CHANGES revision 120631
155714Skris 255714Skris OpenSSL CHANGES 355714Skris _______________ 455714Skris 5167612Ssimon Changes between 0.9.7b and 0.9.7c [30 Sep 2003] 6167612Ssimon 7167612Ssimon *) Fix various bugs revealed by running the NISCC test suite: 8167612Ssimon 9167612Ssimon Stop out of bounds reads in the ASN1 code when presented with 10167612Ssimon invalid tags (CAN-2003-0543 and CAN-2003-0544). 11167612Ssimon 12167612Ssimon Free up ASN1_TYPE correctly if ANY type is invalid (CAN-2003-0545). 13167612Ssimon 14167612Ssimon If verify callback ignores invalid public key errors don't try to check 15167612Ssimon certificate signature with the NULL public key. 16167612Ssimon 17167612Ssimon [Steve Henson] 18167612Ssimon 19167612Ssimon *) New -ignore_err option in ocsp application to stop the server 20167612Ssimon exiting on the first error in a request. 21167612Ssimon [Steve Henson] 22167612Ssimon 23167612Ssimon *) In ssl3_accept() (ssl/s3_srvr.c) only accept a client certificate 24167612Ssimon if the server requested one: as stated in TLS 1.0 and SSL 3.0 25167612Ssimon specifications. 26167612Ssimon [Steve Henson] 27167612Ssimon 28167612Ssimon *) In ssl3_get_client_hello() (ssl/s3_srvr.c), tolerate additional 29167612Ssimon extra data after the compression methods not only for TLS 1.0 30167612Ssimon but also for SSL 3.0 (as required by the specification). 31167612Ssimon [Bodo Moeller; problem pointed out by Matthias Loepfe] 32167612Ssimon 33167612Ssimon *) Change X509_certificate_type() to mark the key as exported/exportable 34167612Ssimon when it's 512 *bits* long, not 512 bytes. 35167612Ssimon [Richard Levitte] 36167612Ssimon 37167612Ssimon *) Change AES_cbc_encrypt() so it outputs exact multiple of 38167612Ssimon blocks during encryption. 39167612Ssimon [Richard Levitte] 40167612Ssimon 41167612Ssimon *) Various fixes to base64 BIO and non blocking I/O. On write 42167612Ssimon flushes were not handled properly if the BIO retried. On read 43167612Ssimon data was not being buffered properly and had various logic bugs. 44167612Ssimon This also affects blocking I/O when the data being decoded is a 45167612Ssimon certain size. 46162911Ssimon [Steve Henson] 47162911Ssimon 48162911Ssimon *) Various S/MIME bugfixes and compatibility changes: 49162911Ssimon output correct application/pkcs7 MIME type if 50162911Ssimon PKCS7_NOOLDMIMETYPE is set. Tolerate some broken signatures. 51162911Ssimon Output CR+LF for EOL if PKCS7_CRLFEOL is set (this makes opening 52162911Ssimon of files as .eml work). Correctly handle very long lines in MIME 53162911Ssimon parser. 54162911Ssimon [Steve Henson] 55162911Ssimon 56162911Ssimon Changes between 0.9.7a and 0.9.7b [10 Apr 2003] 57162911Ssimon 58162911Ssimon *) Countermeasure against the Klima-Pokorny-Rosa extension of 59162911Ssimon Bleichbacher's attack on PKCS #1 v1.5 padding: treat 60162911Ssimon a protocol version number mismatch like a decryption error 61162911Ssimon in ssl3_get_client_key_exchange (ssl/s3_srvr.c). 62162911Ssimon [Bodo Moeller] 63162911Ssimon 64162911Ssimon *) Turn on RSA blinding by default in the default implementation 65162911Ssimon to avoid a timing attack. Applications that don't want it can call 66162911Ssimon RSA_blinding_off() or use the new flag RSA_FLAG_NO_BLINDING. 67162911Ssimon They would be ill-advised to do so in most cases. 68162911Ssimon [Ben Laurie, Steve Henson, Geoff Thorpe, Bodo Moeller] 69162911Ssimon 70162911Ssimon *) Change RSA blinding code so that it works when the PRNG is not 71162911Ssimon seeded (in this case, the secret RSA exponent is abused as 72162911Ssimon an unpredictable seed -- if it is not unpredictable, there 73162911Ssimon is no point in blinding anyway). Make RSA blinding thread-safe 74162911Ssimon by remembering the creator's thread ID in rsa->blinding and 75162911Ssimon having all other threads use local one-time blinding factors 76162911Ssimon (this requires more computation than sharing rsa->blinding, but 77162911Ssimon avoids excessive locking; and if an RSA object is not shared 78162911Ssimon between threads, blinding will still be very fast). 79162911Ssimon [Bodo Moeller] 80162911Ssimon 81162911Ssimon *) Fixed a typo bug that would cause ENGINE_set_default() to set an 82162911Ssimon ENGINE as defaults for all supported algorithms irrespective of 83162911Ssimon the 'flags' parameter. 'flags' is now honoured, so applications 84162911Ssimon should make sure they are passing it correctly. 85162911Ssimon [Geoff Thorpe] 86162911Ssimon 87162911Ssimon *) Target "mingw" now allows native Windows code to be generated in 88162911Ssimon the Cygwin environment as well as with the MinGW compiler. 89162911Ssimon [Ulf Moeller] 90162911Ssimon 91162911Ssimon Changes between 0.9.7 and 0.9.7a [19 Feb 2003] 92162911Ssimon 93162911Ssimon *) In ssl3_get_record (ssl/s3_pkt.c), minimize information leaked 94162911Ssimon via timing by performing a MAC computation even if incorrrect 95162911Ssimon block cipher padding has been found. This is a countermeasure 96162911Ssimon against active attacks where the attacker has to distinguish 97162911Ssimon between bad padding and a MAC verification error. (CAN-2003-0078) 98162911Ssimon 99162911Ssimon [Bodo Moeller; problem pointed out by Brice Canvel (EPFL), 100162911Ssimon Alain Hiltgen (UBS), Serge Vaudenay (EPFL), and 101162911Ssimon Martin Vuagnoux (EPFL, Ilion)] 102162911Ssimon 103162911Ssimon *) Make the no-err option work as intended. The intention with no-err 104162911Ssimon is not to have the whole error stack handling routines removed from 105162911Ssimon libcrypto, it's only intended to remove all the function name and 106162911Ssimon reason texts, thereby removing some of the footprint that may not 107162911Ssimon be interesting if those errors aren't displayed anyway. 108162911Ssimon 109162911Ssimon NOTE: it's still possible for any application or module to have it's 110162911Ssimon own set of error texts inserted. The routines are there, just not 111162911Ssimon used by default when no-err is given. 112162911Ssimon [Richard Levitte] 113162911Ssimon 114162911Ssimon *) Add support for FreeBSD on IA64. 115162911Ssimon [dirk.meyer@dinoex.sub.org via Richard Levitte, resolves #454] 116162911Ssimon 117162911Ssimon *) Adjust DES_cbc_cksum() so it returns the same value as the MIT 118162911Ssimon Kerberos function mit_des_cbc_cksum(). Before this change, 119162911Ssimon the value returned by DES_cbc_cksum() was like the one from 120162911Ssimon mit_des_cbc_cksum(), except the bytes were swapped. 121167612Ssimon [Kevin Greaney <Kevin.Greaney@hp.com> and Richard Levitte] 122162911Ssimon 123162911Ssimon *) Allow an application to disable the automatic SSL chain building. 124162911Ssimon Before this a rather primitive chain build was always performed in 125162911Ssimon ssl3_output_cert_chain(): an application had no way to send the 126162911Ssimon correct chain if the automatic operation produced an incorrect result. 127162911Ssimon 128162911Ssimon Now the chain builder is disabled if either: 129162911Ssimon 130162911Ssimon 1. Extra certificates are added via SSL_CTX_add_extra_chain_cert(). 131162911Ssimon 132162911Ssimon 2. The mode flag SSL_MODE_NO_AUTO_CHAIN is set. 133162911Ssimon 134162911Ssimon The reasoning behind this is that an application would not want the 135162911Ssimon auto chain building to take place if extra chain certificates are 136162911Ssimon present and it might also want a means of sending no additional 137162911Ssimon certificates (for example the chain has two certificates and the 138162911Ssimon root is omitted). 139162911Ssimon [Steve Henson] 140162911Ssimon 141162911Ssimon *) Add the possibility to build without the ENGINE framework. 142162911Ssimon [Steven Reddie <smr@essemer.com.au> via Richard Levitte] 143162911Ssimon 144162911Ssimon *) Under Win32 gmtime() can return NULL: check return value in 145162911Ssimon OPENSSL_gmtime(). Add error code for case where gmtime() fails. 146162911Ssimon [Steve Henson] 147162911Ssimon 148160814Ssimon *) DSA routines: under certain error conditions uninitialized BN objects 149160814Ssimon could be freed. Solution: make sure initialization is performed early 150160814Ssimon enough. (Reported and fix supplied by Ivan D Nestlerode <nestler@MIT.EDU>, 151160814Ssimon Nils Larsch <nla@trustcenter.de> via PR#459) 152160814Ssimon [Lutz Jaenicke] 153160814Ssimon 154160814Ssimon *) Another fix for SSLv2 session ID handling: the session ID was incorrectly 155160814Ssimon checked on reconnect on the client side, therefore session resumption 156160814Ssimon could still fail with a "ssl session id is different" error. This 157160814Ssimon behaviour is masked when SSL_OP_ALL is used due to 158160814Ssimon SSL_OP_MICROSOFT_SESS_ID_BUG being set. 159160814Ssimon Behaviour observed by Crispin Flowerday <crispin@flowerday.cx> as 160160814Ssimon followup to PR #377. 161160814Ssimon [Lutz Jaenicke] 162160814Ssimon 163160814Ssimon *) IA-32 assembler support enhancements: unified ELF targets, support 164160814Ssimon for SCO/Caldera platforms, fix for Cygwin shared build. 165160814Ssimon [Andy Polyakov] 166160814Ssimon 167160814Ssimon *) Add support for FreeBSD on sparc64. As a consequence, support for 168160814Ssimon FreeBSD on non-x86 processors is separate from x86 processors on 169160814Ssimon the config script, much like the NetBSD support. 170160814Ssimon [Richard Levitte & Kris Kennaway <kris@obsecurity.org>] 171160814Ssimon 172160814Ssimon Changes between 0.9.6h and 0.9.7 [31 Dec 2002] 173160814Ssimon 174160814Ssimon [NB: OpenSSL 0.9.6i and later 0.9.6 patch levels were released after 175160814Ssimon OpenSSL 0.9.7.] 176160814Ssimon 177160814Ssimon *) Fix session ID handling in SSLv2 client code: the SERVER FINISHED 178160814Ssimon code (06) was taken as the first octet of the session ID and the last 179160814Ssimon octet was ignored consequently. As a result SSLv2 client side session 180160814Ssimon caching could not have worked due to the session ID mismatch between 181160814Ssimon client and server. 182160814Ssimon Behaviour observed by Crispin Flowerday <crispin@flowerday.cx> as 183160814Ssimon PR #377. 184160814Ssimon [Lutz Jaenicke] 185160814Ssimon 186160814Ssimon *) Change the declaration of needed Kerberos libraries to use EX_LIBS 187160814Ssimon instead of the special (and badly supported) LIBKRB5. LIBKRB5 is 188160814Ssimon removed entirely. 189160814Ssimon [Richard Levitte] 190160814Ssimon 191160814Ssimon *) The hw_ncipher.c engine requires dynamic locks. Unfortunately, it 192160814Ssimon seems that in spite of existing for more than a year, many application 193160814Ssimon author have done nothing to provide the necessary callbacks, which 194160814Ssimon means that this particular engine will not work properly anywhere. 195160814Ssimon This is a very unfortunate situation which forces us, in the name 196160814Ssimon of usability, to give the hw_ncipher.c a static lock, which is part 197160814Ssimon of libcrypto. 198160814Ssimon NOTE: This is for the 0.9.7 series ONLY. This hack will never 199160814Ssimon appear in 0.9.8 or later. We EXPECT application authors to have 200160814Ssimon dealt properly with this when 0.9.8 is released (unless we actually 201160814Ssimon make such changes in the libcrypto locking code that changes will 202160814Ssimon have to be made anyway). 203160814Ssimon [Richard Levitte] 204160814Ssimon 205160814Ssimon *) In asn1_d2i_read_bio() repeatedly call BIO_read() until all content 206160814Ssimon octets have been read, EOF or an error occurs. Without this change 207160814Ssimon some truncated ASN1 structures will not produce an error. 208160814Ssimon [Steve Henson] 209160814Ssimon 210160814Ssimon *) Disable Heimdal support, since it hasn't been fully implemented. 211160814Ssimon Still give the possibility to force the use of Heimdal, but with 212160814Ssimon warnings and a request that patches get sent to openssl-dev. 213160814Ssimon [Richard Levitte] 214160814Ssimon 215160814Ssimon *) Add the VC-CE target, introduce the WINCE sysname, and add 216160814Ssimon INSTALL.WCE and appropriate conditionals to make it build. 217160814Ssimon [Steven Reddie <smr@essemer.com.au> via Richard Levitte] 218160814Ssimon 219160814Ssimon *) Change the DLL names for Cygwin to cygcrypto-x.y.z.dll and 220160814Ssimon cygssl-x.y.z.dll, where x, y and z are the major, minor and 221160814Ssimon edit numbers of the version. 222160814Ssimon [Corinna Vinschen <vinschen@redhat.com> and Richard Levitte] 223160814Ssimon 224160814Ssimon *) Introduce safe string copy and catenation functions 225160814Ssimon (BUF_strlcpy() and BUF_strlcat()). 226160814Ssimon [Ben Laurie (CHATS) and Richard Levitte] 227160814Ssimon 228160814Ssimon *) Avoid using fixed-size buffers for one-line DNs. 229160814Ssimon [Ben Laurie (CHATS)] 230162911Ssimon 231162911Ssimon *) Add BUF_MEM_grow_clean() to avoid information leakage when 232162911Ssimon resizing buffers containing secrets, and use where appropriate. 233160814Ssimon [Ben Laurie (CHATS)] 234160814Ssimon 235160814Ssimon *) Avoid using fixed size buffers for configuration file location. 236160814Ssimon [Ben Laurie (CHATS)] 237160814Ssimon 238160814Ssimon *) Avoid filename truncation for various CA files. 239160814Ssimon [Ben Laurie (CHATS)] 240160814Ssimon 241160814Ssimon *) Use sizeof in preference to magic numbers. 242160814Ssimon [Ben Laurie (CHATS)] 243160814Ssimon 244160814Ssimon *) Avoid filename truncation in cert requests. 245160814Ssimon [Ben Laurie (CHATS)] 246160814Ssimon 247160814Ssimon *) Add assertions to check for (supposedly impossible) buffer 248160814Ssimon overflows. 249160814Ssimon [Ben Laurie (CHATS)] 250160814Ssimon 251160814Ssimon *) Don't cache truncated DNS entries in the local cache (this could 252160814Ssimon potentially lead to a spoofing attack). 253160814Ssimon [Ben Laurie (CHATS)] 254160814Ssimon 255160814Ssimon *) Fix various buffers to be large enough for hex/decimal 256160814Ssimon representations in a platform independent manner. 257160814Ssimon [Ben Laurie (CHATS)] 258160814Ssimon 259160814Ssimon *) Add CRYPTO_realloc_clean() to avoid information leakage when 260160814Ssimon resizing buffers containing secrets, and use where appropriate. 261160814Ssimon [Ben Laurie (CHATS)] 262160814Ssimon 263160814Ssimon *) Add BIO_indent() to avoid much slightly worrying code to do 264160814Ssimon indents. 265160814Ssimon [Ben Laurie (CHATS)] 266160814Ssimon 267160814Ssimon *) Convert sprintf()/BIO_puts() to BIO_printf(). 268160814Ssimon [Ben Laurie (CHATS)] 269160814Ssimon 270160814Ssimon *) buffer_gets() could terminate with the buffer only half 271160814Ssimon full. Fixed. 272160814Ssimon [Ben Laurie (CHATS)] 273160814Ssimon 274160814Ssimon *) Add assertions to prevent user-supplied crypto functions from 275160814Ssimon overflowing internal buffers by having large block sizes, etc. 276160814Ssimon [Ben Laurie (CHATS)] 277160814Ssimon 278160814Ssimon *) New OPENSSL_assert() macro (similar to assert(), but enabled 279160814Ssimon unconditionally). 280160814Ssimon [Ben Laurie (CHATS)] 281160814Ssimon 282160814Ssimon *) Eliminate unused copy of key in RC4. 283160814Ssimon [Ben Laurie (CHATS)] 284160814Ssimon 285160814Ssimon *) Eliminate unused and incorrectly sized buffers for IV in pem.h. 286160814Ssimon [Ben Laurie (CHATS)] 287160814Ssimon 288160814Ssimon *) Fix off-by-one error in EGD path. 289160814Ssimon [Ben Laurie (CHATS)] 290160814Ssimon 291160814Ssimon *) If RANDFILE path is too long, ignore instead of truncating. 292160814Ssimon [Ben Laurie (CHATS)] 293160814Ssimon 294160814Ssimon *) Eliminate unused and incorrectly sized X.509 structure 295160814Ssimon CBCParameter. 296160814Ssimon [Ben Laurie (CHATS)] 297160814Ssimon 298160814Ssimon *) Eliminate unused and dangerous function knumber(). 299160814Ssimon [Ben Laurie (CHATS)] 300160814Ssimon 301160814Ssimon *) Eliminate unused and dangerous structure, KSSL_ERR. 302160814Ssimon [Ben Laurie (CHATS)] 303160814Ssimon 304160814Ssimon *) Protect against overlong session ID context length in an encoded 305160814Ssimon session object. Since these are local, this does not appear to be 306160814Ssimon exploitable. 307160814Ssimon [Ben Laurie (CHATS)] 308160814Ssimon 309160814Ssimon *) Change from security patch (see 0.9.6e below) that did not affect 310160814Ssimon the 0.9.6 release series: 311160814Ssimon 312160814Ssimon Remote buffer overflow in SSL3 protocol - an attacker could 313160814Ssimon supply an oversized master key in Kerberos-enabled versions. 314160814Ssimon (CAN-2002-0657) 315160814Ssimon [Ben Laurie (CHATS)] 316160814Ssimon 317160814Ssimon *) Change the SSL kerb5 codes to match RFC 2712. 318160814Ssimon [Richard Levitte] 319160814Ssimon 320160814Ssimon *) Make -nameopt work fully for req and add -reqopt switch. 321160814Ssimon [Michael Bell <michael.bell@rz.hu-berlin.de>, Steve Henson] 322160814Ssimon 323160814Ssimon *) The "block size" for block ciphers in CFB and OFB mode should be 1. 324160814Ssimon [Steve Henson, reported by Yngve Nysaeter Pettersen <yngve@opera.com>] 325160814Ssimon 326160814Ssimon *) Make sure tests can be performed even if the corresponding algorithms 327160814Ssimon have been removed entirely. This was also the last step to make 328160814Ssimon OpenSSL compilable with DJGPP under all reasonable conditions. 329160814Ssimon [Richard Levitte, Doug Kaufman <dkaufman@rahul.net>] 330160814Ssimon 331160814Ssimon *) Add cipher selection rules COMPLEMENTOFALL and COMPLEMENTOFDEFAULT 332160814Ssimon to allow version independent disabling of normally unselected ciphers, 333160814Ssimon which may be activated as a side-effect of selecting a single cipher. 334160814Ssimon 335160814Ssimon (E.g., cipher list string "RSA" enables ciphersuites that are left 336160814Ssimon out of "ALL" because they do not provide symmetric encryption. 337160814Ssimon "RSA:!COMPLEMEMENTOFALL" avoids these unsafe ciphersuites.) 338160814Ssimon [Lutz Jaenicke, Bodo Moeller] 339160814Ssimon 340160814Ssimon *) Add appropriate support for separate platform-dependent build 341160814Ssimon directories. The recommended way to make a platform-dependent 342160814Ssimon build directory is the following (tested on Linux), maybe with 343160814Ssimon some local tweaks: 344160814Ssimon 345160814Ssimon # Place yourself outside of the OpenSSL source tree. In 346160814Ssimon # this example, the environment variable OPENSSL_SOURCE 347160814Ssimon # is assumed to contain the absolute OpenSSL source directory. 348160814Ssimon mkdir -p objtree/"`uname -s`-`uname -r`-`uname -m`" 349160814Ssimon cd objtree/"`uname -s`-`uname -r`-`uname -m`" 350160814Ssimon (cd $OPENSSL_SOURCE; find . -type f) | while read F; do 351160814Ssimon mkdir -p `dirname $F` 352160814Ssimon ln -s $OPENSSL_SOURCE/$F $F 353160814Ssimon done 354160814Ssimon 355160814Ssimon To be absolutely sure not to disturb the source tree, a "make clean" 356160814Ssimon is a good thing. If it isn't successfull, don't worry about it, 357160814Ssimon it probably means the source directory is very clean. 358160814Ssimon [Richard Levitte] 359160814Ssimon 360160814Ssimon *) Make sure any ENGINE control commands make local copies of string 361160814Ssimon pointers passed to them whenever necessary. Otherwise it is possible 362160814Ssimon the caller may have overwritten (or deallocated) the original string 363160814Ssimon data when a later ENGINE operation tries to use the stored values. 364160814Ssimon [G�tz Babin-Ebell <babinebell@trustcenter.de>] 365160814Ssimon 366160814Ssimon *) Improve diagnostics in file reading and command-line digests. 367160814Ssimon [Ben Laurie aided and abetted by Solar Designer <solar@openwall.com>] 368160814Ssimon 369160814Ssimon *) Add AES modes CFB and OFB to the object database. Correct an 370160814Ssimon error in AES-CFB decryption. 371160814Ssimon [Richard Levitte] 372160814Ssimon 373160814Ssimon *) Remove most calls to EVP_CIPHER_CTX_cleanup() in evp_enc.c, this 374160814Ssimon allows existing EVP_CIPHER_CTX structures to be reused after 375160814Ssimon calling EVP_*Final(). This behaviour is used by encryption 376160814Ssimon BIOs and some applications. This has the side effect that 377160814Ssimon applications must explicitly clean up cipher contexts with 378160814Ssimon EVP_CIPHER_CTX_cleanup() or they will leak memory. 379160814Ssimon [Steve Henson] 380160814Ssimon 381160814Ssimon *) Check the values of dna and dnb in bn_mul_recursive before calling 382160814Ssimon bn_mul_comba (a non zero value means the a or b arrays do not contain 383160814Ssimon n2 elements) and fallback to bn_mul_normal if either is not zero. 384160814Ssimon [Steve Henson] 385160814Ssimon 386160814Ssimon *) Fix escaping of non-ASCII characters when using the -subj option 387160814Ssimon of the "openssl req" command line tool. (Robert Joop <joop@fokus.gmd.de>) 388160814Ssimon [Lutz Jaenicke] 389160814Ssimon 390160814Ssimon *) Make object definitions compliant to LDAP (RFC2256): SN is the short 391160814Ssimon form for "surname", serialNumber has no short form. 392160814Ssimon Use "mail" as the short name for "rfc822Mailbox" according to RFC2798; 393160814Ssimon therefore remove "mail" short name for "internet 7". 394160814Ssimon The OID for unique identifiers in X509 certificates is 395160814Ssimon x500UniqueIdentifier, not uniqueIdentifier. 396160814Ssimon Some more OID additions. (Michael Bell <michael.bell@rz.hu-berlin.de>) 397160814Ssimon [Lutz Jaenicke] 398160814Ssimon 399160814Ssimon *) Add an "init" command to the ENGINE config module and auto initialize 400160814Ssimon ENGINEs. Without any "init" command the ENGINE will be initialized 401160814Ssimon after all ctrl commands have been executed on it. If init=1 the 402160814Ssimon ENGINE is initailized at that point (ctrls before that point are run 403160814Ssimon on the uninitialized ENGINE and after on the initialized one). If 404160814Ssimon init=0 then the ENGINE will not be iniatialized at all. 405160814Ssimon [Steve Henson] 406160814Ssimon 407160814Ssimon *) Fix the 'app_verify_callback' interface so that the user-defined 408160814Ssimon argument is actually passed to the callback: In the 409160814Ssimon SSL_CTX_set_cert_verify_callback() prototype, the callback 410160814Ssimon declaration has been changed from 411160814Ssimon int (*cb)() 412160814Ssimon into 413160814Ssimon int (*cb)(X509_STORE_CTX *,void *); 414160814Ssimon in ssl_verify_cert_chain (ssl/ssl_cert.c), the call 415160814Ssimon i=s->ctx->app_verify_callback(&ctx) 416160814Ssimon has been changed into 417160814Ssimon i=s->ctx->app_verify_callback(&ctx, s->ctx->app_verify_arg). 418160814Ssimon 419160814Ssimon To update applications using SSL_CTX_set_cert_verify_callback(), 420160814Ssimon a dummy argument can be added to their callback functions. 421160814Ssimon [D. K. Smetters <smetters@parc.xerox.com>] 422160814Ssimon 423160814Ssimon *) Added the '4758cca' ENGINE to support IBM 4758 cards. 424160814Ssimon [Maurice Gittens <maurice@gittens.nl>, touchups by Geoff Thorpe] 425160814Ssimon 426160814Ssimon *) Add and OPENSSL_LOAD_CONF define which will cause 427160814Ssimon OpenSSL_add_all_algorithms() to load the openssl.cnf config file. 428160814Ssimon This allows older applications to transparently support certain 429160814Ssimon OpenSSL features: such as crypto acceleration and dynamic ENGINE loading. 430160814Ssimon Two new functions OPENSSL_add_all_algorithms_noconf() which will never 431160814Ssimon load the config file and OPENSSL_add_all_algorithms_conf() which will 432160814Ssimon always load it have also been added. 433160814Ssimon [Steve Henson] 434160814Ssimon 435160814Ssimon *) Add the OFB, CFB and CTR (all with 128 bit feedback) to AES. 436160814Ssimon Adjust NIDs and EVP layer. 437160814Ssimon [Stephen Sprunk <stephen@sprunk.org> and Richard Levitte] 438160814Ssimon 439160814Ssimon *) Config modules support in openssl utility. 440160814Ssimon 441160814Ssimon Most commands now load modules from the config file, 442160814Ssimon though in a few (such as version) this isn't done 443160814Ssimon because it couldn't be used for anything. 444160814Ssimon 445160814Ssimon In the case of ca and req the config file used is 446160814Ssimon the same as the utility itself: that is the -config 447160814Ssimon command line option can be used to specify an 448160814Ssimon alternative file. 449160814Ssimon [Steve Henson] 450160814Ssimon 451160814Ssimon *) Move default behaviour from OPENSSL_config(). If appname is NULL 452160814Ssimon use "openssl_conf" if filename is NULL use default openssl config file. 453160814Ssimon [Steve Henson] 454160814Ssimon 455160814Ssimon *) Add an argument to OPENSSL_config() to allow the use of an alternative 456160814Ssimon config section name. Add a new flag to tolerate a missing config file 457160814Ssimon and move code to CONF_modules_load_file(). 458160814Ssimon [Steve Henson] 459160814Ssimon 460160814Ssimon *) Support for crypto accelerator cards from Accelerated Encryption 461160814Ssimon Processing, www.aep.ie. (Use engine 'aep') 462160814Ssimon The support was copied from 0.9.6c [engine] and adapted/corrected 463160814Ssimon to work with the new engine framework. 464160814Ssimon [AEP Inc. and Richard Levitte] 465160814Ssimon 466160814Ssimon *) Support for SureWare crypto accelerator cards from Baltimore 467160814Ssimon Technologies. (Use engine 'sureware') 468160814Ssimon The support was copied from 0.9.6c [engine] and adapted 469160814Ssimon to work with the new engine framework. 470160814Ssimon [Richard Levitte] 471160814Ssimon 472160814Ssimon *) Have the CHIL engine fork-safe (as defined by nCipher) and actually 473160814Ssimon make the newer ENGINE framework commands for the CHIL engine work. 474160814Ssimon [Toomas Kiisk <vix@cyber.ee> and Richard Levitte] 475160814Ssimon 476160814Ssimon *) Make it possible to produce shared libraries on ReliantUNIX. 477160814Ssimon [Robert Dahlem <Robert.Dahlem@ffm2.siemens.de> via Richard Levitte] 478160814Ssimon 479160814Ssimon *) Add the configuration target debug-linux-ppro. 480160814Ssimon Make 'openssl rsa' use the general key loading routines 481160814Ssimon implemented in apps.c, and make those routines able to 482160814Ssimon handle the key format FORMAT_NETSCAPE and the variant 483160814Ssimon FORMAT_IISSGC. 484160814Ssimon [Toomas Kiisk <vix@cyber.ee> via Richard Levitte] 485160814Ssimon 486160814Ssimon *) Fix a crashbug and a logic bug in hwcrhk_load_pubkey(). 487160814Ssimon [Toomas Kiisk <vix@cyber.ee> via Richard Levitte] 488160814Ssimon 489160814Ssimon *) Add -keyform to rsautl, and document -engine. 490160814Ssimon [Richard Levitte, inspired by Toomas Kiisk <vix@cyber.ee>] 491160814Ssimon 492160814Ssimon *) Change BIO_new_file (crypto/bio/bss_file.c) to use new 493160814Ssimon BIO_R_NO_SUCH_FILE error code rather than the generic 494160814Ssimon ERR_R_SYS_LIB error code if fopen() fails with ENOENT. 495160814Ssimon [Ben Laurie] 496160814Ssimon 497160814Ssimon *) Add new functions 498160814Ssimon ERR_peek_last_error 499160814Ssimon ERR_peek_last_error_line 500160814Ssimon ERR_peek_last_error_line_data. 501160814Ssimon These are similar to 502160814Ssimon ERR_peek_error 503160814Ssimon ERR_peek_error_line 504160814Ssimon ERR_peek_error_line_data, 505160814Ssimon but report on the latest error recorded rather than the first one 506160814Ssimon still in the error queue. 507160814Ssimon [Ben Laurie, Bodo Moeller] 508160814Ssimon 509160814Ssimon *) default_algorithms option in ENGINE config module. This allows things 510160814Ssimon like: 511160814Ssimon default_algorithms = ALL 512160814Ssimon default_algorithms = RSA, DSA, RAND, CIPHERS, DIGESTS 513160814Ssimon [Steve Henson] 514160814Ssimon 515160814Ssimon *) Prelminary ENGINE config module. 516160814Ssimon [Steve Henson] 517160814Ssimon 518160814Ssimon *) New experimental application configuration code. 519160814Ssimon [Steve Henson] 520160814Ssimon 521160814Ssimon *) Change the AES code to follow the same name structure as all other 522160814Ssimon symmetric ciphers, and behave the same way. Move everything to 523160814Ssimon the directory crypto/aes, thereby obsoleting crypto/rijndael. 524160814Ssimon [Stephen Sprunk <stephen@sprunk.org> and Richard Levitte] 525160814Ssimon 526160814Ssimon *) SECURITY: remove unsafe setjmp/signal interaction from ui_openssl.c. 527160814Ssimon [Ben Laurie and Theo de Raadt] 528160814Ssimon 529160814Ssimon *) Add option to output public keys in req command. 530160814Ssimon [Massimiliano Pala madwolf@openca.org] 531160814Ssimon 532160814Ssimon *) Use wNAFs in EC_POINTs_mul() for improved efficiency 533160814Ssimon (up to about 10% better than before for P-192 and P-224). 534160814Ssimon [Bodo Moeller] 535160814Ssimon 536160814Ssimon *) New functions/macros 537160814Ssimon 538160814Ssimon SSL_CTX_set_msg_callback(ctx, cb) 539160814Ssimon SSL_CTX_set_msg_callback_arg(ctx, arg) 540160814Ssimon SSL_set_msg_callback(ssl, cb) 541160814Ssimon SSL_set_msg_callback_arg(ssl, arg) 542160814Ssimon 543160814Ssimon to request calling a callback function 544160814Ssimon 545160814Ssimon void cb(int write_p, int version, int content_type, 546160814Ssimon const void *buf, size_t len, SSL *ssl, void *arg) 547160814Ssimon 548160814Ssimon whenever a protocol message has been completely received 549160814Ssimon (write_p == 0) or sent (write_p == 1). Here 'version' is the 550160814Ssimon protocol version according to which the SSL library interprets 551160814Ssimon the current protocol message (SSL2_VERSION, SSL3_VERSION, or 552160814Ssimon TLS1_VERSION). 'content_type' is 0 in the case of SSL 2.0, or 553160814Ssimon the content type as defined in the SSL 3.0/TLS 1.0 protocol 554160814Ssimon specification (change_cipher_spec(20), alert(21), handshake(22)). 555160814Ssimon 'buf' and 'len' point to the actual message, 'ssl' to the 556160814Ssimon SSL object, and 'arg' is the application-defined value set by 557160814Ssimon SSL[_CTX]_set_msg_callback_arg(). 558160814Ssimon 559160814Ssimon 'openssl s_client' and 'openssl s_server' have new '-msg' options 560160814Ssimon to enable a callback that displays all protocol messages. 561160814Ssimon [Bodo Moeller] 562160814Ssimon 563160814Ssimon *) Change the shared library support so shared libraries are built as 564160814Ssimon soon as the corresponding static library is finished, and thereby get 565160814Ssimon openssl and the test programs linked against the shared library. 566160814Ssimon This still only happens when the keyword "shard" has been given to 567160814Ssimon the configuration scripts. 568160814Ssimon 569160814Ssimon NOTE: shared library support is still an experimental thing, and 570160814Ssimon backward binary compatibility is still not guaranteed. 571160814Ssimon ["Maciej W. Rozycki" <macro@ds2.pg.gda.pl> and Richard Levitte] 572160814Ssimon 573160814Ssimon *) Add support for Subject Information Access extension. 574160814Ssimon [Peter Sylvester <Peter.Sylvester@EdelWeb.fr>] 575160814Ssimon 576160814Ssimon *) Make BUF_MEM_grow() behaviour more consistent: Initialise to zero 577160814Ssimon additional bytes when new memory had to be allocated, not just 578160814Ssimon when reusing an existing buffer. 579160814Ssimon [Bodo Moeller] 580160814Ssimon 581160814Ssimon *) New command line and configuration option 'utf8' for the req command. 582160814Ssimon This allows field values to be specified as UTF8 strings. 583160814Ssimon [Steve Henson] 584160814Ssimon 585160814Ssimon *) Add -multi and -mr options to "openssl speed" - giving multiple parallel 586160814Ssimon runs for the former and machine-readable output for the latter. 587160814Ssimon [Ben Laurie] 588160814Ssimon 589160814Ssimon *) Add '-noemailDN' option to 'openssl ca'. This prevents inclusion 590160814Ssimon of the e-mail address in the DN (i.e., it will go into a certificate 591160814Ssimon extension only). The new configuration file option 'email_in_dn = no' 592160814Ssimon has the same effect. 593160814Ssimon [Massimiliano Pala madwolf@openca.org] 594160814Ssimon 595160814Ssimon *) Change all functions with names starting with des_ to be starting 596160814Ssimon with DES_ instead. Add wrappers that are compatible with libdes, 597160814Ssimon but are named _ossl_old_des_*. Finally, add macros that map the 598160814Ssimon des_* symbols to the corresponding _ossl_old_des_* if libdes 599160814Ssimon compatibility is desired. If OpenSSL 0.9.6c compatibility is 600160814Ssimon desired, the des_* symbols will be mapped to DES_*, with one 601160814Ssimon exception. 602160814Ssimon 603160814Ssimon Since we provide two compatibility mappings, the user needs to 604160814Ssimon define the macro OPENSSL_DES_LIBDES_COMPATIBILITY if libdes 605160814Ssimon compatibility is desired. The default (i.e., when that macro 606160814Ssimon isn't defined) is OpenSSL 0.9.6c compatibility. 607160814Ssimon 608160814Ssimon There are also macros that enable and disable the support of old 609160814Ssimon des functions altogether. Those are OPENSSL_ENABLE_OLD_DES_SUPPORT 610160814Ssimon and OPENSSL_DISABLE_OLD_DES_SUPPORT. If none or both of those 611160814Ssimon are defined, the default will apply: to support the old des routines. 612160814Ssimon 613160814Ssimon In either case, one must include openssl/des.h to get the correct 614160814Ssimon definitions. Do not try to just include openssl/des_old.h, that 615160814Ssimon won't work. 616160814Ssimon 617160814Ssimon NOTE: This is a major break of an old API into a new one. Software 618160814Ssimon authors are encouraged to switch to the DES_ style functions. Some 619160814Ssimon time in the future, des_old.h and the libdes compatibility functions 620160814Ssimon will be disable (i.e. OPENSSL_DISABLE_OLD_DES_SUPPORT will be the 621160814Ssimon default), and then completely removed. 622160814Ssimon [Richard Levitte] 623160814Ssimon 624160814Ssimon *) Test for certificates which contain unsupported critical extensions. 625160814Ssimon If such a certificate is found during a verify operation it is 626160814Ssimon rejected by default: this behaviour can be overridden by either 627160814Ssimon handling the new error X509_V_ERR_UNHANDLED_CRITICAL_EXTENSION or 628160814Ssimon by setting the verify flag X509_V_FLAG_IGNORE_CRITICAL. A new function 629160814Ssimon X509_supported_extension() has also been added which returns 1 if a 630160814Ssimon particular extension is supported. 631160814Ssimon [Steve Henson] 632160814Ssimon 633160814Ssimon *) Modify the behaviour of EVP cipher functions in similar way to digests 634160814Ssimon to retain compatibility with existing code. 635160814Ssimon [Steve Henson] 636160814Ssimon 637160814Ssimon *) Modify the behaviour of EVP_DigestInit() and EVP_DigestFinal() to retain 638160814Ssimon compatibility with existing code. In particular the 'ctx' parameter does 639160814Ssimon not have to be to be initialized before the call to EVP_DigestInit() and 640160814Ssimon it is tidied up after a call to EVP_DigestFinal(). New function 641160814Ssimon EVP_DigestFinal_ex() which does not tidy up the ctx. Similarly function 642160814Ssimon EVP_MD_CTX_copy() changed to not require the destination to be 643160814Ssimon initialized valid and new function EVP_MD_CTX_copy_ex() added which 644160814Ssimon requires the destination to be valid. 645160814Ssimon 646160814Ssimon Modify all the OpenSSL digest calls to use EVP_DigestInit_ex(), 647160814Ssimon EVP_DigestFinal_ex() and EVP_MD_CTX_copy_ex(). 648160814Ssimon [Steve Henson] 649160814Ssimon 650160814Ssimon *) Change ssl3_get_message (ssl/s3_both.c) and the functions using it 651160814Ssimon so that complete 'Handshake' protocol structures are kept in memory 652160814Ssimon instead of overwriting 'msg_type' and 'length' with 'body' data. 653160814Ssimon [Bodo Moeller] 654160814Ssimon 655160814Ssimon *) Add an implementation of SSL_add_dir_cert_subjects_to_stack for Win32. 656160814Ssimon [Massimo Santin via Richard Levitte] 657160814Ssimon 658160814Ssimon *) Major restructuring to the underlying ENGINE code. This includes 659160814Ssimon reduction of linker bloat, separation of pure "ENGINE" manipulation 660160814Ssimon (initialisation, etc) from functionality dealing with implementations 661160814Ssimon of specific crypto iterfaces. This change also introduces integrated 662160814Ssimon support for symmetric ciphers and digest implementations - so ENGINEs 663160814Ssimon can now accelerate these by providing EVP_CIPHER and EVP_MD 664160814Ssimon implementations of their own. This is detailed in crypto/engine/README 665160814Ssimon as it couldn't be adequately described here. However, there are a few 666160814Ssimon API changes worth noting - some RSA, DSA, DH, and RAND functions that 667160814Ssimon were changed in the original introduction of ENGINE code have now 668160814Ssimon reverted back - the hooking from this code to ENGINE is now a good 669160814Ssimon deal more passive and at run-time, operations deal directly with 670160814Ssimon RSA_METHODs, DSA_METHODs (etc) as they did before, rather than 671160814Ssimon dereferencing through an ENGINE pointer any more. Also, the ENGINE 672160814Ssimon functions dealing with BN_MOD_EXP[_CRT] handlers have been removed - 673160814Ssimon they were not being used by the framework as there is no concept of a 674160814Ssimon BIGNUM_METHOD and they could not be generalised to the new 675160814Ssimon 'ENGINE_TABLE' mechanism that underlies the new code. Similarly, 676160814Ssimon ENGINE_cpy() has been removed as it cannot be consistently defined in 677160814Ssimon the new code. 678160814Ssimon [Geoff Thorpe] 679160814Ssimon 680160814Ssimon *) Change ASN1_GENERALIZEDTIME_check() to allow fractional seconds. 681160814Ssimon [Steve Henson] 682160814Ssimon 683160814Ssimon *) Change mkdef.pl to sort symbols that get the same entry number, 684160814Ssimon and make sure the automatically generated functions ERR_load_* 685160814Ssimon become part of libeay.num as well. 686160814Ssimon [Richard Levitte] 687160814Ssimon 688160814Ssimon *) New function SSL_renegotiate_pending(). This returns true once 689160814Ssimon renegotiation has been requested (either SSL_renegotiate() call 690160814Ssimon or HelloRequest/ClientHello receveived from the peer) and becomes 691160814Ssimon false once a handshake has been completed. 692160814Ssimon (For servers, SSL_renegotiate() followed by SSL_do_handshake() 693160814Ssimon sends a HelloRequest, but does not ensure that a handshake takes 694160814Ssimon place. SSL_renegotiate_pending() is useful for checking if the 695160814Ssimon client has followed the request.) 696160814Ssimon [Bodo Moeller] 697160814Ssimon 698160814Ssimon *) New SSL option SSL_OP_NO_SESSION_RESUMPTION_ON_RENEGOTIATION. 699160814Ssimon By default, clients may request session resumption even during 700160814Ssimon renegotiation (if session ID contexts permit); with this option, 701160814Ssimon session resumption is possible only in the first handshake. 702160814Ssimon 703160814Ssimon SSL_OP_ALL is now 0x00000FFFL instead of 0x000FFFFFL. This makes 704160814Ssimon more bits available for options that should not be part of 705160814Ssimon SSL_OP_ALL (such as SSL_OP_NO_SESSION_RESUMPTION_ON_RENEGOTIATION). 706160814Ssimon [Bodo Moeller] 707160814Ssimon 708160814Ssimon *) Add some demos for certificate and certificate request creation. 709160814Ssimon [Steve Henson] 710160814Ssimon 711160814Ssimon *) Make maximum certificate chain size accepted from the peer application 712160814Ssimon settable (SSL*_get/set_max_cert_list()), as proposed by 713160814Ssimon "Douglas E. Engert" <deengert@anl.gov>. 714160814Ssimon [Lutz Jaenicke] 715160814Ssimon 716160814Ssimon *) Add support for shared libraries for Unixware-7 717160814Ssimon (Boyd Lynn Gerber <gerberb@zenez.com>). 718160814Ssimon [Lutz Jaenicke] 719160814Ssimon 720160814Ssimon *) Add a "destroy" handler to ENGINEs that allows structural cleanup to 721160814Ssimon be done prior to destruction. Use this to unload error strings from 722160814Ssimon ENGINEs that load their own error strings. NB: This adds two new API 723160814Ssimon functions to "get" and "set" this destroy handler in an ENGINE. 724160814Ssimon [Geoff Thorpe] 725160814Ssimon 726160814Ssimon *) Alter all existing ENGINE implementations (except "openssl" and 727160814Ssimon "openbsd") to dynamically instantiate their own error strings. This 728160814Ssimon makes them more flexible to be built both as statically-linked ENGINEs 729160814Ssimon and self-contained shared-libraries loadable via the "dynamic" ENGINE. 730160814Ssimon Also, add stub code to each that makes building them as self-contained 731160814Ssimon shared-libraries easier (see README.ENGINE). 732160814Ssimon [Geoff Thorpe] 733160814Ssimon 734160814Ssimon *) Add a "dynamic" ENGINE that provides a mechanism for binding ENGINE 735160814Ssimon implementations into applications that are completely implemented in 736160814Ssimon self-contained shared-libraries. The "dynamic" ENGINE exposes control 737160814Ssimon commands that can be used to configure what shared-library to load and 738160814Ssimon to control aspects of the way it is handled. Also, made an update to 739160814Ssimon the README.ENGINE file that brings its information up-to-date and 740160814Ssimon provides some information and instructions on the "dynamic" ENGINE 741160814Ssimon (ie. how to use it, how to build "dynamic"-loadable ENGINEs, etc). 742160814Ssimon [Geoff Thorpe] 743160814Ssimon 744160814Ssimon *) Make it possible to unload ranges of ERR strings with a new 745160814Ssimon "ERR_unload_strings" function. 746160814Ssimon [Geoff Thorpe] 747160814Ssimon 748160814Ssimon *) Add a copy() function to EVP_MD. 749160814Ssimon [Ben Laurie] 750160814Ssimon 751160814Ssimon *) Make EVP_MD routines take a context pointer instead of just the 752160814Ssimon md_data void pointer. 753160814Ssimon [Ben Laurie] 754160814Ssimon 755160814Ssimon *) Add flags to EVP_MD and EVP_MD_CTX. EVP_MD_FLAG_ONESHOT indicates 756160814Ssimon that the digest can only process a single chunk of data 757160814Ssimon (typically because it is provided by a piece of 758160814Ssimon hardware). EVP_MD_CTX_FLAG_ONESHOT indicates that the application 759160814Ssimon is only going to provide a single chunk of data, and hence the 760160814Ssimon framework needn't accumulate the data for oneshot drivers. 761160814Ssimon [Ben Laurie] 762160814Ssimon 763160814Ssimon *) As with "ERR", make it possible to replace the underlying "ex_data" 764160814Ssimon functions. This change also alters the storage and management of global 765160814Ssimon ex_data state - it's now all inside ex_data.c and all "class" code (eg. 766160814Ssimon RSA, BIO, SSL_CTX, etc) no longer stores its own STACKS and per-class 767160814Ssimon index counters. The API functions that use this state have been changed 768160814Ssimon to take a "class_index" rather than pointers to the class's local STACK 769160814Ssimon and counter, and there is now an API function to dynamically create new 770160814Ssimon classes. This centralisation allows us to (a) plug a lot of the 771160814Ssimon thread-safety problems that existed, and (b) makes it possible to clean 772160814Ssimon up all allocated state using "CRYPTO_cleanup_all_ex_data()". W.r.t. (b) 773160814Ssimon such data would previously have always leaked in application code and 774160814Ssimon workarounds were in place to make the memory debugging turn a blind eye 775160814Ssimon to it. Application code that doesn't use this new function will still 776160814Ssimon leak as before, but their memory debugging output will announce it now 777160814Ssimon rather than letting it slide. 778160814Ssimon 779160814Ssimon Besides the addition of CRYPTO_cleanup_all_ex_data(), another API change 780160814Ssimon induced by the "ex_data" overhaul is that X509_STORE_CTX_init() now 781160814Ssimon has a return value to indicate success or failure. 782160814Ssimon [Geoff Thorpe] 783160814Ssimon 784160814Ssimon *) Make it possible to replace the underlying "ERR" functions such that the 785160814Ssimon global state (2 LHASH tables and 2 locks) is only used by the "default" 786160814Ssimon implementation. This change also adds two functions to "get" and "set" 787160814Ssimon the implementation prior to it being automatically set the first time 788160814Ssimon any other ERR function takes place. Ie. an application can call "get", 789160814Ssimon pass the return value to a module it has just loaded, and that module 790160814Ssimon can call its own "set" function using that value. This means the 791160814Ssimon module's "ERR" operations will use (and modify) the error state in the 792160814Ssimon application and not in its own statically linked copy of OpenSSL code. 793160814Ssimon [Geoff Thorpe] 794160814Ssimon 795160814Ssimon *) Give DH, DSA, and RSA types their own "**_up_ref()" function to increment 796160814Ssimon reference counts. This performs normal REF_PRINT/REF_CHECK macros on 797160814Ssimon the operation, and provides a more encapsulated way for external code 798160814Ssimon (crypto/evp/ and ssl/) to do this. Also changed the evp and ssl code 799160814Ssimon to use these functions rather than manually incrementing the counts. 