the enc_dig_alg field properly based on the signing
1127 *) Allow PKCS#12 password to be set from the command line or the
1128 environment. Let 'ca' get its config file name from the environment
1129 variables "OPENSSL_CONF" or "SSLEAY_CONF" (for consistency with 'req'
1133 *) Allow certificate policies extension to use an IA5STRING for the
1134 organization field. This is contrary to the PKIX definition but
1135 VeriSign uses it and IE5 only recognises this form. Document 'x509'
1139 *) Add PEDANTIC compiler flag to allow compilation with gcc -pedantic,
1140 without disallowing inline assembler and the like for non-pedantic builds.
1143 *) Support Borland C++ builder.
1144 [Janez Jere <jj@void.si>, modified by Ulf Möller]
1149 *) SHA-1 cleanups and performance enhancements.
1150 [Andy Polyakov <appro@fy.chalmers.se>]
1152 *) Sparc v8plus assembler for the bignum library.
1153 [Andy Polyakov <appro@fy.chalmers.se>]
1155 *) Accept any -xxx and +xxx compiler options in Configure.
1158 *) Update HPUX configuration.
1161 *) Add missing sk_<type>_unshift() function to safestack.h
1162 [Ralf S. Engelschall]
1164 *) New function SSL_CTX_use_certificate_chain_file that sets the
1165 "extra_cert"s in addition to the certificate. (This makes sense
1166 only for "PEM" format files, as chains as a whole are not
1170 *) Support verify_depth from the SSL API.
1171 x509_vfy.c had what can be considered an off-by-one-error:
1172 Its depth (which was not part of the external interface)
1173 was actually counting the number of certificates in a chain;
1174 now it really counts the depth.
1177 *) Bugfix in crypto/x509/x509_cmp.c: The SSLerr macro was used
1178 instead of X509err, which often resulted in confusing error
1179 messages since the error codes are not globally unique
1180 (e.g. an alleged error in ssl3_accept when a certificate
1181 didn't match the private key).
1183 *) New function SSL_CTX_set_session_id_context that allows to set a default
1184 value (so that you don't need SSL_set_session_id_context for each
1185 connection using the SSL_CTX).
1188 *) OAEP decoding bug fix.
1191 *) Support INSTALL_PREFIX for package builders, as proposed by
1195 *) New Configure options "threads" and "no-threads". For systems
1196 where the proper compiler options are known (currently Solaris
1197 and Linux), "threads" is the default.
1200 *) New script util/mklink.pl as a faster substitute for util/mklink.sh.
1203 *) Install various scripts to $(OPENSSLDIR)/misc, not to
1204 $(INSTALLTOP)/bin -- they shouldn't clutter directories
1205 such as /usr/local/bin.
1208 *) "make linux-shared" to build shared libraries.
1209 [Niels Poppe <niels@netbox.org>]
1211 *) New Configure option no-<cipher> (rsa, idea, rc5, ...).
1214 *) Add the PKCS#12 API documentation to openssl.txt. Preliminary support for
1215 extension adding in x509 utility.
1218 *) Remove NOPROTO sections and error code comments.
1221 *) Partial rewrite of the DEF file generator to now parse the ANSI
1225 *) New Configure options --prefix=DIR and --openssldir=DIR.
1228 *) Complete rewrite of the error code script(s). It is all now handled
1229 by one script at the top level which handles error code gathering,
1230 header rewriting and C source file generation. It should be much better
1231 than the old method: it now uses a modified version of Ulf's parser to
1232 read the ANSI prototypes in all header files (thus the old K&R definitions
1233 aren't needed for error creation any more) and do a better job of
1234 translating function codes into names. The old 'ASN1 error code imbedded
1235 in a comment' is no longer necessary and it doesn't use .err files which
1236 have now been deleted. Also the error code call doesn't have to appear all
1237 on one line (which resulted in some large lines...).
1240 *) Change #include filenames from <foo.h> to <openssl/foo.h>.
1243 *) Change behaviour of ssl2_read when facing length-0 packets: Don't return
1244 0 (which usually indicates a closed connection), but continue reading.
1247 *) Fix some race conditions.
1250 *) Add support for CRL distribution points extension. Add Certificate
1251 Policies and CRL distribution points documentation.
1254 *) Move the autogenerated header file parts to crypto/opensslconf.h.
1257 *) Fix new 56-bit DES export ciphersuites: they were using 7 bytes instead of
1258 8 of keying material. Merlin has also confirmed interop with this fix
1259 between OpenSSL and Baltimore C/SSL 2.0 and J/SSL 2.0.
1260 [Merlin Hughes <merlin@baltimore.ie>]
1262 *) Fix lots of warnings.
1263 [Richard Levitte <levitte@stacken.kth.se>]
1265 *) In add_cert_dir() in crypto/x509/by_dir.c, break out of the loop if
1266 the directory spec didn't end with a LIST_SEPARATOR_CHAR.
1267 [Richard Levitte <levitte@stacken.kth.se>]
1269 *) Fix problems with sizeof(long) == 8.
1270 [Andy Polyakov <appro@fy.chalmers.se>]
1272 *) Change functions to ANSI C.
1275 *) Fix typos in error codes.
1276 [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>, Ulf Möller]
1278 *) Remove defunct assembler files from Configure.
1281 *) SPARC v8 assembler BIGNUM implementation.
1282 [Andy Polyakov <appro@fy.chalmers.se>]
1284 *) Support for Certificate Policies extension: both print and set.
1285 Various additions to support the r2i method this uses.
1288 *) A lot of constification, and fix a bug in X509_NAME_oneline() that could
1289 return a const string when you are expecting an allocated buffer.