800160814Ssimon 801160814Ssimon Also rename "DSO_up()" function to more descriptive "DSO_up_ref()". 802160814Ssimon [Geoff Thorpe] 803160814Ssimon 804160814Ssimon *) Add EVP test program. 805160814Ssimon [Ben Laurie] 806160814Ssimon 807160814Ssimon *) Add symmetric cipher support to ENGINE. Expect the API to change! 808160814Ssimon [Ben Laurie] 809160814Ssimon 810160814Ssimon *) New CRL functions: X509_CRL_set_version(), X509_CRL_set_issuer_name() 811160814Ssimon X509_CRL_set_lastUpdate(), X509_CRL_set_nextUpdate(), X509_CRL_sort(), 812160814Ssimon X509_REVOKED_set_serialNumber(), and X509_REVOKED_set_revocationDate(). 813160814Ssimon These allow a CRL to be built without having to access X509_CRL fields 814160814Ssimon directly. Modify 'ca' application to use new functions. 815160814Ssimon [Steve Henson] 816160814Ssimon 817160814Ssimon *) Move SSL_OP_TLS_ROLLBACK_BUG out of the SSL_OP_ALL list of recommended 818160814Ssimon bug workarounds. Rollback attack detection is a security feature. 819160814Ssimon The problem will only arise on OpenSSL servers when TLSv1 is not 820160814Ssimon available (sslv3_server_method() or SSL_OP_NO_TLSv1). 821160814Ssimon Software authors not wanting to support TLSv1 will have special reasons 822160814Ssimon for their choice and can explicitly enable this option. 823160814Ssimon [Bodo Moeller, Lutz Jaenicke] 824160814Ssimon 825160814Ssimon *) Rationalise EVP so it can be extended: don't include a union of 826160814Ssimon cipher/digest structures, add init/cleanup functions for EVP_MD_CTX 827160814Ssimon (similar to those existing for EVP_CIPHER_CTX). 828160814Ssimon Usage example: 829160814Ssimon 830160814Ssimon EVP_MD_CTX md; 831160814Ssimon 832160814Ssimon EVP_MD_CTX_init(&md); /* new function call */ 833160814Ssimon EVP_DigestInit(&md, EVP_sha1()); 834160814Ssimon EVP_DigestUpdate(&md, in, len); 835160814Ssimon EVP_DigestFinal(&md, out, NULL); 836160814Ssimon EVP_MD_CTX_cleanup(&md); /* new function call */ 837160814Ssimon 838160814Ssimon [Ben Laurie] 839160814Ssimon 840160814Ssimon *) Make DES key schedule conform to the usual scheme, as well as 841160814Ssimon correcting its structure. This means that calls to DES functions 842160814Ssimon now have to pass a pointer to a des_key_schedule instead of a 843160814Ssimon plain des_key_schedule (which was actually always a pointer 844160814Ssimon anyway): E.g., 845160814Ssimon 846160814Ssimon des_key_schedule ks; 847160814Ssimon 848160814Ssimon des_set_key_checked(..., &ks); 849160814Ssimon des_ncbc_encrypt(..., &ks, ...); 850160814Ssimon 851160814Ssimon (Note that a later change renames 'des_...' into 'DES_...'.) 852160814Ssimon [Ben Laurie] 853160814Ssimon 854160814Ssimon *) Initial reduction of linker bloat: the use of some functions, such as 855160814Ssimon PEM causes large amounts of unused functions to be linked in due to 856160814Ssimon poor organisation. For example pem_all.c contains every PEM function 857160814Ssimon which has a knock on effect of linking in large amounts of (unused) 858160814Ssimon ASN1 code. Grouping together similar functions and splitting unrelated 859160814Ssimon functions prevents this. 860160814Ssimon [Steve Henson] 861160814Ssimon 862160814Ssimon *) Cleanup of EVP macros. 863160814Ssimon [Ben Laurie] 864160814Ssimon 865160814Ssimon *) Change historical references to {NID,SN,LN}_des_ede and ede3 to add the 866160814Ssimon correct _ecb suffix. 867160814Ssimon [Ben Laurie] 868160814Ssimon 869160814Ssimon *) Add initial OCSP responder support to ocsp application. The 870160814Ssimon revocation information is handled using the text based index 871160814Ssimon use by the ca application. The responder can either handle 872160814Ssimon requests generated internally, supplied in files (for example 873160814Ssimon via a CGI script) or using an internal minimal server. 874160814Ssimon [Steve Henson] 875160814Ssimon 876160814Ssimon *) Add configuration choices to get zlib compression for TLS. 877160814Ssimon [Richard Levitte] 878160814Ssimon 879160814Ssimon *) Changes to Kerberos SSL for RFC 2712 compliance: 880160814Ssimon 1. Implemented real KerberosWrapper, instead of just using 881160814Ssimon KRB5 AP_REQ message. [Thanks to Simon Wilkinson <sxw@sxw.org.uk>] 882160814Ssimon 2. Implemented optional authenticator field of KerberosWrapper. 883160814Ssimon 884160814Ssimon Added openssl-style ASN.1 macros for Kerberos ticket, ap_req, 885160814Ssimon and authenticator structs; see crypto/krb5/. 886160814Ssimon 887160814Ssimon Generalized Kerberos calls to support multiple Kerberos libraries. 888160814Ssimon [Vern Staats <staatsvr@asc.hpc.mil>, 889160814Ssimon Jeffrey Altman <jaltman@columbia.edu> 890160814Ssimon via Richard Levitte] 891160814Ssimon 892160814Ssimon *) Cause 'openssl speed' to use fully hard-coded DSA keys as it 893160814Ssimon already does with RSA. testdsa.h now has 'priv_key/pub_key' 894160814Ssimon values for each of the key sizes rather than having just 895160814Ssimon parameters (and 'speed' generating keys each time). 896160814Ssimon [Geoff Thorpe] 897160814Ssimon 898160814Ssimon *) Speed up EVP routines. 899160814Ssimon Before: 900160814Ssimonencrypt 901160814Ssimontype 8 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes 902160814Ssimondes-cbc 4408.85k 5560.51k 5778.46k 5862.20k 5825.16k 903160814Ssimondes-cbc 4389.55k 5571.17k 5792.23k 5846.91k 5832.11k 904160814Ssimondes-cbc 4394.32k 5575.92k 5807.44k 5848.37k 5841.30k 905160814Ssimondecrypt 906160814Ssimondes-cbc 3482.66k 5069.49k 5496.39k 5614.16k 5639.28k 907160814Ssimondes-cbc 3480.74k 5068.76k 5510.34k 5609.87k 5635.52k 908160814Ssimondes-cbc 3483.72k 5067.62k 5504.60k 5708.01k 5724.80k 909160814Ssimon After: 910160814Ssimonencrypt 911160814Ssimondes-cbc 4660.16k 5650.19k 5807.19k 5827.13k 5783.32k 912160814Ssimondecrypt 913160814Ssimondes-cbc 3624.96k 5258.21k 5530.91k 5624.30k 5628.26k 914160814Ssimon [Ben Laurie] 915160814Ssimon 916160814Ssimon *) Added the OS2-EMX target. 917160814Ssimon ["Brian Havard" <brianh@kheldar.apana.org.au> and Richard Levitte] 918160814Ssimon 919160814Ssimon *) Rewrite apps to use NCONF routines instead of the old CONF. New functions 920160814Ssimon to support NCONF routines in extension code. New function CONF_set_nconf() 921160814Ssimon to allow functions which take an NCONF to also handle the old LHASH 922160814Ssimon structure: this means that the old CONF compatible routines can be 923160814Ssimon retained (in particular wrt extensions) without having to duplicate the 924160814Ssimon code. New function X509V3_add_ext_nconf_sk to add extensions to a stack. 925160814Ssimon [Steve Henson] 926160814Ssimon 927160814Ssimon *) Enhance the general user interface with mechanisms for inner control 928160814Ssimon and with possibilities to have yes/no kind of prompts. 929160814Ssimon [Richard Levitte] 930160814Ssimon 931160814Ssimon *) Change all calls to low level digest routines in the library and 932160814Ssimon applications to use EVP. Add missing calls to HMAC_cleanup() and 933160814Ssimon don't assume HMAC_CTX can be copied using memcpy(). 934160814Ssimon [Verdon Walker <VWalker@novell.com>, Steve Henson] 935160814Ssimon 936160814Ssimon *) Add the possibility to control engines through control names but with 937160814Ssimon arbitrary arguments instead of just a string. 938160814Ssimon Change the key loaders to take a UI_METHOD instead of a callback 939160814Ssimon function pointer. NOTE: this breaks binary compatibility with earlier 940160814Ssimon versions of OpenSSL [engine]. 941160814Ssimon Adapt the nCipher code for these new conditions and add a card insertion 942160814Ssimon callback. 943160814Ssimon [Richard Levitte] 944160814Ssimon 945160814Ssimon *) Enhance the general user interface with mechanisms to better support 946160814Ssimon dialog box interfaces, application-defined prompts, the possibility 947160814Ssimon to use defaults (for example default passwords from somewhere else) 948160814Ssimon and interrupts/cancellations. 949160814Ssimon [Richard Levitte] 950160814Ssimon 951160814Ssimon *) Tidy up PKCS#12 attribute handling. Add support for the CSP name 952160814Ssimon attribute in PKCS#12 files, add new -CSP option to pkcs12 utility. 953160814Ssimon [Steve Henson] 954160814Ssimon 955160814Ssimon *) Fix a memory leak in 'sk_dup()' in the case reallocation fails. (Also 956160814Ssimon tidy up some unnecessarily weird code in 'sk_new()'). 957160814Ssimon [Geoff, reported by Diego Tartara <dtartara@novamens.com>] 958160814Ssimon 959160814Ssimon *) Change the key loading routines for ENGINEs to use the same kind 960160814Ssimon callback (pem_password_cb) as all other routines that need this 961160814Ssimon kind of callback. 962160814Ssimon [Richard Levitte] 963160814Ssimon 964160814Ssimon *) Increase ENTROPY_NEEDED to 32 bytes, as Rijndael can operate with 965160814Ssimon 256 bit (=32 byte) keys. Of course seeding with more entropy bytes 966160814Ssimon than this minimum value is recommended. 967160814Ssimon [Lutz Jaenicke] 968160814Ssimon 969160814Ssimon *) New random seeder for OpenVMS, using the system process statistics 970160814Ssimon that are easily reachable. 971160814Ssimon [Richard Levitte] 972160814Ssimon 973160814Ssimon *) Windows apparently can't transparently handle global 974160814Ssimon variables defined in DLLs. Initialisations such as: 975160814Ssimon 976160814Ssimon const ASN1_ITEM *it = &ASN1_INTEGER_it; 977160814Ssimon 978160814Ssimon wont compile. This is used by the any applications that need to 979160814Ssimon declare their own ASN1 modules. This was fixed by adding the option 980160814Ssimon EXPORT_VAR_AS_FN to all Win32 platforms, although this isn't strictly 981160814Ssimon needed for static libraries under Win32. 982160814Ssimon [Steve Henson] 983160814Ssimon 984160814Ssimon *) New functions X509_PURPOSE_set() and X509_TRUST_set() to handle 985160814Ssimon setting of purpose and trust fields. New X509_STORE trust and 986160814Ssimon purpose functions and tidy up setting in other SSL functions. 987160814Ssimon [Steve Henson] 988160814Ssimon 989160814Ssimon *) Add copies of X509_STORE_CTX fields and callbacks to X509_STORE 990160814Ssimon structure. These are inherited by X509_STORE_CTX when it is 991160814Ssimon initialised. This allows various defaults to be set in the 992160814Ssimon X509_STORE structure (such as flags for CRL checking and custom 993160814Ssimon purpose or trust settings) for functions which only use X509_STORE_CTX 994160814Ssimon internally such as S/MIME. 995160814Ssimon 996160814Ssimon Modify X509_STORE_CTX_purpose_inherit() so it only sets purposes and 997160814Ssimon trust settings if they are not set in X509_STORE. This allows X509_STORE 998160814Ssimon purposes and trust (in S/MIME for example) to override any set by default. 999160814Ssimon 1000160814Ssimon Add command line options for CRL checking to smime, s_client and s_server 1001160814Ssimon applications. 1002160814Ssimon [Steve Henson] 1003160814Ssimon 1004160814Ssimon *) Initial CRL based revocation checking. If the CRL checking flag(s) 1005160814Ssimon are set then the CRL is looked up in the X509_STORE structure and 1006160814Ssimon its validity and signature checked, then if the certificate is found 1007160814Ssimon in the CRL the verify fails with a revoked error. 1008160814Ssimon 1009160814Ssimon Various new CRL related callbacks added to X509_STORE_CTX structure. 1010160814Ssimon 1011160814Ssimon Command line options added to 'verify' application to support this. 1012160814Ssimon 1013160814Ssimon This needs some additional work, such as being able to handle multiple 1014160814Ssimon CRLs with different times, extension based lookup (rather than just 1015160814Ssimon by subject name) and ultimately more complete V2 CRL extension 1016160814Ssimon handling. 1017160814Ssimon [Steve Henson] 1018160814Ssimon 1019160814Ssimon *) Add a general user interface API (crypto/ui/). This is designed 1020160814Ssimon to replace things like des_read_password and friends (backward 1021160814Ssimon compatibility functions using this new API are provided). 1022160814Ssimon The purpose is to remove prompting functions from the DES code 1023160814Ssimon section as well as provide for prompting through dialog boxes in 1024160814Ssimon a window system and the like. 1025160814Ssimon [Richard Levitte] 1026160814Ssimon 1027160814Ssimon *) Add "ex_data" support to ENGINE so implementations can add state at a 1028160814Ssimon per-structure level rather than having to store it globally. 1029160814Ssimon [Geoff] 1030160814Ssimon 1031160814Ssimon *) Make it possible for ENGINE structures to be copied when retrieved by 1032160814Ssimon ENGINE_by_id() if the ENGINE specifies a new flag: ENGINE_FLAGS_BY_ID_COPY. 1033160814Ssimon This causes the "original" ENGINE structure to act like a template, 1034160814Ssimon analogous to the RSA vs. RSA_METHOD type of separation. Because of this 1035160814Ssimon operational state can be localised to each ENGINE structure, despite the 1036160814Ssimon fact they all share the same "methods". New ENGINE structures returned in 1037160814Ssimon this case have no functional references and the return value is the single 1038160814Ssimon structural reference. This matches the single structural reference returned 1039160814Ssimon by ENGINE_by_id() normally, when it is incremented on the pre-existing 1040160814Ssimon ENGINE structure. 1041160814Ssimon [Geoff] 1042160814Ssimon 1043160814Ssimon *) Fix ASN1 decoder when decoding type ANY and V_ASN1_OTHER: since this 1044160814Ssimon needs to match any other type at all we need to manually clear the 1045160814Ssimon tag cache. 1046160814Ssimon [Steve Henson] 1047160814Ssimon 1048160814Ssimon *) Changes to the "openssl engine" utility to include; 1049160814Ssimon - verbosity levels ('-v', '-vv', and '-vvv') that provide information 1050167612Ssimon about an ENGINE's available control commands. 1051162911Ssimon - executing control commands from command line arguments using the 1052167612Ssimon '-pre' and '-post' switches. '-post' is only used if '-t' is 1053167612Ssimon specified and the ENGINE is successfully initialised. The syntax for 1054167612Ssimon the individual commands are colon-separated, for example; 1055167612Ssimon openssl engine chil -pre FORK_CHECK:0 -pre SO_PATH:/lib/test.so 1056167612Ssimon [Geoff] 1057167612Ssimon 1058167612Ssimon *) New dynamic control command support for ENGINEs. ENGINEs can now 1059167612Ssimon declare their own commands (numbers), names (strings), descriptions, 1060167612Ssimon and input types for run-time discovery by calling applications. A 1061167612Ssimon subset of these commands are implicitly classed as "executable" 1062167612Ssimon depending on their input type, and only these can be invoked through 1063167612Ssimon the new string-based API function ENGINE_ctrl_cmd_string(). (Eg. this 1064167612Ssimon can be based on user input, config files, etc). The distinction is 1065167612Ssimon that "executable" commands cannot return anything other than a boolean 1066167612Ssimon result and can only support numeric or string input, whereas some 1067167612Ssimon discoverable commands may only be for direct use through 1068167612Ssimon ENGINE_ctrl(), eg. supporting the exchange of binary data, function 1069167612Ssimon pointers, or other custom uses. The "executable" commands are to 1070167612Ssimon support parameterisations of ENGINE behaviour that can be 1071167612Ssimon unambiguously defined by ENGINEs and used consistently across any 1072167612Ssimon OpenSSL-based application. Commands have been added to all the 1073167612Ssimon existing hardware-supporting ENGINEs, noticeably "SO_PATH" to allow 1074167612Ssimon control over shared-library paths without source code alterations. 1075167612Ssimon [Geoff] 1076167612Ssimon 1077167612Ssimon *) Changed all ENGINE implementations to dynamically allocate their 1078167612Ssimon ENGINEs rather than declaring them statically. Apart from this being 1079167612Ssimon necessary with the removal of the ENGINE_FLAGS_MALLOCED distinction, 1080167612Ssimon this also allows the implementations to compile without using the 1081167612Ssimon internal engine_int.h header. 1082167612Ssimon [Geoff] 1083167612Ssimon 1084167612Ssimon *) Minor adjustment to "rand" code. RAND_get_rand_method() now returns a 1085167612Ssimon 'const' value. Any code that should be able to modify a RAND_METHOD 1086167612Ssimon should already have non-const pointers to it (ie. they should only 1087167612Ssimon modify their own ones). 1088167612Ssimon [Geoff] 1089167612Ssimon 1090167612Ssimon *) Made a variety of little tweaks to the ENGINE code. 1091167612Ssimon - "atalla" and "ubsec" string definitions were moved from header files 1092167612Ssimon to C code. "nuron" string definitions were placed in variables 1093162911Ssimon rather than hard-coded - allowing parameterisation of these values 1094162911Ssimon later on via ctrl() commands. 1095162911Ssimon - Removed unused "#if 0"'d code. 1096162911Ssimon - Fixed engine list iteration code so it uses ENGINE_free() to release 1097162911Ssimon structural references. 1098162911Ssimon - Constified the RAND_METHOD element of ENGINE structures. 1099162911Ssimon - Constified various get/set functions as appropriate and added 1100162911Ssimon missing functions (including a catch-all ENGINE_cpy that duplicates 1101162911Ssimon all ENGINE values onto a new ENGINE except reference counts/state). 1102162911Ssimon - Removed NULL parameter checks in get/set functions. Setting a method 1103162911Ssimon or function to NULL is a way of cancelling out a previously set 1104162911Ssimon value. Passing a NULL ENGINE parameter is just plain stupid anyway 1105162911Ssimon and doesn't justify the extra error symbols and code. 1106162911Ssimon - Deprecate the ENGINE_FLAGS_MALLOCED define and move the area for 1107162911Ssimon flags from engine_int.h to engine.h. 1108162911Ssimon - Changed prototypes for ENGINE handler functions (init(), finish(), 1109162911Ssimon ctrl(), key-load functions, etc) to take an (ENGINE*) parameter. 1110162911Ssimon [Geoff] 1111162911Ssimon 1112162911Ssimon *) Implement binary inversion algorithm for BN_mod_inverse in addition 1113162911Ssimon to the algorithm using long division. The binary algorithm can be 1114162911Ssimon used only if the modulus is odd. On 32-bit systems, it is faster 1115162911Ssimon only for relatively small moduli (roughly 20-30% for 128-bit moduli, 1116162911Ssimon roughly 5-15% for 256-bit moduli), so we use it only for moduli 1117162911Ssimon up to 450 bits. In 64-bit environments, the binary algorithm 1118162911Ssimon appears to be advantageous for much longer moduli; here we use it 1119162911Ssimon for moduli up to 2048 bits. 1120162911Ssimon [Bodo Moeller] 1121162911Ssimon 1122162911Ssimon *) Rewrite CHOICE field setting in ASN1_item_ex_d2i(). The old code 1123162911Ssimon could not support the combine flag in choice fields. 1124162911Ssimon [Steve Henson] 1125162911Ssimon 1126162911Ssimon *) Add a 'copy_extensions' option to the 'ca' utility. This copies 1127162911Ssimon extensions from a certificate request to the certificate. 1128162911Ssimon [Steve Henson] 1129162911Ssimon 1130162911Ssimon *) Allow multiple 'certopt' and 'nameopt' options to be separated 1131162911Ssimon by commas. Add 'namopt' and 'certopt' options to the 'ca' config 1132162911Ssimon file: this allows the display of the certificate about to be 1133162911Ssimon signed to be customised, to allow certain fields to be included 1134162911Ssimon or excluded and extension details. The old system didn't display 1135162911Ssimon multicharacter strings properly, omitted fields not in the policy 1136162911Ssimon and couldn't display additional details such as extensions. 1137162911Ssimon [Steve Henson] 1138162911Ssimon 1139162911Ssimon *) Function EC_POINTs_mul for multiple scalar multiplication 1140162911Ssimon of an arbitrary number of elliptic curve points 1141162911Ssimon \sum scalars[i]*points[i], 1142162911Ssimon optionally including the generator defined for the EC_GROUP: 1143162911Ssimon scalar*generator + \sum scalars[i]*points[i]. 1144162911Ssimon 1145162911Ssimon EC_POINT_mul is a simple wrapper function for the typical case 1146160814Ssimon that the point list has just one item (besides the optional 1147160814Ssimon generator). 1148160814Ssimon [Bodo Moeller] 1149160814Ssimon 1150160814Ssimon *) First EC_METHODs for curves over GF(p): 1151160814Ssimon 1152160814Ssimon EC_GFp_simple_method() uses the basic BN_mod_mul and BN_mod_sqr 1153160814Ssimon operations and provides various method functions that can also 1154160814Ssimon operate with faster implementations of modular arithmetic. 1155160814Ssimon 1156160814Ssimon EC_GFp_mont_method() reuses most functions that are part of 1157160814Ssimon EC_GFp_simple_method, but uses Montgomery arithmetic. 1158160814Ssimon 1159160814Ssimon [Bodo Moeller; point addition and point doubling 1160160814Ssimon implementation directly derived from source code provided by 1161160814Ssimon Lenka Fibikova <fibikova@exp-math.uni-essen.de>] 1162160814Ssimon 1163160814Ssimon *) Framework for elliptic curves (crypto/ec/ec.h, crypto/ec/ec_lcl.h, 1164160814Ssimon crypto/ec/ec_lib.c): 1165160814Ssimon 1166160814Ssimon Curves are EC_GROUP objects (with an optional group generator) 1167160814Ssimon based on EC_METHODs that are built into the library. 1168160814Ssimon 1169160814Ssimon Points are EC_POINT objects based on EC_GROUP objects. 1170160814Ssimon 1171160814Ssimon Most of the framework would be able to handle curves over arbitrary 1172160814Ssimon finite fields, but as there are no obvious types for fields other 1173160814Ssimon than GF(p), some functions are limited to that for now. 1174160814Ssimon [Bodo Moeller] 1175160814Ssimon 1176160814Ssimon *) Add the -HTTP option to s_server. It is similar to -WWW, but requires 1177160814Ssimon that the file contains a complete HTTP response. 1178160814Ssimon [Richard Levitte] 1179160814Ssimon 1180160814Ssimon *) Add the ec directory to mkdef.pl and mkfiles.pl. In mkdef.pl 1181160814Ssimon change the def and num file printf format specifier from "%-40sXXX" 1182160814Ssimon to "%-39s XXX". The latter will always guarantee a space after the 1183160814Ssimon field while the former will cause them to run together if the field 1184160814Ssimon is 40 of more characters long. 1185160814Ssimon [Steve Henson] 1186160814Ssimon 1187160814Ssimon *) Constify the cipher and digest 'method' functions and structures 1188160814Ssimon and modify related functions to take constant EVP_MD and EVP_CIPHER 1189160814Ssimon pointers. 1190160814Ssimon [Steve Henson] 1191160814Ssimon 1192160814Ssimon *) Hide BN_CTX structure details in bn_lcl.h instead of publishing them 1193160814Ssimon in <openssl/bn.h>. Also further increase BN_CTX_NUM to 32. 1194160814Ssimon [Bodo Moeller] 1195160814Ssimon 1196160814Ssimon *) Modify EVP_Digest*() routines so they now return values. Although the 1197160814Ssimon internal software routines can never fail additional hardware versions 1198160814Ssimon might. 1199160814Ssimon [Steve Henson] 1200160814Ssimon 1201160814Ssimon *) Clean up crypto/err/err.h and change some error codes to avoid conflicts: 1202160814Ssimon 1203160814Ssimon Previously ERR_R_FATAL was too small and coincided with ERR_LIB_PKCS7 1204160814Ssimon (= ERR_R_PKCS7_LIB); it is now 64 instead of 32. 1205160814Ssimon 1206160814Ssimon ASN1 error codes 1207160814Ssimon ERR_R_NESTED_ASN1_ERROR 1208160814Ssimon ... 1209160814Ssimon ERR_R_MISSING_ASN1_EOS 1210160814Ssimon were 4 .. 9, conflicting with 1211160814Ssimon ERR_LIB_RSA (= ERR_R_RSA_LIB) 1212160814Ssimon ... 1213160814Ssimon ERR_LIB_PEM (= ERR_R_PEM_LIB). 1214160814Ssimon They are now 58 .. 63 (i.e., just below ERR_R_FATAL). 1215160814Ssimon 1216160814Ssimon Add new error code 'ERR_R_INTERNAL_ERROR'. 1217160814Ssimon [Bodo Moeller] 1218160814Ssimon 1219160814Ssimon *) Don't overuse locks in crypto/err/err.c: For data retrieval, CRYPTO_r_lock 1220160814Ssimon suffices. 1221160814Ssimon [Bodo Moeller] 1222160814Ssimon 1223160814Ssimon *) New option '-subj arg' for 'openssl req' and 'openssl ca'. This 1224160814Ssimon sets the subject name for a new request or supersedes the 1225160814Ssimon subject name in a given request. Formats that can be parsed are 1226160814Ssimon 'CN=Some Name, OU=myOU, C=IT' 1227160814Ssimon and 1228160814Ssimon 'CN=Some Name/OU=myOU/C=IT'. 1229160814Ssimon 1230160814Ssimon Add options '-batch' and '-verbose' to 'openssl req'. 1231160814Ssimon [Massimiliano Pala <madwolf@hackmasters.net>] 1232160814Ssimon 1233160814Ssimon *) Introduce the possibility to access global variables through 1234160814Ssimon functions on platform were that's the best way to handle exporting 1235160814Ssimon global variables in shared libraries. To enable this functionality, 1236160814Ssimon one must configure with "EXPORT_VAR_AS_FN" or defined the C macro 1237160814Ssimon "OPENSSL_EXPORT_VAR_AS_FUNCTION" in crypto/opensslconf.h (the latter 1238160814Ssimon is normally done by Configure or something similar). 1239160814Ssimon 1240160814Ssimon To implement a global variable, use the macro OPENSSL_IMPLEMENT_GLOBAL 1241160814Ssimon in the source file (foo.c) like this: 1242160814Ssimon 1243160814Ssimon OPENSSL_IMPLEMENT_GLOBAL(int,foo)=1; 1244160814Ssimon OPENSSL_IMPLEMENT_GLOBAL(double,bar); 1245160814Ssimon 1246160814Ssimon To declare a global variable, use the macros OPENSSL_DECLARE_GLOBAL 1247160814Ssimon and OPENSSL_GLOBAL_REF in the header file (foo.h) like this: 1248160814Ssimon 1249160814Ssimon OPENSSL_DECLARE_GLOBAL(int,foo); 1250160814Ssimon #define foo OPENSSL_GLOBAL_REF(foo) 1251160814Ssimon OPENSSL_DECLARE_GLOBAL(double,bar); 1252160814Ssimon #define bar OPENSSL_GLOBAL_REF(bar) 1253160814Ssimon 1254160814Ssimon The #defines are very important, and therefore so is including the 1255160814Ssimon header file everywhere where the defined globals are used. 1256160814Ssimon 1257160814Ssimon The macro OPENSSL_EXPORT_VAR_AS_FUNCTION also affects the definition 1258160814Ssimon of ASN.1 items, but that structure is a bit different. 1259160814Ssimon 1260160814Ssimon The largest change is in util/mkdef.pl which has been enhanced with 1261160814Ssimon better and easier to understand logic to choose which symbols should 1262160814Ssimon go into the Windows .def files as well as a number of fixes and code 1263160814Ssimon cleanup (among others, algorithm keywords are now sorted 1264160814Ssimon lexicographically to avoid constant rewrites). 1265160814Ssimon [Richard Levitte] 1266160814Ssimon 1267160814Ssimon *) In BN_div() keep a copy of the sign of 'num' before writing the 1268160814Ssimon result to 'rm' because if rm==num the value will be overwritten 1269160814Ssimon and produce the wrong result if 'num' is negative: this caused 1270160814Ssimon problems with BN_mod() and BN_nnmod(). 1271160814Ssimon [Steve Henson] 1272160814Ssimon 1273160814Ssimon *) Function OCSP_request_verify(). This checks the signature on an 1274160814Ssimon OCSP request and verifies the signer certificate. The signer 1275160814Ssimon certificate is just checked for a generic purpose and OCSP request 1276160814Ssimon trust settings. 1277160814Ssimon [Steve Henson] 1278160814Ssimon 1279160814Ssimon *) Add OCSP_check_validity() function to check the validity of OCSP 1280160814Ssimon responses. OCSP responses are prepared in real time and may only 1281160814Ssimon be a few seconds old. Simply checking that the current time lies 1282160814Ssimon between thisUpdate and nextUpdate max reject otherwise valid responses 1283160814Ssimon caused by either OCSP responder or client clock inaccuracy. Instead 1284160814Ssimon we allow thisUpdate and nextUpdate to fall within a certain period of 1285160814Ssimon the current time. The age of the response can also optionally be 1286160814Ssimon checked. Two new options -validity_period and -status_age added to 1287160814Ssimon ocsp utility. 1288160814Ssimon [Steve Henson] 1289160814Ssimon 1290160814Ssimon *) If signature or public key algorithm is unrecognized print out its 1291160814Ssimon OID rather that just UNKNOWN. 1292160814Ssimon [Steve Henson] 1293160814Ssimon 1294160814Ssimon *) Change OCSP_cert_to_id() to tolerate a NULL subject certificate and 1295160814Ssimon OCSP_cert_id_new() a NULL serialNumber. This allows a partial certificate 1296160814Ssimon ID to be generated from the issuer certificate alone which can then be 1297160814Ssimon passed to OCSP_id_issuer_cmp(). 1298160814Ssimon [Steve Henson] 1299142425Snectar 1300142425Snectar *) New compilation option ASN1_ITEM_FUNCTIONS. This causes the new 1301142425Snectar ASN1 modules to export functions returning ASN1_ITEM pointers 1302142425Snectar instead of the ASN1_ITEM structures themselves. This adds several 1303142425Snectar new macros which allow the underlying ASN1 function/structure to 1304142425Snectar be accessed transparently. As a result code should not use ASN1_ITEM 1305142425Snectar references directly (such as &X509_it) but instead use the relevant 1306142425Snectar macros (such as ASN1_ITEM_rptr(X509)). This option is to allow 1307142425Snectar use of the new ASN1 code on platforms where exporting structures 1308142425Snectar is problematical (for example in shared libraries) but exporting 1309142425Snectar functions returning pointers to structures is not. 1310142425Snectar [Steve Henson] 1311142425Snectar 1312142425Snectar *) Add support for overriding the generation of SSL/TLS session IDs. 1313142425Snectar These callbacks can be registered either in an SSL_CTX or per SSL. 1314142425Snectar The purpose of this is to allow applications to control, if they wish, 1315142425Snectar the arbitrary values chosen for use as session IDs, particularly as it 1316142425Snectar can be useful for session caching in multiple-server environments. A 1317142425Snectar command-line switch for testing this (and any client code that wishes 1318142425Snectar to use such a feature) has been added to "s_server". 1319142425Snectar [Geoff Thorpe, Lutz Jaenicke] 1320142425Snectar 1321142425Snectar *) Modify mkdef.pl to recognise and parse preprocessor conditionals 1322142425Snectar of the form '#if defined(...) || defined(...) || ...' and 1323127128Snectar '#if !defined(...) && !defined(...) && ...'. This also avoids 1324127128Snectar the growing number of special cases it was previously handling. 1325127128Snectar [Richard Levitte] 1326160814Ssimon 1327127128Snectar *) Make all configuration macros available for application by making 1328127128Snectar sure they are available in opensslconf.h, by giving them names starting 1329127128Snectar with "OPENSSL_" to avoid conflicts with other packages and by making 1330160814Ssimon sure e_os2.h will cover all platform-specific cases together with 1331127128Snectar opensslconf.h. 1332127128Snectar Additionally, it is now possible to define configuration/platform- 1333127128Snectar specific names (called "system identities"). In the C code, these 1334127128Snectar are prefixed with "OPENSSL_SYSNAME_". e_os2.h will create another 1335127128Snectar macro with the name beginning with "OPENSSL_SYS_", which is determined 1336127128Snectar from "OPENSSL_SYSNAME_*" or compiler-specific macros depending on 1337127128Snectar what is available. 1338127128Snectar [Richard Levitte] 1339127128Snectar 1340127128Snectar *) New option -set_serial to 'req' and 'x509' this allows the serial 1341127128Snectar number to use to be specified on the command line. Previously self 1342127128Snectar signed certificates were hard coded with serial number 0 and the 1343127128Snectar CA options of 'x509' had to use a serial number in a file which was 1344127128Snectar auto incremented. 1345127128Snectar [Steve Henson] 1346127128Snectar 1347127128Snectar *) New options to 'ca' utility to support V2 CRL entry extensions. 1348127128Snectar Currently CRL reason, invalidity date and hold instruction are 1349127128Snectar supported. Add new CRL extensions to V3 code and some new objects. 1350127128Snectar [Steve Henson] 1351127128Snectar 1352127128Snectar *) New function EVP_CIPHER_CTX_set_padding() this is used to 1353127128Snectar disable standard block padding (aka PKCS#5 padding) in the EVP 1354127128Snectar API, which was previously mandatory. This means that the data is 1355127128Snectar not padded in any way and so the total length much be a multiple 1356127128Snectar of the block size, otherwise an error occurs. 1357127128Snectar [Steve Henson] 1358127128Snectar 1359127128Snectar *) Initial (incomplete) OCSP SSL support. 1360127128Snectar [Steve Henson] 1361127128Snectar 1362127128Snectar *) New function OCSP_parse_url(). This splits up a URL into its host, 1363127128Snectar port and path components: primarily to parse OCSP URLs. New -url 1364127128Snectar option to ocsp utility. 1365127128Snectar [Steve Henson] 1366127128Snectar 1367127128Snectar *) New nonce behavior. The return value of OCSP_check_nonce() now 1368120631Snectar reflects the various checks performed. Applications can decide 1369120631Snectar whether to tolerate certain situations such as an absent nonce 1370120631Snectar in a response when one was present in a request: the ocsp application 1371120631Snectar just prints out a warning. New function OCSP_add1_basic_nonce() 1372120631Snectar this is to allow responders to include a nonce in a response even if 1373160814Ssimon the request is nonce-less. 1374120631Snectar [Steve Henson] 1375160814Ssimon 1376120631Snectar *) Disable stdin buffering in load_cert (apps/apps.c) so that no certs are 1377120631Snectar skipped when using openssl x509 multiple times on a single input file, 1378120631Snectar e.g. "(openssl x509 -out cert1; openssl x509 -out cert2) <certs". 1379120631Snectar [Bodo Moeller] 1380120631Snectar 1381120631Snectar *) Make ASN1_UTCTIME_set_string() and ASN1_GENERALIZEDTIME_set_string() 1382120631Snectar set string type: to handle setting ASN1_TIME structures. Fix ca 1383120631Snectar utility to correctly initialize revocation date of CRLs. 1384120631Snectar [Steve Henson] 1385120631Snectar 1386120631Snectar *) New option SSL_OP_CIPHER_SERVER_PREFERENCE allows the server to override 1387120631Snectar the clients preferred ciphersuites and rather use its own preferences. 1388120631Snectar Should help to work around M$ SGC (Server Gated Cryptography) bug in 1389120631Snectar Internet Explorer by ensuring unchanged hash method during stepup. 1390120631Snectar (Also replaces the broken/deactivated SSL_OP_NON_EXPORT_FIRST option.) 1391120631Snectar [Lutz Jaenicke] 1392120631Snectar 1393120631Snectar *) Make mkdef.pl recognise all DECLARE_ASN1 macros, change rijndael 1394120631Snectar to aes and add a new 'exist' option to print out symbols that don't 1395120631Snectar appear to exist. 1396120631Snectar [Steve Henson] 1397120631Snectar 1398120631Snectar *) Additional options to ocsp utility to allow flags to be set and 1399120631Snectar additional certificates supplied. 1400120631Snectar [Steve Henson] 1401120631Snectar 1402120631Snectar *) Add the option -VAfile to 'openssl ocsp', so the user can give the 1403120631Snectar OCSP client a number of certificate to only verify the response 1404120631Snectar signature against. 1405120631Snectar [Richard Levitte] 1406120631Snectar 1407120631Snectar *) Update Rijndael code to version 3.0 and change EVP AES ciphers to 1408120631Snectar handle the new API. Currently only ECB, CBC modes supported. Add new 1409120631Snectar AES OIDs. 1410120631Snectar 1411120631Snectar Add TLS AES ciphersuites as described in RFC3268, "Advanced 1412120631Snectar Encryption Standard (AES) Ciphersuites for Transport Layer 1413120631Snectar Security (TLS)". (In beta versions of OpenSSL 0.9.7, these were 1414120631Snectar not enabled by default and were not part of the "ALL" ciphersuite 1415120631Snectar alias because they were not yet official; they could be 1416120631Snectar explicitly requested by specifying the "AESdraft" ciphersuite 1417120631Snectar group alias. In the final release of OpenSSL 0.9.7, the group 1418120631Snectar alias is called "AES" and is part of "ALL".) 1419120631Snectar [Ben Laurie, Steve Henson, Bodo Moeller] 1420120631Snectar 1421120631Snectar *) New function OCSP_copy_nonce() to copy nonce value (if present) from 1422120631Snectar request to response. 1423120631Snectar [Steve Henson] 1424120631Snectar 1425120631Snectar *) Functions for OCSP responders. OCSP_request_onereq_count(), 1426120631Snectar OCSP_request_onereq_get0(), OCSP_onereq_get0_id() and OCSP_id_get0_info() 1427120631Snectar extract information from a certificate request. OCSP_response_create() 1428120631Snectar creates a response and optionally adds a basic response structure. 1429120631Snectar OCSP_basic_add1_status() adds a complete single response to a basic 1430120631Snectar response and returns the OCSP_SINGLERESP structure just added (to allow 1431120631Snectar extensions to be included for example). OCSP_basic_add1_cert() adds a 1432120631Snectar certificate to a basic response and OCSP_basic_sign() signs a basic 1433120631Snectar response with various flags. New helper functions ASN1_TIME_check() 1434120631Snectar (checks validity of ASN1_TIME structure) and ASN1_TIME_to_generalizedtime() 1435120631Snectar (converts ASN1_TIME to GeneralizedTime). 1436120631Snectar [Steve Henson] 1437120631Snectar 1438120631Snectar *) Various new functions. EVP_Digest() combines EVP_Digest{Init,Update,Final}() 1439120631Snectar in a single operation. X509_get0_pubkey_bitstr() extracts the public_key 1440120631Snectar structure from a certificate. X509_pubkey_digest() digests the public_key 1441120631Snectar contents: this is used in various key identifiers. 1442120631Snectar [Steve Henson] 1443120631Snectar 1444120631Snectar *) Make sk_sort() tolerate a NULL argument. 1445120631Snectar [Steve Henson reported by Massimiliano Pala <madwolf@comune.modena.it>] 1446120631Snectar 1447120631Snectar *) New OCSP verify flag OCSP_TRUSTOTHER. When set the "other" certificates 1448120631Snectar passed by the function are trusted implicitly. If any of them signed the 1449120631Snectar response then it is assumed to be valid and is not verified. 1450120631Snectar [Steve Henson] 1451120631Snectar 1452120631Snectar *) In PKCS7_set_type() initialise content_type in PKCS7_ENC_CONTENT 1453120631Snectar to data. This was previously part of the PKCS7 ASN1 code. This 1454111147Snectar was causing problems with OpenSSL created PKCS#12 and PKCS#7 structures. 1455111147Snectar [Steve Henson, reported by Kenneth R. Robinette 1456111147Snectar <support@securenetterm.com>] 1457111147Snectar 1458111147Snectar *) Add CRYPTO_push_info() and CRYPTO_pop_info() calls to new ASN1 1459111147Snectar routines: without these tracing memory leaks is very painful. 1460160814Ssimon Fix leaks in PKCS12 and PKCS7 routines. 1461111147Snectar [Steve Henson] 1462111147Snectar 1463111147Snectar *) Make X509_time_adj() cope with the new behaviour of ASN1_TIME_new(). 1464111147Snectar Previously it initialised the 'type' argument to V_ASN1_UTCTIME which 1465111147Snectar effectively meant GeneralizedTime would never be used. Now it 1466111147Snectar is initialised to -1 but X509_time_adj() now has to check the value 1467111147Snectar and use ASN1_TIME_set() if the value is not V_ASN1_UTCTIME or 1468111147Snectar V_ASN1_GENERALIZEDTIME, without this it always uses GeneralizedTime. 1469111147Snectar [Steve Henson, reported by Kenneth R. Robinette 1470111147Snectar <support@securenetterm.com>] 1471111147Snectar 1472111147Snectar *) Fixes to BN_to_ASN1_INTEGER when bn is zero. This would previously 1473111147Snectar result in a zero length in the ASN1_INTEGER structure which was 1474111147Snectar not consistent with the structure when d2i_ASN1_INTEGER() was used 1475111147Snectar and would cause ASN1_INTEGER_cmp() to fail. Enhance s2i_ASN1_INTEGER() 1476111147Snectar to cope with hex and negative integers. Fix bug in i2a_ASN1_INTEGER() 1477111147Snectar where it did not print out a minus for negative ASN1_INTEGER. 1478111147Snectar [Steve Henson] 1479111147Snectar 1480111147Snectar *) Add summary printout to ocsp utility. The various functions which 1481111147Snectar convert status values to strings have been renamed to: 1482111147Snectar OCSP_response_status_str(), OCSP_cert_status_str() and 1483111147Snectar OCSP_crl_reason_str() and are no longer static. New options 1484111147Snectar to verify nonce values and to disable verification. OCSP response 1485111147Snectar printout format cleaned up. 1486111147Snectar [Steve Henson] 1487111147Snectar 1488111147Snectar *) Add additional OCSP certificate checks. These are those specified 1489111147Snectar in RFC2560. This consists of two separate checks: the CA of the 1490111147Snectar certificate being checked must either be the OCSP signer certificate 1491111147Snectar or the issuer of the OCSP signer certificate. In the latter case the 1492111147Snectar OCSP signer certificate must contain the OCSP signing extended key 1493111147Snectar usage. This check is performed by attempting to match the OCSP 1494111147Snectar signer or the OCSP signer CA to the issuerNameHash and issuerKeyHash 1495111147Snectar in the OCSP_CERTID structures of the response. 1496111147Snectar [Steve Henson] 1497111147Snectar 1498111147Snectar *) Initial OCSP certificate verification added to OCSP_basic_verify() 1499111147Snectar and related routines. This uses the standard OpenSSL certificate 1500111147Snectar verify routines to perform initial checks (just CA validity) and 1501111147Snectar to obtain the certificate chain. Then additional checks will be 1502111147Snectar performed on the chain. Currently the root CA is checked to see 1503111147Snectar if it is explicitly trusted for OCSP signing. This is used to set 1504111147Snectar a root CA as a global signing root: that is any certificate that 1505111147Snectar chains to that CA is an acceptable OCSP signing certificate. 1506111147Snectar [Steve Henson] 1507111147Snectar 1508111147Snectar *) New '-extfile ...' option to 'openssl ca' for reading X.509v3 1509111147Snectar extensions from a separate configuration file. 1510111147Snectar As when reading extensions from the main configuration file, 1511111147Snectar the '-extensions ...' option may be used for specifying the 1512111147Snectar section to use. 1513111147Snectar [Massimiliano Pala <madwolf@comune.modena.it>] 1514111147Snectar 1515111147Snectar *) New OCSP utility. Allows OCSP requests to be generated or 1516111147Snectar read. The request can be sent to a responder and the output 1517111147Snectar parsed, outputed or printed in text form. Not complete yet: 1518111147Snectar still needs to check the OCSP response validity. 1519111147Snectar [Steve Henson] 1520111147Snectar 1521111147Snectar *) New subcommands for 'openssl ca': 1522111147Snectar 'openssl ca -status <serial>' prints the status of the cert with 1523111147Snectar the given serial number (according to the index file). 1524111147Snectar 'openssl ca -updatedb' updates the expiry status of certificates 1525111147Snectar in the index file. 1526111147Snectar [Massimiliano Pala <madwolf@comune.modena.it>] 1527111147Snectar 1528111147Snectar *) New '-newreq-nodes' command option to CA.pl. This is like 1529111147Snectar '-newreq', but calls 'openssl req' with the '-nodes' option 1530111147Snectar so that the resulting key is not encrypted. 1531111147Snectar [Damien Miller <djm@mindrot.org>] 1532111147Snectar 1533111147Snectar *) New configuration for the GNU Hurd. 1534111147Snectar [Jonathan Bartlett <johnnyb@wolfram.com> via Richard Levitte] 1535109998Smarkm 1536109998Smarkm *) Initial code to implement OCSP basic response verify. This 1537120631Snectar is currently incomplete. Currently just finds the signer's 1538120631Snectar certificate and verifies the signature on the response. 1539120631Snectar [Steve Henson] 1540109998Smarkm 1541109998Smarkm *) New SSLeay_version code SSLEAY_DIR to determine the compiled-in 1542109998Smarkm value of OPENSSLDIR. This is available via the new '-d' option 1543109998Smarkm to 'openssl version', and is also included in 'openssl version -a'. 1544109998Smarkm [Bodo Moeller] 1545109998Smarkm 1546109998Smarkm *) Allowing defining memory allocation callbacks that will be given 1547109998Smarkm file name and line number information in additional arguments 1548109998Smarkm (a const char* and an int). The basic functionality remains, as 1549109998Smarkm well as the original possibility to just replace malloc(), 1550109998Smarkm realloc() and free() by functions that do not know about these 1551109998Smarkm additional arguments. To register and find out the current 1552109998Smarkm settings for extended allocation functions, the following 1553109998Smarkm functions are provided: 1554109998Smarkm 1555109998Smarkm CRYPTO_set_mem_ex_functions 1556109998Smarkm CRYPTO_set_locked_mem_ex_functions 1557109998Smarkm CRYPTO_get_mem_ex_functions 1558109998Smarkm CRYPTO_get_locked_mem_ex_functions 1559109998Smarkm 1560109998Smarkm These work the same way as CRYPTO_set_mem_functions and friends. 1561109998Smarkm CRYPTO_get_[locked_]mem_functions now writes 0 where such an 1562109998Smarkm extended allocation function is enabled. 1563109998Smarkm Similarly, CRYPTO_get_[locked_]mem_ex_functions writes 0 where 1564109998Smarkm a conventional allocation function is enabled. 1565109998Smarkm [Richard Levitte, Bodo Moeller] 1566109998Smarkm 1567109998Smarkm *) Finish off removing the remaining LHASH function pointer casts. 1568109998Smarkm There should no longer be any prototype-casting required when using 1569109998Smarkm the LHASH abstraction, and any casts that remain are "bugs". See 1570109998Smarkm the callback types and macros at the head of lhash.h for details 1571109998Smarkm (and "OBJ_cleanup" in crypto/objects/obj_dat.c as an example). 1572109998Smarkm [Geoff Thorpe] 1573109998Smarkm 1574109998Smarkm *) Add automatic query of EGD sockets in RAND_poll() for the unix variant. 1575109998Smarkm If /dev/[u]random devices are not available or do not return enough 1576109998Smarkm entropy, EGD style sockets (served by EGD or PRNGD) will automatically 1577109998Smarkm be queried. 1578109998Smarkm The locations /var/run/egd-pool, /dev/egd-pool, /etc/egd-pool, and 1579109998Smarkm /etc/entropy will be queried once each in this sequence, quering stops 1580109998Smarkm when enough entropy was collected without querying more sockets. 1581109998Smarkm [Lutz Jaenicke] 1582109998Smarkm 1583109998Smarkm *) Change the Unix RAND_poll() variant to be able to poll several 1584109998Smarkm random devices, as specified by DEVRANDOM, until a sufficient amount 1585109998Smarkm of data has been collected. We spend at most 10 ms on each file 1586109998Smarkm (select timeout) and read in non-blocking mode. DEVRANDOM now 1587109998Smarkm defaults to the list "/dev/urandom", "/dev/random", "/dev/srandom" 1588109998Smarkm (previously it was just the string "/dev/urandom"), so on typical 1589109998Smarkm platforms the 10 ms delay will never occur. 1590109998Smarkm Also separate out the Unix variant to its own file, rand_unix.c. 1591109998Smarkm For VMS, there's a currently-empty rand_vms.c. 1592109998Smarkm [Richard Levitte] 1593109998Smarkm 1594109998Smarkm *) Move OCSP client related routines to ocsp_cl.c. These 1595109998Smarkm provide utility functions which an application needing 1596109998Smarkm to issue a request to an OCSP responder and analyse the 1597109998Smarkm response will typically need: as opposed to those which an 1598109998Smarkm OCSP responder itself would need which will be added later. 1599109998Smarkm 1600109998Smarkm OCSP_request_sign() signs an OCSP request with an API similar 1601109998Smarkm to PKCS7_sign(). OCSP_response_status() returns status of OCSP 1602109998Smarkm response. OCSP_response_get1_basic() extracts basic response 1603109998Smarkm from response. OCSP_resp_find_status(): finds and extracts status 1604109998Smarkm information from an OCSP_CERTID structure (which will be created 1605109998Smarkm when the request structure is built). These are built from lower 1606109998Smarkm level functions which work on OCSP_SINGLERESP structures but 1607109998Smarkm wont normally be used unless the application wishes to examine 1608109998Smarkm extensions in the OCSP response for example. 1609109998Smarkm 1610109998Smarkm Replace nonce routines with a pair of functions. 1611109998Smarkm OCSP_request_add1_nonce() adds a nonce value and optionally 1612109998Smarkm generates a random value. OCSP_check_nonce() checks the 1613109998Smarkm validity of the nonce in an OCSP response. 1614109998Smarkm [Steve Henson] 1615109998Smarkm 1616109998Smarkm *) Change function OCSP_request_add() to OCSP_request_add0_id(). 1617109998Smarkm This doesn't copy the supplied OCSP_CERTID and avoids the 1618109998Smarkm need to free up the newly created id. Change return type 1619109998Smarkm to OCSP_ONEREQ to return the internal OCSP_ONEREQ structure. 1620109998Smarkm This can then be used to add extensions to the request. 1621109998Smarkm Deleted OCSP_request_new(), since most of its functionality 1622109998Smarkm is now in OCSP_REQUEST_new() (and the case insensitive name 1623109998Smarkm clash) apart from the ability to set the request name which 1624109998Smarkm will be added elsewhere. 1625109998Smarkm [Steve Henson] 1626109998Smarkm 1627109998Smarkm *) Update OCSP API. Remove obsolete extensions argument from 1628109998Smarkm various functions. Extensions are now handled using the new 1629109998Smarkm OCSP extension code. New simple OCSP HTTP function which 1630109998Smarkm can be used to send requests and parse the response. 1631109998Smarkm [Steve Henson] 1632109998Smarkm 1633109998Smarkm *) Fix the PKCS#7 (S/MIME) code to work with new ASN1. Two new 1634109998Smarkm ASN1_ITEM structures help with sign and verify. PKCS7_ATTR_SIGN 1635109998Smarkm uses the special reorder version of SET OF to sort the attributes 1636109998Smarkm and reorder them to match the encoded order. This resolves a long 1637109998Smarkm standing problem: a verify on a PKCS7 structure just after signing 1638109998Smarkm it used to fail because the attribute order did not match the 1639109998Smarkm encoded order. PKCS7_ATTR_VERIFY does not reorder the attributes: 1640109998Smarkm it uses the received order. This is necessary to tolerate some broken 1641109998Smarkm software that does not order SET OF. This is handled by encoding 1642109998Smarkm as a SEQUENCE OF but using implicit tagging (with UNIVERSAL class) 1643109998Smarkm to produce the required SET OF. 1644109998Smarkm [Steve Henson] 1645109998Smarkm 1646109998Smarkm *) Have mk1mf.pl generate the macros OPENSSL_BUILD_SHLIBCRYPTO and 1647109998Smarkm OPENSSL_BUILD_SHLIBSSL and use them appropriately in the header 1648109998Smarkm files to get correct declarations of the ASN.1 item variables. 1649109998Smarkm [Richard Levitte] 1650109998Smarkm 1651109998Smarkm *) Rewrite of PKCS#12 code to use new ASN1 functionality. Replace many 1652109998Smarkm PKCS#12 macros with real functions. Fix two unrelated ASN1 bugs: 1653109998Smarkm asn1_check_tlen() would sometimes attempt to use 'ctx' when it was 1654109998Smarkm NULL and ASN1_TYPE was not dereferenced properly in asn1_ex_c2i(). 1655109998Smarkm New ASN1 macro: DECLARE_ASN1_ITEM() which just declares the relevant 1656109998Smarkm ASN1_ITEM and no wrapper functions. 1657109998Smarkm [Steve Henson] 1658109998Smarkm 1659109998Smarkm *) New functions or ASN1_item_d2i_fp() and ASN1_item_d2i_bio(). These 1660109998Smarkm replace the old function pointer based I/O routines. Change most of 1661109998Smarkm the *_d2i_bio() and *_d2i_fp() functions to use these. 1662109998Smarkm [Steve Henson] 1663109998Smarkm 1664109998Smarkm *) Enhance mkdef.pl to be more accepting about spacing in C preprocessor 1665109998Smarkm lines, recognice more "algorithms" that can be deselected, and make 1666109998Smarkm it complain about algorithm deselection that isn't recognised. 1667109998Smarkm [Richard Levitte] 1668109998Smarkm 1669109998Smarkm *) New ASN1 functions to handle dup, sign, verify, digest, pack and 1670109998Smarkm unpack operations in terms of ASN1_ITEM. Modify existing wrappers 1671109998Smarkm to use new functions. Add NO_ASN1_OLD which can be set to remove 1672109998Smarkm some old style ASN1 functions: this can be used to determine if old 1673109998Smarkm code will still work when these eventually go away. 1674109998Smarkm [Steve Henson] 1675109998Smarkm 1676109998Smarkm *) New extension functions for OCSP structures, these follow the 1677160814Ssimon same conventions as certificates and CRLs. 1678109998Smarkm [Steve Henson] 1679109998Smarkm 1680109998Smarkm *) New function X509V3_add1_i2d(). This automatically encodes and 1681109998Smarkm adds an extension. Its behaviour can be customised with various 1682109998Smarkm flags to append, replace or delete. Various wrappers added for 1683109998Smarkm certifcates and CRLs. 1684109998Smarkm [Steve Henson] 1685109998Smarkm 1686109998Smarkm *) Fix to avoid calling the underlying ASN1 print routine when 1687109998Smarkm an extension cannot be parsed. Correct a typo in the 1688109998Smarkm OCSP_SERVICELOC extension. Tidy up print OCSP format. 1689109998Smarkm [Steve Henson] 1690109998Smarkm 1691109998Smarkm *) Make mkdef.pl parse some of the ASN1 macros and add apropriate 1692109998Smarkm entries for variables. 1693109998Smarkm [Steve Henson] 1694109998Smarkm 1695109998Smarkm *) Add functionality to apps/openssl.c for detecting locking 1696109998Smarkm problems: As the program is single-threaded, all we have 1697109998Smarkm to do is register a locking callback using an array for 1698109998Smarkm storing which locks are currently held by the program. 1699109998Smarkm [Bodo Moeller] 1700109998Smarkm 1701109998Smarkm *) Use a lock around the call to CRYPTO_get_ex_new_index() in 1702109998Smarkm SSL_get_ex_data_X509_STORE_idx(), which is used in 1703109998Smarkm ssl_verify_cert_chain() and thus can be called at any time 1704109998Smarkm during TLS/SSL handshakes so that thread-safety is essential. 1705109998Smarkm Unfortunately, the ex_data design is not at all suited 1706109998Smarkm for multi-threaded use, so it probably should be abolished. 1707109998Smarkm [Bodo Moeller] 1708109998Smarkm 1709109998Smarkm *) Added Broadcom "ubsec" ENGINE to OpenSSL. 1710109998Smarkm [Broadcom, tweaked and integrated by Geoff Thorpe] 1711109998Smarkm 1712109998Smarkm *) Move common extension printing code to new function 1713111147Snectar X509V3_print_extensions(). Reorganise OCSP print routines and 1714109998Smarkm implement some needed OCSP ASN1 functions. Add OCSP extensions. 1715109998Smarkm [Steve Henson] 1716109998Smarkm 1717109998Smarkm *) New function X509_signature_print() to remove duplication in some 1718109998Smarkm print routines. 1719109998Smarkm [Steve Henson] 1720109998Smarkm 1721109998Smarkm *) Add a special meaning when SET OF and SEQUENCE OF flags are both 1722109998Smarkm set (this was treated exactly the same as SET OF previously). This 1723109998Smarkm is used to reorder the STACK representing the structure to match the 1724109998Smarkm encoding. This will be used to get round a problem where a PKCS7 1725109998Smarkm structure which was signed could not be verified because the STACK 1726109998Smarkm order did not reflect the encoded order. 1727109998Smarkm [Steve Henson] 1728109998Smarkm 1729109998Smarkm *) Reimplement the OCSP ASN1 module using the new code. 1730109998Smarkm [Steve Henson] 1731109998Smarkm 1732109998Smarkm *) Update the X509V3 code to permit the use of an ASN1_ITEM structure 1733109998Smarkm for its ASN1 operations. The old style function pointers still exist 1734109998Smarkm for now but they will eventually go away. 1735109998Smarkm [Steve Henson] 1736109998Smarkm 1737109998Smarkm *) Merge in replacement ASN1 code from the ASN1 branch. This almost 1738109998Smarkm completely replaces the old ASN1 functionality with a table driven 1739109998Smarkm encoder and decoder which interprets an ASN1_ITEM structure describing 1740109998Smarkm the ASN1 module. Compatibility with the existing ASN1 API (i2d,d2i) is 1741109998Smarkm largely maintained. Almost all of the old asn1_mac.h macro based ASN1 1742109998Smarkm has also been converted to the new form. 1743109998Smarkm [Steve Henson] 1744109998Smarkm 1745109998Smarkm *) Change BN_mod_exp_recp so that negative moduli are tolerated 1746109998Smarkm (the sign is ignored). Similarly, ignore the sign in BN_MONT_CTX_set 1747109998Smarkm so that BN_mod_exp_mont and BN_mod_exp_mont_word work 1748109998Smarkm for negative moduli. 1749109998Smarkm [Bodo Moeller] 1750109998Smarkm 1751109998Smarkm *) Fix BN_uadd and BN_usub: Always return non-negative results instead 1752109998Smarkm of not touching the result's sign bit. 1753109998Smarkm [Bodo Moeller] 1754109998Smarkm 1755109998Smarkm *) BN_div bugfix: If the result is 0, the sign (res->neg) must not be 1756109998Smarkm set. 1757109998Smarkm [Bodo Moeller] 1758109998Smarkm 1759109998Smarkm *) Changed the LHASH code to use prototypes for callbacks, and created 1760109998Smarkm macros to declare and implement thin (optionally static) functions 1761109998Smarkm that provide type-safety and avoid function pointer casting for the 1762109998Smarkm type-specific callbacks. 1763109998Smarkm [Geoff Thorpe] 1764109998Smarkm 1765109998Smarkm *) Added Kerberos Cipher Suites to be used with TLS, as written in 1766109998Smarkm RFC 2712. 1767109998Smarkm [Veers Staats <staatsvr@asc.hpc.mil>, 1768109998Smarkm Jeffrey Altman <jaltman@columbia.edu>, via Richard Levitte] 1769109998Smarkm 1770109998Smarkm *) Reformat the FAQ so the different questions and answers can be divided 1771109998Smarkm in sections depending on the subject. 1772109998Smarkm [Richard Levitte] 1773109998Smarkm 1774109998Smarkm *) Have the zlib compression code load ZLIB.DLL dynamically under 1775109998Smarkm Windows. 1776109998Smarkm [Richard Levitte] 1777109998Smarkm 1778109998Smarkm *) New function BN_mod_sqrt for computing square roots modulo a prime 1779109998Smarkm (using the probabilistic Tonelli-Shanks algorithm unless 1780109998Smarkm p == 3 (mod 4) or p == 5 (mod 8), which are cases that can 1781109998Smarkm be handled deterministically). 1782109998Smarkm [Lenka Fibikova <fibikova@exp-math.uni-essen.de>, Bodo Moeller] 1783109998Smarkm 1784109998Smarkm *) Make BN_mod_inverse faster by explicitly handling small quotients 1785109998Smarkm in the Euclid loop. (Speed gain about 20% for small moduli [256 or 1786109998Smarkm 512 bits], about 30% for larger ones [1024 or 2048 bits].) 1787109998Smarkm [Bodo Moeller] 1788109998Smarkm 1789109998Smarkm *) New function BN_kronecker. 1790109998Smarkm [Bodo Moeller] 1791109998Smarkm 1792109998Smarkm *) Fix BN_gcd so that it works on negative inputs; the result is 1793109998Smarkm positive unless both parameters are zero. 1794109998Smarkm Previously something reasonably close to an infinite loop was 1795109998Smarkm possible because numbers could be growing instead of shrinking 1796109998Smarkm in the implementation of Euclid's algorithm. 1797109998Smarkm [Bodo Moeller] 1798109998Smarkm 1799109998Smarkm *) Fix BN_is_word() and BN_is_one() macros to take into account the 1800109998Smarkm sign of the number in question. 1801109998Smarkm 1802109998Smarkm Fix BN_is_word(a,w) to work correctly for w == 0. 1803109998Smarkm 1804109998Smarkm The old BN_is_word(a,w) macro is now called BN_abs_is_word(a,w) 1805109998Smarkm because its test if the absolute value of 'a' equals 'w'. 1806109998Smarkm Note that BN_abs_is_word does *not* handle w == 0 reliably; 1807109998Smarkm it exists mostly for use in the implementations of BN_is_zero(), 1808109998Smarkm BN_is_one(), and BN_is_word(). 1809109998Smarkm [Bodo Moeller] 1810109998Smarkm 1811109998Smarkm *) New function BN_swap. 1812109998Smarkm [Bodo Moeller] 1813109998Smarkm 1814109998Smarkm *) Use BN_nnmod instead of BN_mod in crypto/bn/bn_exp.c so that 1815109998Smarkm the exponentiation functions are more likely to produce reasonable 1816109998Smarkm results on negative inputs. 1817109998Smarkm [Bodo Moeller] 1818109998Smarkm 1819109998Smarkm *) Change BN_mod_mul so that the result is always non-negative. 1820109998Smarkm Previously, it could be negative if one of the factors was negative; 1821109998Smarkm I don't think anyone really wanted that behaviour. 1822109998Smarkm [Bodo Moeller] 1823109998Smarkm 1824109998Smarkm *) Move BN_mod_... functions into new file crypto/bn/bn_mod.c 1825109998Smarkm (except for exponentiation, which stays in crypto/bn/bn_exp.c, 1826109998Smarkm and BN_mod_mul_reciprocal, which stays in crypto/bn/bn_recp.c) 1827109998Smarkm and add new functions: 1828109998Smarkm 1829109998Smarkm BN_nnmod 1830109998Smarkm BN_mod_sqr 1831109998Smarkm BN_mod_add 1832109998Smarkm BN_mod_add_quick 1833109998Smarkm BN_mod_sub 1834109998Smarkm BN_mod_sub_quick 1835109998Smarkm BN_mod_lshift1 1836109998Smarkm BN_mod_lshift1_quick 1837109998Smarkm BN_mod_lshift 1838109998Smarkm BN_mod_lshift_quick 1839109998Smarkm 1840109998Smarkm These functions always generate non-negative results. 1841109998Smarkm 1842109998Smarkm BN_nnmod otherwise is like BN_mod (if BN_mod computes a remainder r 1843109998Smarkm such that |m| < r < 0, BN_nnmod will output rem + |m| instead). 1844109998Smarkm 1845109998Smarkm BN_mod_XXX_quick(r, a, [b,] m) generates the same result as 1846109998Smarkm BN_mod_XXX(r, a, [b,] m, ctx), but requires that a [and b] 1847109998Smarkm be reduced modulo m. 1848109998Smarkm [Lenka Fibikova <fibikova@exp-math.uni-essen.de>, Bodo Moeller] 1849109998Smarkm 1850109998Smarkm#if 0 1851109998Smarkm The following entry accidentily appeared in the CHANGES file 1852109998Smarkm distributed with OpenSSL 0.9.7. The modifications described in 1853109998Smarkm it do *not* apply to OpenSSL 0.9.7. 1854109998Smarkm 1855109998Smarkm *) Remove a few calls to bn_wexpand() in BN_sqr() (the one in there 1856109998Smarkm was actually never needed) and in BN_mul(). The removal in BN_mul() 1857109998Smarkm required a small change in bn_mul_part_recursive() and the addition 1858109998Smarkm of the functions bn_cmp_part_words(), bn_sub_part_words() and 1859109998Smarkm bn_add_part_words(), which do the same thing as bn_cmp_words(), 1860109998Smarkm bn_sub_words() and bn_add_words() except they take arrays with 1861109998Smarkm differing sizes. 1862109998Smarkm [Richard Levitte] 1863109998Smarkm#endif 1864109998Smarkm 1865109998Smarkm *) In 'openssl passwd', verify passwords read from the terminal 1866109998Smarkm unless the '-salt' option is used (which usually means that 1867109998Smarkm verification would just waste user's time since the resulting 1868109998Smarkm hash is going to be compared with some given password hash) 1869109998Smarkm or the new '-noverify' option is used. 1870109998Smarkm 1871109998Smarkm This is an incompatible change, but it does not affect 1872109998Smarkm non-interactive use of 'openssl passwd' (passwords on the command 1873109998Smarkm line, '-stdin' option, '-in ...' option) and thus should not 1874109998Smarkm cause any problems. 1875109998Smarkm [Bodo Moeller] 1876109998Smarkm 1877109998Smarkm *) Remove all references to RSAref, since there's no more need for it. 1878109998Smarkm [Richard Levitte] 1879109998Smarkm 1880109998Smarkm *) Make DSO load along a path given through an environment variable 1881109998Smarkm (SHLIB_PATH) with shl_load(). 1882109998Smarkm [Richard Levitte] 1883109998Smarkm 1884109998Smarkm *) Constify the ENGINE code as a result of BIGNUM constification. 1885109998Smarkm Also constify the RSA code and most things related to it. In a 1886109998Smarkm few places, most notable in the depth of the ASN.1 code, ugly 1887109998Smarkm casts back to non-const were required (to be solved at a later 1888109998Smarkm time) 1889109998Smarkm [Richard Levitte] 1890109998Smarkm 1891109998Smarkm *) Make it so the openssl application has all engines loaded by default. 1892109998Smarkm [Richard Levitte] 1893109998Smarkm 1894109998Smarkm *) Constify the BIGNUM routines a little more. 1895109998Smarkm [Richard Levitte] 1896109998Smarkm 1897109998Smarkm *) Add the following functions: 1898109998Smarkm 1899109998Smarkm ENGINE_load_cswift() 1900109998Smarkm ENGINE_load_chil() 1901109998Smarkm ENGINE_load_atalla() 1902109998Smarkm ENGINE_load_nuron() 1903109998Smarkm ENGINE_load_builtin_engines() 1904109998Smarkm 1905109998Smarkm That way, an application can itself choose if external engines that 1906109998Smarkm are built-in in OpenSSL shall ever be used or not. The benefit is 1907109998Smarkm that applications won't have to be linked with libdl or other dso 1908109998Smarkm libraries unless it's really needed. 1909109998Smarkm 1910109998Smarkm Changed 'openssl engine' to load all engines on demand. 1911109998Smarkm Changed the engine header files to avoid the duplication of some 1912109998Smarkm declarations (they differed!). 1913109998Smarkm [Richard Levitte] 1914109998Smarkm 1915109998Smarkm *) 'openssl engine' can now list capabilities. 1916109998Smarkm [Richard Levitte] 1917109998Smarkm 1918109998Smarkm *) Better error reporting in 'openssl engine'. 1919109998Smarkm [Richard Levitte] 1920109998Smarkm 1921109998Smarkm *) Never call load_dh_param(NULL) in s_server. 1922109998Smarkm [Bodo Moeller] 1923109998Smarkm 1924109998Smarkm *) Add engine application. It can currently list engines by name and 1925109998Smarkm identity, and test if they are actually available. 1926109998Smarkm [Richard Levitte] 1927109998Smarkm 1928109998Smarkm *) Improve RPM specification file by forcing symbolic linking and making 1929109998Smarkm sure the installed documentation is also owned by root.root. 1930109998Smarkm [Damien Miller <djm@mindrot.org>] 1931109998Smarkm 1932109998Smarkm *) Give the OpenSSL applications more possibilities to make use of 1933109998Smarkm keys (public as well as private) handled by engines. 1934109998Smarkm [Richard Levitte] 1935109998Smarkm 1936109998Smarkm *) Add OCSP code that comes from CertCo. 1937109998Smarkm [Richard Levitte] 1938109998Smarkm 1939109998Smarkm *) Add VMS support for the Rijndael code. 1940109998Smarkm [Richard Levitte] 1941109998Smarkm 1942109998Smarkm *) Added untested support for Nuron crypto accelerator. 1943109998Smarkm [Ben Laurie] 1944109998Smarkm 1945109998Smarkm *) Add support for external cryptographic devices. This code was 1946109998Smarkm previously distributed separately as the "engine" branch. 1947109998Smarkm [Geoff Thorpe, Richard Levitte] 1948109998Smarkm 1949109998Smarkm *) Rework the filename-translation in the DSO code. It is now possible to 1950109998Smarkm have far greater control over how a "name" is turned into a filename 1951109998Smarkm depending on the operating environment and any oddities about the 1952109998Smarkm different shared library filenames on each system. 1953109998Smarkm [Geoff Thorpe] 1954109998Smarkm 1955109998Smarkm *) Support threads on FreeBSD-elf in Configure. 1956109998Smarkm [Richard Levitte] 1957109998Smarkm 1958109998Smarkm *) Fix for SHA1 assembly problem with MASM: it produces 1959109998Smarkm warnings about corrupt line number information when assembling 1960109998Smarkm with debugging information. This is caused by the overlapping 1961109998Smarkm of two sections. 1962109998Smarkm [Bernd Matthes <mainbug@celocom.de>, Steve Henson] 1963109998Smarkm 1964109998Smarkm *) NCONF changes. 1965109998Smarkm NCONF_get_number() has no error checking at all. As a replacement, 1966109998Smarkm NCONF_get_number_e() is defined (_e for "error checking") and is 1967109998Smarkm promoted strongly. The old NCONF_get_number is kept around for 1968109998Smarkm binary backward compatibility. 1969109998Smarkm Make it possible for methods to load from something other than a BIO, 1970109998Smarkm by providing a function pointer that is given a name instead of a BIO. 1971109998Smarkm For example, this could be used to load configuration data from an 1972109998Smarkm LDAP server. 1973109998Smarkm [Richard Levitte] 1974109998Smarkm 1975109998Smarkm *) Fix for non blocking accept BIOs. Added new I/O special reason 1976109998Smarkm BIO_RR_ACCEPT to cover this case. Previously use of accept BIOs 1977109998Smarkm with non blocking I/O was not possible because no retry code was 1978109998Smarkm implemented. Also added new SSL code SSL_WANT_ACCEPT to cover 1979109998Smarkm this case. 1980109998Smarkm [Steve Henson] 1981109998Smarkm 1982109998Smarkm *) Added the beginnings of Rijndael support. 1983109998Smarkm [Ben Laurie] 1984109998Smarkm 1985109998Smarkm *) Fix for bug in DirectoryString mask setting. Add support for 1986109998Smarkm X509_NAME_print_ex() in 'req' and X509_print_ex() function 1987109998Smarkm to allow certificate printing to more controllable, additional 1988109998Smarkm 'certopt' option to 'x509' to allow new printing options to be 1989109998Smarkm set. 1990109998Smarkm [Steve Henson] 1991109998Smarkm 1992109998Smarkm *) Clean old EAY MD5 hack from e_os.h. 1993109998Smarkm [Richard Levitte] 1994109998Smarkm 1995109998Smarkm Changes between 0.9.6j and 0.9.6k [30 Sep 2003] 1996109998Smarkm 1997109998Smarkm *) Fix various bugs revealed by running the NISCC test suite: 1998109998Smarkm 1999109998Smarkm Stop out of bounds reads in the ASN1 code when presented with 2000109998Smarkm invalid tags (CAN-2003-0543 and CAN-2003-0544). 2001109998Smarkm 2002109998Smarkm If verify callback ignores invalid public key errors don't try to check 2003109998Smarkm certificate signature with the NULL public key. 2004109998Smarkm 2005109998Smarkm [Steve Henson] 2006109998Smarkm 2007109998Smarkm *) In ssl3_accept() (ssl/s3_srvr.c) only accept a client certificate 2008109998Smarkm if the server requested one: as stated in TLS 1.0 and SSL 3.0 2009109998Smarkm specifications. 2010109998Smarkm [Steve Henson] 2011109998Smarkm 2012109998Smarkm *) In ssl3_get_client_hello() (ssl/s3_srvr.c), tolerate additional 2013109998Smarkm extra data after the compression methods not only for TLS 1.0 2014109998Smarkm but also for SSL 3.0 (as required by the specification). 2015109998Smarkm [Bodo Moeller; problem pointed out by Matthias Loepfe] 2016109998Smarkm 2017109998Smarkm *) Change X509_certificate_type() to mark the key as exported/exportable 2018109998Smarkm when it's 512 *bits* long, not 512 bytes. 2019109998Smarkm [Richard Levitte] 2020109998Smarkm 2021109998Smarkm Changes between 0.9.6i and 0.9.6j [10 Apr 2003] 2022109998Smarkm 2023109998Smarkm *) Countermeasure against the Klima-Pokorny-Rosa extension of 2024109998Smarkm Bleichbacher's attack on PKCS #1 v1.5 padding: treat 2025109998Smarkm a protocol version number mismatch like a decryption error 2026109998Smarkm in ssl3_get_client_key_exchange (ssl/s3_srvr.c). 2027109998Smarkm [Bodo Moeller] 2028109998Smarkm 2029109998Smarkm *) Turn on RSA blinding by default in the default implementation 2030109998Smarkm to avoid a timing attack. Applications that don't want it can call 2031109998Smarkm RSA_blinding_off() or use the new flag RSA_FLAG_NO_BLINDING. 2032109998Smarkm They would be ill-advised to do so in most cases. 2033109998Smarkm [Ben Laurie, Steve Henson, Geoff Thorpe, Bodo Moeller] 2034109998Smarkm 2035109998Smarkm *) Change RSA blinding code so that it works when the PRNG is not 2036109998Smarkm seeded (in this case, the secret RSA exponent is abused as 2037109998Smarkm an unpredictable seed -- if it is not unpredictable, there 2038109998Smarkm is no point in blinding anyway). Make RSA blinding thread-safe 2039109998Smarkm by remembering the creator's thread ID in rsa->blinding and 2040109998Smarkm having all other threads use local one-time blinding factors 2041109998Smarkm (this requires more computation than sharing rsa->blinding, but 2042109998Smarkm avoids excessive locking; and if an RSA object is not shared 2043109998Smarkm between threads, blinding will still be very fast). 2044109998Smarkm [Bodo Moeller] 2045109998Smarkm 2046109998Smarkm Changes between 0.9.6h and 0.9.6i [19 Feb 2003] 2047109998Smarkm 2048109998Smarkm *) In ssl3_get_record (ssl/s3_pkt.c), minimize information leaked 2049109998Smarkm via timing by performing a MAC computation even if incorrrect 2050109998Smarkm block cipher padding has been found. This is a countermeasure 2051109998Smarkm against active attacks where the attacker has to distinguish 2052109998Smarkm between bad padding and a MAC verification error. (CAN-2003-0078) 2053109998Smarkm 2054109998Smarkm [Bodo Moeller; problem pointed out by Brice Canvel (EPFL), 2055109998Smarkm Alain Hiltgen (UBS), Serge Vaudenay (EPFL), and 2056109998Smarkm Martin Vuagnoux (EPFL, Ilion)] 2057109998Smarkm 2058109998Smarkm Changes between 0.9.6g and 0.9.6h [5 Dec 2002] 2059109998Smarkm 2060109998Smarkm *) New function OPENSSL_cleanse(), which is used to cleanse a section of 2061109998Smarkm memory from it's contents. This is done with a counter that will 2062109998Smarkm place alternating values in each byte. This can be used to solve 2063109998Smarkm two issues: 1) the removal of calls to memset() by highly optimizing 2064109998Smarkm compilers, and 2) cleansing with other values than 0, since those can 2065109998Smarkm be read through on certain media, for example a swap space on disk. 2066109998Smarkm [Geoff Thorpe] 2067109998Smarkm 2068109998Smarkm *) Bugfix: client side session caching did not work with external caching, 2069109998Smarkm because the session->cipher setting was not restored when reloading 2070109998Smarkm from the external cache. This problem was masked, when 2071109998Smarkm SSL_OP_NETSCAPE_REUSE_CIPHER_CHANGE_BUG (part of SSL_OP_ALL) was set. 2072109998Smarkm (Found by Steve Haslam <steve@araqnid.ddts.net>.) 2073109998Smarkm [Lutz Jaenicke] 2074109998Smarkm 2075109998Smarkm *) Fix client_certificate (ssl/s2_clnt.c): The permissible total 2076109998Smarkm length of the REQUEST-CERTIFICATE message is 18 .. 34, not 17 .. 33. 2077109998Smarkm [Zeev Lieber <zeev-l@yahoo.com>] 2078109998Smarkm 2079109998Smarkm *) Undo an undocumented change introduced in 0.9.6e which caused 2080109998Smarkm repeated calls to OpenSSL_add_all_ciphers() and 2081109998Smarkm OpenSSL_add_all_digests() to be ignored, even after calling 2082109998Smarkm EVP_cleanup(). 2083109998Smarkm [Richard Levitte] 2084109998Smarkm 2085109998Smarkm *) Change the default configuration reader to deal with last line not 2086109998Smarkm being properly terminated. 2087109998Smarkm [Richard Levitte] 2088109998Smarkm 2089109998Smarkm *) Change X509_NAME_cmp() so it applies the special rules on handling 2090109998Smarkm DN values that are of type PrintableString, as well as RDNs of type 2091109998Smarkm emailAddress where the value has the type ia5String. 2092109998Smarkm [stefank@valicert.com via Richard Levitte] 2093109998Smarkm 2094109998Smarkm *) Add a SSL_SESS_CACHE_NO_INTERNAL_STORE flag to take over half 2095109998Smarkm the job SSL_SESS_CACHE_NO_INTERNAL_LOOKUP was inconsistently 2096109998Smarkm doing, define a new flag (SSL_SESS_CACHE_NO_INTERNAL) to be 2097109998Smarkm the bitwise-OR of the two for use by the majority of applications 2098109998Smarkm wanting this behaviour, and update the docs. The documented 2099109998Smarkm behaviour and actual behaviour were inconsistent and had been 2100109998Smarkm changing anyway, so this is more a bug-fix than a behavioural 2101109998Smarkm change. 2102109998Smarkm [Geoff Thorpe, diagnosed by Nadav Har'El] 2103109998Smarkm 2104109998Smarkm *) Don't impose a 16-byte length minimum on session IDs in ssl/s3_clnt.c 2105109998Smarkm (the SSL 3.0 and TLS 1.0 specifications allow any length up to 32 bytes). 2106109998Smarkm [Bodo Moeller] 2107109998Smarkm 2108109998Smarkm *) Fix initialization code race conditions in 2109109998Smarkm SSLv23_method(), SSLv23_client_method(), SSLv23_server_method(), 2110109998Smarkm SSLv2_method(), SSLv2_client_method(), SSLv2_server_method(), 2111109998Smarkm SSLv3_method(), SSLv3_client_method(), SSLv3_server_method(), 2112109998Smarkm TLSv1_method(), TLSv1_client_method(), TLSv1_server_method(), 2113109998Smarkm ssl2_get_cipher_by_char(), 2114109998Smarkm ssl3_get_cipher_by_char(). 2115109998Smarkm [Patrick McCormick <patrick@tellme.com>, Bodo Moeller] 2116109998Smarkm 2117109998Smarkm *) Reorder cleanup sequence in SSL_CTX_free(): only remove the ex_data after 2118109998Smarkm the cached sessions are flushed, as the remove_cb() might use ex_data 2119109998Smarkm contents. Bug found by Sam Varshavchik <mrsam@courier-mta.com> 2120109998Smarkm (see [openssl.org #212]). 2121109998Smarkm [Geoff Thorpe, Lutz Jaenicke] 2122109998Smarkm 2123109998Smarkm *) Fix typo in OBJ_txt2obj which incorrectly passed the content 2124109998Smarkm length, instead of the encoding length to d2i_ASN1_OBJECT. 2125109998Smarkm [Steve Henson] 2126109998Smarkm 2127109998Smarkm Changes between 0.9.6f and 0.9.6g [9 Aug 2002] 2128109998Smarkm 2129109998Smarkm *) [In 0.9.6g-engine release:] 2130109998Smarkm Fix crypto/engine/vendor_defns/cswift.h for WIN32 (use '_stdcall'). 2131109998Smarkm [Lynn Gazis <lgazis@rainbow.com>] 2132109998Smarkm 2133109998Smarkm Changes between 0.9.6e and 0.9.6f [8 Aug 2002] 2134109998Smarkm 2135109998Smarkm *) Fix ASN1 checks. Check for overflow by comparing with LONG_MAX 2136109998Smarkm and get fix the header length calculation. 2137109998Smarkm [Florian Weimer <Weimer@CERT.Uni-Stuttgart.DE>, 2138109998Smarkm Alon Kantor <alonk@checkpoint.com> (and others), 2139109998Smarkm Steve Henson] 2140109998Smarkm 2141109998Smarkm *) Use proper error handling instead of 'assertions' in buffer 2142109998Smarkm overflow checks added in 0.9.6e. This prevents DoS (the 2143109998Smarkm assertions could call abort()). 2144109998Smarkm [Arne Ansper <arne@ats.cyber.ee>, Bodo Moeller] 2145109998Smarkm 2146109998Smarkm Changes between 0.9.6d and 0.9.6e [30 Jul 2002] 2147109998Smarkm 2148109998Smarkm *) Add various sanity checks to asn1_get_length() to reject 2149109998Smarkm the ASN1 length bytes if they exceed sizeof(long), will appear 2150109998Smarkm negative or the content length exceeds the length of the 2151109998Smarkm supplied buffer. 2152109998Smarkm [Steve Henson, Adi Stav <stav@mercury.co.il>, James Yonan <jim@ntlp.com>] 2153109998Smarkm 2154109998Smarkm *) Fix cipher selection routines: ciphers without encryption had no flags 2155109998Smarkm for the cipher strength set and where therefore not handled correctly 2156109998Smarkm by the selection routines (PR #130). 2157109998Smarkm [Lutz Jaenicke] 2158109998Smarkm 2159109998Smarkm *) Fix EVP_dsa_sha macro. 2160109998Smarkm [Nils Larsch] 2161109998Smarkm 2162109998Smarkm *) New option 2163109998Smarkm SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS 2164109998Smarkm for disabling the SSL 3.0/TLS 1.0 CBC vulnerability countermeasure 2165109998Smarkm that was added in OpenSSL 0.9.6d. 2166109998Smarkm 2167109998Smarkm As the countermeasure turned out to be incompatible with some 2168109998Smarkm broken SSL implementations, the new option is part of SSL_OP_ALL. 2169109998Smarkm SSL_OP_ALL is usually employed when compatibility with weird SSL 2170109998Smarkm implementations is desired (e.g. '-bugs' option to 's_client' and 2171109998Smarkm 's_server'), so the new option is automatically set in many 2172109998Smarkm applications. 2173109998Smarkm [Bodo Moeller] 2174109998Smarkm 2175109998Smarkm *) Changes in security patch: 2176109998Smarkm 2177109998Smarkm Changes marked "(CHATS)" were sponsored by the Defense Advanced 2178109998Smarkm Research Projects Agency (DARPA) and Air Force Research Laboratory, 2179109998Smarkm Air Force Materiel Command, USAF, under agreement number 2180109998Smarkm F30602-01-2-0537. 2181109998Smarkm 2182109998Smarkm *) Add various sanity checks to asn1_get_length() to reject 2183109998Smarkm the ASN1 length bytes if they exceed sizeof(long), will appear 2184109998Smarkm negative or the content length exceeds the length of the 2185109998Smarkm supplied buffer. (CAN-2002-0659) 2186109998Smarkm [Steve Henson, Adi Stav <stav@mercury.co.il>, James Yonan <jim@ntlp.com>] 2187109998Smarkm 2188109998Smarkm *) Assertions for various potential buffer overflows, not known to 2189109998Smarkm happen in practice. 2190109998Smarkm [Ben Laurie (CHATS)] 2191109998Smarkm 2192109998Smarkm *) Various temporary buffers to hold ASCII versions of integers were 2193109998Smarkm too small for 64 bit platforms. (CAN-2002-0655) 2194109998Smarkm [Matthew Byng-Maddick <mbm@aldigital.co.uk> and Ben Laurie (CHATS)> 2195109998Smarkm 2196109998Smarkm *) Remote buffer overflow in SSL3 protocol - an attacker could 2197109998Smarkm supply an oversized session ID to a client. (CAN-2002-0656) 2198109998Smarkm [Ben Laurie (CHATS)] 2199109998Smarkm 2200109998Smarkm *) Remote buffer overflow in SSL2 protocol - an attacker could 2201109998Smarkm supply an oversized client master key. (CAN-2002-0656) 2202109998Smarkm [Ben Laurie (CHATS)] 2203109998Smarkm 2204109998Smarkm Changes between 0.9.6c and 0.9.6d [9 May 2002] 2205109998Smarkm 2206109998Smarkm *) Fix crypto/asn1/a_sign.c so that 'parameters' is omitted (not 2207109998Smarkm encoded as NULL) with id-dsa-with-sha1. 2208109998Smarkm [Nils Larsch <nla@trustcenter.de>; problem pointed out by Bodo Moeller] 2209109998Smarkm 2210109998Smarkm *) Check various X509_...() return values in apps/req.c. 2211109998Smarkm [Nils Larsch <nla@trustcenter.de>] 2212109998Smarkm 2213109998Smarkm *) Fix BASE64 decode (EVP_DecodeUpdate) for data with CR/LF ended lines: 2214109998Smarkm an end-of-file condition would erronously be flagged, when the CRLF 2215109998Smarkm was just at the end of a processed block. The bug was discovered when 2216109998Smarkm processing data through a buffering memory BIO handing the data to a 2217109998Smarkm BASE64-decoding BIO. Bug fund and patch submitted by Pavel Tsekov 2218109998Smarkm <ptsekov@syntrex.com> and Nedelcho Stanev. 2219109998Smarkm [Lutz Jaenicke] 2220109998Smarkm 2221109998Smarkm *) Implement a countermeasure against a vulnerability recently found 2222109998Smarkm in CBC ciphersuites in SSL 3.0/TLS 1.0: Send an empty fragment 2223109998Smarkm before application data chunks to avoid the use of known IVs 2224109998Smarkm with data potentially chosen by the attacker. 2225109998Smarkm [Bodo Moeller] 2226109998Smarkm 2227109998Smarkm *) Fix length checks in ssl3_get_client_hello(). 2228109998Smarkm [Bodo Moeller] 2229109998Smarkm 2230109998Smarkm *) TLS/SSL library bugfix: use s->s3->in_read_app_data differently 2231109998Smarkm to prevent ssl3_read_internal() from incorrectly assuming that 2232109998Smarkm ssl3_read_bytes() found application data while handshake 2233109998Smarkm processing was enabled when in fact s->s3->in_read_app_data was 2234109998Smarkm merely automatically cleared during the initial handshake. 2235109998Smarkm [Bodo Moeller; problem pointed out by Arne Ansper <arne@ats.cyber.ee>] 2236109998Smarkm 2237109998Smarkm *) Fix object definitions for Private and Enterprise: they were not 2238109998Smarkm recognized in their shortname (=lowercase) representation. Extend 2239109998Smarkm obj_dat.pl to issue an error when using undefined keywords instead 2240109998Smarkm of silently ignoring the problem (Svenning Sorensen 2241109998Smarkm <sss@sss.dnsalias.net>). 2242109998Smarkm [Lutz Jaenicke] 2243109998Smarkm 2244109998Smarkm *) Fix DH_generate_parameters() so that it works for 'non-standard' 2245109998Smarkm generators, i.e. generators other than 2 and 5. (Previously, the 2246109998Smarkm code did not properly initialise the 'add' and 'rem' values to 2247109998Smarkm BN_generate_prime().) 2248109998Smarkm 2249109998Smarkm In the new general case, we do not insist that 'generator' is 2250109998Smarkm actually a primitive root: This requirement is rather pointless; 2251109998Smarkm a generator of the order-q subgroup is just as good, if not 2252109998Smarkm better. 2253109998Smarkm [Bodo Moeller] 2254109998Smarkm 2255109998Smarkm *) Map new X509 verification errors to alerts. Discovered and submitted by 2256109998Smarkm Tom Wu <tom@arcot.com>. 2257109998Smarkm [Lutz Jaenicke] 2258109998Smarkm 2259109998Smarkm *) Fix ssl3_pending() (ssl/s3_lib.c) to prevent SSL_pending() from 2260109998Smarkm returning non-zero before the data has been completely received 2261109998Smarkm when using non-blocking I/O. 2262109998Smarkm [Bodo Moeller; problem pointed out by John Hughes] 2263109998Smarkm 2264109998Smarkm *) Some of the ciphers missed the strength entry (SSL_LOW etc). 2265109998Smarkm [Ben Laurie, Lutz Jaenicke] 2266109998Smarkm 2267109998Smarkm *) Fix bug in SSL_clear(): bad sessions were not removed (found by 2268109998Smarkm Yoram Zahavi <YoramZ@gilian.com>). 2269109998Smarkm [Lutz Jaenicke] 2270109998Smarkm 2271109998Smarkm *) Add information about CygWin 1.3 and on, and preserve proper 2272109998Smarkm configuration for the versions before that. 2273109998Smarkm [Corinna Vinschen <vinschen@redhat.com> and Richard Levitte] 2274109998Smarkm 2275109998Smarkm *) Make removal from session cache (SSL_CTX_remove_session()) more robust: 2276109998Smarkm check whether we deal with a copy of a session and do not delete from 2277109998Smarkm the cache in this case. Problem reported by "Izhar Shoshani Levi" 2278109998Smarkm <izhar@checkpoint.com>. 2279109998Smarkm [Lutz Jaenicke] 2280109998Smarkm 2281109998Smarkm *) Do not store session data into the internal session cache, if it 2282109998Smarkm is never intended to be looked up (SSL_SESS_CACHE_NO_INTERNAL_LOOKUP 2283109998Smarkm flag is set). Proposed by Aslam <aslam@funk.com>. 2284109998Smarkm [Lutz Jaenicke] 2285109998Smarkm 2286109998Smarkm *) Have ASN1_BIT_STRING_set_bit() really clear a bit when the requested 2287109998Smarkm value is 0. 2288109998Smarkm [Richard Levitte] 2289109998Smarkm 2290109998Smarkm *) [In 0.9.6d-engine release:] 2291109998Smarkm Fix a crashbug and a logic bug in hwcrhk_load_pubkey(). 2292109998Smarkm [Toomas Kiisk <vix@cyber.ee> via Richard Levitte] 2293109998Smarkm 2294109998Smarkm *) Add the configuration target linux-s390x. 2295109998Smarkm [Neale Ferguson <Neale.Ferguson@SoftwareAG-USA.com> via Richard Levitte] 2296109998Smarkm 2297109998Smarkm *) The earlier bugfix for the SSL3_ST_SW_HELLO_REQ_C case of 2298109998Smarkm ssl3_accept (ssl/s3_srvr.c) incorrectly used a local flag 2299109998Smarkm variable as an indication that a ClientHello message has been 2300109998Smarkm received. As the flag value will be lost between multiple 2301109998Smarkm invocations of ssl3_accept when using non-blocking I/O, the 2302109998Smarkm function may not be aware that a handshake has actually taken 2303109998Smarkm place, thus preventing a new session from being added to the 2304109998Smarkm session cache. 2305109998Smarkm 2306109998Smarkm To avoid this problem, we now set s->new_session to 2 instead of 2307109998Smarkm using a local variable. 2308109998Smarkm [Lutz Jaenicke, Bodo Moeller] 2309109998Smarkm 2310109998Smarkm *) Bugfix: Return -1 from ssl3_get_server_done (ssl3/s3_clnt.c) 2311109998Smarkm if the SSL_R_LENGTH_MISMATCH error is detected. 2312109998Smarkm [Geoff Thorpe, Bodo Moeller] 2313109998Smarkm 2314109998Smarkm *) New 'shared_ldflag' column in Configure platform table. 2315109998Smarkm [Richard Levitte] 2316109998Smarkm 2317109998Smarkm *) Fix EVP_CIPHER_mode macro. 2318109998Smarkm ["Dan S. Camper" <dan@bti.net>] 2319109998Smarkm 2320109998Smarkm *) Fix ssl3_read_bytes (ssl/s3_pkt.c): To ignore messages of unknown 2321109998Smarkm type, we must throw them away by setting rr->length to 0. 2322109998Smarkm [D P Chang <dpc@qualys.com>] 2323109998Smarkm 2324109998Smarkm Changes between 0.9.6b and 0.9.6c [21 dec 2001] 2325109998Smarkm 2326109998Smarkm *) Fix BN_rand_range bug pointed out by Dominikus Scherkl 2327109998Smarkm <Dominikus.Scherkl@biodata.com>. (The previous implementation 2328109998Smarkm worked incorrectly for those cases where range = 10..._2 and 2329109998Smarkm 3*range is two bits longer than range.) 2330109998Smarkm [Bodo Moeller] 2331109998Smarkm 2332109998Smarkm *) Only add signing time to PKCS7 structures if it is not already 2333109998Smarkm present. 2334109998Smarkm [Steve Henson] 2335109998Smarkm 2336109998Smarkm *) Fix crypto/objects/objects.h: "ld-ce" should be "id-ce", 2337109998Smarkm OBJ_ld_ce should be OBJ_id_ce. 2338109998Smarkm Also some ip-pda OIDs in crypto/objects/objects.txt were 2339109998Smarkm incorrect (cf. RFC 3039). 2340109998Smarkm [Matt Cooper, Frederic Giudicelli, Bodo Moeller] 2341109998Smarkm 2342109998Smarkm *) Release CRYPTO_LOCK_DYNLOCK when CRYPTO_destroy_dynlockid() 2343109998Smarkm returns early because it has nothing to do. 2344109998Smarkm [Andy Schneider <andy.schneider@bjss.co.uk>] 2345109998Smarkm 2346109998Smarkm *) [In 0.9.6c-engine release:] 2347109998Smarkm Fix mutex callback return values in crypto/engine/hw_ncipher.c. 2348109998Smarkm [Andy Schneider <andy.schneider@bjss.co.uk>] 2349109998Smarkm 2350109998Smarkm *) [In 0.9.6c-engine release:] 2351109998Smarkm Add support for Cryptographic Appliance's keyserver technology. 2352109998Smarkm (Use engine 'keyclient') 2353109998Smarkm [Cryptographic Appliances and Geoff Thorpe] 2354109998Smarkm 2355109998Smarkm *) Add a configuration entry for OS/390 Unix. The C compiler 'c89' 2356109998Smarkm is called via tools/c89.sh because arguments have to be 2357109998Smarkm rearranged (all '-L' options must appear before the first object 2358109998Smarkm modules). 2359109998Smarkm [Richard Shapiro <rshapiro@abinitio.com>] 2360109998Smarkm 2361109998Smarkm *) [In 0.9.6c-engine release:] 2362109998Smarkm Add support for Broadcom crypto accelerator cards, backported 2363109998Smarkm from 0.9.7. 2364109998Smarkm [Broadcom, Nalin Dahyabhai <nalin@redhat.com>, Mark Cox] 2365109998Smarkm 2366109998Smarkm *) [In 0.9.6c-engine release:] 2367109998Smarkm Add support for SureWare crypto accelerator cards from 2368109998Smarkm Baltimore Technologies. (Use engine 'sureware') 2369109998Smarkm [Baltimore Technologies and Mark Cox] 2370109998Smarkm 2371109998Smarkm *) [In 0.9.6c-engine release:] 2372109998Smarkm Add support for crypto accelerator cards from Accelerated 2373109998Smarkm Encryption Processing, www.aep.ie. (Use engine 'aep') 2374109998Smarkm [AEP Inc. and Mark Cox] 2375109998Smarkm 2376109998Smarkm *) Add a configuration entry for gcc on UnixWare. 