1292 *) Add support for ASN1 types UTF8String and VISIBLESTRING, also the CHOICE
1293 types DirectoryString and DisplayText.
1296 *) Add code to allow r2i extensions to access the configuration database,
1297 add an LHASH database driver and add several ctx helper functions.
1300 *) Fix an evil bug in bn_expand2() which caused various BN functions to
1301 fail when they extended the size of a BIGNUM.
1304 *) Various utility functions to handle SXNet extension. Modify mkdef.pl to
1305 support typesafe stack.
1308 *) Fix typo in SSL_[gs]et_options().
1309 [Nils Frostberg <nils@medcom.se>]
1311 *) Delete various functions and files that belonged to the (now obsolete)
1312 old X509V3 handling code.
1315 *) New Configure option "rsaref".
1318 *) Don't auto-generate pem.h.
1321 *) Introduce type-safe ASN.1 SETs.
1324 *) Convert various additional casted stacks to type-safe STACK_OF() variants.
1325 [Ben Laurie, Ralf S. Engelschall, Steve Henson]
1327 *) Introduce type-safe STACKs. This will almost certainly break lots of code
1328 that links with OpenSSL (well at least cause lots of warnings), but fear
1329 not: the conversion is trivial, and it eliminates loads of evil casts. A
1330 few STACKed things have been converted already. Feel free to convert more.
1331 In the fullness of time, I'll do away with the STACK type altogether.
1334 *) Add `openssl ca -revoke <certfile>' facility which revokes a certificate
1335 specified in <certfile> by updating the entry in the index.txt file.
1336 This way one no longer has to edit the index.txt file manually for
1337 revoking a certificate. The -revoke option does the gory details now.
1338 [Massimiliano Pala <madwolf@openca.org>, Ralf S. Engelschall]
1340 *) Fix `openssl crl -noout -text' combination where `-noout' killed the
1341 `-text' option at all and this way the `-noout -text' combination was
1342 inconsistent in `openssl crl' with the friends in `openssl x509|rsa|dsa'.
1343 [Ralf S. Engelschall]
1345 *) Make sure a corresponding plain text error message exists for the
1346 X509_V_ERR_CERT_REVOKED/23 error number which can occur when a
1347 verify callback function determined that a certificate was revoked.
1348 [Ralf S. Engelschall]
1350 *) Bugfix: In test/testenc, don't test "openssl <cipher>" for
1351 ciphers that were excluded, e.g. by -DNO_IDEA. Also, test
1352 all available cipers including rc5, which was forgotten until now.
1353 In order to let the testing shell script know which algorithms
1354 are available, a new (up to now undocumented) command
1355 "openssl list-cipher-commands" is used.
1358 *) Bugfix: s_client occasionally would sleep in select() when
1359 it should have checked SSL_pending() first.
1362 *) New functions DSA_do_sign and DSA_do_verify to provide access to
1363 the raw DSA values prior to ASN.1 encoding.
1366 *) Tweaks to Configure
1367 [Niels Poppe <niels@netbox.org>]
1369 *) Add support for PKCS#5 v2.0 ASN1 PBES2 structures. No other support,
1373 *) New variables $(RANLIB) and $(PERL) in the Makefiles.
1376 *) New config option to avoid instructions that are illegal on the 80386.
1377 The default code is faster, but requires at least a 486.
1380 *) Got rid of old SSL2_CLIENT_VERSION (inconsistently used) and
1381 SSL2_SERVER_VERSION (not used at all) macros, which are now the
1382 same as SSL2_VERSION anyway.
1385 *) New "-showcerts" option for s_client.
1388 *) Still more PKCS#12 integration. Add pkcs12 application to openssl
1389 application. Various cleanups and fixes.
1392 *) More PKCS#12 integration. Add new pkcs12 directory with Makefile.ssl and
1393 modify error routines to work internally. Add error codes and PBE init
1394 to library startup routines.
1397 *) Further PKCS#12 integration. Added password based encryption, PKCS#8 and
1398 packing functions to asn1 and evp. Changed function names and error
1399 codes along the way.
1402 *) PKCS12 integration: and so it begins... First of several patches to
1403 slowly integrate PKCS#12 functionality into OpenSSL. Add PKCS#12
1404 objects to objects.h
1407 *) Add a new 'indent' option to some X509V3 extension code. Initial ASN1
1408 and display support for Thawte strong extranet extension.
1411 *) Add LinuxPPC support.
1412 [Jeff Dubrule <igor@pobox.org>]
1414 *) Get rid of redundant BN file bn_mulw.c, and rename bn_div64 to
1415 bn_div_words in alpha.s.
1416 [Hannes Reinecke <H.Reinecke@hw.ac.uk> and Ben Laurie]
1418 *) Make sure the RSA OAEP test is skipped under -DRSAref because
1419 OAEP isn't supported when OpenSSL is built with RSAref.
1420 [Ulf Moeller <ulf@fitug.de>]
1422 *) Move definitions of IS_SET/IS_SEQUENCE inside crypto/asn1/asn1.h
1423 so they no longer are missing under -DNOPROTO.
1424 [Soren S. Jorvang <soren@t.dk>]
1427 Changes between 0.9.1c and 0.9.2b [22 Mar 1999]
1429 *) Make SSL_get_peer_cert_chain() work in servers. Unfortunately, it still
1430 doesn't work when the session is reused. Coming soon!
1433 *) Fix a security hole, that allows sessions to be reused in the wrong
1434 context thus bypassing client cert protection! All software that uses
1435 client certs and session caches in multiple contexts NEEDS PATCHING to
1436 allow session reuse! A fuller solution is in the works.
1437 [Ben Laurie, problem pointed out by Holger Reif, Bodo Moeller (and ???)]