2377109998Smarkm [Gary Benson <gbenson@redhat.com>] 2378109998Smarkm 2379109998Smarkm *) Change ssl/s2_clnt.c and ssl/s2_srvr.c so that received handshake 2380109998Smarkm messages are stored in a single piece (fixed-length part and 2381109998Smarkm variable-length part combined) and fix various bugs found on the way. 2382109998Smarkm [Bodo Moeller] 2383109998Smarkm 2384109998Smarkm *) Disable caching in BIO_gethostbyname(), directly use gethostbyname() 2385109998Smarkm instead. BIO_gethostbyname() does not know what timeouts are 2386109998Smarkm appropriate, so entries would stay in cache even when they have 2387109998Smarkm become invalid. 2388109998Smarkm [Bodo Moeller; problem pointed out by Rich Salz <rsalz@zolera.com> 2389109998Smarkm 2390109998Smarkm *) Change ssl23_get_client_hello (ssl/s23_srvr.c) behaviour when 2391109998Smarkm faced with a pathologically small ClientHello fragment that does 2392109998Smarkm not contain client_version: Instead of aborting with an error, 2393109998Smarkm simply choose the highest available protocol version (i.e., 2394109998Smarkm TLS 1.0 unless it is disabled). In practice, ClientHello 2395109998Smarkm messages are never sent like this, but this change gives us 2396109998Smarkm strictly correct behaviour at least for TLS. 2397109998Smarkm [Bodo Moeller] 2398109998Smarkm 2399109998Smarkm *) Fix SSL handshake functions and SSL_clear() such that SSL_clear() 2400109998Smarkm never resets s->method to s->ctx->method when called from within 2401109998Smarkm one of the SSL handshake functions. 2402109998Smarkm [Bodo Moeller; problem pointed out by Niko Baric] 2403109998Smarkm 2404109998Smarkm *) In ssl3_get_client_hello (ssl/s3_srvr.c), generate a fatal alert 2405109998Smarkm (sent using the client's version number) if client_version is 2406109998Smarkm smaller than the protocol version in use. Also change 2407109998Smarkm ssl23_get_client_hello (ssl/s23_srvr.c) to select TLS 1.0 if 2408109998Smarkm the client demanded SSL 3.0 but only TLS 1.0 is enabled; then 2409109998Smarkm the client will at least see that alert. 2410109998Smarkm [Bodo Moeller] 2411109998Smarkm 2412109998Smarkm *) Fix ssl3_get_message (ssl/s3_both.c) to handle message fragmentation 2413109998Smarkm correctly. 2414109998Smarkm [Bodo Moeller] 2415109998Smarkm 2416109998Smarkm *) Avoid infinite loop in ssl3_get_message (ssl/s3_both.c) if a 2417109998Smarkm client receives HelloRequest while in a handshake. 2418109998Smarkm [Bodo Moeller; bug noticed by Andy Schneider <andy.schneider@bjss.co.uk>] 2419109998Smarkm 2420109998Smarkm *) Bugfix in ssl3_accept (ssl/s3_srvr.c): Case SSL3_ST_SW_HELLO_REQ_C 2421109998Smarkm should end in 'break', not 'goto end' which circuments various 2422109998Smarkm cleanups done in state SSL_ST_OK. But session related stuff 2423109998Smarkm must be disabled for SSL_ST_OK in the case that we just sent a 2424109998Smarkm HelloRequest. 2425109998Smarkm 2426109998Smarkm Also avoid some overhead by not calling ssl_init_wbio_buffer() 2427109998Smarkm before just sending a HelloRequest. 2428109998Smarkm [Bodo Moeller, Eric Rescorla <ekr@rtfm.com>] 2429109998Smarkm 2430109998Smarkm *) Fix ssl/s3_enc.c, ssl/t1_enc.c and ssl/s3_pkt.c so that we don't 2431109998Smarkm reveal whether illegal block cipher padding was found or a MAC 2432109998Smarkm verification error occured. (Neither SSLerr() codes nor alerts 2433109998Smarkm are directly visible to potential attackers, but the information 2434109998Smarkm may leak via logfiles.) 2435109998Smarkm 2436109998Smarkm Similar changes are not required for the SSL 2.0 implementation 2437109998Smarkm because the number of padding bytes is sent in clear for SSL 2.0, 2438109998Smarkm and the extra bytes are just ignored. However ssl/s2_pkt.c 2439109998Smarkm failed to verify that the purported number of padding bytes is in 2440109998Smarkm the legal range. 2441109998Smarkm [Bodo Moeller] 2442109998Smarkm 2443109998Smarkm *) Add OpenUNIX-8 support including shared libraries 2444109998Smarkm (Boyd Lynn Gerber <gerberb@zenez.com>). 2445109998Smarkm [Lutz Jaenicke] 2446109998Smarkm 2447109998Smarkm *) Improve RSA_padding_check_PKCS1_OAEP() check again to avoid 2448109998Smarkm 'wristwatch attack' using huge encoding parameters (cf. 2449109998Smarkm James H. Manger's CRYPTO 2001 paper). Note that the 2450109998Smarkm RSA_PKCS1_OAEP_PADDING case of RSA_private_decrypt() does not use 2451109998Smarkm encoding parameters and hence was not vulnerable. 2452109998Smarkm [Bodo Moeller] 2453109998Smarkm 2454109998Smarkm *) BN_sqr() bug fix. 2455109998Smarkm [Ulf M�ller, reported by Jim Ellis <jim.ellis@cavium.com>] 2456109998Smarkm 2457109998Smarkm *) Rabin-Miller test analyses assume uniformly distributed witnesses, 2458109998Smarkm so use BN_pseudo_rand_range() instead of using BN_pseudo_rand() 2459109998Smarkm followed by modular reduction. 2460109998Smarkm [Bodo Moeller; pointed out by Adam Young <AYoung1@NCSUS.JNJ.COM>] 2461109998Smarkm 2462109998Smarkm *) Add BN_pseudo_rand_range() with obvious functionality: BN_rand_range() 2463109998Smarkm equivalent based on BN_pseudo_rand() instead of BN_rand(). 2464109998Smarkm [Bodo Moeller] 2465109998Smarkm 2466109998Smarkm *) s3_srvr.c: allow sending of large client certificate lists (> 16 kB). 2467109998Smarkm This function was broken, as the check for a new client hello message 2468109998Smarkm to handle SGC did not allow these large messages. 2469109998Smarkm (Tracked down by "Douglas E. Engert" <deengert@anl.gov>.) 2470109998Smarkm [Lutz Jaenicke] 2471109998Smarkm 2472109998Smarkm *) Add alert descriptions for TLSv1 to SSL_alert_desc_string[_long](). 2473109998Smarkm [Lutz Jaenicke] 2474109998Smarkm 2475109998Smarkm *) Fix buggy behaviour of BIO_get_num_renegotiates() and BIO_ctrl() 2476109998Smarkm for BIO_C_GET_WRITE_BUF_SIZE ("Stephen Hinton" <shinton@netopia.com>). 2477109998Smarkm [Lutz Jaenicke] 2478109998Smarkm 2479109998Smarkm *) Rework the configuration and shared library support for Tru64 Unix. 2480109998Smarkm The configuration part makes use of modern compiler features and 2481109998Smarkm still retains old compiler behavior for those that run older versions 2482109998Smarkm of the OS. The shared library support part includes a variant that 2483109998Smarkm uses the RPATH feature, and is available through the special 2484109998Smarkm configuration target "alpha-cc-rpath", which will never be selected 2485109998Smarkm automatically. 2486109998Smarkm [Tim Mooney <mooney@dogbert.cc.ndsu.NoDak.edu> via Richard Levitte] 2487109998Smarkm 2488109998Smarkm *) In ssl3_get_key_exchange (ssl/s3_clnt.c), call ssl3_get_message() 2489109998Smarkm with the same message size as in ssl3_get_certificate_request(). 2490109998Smarkm Otherwise, if no ServerKeyExchange message occurs, CertificateRequest 2491109998Smarkm messages might inadvertently be reject as too long. 2492109998Smarkm [Petr Lampa <lampa@fee.vutbr.cz>] 2493109998Smarkm 2494109998Smarkm *) Enhanced support for IA-64 Unix platforms (well, Linux and HP-UX). 2495109998Smarkm [Andy Polyakov] 2496109998Smarkm 2497109998Smarkm *) Modified SSL library such that the verify_callback that has been set 2498109998Smarkm specificly for an SSL object with SSL_set_verify() is actually being 2499109998Smarkm used. Before the change, a verify_callback set with this function was 2500109998Smarkm ignored and the verify_callback() set in the SSL_CTX at the time of 2501109998Smarkm the call was used. New function X509_STORE_CTX_set_verify_cb() introduced 2502109998Smarkm to allow the necessary settings. 2503109998Smarkm [Lutz Jaenicke] 2504109998Smarkm 2505109998Smarkm *) Initialize static variable in crypto/dsa/dsa_lib.c and crypto/dh/dh_lib.c 2506109998Smarkm explicitly to NULL, as at least on Solaris 8 this seems not always to be 2507109998Smarkm done automatically (in contradiction to the requirements of the C 2508109998Smarkm standard). This made problems when used from OpenSSH. 2509109998Smarkm [Lutz Jaenicke] 2510109998Smarkm 2511109998Smarkm *) In OpenSSL 0.9.6a and 0.9.6b, crypto/dh/dh_key.c ignored 2512109998Smarkm dh->length and always used 2513109998Smarkm 2514109998Smarkm BN_rand_range(priv_key, dh->p). 2515109998Smarkm 2516109998Smarkm BN_rand_range() is not necessary for Diffie-Hellman, and this 2517109998Smarkm specific range makes Diffie-Hellman unnecessarily inefficient if 2518109998Smarkm dh->length (recommended exponent length) is much smaller than the 2519109998Smarkm length of dh->p. We could use BN_rand_range() if the order of 2520109998Smarkm the subgroup was stored in the DH structure, but we only have 2521109998Smarkm dh->length. 2522109998Smarkm 2523109998Smarkm So switch back to 2524109998Smarkm 2525109998Smarkm BN_rand(priv_key, l, ...) 2526109998Smarkm 2527109998Smarkm where 'l' is dh->length if this is defined, or BN_num_bits(dh->p)-1 2528109998Smarkm otherwise. 2529109998Smarkm [Bodo Moeller] 2530109998Smarkm 2531109998Smarkm *) In 2532109998Smarkm 2533109998Smarkm RSA_eay_public_encrypt 2534109998Smarkm RSA_eay_private_decrypt 2535109998Smarkm RSA_eay_private_encrypt (signing) 2536109998Smarkm RSA_eay_public_decrypt (signature verification) 2537109998Smarkm 2538109998Smarkm (default implementations for RSA_public_encrypt, 2539109998Smarkm RSA_private_decrypt, RSA_private_encrypt, RSA_public_decrypt), 2540109998Smarkm always reject numbers >= n. 2541109998Smarkm [Bodo Moeller] 2542109998Smarkm 2543109998Smarkm *) In crypto/rand/md_rand.c, use a new short-time lock CRYPTO_LOCK_RAND2 2544109998Smarkm to synchronize access to 'locking_thread'. This is necessary on 2545109998Smarkm systems where access to 'locking_thread' (an 'unsigned long' 2546109998Smarkm variable) is not atomic. 2547109998Smarkm [Bodo Moeller] 2548109998Smarkm 2549109998Smarkm *) In crypto/rand/md_rand.c, set 'locking_thread' to current thread's ID 2550109998Smarkm *before* setting the 'crypto_lock_rand' flag. The previous code had 2551109998Smarkm a race condition if 0 is a valid thread ID. 2552109998Smarkm [Travis Vitek <vitek@roguewave.com>] 2553109998Smarkm 2554109998Smarkm *) Add support for shared libraries under Irix. 2555109998Smarkm [Albert Chin-A-Young <china@thewrittenword.com>] 2556109998Smarkm 2557109998Smarkm *) Add configuration option to build on Linux on both big-endian and 2558109998Smarkm little-endian MIPS. 2559109998Smarkm [Ralf Baechle <ralf@uni-koblenz.de>] 2560109998Smarkm 2561109998Smarkm *) Add the possibility to create shared libraries on HP-UX. 2562109998Smarkm [Richard Levitte] 2563109998Smarkm 2564109998Smarkm Changes between 0.9.6a and 0.9.6b [9 Jul 2001] 2565109998Smarkm 2566109998Smarkm *) Change ssleay_rand_bytes (crypto/rand/md_rand.c) 2567109998Smarkm to avoid a SSLeay/OpenSSL PRNG weakness pointed out by 2568109998Smarkm Markku-Juhani O. Saarinen <markku-juhani.saarinen@nokia.com>: 2569109998Smarkm PRNG state recovery was possible based on the output of 2570109998Smarkm one PRNG request appropriately sized to gain knowledge on 2571109998Smarkm 'md' followed by enough consecutive 1-byte PRNG requests 2572109998Smarkm to traverse all of 'state'. 2573109998Smarkm 2574109998Smarkm 1. When updating 'md_local' (the current thread's copy of 'md') 2575109998Smarkm during PRNG output generation, hash all of the previous 2576109998Smarkm 'md_local' value, not just the half used for PRNG output. 2577109998Smarkm 2578109998Smarkm 2. Make the number of bytes from 'state' included into the hash 2579109998Smarkm independent from the number of PRNG bytes requested. 2580109998Smarkm 2581109998Smarkm The first measure alone would be sufficient to avoid 2582109998Smarkm Markku-Juhani's attack. (Actually it had never occurred 2583109998Smarkm to me that the half of 'md_local' used for chaining was the 2584109998Smarkm half from which PRNG output bytes were taken -- I had always 2585109998Smarkm assumed that the secret half would be used.) The second 2586109998Smarkm measure makes sure that additional data from 'state' is never 2587109998Smarkm mixed into 'md_local' in small portions; this heuristically 2588109998Smarkm further strengthens the PRNG. 2589109998Smarkm [Bodo Moeller] 2590109998Smarkm 2591109998Smarkm *) Fix crypto/bn/asm/mips3.s. 2592109998Smarkm [Andy Polyakov] 2593109998Smarkm 2594109998Smarkm *) When only the key is given to "enc", the IV is undefined. Print out 2595109998Smarkm an error message in this case. 2596109998Smarkm [Lutz Jaenicke] 2597109998Smarkm 2598109998Smarkm *) Handle special case when X509_NAME is empty in X509 printing routines. 2599109998Smarkm [Steve Henson] 2600109998Smarkm 2601109998Smarkm *) In dsa_do_verify (crypto/dsa/dsa_ossl.c), verify that r and s are 2602109998Smarkm positive and less than q. 2603109998Smarkm [Bodo Moeller] 2604109998Smarkm 2605109998Smarkm *) Don't change *pointer in CRYPTO_add_lock() is add_lock_callback is 2606109998Smarkm used: it isn't thread safe and the add_lock_callback should handle 2607109998Smarkm that itself. 2608109998Smarkm [Paul Rose <Paul.Rose@bridge.com>] 2609109998Smarkm 2610109998Smarkm *) Verify that incoming data obeys the block size in 2611109998Smarkm ssl3_enc (ssl/s3_enc.c) and tls1_enc (ssl/t1_enc.c). 2612109998Smarkm [Bodo Moeller] 2613109998Smarkm 2614109998Smarkm *) Fix OAEP check. 2615109998Smarkm [Ulf M�ller, Bodo M�ller] 2616109998Smarkm 2617109998Smarkm *) The countermeasure against Bleichbacher's attack on PKCS #1 v1.5 2618109998Smarkm RSA encryption was accidentally removed in s3_srvr.c in OpenSSL 0.9.5 2619109998Smarkm when fixing the server behaviour for backwards-compatible 'client 2620109998Smarkm hello' messages. (Note that the attack is impractical against 2621109998Smarkm SSL 3.0 and TLS 1.0 anyway because length and version checking 2622109998Smarkm means that the probability of guessing a valid ciphertext is 2623109998Smarkm around 2^-40; see section 5 in Bleichenbacher's CRYPTO '98 2624109998Smarkm paper.) 2625109998Smarkm 2626109998Smarkm Before 0.9.5, the countermeasure (hide the error by generating a 2627109998Smarkm random 'decryption result') did not work properly because 2628109998Smarkm ERR_clear_error() was missing, meaning that SSL_get_error() would 2629109998Smarkm detect the supposedly ignored error. 2630109998Smarkm 2631109998Smarkm Both problems are now fixed. 2632109998Smarkm [Bodo Moeller] 2633109998Smarkm 2634109998Smarkm *) In crypto/bio/bf_buff.c, increase DEFAULT_BUFFER_SIZE to 4096 2635109998Smarkm (previously it was 1024). 2636109998Smarkm [Bodo Moeller] 2637109998Smarkm 2638109998Smarkm *) Fix for compatibility mode trust settings: ignore trust settings 2639109998Smarkm unless some valid trust or reject settings are present. 2640109998Smarkm [Steve Henson] 2641109998Smarkm 2642109998Smarkm *) Fix for blowfish EVP: its a variable length cipher. 2643109998Smarkm [Steve Henson] 2644109998Smarkm 2645109998Smarkm *) Fix various bugs related to DSA S/MIME verification. Handle missing 2646109998Smarkm parameters in DSA public key structures and return an error in the 2647109998Smarkm DSA routines if parameters are absent. 2648109998Smarkm [Steve Henson] 2649109998Smarkm 2650109998Smarkm *) In versions up to 0.9.6, RAND_file_name() resorted to file ".rnd" 2651109998Smarkm in the current directory if neither $RANDFILE nor $HOME was set. 2652109998Smarkm RAND_file_name() in 0.9.6a returned NULL in this case. This has 2653109998Smarkm caused some confusion to Windows users who haven't defined $HOME. 2654109998Smarkm Thus RAND_file_name() is changed again: e_os.h can define a 2655109998Smarkm DEFAULT_HOME, which will be used if $HOME is not set. 2656109998Smarkm For Windows, we use "C:"; on other platforms, we still require 2657109998Smarkm environment variables. 2658109998Smarkm 2659109998Smarkm *) Move 'if (!initialized) RAND_poll()' into regions protected by 2660109998Smarkm CRYPTO_LOCK_RAND. This is not strictly necessary, but avoids 2661109998Smarkm having multiple threads call RAND_poll() concurrently. 2662109998Smarkm [Bodo Moeller] 2663109998Smarkm 2664109998Smarkm *) In crypto/rand/md_rand.c, replace 'add_do_not_lock' flag by a 2665109998Smarkm combination of a flag and a thread ID variable. 2666109998Smarkm Otherwise while one thread is in ssleay_rand_bytes (which sets the 2667109998Smarkm flag), *other* threads can enter ssleay_add_bytes without obeying 2668109998Smarkm the CRYPTO_LOCK_RAND lock (and may even illegally release the lock 2669109998Smarkm that they do not hold after the first thread unsets add_do_not_lock). 2670109998Smarkm [Bodo Moeller] 2671109998Smarkm 2672109998Smarkm *) Change bctest again: '-x' expressions are not available in all 2673109998Smarkm versions of 'test'. 2674109998Smarkm [Bodo Moeller] 2675109998Smarkm 2676109998Smarkm Changes between 0.9.6 and 0.9.6a [5 Apr 2001] 2677109998Smarkm 2678109998Smarkm *) Fix a couple of memory leaks in PKCS7_dataDecode() 2679109998Smarkm [Steve Henson, reported by Heyun Zheng <hzheng@atdsprint.com>] 2680109998Smarkm 2681109998Smarkm *) Change Configure and Makefiles to provide EXE_EXT, which will contain 2682109998Smarkm the default extension for executables, if any. Also, make the perl 2683109998Smarkm scripts that use symlink() to test if it really exists and use "cp" 2684109998Smarkm if it doesn't. All this made OpenSSL compilable and installable in 2685109998Smarkm CygWin. 2686109998Smarkm [Richard Levitte] 2687109998Smarkm 2688109998Smarkm *) Fix for asn1_GetSequence() for indefinite length constructed data. 2689109998Smarkm If SEQUENCE is length is indefinite just set c->slen to the total 2690109998Smarkm amount of data available. 2691109998Smarkm [Steve Henson, reported by shige@FreeBSD.org] 2692109998Smarkm [This change does not apply to 0.9.7.] 2693109998Smarkm 2694109998Smarkm *) Change bctest to avoid here-documents inside command substitution 2695109998Smarkm (workaround for FreeBSD /bin/sh bug). 2696109998Smarkm For compatibility with Ultrix, avoid shell functions (introduced 2697109998Smarkm in the bctest version that searches along $PATH). 2698109998Smarkm [Bodo Moeller] 2699109998Smarkm 2700109998Smarkm *) Rename 'des_encrypt' to 'des_encrypt1'. This avoids the clashes 2701109998Smarkm with des_encrypt() defined on some operating systems, like Solaris 2702109998Smarkm and UnixWare. 2703109998Smarkm [Richard Levitte] 2704109998Smarkm 2705109998Smarkm *) Check the result of RSA-CRT (see D. Boneh, R. DeMillo, R. Lipton: 2706109998Smarkm On the Importance of Eliminating Errors in Cryptographic 2707109998Smarkm Computations, J. Cryptology 14 (2001) 2, 101-119, 2708109998Smarkm http://theory.stanford.edu/~dabo/papers/faults.ps.gz). 2709109998Smarkm [Ulf Moeller] 2710109998Smarkm 2711109998Smarkm *) MIPS assembler BIGNUM division bug fix. 2712109998Smarkm [Andy Polyakov] 2713109998Smarkm 2714109998Smarkm *) Disabled incorrect Alpha assembler code. 2715109998Smarkm [Richard Levitte] 2716109998Smarkm 2717109998Smarkm *) Fix PKCS#7 decode routines so they correctly update the length 2718109998Smarkm after reading an EOC for the EXPLICIT tag. 2719109998Smarkm [Steve Henson] 2720109998Smarkm [This change does not apply to 0.9.7.] 2721109998Smarkm 2722109998Smarkm *) Fix bug in PKCS#12 key generation routines. This was triggered 2723109998Smarkm if a 3DES key was generated with a 0 initial byte. Include 2724109998Smarkm PKCS12_BROKEN_KEYGEN compilation option to retain the old 2725109998Smarkm (but broken) behaviour. 2726109998Smarkm [Steve Henson] 2727109998Smarkm 2728109998Smarkm *) Enhance bctest to search for a working bc along $PATH and print 2729109998Smarkm it when found. 2730109998Smarkm [Tim Rice <tim@multitalents.net> via Richard Levitte] 2731109998Smarkm 2732109998Smarkm *) Fix memory leaks in err.c: free err_data string if necessary; 2733109998Smarkm don't write to the wrong index in ERR_set_error_data. 2734109998Smarkm [Bodo Moeller] 2735109998Smarkm 2736109998Smarkm *) Implement ssl23_peek (analogous to ssl23_read), which previously 2737109998Smarkm did not exist. 2738109998Smarkm [Bodo Moeller] 2739109998Smarkm 2740109998Smarkm *) Replace rdtsc with _emit statements for VC++ version 5. 2741109998Smarkm [Jeremy Cooper <jeremy@baymoo.org>] 2742109998Smarkm 2743109998Smarkm *) Make it possible to reuse SSLv2 sessions. 2744109998Smarkm [Richard Levitte] 2745109998Smarkm 2746109998Smarkm *) In copy_email() check for >= 0 as a return value for 2747109998Smarkm X509_NAME_get_index_by_NID() since 0 is a valid index. 2748109998Smarkm [Steve Henson reported by Massimiliano Pala <madwolf@opensca.org>] 2749109998Smarkm 2750109998Smarkm *) Avoid coredump with unsupported or invalid public keys by checking if 2751109998Smarkm X509_get_pubkey() fails in PKCS7_verify(). Fix memory leak when 2752109998Smarkm PKCS7_verify() fails with non detached data. 2753109998Smarkm [Steve Henson] 2754109998Smarkm 2755109998Smarkm *) Don't use getenv in library functions when run as setuid/setgid. 2756109998Smarkm New function OPENSSL_issetugid(). 2757109998Smarkm [Ulf Moeller] 2758109998Smarkm 2759109998Smarkm *) Avoid false positives in memory leak detection code (crypto/mem_dbg.c) 2760109998Smarkm due to incorrect handling of multi-threading: 2761109998Smarkm 2762109998Smarkm 1. Fix timing glitch in the MemCheck_off() portion of CRYPTO_mem_ctrl(). 2763109998Smarkm 2764109998Smarkm 2. Fix logical glitch in is_MemCheck_on() aka CRYPTO_is_mem_check_on(). 2765109998Smarkm 2766109998Smarkm 3. Count how many times MemCheck_off() has been called so that 2767109998Smarkm nested use can be treated correctly. This also avoids 2768109998Smarkm inband-signalling in the previous code (which relied on the 2769109998Smarkm assumption that thread ID 0 is impossible). 2770109998Smarkm [Bodo Moeller] 2771109998Smarkm 2772109998Smarkm *) Add "-rand" option also to s_client and s_server. 2773109998Smarkm [Lutz Jaenicke] 2774109998Smarkm 2775109998Smarkm *) Fix CPU detection on Irix 6.x. 2776109998Smarkm [Kurt Hockenbury <khockenb@stevens-tech.edu> and 2777109998Smarkm "Bruce W. Forsberg" <bruce.forsberg@baesystems.com>] 2778109998Smarkm 2779109998Smarkm *) Fix X509_NAME bug which produced incorrect encoding if X509_NAME 2780109998Smarkm was empty. 2781109998Smarkm [Steve Henson] 2782109998Smarkm [This change does not apply to 0.9.7.] 2783109998Smarkm 2784109998Smarkm *) Use the cached encoding of an X509_NAME structure rather than 2785109998Smarkm copying it. This is apparently the reason for the libsafe "errors" 2786109998Smarkm but the code is actually correct. 2787109998Smarkm [Steve Henson] 2788109998Smarkm 2789109998Smarkm *) Add new function BN_rand_range(), and fix DSA_sign_setup() to prevent 2790109998Smarkm Bleichenbacher's DSA attack. 2791109998Smarkm Extend BN_[pseudo_]rand: As before, top=1 forces the highest two bits 2792109998Smarkm to be set and top=0 forces the highest bit to be set; top=-1 is new 2793109998Smarkm and leaves the highest bit random. 2794109998Smarkm [Ulf Moeller, Bodo Moeller] 2795109998Smarkm 2796109998Smarkm *) In the NCONF_...-based implementations for CONF_... queries 2797109998Smarkm (crypto/conf/conf_lib.c), if the input LHASH is NULL, avoid using 2798109998Smarkm a temporary CONF structure with the data component set to NULL 2799109998Smarkm (which gives segmentation faults in lh_retrieve). 2800109998Smarkm Instead, use NULL for the CONF pointer in CONF_get_string and 2801109998Smarkm CONF_get_number (which may use environment variables) and directly 2802109998Smarkm return NULL from CONF_get_section. 2803109998Smarkm [Bodo Moeller] 2804109998Smarkm 2805109998Smarkm *) Fix potential buffer overrun for EBCDIC. 2806109998Smarkm [Ulf Moeller] 2807109998Smarkm 2808109998Smarkm *) Tolerate nonRepudiation as being valid for S/MIME signing and certSign 2809109998Smarkm keyUsage if basicConstraints absent for a CA. 2810109998Smarkm [Steve Henson] 2811109998Smarkm 2812109998Smarkm *) Make SMIME_write_PKCS7() write mail header values with a format that 2813109998Smarkm is more generally accepted (no spaces before the semicolon), since 2814109998Smarkm some programs can't parse those values properly otherwise. Also make 2815109998Smarkm sure BIO's that break lines after each write do not create invalid 2816109998Smarkm headers. 2817109998Smarkm [Richard Levitte] 2818109998Smarkm 2819109998Smarkm *) Make the CRL encoding routines work with empty SEQUENCE OF. The 2820109998Smarkm macros previously used would not encode an empty SEQUENCE OF 2821109998Smarkm and break the signature. 2822109998Smarkm [Steve Henson] 2823109998Smarkm [This change does not apply to 0.9.7.] 2824109998Smarkm 2825109998Smarkm *) Zero the premaster secret after deriving the master secret in 2826109998Smarkm DH ciphersuites. 2827109998Smarkm [Steve Henson] 2828109998Smarkm 2829109998Smarkm *) Add some EVP_add_digest_alias registrations (as found in 2830109998Smarkm OpenSSL_add_all_digests()) to SSL_library_init() 2831109998Smarkm aka OpenSSL_add_ssl_algorithms(). This provides improved 2832109998Smarkm compatibility with peers using X.509 certificates 2833109998Smarkm with unconventional AlgorithmIdentifier OIDs. 2834109998Smarkm [Bodo Moeller] 2835109998Smarkm 2836109998Smarkm *) Fix for Irix with NO_ASM. 2837109998Smarkm ["Bruce W. Forsberg" <bruce.forsberg@baesystems.com>] 2838109998Smarkm 2839109998Smarkm *) ./config script fixes. 2840109998Smarkm [Ulf Moeller, Richard Levitte] 2841109998Smarkm 2842109998Smarkm *) Fix 'openssl passwd -1'. 2843109998Smarkm [Bodo Moeller] 2844109998Smarkm 2845109998Smarkm *) Change PKCS12_key_gen_asc() so it can cope with non null 2846109998Smarkm terminated strings whose length is passed in the passlen 2847109998Smarkm parameter, for example from PEM callbacks. This was done 2848109998Smarkm by adding an extra length parameter to asc2uni(). 2849109998Smarkm [Steve Henson, reported by <oddissey@samsung.co.kr>] 2850109998Smarkm 2851109998Smarkm *) Fix C code generated by 'openssl dsaparam -C': If a BN_bin2bn 2852109998Smarkm call failed, free the DSA structure. 2853109998Smarkm [Bodo Moeller] 2854109998Smarkm 2855109998Smarkm *) Fix to uni2asc() to cope with zero length Unicode strings. 2856109998Smarkm These are present in some PKCS#12 files. 2857109998Smarkm [Steve Henson] 2858109998Smarkm 2859109998Smarkm *) Increase s2->wbuf allocation by one byte in ssl2_new (ssl/s2_lib.c). 2860109998Smarkm Otherwise do_ssl_write (ssl/s2_pkt.c) will write beyond buffer limits 2861109998Smarkm when writing a 32767 byte record. 2862109998Smarkm [Bodo Moeller; problem reported by Eric Day <eday@concentric.net>] 2863109998Smarkm 2864109998Smarkm *) In RSA_eay_public_{en,ed}crypt and RSA_eay_mod_exp (rsa_eay.c), 2865109998Smarkm obtain lock CRYPTO_LOCK_RSA before setting rsa->_method_mod_{n,p,q}. 2866109998Smarkm 2867109998Smarkm (RSA objects have a reference count access to which is protected 2868109998Smarkm by CRYPTO_LOCK_RSA [see rsa_lib.c, s3_srvr.c, ssl_cert.c, ssl_rsa.c], 2869109998Smarkm so they are meant to be shared between threads.) 2870109998Smarkm [Bodo Moeller, Geoff Thorpe; original patch submitted by 2871109998Smarkm "Reddie, Steven" <Steven.Reddie@ca.com>] 2872109998Smarkm 2873109998Smarkm *) Fix a deadlock in CRYPTO_mem_leaks(). 2874109998Smarkm [Bodo Moeller] 2875109998Smarkm 2876109998Smarkm *) Use better test patterns in bntest. 2877109998Smarkm [Ulf M�ller] 2878109998Smarkm 2879109998Smarkm *) rand_win.c fix for Borland C. 2880109998Smarkm [Ulf M�ller] 2881109998Smarkm 2882109998Smarkm *) BN_rshift bugfix for n == 0. 2883109998Smarkm [Bodo Moeller] 2884109998Smarkm 2885109998Smarkm *) Add a 'bctest' script that checks for some known 'bc' bugs 2886109998Smarkm so that 'make test' does not abort just because 'bc' is broken. 2887109998Smarkm [Bodo Moeller] 2888109998Smarkm 2889109998Smarkm *) Store verify_result within SSL_SESSION also for client side to 2890109998Smarkm avoid potential security hole. (Re-used sessions on the client side 2891109998Smarkm always resulted in verify_result==X509_V_OK, not using the original 2892109998Smarkm result of the server certificate verification.) 2893109998Smarkm [Lutz Jaenicke] 2894109998Smarkm 2895109998Smarkm *) Fix ssl3_pending: If the record in s->s3->rrec is not of type 2896109998Smarkm SSL3_RT_APPLICATION_DATA, return 0. 2897109998Smarkm Similarly, change ssl2_pending to return 0 if SSL_in_init(s) is true. 2898109998Smarkm [Bodo Moeller] 2899109998Smarkm 2900109998Smarkm *) Fix SSL_peek: 2901109998Smarkm Both ssl2_peek and ssl3_peek, which were totally broken in earlier 2902109998Smarkm releases, have been re-implemented by renaming the previous 2903109998Smarkm implementations of ssl2_read and ssl3_read to ssl2_read_internal 2904109998Smarkm and ssl3_read_internal, respectively, and adding 'peek' parameters 2905109998Smarkm to them. The new ssl[23]_{read,peek} functions are calls to 2906109998Smarkm ssl[23]_read_internal with the 'peek' flag set appropriately. 2907109998Smarkm A 'peek' parameter has also been added to ssl3_read_bytes, which 2908109998Smarkm does the actual work for ssl3_read_internal. 2909109998Smarkm [Bodo Moeller] 2910109998Smarkm 2911109998Smarkm *) Initialise "ex_data" member of RSA/DSA/DH structures prior to calling 2912109998Smarkm the method-specific "init()" handler. Also clean up ex_data after 2913109998Smarkm calling the method-specific "finish()" handler. Previously, this was 2914109998Smarkm happening the other way round. 2915109998Smarkm [Geoff Thorpe] 2916109998Smarkm 2917109998Smarkm *) Increase BN_CTX_NUM (the number of BIGNUMs in a BN_CTX) to 16. 2918109998Smarkm The previous value, 12, was not always sufficient for BN_mod_exp(). 2919109998Smarkm [Bodo Moeller] 2920109998Smarkm 2921109998Smarkm *) Make sure that shared libraries get the internal name engine with 2922109998Smarkm the full version number and not just 0. This should mark the 2923109998Smarkm shared libraries as not backward compatible. Of course, this should 2924109998Smarkm be changed again when we can guarantee backward binary compatibility. 2925109998Smarkm [Richard Levitte] 2926109998Smarkm 2927109998Smarkm *) Fix typo in get_cert_by_subject() in by_dir.c 2928109998Smarkm [Jean-Marc Desperrier <jean-marc.desperrier@certplus.com>] 2929109998Smarkm 2930109998Smarkm *) Rework the system to generate shared libraries: 2931109998Smarkm 2932109998Smarkm - Make note of the expected extension for the shared libraries and 2933109998Smarkm if there is a need for symbolic links from for example libcrypto.so.0 2934109998Smarkm to libcrypto.so.0.9.7. There is extended info in Configure for 2935109998Smarkm that. 2936109998Smarkm 2937109998Smarkm - Make as few rebuilds of the shared libraries as possible. 2938109998Smarkm 2939109998Smarkm - Still avoid linking the OpenSSL programs with the shared libraries. 2940109998Smarkm 2941109998Smarkm - When installing, install the shared libraries separately from the 2942109998Smarkm static ones. 2943109998Smarkm [Richard Levitte] 2944109998Smarkm 2945109998Smarkm *) Fix SSL_CTX_set_read_ahead macro to actually use its argument. 2946109998Smarkm 2947109998Smarkm Copy SSL_CTX's read_ahead flag to SSL object directly in SSL_new 2948109998Smarkm and not in SSL_clear because the latter is also used by the 2949109998Smarkm accept/connect functions; previously, the settings made by 2950109998Smarkm SSL_set_read_ahead would be lost during the handshake. 2951109998Smarkm [Bodo Moeller; problems reported by Anders Gertz <gertz@epact.se>] 2952109998Smarkm 2953109998Smarkm *) Correct util/mkdef.pl to be selective about disabled algorithms. 2954109998Smarkm Previously, it would create entries for disableed algorithms no 2955109998Smarkm matter what. 2956109998Smarkm [Richard Levitte] 2957109998Smarkm 2958109998Smarkm *) Added several new manual pages for SSL_* function. 2959109998Smarkm [Lutz Jaenicke] 2960109998Smarkm 2961109998Smarkm Changes between 0.9.5a and 0.9.6 [24 Sep 2000] 2962109998Smarkm 2963109998Smarkm *) In ssl23_get_client_hello, generate an error message when faced 2964109998Smarkm with an initial SSL 3.0/TLS record that is too small to contain the 2965109998Smarkm first two bytes of the ClientHello message, i.e. client_version. 2966109998Smarkm (Note that this is a pathologic case that probably has never happened 2967109998Smarkm in real life.) The previous approach was to use the version number 2968109998Smarkm from the record header as a substitute; but our protocol choice 2969109998Smarkm should not depend on that one because it is not authenticated 2970109998Smarkm by the Finished messages. 2971109998Smarkm [Bodo Moeller] 2972109998Smarkm 2973109998Smarkm *) More robust randomness gathering functions for Windows. 2974109998Smarkm [Jeffrey Altman <jaltman@columbia.edu>] 2975109998Smarkm 2976109998Smarkm *) For compatibility reasons if the flag X509_V_FLAG_ISSUER_CHECK is 2977109998Smarkm not set then we don't setup the error code for issuer check errors 2978109998Smarkm to avoid possibly overwriting other errors which the callback does 2979109998Smarkm handle. If an application does set the flag then we assume it knows 2980109998Smarkm what it is doing and can handle the new informational codes 2981109998Smarkm appropriately. 2982109998Smarkm [Steve Henson] 2983109998Smarkm 2984109998Smarkm *) Fix for a nasty bug in ASN1_TYPE handling. ASN1_TYPE is used for 2985109998Smarkm a general "ANY" type, as such it should be able to decode anything 2986109998Smarkm including tagged types. However it didn't check the class so it would 2987109998Smarkm wrongly interpret tagged types in the same way as their universal 2988109998Smarkm counterpart and unknown types were just rejected. Changed so that the 2989109998Smarkm tagged and unknown types are handled in the same way as a SEQUENCE: 2990109998Smarkm that is the encoding is stored intact. There is also a new type 2991109998Smarkm "V_ASN1_OTHER" which is used when the class is not universal, in this 2992109998Smarkm case we have no idea what the actual type is so we just lump them all 2993109998Smarkm together. 2994109998Smarkm [Steve Henson] 2995109998Smarkm 2996109998Smarkm *) On VMS, stdout may very well lead to a file that is written to 2997109998Smarkm in a record-oriented fashion. That means that every write() will 2998109998Smarkm write a separate record, which will be read separately by the 2999109998Smarkm programs trying to read from it. This can be very confusing. 3000109998Smarkm 3001109998Smarkm The solution is to put a BIO filter in the way that will buffer 3002109998Smarkm text until a linefeed is reached, and then write everything a 3003109998Smarkm line at a time, so every record written will be an actual line, 3004109998Smarkm not chunks of lines and not (usually doesn't happen, but I've 3005109998Smarkm seen it once) several lines in one record. BIO_f_linebuffer() is 3006109998Smarkm the answer. 3007109998Smarkm 3008109998Smarkm Currently, it's a VMS-only method, because that's where it has 3009109998Smarkm been tested well enough. 3010109998Smarkm [Richard Levitte] 3011109998Smarkm 3012109998Smarkm *) Remove 'optimized' squaring variant in BN_mod_mul_montgomery, 3013109998Smarkm it can return incorrect results. 3014109998Smarkm (Note: The buggy variant was not enabled in OpenSSL 0.9.5a, 3015109998Smarkm but it was in 0.9.6-beta[12].) 3016109998Smarkm [Bodo Moeller] 3017109998Smarkm 3018109998Smarkm *) Disable the check for content being present when verifying detached 3019109998Smarkm signatures in pk7_smime.c. Some versions of Netscape (wrongly) 3020109998Smarkm include zero length content when signing messages. 3021109998Smarkm [Steve Henson] 3022109998Smarkm 3023109998Smarkm *) New BIO_shutdown_wr macro, which invokes the BIO_C_SHUTDOWN_WR 3024109998Smarkm BIO_ctrl (for BIO pairs). 3025109998Smarkm [Bodo M�ller] 3026109998Smarkm 3027109998Smarkm *) Add DSO method for VMS. 3028109998Smarkm [Richard Levitte] 3029109998Smarkm 3030109998Smarkm *) Bug fix: Montgomery multiplication could produce results with the 3031109998Smarkm wrong sign. 3032109998Smarkm [Ulf M�ller] 3033109998Smarkm 3034109998Smarkm *) Add RPM specification openssl.spec and modify it to build three 3035109998Smarkm packages. The default package contains applications, application 3036109998Smarkm documentation and run-time libraries. The devel package contains 3037109998Smarkm include files, static libraries and function documentation. The 3038109998Smarkm doc package contains the contents of the doc directory. The original 3039109998Smarkm openssl.spec was provided by Damien Miller <djm@mindrot.org>. 3040109998Smarkm [Richard Levitte] 3041109998Smarkm 3042109998Smarkm *) Add a large number of documentation files for many SSL routines. 3043109998Smarkm [Lutz Jaenicke <Lutz.Jaenicke@aet.TU-Cottbus.DE>] 3044109998Smarkm 3045109998Smarkm *) Add a configuration entry for Sony News 4. 3046109998Smarkm [NAKAJI Hiroyuki <nakaji@tutrp.tut.ac.jp>] 3047109998Smarkm 3048109998Smarkm *) Don't set the two most significant bits to one when generating a 3049109998Smarkm random number < q in the DSA library. 3050109998Smarkm [Ulf M�ller] 3051109998Smarkm 3052109998Smarkm *) New SSL API mode 'SSL_MODE_AUTO_RETRY'. This disables the default 3053109998Smarkm behaviour that SSL_read may result in SSL_ERROR_WANT_READ (even if 3054109998Smarkm the underlying transport is blocking) if a handshake took place. 3055109998Smarkm (The default behaviour is needed by applications such as s_client 3056109998Smarkm and s_server that use select() to determine when to use SSL_read; 3057109998Smarkm but for applications that know in advance when to expect data, it 3058109998Smarkm just makes things more complicated.) 3059109998Smarkm [Bodo Moeller] 3060109998Smarkm 3061109998Smarkm *) Add RAND_egd_bytes(), which gives control over the number of bytes read 3062109998Smarkm from EGD. 3063109998Smarkm [Ben Laurie] 3064109998Smarkm 3065109998Smarkm *) Add a few more EBCDIC conditionals that make `req' and `x509' 3066109998Smarkm work better on such systems. 3067109998Smarkm [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>] 3068109998Smarkm 3069109998Smarkm *) Add two demo programs for PKCS12_parse() and PKCS12_create(). 3070109998Smarkm Update PKCS12_parse() so it copies the friendlyName and the 3071109998Smarkm keyid to the certificates aux info. 3072109998Smarkm [Steve Henson] 3073109998Smarkm 3074109998Smarkm *) Fix bug in PKCS7_verify() which caused an infinite loop 3075109998Smarkm if there was more than one signature. 3076109998Smarkm [Sven Uszpelkat <su@celocom.de>] 3077109998Smarkm 3078109998Smarkm *) Major change in util/mkdef.pl to include extra information 3079109998Smarkm about each symbol, as well as presentig variables as well 3080109998Smarkm as functions. This change means that there's n more need 3081109998Smarkm to rebuild the .num files when some algorithms are excluded. 3082109998Smarkm [Richard Levitte] 3083109998Smarkm 3084109998Smarkm *) Allow the verify time to be set by an application, 3085109998Smarkm rather than always using the current time. 3086109998Smarkm [Steve Henson] 3087109998Smarkm 3088109998Smarkm *) Phase 2 verify code reorganisation. The certificate 3089109998Smarkm verify code now looks up an issuer certificate by a 3090109998Smarkm number of criteria: subject name, authority key id 3091109998Smarkm and key usage. It also verifies self signed certificates 3092109998Smarkm by the same criteria. The main comparison function is 3093109998Smarkm X509_check_issued() which performs these checks. 3094109998Smarkm 3095109998Smarkm Lot of changes were necessary in order to support this 3096109998Smarkm without completely rewriting the lookup code. 3097109998Smarkm 3098109998Smarkm Authority and subject key identifier are now cached. 3099109998Smarkm 3100109998Smarkm The LHASH 'certs' is X509_STORE has now been replaced 3101109998Smarkm by a STACK_OF(X509_OBJECT). This is mainly because an 3102109998Smarkm LHASH can't store or retrieve multiple objects with 3103109998Smarkm the same hash value. 3104109998Smarkm 3105109998Smarkm As a result various functions (which were all internal 3106109998Smarkm use only) have changed to handle the new X509_STORE 3107109998Smarkm structure. This will break anything that messed round 3108109998Smarkm with X509_STORE internally. 3109109998Smarkm 3110109998Smarkm The functions X509_STORE_add_cert() now checks for an 3111109998Smarkm exact match, rather than just subject name. 3112109998Smarkm 3113109998Smarkm The X509_STORE API doesn't directly support the retrieval 3114109998Smarkm of multiple certificates matching a given criteria, however 3115109998Smarkm this can be worked round by performing a lookup first 3116109998Smarkm (which will fill the cache with candidate certificates) 3117109998Smarkm and then examining the cache for matches. This is probably 3118109998Smarkm the best we can do without throwing out X509_LOOKUP 3119109998Smarkm entirely (maybe later...). 3120109998Smarkm 3121109998Smarkm The X509_VERIFY_CTX structure has been enhanced considerably. 3122109998Smarkm 3123109998Smarkm All certificate lookup operations now go via a get_issuer() 3124109998Smarkm callback. Although this currently uses an X509_STORE it 3125109998Smarkm can be replaced by custom lookups. This is a simple way 3126109998Smarkm to bypass the X509_STORE hackery necessary to make this 3127109998Smarkm work and makes it possible to use more efficient techniques 3128109998Smarkm in future. A very simple version which uses a simple 3129109998Smarkm STACK for its trusted certificate store is also provided 3130109998Smarkm using X509_STORE_CTX_trusted_stack(). 3131109998Smarkm 3132109998Smarkm The verify_cb() and verify() callbacks now have equivalents 3133109998Smarkm in the X509_STORE_CTX structure. 3134109998Smarkm 3135109998Smarkm X509_STORE_CTX also has a 'flags' field which can be used 3136109998Smarkm to customise the verify behaviour. 3137109998Smarkm [Steve Henson] 3138109998Smarkm 3139109998Smarkm *) Add new PKCS#7 signing option PKCS7_NOSMIMECAP which 3140109998Smarkm excludes S/MIME capabilities. 3141109998Smarkm [Steve Henson] 3142109998Smarkm 3143109998Smarkm *) When a certificate request is read in keep a copy of the 3144109998Smarkm original encoding of the signed data and use it when outputing 3145109998Smarkm again. Signatures then use the original encoding rather than 3146109998Smarkm a decoded, encoded version which may cause problems if the 3147109998Smarkm request is improperly encoded. 3148109998Smarkm [Steve Henson] 3149109998Smarkm 3150109998Smarkm *) For consistency with other BIO_puts implementations, call 3151109998Smarkm buffer_write(b, ...) directly in buffer_puts instead of calling 3152109998Smarkm BIO_write(b, ...). 3153109998Smarkm 3154109998Smarkm In BIO_puts, increment b->num_write as in BIO_write. 3155109998Smarkm [Peter.Sylvester@EdelWeb.fr] 3156109998Smarkm 3157109998Smarkm *) Fix BN_mul_word for the case where the word is 0. (We have to use 3158109998Smarkm BN_zero, we may not return a BIGNUM with an array consisting of 3159109998Smarkm words set to zero.) 3160109998Smarkm [Bodo Moeller] 3161109998Smarkm 3162109998Smarkm *) Avoid calling abort() from within the library when problems are 3163109998Smarkm detected, except if preprocessor symbols have been defined 3164109998Smarkm (such as REF_CHECK, BN_DEBUG etc.). 3165109998Smarkm [Bodo Moeller] 3166109998Smarkm 3167109998Smarkm *) New openssl application 'rsautl'. This utility can be 3168109998Smarkm used for low level RSA operations. DER public key 3169109998Smarkm BIO/fp routines also added. 3170109998Smarkm [Steve Henson] 3171109998Smarkm 3172109998Smarkm *) New Configure entry and patches for compiling on QNX 4. 3173109998Smarkm [Andreas Schneider <andreas@ds3.etech.fh-hamburg.de>] 3174109998Smarkm 3175109998Smarkm *) A demo state-machine implementation was sponsored by 3176109998Smarkm Nuron (http://www.nuron.com/) and is now available in 3177109998Smarkm demos/state_machine. 3178109998Smarkm [Ben Laurie] 3179109998Smarkm 3180109998Smarkm *) New options added to the 'dgst' utility for signature 3181109998Smarkm generation and verification. 3182109998Smarkm [Steve Henson] 3183109998Smarkm 3184109998Smarkm *) Unrecognized PKCS#7 content types are now handled via a 3185109998Smarkm catch all ASN1_TYPE structure. This allows unsupported 3186109998Smarkm types to be stored as a "blob" and an application can 3187109998Smarkm encode and decode it manually. 3188109998Smarkm [Steve Henson] 3189109998Smarkm 3190109998Smarkm *) Fix various signed/unsigned issues to make a_strex.c 3191109998Smarkm compile under VC++. 3192109998Smarkm [Oscar Jacobsson <oscar.jacobsson@celocom.com>] 3193109998Smarkm 3194109998Smarkm *) ASN1 fixes. i2d_ASN1_OBJECT was not returning the correct 3195109998Smarkm length if passed a buffer. ASN1_INTEGER_to_BN failed 3196109998Smarkm if passed a NULL BN and its argument was negative. 3197109998Smarkm [Steve Henson, pointed out by Sven Heiberg <sven@tartu.cyber.ee>] 3198109998Smarkm 3199109998Smarkm *) Modification to PKCS#7 encoding routines to output definite 3200109998Smarkm length encoding. Since currently the whole structures are in 3201109998Smarkm memory there's not real point in using indefinite length 3202109998Smarkm constructed encoding. However if OpenSSL is compiled with 3203109998Smarkm the flag PKCS7_INDEFINITE_ENCODING the old form is used. 3204109998Smarkm [Steve Henson] 3205109998Smarkm 3206109998Smarkm *) Added BIO_vprintf() and BIO_vsnprintf(). 3207109998Smarkm [Richard Levitte] 3208109998Smarkm 3209109998Smarkm *) Added more prefixes to parse for in the the strings written 3210109998Smarkm through a logging bio, to cover all the levels that are available 3211109998Smarkm through syslog. The prefixes are now: 3212109998Smarkm 3213111147Snectar PANIC, EMERG, EMR => LOG_EMERG 3214111147Snectar ALERT, ALR => LOG_ALERT 3215111147Snectar CRIT, CRI => LOG_CRIT 3216111147Snectar ERROR, ERR => LOG_ERR 3217111147Snectar WARNING, WARN, WAR => LOG_WARNING 3218109998Smarkm NOTICE, NOTE, NOT => LOG_NOTICE 3219109998Smarkm INFO, INF => LOG_INFO 3220109998Smarkm DEBUG, DBG => LOG_DEBUG 3221109998Smarkm 3222109998Smarkm and as before, if none of those prefixes are present at the 3223109998Smarkm beginning of the string, LOG_ERR is chosen. 3224109998Smarkm 3225109998Smarkm On Win32, the LOG_* levels are mapped according to this: 3226111147Snectar 3227109998Smarkm LOG_EMERG, LOG_ALERT, LOG_CRIT, LOG_ERR => EVENTLOG_ERROR_TYPE 3228109998Smarkm LOG_WARNING => EVENTLOG_WARNING_TYPE 3229109998Smarkm LOG_NOTICE, LOG_INFO, LOG_DEBUG => EVENTLOG_INFORMATION_TYPE 3230109998Smarkm 3231109998Smarkm [Richard Levitte] 3232109998Smarkm 3233109998Smarkm *) Made it possible to reconfigure with just the configuration 3234109998Smarkm argument "reconf" or "reconfigure". The command line arguments 3235109998Smarkm are stored in Makefile.ssl in the variable CONFIGURE_ARGS, 3236109998Smarkm and are retrieved from there when reconfiguring. 3237109998Smarkm [Richard Levitte] 3238109998Smarkm 3239109998Smarkm *) MD4 implemented. 3240109998Smarkm [Assar Westerlund <assar@sics.se>, Richard Levitte] 3241109998Smarkm 3242109998Smarkm *) Add the arguments -CAfile and -CApath to the pkcs12 utility. 3243109998Smarkm [Richard Levitte] 3244109998Smarkm 3245109998Smarkm *) The obj_dat.pl script was messing up the sorting of object 3246109998Smarkm names. The reason was that it compared the quoted version 3247109998Smarkm of strings as a result "OCSP" > "OCSP Signing" because 3248109998Smarkm " > SPACE. Changed script to store unquoted versions of 3249109998Smarkm names and add quotes on output. It was also omitting some 3250109998Smarkm names from the lookup table if they were given a default 3251109998Smarkm value (that is if SN is missing it is given the same 3252109998Smarkm value as LN and vice versa), these are now added on the 3253109998Smarkm grounds that if an object has a name we should be able to 3254109998Smarkm look it up. Finally added warning output when duplicate 3255109998Smarkm short or long names are found. 3256109998Smarkm [Steve Henson] 3257109998Smarkm 3258109998Smarkm *) Changes needed for Tandem NSK. 3259109998Smarkm [Scott Uroff <scott@xypro.com>] 3260109998Smarkm 3261109998Smarkm *) Fix SSL 2.0 rollback checking: Due to an off-by-one error in 3262109998Smarkm RSA_padding_check_SSLv23(), special padding was never detected 3263109998Smarkm and thus the SSL 3.0/TLS 1.0 countermeasure against protocol 3264109998Smarkm version rollback attacks was not effective. 3265109998Smarkm 3266109998Smarkm In s23_clnt.c, don't use special rollback-attack detection padding 3267109998Smarkm (RSA_SSLV23_PADDING) if SSL 2.0 is the only protocol enabled in the 3268109998Smarkm client; similarly, in s23_srvr.c, don't do the rollback check if 3269109998Smarkm SSL 2.0 is the only protocol enabled in the server. 3270109998Smarkm [Bodo Moeller] 3271109998Smarkm 3272109998Smarkm *) Make it possible to get hexdumps of unprintable data with 'openssl 3273109998Smarkm asn1parse'. By implication, the functions ASN1_parse_dump() and 3274109998Smarkm BIO_dump_indent() are added. 3275109998Smarkm [Richard Levitte] 3276109998Smarkm 3277109998Smarkm *) New functions ASN1_STRING_print_ex() and X509_NAME_print_ex() 3278109998Smarkm these print out strings and name structures based on various 3279109998Smarkm flags including RFC2253 support and proper handling of 3280109998Smarkm multibyte characters. Added options to the 'x509' utility 3281109998Smarkm to allow the various flags to be set. 3282109998Smarkm [Steve Henson] 3283109998Smarkm 3284109998Smarkm *) Various fixes to use ASN1_TIME instead of ASN1_UTCTIME. 3285109998Smarkm Also change the functions X509_cmp_current_time() and 3286109998Smarkm X509_gmtime_adj() work with an ASN1_TIME structure, 3287109998Smarkm this will enable certificates using GeneralizedTime in validity 3288109998Smarkm dates to be checked. 3289109998Smarkm [Steve Henson] 3290109998Smarkm 3291109998Smarkm *) Make the NEG_PUBKEY_BUG code (which tolerates invalid 3292109998Smarkm negative public key encodings) on by default, 3293109998Smarkm NO_NEG_PUBKEY_BUG can be set to disable it. 3294109998Smarkm [Steve Henson] 3295109998Smarkm 3296109998Smarkm *) New function c2i_ASN1_OBJECT() which acts on ASN1_OBJECT 3297109998Smarkm content octets. An i2c_ASN1_OBJECT is unnecessary because 3298109998Smarkm the encoding can be trivially obtained from the structure. 3299109998Smarkm [Steve Henson] 3300109998Smarkm 3301109998Smarkm *) crypto/err.c locking bugfix: Use write locks (CRYPTO_w_[un]lock), 3302109998Smarkm not read locks (CRYPTO_r_[un]lock). 3303109998Smarkm [Bodo Moeller] 3304109998Smarkm 3305109998Smarkm *) A first attempt at creating official support for shared 3306109998Smarkm libraries through configuration. I've kept it so the 3307109998Smarkm default is static libraries only, and the OpenSSL programs 3308109998Smarkm are always statically linked for now, but there are 3309109998Smarkm preparations for dynamic linking in place. 3310109998Smarkm This has been tested on Linux and Tru64. 3311109998Smarkm [Richard Levitte] 3312109998Smarkm 3313109998Smarkm *) Randomness polling function for Win9x, as described in: 3314109998Smarkm Peter Gutmann, Software Generation of Practically Strong 3315109998Smarkm Random Numbers. 3316109998Smarkm [Ulf M�ller] 3317109998Smarkm 3318109998Smarkm *) Fix so PRNG is seeded in req if using an already existing 3319109998Smarkm DSA key. 3320109998Smarkm [Steve Henson] 3321109998Smarkm 3322109998Smarkm *) New options to smime application. -inform and -outform 3323109998Smarkm allow alternative formats for the S/MIME message including 3324109998Smarkm PEM and DER. The -content option allows the content to be 3325109998Smarkm specified separately. This should allow things like Netscape 3326109998Smarkm form signing output easier to verify. 3327109998Smarkm [Steve Henson] 3328109998Smarkm 3329109998Smarkm *) Fix the ASN1 encoding of tags using the 'long form'. 3330109998Smarkm [Steve Henson] 3331109998Smarkm 3332109998Smarkm *) New ASN1 functions, i2c_* and c2i_* for INTEGER and BIT 3333109998Smarkm STRING types. These convert content octets to and from the 3334109998Smarkm underlying type. The actual tag and length octets are 3335109998Smarkm already assumed to have been read in and checked. These 3336109998Smarkm are needed because all other string types have virtually 3337109998Smarkm identical handling apart from the tag. By having versions 3338109998Smarkm of the ASN1 functions that just operate on content octets 3339109998Smarkm IMPLICIT tagging can be handled properly. It also allows 3340109998Smarkm the ASN1_ENUMERATED code to be cut down because ASN1_ENUMERATED 3341109998Smarkm and ASN1_INTEGER are identical apart from the tag. 3342109998Smarkm [Steve Henson] 3343109998Smarkm 3344109998Smarkm *) Change the handling of OID objects as follows: 3345109998Smarkm 3346109998Smarkm - New object identifiers are inserted in objects.txt, following 3347109998Smarkm the syntax given in objects.README. 3348109998Smarkm - objects.pl is used to process obj_mac.num and create a new 3349109998Smarkm obj_mac.h. 3350109998Smarkm - obj_dat.pl is used to create a new obj_dat.h, using the data in 3351109998Smarkm obj_mac.h. 3352109998Smarkm 3353109998Smarkm This is currently kind of a hack, and the perl code in objects.pl 3354109998Smarkm isn't very elegant, but it works as I intended. The simplest way 3355109998Smarkm to check that it worked correctly is to look in obj_dat.h and 3356109998Smarkm check the array nid_objs and make sure the objects haven't moved 3357109998Smarkm around (this is important!). Additions are OK, as well as 3358142425Snectar consistent name changes. 3359142425Snectar [Richard Levitte] 3360142425Snectar 3361160814Ssimon *) Add BSD-style MD5-based passwords to 'openssl passwd' (option '-1'). 3362142425Snectar [Bodo Moeller] 3363142425Snectar 3364142425Snectar *) Addition of the command line parameter '-rand file' to 'openssl req'. 3365142425Snectar The given file adds to whatever has already been seeded into the 3366142425Snectar random pool through the RANDFILE configuration file option or 3367142425Snectar environment variable, or the default random state file. 3368142425Snectar [Richard Levitte] 3369160814Ssimon 3370142425Snectar *) mkstack.pl now sorts each macro group into lexical order. 3371142425Snectar Previously the output order depended on the order the files 3372120631Snectar appeared in the directory, resulting in needless rewriting 3373120631Snectar of safestack.h . 3374120631Snectar [Steve Henson] 3375120631Snectar 3376120631Snectar *) Patches to make OpenSSL compile under Win32 again. Mostly 3377160814Ssimon work arounds for the VC++ problem that it treats func() as 3378120631Snectar func(void). Also stripped out the parts of mkdef.pl that 3379120631Snectar added extra typesafe functions: these no longer exist. 3380120631Snectar [Steve Henson] 3381120631Snectar 3382120631Snectar *) Reorganisation of the stack code. The macros are now all 3383120631Snectar collected in safestack.h . Each macro is defined in terms of 3384120631Snectar a "stack macro" of the form SKM_<name>(type, a, b). The 3385120631Snectar DEBUG_SAFESTACK is now handled in terms of function casts, 3386120631Snectar this has the advantage of retaining type safety without the 3387120631Snectar use of additional functions. If DEBUG_SAFESTACK is not defined 3388120631Snectar then the non typesafe macros are used instead. Also modified the 3389120631Snectar mkstack.pl script to handle the new form. Needs testing to see 3390120631Snectar if which (if any) compilers it chokes and maybe make DEBUG_SAFESTACK 3391120631Snectar the default if no major problems. Similar behaviour for ASN1_SET_OF 3392120631Snectar and PKCS12_STACK_OF. 3393120631Snectar [Steve Henson] 3394120631Snectar 3395120631Snectar *) When some versions of IIS use the 'NET' form of private key the 3396120631Snectar key derivation algorithm is different. Normally MD5(password) is 3397120631Snectar used as a 128 bit RC4 key. In the modified case 3398120631Snectar MD5(MD5(password) + "SGCKEYSALT") is used insted. Added some 3399120631Snectar new functions i2d_RSA_NET(), d2i_RSA_NET() etc which are the same 3400120631Snectar as the old Netscape_RSA functions except they have an additional 3401120631Snectar 'sgckey' parameter which uses the modified algorithm. Also added 3402120631Snectar an -sgckey command line option to the rsa utility. Thanks to 3403120631Snectar Adrian Peck <bertie@ncipher.com> for posting details of the modified 3404120631Snectar algorithm to openssl-dev. 3405120631Snectar [Steve Henson] 3406120631Snectar 3407120631Snectar *) The evp_local.h macros were using 'c.##kname' which resulted in 3408120631Snectar invalid expansion on some systems (SCO 5.0.5 for example). 3409120631Snectar Corrected to 'c.kname'. 3410120631Snectar [Phillip Porch <root@theporch.com>] 3411120631Snectar 3412120631Snectar *) New X509_get1_email() and X509_REQ_get1_email() functions that return 3413120631Snectar a STACK of email addresses from a certificate or request, these look 3414120631Snectar in the subject name and the subject alternative name extensions and 3415120631Snectar omit any duplicate addresses. 3416120631Snectar [Steve Henson] 3417120631Snectar 3418120631Snectar *) Re-implement BN_mod_exp2_mont using independent (and larger) windows. 3419120631Snectar This makes DSA verification about 2 % faster. 3420120631Snectar [Bodo Moeller] 3421120631Snectar 3422120631Snectar *) Increase maximum window size in BN_mod_exp_... to 6 bits instead of 5 3423111147Snectar (meaning that now 2^5 values will be precomputed, which is only 4 KB 3424111147Snectar plus overhead for 1024 bit moduli). 3425111147Snectar This makes exponentiations about 0.5 % faster for 1024 bit 3426111147Snectar exponents (as measured by "openssl speed rsa2048"). 3427111147Snectar [Bodo Moeller] 3428111147Snectar 3429160814Ssimon *) Rename memory handling macros to avoid conflicts with other 3430111147Snectar software: 3431111147Snectar Malloc => OPENSSL_malloc 3432111147Snectar Malloc_locked => OPENSSL_malloc_locked 3433111147Snectar Realloc => OPENSSL_realloc 3434111147Snectar Free => OPENSSL_free 3435109998Smarkm [Richard Levitte] 3436109998Smarkm 3437109998Smarkm *) New function BN_mod_exp_mont_word for small bases (roughly 15% 3438109998Smarkm faster than BN_mod_exp_mont, i.e. 7% for a full DH exchange). 3439109998Smarkm [Bodo Moeller] 3440109998Smarkm 3441109998Smarkm *) CygWin32 support. 3442109998Smarkm [John Jarvie <jjarvie@newsguy.com>] 3443109998Smarkm 3444109998Smarkm *) The type-safe stack code has been rejigged. It is now only compiled 3445109998Smarkm in when OpenSSL is configured with the DEBUG_SAFESTACK option and 3446109998Smarkm by default all type-specific stack functions are "#define"d back to 3447109998Smarkm standard stack functions. This results in more streamlined output 3448109998Smarkm but retains the type-safety checking possibilities of the original 3449109998Smarkm approach. 3450109998Smarkm [Geoff Thorpe] 3451109998Smarkm 3452109998Smarkm *) The STACK code has been cleaned up, and certain type declarations 3453109998Smarkm that didn't make a lot of sense have been brought in line. This has 3454109998Smarkm also involved a cleanup of sorts in safestack.h to more correctly 3455109998Smarkm map type-safe stack functions onto their plain stack counterparts. 3456109998Smarkm This work has also resulted in a variety of "const"ifications of 3457109998Smarkm lots of the code, especially "_cmp" operations which should normally 3458109998Smarkm be prototyped with "const" parameters anyway. 3459109998Smarkm [Geoff Thorpe] 3460109998Smarkm 3461109998Smarkm *) When generating bytes for the first time in md_rand.c, 'stir the pool' 3462109998Smarkm by seeding with STATE_SIZE dummy bytes (with zero entropy count). 3463109998Smarkm (The PRNG state consists of two parts, the large pool 'state' and 'md', 3464109998Smarkm where all of 'md' is used each time the PRNG is used, but 'state' 3465109998Smarkm is used only indexed by a cyclic counter. As entropy may not be 3466109998Smarkm well distributed from the beginning, 'md' is important as a 3467109998Smarkm chaining variable. However, the output function chains only half 3468109998Smarkm of 'md', i.e. 80 bits. ssleay_rand_add, on the other hand, chains 3469109998Smarkm all of 'md', and seeding with STATE_SIZE dummy bytes will result 3470109998Smarkm in all of 'state' being rewritten, with the new values depending 3471109998Smarkm on virtually all of 'md'. This overcomes the 80 bit limitation.) 3472109998Smarkm [Bodo Moeller] 3473109998Smarkm 3474109998Smarkm *) In ssl/s2_clnt.c and ssl/s3_clnt.c, call ERR_clear_error() when 3475109998Smarkm the handshake is continued after ssl_verify_cert_chain(); 3476109998Smarkm otherwise, if SSL_VERIFY_NONE is set, remaining error codes 3477109998Smarkm can lead to 'unexplainable' connection aborts later. 3478109998Smarkm [Bodo Moeller; problem tracked down by Lutz Jaenicke] 3479109998Smarkm 3480109998Smarkm *) Major EVP API cipher revision. 3481109998Smarkm Add hooks for extra EVP features. This allows various cipher 3482109998Smarkm parameters to be set in the EVP interface. Support added for variable 3483109998Smarkm key length ciphers via the EVP_CIPHER_CTX_set_key_length() function and 3484109998Smarkm setting of RC2 and RC5 parameters. 3485109998Smarkm 3486109998Smarkm Modify EVP_OpenInit() and EVP_SealInit() to cope with variable key length 3487109998Smarkm ciphers. 3488109998Smarkm 3489109998Smarkm Remove lots of duplicated code from the EVP library. For example *every* 3490109998Smarkm cipher init() function handles the 'iv' in the same way according to the 3491109998Smarkm cipher mode. They also all do nothing if the 'key' parameter is NULL and 3492109998Smarkm for CFB and OFB modes they zero ctx->num. 3493109998Smarkm 3494109998Smarkm New functionality allows removal of S/MIME code RC2 hack. 3495109998Smarkm 3496109998Smarkm Most of the routines have the same form and so can be declared in terms 3497109998Smarkm of macros. 3498109998Smarkm 3499109998Smarkm By shifting this to the top level EVP_CipherInit() it can be removed from 3500109998Smarkm all individual ciphers. If the cipher wants to handle IVs or keys 3501109998Smarkm differently it can set the EVP_CIPH_CUSTOM_IV or EVP_CIPH_ALWAYS_CALL_INIT 3502109998Smarkm flags. 3503109998Smarkm 3504101618Snectar Change lots of functions like EVP_EncryptUpdate() to now return a 3505101618Snectar value: although software versions of the algorithms cannot fail 3506101618Snectar any installed hardware versions can. 3507101618Snectar [Steve Henson] 3508101618Snectar 3509101618Snectar *) Implement SSL_OP_TLS_ROLLBACK_BUG: In ssl3_get_client_key_exchange, if 3510101613Snectar this option is set, tolerate broken clients that send the negotiated 3511101613Snectar protocol version number instead of the requested protocol version 3512101613Snectar number. 3513101613Snectar [Bodo Moeller] 3514101613Snectar 3515101613Snectar *) Call dh_tmp_cb (set by ..._TMP_DH_CB) with correct 'is_export' flag; 3516101613Snectar i.e. non-zero for export ciphersuites, zero otherwise. 3517101613Snectar Previous versions had this flag inverted, inconsistent with 3518101613Snectar rsa_tmp_cb (..._TMP_RSA_CB). 3519101613Snectar [Bodo Moeller; problem reported by Amit Chopra] 3520101613Snectar 3521101613Snectar *) Add missing DSA library text string. Work around for some IIS 3522101613Snectar key files with invalid SEQUENCE encoding. 3523100936Snectar [Steve Henson] 3524100936Snectar 3525109998Smarkm *) Add a document (doc/standards.txt) that list all kinds of standards 3526109998Smarkm and so on that are implemented in OpenSSL. 3527109998Smarkm [Richard Levitte] 3528109998Smarkm 3529109998Smarkm *) Enhance c_rehash script. Old version would mishandle certificates 3530109998Smarkm with the same subject name hash and wouldn't handle CRLs at all. 3531100936Snectar Added -fingerprint option to crl utility, to support new c_rehash 3532100936Snectar features. 3533100936Snectar [Steve Henson] 3534100936Snectar 3535100936Snectar *) Eliminate non-ANSI declarations in crypto.h and stack.h. 3536100936Snectar [Ulf M�ller] 3537100936Snectar 3538100936Snectar *) Fix for SSL server purpose checking. Server checking was 3539100936Snectar rejecting certificates which had extended key usage present 3540100936Snectar but no ssl client purpose. 3541100936Snectar [Steve Henson, reported by Rene Grosser <grosser@hisolutions.com>] 3542100936Snectar 3543100936Snectar *) Make PKCS#12 code work with no password. The PKCS#12 spec 3544100936Snectar is a little unclear about how a blank password is handled. 3545100936Snectar Since the password in encoded as a BMPString with terminating 3546100936Snectar double NULL a zero length password would end up as just the 3547100936Snectar double NULL. However no password at all is different and is 3548100936Snectar handled differently in the PKCS#12 key generation code. NS 3549100936Snectar treats a blank password as zero length. MSIE treats it as no 3550100936Snectar password on export: but it will try both on import. We now do 3551100936Snectar the same: PKCS12_parse() tries zero length and no password if 3552100936Snectar the password is set to "" or NULL (NULL is now a valid password: 3553100936Snectar it wasn't before) as does the pkcs12 application. 3554100936Snectar [Steve Henson] 3555100936Snectar 3556100936Snectar *) Bugfixes in apps/x509.c: Avoid a memory leak; and don't use 3557100936Snectar perror when PEM_read_bio_X509_REQ fails, the error message must 3558100936Snectar be obtained from the error queue. 3559100936Snectar [Bodo Moeller] 3560100936Snectar 3561100936Snectar *) Avoid 'thread_hash' memory leak in crypto/err/err.c by freeing 3562160814Ssimon it in ERR_remove_state if appropriate, and change ERR_get_state 3563100936Snectar accordingly to avoid race conditions (this is necessary because 3564100936Snectar thread_hash is no longer constant once set). 3565100936Snectar [Bodo Moeller] 3566100936Snectar 3567100936Snectar *) Bugfix for linux-elf makefile.one. 3568100936Snectar [Ulf M�ller] 3569100936Snectar 3570160814Ssimon *) RSA_get_default_method() will now cause a default 3571100936Snectar RSA_METHOD to be chosen if one doesn't exist already. 3572100936Snectar Previously this was only set during a call to RSA_new() 3573100936Snectar or RSA_new_method(NULL) meaning it was possible for 3574160814Ssimon RSA_get_default_method() to return NULL. 3575100936Snectar [Geoff Thorpe] 3576100936Snectar 3577100936Snectar *) Added native name translation to the existing DSO code 3578160814Ssimon that will convert (if the flag to do so is set) filenames 3579100936Snectar that are sufficiently small and have no path information 3580100936Snectar into a canonical native form. Eg. "blah" converted to 3581100928Snectar "libblah.so" or "blah.dll" etc. 3582100928Snectar [Geoff Thorpe] 3583100928Snectar 3584100928Snectar *) New function ERR_error_string_n(e, buf, len) which is like 3585100928Snectar ERR_error_string(e, buf), but writes at most 'len' bytes 3586100928Snectar including the 0 terminator. For ERR_error_string_n, 'buf' 3587100928Snectar may not be NULL. 3588100928Snectar [Damien Miller <djm@mindrot.org>, Bodo Moeller] 3589100928Snectar 3590100928Snectar *) CONF library reworked to become more general. A new CONF 3591100928Snectar configuration file reader "class" is implemented as well as a 3592100928Snectar new functions (NCONF_*, for "New CONF") to handle it. The now 3593100928Snectar old CONF_* functions are still there, but are reimplemented to 3594100928Snectar work in terms of the new functions. Also, a set of functions 3595100928Snectar to handle the internal storage of the configuration data is 3596100928Snectar provided to make it easier to write new configuration file 3597100928Snectar reader "classes" (I can definitely see something reading a 3598100928Snectar configuration file in XML format, for example), called _CONF_*, 3599100928Snectar or "the configuration storage API"... 3600100928Snectar 3601100928Snectar The new configuration file reading functions are: 3602100928Snectar 3603100928Snectar NCONF_new, NCONF_free, NCONF_load, NCONF_load_fp, NCONF_load_bio, 3604100928Snectar NCONF_get_section, NCONF_get_string, NCONF_get_numbre 3605100928Snectar 3606100928Snectar NCONF_default, NCONF_WIN32 3607100928Snectar 3608100928Snectar NCONF_dump_fp, NCONF_dump_bio 3609100928Snectar 3610100928Snectar NCONF_default and NCONF_WIN32 are method (or "class") choosers, 3611100928Snectar NCONF_new creates a new CONF object. This works in the same way 3612100928Snectar as other interfaces in OpenSSL, like the BIO interface. 3613100928Snectar NCONF_dump_* dump the internal storage of the configuration file, 3614100928Snectar which is useful for debugging. All other functions take the same 3615100928Snectar arguments as the old CONF_* functions wth the exception of the 3616100928Snectar first that must be a `CONF *' instead of a `LHASH *'. 3617100928Snectar 3618100928Snectar To make it easer to use the new classes with the old CONF_* functions, 3619100928Snectar the function CONF_set_default_method is provided. 3620100928Snectar [Richard Levitte] 3621100928Snectar 3622100928Snectar *) Add '-tls1' option to 'openssl ciphers', which was already 3623100928Snectar mentioned in the documentation but had not been implemented. 3624100928Snectar (This option is not yet really useful because even the additional 3625100928Snectar experimental TLS 1.0 ciphers are currently treated as SSL 3.0 ciphers.) 3626100928Snectar [Bodo Moeller] 3627100928Snectar 3628100928Snectar *) Initial DSO code added into libcrypto for letting OpenSSL (and 3629100928Snectar OpenSSL-based applications) load shared libraries and bind to 3630100928Snectar them in a portable way. 3631100928Snectar [Geoff Thorpe, with contributions from Richard Levitte] 3632100928Snectar 3633100928Snectar Changes between 0.9.5 and 0.9.5a [1 Apr 2000] 3634100928Snectar 3635100928Snectar *) Make sure _lrotl and _lrotr are only used with MSVC. 3636100928Snectar 3637100928Snectar *) Use lock CRYPTO_LOCK_RAND correctly in ssleay_rand_status 3638100928Snectar (the default implementation of RAND_status). 3639100928Snectar 3640100928Snectar *) Rename openssl x509 option '-crlext', which was added in 0.9.5, 3641100928Snectar to '-clrext' (= clear extensions), as intended and documented. 3642100928Snectar [Bodo Moeller; inconsistency pointed out by Michael Attili 3643100928Snectar <attili@amaxo.com>] 3644100928Snectar 3645100928Snectar *) Fix for HMAC. It wasn't zeroing the rest of the block if the key length 3646100928Snectar was larger than the MD block size. 3647100928Snectar [Steve Henson, pointed out by Yost William <YostW@tce.com>] 3648100928Snectar 3649100928Snectar *) Modernise PKCS12_parse() so it uses STACK_OF(X509) for its ca argument 3650100928Snectar fix a leak when the ca argument was passed as NULL. Stop X509_PUBKEY_set() 3651100928Snectar using the passed key: if the passed key was a private key the result 3652100928Snectar of X509_print(), for example, would be to print out all the private key 3653100928Snectar components. 3654100928Snectar [Steve Henson] 3655100928Snectar 3656100928Snectar *) des_quad_cksum() byte order bug fix. 3657100928Snectar [Ulf M�ller, using the problem description in krb4-0.9.7, where 3658100928Snectar the solution is attributed to Derrick J Brashear <shadow@DEMENTIA.ORG>] 3659100928Snectar 3660100928Snectar *) Fix so V_ASN1_APP_CHOOSE works again: however its use is strongly 3661100928Snectar discouraged. 3662100928Snectar [Steve Henson, pointed out by Brian Korver <briank@cs.stanford.edu>] 3663100928Snectar 3664100928Snectar *) For easily testing in shell scripts whether some command 3665100928Snectar 'openssl XXX' exists, the new pseudo-command 'openssl no-XXX' 3666100928Snectar returns with exit code 0 iff no command of the given name is available. 3667109998Smarkm 'no-XXX' is printed in this case, 'XXX' otherwise. In both cases, 3668109998Smarkm the output goes to stdout and nothing is printed to stderr. 3669100928Snectar Additional arguments are always ignored. 3670100928Snectar 3671100928Snectar Since for each cipher there is a command of the same name, 3672100928Snectar the 'no-cipher' compilation switches can be tested this way. 3673100928Snectar 3674100928Snectar ('openssl no-XXX' is not able to detect pseudo-commands such 3675100928Snectar as 'quit', 'list-XXX-commands', or 'no-XXX' itself.) 3676100928Snectar [Bodo Moeller] 3677100928Snectar 3678100928Snectar *) Update test suite so that 'make test' succeeds in 'no-rsa' configuration. 3679100928Snectar [Bodo Moeller] 3680100928Snectar 3681100928Snectar *) For SSL_[CTX_]set_tmp_dh, don't create a DH key if SSL_OP_SINGLE_DH_USE 3682100928Snectar is set; it will be thrown away anyway because each handshake creates 3683100928Snectar its own key. 3684100928Snectar ssl_cert_dup, which is used by SSL_new, now copies DH keys in addition 3685100928Snectar to parameters -- in previous versions (since OpenSSL 0.9.3) the 3686100928Snectar 'default key' from SSL_CTX_set_tmp_dh would always be lost, meanining 3687100928Snectar you effectivly got SSL_OP_SINGLE_DH_USE when using this macro. 3688100928Snectar [Bodo Moeller] 3689100928Snectar 3690100928Snectar *) New s_client option -ign_eof: EOF at stdin is ignored, and 3691100928Snectar 'Q' and 'R' lose their special meanings (quit/renegotiate). 3692100928Snectar This is part of what -quiet does; unlike -quiet, -ign_eof 3693100928Snectar does not suppress any output. 3694100928Snectar [Richard Levitte] 3695100928Snectar 3696100928Snectar *) Add compatibility options to the purpose and trust code. The 3697100928Snectar purpose X509_PURPOSE_ANY is "any purpose" which automatically 3698100928Snectar accepts a certificate or CA, this was the previous behaviour, 3699100928Snectar with all the associated security issues. 3700100928Snectar 370189837Skris X509_TRUST_COMPAT is the old trust behaviour: only and 370289837Skris automatically trust self signed roots in certificate store. A 370389837Skris new trust setting X509_TRUST_DEFAULT is used to specify that 370489837Skris a purpose has no associated trust setting and it should instead 370589837Skris use the value in the default purpose. 370689837Skris [Steve Henson] 370789837Skris 370889837Skris *) Fix the PKCS#8 DSA private key code so it decodes keys again 370989837Skris and fix a memory leak. 371089837Skris [Steve Henson] 371189837Skris 371289837Skris *) In util/mkerr.pl (which implements 'make errors'), preserve 371389837Skris reason strings from the previous version of the .c file, as 371489837Skris the default to have only downcase letters (and digits) in 371589837Skris automatically generated reasons codes is not always appropriate. 371689837Skris [Bodo Moeller] 371789837Skris 371889837Skris *) In ERR_load_ERR_strings(), build an ERR_LIB_SYS error reason table 371989837Skris using strerror. Previously, ERR_reason_error_string() returned 372089837Skris library names as reason strings for SYSerr; but SYSerr is a special 372189837Skris case where small numbers are errno values, not library numbers. 372289837Skris [Bodo Moeller] 372389837Skris 372489837Skris *) Add '-dsaparam' option to 'openssl dhparam' application. This 372589837Skris converts DSA parameters into DH parameters. (When creating parameters, 372689837Skris DSA_generate_parameters is used.) 372789837Skris [Bodo Moeller] 372889837Skris 372989837Skris *) Include 'length' (recommended exponent length) in C code generated 373089837Skris by 'openssl dhparam -C'. 373189837Skris [Bodo Moeller] 373289837Skris 373389837Skris *) The second argument to set_label in perlasm was already being used 373489837Skris so couldn't be used as a "file scope" flag. Moved to third argument 373589837Skris which was free. 373689837Skris [Steve Henson] 373789837Skris 373889837Skris *) In PEM_ASN1_write_bio and some other functions, use RAND_pseudo_bytes 373989837Skris instead of RAND_bytes for encryption IVs and salts. 374089837Skris [Bodo Moeller] 374189837Skris 374289837Skris *) Include RAND_status() into RAND_METHOD instead of implementing 374389837Skris it only for md_rand.c Otherwise replacing the PRNG by calling 374489837Skris RAND_set_rand_method would be impossible. 374589837Skris [Bodo Moeller] 374689837Skris 374789837Skris *) Don't let DSA_generate_key() enter an infinite loop if the random 374889837Skris number generation fails. 374989837Skris [Bodo Moeller] 375089837Skris 375189837Skris *) New 'rand' application for creating pseudo-random output. 375289837Skris [Bodo Moeller] 375389837Skris 375489837Skris *) Added configuration support for Linux/IA64 375589837Skris [Rolf Haberrecker <rolf@suse.de>] 375689837Skris 375789837Skris *) Assembler module support for Mingw32. 375889837Skris [Ulf M�ller] 375989837Skris 376089837Skris *) Shared library support for HPUX (in shlib/). 376189837Skris [Lutz Jaenicke <Lutz.Jaenicke@aet.TU-Cottbus.DE> and Anonymous] 376289837Skris 376389837Skris *) Shared library support for Solaris gcc. 376489837Skris [Lutz Behnke <behnke@trustcenter.de>] 376589837Skris 376689837Skris Changes between 0.9.4 and 0.9.5 [28 Feb 2000] 376789837Skris 376889837Skris *) PKCS7_encrypt() was adding text MIME headers twice because they 376989837Skris were added manually and by SMIME_crlf_copy(). 377089837Skris [Steve Henson] 377189837Skris 377289837Skris *) In bntest.c don't call BN_rand with zero bits argument. 377389837Skris [Steve Henson, pointed out by Andrew W. Gray <agray@iconsinc.com>] 377489837Skris 377589837Skris *) BN_mul bugfix: In bn_mul_part_recursion() only the a>a[n] && b>b[n] 377689837Skris case was implemented. This caused BN_div_recp() to fail occasionally. 377789837Skris [Ulf M�ller] 377889837Skris 377989837Skris *) Add an optional second argument to the set_label() in the perl 378089837Skris assembly language builder. If this argument exists and is set 378189837Skris to 1 it signals that the assembler should use a symbol whose 378289837Skris scope is the entire file, not just the current function. This 378389837Skris is needed with MASM which uses the format label:: for this scope. 378489837Skris [Steve Henson, pointed out by Peter Runestig <peter@runestig.com>] 378589837Skris 378689837Skris *) Change the ASN1 types so they are typedefs by default. Before 378789837Skris almost all types were #define'd to ASN1_STRING which was causing 378889837Skris STACK_OF() problems: you couldn't declare STACK_OF(ASN1_UTF8STRING) 378989837Skris for example. 379089837Skris [Steve Henson] 379189837Skris 379289837Skris *) Change names of new functions to the new get1/get0 naming 379389837Skris convention: After 'get1', the caller owns a reference count 379489837Skris and has to call ..._free; 'get0' returns a pointer to some 379589837Skris data structure without incrementing reference counters. 379689837Skris (Some of the existing 'get' functions increment a reference 379789837Skris counter, some don't.) 379889837Skris Similarly, 'set1' and 'add1' functions increase reference 379989837Skris counters or duplicate objects. 380089837Skris [Steve Henson] 380189837Skris 380289837Skris *) Allow for the possibility of temp RSA key generation failure: 380389837Skris the code used to assume it always worked and crashed on failure. 380489837Skris [Steve Henson] 380589837Skris 380689837Skris *) Fix potential buffer overrun problem in BIO_printf(). 380789837Skris [Ulf M�ller, using public domain code by Patrick Powell; problem 380889837Skris pointed out by David Sacerdote <das33@cornell.edu>] 380989837Skris 381089837Skris *) Support EGD <http://www.lothar.com/tech/crypto/>. New functions 381189837Skris RAND_egd() and RAND_status(). In the command line application, 381289837Skris the EGD socket can be specified like a seed file using RANDFILE 381389837Skris or -rand. 381489837Skris [Ulf M�ller] 381589837Skris 381689837Skris *) Allow the string CERTIFICATE to be tolerated in PKCS#7 structures. 381789837Skris Some CAs (e.g. Verisign) distribute certificates in this form. 381889837Skris [Steve Henson] 381989837Skris 382089837Skris *) Remove the SSL_ALLOW_ADH compile option and set the default cipher 382189837Skris list to exclude them. This means that no special compilation option 382289837Skris is needed to use anonymous DH: it just needs to be included in the 382389837Skris cipher list. 382489837Skris [Steve Henson] 382589837Skris 382689837Skris *) Change the EVP_MD_CTX_type macro so its meaning consistent with 382789837Skris EVP_MD_type. The old functionality is available in a new macro called 382889837Skris EVP_MD_md(). Change code that uses it and update docs. 382989837Skris [Steve Henson] 383089837Skris 383189837Skris *) ..._ctrl functions now have corresponding ..._callback_ctrl functions 383289837Skris where the 'void *' argument is replaced by a function pointer argument. 383389837Skris Previously 'void *' was abused to point to functions, which works on 383489837Skris many platforms, but is not correct. As these functions are usually 383589837Skris called by macros defined in OpenSSL header files, most source code 383689837Skris should work without changes. 383789837Skris [Richard Levitte] 383889837Skris 383989837Skris *) <openssl/opensslconf.h> (which is created by Configure) now contains 384089837Skris sections with information on -D... compiler switches used for 384189837Skris compiling the library so that applications can see them. To enable 384289837Skris one of these sections, a pre-processor symbol OPENSSL_..._DEFINES 384389837Skris must be defined. E.g., 384489837Skris #define OPENSSL_ALGORITHM_DEFINES 384589837Skris #include <openssl/opensslconf.h> 384689837Skris defines all pertinent NO_<algo> symbols, such as NO_IDEA, NO_RSA, etc. 384789837Skris [Richard Levitte, Ulf and Bodo M�ller] 384889837Skris 384989837Skris *) Bugfix: Tolerate fragmentation and interleaving in the SSL 3/TLS 385089837Skris record layer. 385189837Skris [Bodo Moeller] 385289837Skris 385389837Skris *) Change the 'other' type in certificate aux info to a STACK_OF 385489837Skris X509_ALGOR. Although not an AlgorithmIdentifier as such it has 385589837Skris the required ASN1 format: arbitrary types determined by an OID. 385689837Skris [Steve Henson] 385789837Skris 385889837Skris *) Add some PEM_write_X509_REQ_NEW() functions and a command line 385989837Skris argument to 'req'. This is not because the function is newer or 386089837Skris better than others it just uses the work 'NEW' in the certificate 386189837Skris request header lines. Some software needs this. 386289837Skris [Steve Henson] 386389837Skris 386489837Skris *) Reorganise password command line arguments: now passwords can be 386589837Skris obtained from various sources. Delete the PEM_cb function and make 386689837Skris it the default behaviour: i.e. if the callback is NULL and the 386789837Skris usrdata argument is not NULL interpret it as a null terminated pass 386889837Skris phrase. If usrdata and the callback are NULL then the pass phrase 386989837Skris is prompted for as usual. 387089837Skris [Steve Henson] 387189837Skris 387289837Skris *) Add support for the Compaq Atalla crypto accelerator. If it is installed, 387389837Skris the support is automatically enabled. The resulting binaries will 387489837Skris autodetect the card and use it if present. 387589837Skris [Ben Laurie and Compaq Inc.] 387689837Skris 387789837Skris *) Work around for Netscape hang bug. This sends certificate request 387889837Skris and server done in one record. Since this is perfectly legal in the 387989837Skris SSL/TLS protocol it isn't a "bug" option and is on by default. See 388089837Skris the bugs/SSLv3 entry for more info. 388189837Skris [Steve Henson] 388289837Skris 388389837Skris *) HP-UX tune-up: new unified configs, HP C compiler bug workaround. 388489837Skris [Andy Polyakov] 388589837Skris 388689837Skris *) Add -rand argument to smime and pkcs12 applications and read/write 388789837Skris of seed file. 388889837Skris [Steve Henson] 388989837Skris 389089837Skris *) New 'passwd' tool for crypt(3) and apr1 password hashes. 389189837Skris [Bodo Moeller] 389289837Skris 389389837Skris *) Add command line password options to the remaining applications. 389489837Skris [Steve Henson] 389589837Skris 389689837Skris *) Bug fix for BN_div_recp() for numerators with an even number of 389789837Skris bits. 389889837Skris [Ulf M�ller] 389989837Skris 390089837Skris *) More tests in bntest.c, and changed test_bn output. 390189837Skris [Ulf M�ller] 390289837Skris 390389837Skris *) ./config recognizes MacOS X now. 390489837Skris [Andy Polyakov] 390589837Skris 390689837Skris *) Bug fix for BN_div() when the first words of num and divsor are 390789837Skris equal (it gave wrong results if (rem=(n1-q*d0)&BN_MASK2) < d0). 390889837Skris [Ulf M�ller] 390989837Skris 391089837Skris *) Add support for various broken PKCS#8 formats, and command line 391189837Skris options to produce them. 391289837Skris [Steve Henson] 391389837Skris 391489837Skris *) New functions BN_CTX_start(), BN_CTX_get() and BT_CTX_end() to 391589837Skris get temporary BIGNUMs from a BN_CTX. 391689837Skris [Ulf M�ller] 391789837Skris 391889837Skris *) Correct return values in BN_mod_exp_mont() and BN_mod_exp2_mont() 391989837Skris for p == 0. 392089837Skris [Ulf M�ller] 392189837Skris 392289837Skris *) Change the SSLeay_add_all_*() functions to OpenSSL_add_all_*() and 392389837Skris include a #define from the old name to the new. The original intent 392489837Skris was that statically linked binaries could for example just call 392589837Skris SSLeay_add_all_ciphers() to just add ciphers to the table and not 392689837Skris link with digests. This never worked becayse SSLeay_add_all_digests() 392789837Skris and SSLeay_add_all_ciphers() were in the same source file so calling 392889837Skris one would link with the other. They are now in separate source files. 392989837Skris [Steve Henson] 393089837Skris 393189837Skris *) Add a new -notext option to 'ca' and a -pubkey option to 'spkac'. 393289837Skris [Steve Henson] 393389837Skris 393489837Skris *) Use a less unusual form of the Miller-Rabin primality test (it used 393589837Skris a binary algorithm for exponentiation integrated into the Miller-Rabin 393689837Skris loop, our standard modexp algorithms are faster). 393789837Skris [Bodo Moeller] 393889837Skris 393989837Skris *) Support for the EBCDIC character set completed. 394089837Skris [Martin Kraemer <Martin.Kraemer@Mch.SNI.De>] 394179998Skris 394279998Skris *) Source code cleanups: use const where appropriate, eliminate casts, 394379998Skris use void * instead of char * in lhash. 394479998Skris [Ulf M�ller] 394579998Skris 394679998Skris *) Bugfix: ssl3_send_server_key_exchange was not restartable 394779998Skris (the state was not changed to SSL3_ST_SW_KEY_EXCH_B, and because of 394879998Skris this the server could overwrite ephemeral keys that the client 394979998Skris has already seen). 395079998Skris [Bodo Moeller] 395179998Skris 395279998Skris *) Turn DSA_is_prime into a macro that calls BN_is_prime, 395379998Skris using 50 iterations of the Rabin-Miller test. 395479998Skris 395579998Skris DSA_generate_parameters now uses BN_is_prime_fasttest (with 50 395679998Skris iterations of the Rabin-Miller test as required by the appendix 395779998Skris to FIPS PUB 186[-1]) instead of DSA_is_prime. 395879998Skris As BN_is_prime_fasttest includes trial division, DSA parameter 395979998Skris generation becomes much faster. 396079998Skris 396179998Skris This implies a change for the callback functions in DSA_is_prime 396279998Skris and DSA_generate_parameters: The callback function is called once 396379998Skris for each positive witness in the Rabin-Miller test, not just 396479998Skris occasionally in the inner loop; and the parameters to the 396579998Skris callback function now provide an iteration count for the outer 396679998Skris loop rather than for the current invocation of the inner loop. 396779998Skris DSA_generate_parameters additionally can call the callback 396879998Skris function with an 'iteration count' of -1, meaning that a 396979998Skris candidate has passed the trial division test (when q is generated 397079998Skris from an application-provided seed, trial division is skipped). 397179998Skris [Bodo Moeller] 397279998Skris 397379998Skris *) New function BN_is_prime_fasttest that optionally does trial 397479998Skris division before starting the Rabin-Miller test and has 397579998Skris an additional BN_CTX * argument (whereas BN_is_prime always 397679998Skris has to allocate at least one BN_CTX). 397779998Skris 'callback(1, -1, cb_arg)' is called when a number has passed the 397879998Skris trial division stage. 397979998Skris [Bodo Moeller] 398079998Skris 398179998Skris *) Fix for bug in CRL encoding. The validity dates weren't being handled 398279998Skris as ASN1_TIME. 398379998Skris [Steve Henson] 398479998Skris 398579998Skris *) New -pkcs12 option to CA.pl script to write out a PKCS#12 file. 398679998Skris [Steve Henson] 398779998Skris 398879998Skris *) New function BN_pseudo_rand(). 398979998Skris [Ulf M�ller] 399079998Skris 399179998Skris *) Clean up BN_mod_mul_montgomery(): replace the broken (and unreadable) 399279998Skris bignum version of BN_from_montgomery() with the working code from 399379998Skris SSLeay 0.9.0 (the word based version is faster anyway), and clean up 399479998Skris the comments. 399589837Skris [Ulf M�ller] 399679998Skris 399779998Skris *) Avoid a race condition in s2_clnt.c (function get_server_hello) that 399879998Skris made it impossible to use the same SSL_SESSION data structure in 399979998Skris SSL2 clients in multiple threads. 400079998Skris [Bodo Moeller] 400179998Skris 400279998Skris *) The return value of RAND_load_file() no longer counts bytes obtained 400379998Skris by stat(). RAND_load_file(..., -1) is new and uses the complete file 400479998Skris to seed the PRNG (previously an explicit byte count was required). 400579998Skris [Ulf M�ller, Bodo M�ller] 400679998Skris 400779998Skris *) Clean up CRYPTO_EX_DATA functions, some of these didn't have prototypes 400879998Skris used (char *) instead of (void *) and had casts all over the place. 400979998Skris [Steve Henson] 401079998Skris 401179998Skris *) Make BN_generate_prime() return NULL on error if ret!