1439 *) Some more source tree cleanups (removed obsolete files
1440 crypto/bf/asm/bf586.pl, test/test.txt and crypto/sha/asm/f.s; changed
1441 permission on "config" script to be executable) and a fix for the INSTALL
1443 [Ulf Moeller <ulf@fitug.de>]
1445 *) Remove some legacy and erroneous uses of malloc, free instead of
1447 [Lennart Bang <lob@netstream.se>, with minor changes by Steve]
1449 *) Make rsa_oaep_test return non-zero on error.
1450 [Ulf Moeller <ulf@fitug.de>]
1452 *) Add support for native Solaris shared libraries. Configure
1453 solaris-sparc-sc4-pic, make, then run shlib/solaris-sc4.sh. It'd be nice
1454 if someone would make that last step automatic.
1455 [Matthias Loepfe <Matthias.Loepfe@AdNovum.CH>]
1457 *) ctx_size was not built with the right compiler during "make links". Fixed.
1460 *) Change the meaning of 'ALL' in the cipher list. It now means "everything
1461 except NULL ciphers". This means the default cipher list will no longer
1462 enable NULL ciphers. They need to be specifically enabled e.g. with
1463 the string "DEFAULT:eNULL".
1466 *) Fix to RSA private encryption routines: if p < q then it would
1467 occasionally produce an invalid result. This will only happen with
1468 externally generated keys because OpenSSL (and SSLeay) ensure p > q.
1471 *) Be less restrictive and allow also `perl util/perlpath.pl
1472 /path/to/bin/perl' in addition to `perl util/perlpath.pl /path/to/bin',
1473 because this way one can also use an interpreter named `perl5' (which is
1474 usually the name of Perl 5.xxx on platforms where an Perl 4.x is still
1475 installed as `perl').
1476 [Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
1478 *) Let util/clean-depend.pl work also with older Perl 5.00x versions.
1479 [Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
1481 *) Fix Makefile.org so CC,CFLAG etc are passed to 'make links' add
1482 advapi32.lib to Win32 build and change the pem test comparision
1483 to fc.exe (thanks to Ulrich Kroener <kroneru@yahoo.com> for the
1484 suggestion). Fix misplaced ASNI prototypes and declarations in evp.h
1485 and crypto/des/ede_cbcm_enc.c.
1488 *) DES quad checksum was broken on big-endian architectures. Fixed.
1491 *) Comment out two functions in bio.h that aren't implemented. Fix up the
1492 Win32 test batch file so it (might) work again. The Win32 test batch file
1493 is horrible: I feel ill....
1496 *) Move various #ifdefs around so NO_SYSLOG, NO_DIRENT etc are now selected
1497 in e_os.h. Audit of header files to check ANSI and non ANSI
1498 sections: 10 functions were absent from non ANSI section and not exported
1499 from Windows DLLs. Fixed up libeay.num for new functions.
1502 *) Make `openssl version' output lines consistent.
1503 [Ralf S. Engelschall]
1505 *) Fix Win32 symbol export lists for BIO functions: Added
1506 BIO_get_ex_new_index, BIO_get_ex_num, BIO_get_ex_data and BIO_set_ex_data
1507 to ms/libeay{16,32}.def.
1508 [Ralf S. Engelschall]
1510 *) Second round of fixing the OpenSSL perl/ stuff. It now at least compiled
1511 fine under Unix and passes some trivial tests I've now added. But the
1512 whole stuff is horribly incomplete, so a README.1ST with a disclaimer was
1513 added to make sure no one expects that this stuff really works in the
1514 OpenSSL 0.9.2 release. Additionally I've started to clean the XS sources
1515 up and fixed a few little bugs and inconsistencies in OpenSSL.{pm,xs} and
1517 [Ralf S. Engelschall]
1519 *) Fix the generation of two part addresses in perl.
1520 [Kenji Miyake <kenji@miyake.org>, integrated by Ben Laurie]
1522 *) Add config entry for Linux on MIPS.
1523 [John Tobey <jtobey@channel1.com>]
1525 *) Make links whenever Configure is run, unless we are on Windoze.
1528 *) Permit extensions to be added to CRLs using crl_section in openssl.cnf.
1529 Currently only issuerAltName and AuthorityKeyIdentifier make any sense
1533 *) Add a useful kludge to allow package maintainers to specify compiler and
1534 other platforms details on the command line without having to patch the
1535 Configure script everytime: One now can use ``perl Configure
1536 <id>:<details>'', i.e. platform ids are allowed to have details appended
1537 to them (seperated by colons). This is treated as there would be a static
1538 pre-configured entry in Configure's %table under key <id> with value
1539 <details> and ``perl Configure <id>'' is called. So, when you want to
1540 perform a quick test-compile under FreeBSD 3.1 with pgcc and without
1541 assembler stuff you can use ``perl Configure "FreeBSD-elf:pgcc:-O6:::"''
1542 now, which overrides the FreeBSD-elf entry on-the-fly.
1543 [Ralf S. Engelschall]
1545 *) Disable new TLS1 ciphersuites by default: they aren't official yet.
1548 *) Allow DSO flags like -fpic, -fPIC, -KPIC etc. to be specified
1549 on the `perl Configure ...' command line. This way one can compile
1550 OpenSSL libraries with Position Independent Code (PIC) which is needed
1551 for linking it into DSOs.
1552 [Ralf S. Engelschall]
1554 *) Remarkably, export ciphers were totally broken and no-one had noticed!
1558 *) Cleaned up the LICENSE document: The official contact for any license
1559 questions now is the OpenSSL core team under openssl-core@openssl.org.