=NULL. 401279998Skris [Ulf M�ller] 401379998Skris 401479998Skris *) Retain source code compatibility for BN_prime_checks macro: 401579998Skris BN_is_prime(..., BN_prime_checks, ...) now uses 401679998Skris BN_prime_checks_for_size to determine the appropriate number of 401779998Skris Rabin-Miller iterations. 401879998Skris [Ulf M�ller] 401979998Skris 402079998Skris *) Diffie-Hellman uses "safe" primes: DH_check() return code renamed to 402179998Skris DH_CHECK_P_NOT_SAFE_PRIME. 402279998Skris (Check if this is true? OpenPGP calls them "strong".) 402379998Skris [Ulf M�ller] 402479998Skris 402579998Skris *) Merge the functionality of "dh" and "gendh" programs into a new program 402679998Skris "dhparam". The old programs are retained for now but will handle DH keys 402779998Skris (instead of parameters) in future. 402879998Skris [Steve Henson] 402979998Skris 403079998Skris *) Make the ciphers, s_server and s_client programs check the return values 403179998Skris when a new cipher list is set. 403279998Skris [Steve Henson] 403379998Skris 403479998Skris *) Enhance the SSL/TLS cipher mechanism to correctly handle the TLS 56bit 403579998Skris ciphers. Before when the 56bit ciphers were enabled the sorting was 403679998Skris wrong. 403779998Skris 403879998Skris The syntax for the cipher sorting has been extended to support sorting by 403979998Skris cipher-strength (using the strength_bits hard coded in the tables). 404079998Skris The new command is "@STRENGTH" (see also doc/apps/ciphers.pod). 404179998Skris 404279998Skris Fix a bug in the cipher-command parser: when supplying a cipher command 404379998Skris string with an "undefined" symbol (neither command nor alphanumeric 404479998Skris [A-Za-z0-9], ssl_set_cipher_list used to hang in an endless loop. Now 404589837Skris an error is flagged. 404679998Skris 404779998Skris Due to the strength-sorting extension, the code of the 404879998Skris ssl_create_cipher_list() function was completely rearranged. I hope that 404979998Skris the readability was also increased :-) 405079998Skris [Lutz Jaenicke <Lutz.Jaenicke@aet.TU-Cottbus.DE>] 405179998Skris 405279998Skris *) Minor change to 'x509' utility. The -CAcreateserial option now uses 1 405376866Skris for the first serial number and places 2 in the serial number file. This 405472613Skris avoids problems when the root CA is created with serial number zero and 405576866Skris the first user certificate has the same issuer name and serial number 405676866Skris as the root CA. 405776866Skris [Steve Henson] 405876866Skris 405976866Skris *) Fixes to X509_ATTRIBUTE utilities, change the 'req' program so it uses 406076866Skris the new code. Add documentation for this stuff. 406176866Skris [Steve Henson] 406276866Skris 406376866Skris *) Changes to X509_ATTRIBUTE utilities. These have been renamed from 406476866Skris X509_*() to X509at_*() on the grounds that they don't handle X509 406576866Skris structures and behave in an analagous way to the X509v3 functions: 406676866Skris they shouldn't be called directly but wrapper functions should be used 406776866Skris instead. 406876866Skris 406976866Skris So we also now have some wrapper functions that call the X509at functions 407076866Skris when passed certificate requests. (TO DO: similar things can be done with 407176866Skris PKCS#7 signed and unsigned attributes, PKCS#12 attributes and a few other 407276866Skris things. Some of these need some d2i or i2d and print functionality 407376866Skris because they handle more complex structures.) 407476866Skris [Steve Henson] 407576866Skris 407676866Skris *) Add missing #ifndefs that caused missing symbols when building libssl 407776866Skris as a shared library without RSA. Use #ifndef NO_SSL2 instead of 407876866Skris NO_RSA in ssl/s2*.c. 407976866Skris [Kris Kennaway <kris@hub.freebsd.org>, modified by Ulf M�ller] 408076866Skris 408176866Skris *) Precautions against using the PRNG uninitialized: RAND_bytes() now 408276866Skris has a return value which indicates the quality of the random data 408376866Skris (1 = ok, 0 = not seeded). Also an error is recorded on the thread's 408476866Skris error queue. New function RAND_pseudo_bytes() generates output that is 408576866Skris guaranteed to be unique but not unpredictable. RAND_add is like 408676866Skris RAND_seed, but takes an extra argument for an entropy estimate 408776866Skris (RAND_seed always assumes full entropy). 408876866Skris [Ulf M�ller] 408976866Skris 409076866Skris *) Do more iterations of Rabin-Miller probable prime test (specifically, 409176866Skris 3 for 1024-bit primes, 6 for 512-bit primes, 12 for 256-bit primes 409276866Skris instead of only 2 for all lengths; see BN_prime_checks_for_size definition 409376866Skris in crypto/bn/bn_prime.c for the complete table). This guarantees a 409476866Skris false-positive rate of at most 2^-80 for random input. 409576866Skris [Bodo Moeller] 409676866Skris 409776866Skris *) Rewrite ssl3_read_n (ssl/s3_pkt.c) avoiding a couple of bugs. 409876866Skris [Bodo Moeller] 409976866Skris 410076866Skris *) New function X509_CTX_rget_chain() (renamed to X509_CTX_get1_chain 410176866Skris in the 0.9.5 release), this returns the chain 410276866Skris from an X509_CTX structure with a dup of the stack and all 410376866Skris the X509 reference counts upped: so the stack will exist 410476866Skris after X509_CTX_cleanup() has been called. Modify pkcs12.c 410576866Skris to use this. 410676866Skris 410776866Skris Also make SSL_SESSION_print() print out the verify return 410876866Skris code. 410976866Skris [Steve Henson] 411076866Skris 411176866Skris *) Add manpage for the pkcs12 command. Also change the default 411276866Skris behaviour so MAC iteration counts are used unless the new 411376866Skris -nomaciter option is used. This improves file security and 411476866Skris only older versions of MSIE (4.0 for example) need it. 411576866Skris [Steve Henson] 411676866Skris 411776866Skris *) Honor the no-xxx Configure options when creating .DEF files. 411876866Skris [Ulf M�ller] 411976866Skris 412076866Skris *) Add PKCS#10 attributes to field table: challengePassword, 412176866Skris unstructuredName and unstructuredAddress. These are taken from 412276866Skris draft PKCS#9 v2.0 but are compatible with v1.2 provided no 412376866Skris international characters are used. 412476866Skris 412576866Skris More changes to X509_ATTRIBUTE code: allow the setting of types 412676866Skris based on strings. Remove the 'loc' parameter when adding 412776866Skris attributes because these will be a SET OF encoding which is sorted 412876866Skris in ASN1 order. 412976866Skris [Steve Henson] 413076866Skris 413176866Skris *) Initial changes to the 'req' utility to allow request generation 413276866Skris automation. This will allow an application to just generate a template 413376866Skris file containing all the field values and have req construct the 413476866Skris request. 413576866Skris 413676866Skris Initial support for X509_ATTRIBUTE handling. Stacks of these are 413776866Skris used all over the place including certificate requests and PKCS#7 413876866Skris structures. They are currently handled manually where necessary with 413976866Skris some primitive wrappers for PKCS#7. The new functions behave in a 414076866Skris manner analogous to the X509 extension functions: they allow 414176866Skris attributes to be looked up by NID and added. 414276866Skris 414376866Skris Later something similar to the X509V3 code would be desirable to 414476866Skris automatically handle the encoding, decoding and printing of the 414576866Skris more complex types. The string types like challengePassword can 414676866Skris be handled by the string table functions. 414776866Skris 414876866Skris Also modified the multi byte string table handling. Now there is 414976866Skris a 'global mask' which masks out certain types. The table itself 415076866Skris can use the flag STABLE_NO_MASK to ignore the mask setting: this 415176866Skris is useful when for example there is only one permissible type 415276866Skris (as in countryName) and using the mask might result in no valid 415376866Skris types at all. 415476866Skris [Steve Henson] 415576866Skris 415676866Skris *) Clean up 'Finished' handling, and add functions SSL_get_finished and 415776866Skris SSL_get_peer_finished to allow applications to obtain the latest 415876866Skris Finished messages sent to the peer or expected from the peer, 415976866Skris respectively. (SSL_get_peer_finished is usually the Finished message 416076866Skris actually received from the peer, otherwise the protocol will be aborted.) 416176866Skris 416276866Skris As the Finished message are message digests of the complete handshake 416376866Skris (with a total of 192 bits for TLS 1.0 and more for SSL 3.0), they can 416476866Skris be used for external authentication procedures when the authentication 416576866Skris provided by SSL/TLS is not desired or is not enough. 416672613Skris [Bodo Moeller] 416772613Skris 416876866Skris *) Enhanced support for Alpha Linux is added. Now ./config checks if 416976866Skris the host supports BWX extension and if Compaq C is present on the 417076866Skris $PATH. Just exploiting of the BWX extension results in 20-30% 417176866Skris performance kick for some algorithms, e.g. DES and RC4 to mention 417272613Skris a couple. Compaq C in turn generates ~20% faster code for MD5 and 417372613Skris SHA1. 417472613Skris [Andy Polyakov] 417572613Skris 417672613Skris *) Add support for MS "fast SGC". This is arguably a violation of the 417772613Skris SSL3/TLS protocol. Netscape SGC does two handshakes: the first with 417872613Skris weak crypto and after checking the certificate is SGC a second one 417972613Skris with strong crypto. MS SGC stops the first handshake after receiving 418072613Skris the server certificate message and sends a second client hello. Since 418172613Skris a server will typically do all the time consuming operations before 418272613Skris expecting any further messages from the client (server key exchange 418372613Skris is the most expensive) there is little difference between the two. 418472613Skris 418572613Skris To get OpenSSL to support MS SGC we have to permit a second client 418672613Skris hello message after we have sent server done. In addition we have to 418772613Skris reset the MAC if we do get this second client hello. 418872613Skris [Steve Henson] 418972613Skris 419072613Skris *) Add a function 'd2i_AutoPrivateKey()' this will automatically decide 419172613Skris if a DER encoded private key is RSA or DSA traditional format. Changed 419272613Skris d2i_PrivateKey_bio() to use it. This is only needed for the "traditional" 419372613Skris format DER encoded private key. Newer code should use PKCS#8 format which 419472613Skris has the key type encoded in the ASN1 structure. Added DER private key 419572613Skris support to pkcs8 application. 419672613Skris [Steve Henson] 419772613Skris 419872613Skris *) SSL 3/TLS 1 servers now don't request certificates when an anonymous 419972613Skris ciphersuites has been selected (as required by the SSL 3/TLS 1 420076866Skris specifications). Exception: When SSL_VERIFY_FAIL_IF_NO_PEER_CERT 420172613Skris is set, we interpret this as a request to violate the specification 420272613Skris (the worst that can happen is a handshake failure, and 'correct' 420372613Skris behaviour would result in a handshake failure anyway). 420472613Skris [Bodo Moeller] 420572613Skris 420672613Skris *) In SSL_CTX_add_session, take into account that there might be multiple 420772613Skris SSL_SESSION structures with the same session ID (e.g. when two threads 420872613Skris concurrently obtain them from an external cache). 420972613Skris The internal cache can handle only one SSL_SESSION with a given ID, 421072613Skris so if there's a conflict, we now throw out the old one to achieve 421172613Skris consistency. 421272613Skris [Bodo Moeller] 421372613Skris 421472613Skris *) Add OIDs for idea and blowfish in CBC mode. This will allow both 421572613Skris to be used in PKCS#5 v2.0 and S/MIME. Also add checking to 421672613Skris some routines that use cipher OIDs: some ciphers do not have OIDs 421772613Skris defined and so they cannot be used for S/MIME and PKCS#5 v2.0 for 421872613Skris example. 421972613Skris [Steve Henson] 422072613Skris 422172613Skris *) Simplify the trust setting structure and code. Now we just have 422272613Skris two sequences of OIDs for trusted and rejected settings. These will 422372613Skris typically have values the same as the extended key usage extension 422472613Skris and any application specific purposes. 422572613Skris 422672613Skris The trust checking code now has a default behaviour: it will just 422772613Skris check for an object with the same NID as the passed id. Functions can 422872613Skris be provided to override either the default behaviour or the behaviour 422972613Skris for a given id. SSL client, server and email already have functions 423072613Skris in place for compatibility: they check the NID and also return "trusted" 423172613Skris if the certificate is self signed. 423272613Skris [Steve Henson] 423372613Skris 423472613Skris *) Add d2i,i2d bio/fp functions for PrivateKey: these convert the 423572613Skris traditional format into an EVP_PKEY structure. 423672613Skris [Steve Henson] 423772613Skris 423872613Skris *) Add a password callback function PEM_cb() which either prompts for 423972613Skris a password if usr_data is NULL or otherwise assumes it is a null 424072613Skris terminated password. Allow passwords to be passed on command line 424172613Skris environment or config files in a few more utilities. 424272613Skris [Steve Henson] 424372613Skris 424472613Skris *) Add a bunch of DER and PEM functions to handle PKCS#8 format private 424572613Skris keys. Add some short names for PKCS#8 PBE algorithms and allow them 424672613Skris to be specified on the command line for the pkcs8 and pkcs12 utilities. 424772613Skris Update documentation. 424872613Skris [Steve Henson] 424972613Skris 425072613Skris *) Support for ASN1 "NULL" type. This could be handled before by using 425172613Skris ASN1_TYPE but there wasn't any function that would try to read a NULL 425272613Skris and produce an error if it couldn't. For compatibility we also have 425376866Skris ASN1_NULL_new() and ASN1_NULL_free() functions but these are faked and 425476866Skris don't allocate anything because they don't need to. 425576866Skris [Steve Henson] 425672613Skris 425772613Skris *) Initial support for MacOS is now provided. Examine INSTALL.MacOS 425872613Skris for details. 425972613Skris [Andy Polyakov, Roy Woods <roy@centicsystems.ca>] 426072613Skris 426172613Skris *) Rebuild of the memory allocation routines used by OpenSSL code and 426276866Skris possibly others as well. The purpose is to make an interface that 426376866Skris provide hooks so anyone can build a separate set of allocation and 426476866Skris deallocation routines to be used by OpenSSL, for example memory 426576866Skris pool implementations, or something else, which was previously hard 426672613Skris since Malloc(), Realloc() and Free() were defined as macros having 426772613Skris the values malloc, realloc and free, respectively (except for Win32 426872613Skris compilations). The same is provided for memory debugging code. 426972613Skris OpenSSL already comes with functionality to find memory leaks, but 427072613Skris this gives people a chance to debug other memory problems. 427172613Skris 427272613Skris With these changes, a new set of functions and macros have appeared: 427372613Skris 427472613Skris CRYPTO_set_mem_debug_functions() [F] 427572613Skris CRYPTO_get_mem_debug_functions() [F] 427672613Skris CRYPTO_dbg_set_options() [F] 427772613Skris CRYPTO_dbg_get_options() [F] 427872613Skris CRYPTO_malloc_debug_init() [M] 427972613Skris 428072613Skris The memory debug functions are NULL by default, unless the library 428172613Skris is compiled with CRYPTO_MDEBUG or friends is defined. If someone 428272613Skris wants to debug memory anyway, CRYPTO_malloc_debug_init() (which 428372613Skris gives the standard debugging functions that come with OpenSSL) or 428472613Skris CRYPTO_set_mem_debug_functions() (tells OpenSSL to use functions 428572613Skris provided by the library user) must be used. When the standard 428672613Skris debugging functions are used, CRYPTO_dbg_set_options can be used to 428772613Skris request additional information: 428876866Skris CRYPTO_dbg_set_options(V_CYRPTO_MDEBUG_xxx) corresponds to setting 428976866Skris the CRYPTO_MDEBUG_xxx macro when compiling the library. 429076866Skris 429176866Skris Also, things like CRYPTO_set_mem_functions will always give the 429276866Skris expected result (the new set of functions is used for allocation 429376866Skris and deallocation) at all times, regardless of platform and compiler 429472613Skris options. 429572613Skris 429672613Skris To finish it up, some functions that were never use in any other 429772613Skris way than through macros have a new API and new semantic: 429876866Skris 429976866Skris CRYPTO_dbg_malloc() 430076866Skris CRYPTO_dbg_realloc() 430176866Skris CRYPTO_dbg_free() 430276866Skris 430376866Skris All macros of value have retained their old syntax. 430472613Skris [Richard Levitte and Bodo Moeller] 430572613Skris 430672613Skris *) Some S/MIME fixes. The OID for SMIMECapabilities was wrong, the 430776866Skris ordering of SMIMECapabilities wasn't in "strength order" and there 430876866Skris was a missing NULL in the AlgorithmIdentifier for the SHA1 signature 430976866Skris algorithm. 431076866Skris [Steve Henson] 431176866Skris 431276866Skris *) Some ASN1 types with illegal zero length encoding (INTEGER, 431376866Skris ENUMERATED and OBJECT IDENTIFIER) choked the ASN1 routines. 431476866Skris [Frans Heymans <fheymans@isaserver.be>, modified by Steve Henson] 431576866Skris 431676866Skris *) Merge in my S/MIME library for OpenSSL. This provides a simple 431776866Skris S/MIME API on top of the PKCS#7 code, a MIME parser (with enough 431876866Skris functionality to handle multipart/signed properly) and a utility 431976866Skris called 'smime' to call all this stuff. This is based on code I 432076866Skris originally wrote for Celo who have kindly allowed it to be 432176866Skris included in OpenSSL. 432272613Skris [Steve Henson] 432372613Skris 432472613Skris *) Add variants des_set_key_checked and des_set_key_unchecked of 432572613Skris des_set_key (aka des_key_sched). Global variable des_check_key 432672613Skris decides which of these is called by des_set_key; this way 432772613Skris des_check_key behaves as it always did, but applications and 432872613Skris the library itself, which was buggy for des_check_key == 1, 432972613Skris have a cleaner way to pick the version they need. 433072613Skris [Bodo Moeller] 433172613Skris 433272613Skris *) New function PKCS12_newpass() which changes the password of a 433372613Skris PKCS12 structure. 433472613Skris [Steve Henson] 433572613Skris 433672613Skris *) Modify X509_TRUST and X509_PURPOSE so it also uses a static and 433772613Skris dynamic mix. In both cases the ids can be used as an index into the 433868651Skris table. Also modified the X509_TRUST_add() and X509_PURPOSE_add() 433968651Skris functions so they accept a list of the field values and the 434068651Skris application doesn't need to directly manipulate the X509_TRUST 434168651Skris structure. 434268651Skris [Steve Henson] 434368651Skris 434468651Skris *) Modify the ASN1_STRING_TABLE stuff so it also uses bsearch and doesn't 434568651Skris need initialising. 434668651Skris [Steve Henson] 434768651Skris 434868651Skris *) Modify the way the V3 extension code looks up extensions. This now 434968651Skris works in a similar way to the object code: we have some "standard" 435068651Skris extensions in a static table which is searched with OBJ_bsearch() 435168651Skris and the application can add dynamic ones if needed. The file 435268651Skris crypto/x509v3/ext_dat.h now has the info: this file needs to be 435368651Skris updated whenever a new extension is added to the core code and kept 435468651Skris in ext_nid order. There is a simple program 'tabtest.c' which checks 435568651Skris this. New extensions are not added too often so this file can readily 435668651Skris be maintained manually. 435768651Skris 435868651Skris There are two big advantages in doing things this way. The extensions 435968651Skris can be looked up immediately and no longer need to be "added" using 436068651Skris X509V3_add_standard_extensions(): this function now does nothing. 436168651Skris [Side note: I get *lots* of email saying the extension code doesn't 436268651Skris work because people forget to call this function] 436368651Skris Also no dynamic allocation is done unless new extensions are added: 436468651Skris so if we don't add custom extensions there is no need to call 436568651Skris X509V3_EXT_cleanup(). 436668651Skris [Steve Henson] 436768651Skris 436868651Skris *) Modify enc utility's salting as follows: make salting the default. Add a 436968651Skris magic header, so unsalted files fail gracefully instead of just decrypting 437068651Skris to garbage. This is because not salting is a big security hole, so people 437168651Skris should be discouraged from doing it. 437268651Skris [Ben Laurie] 437368651Skris 437468651Skris *) Fixes and enhancements to the 'x509' utility. It allowed a message 437568651Skris digest to be passed on the command line but it only used this 437668651Skris parameter when signing a certificate. Modified so all relevant 437768651Skris operations are affected by the digest parameter including the 437868651Skris -fingerprint and -x509toreq options. Also -x509toreq choked if a 437968651Skris DSA key was used because it didn't fix the digest. 438068651Skris [Steve Henson] 438168651Skris 438268651Skris *) Initial certificate chain verify code. Currently tests the untrusted 438368651Skris certificates for consistency with the verify purpose (which is set 438468651Skris when the X509_STORE_CTX structure is set up) and checks the pathlength. 438568651Skris 438668651Skris There is a NO_CHAIN_VERIFY compilation option to keep the old behaviour: 438768651Skris this is because it will reject chains with invalid extensions whereas 438868651Skris every previous version of OpenSSL and SSLeay made no checks at all. 438968651Skris 439068651Skris Trust code: checks the root CA for the relevant trust settings. Trust 439168651Skris settings have an initial value consistent with the verify purpose: e.g. 439268651Skris if the verify purpose is for SSL client use it expects the CA to be 439368651Skris trusted for SSL client use. However the default value can be changed to 439468651Skris permit custom trust settings: one example of this would be to only trust 439568651Skris certificates from a specific "secure" set of CAs. 439668651Skris 439768651Skris Also added X509_STORE_CTX_new() and X509_STORE_CTX_free() functions 439868651Skris which should be used for version portability: especially since the 439968651Skris verify structure is likely to change more often now. 440068651Skris 440168651Skris SSL integration. Add purpose and trust to SSL_CTX and SSL and functions 440268651Skris to set them. If not set then assume SSL clients will verify SSL servers 440368651Skris and vice versa. 440468651Skris 440568651Skris Two new options to the verify program: -untrusted allows a set of 440668651Skris untrusted certificates to be passed in and -purpose which sets the 440768651Skris intended purpose of the certificate. If a purpose is set then the 440868651Skris new chain verify code is used to check extension consistency. 440968651Skris [Steve Henson] 441068651Skris 441168651Skris *) Support for the authority information access extension. 441268651Skris [Steve Henson] 441368651Skris 441468651Skris *) Modify RSA and DSA PEM read routines to transparently handle 441568651Skris PKCS#8 format private keys. New *_PUBKEY_* functions that handle 441668651Skris public keys in a format compatible with certificate 441768651Skris SubjectPublicKeyInfo structures. Unfortunately there were already 441868651Skris functions called *_PublicKey_* which used various odd formats so 441968651Skris these are retained for compatibility: however the DSA variants were 442068651Skris never in a public release so they have been deleted. Changed dsa/rsa 442168651Skris utilities to handle the new format: note no releases ever handled public 442268651Skris keys so we should be OK. 442368651Skris 442468651Skris The primary motivation for this change is to avoid the same fiasco 442568651Skris that dogs private keys: there are several incompatible private key 442668651Skris formats some of which are standard and some OpenSSL specific and 442768651Skris require various evil hacks to allow partial transparent handling and 442868651Skris even then it doesn't work with DER formats. Given the option anything 442968651Skris other than PKCS#8 should be dumped: but the other formats have to 443068651Skris stay in the name of compatibility. 443168651Skris 443268651Skris With public keys and the benefit of hindsight one standard format 443368651Skris is used which works with EVP_PKEY, RSA or DSA structures: though 443468651Skris it clearly returns an error if you try to read the wrong kind of key. 443568651Skris 443668651Skris Added a -pubkey option to the 'x509' utility to output the public key. 443768651Skris Also rename the EVP_PKEY_get_*() to EVP_PKEY_rget_*() 443868651Skris (renamed to EVP_PKEY_get1_*() in the OpenSSL 0.9.5 release) and add 443968651Skris EVP_PKEY_rset_*() functions (renamed to EVP_PKEY_set1_*()) 444068651Skris that do the same as the EVP_PKEY_assign_*() except they up the 444168651Skris reference count of the added key (they don't "swallow" the 444268651Skris supplied key). 444368651Skris [Steve Henson] 444468651Skris 444568651Skris *) Fixes to crypto/x509/by_file.c the code to read in certificates and 444668651Skris CRLs would fail if the file contained no certificates or no CRLs: 444768651Skris added a new function to read in both types and return the number 444868651Skris read: this means that if none are read it will be an error. The 444968651Skris DER versions of the certificate and CRL reader would always fail 445068651Skris because it isn't possible to mix certificates and CRLs in DER format 445168651Skris without choking one or the other routine. Changed this to just read 445268651Skris a certificate: this is the best we can do. Also modified the code 445368651Skris in apps/verify.c to take notice of return codes: it was previously 445468651Skris attempting to read in certificates from NULL pointers and ignoring 445568651Skris any errors: this is one reason why the cert and CRL reader seemed 445668651Skris to work. It doesn't check return codes from the default certificate 445768651Skris routines: these may well fail if the certificates aren't installed. 445868651Skris [Steve Henson] 445968651Skris 446068651Skris *) Code to support otherName option in GeneralName. 446168651Skris [Steve Henson] 446268651Skris 446368651Skris *) First update to verify code. Change the verify utility 446468651Skris so it warns if it is passed a self signed certificate: 446568651Skris for consistency with the normal behaviour. X509_verify 446668651Skris has been modified to it will now verify a self signed 446768651Skris certificate if *exactly* the same certificate appears 446868651Skris in the store: it was previously impossible to trust a 446968651Skris single self signed certificate. This means that: 447068651Skris openssl verify ss.pem 447168651Skris now gives a warning about a self signed certificate but 447268651Skris openssl verify -CAfile ss.pem ss.pem 447368651Skris is OK. 447468651Skris [Steve Henson] 447568651Skris 447668651Skris *) For servers, store verify_result in SSL_SESSION data structure 447768651Skris (and add it to external session representation). 447868651Skris This is needed when client certificate verifications fails, 447968651Skris but an application-provided verification callback (set by 448068651Skris SSL_CTX_set_cert_verify_callback) allows accepting the session 448168651Skris anyway (i.e. leaves x509_store_ctx->error != X509_V_OK 448268651Skris but returns 1): When the session is reused, we have to set 448368651Skris ssl->verify_result to the appropriate error code to avoid 448468651Skris security holes. 448568651Skris [Bodo Moeller, problem pointed out by Lutz Jaenicke] 448668651Skris 448768651Skris *) Fix a bug in the new PKCS#7 code: it didn't consider the 448868651Skris case in PKCS7_dataInit() where the signed PKCS7 structure 448968651Skris didn't contain any existing data because it was being created. 449068651Skris [Po-Cheng Chen <pocheng@nst.com.tw>, slightly modified by Steve Henson] 449168651Skris 449268651Skris *) Add a salt to the key derivation routines in enc.c. This 449368651Skris forms the first 8 bytes of the encrypted file. Also add a 449468651Skris -S option to allow a salt to be input on the command line. 449568651Skris [Steve Henson] 449668651Skris 449768651Skris *) New function X509_cmp(). Oddly enough there wasn't a function 449868651Skris to compare two certificates. We do this by working out the SHA1 449968651Skris hash and comparing that. X509_cmp() will be needed by the trust 450068651Skris code. 450168651Skris [Steve Henson] 450268651Skris 450368651Skris *) SSL_get1_session() is like SSL_get_session(), but increments 450468651Skris the reference count in the SSL_SESSION returned. 450568651Skris [Geoff Thorpe <geoff@eu.c2.net>] 450668651Skris 450768651Skris *) Fix for 'req': it was adding a null to request attributes. 450868651Skris Also change the X509_LOOKUP and X509_INFO code to handle 450968651Skris certificate auxiliary information. 451068651Skris [Steve Henson] 451168651Skris 451268651Skris *) Add support for 40 and 64 bit RC2 and RC4 algorithms: document 451368651Skris the 'enc' command. 451468651Skris [Steve Henson] 451568651Skris 451668651Skris *) Add the possibility to add extra information to the memory leak 451768651Skris detecting output, to form tracebacks, showing from where each 451868651Skris allocation was originated: CRYPTO_push_info("constant string") adds 451968651Skris the string plus current file name and line number to a per-thread 452068651Skris stack, CRYPTO_pop_info() does the obvious, CRYPTO_remove_all_info() 452168651Skris is like calling CYRPTO_pop_info() until the stack is empty. 452268651Skris Also updated memory leak detection code to be multi-thread-safe. 452368651Skris [Richard Levitte] 452468651Skris 452568651Skris *) Add options -text and -noout to pkcs7 utility and delete the 452668651Skris encryption options which never did anything. Update docs. 452768651Skris [Steve Henson] 452868651Skris 452968651Skris *) Add options to some of the utilities to allow the pass phrase 453068651Skris to be included on either the command line (not recommended on 453168651Skris OSes like Unix) or read from the environment. Update the 453268651Skris manpages and fix a few bugs. 453368651Skris [Steve Henson] 453468651Skris 453568651Skris *) Add a few manpages for some of the openssl commands. 453668651Skris [Steve Henson] 453768651Skris 453868651Skris *) Fix the -revoke option in ca. It was freeing up memory twice, 453968651Skris leaking and not finding already revoked certificates. 454068651Skris [Steve Henson] 454168651Skris 454268651Skris *) Extensive changes to support certificate auxiliary information. 454368651Skris This involves the use of X509_CERT_AUX structure and X509_AUX 454468651Skris functions. An X509_AUX function such as PEM_read_X509_AUX() 454568651Skris can still read in a certificate file in the usual way but it 454668651Skris will also read in any additional "auxiliary information". By 454768651Skris doing things this way a fair degree of compatibility can be 454868651Skris retained: existing certificates can have this information added 454968651Skris using the new 'x509' options. 455068651Skris 455168651Skris Current auxiliary information includes an "alias" and some trust 455268651Skris settings. The trust settings will ultimately be used in enhanced 455368651Skris certificate chain verification routines: currently a certificate 455468651Skris can only be trusted if it is self signed and then it is trusted 455568651Skris for all purposes. 455668651Skris [Steve Henson] 455768651Skris 455868651Skris *) Fix assembler for Alpha (tested only on DEC OSF not Linux or *BSD). 455968651Skris The problem was that one of the replacement routines had not been working 456068651Skris since SSLeay releases. For now the offending routine has been replaced 456168651Skris with non-optimised assembler. Even so, this now gives around 95% 456268651Skris performance improvement for 1024 bit RSA signs. 456368651Skris [Mark Cox] 456468651Skris 456568651Skris *) Hack to fix PKCS#7 decryption when used with some unorthodox RC2 456668651Skris handling. Most clients have the effective key size in bits equal to 456768651Skris the key length in bits: so a 40 bit RC2 key uses a 40 bit (5 byte) key. 456868651Skris A few however don't do this and instead use the size of the decrypted key 456968651Skris to determine the RC2 key length and the AlgorithmIdentifier to determine 457068651Skris the effective key length. In this case the effective key length can still 457168651Skris be 40 bits but the key length can be 168 bits for example. This is fixed 457268651Skris by manually forcing an RC2 key into the EVP_PKEY structure because the 457368651Skris EVP code can't currently handle unusual RC2 key sizes: it always assumes 457468651Skris the key length and effective key length are equal. 457568651Skris [Steve Henson] 457668651Skris 457768651Skris *) Add a bunch of functions that should simplify the creation of 457868651Skris X509_NAME structures. Now you should be able to do: 457968651Skris X509_NAME_add_entry_by_txt(nm, "CN", MBSTRING_ASC, "Steve", -1, -1, 0); 458068651Skris and have it automatically work out the correct field type and fill in 458168651Skris the structures. The more adventurous can try: 458268651Skris X509_NAME_add_entry_by_txt(nm, field, MBSTRING_UTF8, str, -1, -1, 0); 458368651Skris and it will (hopefully) work out the correct multibyte encoding. 458468651Skris [Steve Henson] 458568651Skris 458668651Skris *) Change the 'req' utility to use the new field handling and multibyte 458768651Skris copy routines. Before the DN field creation was handled in an ad hoc 458868651Skris way in req, ca, and x509 which was rather broken and didn't support 458968651Skris BMPStrings or UTF8Strings. Since some software doesn't implement 459068651Skris BMPStrings or UTF8Strings yet, they can be enabled using the config file 459168651Skris using the dirstring_type option. See the new comment in the default 459268651Skris openssl.cnf for more info. 459368651Skris [Steve Henson] 459468651Skris 459568651Skris *) Make crypto/rand/md_rand.c more robust: 459668651Skris - Assure unique random numbers after fork(). 459768651Skris - Make sure that concurrent threads access the global counter and 459868651Skris md serializably so that we never lose entropy in them 459968651Skris or use exactly the same state in multiple threads. 460068651Skris Access to the large state is not always serializable because 460168651Skris the additional locking could be a performance killer, and 460268651Skris md should be large enough anyway. 460368651Skris [Bodo Moeller] 460468651Skris 460568651Skris *) New file apps/app_rand.c with commonly needed functionality 460668651Skris for handling the random seed file. 460768651Skris 460868651Skris Use the random seed file in some applications that previously did not: 460968651Skris ca, 461068651Skris dsaparam -genkey (which also ignored its '-rand' option), 461168651Skris s_client, 461268651Skris s_server, 461368651Skris x509 (when signing). 461468651Skris Except on systems with /dev/urandom, it is crucial to have a random 461568651Skris seed file at least for key creation, DSA signing, and for DH exchanges; 461668651Skris for RSA signatures we could do without one. 461768651Skris 461868651Skris gendh and gendsa (unlike genrsa) used to read only the first byte 461968651Skris of each file listed in the '-rand' option. The function as previously 462068651Skris found in genrsa is now in app_rand.c and is used by all programs 462168651Skris that support '-rand'. 462268651Skris [Bodo Moeller] 462368651Skris 462468651Skris *) In RAND_write_file, use mode 0600 for creating files; 462568651Skris don't just chmod when it may be too late. 462668651Skris [Bodo Moeller] 462768651Skris 462868651Skris *) Report an error from X509_STORE_load_locations 462968651Skris when X509_LOOKUP_load_file or X509_LOOKUP_add_dir failed. 463068651Skris [Bill Perry] 463168651Skris 463268651Skris *) New function ASN1_mbstring_copy() this copies a string in either 463368651Skris ASCII, Unicode, Universal (4 bytes per character) or UTF8 format 463468651Skris into an ASN1_STRING type. A mask of permissible types is passed 463568651Skris and it chooses the "minimal" type to use or an error if not type 463668651Skris is suitable. 463768651Skris [Steve Henson] 463868651Skris 463968651Skris *) Add function equivalents to the various macros in asn1.h. The old 464068651Skris macros are retained with an M_ prefix. Code inside the library can 464168651Skris use the M_ macros. External code (including the openssl utility) 464268651Skris should *NOT* in order to be "shared library friendly". 464368651Skris [Steve Henson] 464468651Skris 464568651Skris *) Add various functions that can check a certificate's extensions 464668651Skris to see if it usable for various purposes such as SSL client, 464768651Skris server or S/MIME and CAs of these types. This is currently 464868651Skris VERY EXPERIMENTAL but will ultimately be used for certificate chain 464968651Skris verification. Also added a -purpose flag to x509 utility to 465068651Skris print out all the purposes. 465168651Skris [Steve Henson] 465268651Skris 465368651Skris *) Add a CRYPTO_EX_DATA to X509 certificate structure and associated 465468651Skris functions. 465568651Skris [Steve Henson] 465668651Skris 465768651Skris *) New X509V3_{X509,CRL,REVOKED}_get_d2i() functions. These will search 465868651Skris for, obtain and decode and extension and obtain its critical flag. 465968651Skris This allows all the necessary extension code to be handled in a 466068651Skris single function call. 466168651Skris [Steve Henson] 466268651Skris 466368651Skris *) RC4 tune-up featuring 30-40% performance improvement on most RISC 466468651Skris platforms. See crypto/rc4/rc4_enc.c for further details. 466568651Skris [Andy Polyakov] 466668651Skris 466768651Skris *) New -noout option to asn1parse. This causes no output to be produced 466868651Skris its main use is when combined with -strparse and -out to extract data 466968651Skris from a file (which may not be in ASN.1 format). 467068651Skris [Steve Henson] 467168651Skris 467268651Skris *) Fix for pkcs12 program. It was hashing an invalid certificate pointer 467368651Skris when producing the local key id. 467468651Skris [Richard Levitte <levitte@stacken.kth.se>] 467568651Skris 467668651Skris *) New option -dhparam in s_server. This allows a DH parameter file to be 467768651Skris stated explicitly. If it is not stated then it tries the first server 467868651Skris certificate file. The previous behaviour hard coded the filename 467968651Skris "server.pem". 468068651Skris [Steve Henson] 468168651Skris 468268651Skris *) Add -pubin and -pubout options to the rsa and dsa commands. These allow 468368651Skris a public key to be input or output. For example: 468468651Skris openssl rsa -in key.pem -pubout -out pubkey.pem 468568651Skris Also added necessary DSA public key functions to handle this. 468668651Skris [Steve Henson] 468789837Skris 468868651Skris *) Fix so PKCS7_dataVerify() doesn't crash if no certificates are contained 468968651Skris in the message. This was handled by allowing 469068651Skris X509_find_by_issuer_and_serial() to tolerate a NULL passed to it. 469168651Skris [Steve Henson, reported by Sampo Kellomaki <sampo@mail.neuronio.pt>] 469268651Skris 469368651Skris *) Fix for bug in d2i_ASN1_bytes(): other ASN1 functions add an extra null 469468651Skris to the end of the strings whereas this didn't. This would cause problems 469568651Skris if strings read with d2i_ASN1_bytes() were later modified. 469668651Skris [Steve Henson, reported by Arne Ansper <arne@ats.cyber.ee>] 469768651Skris 469868651Skris *) Fix for base64 decode bug. When a base64 bio reads only one line of 469968651Skris data and it contains EOF it will end up returning an error. This is 470068651Skris caused by input 46 bytes long. The cause is due to the way base64 470168651Skris BIOs find the start of base64 encoded data. They do this by trying a 470268651Skris trial decode on each line until they find one that works. When they 470368651Skris do a flag is set and it starts again knowing it can pass all the 470468651Skris data directly through the decoder. Unfortunately it doesn't reset 470568651Skris the context it uses. This means that if EOF is reached an attempt 470668651Skris is made to pass two EOFs through the context and this causes the 470768651Skris resulting error. This can also cause other problems as well. As is 470868651Skris usual with these problems it takes *ages* to find and the fix is 470968651Skris trivial: move one line. 471068651Skris [Steve Henson, reported by ian@uns.ns.ac.yu (Ivan Nejgebauer) ] 471168651Skris 471268651Skris *) Ugly workaround to get s_client and s_server working under Windows. The 471368651Skris old code wouldn't work because it needed to select() on sockets and the 471468651Skris tty (for keypresses and to see if data could be written). Win32 only 471568651Skris supports select() on sockets so we select() with a 1s timeout on the 471668651Skris sockets and then see if any characters are waiting to be read, if none 471768651Skris are present then we retry, we also assume we can always write data to 471868651Skris the tty. This isn't nice because the code then blocks until we've 471968651Skris received a complete line of data and it is effectively polling the 472068651Skris keyboard at 1s intervals: however it's quite a bit better than not 472168651Skris working at all :-) A dedicated Windows application might handle this 472268651Skris with an event loop for example. 472368651Skris [Steve Henson] 472468651Skris 472568651Skris *) Enhance RSA_METHOD structure. Now there are two extra methods, rsa_sign 472668651Skris and rsa_verify. When the RSA_FLAGS_SIGN_VER option is set these functions 472768651Skris will be called when RSA_sign() and RSA_verify() are used. This is useful 472868651Skris if rsa_pub_dec() and rsa_priv_enc() equivalents are not available. 472968651Skris For this to work properly RSA_public_decrypt() and RSA_private_encrypt() 473068651Skris should *not* be used: RSA_sign() and RSA_verify() must be used instead. 473168651Skris This necessitated the support of an extra signature type NID_md5_sha1 473268651Skris for SSL signatures and modifications to the SSL library to use it instead 473368651Skris of calling RSA_public_decrypt() and RSA_private_encrypt(). 473468651Skris [Steve Henson] 473568651Skris 473668651Skris *) Add new -verify -CAfile and -CApath options to the crl program, these 473768651Skris will lookup a CRL issuers certificate and verify the signature in a 473868651Skris similar way to the verify program. Tidy up the crl program so it 473968651Skris no longer accesses structures directly. Make the ASN1 CRL parsing a bit 474068651Skris less strict. It will now permit CRL extensions even if it is not 474168651Skris a V2 CRL: this will allow it to tolerate some broken CRLs. 474268651Skris [Steve Henson] 474368651Skris 474468651Skris *) Initialize all non-automatic variables each time one of the openssl 474568651Skris sub-programs is started (this is necessary as they may be started 474668651Skris multiple times from the "OpenSSL>" prompt). 474768651Skris [Lennart Bang, Bodo Moeller] 474868651Skris 474968651Skris *) Preliminary compilation option RSA_NULL which disables RSA crypto without 475068651Skris removing all other RSA functionality (this is what NO_RSA does). This 475168651Skris is so (for example) those in the US can disable those operations covered 475268651Skris by the RSA patent while allowing storage and parsing of RSA keys and RSA 475368651Skris key generation. 475468651Skris [Steve Henson] 475568651Skris 475668651Skris *) Non-copying interface to BIO pairs. 475768651Skris (still largely untested) 475868651Skris [Bodo Moeller] 475968651Skris 476068651Skris *) New function ANS1_tag2str() to convert an ASN1 tag to a descriptive 476168651Skris ASCII string. This was handled independently in various places before. 476268651Skris [Steve Henson] 476368651Skris 476468651Skris *) New functions UTF8_getc() and UTF8_putc() that parse and generate 476568651Skris UTF8 strings a character at a time. 476668651Skris [Steve Henson] 476768651Skris 476868651Skris *) Use client_version from client hello to select the protocol 476968651Skris (s23_srvr.c) and for RSA client key exchange verification 477068651Skris (s3_srvr.c), as required by the SSL 3.0/TLS 1.0 specifications. 