1560 And add a paragraph about the dual-license situation to make sure people
1561 recognize that _BOTH_ the OpenSSL license _AND_ the SSLeay license apply
1562 to the OpenSSL toolkit.
1563 [Ralf S. Engelschall]
1565 *) General source tree makefile cleanups: Made `making xxx in yyy...'
1566 display consistent in the source tree and replaced `/bin/rm' by `rm'.
1567 Additonally cleaned up the `make links' target: Remove unnecessary
1568 semicolons, subsequent redundant removes, inline point.sh into mklink.sh
1569 to speed processing and no longer clutter the display with confusing
1570 stuff. Instead only the actually done links are displayed.
1571 [Ralf S. Engelschall]
1573 *) Permit null encryption ciphersuites, used for authentication only. It used
1574 to be necessary to set the preprocessor define SSL_ALLOW_ENULL to do this.
1575 It is now necessary to set SSL_FORBID_ENULL to prevent the use of null
1579 *) Add a bunch of fixes to the PKCS#7 stuff. It used to sometimes reorder
1580 signed attributes when verifying signatures (this would break them),
1581 the detached data encoding was wrong and public keys obtained using
1582 X509_get_pubkey() weren't freed.
1585 *) Add text documentation for the BUFFER functions. Also added a work around
1586 to a Win95 console bug. This was triggered by the password read stuff: the
1587 last character typed gets carried over to the next fread(). If you were
1588 generating a new cert request using 'req' for example then the last
1589 character of the passphrase would be CR which would then enter the first
1593 *) Added the new `Includes OpenSSL Cryptography Software' button as
1594 doc/openssl_button.{gif,html} which is similar in style to the old SSLeay
1595 button and can be used by applications based on OpenSSL to show the
1596 relationship to the OpenSSL project.
1597 [Ralf S. Engelschall]
1599 *) Remove confusing variables in function signatures in files
1600 ssl/ssl_lib.c and ssl/ssl.h.
1601 [Lennart Bong <lob@kulthea.stacken.kth.se>]
1603 *) Don't install bss_file.c under PREFIX/include/
1604 [Lennart Bong <lob@kulthea.stacken.kth.se>]
1606 *) Get the Win32 compile working again. Modify mkdef.pl so it can handle
1607 functions that return function pointers and has support for NT specific
1608 stuff. Fix mk1mf.pl and VC-32.pl to support NT differences also. Various
1609 #ifdef WIN32 and WINNTs sprinkled about the place and some changes from
1610 unsigned to signed types: this was killing the Win32 compile.
1613 *) Add new certificate file to stack functions,
1614 SSL_add_dir_cert_subjects_to_stack() and
1615 SSL_add_file_cert_subjects_to_stack(). These largely supplant
1616 SSL_load_client_CA_file(), and can be used to add multiple certs easily
1617 to a stack (usually this is then handed to SSL_CTX_set_client_CA_list()).
1618 This means that Apache-SSL and similar packages don't have to mess around
1619 to add as many CAs as they want to the preferred list.
1622 *) Experiment with doxygen documentation. Currently only partially applied to
1624 See http://www.stack.nl/~dimitri/doxygen/index.html, and run doxygen with
1625 openssl.doxy as the configuration file.
1628 *) Get rid of remaining C++-style comments which strict C compilers hate.
1629 [Ralf S. Engelschall, pointed out by Carlos Amengual]
1631 *) Changed BN_RECURSION in bn_mont.c to BN_RECURSION_MONT so it is not
1632 compiled in by default: it has problems with large keys.
1635 *) Add a bunch of SSL_xxx() functions for configuring the temporary RSA and
1636 DH private keys and/or callback functions which directly correspond to
1637 their SSL_CTX_xxx() counterparts but work on a per-connection basis. This
1638 is needed for applications which have to configure certificates on a
1639 per-connection basis (e.g. Apache+mod_ssl) instead of a per-context basis
1641 For the RSA certificate situation is makes no difference, but
1642 for the DSA certificate situation this fixes the "no shared cipher"
1643 problem where the OpenSSL cipher selection procedure failed because the
1644 temporary keys were not overtaken from the context and the API provided
1645 no way to reconfigure them.
1646 The new functions now let applications reconfigure the stuff and they
1647 are in detail: SSL_need_tmp_RSA, SSL_set_tmp_rsa, SSL_set_tmp_dh,
1648 SSL_set_tmp_rsa_callback and SSL_set_tmp_dh_callback. Additionally a new
1649 non-public-API function ssl_cert_instantiate() is used as a helper
1650 function and also to reduce code redundancy inside ssl_rsa.c.
1651 [Ralf S. Engelschall]
1653 *) Move s_server -dcert and -dkey options out of the undocumented feature
1654 area because they are useful for the DSA situation and should be
1655 recognized by the users.
1656 [Ralf S. Engelschall]
1658 *) Fix the cipher decision scheme for export ciphers: the export bits are
1659 *not* within SSL_MKEY_MASK or SSL_AUTH_MASK, they are within
1660 SSL_EXP_MASK. So, the original variable has to be used instead of the
1661 already masked variable.
1662 [Richard Levitte <levitte@stacken.kth.se>]
1664 *) Fix 'port' variable from `int' to `unsigned int' in crypto/bio/b_sock.c
1665 [Richard Levitte <levitte@stacken.kth.se>]
1667 *) Change type of another md_len variable in pk7_doit.c:PKCS7_dataFinal()
1668 from `int' to `unsigned int' because it's a length and initialized by
1669 EVP_DigestFinal() which expects an `unsigned int *'.
1670 [Richard Levitte <levitte@stacken.kth.se>]
1672 *) Don't hard-code path to Perl interpreter on shebang line of Configure
1673 script. Instead use the usual Shell->Perl transition trick.
1674 [Ralf S. Engelschall]
1676 *) Make `openssl x509 -noout -modulus' functional also for DSA certificates
1677 (in addition to RSA certificates) to match the behaviour of `openssl dsa
1678 -noout -modulus' as it's already the case for `openssl rsa -noout
1679 -modulus'. For RSA the -modulus is the real "modulus" while for DSA
1680 currently the public key is printed (a decision which was already done by
1681 `openssl dsa -modulus' in the past) which serves a similar purpose.
1682 Additionally the NO_RSA no longer completely removes the whole -modulus
1683 option; it now only avoids using the RSA stuff. Same applies to NO_DSA
1685 [Ralf S. Engelschall]
1687 *) Add Arne Ansper's reliable BIO - this is an encrypted, block-digested
1688 BIO. See the source (crypto/evp/bio_ok.c) for more info.
1689 [Arne Ansper <arne@ats.cyber.ee>]
1691 *) Dump the old yucky req code that tried (and failed) to allow raw OIDs
1692 to be added. Now both 'req' and 'ca' can use new objects defined in the
1696 *) Add cool BIO that does syslog (or event log on NT).
1697 [Arne Ansper <arne@ats.cyber.ee>, integrated by Ben Laurie]
1699 *) Add support for new TLS ciphersuites, TLS_RSA_EXPORT56_WITH_RC4_56_MD5,
1700 TLS_RSA_EXPORT56_WITH_RC2_CBC_56_MD5 and
1701 TLS_RSA_EXPORT56_WITH_DES_CBC_SHA, as specified in "56-bit Export Cipher
1702 Suites For TLS", draft-ietf-tls-56-bit-ciphersuites-00.txt.
1705 *) Add preliminary config info for new extension code.
1708 *) Make RSA_NO_PADDING really use no padding.
1709 [Ulf Moeller <ulf@fitug.de>]
1711 *) Generate errors when private/public key check is done.
1714 *) Overhaul for 'crl' utility. New function X509_CRL_print. Partial support
1715 for some CRL extensions and new objects added.
1718 *) Really fix the ASN1 IMPLICIT bug this time... Partial support for private
1719 key usage extension and fuller support for authority key id.
1722 *) Add OAEP encryption for the OpenSSL crypto library. OAEP is the improved
1723 padding method for RSA, which is recommended for new applications in PKCS
1724 #1 v2.0 (RFC 2437, October 1998).
1725 OAEP (Optimal Asymmetric Encryption Padding) has better theoretical
1726 foundations than the ad-hoc padding used in PKCS #1 v1.5. It is secure
1727 against Bleichbacher's attack on RSA.
1728 [Ulf Moeller <ulf@fitug.de>, reformatted, corrected and integrated by
1731 *) Updates to the new SSL compression code
1732 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
1734 *) Fix so that the version number in the master secret, when passed
1735 via RSA, checks that if TLS was proposed, but we roll back to SSLv3
1736 (because the server will not accept higher), that the version number
1737 is 0x03,0x01, not 0x03,0x00
1738 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
1740 *) Run extensive memory leak checks on SSL apps. Fixed *lots* of memory
1741 leaks in ssl/ relating to new X509_get_pubkey() behaviour. Also fixes
1742 in apps/ and an unrelated leak in crypto/dsa/dsa_vrf.c
1745 *) Support for RAW extensions where an arbitrary extension can be
1746 created by including its DER encoding. See apps/openssl.cnf for
1750 *) Make sure latest Perl versions don't interpret some generated C array
1751 code as Perl array code in the crypto/err/err_genc.pl script.
1752 [Lars Weber <3weber@informatik.uni-hamburg.de>]
1754 *) Modify ms/do_ms.bat to not generate assembly language makefiles since
1755 not many people have the assembler. Various Win32 compilation fixes and
1756 update to the INSTALL.W32 file with (hopefully) more accurate Win32
1760 *) Modify configure script 'Configure' to automatically create crypto/date.h
1761 file under Win32 and also build pem.h from pem.org. New script
1762 util/mkfiles.pl to create the MINFO file on environments that can't do a
1763 'make files': perl util/mkfiles.pl >MINFO should work.
1766 *) Major rework of DES function declarations, in the pursuit of correctness
1767 and purity. As a result, many evil casts evaporated, and some weirdness,
1768 too. You may find this causes warnings in your code. Zapping your evil
1769 casts will probably fix them. Mostly.
1772 *) Fix for a typo in asn1.h. Bug fix to object creation script
1773 obj_dat.pl. It considered a zero in an object definition to mean
1774 "end of object": none of the objects in objects.h have any zeros
1775 so it wasn't spotted.
1776 [Steve Henson, reported by Erwann ABALEA <eabalea@certplus.com>]
1778 *) Add support for Triple DES Cipher Block Chaining with Output Feedback
1779 Masking (CBCM). In the absence of test vectors, the best I have been able
1780 to do is check that the decrypt undoes the encrypt, so far. Send me test
1781 vectors if you have them.
1784 *) Correct calculation of key length for export ciphers (too much space was
1785 allocated for null ciphers). This has not been tested!
1788 *) Modifications to the mkdef.pl for Win32 DEF file creation. The usage
1789 message is now correct (it understands "crypto" and "ssl" on its
1790 command line). There is also now an "update" option. This will update
1791 the util/ssleay.num and util/libeay.num files with any new functions.
1793 perl util/mkdef.pl crypto ssl update
1794 it will update them.