477168651Skris [Bodo Moeller] 477268651Skris 477368651Skris *) Add various utility functions to handle SPKACs, these were previously 477468651Skris handled by poking round in the structure internals. Added new function 477568651Skris NETSCAPE_SPKI_print() to print out SPKAC and a new utility 'spkac' to 477668651Skris print, verify and generate SPKACs. Based on an original idea from 477768651Skris Massimiliano Pala <madwolf@comune.modena.it> but extensively modified. 477868651Skris [Steve Henson] 477968651Skris 478068651Skris *) RIPEMD160 is operational on all platforms and is back in 'make test'. 478168651Skris [Andy Polyakov] 478268651Skris 478368651Skris *) Allow the config file extension section to be overwritten on the 478468651Skris command line. Based on an original idea from Massimiliano Pala 478568651Skris <madwolf@comune.modena.it>. The new option is called -extensions 478668651Skris and can be applied to ca, req and x509. Also -reqexts to override 478768651Skris the request extensions in req and -crlexts to override the crl extensions 478868651Skris in ca. 478968651Skris [Steve Henson] 479068651Skris 479168651Skris *) Add new feature to the SPKAC handling in ca. Now you can include 479268651Skris the same field multiple times by preceding it by "XXXX." for example: 479368651Skris 1.OU="Unit name 1" 479468651Skris 2.OU="Unit name 2" 479568651Skris this is the same syntax as used in the req config file. 479668651Skris [Steve Henson] 479768651Skris 479868651Skris *) Allow certificate extensions to be added to certificate requests. These 479968651Skris are specified in a 'req_extensions' option of the req section of the 480068651Skris config file. They can be printed out with the -text option to req but 480168651Skris are otherwise ignored at present. 480268651Skris [Steve Henson] 480368651Skris 480468651Skris *) Fix a horrible bug in enc_read() in crypto/evp/bio_enc.c: if the first 480568651Skris data read consists of only the final block it would not decrypted because 480668651Skris EVP_CipherUpdate() would correctly report zero bytes had been decrypted. 480768651Skris A misplaced 'break' also meant the decrypted final block might not be 480868651Skris copied until the next read. 480968651Skris [Steve Henson] 481068651Skris 481168651Skris *) Initial support for DH_METHOD. Again based on RSA_METHOD. Also added 481268651Skris a few extra parameters to the DH structure: these will be useful if 481368651Skris for example we want the value of 'q' or implement X9.42 DH. 481468651Skris [Steve Henson] 481568651Skris 481668651Skris *) Initial support for DSA_METHOD. This is based on the RSA_METHOD and 481768651Skris provides hooks that allow the default DSA functions or functions on a 481868651Skris "per key" basis to be replaced. This allows hardware acceleration and 481968651Skris hardware key storage to be handled without major modification to the 482068651Skris library. Also added low level modexp hooks and CRYPTO_EX structure and 482168651Skris associated functions. 482268651Skris [Steve Henson] 482368651Skris 482468651Skris *) Add a new flag to memory BIOs, BIO_FLAG_MEM_RDONLY. This marks the BIO 482568651Skris as "read only": it can't be written to and the buffer it points to will 482668651Skris not be freed. Reading from a read only BIO is much more efficient than 482768651Skris a normal memory BIO. This was added because there are several times when 482868651Skris an area of memory needs to be read from a BIO. The previous method was 482968651Skris to create a memory BIO and write the data to it, this results in two 483068651Skris copies of the data and an O(n^2) reading algorithm. There is a new 483168651Skris function BIO_new_mem_buf() which creates a read only memory BIO from 483268651Skris an area of memory. Also modified the PKCS#7 routines to use read only 483368651Skris memory BIOs. 483468651Skris [Steve Henson] 483568651Skris 483668651Skris *) Bugfix: ssl23_get_client_hello did not work properly when called in 483768651Skris state SSL23_ST_SR_CLNT_HELLO_B, i.e. when the first 7 bytes of 483868651Skris a SSLv2-compatible client hello for SSLv3 or TLSv1 could be read, 483968651Skris but a retry condition occured while trying to read the rest. 484068651Skris [Bodo Moeller] 484168651Skris 484268651Skris *) The PKCS7_ENC_CONTENT_new() function was setting the content type as 484368651Skris NID_pkcs7_encrypted by default: this was wrong since this should almost 484468651Skris always be NID_pkcs7_data. Also modified the PKCS7_set_type() to handle 484568651Skris the encrypted data type: this is a more sensible place to put it and it 484668651Skris allows the PKCS#12 code to be tidied up that duplicated this 484768651Skris functionality. 484868651Skris [Steve Henson] 484968651Skris 485068651Skris *) Changed obj_dat.pl script so it takes its input and output files on 485168651Skris the command line. This should avoid shell escape redirection problems 485268651Skris under Win32. 485368651Skris [Steve Henson] 485468651Skris 485568651Skris *) Initial support for certificate extension requests, these are included 485668651Skris in things like Xenroll certificate requests. Included functions to allow 485768651Skris extensions to be obtained and added. 485868651Skris [Steve Henson] 485968651Skris 486068651Skris *) -crlf option to s_client and s_server for sending newlines as 486168651Skris CRLF (as required by many protocols). 486268651Skris [Bodo Moeller] 486368651Skris 486468651Skris Changes between 0.9.3a and 0.9.4 [09 Aug 1999] 486568651Skris 486668651Skris *) Install libRSAglue.a when OpenSSL is built with RSAref. 486768651Skris [Ralf S. Engelschall] 486868651Skris 486968651Skris *) A few more ``#ifndef NO_FP_API / #endif'' pairs for consistency. 487068651Skris [Andrija Antonijevic <TheAntony2@bigfoot.com>] 487168651Skris 487268651Skris *) Fix -startdate and -enddate (which was missing) arguments to 'ca' 487368651Skris program. 487468651Skris [Steve Henson] 487568651Skris 487668651Skris *) New function DSA_dup_DH, which duplicates DSA parameters/keys as 487768651Skris DH parameters/keys (q is lost during that conversion, but the resulting 487868651Skris DH parameters contain its length). 487968651Skris 488068651Skris For 1024-bit p, DSA_generate_parameters followed by DSA_dup_DH is 488168651Skris much faster than DH_generate_parameters (which creates parameters 488268651Skris where p = 2*q + 1), and also the smaller q makes DH computations 488368651Skris much more efficient (160-bit exponentiation instead of 1024-bit 488468651Skris exponentiation); so this provides a convenient way to support DHE 488568651Skris ciphersuites in SSL/TLS servers (see ssl/ssltest.c). It is of 488668651Skris utter importance to use 488768651Skris SSL_CTX_set_options(s_ctx, SSL_OP_SINGLE_DH_USE); 488868651Skris or 488968651Skris SSL_set_options(s_ctx, SSL_OP_SINGLE_DH_USE); 489068651Skris when such DH parameters are used, because otherwise small subgroup 489168651Skris attacks may become possible! 489268651Skris [Bodo Moeller] 489368651Skris 489468651Skris *) Avoid memory leak in i2d_DHparams. 489568651Skris [Bodo Moeller] 489668651Skris 489768651Skris *) Allow the -k option to be used more than once in the enc program: 489868651Skris this allows the same encrypted message to be read by multiple recipients. 489968651Skris [Steve Henson] 490068651Skris 490168651Skris *) New function OBJ_obj2txt(buf, buf_len, a, no_name), this converts 490268651Skris an ASN1_OBJECT to a text string. If the "no_name" parameter is set then 490368651Skris it will always use the numerical form of the OID, even if it has a short 490468651Skris or long name. 490568651Skris [Steve Henson] 490668651Skris 490768651Skris *) Added an extra RSA flag: RSA_FLAG_EXT_PKEY. Previously the rsa_mod_exp 490868651Skris method only got called if p,q,dmp1,dmq1,iqmp components were present, 490968651Skris otherwise bn_mod_exp was called. In the case of hardware keys for example 491068651Skris no private key components need be present and it might store extra data 491168651Skris in the RSA structure, which cannot be accessed from bn_mod_exp. 491268651Skris By setting RSA_FLAG_EXT_PKEY rsa_mod_exp will always be called for 491368651Skris private key operations. 491468651Skris [Steve Henson] 491568651Skris 491668651Skris *) Added support for SPARC Linux. 491768651Skris [Andy Polyakov] 491868651Skris 491968651Skris *) pem_password_cb function type incompatibly changed from 492068651Skris typedef int pem_password_cb(char *buf, int size, int rwflag); 492168651Skris to 492268651Skris ....(char *buf, int size, int rwflag, void *userdata); 492368651Skris so that applications can pass data to their callbacks: 492468651Skris The PEM[_ASN1]_{read,write}... functions and macros now take an 492568651Skris additional void * argument, which is just handed through whenever 492668651Skris the password callback is called. 492768651Skris [Damien Miller <dmiller@ilogic.com.au>; tiny changes by Bodo Moeller] 492868651Skris 492968651Skris New function SSL_CTX_set_default_passwd_cb_userdata. 493068651Skris 493168651Skris Compatibility note: As many C implementations push function arguments 493268651Skris onto the stack in reverse order, the new library version is likely to 493368651Skris interoperate with programs that have been compiled with the old 493468651Skris pem_password_cb definition (PEM_whatever takes some data that 493568651Skris happens to be on the stack as its last argument, and the callback 493668651Skris just ignores this garbage); but there is no guarantee whatsoever that 493768651Skris this will work. 493868651Skris 493968651Skris *) The -DPLATFORM="\"$(PLATFORM)\"" definition and the similar -DCFLAGS=... 494068651Skris (both in crypto/Makefile.ssl for use by crypto/cversion.c) caused 494168651Skris problems not only on Windows, but also on some Unix platforms. 494268651Skris To avoid problematic command lines, these definitions are now in an 494368651Skris auto-generated file crypto/buildinf.h (created by crypto/Makefile.ssl 494468651Skris for standard "make" builds, by util/mk1mf.pl for "mk1mf" builds). 494568651Skris [Bodo Moeller] 494668651Skris 494768651Skris *) MIPS III/IV assembler module is reimplemented. 494868651Skris [Andy Polyakov] 494968651Skris 495068651Skris *) More DES library cleanups: remove references to srand/rand and 495168651Skris delete an unused file. 495268651Skris [Ulf M�ller] 495368651Skris 495468651Skris *) Add support for the the free Netwide assembler (NASM) under Win32, 495568651Skris since not many people have MASM (ml) and it can be hard to obtain. 495668651Skris This is currently experimental but it seems to work OK and pass all 495768651Skris the tests. Check out INSTALL.W32 for info. 495868651Skris [Steve Henson] 495968651Skris 496068651Skris *) Fix memory leaks in s3_clnt.c: All non-anonymous SSL3/TLS1 connections 496168651Skris without temporary keys kept an extra copy of the server key, 496268651Skris and connections with temporary keys did not free everything in case 496368651Skris of an error. 496468651Skris [Bodo Moeller] 496568651Skris 496668651Skris *) New function RSA_check_key and new openssl rsa option -check 496768651Skris for verifying the consistency of RSA keys. 496868651Skris [Ulf Moeller, Bodo Moeller] 496968651Skris 497068651Skris *) Various changes to make Win32 compile work: 497168651Skris 1. Casts to avoid "loss of data" warnings in p5_crpt2.c 497268651Skris 2. Change unsigned int to int in b_dump.c to avoid "signed/unsigned 497368651Skris comparison" warnings. 497468651Skris 3. Add sk_<TYPE>_sort to DEF file generator and do make update. 497568651Skris [Steve Henson] 497668651Skris 497768651Skris *) Add a debugging option to PKCS#5 v2 key generation function: when 497868651Skris you #define DEBUG_PKCS5V2 passwords, salts, iteration counts and 497968651Skris derived keys are printed to stderr. 498068651Skris [Steve Henson] 498168651Skris 498268651Skris *) Copy the flags in ASN1_STRING_dup(). 498368651Skris [Roman E. Pavlov <pre@mo.msk.ru>] 498468651Skris 498568651Skris *) The x509 application mishandled signing requests containing DSA 498668651Skris keys when the signing key was also DSA and the parameters didn't match. 498768651Skris 498868651Skris It was supposed to omit the parameters when they matched the signing key: 498968651Skris the verifying software was then supposed to automatically use the CA's 499068651Skris parameters if they were absent from the end user certificate. 499168651Skris 499268651Skris Omitting parameters is no longer recommended. The test was also 499368651Skris the wrong way round! This was probably due to unusual behaviour in 499468651Skris EVP_cmp_parameters() which returns 1 if the parameters match. 499568651Skris This meant that parameters were omitted when they *didn't* match and 499668651Skris the certificate was useless. Certificates signed with 'ca' didn't have 499768651Skris this bug. 499868651Skris [Steve Henson, reported by Doug Erickson <Doug.Erickson@Part.NET>] 499968651Skris 500068651Skris *) Memory leak checking (-DCRYPTO_MDEBUG) had some problems. 500168651Skris The interface is as follows: 500268651Skris Applications can use 500368651Skris CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ON) aka MemCheck_start(), 500468651Skris CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_OFF) aka MemCheck_stop(); 500568651Skris "off" is now the default. 500668651Skris The library internally uses 500768651Skris CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_DISABLE) aka MemCheck_off(), 500868651Skris CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ENABLE) aka MemCheck_on() 500968651Skris to disable memory-checking temporarily. 501059191Skris 501159191Skris Some inconsistent states that previously were possible (and were 501259191Skris even the default) are now avoided. 501359191Skris 501459191Skris -DCRYPTO_MDEBUG_TIME is new and additionally stores the current time 501559191Skris with each memory chunk allocated; this is occasionally more helpful 501659191Skris than just having a counter. 501759191Skris 501859191Skris -DCRYPTO_MDEBUG_THREAD is also new and adds the thread ID. 501959191Skris 502059191Skris -DCRYPTO_MDEBUG_ALL enables all of the above, plus any future 502159191Skris extensions. 502259191Skris [Bodo Moeller] 502359191Skris 502459191Skris *) Introduce "mode" for SSL structures (with defaults in SSL_CTX), 502559191Skris which largely parallels "options", but is for changing API behaviour, 502659191Skris whereas "options" are about protocol behaviour. 502759191Skris Initial "mode" flags are: 502859191Skris 502959191Skris SSL_MODE_ENABLE_PARTIAL_WRITE Allow SSL_write to report success when 503059191Skris a single record has been written. 503159191Skris SSL_MODE_ACCEPT_MOVING_WRITE_BUFFER Don't insist that SSL_write 503259191Skris retries use the same buffer location. 503359191Skris (But all of the contents must be 503459191Skris copied!) 503559191Skris [Bodo Moeller] 503659191Skris 503759191Skris *) Bugfix: SSL_set_options ignored its parameter, only SSL_CTX_set_options 503859191Skris worked. 503959191Skris 504059191Skris *) Fix problems with no-hmac etc. 504159191Skris [Ulf M�ller, pointed out by Brian Wellington <bwelling@tislabs.com>] 504259191Skris 504359191Skris *) New functions RSA_get_default_method(), RSA_set_method() and 504459191Skris RSA_get_method(). These allows replacement of RSA_METHODs without having 504559191Skris to mess around with the internals of an RSA structure. 504659191Skris [Steve Henson] 504759191Skris 504859191Skris *) Fix memory leaks in DSA_do_sign and DSA_is_prime. 504959191Skris Also really enable memory leak checks in openssl.c and in some 505059191Skris test programs. 505159191Skris [Chad C. Mulligan, Bodo Moeller] 505259191Skris 505359191Skris *) Fix a bug in d2i_ASN1_INTEGER() and i2d_ASN1_INTEGER() which can mess 505459191Skris up the length of negative integers. This has now been simplified to just 505559191Skris store the length when it is first determined and use it later, rather 505659191Skris than trying to keep track of where data is copied and updating it to 505759191Skris point to the end. 505859191Skris [Steve Henson, reported by Brien Wheeler 505959191Skris <bwheeler@authentica-security.com>] 506059191Skris 506159191Skris *) Add a new function PKCS7_signatureVerify. This allows the verification 506259191Skris of a PKCS#7 signature but with the signing certificate passed to the 506359191Skris function itself. This contrasts with PKCS7_dataVerify which assumes the 506459191Skris certificate is present in the PKCS#7 structure. This isn't always the 506559191Skris case: certificates can be omitted from a PKCS#7 structure and be 506659191Skris distributed by "out of band" means (such as a certificate database). 506759191Skris [Steve Henson] 506859191Skris 506959191Skris *) Complete the PEM_* macros with DECLARE_PEM versions to replace the 507059191Skris function prototypes in pem.h, also change util/mkdef.pl to add the 507159191Skris necessary function names. 507259191Skris [Steve Henson] 507359191Skris 507459191Skris *) mk1mf.pl (used by Windows builds) did not properly read the 507559191Skris options set by Configure in the top level Makefile, and Configure 507659191Skris was not even able to write more than one option correctly. 507759191Skris Fixed, now "no-idea no-rc5 -DCRYPTO_MDEBUG" etc. works as intended. 507859191Skris [Bodo Moeller] 507959191Skris 508059191Skris *) New functions CONF_load_bio() and CONF_load_fp() to allow a config 508159191Skris file to be loaded from a BIO or FILE pointer. The BIO version will 508259191Skris for example allow memory BIOs to contain config info. 508359191Skris [Steve Henson] 508459191Skris 508559191Skris *) New function "CRYPTO_num_locks" that returns CRYPTO_NUM_LOCKS. 508659191Skris Whoever hopes to achieve shared-library compatibility across versions 508759191Skris must use this, not the compile-time macro. 508859191Skris (Exercise 0.9.4: Which is the minimum library version required by 508959191Skris such programs?) 509059191Skris Note: All this applies only to multi-threaded programs, others don't 509159191Skris need locks. 509259191Skris [Bodo Moeller] 509359191Skris 509459191Skris *) Add missing case to s3_clnt.c state machine -- one of the new SSL tests 509559191Skris through a BIO pair triggered the default case, i.e. 509659191Skris SSLerr(...,SSL_R_UNKNOWN_STATE). 509759191Skris [Bodo Moeller] 509859191Skris 509959191Skris *) New "BIO pair" concept (crypto/bio/bss_bio.c) so that applications 510059191Skris can use the SSL library even if none of the specific BIOs is 510159191Skris appropriate. 510259191Skris [Bodo Moeller] 510359191Skris 510459191Skris *) Fix a bug in i2d_DSAPublicKey() which meant it returned the wrong value 510559191Skris for the encoded length. 510659191Skris [Jeon KyoungHo <khjeon@sds.samsung.co.kr>] 510759191Skris 510859191Skris *) Add initial documentation of the X509V3 functions. 510959191Skris [Steve Henson] 511059191Skris 511159191Skris *) Add a new pair of functions PEM_write_PKCS8PrivateKey() and 511259191Skris PEM_write_bio_PKCS8PrivateKey() that are equivalent to 511359191Skris PEM_write_PrivateKey() and PEM_write_bio_PrivateKey() but use the more 511459191Skris secure PKCS#8 private key format with a high iteration count. 511559191Skris [Steve Henson] 511659191Skris 511759191Skris *) Fix determination of Perl interpreter: A perl or perl5 511859191Skris _directory_ in $PATH was also accepted as the interpreter. 511959191Skris [Ralf S. Engelschall] 512059191Skris 512159191Skris *) Fix demos/sign/sign.c: well there wasn't anything strictly speaking 512259191Skris wrong with it but it was very old and did things like calling 512359191Skris PEM_ASN1_read() directly and used MD5 for the hash not to mention some 512459191Skris unusual formatting. 512559191Skris [Steve Henson] 512659191Skris 512759191Skris *) Fix demos/selfsign.c: it used obsolete and deleted functions, changed 512859191Skris to use the new extension code. 512959191Skris [Steve Henson] 513059191Skris 513159191Skris *) Implement the PEM_read/PEM_write functions in crypto/pem/pem_all.c 513259191Skris with macros. This should make it easier to change their form, add extra 513359191Skris arguments etc. Fix a few PEM prototypes which didn't have cipher as a 513459191Skris constant. 513559191Skris [Steve Henson] 513659191Skris 513759191Skris *) Add to configuration table a new entry that can specify an alternative 513859191Skris name for unistd.h (for pre-POSIX systems); we need this for NeXTstep, 513959191Skris according to Mark Crispin <MRC@Panda.COM>. 514059191Skris [Bodo Moeller] 514159191Skris 514259191Skris#if 0 514359191Skris *) DES CBC did not update the IV. Weird. 514459191Skris [Ben Laurie] 514559191Skris#else 514659191Skris des_cbc_encrypt does not update the IV, but des_ncbc_encrypt does. 514759191Skris Changing the behaviour of the former might break existing programs -- 514859191Skris where IV updating is needed, des_ncbc_encrypt can be used. 514959191Skris#endif 515059191Skris 515159191Skris *) When bntest is run from "make test" it drives bc to check its 515259191Skris calculations, as well as internally checking them. If an internal check 515359191Skris fails, it needs to cause bc to give a non-zero result or make test carries 515459191Skris on without noticing the failure. Fixed. 515559191Skris [Ben Laurie] 515659191Skris 515759191Skris *) DES library cleanups. 515859191Skris [Ulf M�ller] 515959191Skris 516059191Skris *) Add support for PKCS#5 v2.0 PBE algorithms. This will permit PKCS#8 to be 516159191Skris used with any cipher unlike PKCS#5 v1.5 which can at most handle 64 bit 516259191Skris ciphers. NOTE: although the key derivation function has been verified 516359191Skris against some published test vectors it has not been extensively tested 516459191Skris yet. Added a -v2 "cipher" option to pkcs8 application to allow the use 516559191Skris of v2.0. 516659191Skris [Steve Henson] 516759191Skris 516859191Skris *) Instead of "mkdir -p", which is not fully portable, use new 516959191Skris Perl script "util/mkdir-p.pl". 517059191Skris [Bodo Moeller] 517159191Skris 517259191Skris *) Rewrite the way password based encryption (PBE) is handled. It used to 517359191Skris assume that the ASN1 AlgorithmIdentifier parameter was a PBEParameter 517459191Skris structure. This was true for the PKCS#5 v1.5 and PKCS#12 PBE algorithms 517559191Skris but doesn't apply to PKCS#5 v2.0 where it can be something else. Now 517659191Skris the 'parameter' field of the AlgorithmIdentifier is passed to the 517759191Skris underlying key generation function so it must do its own ASN1 parsing. 517859191Skris This has also changed the EVP_PBE_CipherInit() function which now has a 517959191Skris 'parameter' argument instead of literal salt and iteration count values 518059191Skris and the function EVP_PBE_ALGOR_CipherInit() has been deleted. 518159191Skris [Steve Henson] 518259191Skris 518359191Skris *) Support for PKCS#5 v1.5 compatible password based encryption algorithms 518459191Skris and PKCS#8 functionality. New 'pkcs8' application linked to openssl. 518559191Skris Needed to change the PEM_STRING_EVP_PKEY value which was just "PRIVATE 518659191Skris KEY" because this clashed with PKCS#8 unencrypted string. Since this 518759191Skris value was just used as a "magic string" and not used directly its 518859191Skris value doesn't matter. 518959191Skris [Steve Henson] 519059191Skris 519159191Skris *) Introduce some semblance of const correctness to BN. Shame C doesn't 519259191Skris support mutable. 519359191Skris [Ben Laurie] 519459191Skris 519559191Skris *) "linux-sparc64" configuration (ultrapenguin). 519659191Skris [Ray Miller <ray.miller@oucs.ox.ac.uk>] 519759191Skris "linux-sparc" configuration. 519859191Skris [Christian Forster <fo@hawo.stw.uni-erlangen.de>] 519959191Skris 520059191Skris *) config now generates no-xxx options for missing ciphers. 520159191Skris [Ulf M�ller] 520259191Skris 520359191Skris *) Support the EBCDIC character set (work in progress). 520459191Skris File ebcdic.c not yet included because it has a different license. 520559191Skris [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>] 520659191Skris 520759191Skris *) Support BS2000/OSD-POSIX. 520859191Skris [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>] 520959191Skris 521059191Skris *) Make callbacks for key generation use void * instead of char *. 521159191Skris [Ben Laurie] 521259191Skris 521359191Skris *) Make S/MIME samples compile (not yet tested). 521459191Skris [Ben Laurie] 521559191Skris 521659191Skris *) Additional typesafe stacks. 521759191Skris [Ben Laurie] 521859191Skris 521959191Skris *) New configuration variants "bsdi-elf-gcc" (BSD/OS 4.x). 522059191Skris [Bodo Moeller] 522159191Skris 522259191Skris 522359191Skris Changes between 0.9.3 and 0.9.3a [29 May 1999] 522459191Skris 522559191Skris *) New configuration variant "sco5-gcc". 522659191Skris 522759191Skris *) Updated some demos. 522859191Skris [Sean O Riordain, Wade Scholine] 522959191Skris 523059191Skris *) Add missing BIO_free at exit of pkcs12 application. 523159191Skris [Wu Zhigang] 523259191Skris 523359191Skris *) Fix memory leak in conf.c. 523459191Skris [Steve Henson] 523559191Skris 523659191Skris *) Updates for Win32 to assembler version of MD5. 523759191Skris [Steve Henson] 523859191Skris 523959191Skris *) Set #! path to perl in apps/der_chop to where we found it 524059191Skris instead of using a fixed path. 524159191Skris [Bodo Moeller] 524259191Skris 524359191Skris *) SHA library changes for irix64-mips4-cc. 524459191Skris [Andy Polyakov] 524559191Skris 524659191Skris *) Improvements for VMS support. 524759191Skris [Richard Levitte] 524859191Skris 524959191Skris 525059191Skris Changes between 0.9.2b and 0.9.3 [24 May 1999] 525159191Skris 525259191Skris *) Bignum library bug fix. IRIX 6 passes "make test" now! 525359191Skris This also avoids the problems with SC4.2 and unpatched SC5. 525459191Skris [Andy Polyakov <appro@fy.chalmers.se>] 525559191Skris 525659191Skris *) New functions sk_num, sk_value and sk_set to replace the previous macros. 525759191Skris These are required because of the typesafe stack would otherwise break 525859191Skris existing code. If old code used a structure member which used to be STACK 525959191Skris and is now STACK_OF (for example cert in a PKCS7_SIGNED structure) with 526059191Skris sk_num or sk_value it would produce an error because the num, data members 526159191Skris are not present in STACK_OF. Now it just produces a warning. sk_set 526259191Skris replaces the old method of assigning a value to sk_value 526359191Skris (e.g. sk_value(x, i) = y) which the library used in a few cases. Any code 526459191Skris that does this will no longer work (and should use sk_set instead) but 526559191Skris this could be regarded as a "questionable" behaviour anyway. 526659191Skris [Steve Henson] 526759191Skris 526859191Skris *) Fix most of the other PKCS#7 bugs. The "experimental" code can now 526959191Skris correctly handle encrypted S/MIME data. 527059191Skris [Steve Henson] 527159191Skris 527259191Skris *) Change type of various DES function arguments from des_cblock 527359191Skris (which means, in function argument declarations, pointer to char) 527459191Skris to des_cblock * (meaning pointer to array with 8 char elements), 527559191Skris which allows the compiler to do more typechecking; it was like 527659191Skris that back in SSLeay, but with lots of ugly casts. 527759191Skris 527859191Skris Introduce new type const_des_cblock. 527959191Skris [Bodo Moeller] 528059191Skris 528159191Skris *) Reorganise the PKCS#7 library and get rid of some of the more obvious 528259191Skris problems: find RecipientInfo structure that matches recipient certificate 528359191Skris and initialise the ASN1 structures properly based on passed cipher. 528459191Skris [Steve Henson] 528559191Skris 528659191Skris *) Belatedly make the BN tests actually check the results. 528759191Skris [Ben Laurie] 528859191Skris 528959191Skris *) Fix the encoding and decoding of negative ASN1 INTEGERS and conversion 529059191Skris to and from BNs: it was completely broken. New compilation option 529159191Skris NEG_PUBKEY_BUG to allow for some broken certificates that encode public 529259191Skris key elements as negative integers. 529359191Skris [Steve Henson] 529459191Skris 529559191Skris *) Reorganize and speed up MD5. 529659191Skris [Andy Polyakov <appro@fy.chalmers.se>] 529759191Skris 529859191Skris *) VMS support. 529959191Skris [Richard Levitte <richard@levitte.org>] 530059191Skris 530159191Skris *) New option -out to asn1parse to allow the parsed structure to be 530259191Skris output to a file. This is most useful when combined with the -strparse 530359191Skris option to examine the output of things like OCTET STRINGS. 530459191Skris [Steve Henson] 530559191Skris 530659191Skris *) Make SSL library a little more fool-proof by not requiring any longer 530759191Skris that SSL_set_{accept,connect}_state be called before 530859191Skris SSL_{accept,connect} may be used (SSL_set_..._state is omitted 530959191Skris in many applications because usually everything *appeared* to work as 531059191Skris intended anyway -- now it really works as intended). 531159191Skris [Bodo Moeller] 531259191Skris 531359191Skris *) Move openssl.cnf out of lib/. 531459191Skris [Ulf M�ller] 531559191Skris 531659191Skris *) Fix various things to let OpenSSL even pass ``egcc -pipe -O2 -Wall 531759191Skris -Wshadow -Wpointer-arith -Wcast-align -Wmissing-prototypes 531859191Skris -Wmissing-declarations -Wnested-externs -Winline'' with EGCS 1.1.2+ 531959191Skris [Ralf S. Engelschall] 532059191Skris 532159191Skris *) Various fixes to the EVP and PKCS#7 code. It may now be able to 532259191Skris handle PKCS#7 enveloped data properly. 532359191Skris [Sebastian Akerman <sak@parallelconsulting.com>, modified by Steve] 532459191Skris 532559191Skris *) Create a duplicate of the SSL_CTX's CERT in SSL_new instead of 532659191Skris copying pointers. The cert_st handling is changed by this in 532759191Skris various ways (and thus what used to be known as ctx->default_cert 532859191Skris is now called ctx->cert, since we don't resort to s->ctx->[default_]cert 532959191Skris any longer when s->cert does not give us what we need). 533059191Skris ssl_cert_instantiate becomes obsolete by this change. 533159191Skris As soon as we've got the new code right (possibly it already is?), 533259191Skris we have solved a couple of bugs of the earlier code where s->cert 533359191Skris was used as if it could not have been shared with other SSL structures. 533459191Skris 533559191Skris Note that using the SSL API in certain dirty ways now will result 533659191Skris in different behaviour than observed with earlier library versions: 533759191Skris Changing settings for an SSL_CTX *ctx after having done s = SSL_new(ctx) 533859191Skris does not influence s as it used to. 533959191Skris 534059191Skris In order to clean up things more thoroughly, inside SSL_SESSION 534159191Skris we don't use CERT any longer, but a new structure SESS_CERT 534259191Skris that holds per-session data (if available); currently, this is 534359191Skris the peer's certificate chain and, for clients, the server's certificate 534459191Skris and temporary key. CERT holds only those values that can have 534559191Skris meaningful defaults in an SSL_CTX. 534659191Skris [Bodo Moeller] 534759191Skris 534859191Skris *) New function X509V3_EXT_i2d() to create an X509_EXTENSION structure 534959191Skris from the internal representation. Various PKCS#7 fixes: remove some 535059191Skris evil casts and set the enc_dig_alg field properly based on the signing 535159191Skris key type. 535259191Skris [Steve Henson] 535359191Skris 535459191Skris *) Allow PKCS#12 password to be set from the command line or the 535559191Skris environment. Let 'ca' get its config file name from the environment 535659191Skris variables "OPENSSL_CONF" or "SSLEAY_CONF" (for consistency with 'req' 535759191Skris and 'x509'). 535859191Skris [Steve Henson] 535959191Skris 536059191Skris *) Allow certificate policies extension to use an IA5STRING for the 536159191Skris organization field. This is contrary to the PKIX definition but 536259191Skris VeriSign uses it and IE5 only recognises this form. Document 'x509' 536359191Skris extension option. 536459191Skris [Steve Henson] 536559191Skris 536659191Skris *) Add PEDANTIC compiler flag to allow compilation with gcc -pedantic, 536759191Skris without disallowing inline assembler and the like for non-pedantic builds. 536859191Skris [Ben Laurie] 536959191Skris 537059191Skris *) Support Borland C++ builder. 537159191Skris [Janez Jere <jj@void.si>, modified by Ulf M�ller] 537259191Skris 537359191Skris *) Support Mingw32. 537459191Skris [Ulf M�ller] 537559191Skris 537659191Skris *) SHA-1 cleanups and performance enhancements. 537759191Skris [Andy Polyakov <appro@fy.chalmers.se>] 537859191Skris 537959191Skris *) Sparc v8plus assembler for the bignum library. 538059191Skris [Andy Polyakov <appro@fy.chalmers.se>] 538159191Skris 538259191Skris *) Accept any -xxx and +xxx compiler options in Configure. 538359191Skris [Ulf M�ller] 538459191Skris 538559191Skris *) Update HPUX configuration. 538659191Skris [Anonymous] 538759191Skris 538859191Skris *) Add missing sk_<type>_unshift() function to safestack.h 538959191Skris [Ralf S. Engelschall] 539059191Skris 539159191Skris *) New function SSL_CTX_use_certificate_chain_file that sets the 539259191Skris "extra_cert"s in addition to the certificate. (This makes sense 539359191Skris only for "PEM" format files, as chains as a whole are not 539459191Skris DER-encoded.) 539559191Skris [Bodo Moeller] 539659191Skris 539759191Skris *) Support verify_depth from the SSL API. 539859191Skris x509_vfy.c had what can be considered an off-by-one-error: 539959191Skris Its depth (which was not part of the external interface) 540059191Skris was actually counting the number of certificates in a chain; 540159191Skris now it really counts the depth. 540259191Skris [Bodo Moeller] 540359191Skris 540459191Skris *) Bugfix in crypto/x509/x509_cmp.c: The SSLerr macro was used 540559191Skris instead of X509err, which often resulted in confusing error 540659191Skris messages since the error codes are not globally unique 540759191Skris (e.g. an alleged error in ssl3_accept when a certificate 540859191Skris didn't match the private key). 540959191Skris 541059191Skris *) New function SSL_CTX_set_session_id_context that allows to set a default 541159191Skris value (so that you don't need SSL_set_session_id_context for each 541259191Skris connection using the SSL_CTX). 541359191Skris [Bodo Moeller] 541459191Skris 541559191Skris *) OAEP decoding bug fix. 541659191Skris [Ulf M�ller] 541759191Skris 541859191Skris *) Support INSTALL_PREFIX for package builders, as proposed by 541959191Skris David Harris. 542059191Skris [Bodo Moeller] 542159191Skris 542259191Skris *) New Configure options "threads" and "no-threads". For systems 542359191Skris where the proper compiler options are known (currently Solaris 542459191Skris and Linux), "threads" is the default. 542559191Skris [Bodo Moeller] 542659191Skris 542759191Skris *) New script util/mklink.pl as a faster substitute for util/mklink.sh. 542859191Skris [Bodo Moeller] 542959191Skris 543059191Skris *) Install various scripts to $(OPENSSLDIR)/misc, not to 543159191Skris $(INSTALLTOP)/bin -- they shouldn't clutter directories 543259191Skris such as /usr/local/bin. 543359191Skris [Bodo Moeller] 543459191Skris 543559191Skris *) "make linux-shared" to build shared libraries. 543659191Skris [Niels Poppe <niels@netbox.org>] 543759191Skris 543859191Skris *) New Configure option no-<cipher> (rsa, idea, rc5, ...). 543959191Skris [Ulf M�ller] 544059191Skris 544159191Skris *) Add the PKCS#12 API documentation to openssl.txt. Preliminary support for 544259191Skris extension adding in x509 utility. 544359191Skris [Steve Henson] 544459191Skris 544559191Skris *) Remove NOPROTO sections and error code comments. 544659191Skris [Ulf M�ller] 544759191Skris 544859191Skris *) Partial rewrite of the DEF file generator to now parse the ANSI 544959191Skris prototypes. 545059191Skris [Steve Henson] 545159191Skris 545259191Skris *) New Configure options --prefix=DIR and --openssldir=DIR. 545359191Skris [Ulf M�ller] 545459191Skris 545559191Skris *) Complete rewrite of the error code script(s). It is all now handled 545659191Skris by one script at the top level which handles error code gathering, 545759191Skris header rewriting and C source file generation. It should be much better 545859191Skris than the old method: it now uses a modified version of Ulf's parser to 545959191Skris read the ANSI prototypes in all header files (thus the old K&R definitions 546059191Skris aren't needed for error creation any more) and do a better job of 546159191Skris translating function codes into names. The old 'ASN1 error code imbedded 546259191Skris in a comment' is no longer necessary and it doesn't use .err files which 546359191Skris have now been deleted. Also the error code call doesn't have to appear all 546459191Skris on one line (which resulted in some large lines...). 546559191Skris [Steve Henson] 546659191Skris 546759191Skris *) Change #include filenames from <foo.h> to <openssl/foo.h>. 546859191Skris [Bodo Moeller] 546959191Skris 547059191Skris *) Change behaviour of ssl2_read when facing length-0 packets: Don't return 547159191Skris 0 (which usually indicates a closed connection), but continue reading. 547259191Skris [Bodo Moeller] 547359191Skris 547459191Skris *) Fix some race conditions. 547559191Skris [Bodo Moeller] 547659191Skris 547759191Skris *) Add support for CRL distribution points extension. Add Certificate 547859191Skris Policies and CRL distribution points documentation. 547959191Skris [Steve Henson] 548059191Skris 548159191Skris *) Move the autogenerated header file parts to crypto/opensslconf.h. 548259191Skris [Ulf M�ller] 548359191Skris 548459191Skris *) Fix new 56-bit DES export ciphersuites: they were using 7 bytes instead of 548559191Skris 8 of keying material. Merlin has also confirmed interop with this fix 548659191Skris between OpenSSL and Baltimore C/SSL 2.0 and J/SSL 2.0. 548759191Skris [Merlin Hughes <merlin@baltimore.ie>] 548859191Skris 548959191Skris *) Fix lots of warnings. 549059191Skris [Richard Levitte <levitte@stacken.kth.se>] 549159191Skris 549259191Skris *) In add_cert_dir() in crypto/x509/by_dir.c, break out of the loop if 549359191Skris the directory spec didn't end with a LIST_SEPARATOR_CHAR. 549459191Skris [Richard Levitte <levitte@stacken.kth.se>] 549559191Skris 549659191Skris *) Fix problems with sizeof(long) == 8. 549759191Skris [Andy Polyakov <appro@fy.chalmers.se>] 549859191Skris 549959191Skris *) Change functions to ANSI C. 550059191Skris [Ulf M�ller] 550159191Skris 550259191Skris *) Fix typos in error codes. 550359191Skris [Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>, Ulf M�ller] 550459191Skris 550559191Skris *) Remove defunct assembler files from Configure. 550659191Skris [Ulf M�ller] 550759191Skris 550859191Skris *) SPARC v8 assembler BIGNUM implementation. 550959191Skris [Andy Polyakov <appro@fy.chalmers.se>] 551059191Skris 551159191Skris *) Support for Certificate Policies extension: both print and set. 551259191Skris Various additions to support the r2i method this uses. 551359191Skris [Steve Henson] 551459191Skris 551559191Skris *) A lot of constification, and fix a bug in X509_NAME_oneline() that could 551659191Skris return a const string when you are expecting an allocated buffer. 551759191Skris [Ben Laurie] 551859191Skris 551959191Skris *) Add support for ASN1 types UTF8String and VISIBLESTRING, also the CHOICE 552059191Skris types DirectoryString and DisplayText. 552159191Skris [Steve Henson] 552259191Skris 552359191Skris *) Add code to allow r2i extensions to access the configuration database, 552459191Skris add an LHASH database driver and add several ctx helper functions. 552559191Skris [Steve Henson] 552659191Skris 552759191Skris *) Fix an evil bug in bn_expand2() which caused various BN functions to 552859191Skris fail when they extended the size of a BIGNUM. 552959191Skris [Steve Henson] 553059191Skris 553159191Skris *) Various utility functions to handle SXNet extension. Modify mkdef.pl to 553259191Skris support typesafe stack. 553359191Skris [Steve Henson] 553459191Skris 553559191Skris *) Fix typo in SSL_[gs]et_options(). 553659191Skris [Nils Frostberg <nils@medcom.se>] 553759191Skris 553859191Skris *) Delete various functions and files that belonged to the (now obsolete) 553959191Skris old X509V3 handling code. 554059191Skris [Steve Henson] 554159191Skris 554259191Skris *) New Configure option "rsaref". 554359191Skris [Ulf M�ller] 554459191Skris 554559191Skris *) Don't auto-generate pem.h. 554659191Skris [Bodo Moeller] 554759191Skris 554859191Skris *) Introduce type-safe ASN.1 SETs. 554959191Skris [Ben Laurie] 555059191Skris 555159191Skris *) Convert various additional casted stacks to type-safe STACK_OF() variants. 555259191Skris [Ben Laurie, Ralf S. Engelschall, Steve Henson] 555359191Skris 555459191Skris *) Introduce type-safe STACKs. This will almost certainly break lots of code 555559191Skris that links with OpenSSL (well at least cause lots of warnings), but fear 555659191Skris not: the conversion is trivial, and it eliminates loads of evil casts. A 555759191Skris few STACKed things have been converted already. Feel free to convert more. 555859191Skris In the fullness of time, I'll do away with the STACK type altogether. 555959191Skris [Ben Laurie] 556059191Skris 556159191Skris *) Add `openssl ca -revoke <certfile>' facility which revokes a certificate 556259191Skris specified in <certfile> by updating the entry in the index.txt file. 556359191Skris This way one no longer has to edit the index.txt file manually for 556459191Skris revoking a certificate. The -revoke option does the gory details now. 556559191Skris [Massimiliano Pala <madwolf@openca.org>, Ralf S. Engelschall] 556659191Skris 556759191Skris *) Fix `openssl crl -noout -text' combination where `-noout' killed the 556859191Skris `-text' option at all and this way the `-noout -text' combination was 556959191Skris inconsistent in `openssl crl' with the friends in `openssl x509|rsa|dsa'. 557059191Skris [Ralf S. Engelschall] 557159191Skris 557259191Skris *) Make sure a corresponding plain text error message exists for the 557359191Skris X509_V_ERR_CERT_REVOKED/23 error number which can occur when a 557459191Skris verify callback function determined that a certificate was revoked. 557559191Skris [Ralf S. Engelschall] 557659191Skris 557759191Skris *) Bugfix: In test/testenc, don't test "openssl <cipher>" for 557859191Skris ciphers that were excluded, e.g. by -DNO_IDEA. Also, test 557959191Skris all available cipers including rc5, which was forgotten until now. 558059191Skris In order to let the testing shell script know which algorithms 558159191Skris are available, a new (up to now undocumented) command 558259191Skris "openssl list-cipher-commands" is used. 558359191Skris [Bodo Moeller] 558459191Skris 558559191Skris *) Bugfix: s_client occasionally would sleep in select() when 558659191Skris it should have checked SSL_pending() first. 558759191Skris [Bodo Moeller] 558859191Skris 558959191Skris *) New functions DSA_do_sign and DSA_do_verify to provide access to 559059191Skris the raw DSA values prior to ASN.1 encoding. 559159191Skris [Ulf M�ller] 559259191Skris 559359191Skris *) Tweaks to Configure 559459191Skris [Niels Poppe <niels@netbox.org>] 559559191Skris 559659191Skris *) Add support for PKCS#5 v2.0 ASN1 PBES2 structures. No other support, 559759191Skris yet... 559859191Skris [Steve Henson] 559959191Skris 560059191Skris *) New variables $(RANLIB) and $(PERL) in the Makefiles. 560159191Skris [Ulf M�ller] 560259191Skris 560359191Skris *) New config option to avoid instructions that are illegal on the 80386. 560459191Skris The default code is faster, but requires at least a 486. 560559191Skris [Ulf M�ller] 560659191Skris 560759191Skris *) Got rid of old SSL2_CLIENT_VERSION (inconsistently used) and 560859191Skris SSL2_SERVER_VERSION (not used at all) macros, which are now the 560959191Skris same as SSL2_VERSION anyway. 561059191Skris [Bodo Moeller] 561159191Skris 561259191Skris *) New "-showcerts" option for s_client. 561359191Skris [Bodo Moeller] 561459191Skris 561559191Skris *) Still more PKCS#12 integration. Add pkcs12 application to openssl 561659191Skris application. Various cleanups and fixes. 561759191Skris [Steve Henson] 561859191Skris 561959191Skris *) More PKCS#12 integration. Add new pkcs12 directory with Makefile.ssl and 562059191Skris modify error routines to work internally. Add error codes and PBE init 562159191Skris to library startup routines. 