1797 *) Overhauled the Perl interface (perl/*):
1798 - ported BN stuff to OpenSSL's different BN library
1799 - made the perl/ source tree CVS-aware
1800 - renamed the package from SSLeay to OpenSSL (the files still contain
1801 their history because I've copied them in the repository)
1802 - removed obsolete files (the test scripts will be replaced
1803 by better Test::Harness variants in the future)
1804 [Ralf S. Engelschall]
1806 *) First cut for a very conservative source tree cleanup:
1807 1. merge various obsolete readme texts into doc/ssleay.txt
1808 where we collect the old documents and readme texts.
1809 2. remove the first part of files where I'm already sure that we no
1810 longer need them because of three reasons: either they are just temporary
1811 files which were left by Eric or they are preserved original files where
1812 I've verified that the diff is also available in the CVS via "cvs diff
1813 -rSSLeay_0_8_1b" or they were renamed (as it was definitely the case for
1814 the crypto/md/ stuff).
1815 [Ralf S. Engelschall]
1817 *) More extension code. Incomplete support for subject and issuer alt
1818 name, issuer and authority key id. Change the i2v function parameters
1819 and add an extra 'crl' parameter in the X509V3_CTX structure: guess
1820 what that's for :-) Fix to ASN1 macro which messed up
1821 IMPLICIT tag and add f_enum.c which adds a2i, i2a for ENUMERATED.
1824 *) Preliminary support for ENUMERATED type. This is largely copied from the
1828 *) Add new function, EVP_MD_CTX_copy() to replace frequent use of memcpy.
1829 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
1831 *) Make sure `make rehash' target really finds the `openssl' program.
1832 [Ralf S. Engelschall, Matthias Loepfe <Matthias.Loepfe@adnovum.ch>]
1834 *) Squeeze another 7% of speed out of MD5 assembler, at least on a P2. I'd
1835 like to hear about it if this slows down other processors.
1838 *) Add CygWin32 platform information to Configure script.
1839 [Alan Batie <batie@aahz.jf.intel.com>]
1841 *) Fixed ms/32all.bat script: `no_asm' -> `no-asm'
1842 [Rainer W. Gerling <gerling@mpg-gv.mpg.de>]
1844 *) New program nseq to manipulate netscape certificate sequences
1847 *) Modify crl2pkcs7 so it supports multiple -certfile arguments. Fix a
1851 *) Fixes to BN code. Previously the default was to define BN_RECURSION
1852 but the BN code had some problems that would cause failures when
1853 doing certificate verification and some other functions.
1854 [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)]
1856 *) Add ASN1 and PEM code to support netscape certificate sequences.
1859 *) Add ASN1 and PEM code to support netscape certificate sequences.
1862 *) Add several PKIX and private extended key usage OIDs.
1865 *) Modify the 'ca' program to handle the new extension code. Modify
1866 openssl.cnf for new extension format, add comments.
1869 *) More X509 V3 changes. Fix typo in v3_bitstr.c. Add support to 'req'
1870 and add a sample to openssl.cnf so req -x509 now adds appropriate
1874 *) Continued X509 V3 changes. Add to other makefiles, integrate with the
1875 error code, add initial support to X509_print() and x509 application.
1878 *) Takes a deep breath and start addding X509 V3 extension support code. Add
1879 files in crypto/x509v3. Move original stuff to crypto/x509v3/old. All this
1880 stuff is currently isolated and isn't even compiled yet.
1883 *) Continuing patches for GeneralizedTime. Fix up certificate and CRL
1884 ASN1 to use ASN1_TIME and modify print routines to use ASN1_TIME_print.
1885 Removed the versions check from X509 routines when loading extensions:
1886 this allows certain broken certificates that don't set the version
1887 properly to be processed.
1890 *) Deal with irritating shit to do with dependencies, in YAAHW (Yet Another
1891 Ad Hoc Way) - Makefile.ssls now all contain local dependencies, which
1892 can still be regenerated with "make depend".
1895 *) Spelling mistake in C version of CAST-128.
1896 [Ben Laurie, reported by Jeremy Hylton <jeremy@cnri.reston.va.us>]
1898 *) Changes to the error generation code. The perl script err-code.pl
1899 now reads in the old error codes and retains the old numbers, only
1900 adding new ones if necessary. It also only changes the .err files if new
1901 codes are added. The makefiles have been modified to only insert errors
1902 when needed (to avoid needlessly modifying header files). This is done
1903 by only inserting errors if the .err file is newer than the auto generated
1904 C file. To rebuild all the error codes from scratch (the old behaviour)
1905 either modify crypto/Makefile.ssl to pass the -regen flag to err_code.pl
1906 or delete all the .err files.
1909 *) CAST-128 was incorrectly implemented for short keys. The C version has
1910 been fixed, but is untested. The assembler versions are also fixed, but
1911 new assembler HAS NOT BEEN GENERATED FOR WIN32 - the Makefile needs fixing
1912 to regenerate it if needed.
1913 [Ben Laurie, reported (with fix for C version) by Jun-ichiro itojun
1914 Hagino <itojun@kame.net>]
1916 *) File was opened incorrectly in randfile.c.
1917 [Ulf Möller <ulf@fitug.de>]
1919 *) Beginning of support for GeneralizedTime. d2i, i2d, check and print
1920 functions. Also ASN1_TIME suite which is a CHOICE of UTCTime or
1921 GeneralizedTime. ASN1_TIME is the proper type used in certificates et
1922 al: it's just almost always a UTCTime. Note this patch adds new error
1923 codes so do a "make errors" if there are problems.
1926 *) Correct Linux 1 recognition in config.
1927 [Ulf Möller <ulf@fitug.de>]
1929 *) Remove pointless MD5 hash when using DSA keys in ca.
1930 [Anonymous <nobody@replay.com>]
1932 *) Generate an error if given an empty string as a cert directory. Also
1933 generate an error if handed NULL (previously returned 0 to indicate an
1934 error, but didn't set one).
1935 [Ben Laurie, reported by Anonymous <nobody@replay.com>]
1937 *) Add prototypes to SSL methods. Make SSL_write's buffer const, at last.
1940 *) Fix the dummy function BN_ref_mod_exp() in rsaref.c to have the correct
1941 parameters. This was causing a warning which killed off the Win32 compile.
1944 *) Remove C++ style comments from crypto/bn/bn_local.h.
1945 [Neil Costigan <neil.costigan@celocom.com>]
1947 *) The function OBJ_txt2nid was broken. It was supposed to return a nid
1948 based on a text string, looking up short and long names and finally
1949 "dot" format. The "dot" format stuff didn't work. Added new function
1950 OBJ_txt2obj to do the same but return an ASN1_OBJECT and rewrote
1951 OBJ_txt2nid to use it. OBJ_txt2obj can also return objects even if the
1952 OID is not part of the table.
1955 *) Add prototypes to X509 lookup/verify methods, fixing a bug in
1956 X509_LOOKUP_by_alias().
1959 *) Sort openssl functions by name.
1962 *) Get the gendsa program working (hopefully) and add it to app list. Remove
1963 encryption from sample DSA keys (in case anyone is interested the password
1967 *) Make _all_ *_free functions accept a NULL pointer.
1968 [Frans Heymans <fheymans@isaserver.be>]
1970 *) If a DH key is generated in s3_srvr.c, don't blow it by trying to use
1972 [Anonymous <nobody@replay.com>]
1974 *) s_server should send the CAfile as acceptable CAs, not its own cert.
1975 [Bodo Moeller <3moeller@informatik.uni-hamburg.de>]
1977 *) Don't blow it for numeric -newkey arguments to apps/req.
1978 [Bodo Moeller <3moeller@informatik.uni-hamburg.de>]
1980 *) Temp key "for export" tests were wrong in s3_srvr.c.
1981 [Anonymous <nobody@replay.com>]
1983 *) Add prototype for temp key callback functions
1984 SSL_CTX_set_tmp_{rsa,dh}_callback().
1987 *) Make DH_free() tolerate being passed a NULL pointer (like RSA_free() and
1988 DSA_free()). Make X509_PUBKEY_set() check for errors in d2i_PublicKey().
1991 *) X509_name_add_entry() freed the wrong thing after an error.
1992 [Arne Ansper <arne@ats.cyber.ee>]
1994 *) rsa_eay.c would attempt to free a NULL context.
1995 [Arne Ansper <arne@ats.cyber.ee>]
1997 *) BIO_s_socket() had a broken should_retry() on Windoze.
1998 [Arne Ansper <arne@ats.cyber.ee>]
2000 *) BIO_f_buffer() didn't pass on BIO_CTRL_FLUSH.
2001 [Arne Ansper <arne@ats.cyber.ee>]
2003 *) Make sure the already existing X509_STORE->depth variable is initialized
2004 in X509_STORE_new(), but document the fact that this variable is still
2005 unused in the certificate verification process.
2006 [Ralf S. Engelschall]
2008 *) Fix the various library and apps files to free up pkeys obtained from
2009 X509_PUBKEY_get() et al. Also allow x509.c to handle netscape extensions.
2012 *) Fix reference counting in X509_PUBKEY_get(). This makes
2013 demos/maurice/example2.c work, amongst others, probably.
2014 [Steve Henson and Ben Laurie]
2016 *) First cut of a cleanup for apps/. First the `ssleay' program is now named
2017 `openssl' and second, the shortcut symlinks for the `openssl <command>'
2018 are no longer created. This way we have a single and consistent command
2019 line interface `openssl <command>', similar to `cvs <command>'.
2020 [Ralf S. Engelschall, Paul Sutton and Ben Laurie]
2022 *) ca.c: move test for DSA keys inside #ifndef NO_DSA. Make pubkey
2023 BIT STRING wrapper always have zero unused bits.
2026 *) Add CA.pl, perl version of CA.sh, add extended key usage OID.
2029 *) Make the top-level INSTALL documentation easier to understand.
2032 *) Makefiles updated to exit if an error occurs in a sub-directory
2033 make (including if user presses ^C) [Paul Sutton]
2035 *) Make Montgomery context stuff explicit in RSA data structure.
2038 *) Fix build order of pem and err to allow for generated pem.h.
2041 *) Fix renumbering bug in X509_NAME_delete_entry().
2044 *) Enhanced the err-ins.pl script so it makes the error library number
2045 global and can add a library name. This is needed for external ASN1 and
2046 other error libraries.
2049 *) Fixed sk_insert which never worked properly.
2052 *) Fix ASN1 macros so they can handle indefinite length construted
2053 EXPLICIT tags. Some non standard certificates use these: they can now
2057 *) Merged the various old/obsolete SSLeay documentation files (doc/xxx.doc)
2058 into a single doc/ssleay.txt bundle. This way the information is still
2059 preserved but no longer messes up this directory. Now it's new room for
2060 the new set of documenation files.
2061 [Ralf S. Engelschall]
2063 *) SETs were incorrectly DER encoded. This was a major pain, because they
2064 shared code with SEQUENCEs, which aren't coded the same. This means that
2065 almost everything to do with SETs or SEQUENCEs has either changed name or
2066 number of arguments.
2067 [Ben Laurie, based on a partial fix by GP Jayan <gp@nsj.co.jp>]
2069 *) Fix test data to work with the above.
2072 *) Fix the RSA header declarations that hid a bug I fixed in 0.9.0b but
2073 was already fixed by Eric for 0.9.1 it seems.
2074 [Ben Laurie - pointed out by Ulf Möller <ulf@fitug.de>]
2076 *) Autodetect FreeBSD3.