562259191Skris [Steve Henson] 562359191Skris 562459191Skris *) Further PKCS#12 integration. Added password based encryption, PKCS#8 and 562559191Skris packing functions to asn1 and evp. Changed function names and error 562659191Skris codes along the way. 562759191Skris [Steve Henson] 562859191Skris 562959191Skris *) PKCS12 integration: and so it begins... First of several patches to 563059191Skris slowly integrate PKCS#12 functionality into OpenSSL. Add PKCS#12 563159191Skris objects to objects.h 563259191Skris [Steve Henson] 563359191Skris 563459191Skris *) Add a new 'indent' option to some X509V3 extension code. Initial ASN1 563559191Skris and display support for Thawte strong extranet extension. 563659191Skris [Steve Henson] 563759191Skris 563859191Skris *) Add LinuxPPC support. 563959191Skris [Jeff Dubrule <igor@pobox.org>] 564059191Skris 564159191Skris *) Get rid of redundant BN file bn_mulw.c, and rename bn_div64 to 564259191Skris bn_div_words in alpha.s. 564359191Skris [Hannes Reinecke <H.Reinecke@hw.ac.uk> and Ben Laurie] 564459191Skris 564559191Skris *) Make sure the RSA OAEP test is skipped under -DRSAref because 564659191Skris OAEP isn't supported when OpenSSL is built with RSAref. 564759191Skris [Ulf Moeller <ulf@fitug.de>] 564859191Skris 564959191Skris *) Move definitions of IS_SET/IS_SEQUENCE inside crypto/asn1/asn1.h 565059191Skris so they no longer are missing under -DNOPROTO. 565168651Skris [Soren S. Jorvang <soren@t.dk>] 565268651Skris 565368651Skris 565468651Skris Changes between 0.9.1c and 0.9.2b [22 Mar 1999] 565568651Skris 565659191Skris *) Make SSL_get_peer_cert_chain() work in servers. Unfortunately, it still 565759191Skris doesn't work when the session is reused. Coming soon! 565859191Skris [Ben Laurie] 565959191Skris 566059191Skris *) Fix a security hole, that allows sessions to be reused in the wrong 566159191Skris context thus bypassing client cert protection! All software that uses 566259191Skris client certs and session caches in multiple contexts NEEDS PATCHING to 566359191Skris allow session reuse! A fuller solution is in the works. 566459191Skris [Ben Laurie, problem pointed out by Holger Reif, Bodo Moeller (and ???)] 566559191Skris 566659191Skris *) Some more source tree cleanups (removed obsolete files 566759191Skris crypto/bf/asm/bf586.pl, test/test.txt and crypto/sha/asm/f.s; changed 566859191Skris permission on "config" script to be executable) and a fix for the INSTALL 566959191Skris document. 567059191Skris [Ulf Moeller <ulf@fitug.de>] 567159191Skris 567259191Skris *) Remove some legacy and erroneous uses of malloc, free instead of 567359191Skris Malloc, Free. 567459191Skris [Lennart Bang <lob@netstream.se>, with minor changes by Steve] 567559191Skris 567659191Skris *) Make rsa_oaep_test return non-zero on error. 567759191Skris [Ulf Moeller <ulf@fitug.de>] 567859191Skris 567959191Skris *) Add support for native Solaris shared libraries. Configure 568059191Skris solaris-sparc-sc4-pic, make, then run shlib/solaris-sc4.sh. It'd be nice 568159191Skris if someone would make that last step automatic. 568259191Skris [Matthias Loepfe <Matthias.Loepfe@AdNovum.CH>] 568359191Skris 568459191Skris *) ctx_size was not built with the right compiler during "make links". Fixed. 568559191Skris [Ben Laurie] 568659191Skris 568759191Skris *) Change the meaning of 'ALL' in the cipher list. It now means "everything 568859191Skris except NULL ciphers". This means the default cipher list will no longer 568959191Skris enable NULL ciphers. They need to be specifically enabled e.g. with 569059191Skris the string "DEFAULT:eNULL". 569159191Skris [Steve Henson] 569259191Skris 569359191Skris *) Fix to RSA private encryption routines: if p < q then it would 569459191Skris occasionally produce an invalid result. This will only happen with 569559191Skris externally generated keys because OpenSSL (and SSLeay) ensure p > q. 569659191Skris [Steve Henson] 569759191Skris 569859191Skris *) Be less restrictive and allow also `perl util/perlpath.pl 569959191Skris /path/to/bin/perl' in addition to `perl util/perlpath.pl /path/to/bin', 570059191Skris because this way one can also use an interpreter named `perl5' (which is 570159191Skris usually the name of Perl 5.xxx on platforms where an Perl 4.x is still 570259191Skris installed as `perl'). 570359191Skris [Matthias Loepfe <Matthias.Loepfe@adnovum.ch>] 570459191Skris 570559191Skris *) Let util/clean-depend.pl work also with older Perl 5.00x versions. 570659191Skris [Matthias Loepfe <Matthias.Loepfe@adnovum.ch>] 570759191Skris 570859191Skris *) Fix Makefile.org so CC,CFLAG etc are passed to 'make links' add 570959191Skris advapi32.lib to Win32 build and change the pem test comparision 571059191Skris to fc.exe (thanks to Ulrich Kroener <kroneru@yahoo.com> for the 571159191Skris suggestion). Fix misplaced ASNI prototypes and declarations in evp.h 571259191Skris and crypto/des/ede_cbcm_enc.c. 571359191Skris [Steve Henson] 571459191Skris 571559191Skris *) DES quad checksum was broken on big-endian architectures. Fixed. 571659191Skris [Ben Laurie] 571759191Skris 571859191Skris *) Comment out two functions in bio.h that aren't implemented. Fix up the 571959191Skris Win32 test batch file so it (might) work again. The Win32 test batch file 572059191Skris is horrible: I feel ill.... 572159191Skris [Steve Henson] 572259191Skris 572359191Skris *) Move various #ifdefs around so NO_SYSLOG, NO_DIRENT etc are now selected 572459191Skris in e_os.h. Audit of header files to check ANSI and non ANSI 572559191Skris sections: 10 functions were absent from non ANSI section and not exported 572659191Skris from Windows DLLs. Fixed up libeay.num for new functions. 572759191Skris [Steve Henson] 572859191Skris 572959191Skris *) Make `openssl version' output lines consistent. 573059191Skris [Ralf S. Engelschall] 573159191Skris 573259191Skris *) Fix Win32 symbol export lists for BIO functions: Added 573359191Skris BIO_get_ex_new_index, BIO_get_ex_num, BIO_get_ex_data and BIO_set_ex_data 573459191Skris to ms/libeay{16,32}.def. 573559191Skris [Ralf S. Engelschall] 573659191Skris 573759191Skris *) Second round of fixing the OpenSSL perl/ stuff. It now at least compiled 573859191Skris fine under Unix and passes some trivial tests I've now added. But the 573959191Skris whole stuff is horribly incomplete, so a README.1ST with a disclaimer was 574059191Skris added to make sure no one expects that this stuff really works in the 574159191Skris OpenSSL 0.9.2 release. Additionally I've started to clean the XS sources 574259191Skris up and fixed a few little bugs and inconsistencies in OpenSSL.{pm,xs} and 574359191Skris openssl_bio.xs. 574459191Skris [Ralf S. Engelschall] 574559191Skris 574659191Skris *) Fix the generation of two part addresses in perl. 574759191Skris [Kenji Miyake <kenji@miyake.org>, integrated by Ben Laurie] 574859191Skris 574959191Skris *) Add config entry for Linux on MIPS. 575059191Skris [John Tobey <jtobey@channel1.com>] 575159191Skris 575259191Skris *) Make links whenever Configure is run, unless we are on Windoze. 575359191Skris [Ben Laurie] 575459191Skris 575559191Skris *) Permit extensions to be added to CRLs using crl_section in openssl.cnf. 575659191Skris Currently only issuerAltName and AuthorityKeyIdentifier make any sense 575759191Skris in CRLs. 575859191Skris [Steve Henson] 575959191Skris 576059191Skris *) Add a useful kludge to allow package maintainers to specify compiler and 576159191Skris other platforms details on the command line without having to patch the 576259191Skris Configure script everytime: One now can use ``perl Configure 576359191Skris <id>:<details>'', i.e. platform ids are allowed to have details appended 576459191Skris to them (seperated by colons). This is treated as there would be a static 576559191Skris pre-configured entry in Configure's %table under key <id> with value 576659191Skris <details> and ``perl Configure <id>'' is called. So, when you want to 576759191Skris perform a quick test-compile under FreeBSD 3.1 with pgcc and without 576859191Skris assembler stuff you can use ``perl Configure "FreeBSD-elf:pgcc:-O6:::"'' 576959191Skris now, which overrides the FreeBSD-elf entry on-the-fly. 577059191Skris [Ralf S. Engelschall] 577159191Skris 577259191Skris *) Disable new TLS1 ciphersuites by default: they aren't official yet. 577359191Skris [Ben Laurie] 577459191Skris 577559191Skris *) Allow DSO flags like -fpic, -fPIC, -KPIC etc. to be specified 577659191Skris on the `perl Configure ...' command line. This way one can compile 577759191Skris OpenSSL libraries with Position Independent Code (PIC) which is needed 577859191Skris for linking it into DSOs. 577959191Skris [Ralf S. Engelschall] 578059191Skris 578159191Skris *) Remarkably, export ciphers were totally broken and no-one had noticed! 578259191Skris Fixed. 578359191Skris [Ben Laurie] 578459191Skris 578559191Skris *) Cleaned up the LICENSE document: The official contact for any license 578659191Skris questions now is the OpenSSL core team under openssl-core@openssl.org. 578759191Skris And add a paragraph about the dual-license situation to make sure people 578859191Skris recognize that _BOTH_ the OpenSSL license _AND_ the SSLeay license apply 578959191Skris to the OpenSSL toolkit. 579059191Skris [Ralf S. Engelschall] 579159191Skris 579259191Skris *) General source tree makefile cleanups: Made `making xxx in yyy...' 579359191Skris display consistent in the source tree and replaced `/bin/rm' by `rm'. 579459191Skris Additonally cleaned up the `make links' target: Remove unnecessary 579559191Skris semicolons, subsequent redundant removes, inline point.sh into mklink.sh 579659191Skris to speed processing and no longer clutter the display with confusing 579759191Skris stuff. Instead only the actually done links are displayed. 579859191Skris [Ralf S. Engelschall] 579959191Skris 580059191Skris *) Permit null encryption ciphersuites, used for authentication only. It used 580159191Skris to be necessary to set the preprocessor define SSL_ALLOW_ENULL to do this. 580259191Skris It is now necessary to set SSL_FORBID_ENULL to prevent the use of null 580359191Skris encryption. 580459191Skris [Ben Laurie] 580559191Skris 580659191Skris *) Add a bunch of fixes to the PKCS#7 stuff. It used to sometimes reorder 580759191Skris signed attributes when verifying signatures (this would break them), 580859191Skris the detached data encoding was wrong and public keys obtained using 580959191Skris X509_get_pubkey() weren't freed. 581059191Skris [Steve Henson] 581159191Skris 581259191Skris *) Add text documentation for the BUFFER functions. Also added a work around 581359191Skris to a Win95 console bug. This was triggered by the password read stuff: the 581459191Skris last character typed gets carried over to the next fread(). If you were 581559191Skris generating a new cert request using 'req' for example then the last 581659191Skris character of the passphrase would be CR which would then enter the first 581759191Skris field as blank. 581859191Skris [Steve Henson] 581959191Skris 582059191Skris *) Added the new `Includes OpenSSL Cryptography Software' button as 582159191Skris doc/openssl_button.{gif,html} which is similar in style to the old SSLeay 582259191Skris button and can be used by applications based on OpenSSL to show the 582359191Skris relationship to the OpenSSL project. 582459191Skris [Ralf S. Engelschall] 582559191Skris 582659191Skris *) Remove confusing variables in function signatures in files 582759191Skris ssl/ssl_lib.c and ssl/ssl.h. 582859191Skris [Lennart Bong <lob@kulthea.stacken.kth.se>] 582959191Skris 583059191Skris *) Don't install bss_file.c under PREFIX/include/ 583159191Skris [Lennart Bong <lob@kulthea.stacken.kth.se>] 583259191Skris 583359191Skris *) Get the Win32 compile working again. Modify mkdef.pl so it can handle 583459191Skris functions that return function pointers and has support for NT specific 583559191Skris stuff. Fix mk1mf.pl and VC-32.pl to support NT differences also. Various 583659191Skris #ifdef WIN32 and WINNTs sprinkled about the place and some changes from 583759191Skris unsigned to signed types: this was killing the Win32 compile. 583859191Skris [Steve Henson] 583959191Skris 584059191Skris *) Add new certificate file to stack functions, 584159191Skris SSL_add_dir_cert_subjects_to_stack() and 584259191Skris SSL_add_file_cert_subjects_to_stack(). These largely supplant 584359191Skris SSL_load_client_CA_file(), and can be used to add multiple certs easily 584459191Skris to a stack (usually this is then handed to SSL_CTX_set_client_CA_list()). 584559191Skris This means that Apache-SSL and similar packages don't have to mess around 584659191Skris to add as many CAs as they want to the preferred list. 584759191Skris [Ben Laurie] 584859191Skris 584959191Skris *) Experiment with doxygen documentation. Currently only partially applied to 585059191Skris ssl/ssl_lib.c. 585159191Skris See http://www.stack.nl/~dimitri/doxygen/index.html, and run doxygen with 585259191Skris openssl.doxy as the configuration file. 585359191Skris [Ben Laurie] 585459191Skris 585559191Skris *) Get rid of remaining C++-style comments which strict C compilers hate. 585659191Skris [Ralf S. Engelschall, pointed out by Carlos Amengual] 585759191Skris 585859191Skris *) Changed BN_RECURSION in bn_mont.c to BN_RECURSION_MONT so it is not 585959191Skris compiled in by default: it has problems with large keys. 586059191Skris [Steve Henson] 586159191Skris 586259191Skris *) Add a bunch of SSL_xxx() functions for configuring the temporary RSA and 586359191Skris DH private keys and/or callback functions which directly correspond to 586459191Skris their SSL_CTX_xxx() counterparts but work on a per-connection basis. This 586559191Skris is needed for applications which have to configure certificates on a 586659191Skris per-connection basis (e.g. Apache+mod_ssl) instead of a per-context basis 586759191Skris (e.g. s_server). 586859191Skris For the RSA certificate situation is makes no difference, but 586959191Skris for the DSA certificate situation this fixes the "no shared cipher" 587059191Skris problem where the OpenSSL cipher selection procedure failed because the 587159191Skris temporary keys were not overtaken from the context and the API provided 587259191Skris no way to reconfigure them. 587359191Skris The new functions now let applications reconfigure the stuff and they 587459191Skris are in detail: SSL_need_tmp_RSA, SSL_set_tmp_rsa, SSL_set_tmp_dh, 587559191Skris SSL_set_tmp_rsa_callback and SSL_set_tmp_dh_callback. Additionally a new 587659191Skris non-public-API function ssl_cert_instantiate() is used as a helper 587759191Skris function and also to reduce code redundancy inside ssl_rsa.c. 587859191Skris [Ralf S. Engelschall] 587959191Skris 588059191Skris *) Move s_server -dcert and -dkey options out of the undocumented feature 588159191Skris area because they are useful for the DSA situation and should be 588259191Skris recognized by the users. 588359191Skris [Ralf S. Engelschall] 588459191Skris 588559191Skris *) Fix the cipher decision scheme for export ciphers: the export bits are 588659191Skris *not* within SSL_MKEY_MASK or SSL_AUTH_MASK, they are within 588759191Skris SSL_EXP_MASK. So, the original variable has to be used instead of the 588859191Skris already masked variable. 588959191Skris [Richard Levitte <levitte@stacken.kth.se>] 589059191Skris 589159191Skris *) Fix 'port' variable from `int' to `unsigned int' in crypto/bio/b_sock.c 589259191Skris [Richard Levitte <levitte@stacken.kth.se>] 589359191Skris 589459191Skris *) Change type of another md_len variable in pk7_doit.c:PKCS7_dataFinal() 589559191Skris from `int' to `unsigned int' because it's a length and initialized by 589659191Skris EVP_DigestFinal() which expects an `unsigned int *'. 589759191Skris [Richard Levitte <levitte@stacken.kth.se>] 589859191Skris 589959191Skris *) Don't hard-code path to Perl interpreter on shebang line of Configure 590059191Skris script. Instead use the usual Shell->Perl transition trick. 590159191Skris [Ralf S. Engelschall] 590259191Skris 590359191Skris *) Make `openssl x509 -noout -modulus' functional also for DSA certificates 590459191Skris (in addition to RSA certificates) to match the behaviour of `openssl dsa 590559191Skris -noout -modulus' as it's already the case for `openssl rsa -noout 590659191Skris -modulus'. For RSA the -modulus is the real "modulus" while for DSA 590759191Skris currently the public key is printed (a decision which was already done by 590859191Skris `openssl dsa -modulus' in the past) which serves a similar purpose. 590959191Skris Additionally the NO_RSA no longer completely removes the whole -modulus 591059191Skris option; it now only avoids using the RSA stuff. Same applies to NO_DSA 591159191Skris now, too. 591259191Skris [Ralf S. Engelschall] 591359191Skris 591459191Skris *) Add Arne Ansper's reliable BIO - this is an encrypted, block-digested 591559191Skris BIO. See the source (crypto/evp/bio_ok.c) for more info. 591659191Skris [Arne Ansper <arne@ats.cyber.ee>] 591759191Skris 591859191Skris *) Dump the old yucky req code that tried (and failed) to allow raw OIDs 591959191Skris to be added. Now both 'req' and 'ca' can use new objects defined in the 592059191Skris config file. 592159191Skris [Steve Henson] 592259191Skris 592359191Skris *) Add cool BIO that does syslog (or event log on NT). 592459191Skris [Arne Ansper <arne@ats.cyber.ee>, integrated by Ben Laurie] 592559191Skris 592659191Skris *) Add support for new TLS ciphersuites, TLS_RSA_EXPORT56_WITH_RC4_56_MD5, 592759191Skris TLS_RSA_EXPORT56_WITH_RC2_CBC_56_MD5 and 592859191Skris TLS_RSA_EXPORT56_WITH_DES_CBC_SHA, as specified in "56-bit Export Cipher 592959191Skris Suites For TLS", draft-ietf-tls-56-bit-ciphersuites-00.txt. 593059191Skris [Ben Laurie] 593159191Skris 593259191Skris *) Add preliminary config info for new extension code. 593359191Skris [Steve Henson] 593459191Skris 593559191Skris *) Make RSA_NO_PADDING really use no padding. 593659191Skris [Ulf Moeller <ulf@fitug.de>] 593759191Skris 593859191Skris *) Generate errors when private/public key check is done. 593959191Skris [Ben Laurie] 594059191Skris 594159191Skris *) Overhaul for 'crl' utility. New function X509_CRL_print. Partial support 594259191Skris for some CRL extensions and new objects added. 594359191Skris [Steve Henson] 594459191Skris 594559191Skris *) Really fix the ASN1 IMPLICIT bug this time... Partial support for private 594659191Skris key usage extension and fuller support for authority key id. 594759191Skris [Steve Henson] 594859191Skris 594959191Skris *) Add OAEP encryption for the OpenSSL crypto library. OAEP is the improved 595059191Skris padding method for RSA, which is recommended for new applications in PKCS 595159191Skris #1 v2.0 (RFC 2437, October 1998). 595259191Skris OAEP (Optimal Asymmetric Encryption Padding) has better theoretical 595359191Skris foundations than the ad-hoc padding used in PKCS #1 v1.5. It is secure 595459191Skris against Bleichbacher's attack on RSA. 595559191Skris [Ulf Moeller <ulf@fitug.de>, reformatted, corrected and integrated by 595659191Skris Ben Laurie] 595759191Skris 595859191Skris *) Updates to the new SSL compression code 595959191Skris [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)] 596059191Skris 596159191Skris *) Fix so that the version number in the master secret, when passed 596259191Skris via RSA, checks that if TLS was proposed, but we roll back to SSLv3 596359191Skris (because the server will not accept higher), that the version number 596459191Skris is 0x03,0x01, not 0x03,0x00 596559191Skris [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)] 596659191Skris 596759191Skris *) Run extensive memory leak checks on SSL apps. Fixed *lots* of memory 596859191Skris leaks in ssl/ relating to new X509_get_pubkey() behaviour. Also fixes 596959191Skris in apps/ and an unrelated leak in crypto/dsa/dsa_vrf.c 597059191Skris [Steve Henson] 597159191Skris 597259191Skris *) Support for RAW extensions where an arbitrary extension can be 597359191Skris created by including its DER encoding. See apps/openssl.cnf for 597459191Skris an example. 597559191Skris [Steve Henson] 597659191Skris 597759191Skris *) Make sure latest Perl versions don't interpret some generated C array 597859191Skris code as Perl array code in the crypto/err/err_genc.pl script. 597959191Skris [Lars Weber <3weber@informatik.uni-hamburg.de>] 598059191Skris 598159191Skris *) Modify ms/do_ms.bat to not generate assembly language makefiles since 598259191Skris not many people have the assembler. Various Win32 compilation fixes and 598359191Skris update to the INSTALL.W32 file with (hopefully) more accurate Win32 598459191Skris build instructions. 598559191Skris [Steve Henson] 598659191Skris 598759191Skris *) Modify configure script 'Configure' to automatically create crypto/date.h 598859191Skris file under Win32 and also build pem.h from pem.org. New script 598959191Skris util/mkfiles.pl to create the MINFO file on environments that can't do a 599059191Skris 'make files': perl util/mkfiles.pl >MINFO should work. 599159191Skris [Steve Henson] 599259191Skris 599359191Skris *) Major rework of DES function declarations, in the pursuit of correctness 599459191Skris and purity. As a result, many evil casts evaporated, and some weirdness, 599559191Skris too. You may find this causes warnings in your code. Zapping your evil 599659191Skris casts will probably fix them. Mostly. 599759191Skris [Ben Laurie] 599859191Skris 599959191Skris *) Fix for a typo in asn1.h. Bug fix to object creation script 600059191Skris obj_dat.pl. It considered a zero in an object definition to mean 600159191Skris "end of object": none of the objects in objects.h have any zeros 600259191Skris so it wasn't spotted. 600359191Skris [Steve Henson, reported by Erwann ABALEA <eabalea@certplus.com>] 600459191Skris 600559191Skris *) Add support for Triple DES Cipher Block Chaining with Output Feedback 600659191Skris Masking (CBCM). In the absence of test vectors, the best I have been able 600759191Skris to do is check that the decrypt undoes the encrypt, so far. Send me test 600859191Skris vectors if you have them. 600959191Skris [Ben Laurie] 601059191Skris 601159191Skris *) Correct calculation of key length for export ciphers (too much space was 601259191Skris allocated for null ciphers). This has not been tested! 601359191Skris [Ben Laurie] 601459191Skris 601559191Skris *) Modifications to the mkdef.pl for Win32 DEF file creation. The usage 601659191Skris message is now correct (it understands "crypto" and "ssl" on its 601759191Skris command line). There is also now an "update" option. This will update 601859191Skris the util/ssleay.num and util/libeay.num files with any new functions. 601959191Skris If you do a: 602059191Skris perl util/mkdef.pl crypto ssl update 602159191Skris it will update them. 602259191Skris [Steve Henson] 602359191Skris 602459191Skris *) Overhauled the Perl interface (perl/*): 602559191Skris - ported BN stuff to OpenSSL's different BN library 602659191Skris - made the perl/ source tree CVS-aware 602759191Skris - renamed the package from SSLeay to OpenSSL (the files still contain 602859191Skris their history because I've copied them in the repository) 602959191Skris - removed obsolete files (the test scripts will be replaced 603059191Skris by better Test::Harness variants in the future) 603159191Skris [Ralf S. Engelschall] 603259191Skris 603359191Skris *) First cut for a very conservative source tree cleanup: 603459191Skris 1. merge various obsolete readme texts into doc/ssleay.txt 603559191Skris where we collect the old documents and readme texts. 603659191Skris 2. remove the first part of files where I'm already sure that we no 603759191Skris longer need them because of three reasons: either they are just temporary 603859191Skris files which were left by Eric or they are preserved original files where 603959191Skris I've verified that the diff is also available in the CVS via "cvs diff 604059191Skris -rSSLeay_0_8_1b" or they were renamed (as it was definitely the case for 604159191Skris the crypto/md/ stuff). 604259191Skris [Ralf S. Engelschall] 604359191Skris 604459191Skris *) More extension code. Incomplete support for subject and issuer alt 604559191Skris name, issuer and authority key id. Change the i2v function parameters 604659191Skris and add an extra 'crl' parameter in the X509V3_CTX structure: guess 604759191Skris what that's for :-) Fix to ASN1 macro which messed up 604859191Skris IMPLICIT tag and add f_enum.c which adds a2i, i2a for ENUMERATED. 604959191Skris [Steve Henson] 605059191Skris 605159191Skris *) Preliminary support for ENUMERATED type. This is largely copied from the 605259191Skris INTEGER code. 605359191Skris [Steve Henson] 605459191Skris 605559191Skris *) Add new function, EVP_MD_CTX_copy() to replace frequent use of memcpy. 605659191Skris [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)] 605759191Skris 605859191Skris *) Make sure `make rehash' target really finds the `openssl' program. 605959191Skris [Ralf S. Engelschall, Matthias Loepfe <Matthias.Loepfe@adnovum.ch>] 606059191Skris 606159191Skris *) Squeeze another 7% of speed out of MD5 assembler, at least on a P2. I'd 606259191Skris like to hear about it if this slows down other processors. 606359191Skris [Ben Laurie] 606459191Skris 606559191Skris *) Add CygWin32 platform information to Configure script. 606659191Skris [Alan Batie <batie@aahz.jf.intel.com>] 606759191Skris 606859191Skris *) Fixed ms/32all.bat script: `no_asm' -> `no-asm' 606959191Skris [Rainer W. Gerling <gerling@mpg-gv.mpg.de>] 607059191Skris 607159191Skris *) New program nseq to manipulate netscape certificate sequences 607259191Skris [Steve Henson] 607359191Skris 607459191Skris *) Modify crl2pkcs7 so it supports multiple -certfile arguments. Fix a 607559191Skris few typos. 607659191Skris [Steve Henson] 607759191Skris 607859191Skris *) Fixes to BN code. Previously the default was to define BN_RECURSION 607959191Skris but the BN code had some problems that would cause failures when 608059191Skris doing certificate verification and some other functions. 608159191Skris [Eric A. Young, (from changes to C2Net SSLeay, integrated by Mark Cox)] 608259191Skris 608359191Skris *) Add ASN1 and PEM code to support netscape certificate sequences. 608459191Skris [Steve Henson] 608559191Skris 608659191Skris *) Add ASN1 and PEM code to support netscape certificate sequences. 608759191Skris [Steve Henson] 608859191Skris 608959191Skris *) Add several PKIX and private extended key usage OIDs. 609059191Skris [Steve Henson] 609159191Skris 609259191Skris *) Modify the 'ca' program to handle the new extension code. Modify 609359191Skris openssl.cnf for new extension format, add comments. 609459191Skris [Steve Henson] 609559191Skris 609659191Skris *) More X509 V3 changes. Fix typo in v3_bitstr.c. Add support to 'req' 609759191Skris and add a sample to openssl.cnf so req -x509 now adds appropriate 609859191Skris CA extensions. 609959191Skris [Steve Henson] 610059191Skris 610159191Skris *) Continued X509 V3 changes. Add to other makefiles, integrate with the 610259191Skris error code, add initial support to X509_print() and x509 application. 610359191Skris [Steve Henson] 610459191Skris 610559191Skris *) Takes a deep breath and start addding X509 V3 extension support code. Add 610659191Skris files in crypto/x509v3. Move original stuff to crypto/x509v3/old. All this 610759191Skris stuff is currently isolated and isn't even compiled yet. 610859191Skris [Steve Henson] 610959191Skris 611059191Skris *) Continuing patches for GeneralizedTime. Fix up certificate and CRL 611159191Skris ASN1 to use ASN1_TIME and modify print routines to use ASN1_TIME_print. 611259191Skris Removed the versions check from X509 routines when loading extensions: 611359191Skris this allows certain broken certificates that don't set the version 611459191Skris properly to be processed. 611559191Skris [Steve Henson] 611659191Skris 611759191Skris *) Deal with irritating shit to do with dependencies, in YAAHW (Yet Another 611859191Skris Ad Hoc Way) - Makefile.ssls now all contain local dependencies, which 611959191Skris can still be regenerated with "make depend". 612059191Skris [Ben Laurie] 612159191Skris 612259191Skris *) Spelling mistake in C version of CAST-128. 612359191Skris [Ben Laurie, reported by Jeremy Hylton <jeremy@cnri.reston.va.us>] 612459191Skris 612559191Skris *) Changes to the error generation code. The perl script err-code.pl 612659191Skris now reads in the old error codes and retains the old numbers, only 612759191Skris adding new ones if necessary. It also only changes the .err files if new 612859191Skris codes are added. The makefiles have been modified to only insert errors 612959191Skris when needed (to avoid needlessly modifying header files). This is done 613059191Skris by only inserting errors if the .err file is newer than the auto generated 613159191Skris C file. To rebuild all the error codes from scratch (the old behaviour) 613259191Skris either modify crypto/Makefile.ssl to pass the -regen flag to err_code.pl 613359191Skris or delete all the .err files. 613459191Skris [Steve Henson] 613559191Skris 613659191Skris *) CAST-128 was incorrectly implemented for short keys. The C version has 613759191Skris been fixed, but is untested. The assembler versions are also fixed, but 613859191Skris new assembler HAS NOT BEEN GENERATED FOR WIN32 - the Makefile needs fixing 613959191Skris to regenerate it if needed. 614059191Skris [Ben Laurie, reported (with fix for C version) by Jun-ichiro itojun 614159191Skris Hagino <itojun@kame.net>] 614259191Skris 614359191Skris *) File was opened incorrectly in randfile.c. 614459191Skris [Ulf M�ller <ulf@fitug.de>] 614559191Skris 614659191Skris *) Beginning of support for GeneralizedTime. d2i, i2d, check and print 614759191Skris functions. Also ASN1_TIME suite which is a CHOICE of UTCTime or 614859191Skris GeneralizedTime. ASN1_TIME is the proper type used in certificates et 614959191Skris al: it's just almost always a UTCTime. Note this patch adds new error 615059191Skris codes so do a "make errors" if there are problems. 615159191Skris [Steve Henson] 615259191Skris 615359191Skris *) Correct Linux 1 recognition in config. 615459191Skris [Ulf M�ller <ulf@fitug.de>] 615559191Skris 615659191Skris *) Remove pointless MD5 hash when using DSA keys in ca. 615759191Skris [Anonymous <nobody@replay.com>] 615859191Skris 615959191Skris *) Generate an error if given an empty string as a cert directory. Also 616059191Skris generate an error if handed NULL (previously returned 0 to indicate an 616159191Skris error, but didn't set one). 616259191Skris [Ben Laurie, reported by Anonymous <nobody@replay.com>] 616359191Skris 616459191Skris *) Add prototypes to SSL methods. Make SSL_write's buffer const, at last. 616559191Skris [Ben Laurie] 616659191Skris 616759191Skris *) Fix the dummy function BN_ref_mod_exp() in rsaref.c to have the correct 616859191Skris parameters. This was causing a warning which killed off the Win32 compile. 616959191Skris [Steve Henson] 617059191Skris 617159191Skris *) Remove C++ style comments from crypto/bn/bn_local.h. 617259191Skris [Neil Costigan <neil.costigan@celocom.com>] 617359191Skris 617459191Skris *) The function OBJ_txt2nid was broken. It was supposed to return a nid 617559191Skris based on a text string, looking up short and long names and finally 617659191Skris "dot" format. The "dot" format stuff didn't work. Added new function 617759191Skris OBJ_txt2obj to do the same but return an ASN1_OBJECT and rewrote 617859191Skris OBJ_txt2nid to use it. OBJ_txt2obj can also return objects even if the 617959191Skris OID is not part of the table. 618059191Skris [Steve Henson] 618159191Skris 618259191Skris *) Add prototypes to X509 lookup/verify methods, fixing a bug in 618359191Skris X509_LOOKUP_by_alias(). 618459191Skris [Ben Laurie] 618559191Skris 618659191Skris *) Sort openssl functions by name. 618759191Skris [Ben Laurie] 618859191Skris 618959191Skris *) Get the gendsa program working (hopefully) and add it to app list. Remove 619059191Skris encryption from sample DSA keys (in case anyone is interested the password 619159191Skris was "1234"). 619259191Skris [Steve Henson] 619359191Skris 619459191Skris *) Make _all_ *_free functions accept a NULL pointer. 619559191Skris [Frans Heymans <fheymans@isaserver.be>] 619659191Skris 619759191Skris *) If a DH key is generated in s3_srvr.c, don't blow it by trying to use 619859191Skris NULL pointers. 619959191Skris [Anonymous <nobody@replay.com>] 620059191Skris 620159191Skris *) s_server should send the CAfile as acceptable CAs, not its own cert. 620259191Skris [Bodo Moeller <3moeller@informatik.uni-hamburg.de>] 620359191Skris 620459191Skris *) Don't blow it for numeric -newkey arguments to apps/req. 620559191Skris [Bodo Moeller <3moeller@informatik.uni-hamburg.de>] 620659191Skris 620759191Skris *) Temp key "for export" tests were wrong in s3_srvr.c. 620859191Skris [Anonymous <nobody@replay.com>] 620959191Skris 621059191Skris *) Add prototype for temp key callback functions 621159191Skris SSL_CTX_set_tmp_{rsa,dh}_callback(). 621259191Skris [Ben Laurie] 621359191Skris 621459191Skris *) Make DH_free() tolerate being passed a NULL pointer (like RSA_free() and 621559191Skris DSA_free()). Make X509_PUBKEY_set() check for errors in d2i_PublicKey(). 621659191Skris [Steve Henson] 621759191Skris 621859191Skris *) X509_name_add_entry() freed the wrong thing after an error. 621959191Skris [Arne Ansper <arne@ats.cyber.ee>] 622059191Skris 622159191Skris *) rsa_eay.c would attempt to free a NULL context. 622259191Skris [Arne Ansper <arne@ats.cyber.ee>] 622359191Skris 622459191Skris *) BIO_s_socket() had a broken should_retry() on Windoze. 622559191Skris [Arne Ansper <arne@ats.cyber.ee>] 622659191Skris 622759191Skris *) BIO_f_buffer() didn't pass on BIO_CTRL_FLUSH. 622859191Skris [Arne Ansper <arne@ats.cyber.ee>] 622959191Skris 623059191Skris *) Make sure the already existing X509_STORE->depth variable is initialized 623159191Skris in X509_STORE_new(), but document the fact that this variable is still 623259191Skris unused in the certificate verification process. 623359191Skris [Ralf S. Engelschall] 623459191Skris 623559191Skris *) Fix the various library and apps files to free up pkeys obtained from 623659191Skris X509_PUBKEY_get() et al. Also allow x509.c to handle netscape extensions. 623759191Skris [Steve Henson] 623859191Skris 623959191Skris *) Fix reference counting in X509_PUBKEY_get(). This makes 624059191Skris demos/maurice/example2.c work, amongst others, probably. 624155714Skris [Steve Henson and Ben Laurie] 624255714Skris 624355714Skris *) First cut of a cleanup for apps/. First the `ssleay' program is now named 624455714Skris `openssl' and second, the shortcut symlinks for the `openssl <command>' 624555714Skris are no longer created. This way we have a single and consistent command 624655714Skris line interface `openssl <command>', similar to `cvs <command>'. 624755714Skris [Ralf S. Engelschall, Paul Sutton and Ben Laurie] 624855714Skris 624955714Skris *) ca.c: move test for DSA keys inside #ifndef NO_DSA. Make pubkey 625055714Skris BIT STRING wrapper always have zero unused bits. 625155714Skris [Steve Henson] 625255714Skris 625355714Skris *) Add CA.pl, perl version of CA.sh, add extended key usage OID. 625455714Skris [Steve Henson] 625555714Skris 625655714Skris *) Make the top-level INSTALL documentation easier to understand. 625755714Skris [Paul Sutton] 625855714Skris 625955714Skris *) Makefiles updated to exit if an error occurs in a sub-directory 626055714Skris make (including if user presses ^C) [Paul Sutton] 626155714Skris 626255714Skris *) Make Montgomery context stuff explicit in RSA data structure. 626355714Skris [Ben Laurie] 626455714Skris 626555714Skris *) Fix build order of pem and err to allow for generated pem.h. 626655714Skris [Ben Laurie] 626755714Skris 626855714Skris *) Fix renumbering bug in X509_NAME_delete_entry(). 626955714Skris [Ben Laurie] 627055714Skris 627155714Skris *) Enhanced the err-ins.pl script so it makes the error library number 627255714Skris global and can add a library name. This is needed for external ASN1 and 627355714Skris other error libraries. 627455714Skris [Steve Henson] 627555714Skris 627655714Skris *) Fixed sk_insert which never worked properly. 627755714Skris [Steve Henson] 627855714Skris 627955714Skris *) Fix ASN1 macros so they can handle indefinite length construted 628055714Skris EXPLICIT tags. Some non standard certificates use these: they can now 628155714Skris be read in. 628255714Skris [Steve Henson] 628355714Skris 628455714Skris *) Merged the various old/obsolete SSLeay documentation files (doc/xxx.doc) 628555714Skris into a single doc/ssleay.txt bundle. This way the information is still 628655714Skris preserved but no longer messes up this directory. Now it's new room for 628755714Skris the new set of documenation files. 628859191Skris [Ralf S. Engelschall] 628959191Skris 629059191Skris *) SETs were incorrectly DER encoded. This was a major pain, because they 629155714Skris shared code with SEQUENCEs, which aren't coded the same. This means that 629255714Skris almost everything to do with SETs or SEQUENCEs has either changed name or 629355714Skris number of arguments. 629455714Skris [Ben Laurie, based on a partial fix by GP Jayan <gp@nsj.co.jp>] 629555714Skris 629655714Skris *) Fix test data to work with the above. 629755714Skris [Ben Laurie] 629855714Skris 629955714Skris *) Fix the RSA header declarations that hid a bug I fixed in 0.9.0b but 630055714Skris was already fixed by Eric for 0.9.1 it seems. 630155714Skris [Ben Laurie - pointed out by Ulf M�ller <ulf@fitug.de>] 630255714Skris 630355714Skris *) Autodetect FreeBSD3. 630459191Skris [Ben Laurie] 630555714Skris 630655714Skris *) Fix various bugs in Configure. This affects the following platforms: 630755714Skris nextstep 630855714Skris ncr-scde 630955714Skris unixware-2.0 631055714Skris unixware-2.0-pentium 631155714Skris sco5-cc. 631255714Skris [Ben Laurie] 631355714Skris 631455714Skris *) Eliminate generated files from CVS. Reorder tests to regenerate files 631555714Skris before they are needed. 631655714Skris [Ben Laurie] 631755714Skris 631855714Skris *) Generate Makefile.ssl from Makefile.org (to keep CVS happy). 631955714Skris [Ben Laurie] 632055714Skris 632155714Skris 632255714Skris Changes between 0.9.1b and 0.9.1c [23-Dec-1998] 632355714Skris 632455714Skris *) Added OPENSSL_VERSION_NUMBER to crypto/crypto.h and 632555714Skris changed SSLeay to OpenSSL in version strings. 632655714Skris [Ralf S. Engelschall] 632755714Skris 632855714Skris *) Some fixups to the top-level documents. 632955714Skris [Paul Sutton] 633055714Skris 633155714Skris *) Fixed the nasty bug where rsaref.h was not found under compile-time 633255714Skris because the symlink to include/ was missing. 633355714Skris [Ralf S. Engelschall] 633455714Skris 633555714Skris *) Incorporated the popular no-RSA/DSA-only patches 633655714Skris which allow to compile a RSA-free SSLeay. 633755714Skris [Andrew Cooke / Interrader Ldt., Ralf S. Engelschall] 633855714Skris 633955714Skris *) Fixed nasty rehash problem under `make -f Makefile.ssl links' 634055714Skris when "ssleay" is still not found. 634155714Skris [Ralf S. Engelschall] 634255714Skris 634355714Skris *) Added more platforms to Configure: Cray T3E, HPUX 11, 634455714Skris [Ralf S. Engelschall, Beckmann <beckman@acl.lanl.gov>] 634555714Skris 634655714Skris *) Updated the README file. 634755714Skris [Ralf S. Engelschall] 634855714Skris 634955714Skris *) Added various .cvsignore files in the CVS repository subdirs 635055714Skris to make a "cvs update" really silent. 635155714Skris [Ralf S. Engelschall] 635255714Skris 635355714Skris *) Recompiled the error-definition header files and added 635455714Skris missing symbols to the Win32 linker tables. 635555714Skris [Ralf S. Engelschall] 635655714Skris 635755714Skris *) Cleaned up the top-level documents; 635855714Skris o new files: CHANGES and LICENSE 635955714Skris o merged VERSION, HISTORY* and README* files a CHANGES.SSLeay 636055714Skris o merged COPYRIGHT into LICENSE 636155714Skris o removed obsolete TODO file 636255714Skris o renamed MICROSOFT to INSTALL.W32 636355714Skris [Ralf S. Engelschall] 636455714Skris 636555714Skris *) Removed dummy files from the 0.9.1b source tree: 636655714Skris crypto/asn1/x crypto/bio/cd crypto/bio/fg crypto/bio/grep crypto/bio/vi 636755714Skris crypto/bn/asm/......add.c crypto/bn/asm/a.out crypto/dsa/f crypto/md5/f 636855714Skris crypto/pem/gmon.out crypto/perlasm/f crypto/pkcs7/build crypto/rsa/f 636955714Skris crypto/sha/asm/f crypto/threads/f ms/zzz ssl/f ssl/f.mak test/f 637055714Skris util/f.mak util/pl/f util/pl/f.mak crypto/bf/bf_locl.old apps/f 637155714Skris [Ralf S. Engelschall] 637255714Skris 637355714Skris *) Added various platform portability fixes. 637455714Skris [Mark J. Cox] 637555714Skris 637655714Skris *) The Genesis of the OpenSSL rpject: 637755714Skris We start with the latest (unreleased) SSLeay version 0.9.1b which Eric A. 637855714Skris Young and Tim J. Hudson created while they were working for C2Net until 637955714Skris summer 1998. 638055714Skris [The OpenSSL Project] 638155714Skris 638255714Skris 638355714Skris Changes between 0.9.0b and 0.9.1b [not released] 638455714Skris 638555714Skris *) Updated a few CA certificates under certs/ 638655714Skris [Eric A. Young] 638755714Skris 638855714Skris *) Changed some BIGNUM api stuff. 638955714Skris [Eric A. Young] 639055714Skris 639155714Skris *) Various platform ports: OpenBSD, Ultrix, IRIX 64bit, NetBSD, 639255714Skris DGUX x86, Linux Alpha, etc. 639355714Skris [Eric A. Young] 639455714Skris 639555714Skris *) New COMP library [crypto/comp/] for SSL Record Layer Compression: 639655714Skris RLE (dummy implemented) and ZLIB (really implemented when ZLIB is 639755714Skris available). 639855714Skris [Eric A. Young] 639955714Skris 640055714Skris *) Add -strparse option to asn1pars program which parses nested 640155714Skris binary structures 640255714Skris [Dr Stephen Henson <shenson@bigfoot.com>] 640355714Skris 640455714Skris *) Added "oid_file" to ssleay.cnf for "ca" and "req" programs. 640555714Skris [Eric A. Young] 640655714Skris 640755714Skris *) DSA fix for "ca" program. 640855714Skris [Eric A. Young] 640955714Skris 641055714Skris *) Added "-genkey" option to "dsaparam" program. 641155714Skris [Eric A. Young] 641255714Skris 641355714Skris *) Added RIPE MD160 (rmd160) message digest. 641479998Skris [Eric A. Young] 641555714Skris 641655714Skris *) Added -a (all) option to "ssleay version" command. 641755714Skris [Eric A. Young] 641855714Skris 641955714Skris *) Added PLATFORM define which is the id given to Configure. 642055714Skris [Eric A. Young] 642155714Skris 642255714Skris *) Added MemCheck_XXXX functions to crypto/mem.c for memory checking. 642355714Skris [Eric A. Young] 642455714Skris 642555714Skris *) Extended the ASN.1 parser routines. 642655714Skris [Eric A. Young] 642755714Skris 642855714Skris *) Extended BIO routines to support REUSEADDR, seek, tell, etc. 642955714Skris [Eric A. Young] 643055714Skris 643155714Skris *) Added a BN_CTX to the BN library. 643255714Skris [Eric A. Young] 643355714Skris 643455714Skris *) Fixed the weak key values in DES library 643555714Skris [Eric A. Young] 643655714Skris 643755714Skris *) Changed API in EVP library for cipher aliases. 643855714Skris [Eric A. Young] 643955714Skris 644055714Skris *) Added support for RC2/64bit cipher. 644155714Skris [Eric A. Young] 644255714Skris 644355714Skris *) Converted the lhash library to the crypto/mem.c functions. 644455714Skris [Eric A. Young] 644555714Skris 644655714Skris *) Added more recognized ASN.1 object ids. 644755714Skris [Eric A. Young] 644855714Skris 644955714Skris *) Added more RSA padding checks for SSL/TLS. 645055714Skris [Eric A. Young] 645155714Skris 645255714Skris *) Added BIO proxy/filter functionality. 645355714Skris [Eric A. Young] 645455714Skris 645555714Skris *) Added extra_certs to SSL_CTX which can be used 645655714Skris send extra CA certificates to the client in the CA cert chain sending 645755714Skris process. It can be configured with SSL_CTX_add_extra_chain_cert(). 645855714Skris [Eric A. Young] 645955714Skris 646055714Skris *) Now Fortezza is denied in the authentication phase because 646155714Skris this is key exchange mechanism is not supported by SSLeay at all. 646255714Skris [Eric A. Young] 646355714Skris 646455714Skris *) Additional PKCS1 checks. 646555714Skris [Eric A. Young] 646655714Skris 646755714Skris *) Support the string "TLSv1" for all TLS v1 ciphers. 646855714Skris [Eric A. Young] 646955714Skris 647055714Skris *) Added function SSL_get_ex_data_X509_STORE_CTX_idx() which gives the 647155714Skris ex_data index of the SSL context in the X509_STORE_CTX ex_data. 647255714Skris [Eric A. Young] 647355714Skris 647455714Skris *) Fixed a few memory leaks. 647555714Skris [Eric A. Young] 647655714Skris 647755714Skris *) Fixed various code and comment typos. 647855714Skris [Eric A. Young] 647955714Skris 648055714Skris *) A minor bug in ssl/s3_clnt.c where there would always be 4 0 648155714Skris bytes sent in the client random. 648255714Skris [Edward Bishop <ebishop@spyglass.com>] 648355714Skris 648455714Skris