2079 *) Fix various bugs in Configure. This affects the following platforms:
2083 unixware-2.0-pentium
2087 *) Eliminate generated files from CVS. Reorder tests to regenerate files
2088 before they are needed.
2091 *) Generate Makefile.ssl from Makefile.org (to keep CVS happy).
2095 Changes between 0.9.1b and 0.9.1c [23-Dec-1998]
2097 *) Added OPENSSL_VERSION_NUMBER to crypto/crypto.h and
2098 changed SSLeay to OpenSSL in version strings.
2099 [Ralf S. Engelschall]
2101 *) Some fixups to the top-level documents.
2104 *) Fixed the nasty bug where rsaref.h was not found under compile-time
2105 because the symlink to include/ was missing.
2106 [Ralf S. Engelschall]
2108 *) Incorporated the popular no-RSA/DSA-only patches
2109 which allow to compile a RSA-free SSLeay.
2110 [Andrew Cooke / Interrader Ldt., Ralf S. Engelschall]
2112 *) Fixed nasty rehash problem under `make -f Makefile.ssl links'
2113 when "ssleay" is still not found.
2114 [Ralf S. Engelschall]
2116 *) Added more platforms to Configure: Cray T3E, HPUX 11,
2117 [Ralf S. Engelschall, Beckmann <beckman@acl.lanl.gov>]
2119 *) Updated the README file.
2120 [Ralf S. Engelschall]
2122 *) Added various .cvsignore files in the CVS repository subdirs
2123 to make a "cvs update" really silent.
2124 [Ralf S. Engelschall]
2126 *) Recompiled the error-definition header files and added
2127 missing symbols to the Win32 linker tables.
2128 [Ralf S. Engelschall]
2130 *) Cleaned up the top-level documents;
2131 o new files: CHANGES and LICENSE
2132 o merged VERSION, HISTORY* and README* files a CHANGES.SSLeay
2133 o merged COPYRIGHT into LICENSE
2134 o removed obsolete TODO file
2135 o renamed MICROSOFT to INSTALL.W32
2136 [Ralf S. Engelschall]
2138 *) Removed dummy files from the 0.9.1b source tree:
2139 crypto/asn1/x crypto/bio/cd crypto/bio/fg crypto/bio/grep crypto/bio/vi
2140 crypto/bn/asm/......add.c crypto/bn/asm/a.out crypto/dsa/f crypto/md5/f
2141 crypto/pem/gmon.out crypto/perlasm/f crypto/pkcs7/build crypto/rsa/f
2142 crypto/sha/asm/f crypto/threads/f ms/zzz ssl/f ssl/f.mak test/f
2143 util/f.mak util/pl/f util/pl/f.mak crypto/bf/bf_locl.old apps/f
2144 [Ralf S. Engelschall]
2146 *) Added various platform portability fixes.
2149 *) The Genesis of the OpenSSL rpject:
2150 We start with the latest (unreleased) SSLeay version 0.9.1b which Eric A.
2151 Young and Tim J. Hudson created while they were working for C2Net until
2153 [The OpenSSL Project]
2156 Changes between 0.9.0b and 0.9.1b [not released]
2158 *) Updated a few CA certificates under certs/
2161 *) Changed some BIGNUM api stuff.
2164 *) Various platform ports: OpenBSD, Ultrix, IRIX 64bit, NetBSD,
2165 DGUX x86, Linux Alpha, etc.
2168 *) New COMP library [crypto/comp/] for SSL Record Layer Compression:
2169 RLE (dummy implemented) and ZLIB (really implemented when ZLIB is
2173 *) Add -strparse option to asn1pars program which parses nested
2175 [Dr Stephen Henson <shenson@bigfoot.com>]
2177 *) Added "oid_file" to ssleay.cnf for "ca" and "req" programs.
2180 *) DSA fix for "ca" program.
2183 *) Added "-genkey" option to "dsaparam" program.
2186 *) Added RIPE MD160 (rmd160) message digest.
2189 *) Added -a (all) option to "ssleay version" command.
2192 *) Added PLATFORM define which is the id given to Configure.
2195 *) Added MemCheck_XXXX functions to crypto/mem.c for memory checking.
2198 *) Extended the ASN.1 parser routines.
2201 *) Extended BIO routines to support REUSEADDR, seek, tell, etc.
2204 *) Added a BN_CTX to the BN library.
2207 *) Fixed the weak key values in DES library
2210 *) Changed API in EVP library for cipher aliases.
2213 *) Added support for RC2/64bit cipher.
2216 *) Converted the lhash library to the crypto/mem.c functions.
2219 *) Added more recognized ASN.1 object ids.
2222 *) Added more RSA padding checks for SSL/TLS.
2225 *) Added BIO proxy/filter functionality.
2228 *) Added extra_certs to SSL_CTX which can be used
2229 send extra CA certificates to the client in the CA cert chain sending
2230 process. It can be configured with SSL_CTX_add_extra_chain_cert().
2233 *) Now Fortezza is denied in the authentication phase because
2234 this is key exchange mechanism is not supported by SSLeay at all.
2237 *) Additional PKCS1 checks.
2240 *) Support the string "TLSv1" for all TLS v1 ciphers.
2243 *) Added function SSL_get_ex_data_X509_STORE_CTX_idx() which gives the
2244 ex_data index of the SSL context in the X509_STORE_CTX ex_data.
2247 *) Fixed a few memory leaks.
2250 *) Fixed various code and comment typos.
2253 *) A minor bug in ssl/s3_clnt.c where there would always be 4 0
2254 bytes sent in the client random.
2255 [Edward Bishop <ebishop@spyglass.com>]