Back to home page

openssl Cross Reference



0001 This file contains the changes for the SSLeay library up to version
0002 0.9.0b. For later changes, see the file "CHANGES".
0004   SSLeay CHANGES
0005   ______________
0007 Changes between 0.8.x and 0.9.0b
0009 10-Apr-1998
0011 I said the next version would go out at easter, and so it shall.
0012 I expect a 0.9.1 will follow with portability fixes in the next few weeks.
0014 This is a quick, meet the deadline.  Look to ssl-users for comments on what
0015 is new etc.
0017 eric (about to go bushwalking for the 4 day easter break :-)
0019 16-Mar-98
0020     - Patch for Cray T90 from Wayne Schroeder <schroede@SDSC.EDU>
0021     - Lots and lots of changes
0023 29-Jan-98
0024     - ASN1_BIT_STRING_set_bit()/ASN1_BIT_STRING_get_bit() from
0025       Goetz Babin-Ebell <>.
0026     - SSL_version() now returns SSL2_VERSION, SSL3_VERSION or
0027       TLS1_VERSION.
0029 7-Jan-98
0030     - Finally reworked the cipher string to ciphers again, so it
0031       works correctly
0032     - All the app_data stuff is now ex_data with funcion calls to access.
0033       The index is supplied by a function and 'methods' can be setup
0034       for the types that are called on XXX_new/XXX_free.  This lets
0035       applications get notified on creation and destruction.  Some of
0036       the RSA methods could be implemented this way and I may do so.
0037     - Oh yes, SSL under perl5 is working at the basic level.
0039 15-Dec-97
0040     - Warning - the gethostbyname cache is not fully thread safe,
0041       but it should work well enough.
0042     - Major internal reworking of the app_data stuff.  More functions
0043       but if you were accessing ->app_data directly, things will
0044       stop working.
0045     - The perlv5 stuff is working.  Currently on message digests,
0046       ciphers and the bignum library.
0048 9-Dec-97
0049     - Modified re-negotiation so that server initated re-neg
0050       will cause a SSL_read() to return -1 should retry.
0051       The danger otherwise was that the server and the
0052       client could end up both trying to read when using non-blocking
0053       sockets.
0055 4-Dec-97
0056     - Lots of small changes
0057     - Fix for binaray mode in Windows for the FILE BIO, thanks to
0058       Bob Denny <>
0060 17-Nov-97
0061     - Quite a few internal cleanups, (removal of errno, and using macros
0062       defined in e_os.h).
0063     - A bug in ca.c, pointed out by, where
0064       the automactic naming out output files was being stuffed up.
0066 29-Oct-97
0067     - The Cast5 cipher has been added.  MD5 and SHA-1 are now in assember
0068       for x86.
0070 21-Oct-97
0071     - Fixed a bug in the BIO_gethostbyname() cache.
0073 15-Oct-97
0074     - cbc mode for blowfish/des/3des is now in assember.  Blowfish asm
0075       has also been improved.  At this point in time, on the pentium,
0076       md5 is %80 faster, the unoptimesed sha-1 is %79 faster,
0077       des-cbc is %28 faster, des-ede3-cbc is %9 faster and blowfish-cbc
0078       is %62 faster.
0080 12-Oct-97
0081     - MEM_BUF_grow() has been fixed so that it always sets the buf->length
0082       to the value we are 'growing' to.  Think of MEM_BUF_grow() as the
0083       way to set the length value correctly.
0085 10-Oct-97
0086     - I now hash for certificate lookup on the raw DER encoded RDN (md5).
0087       This breaks things again :-(.  This is efficent since I cache
0088       the DER encoding of the RDN.
0089     - The text DN now puts in the numeric OID instead of UNKNOWN.
0090     - req can now process arbitary OIDs in the config file.
0091     - I've been implementing md5 in x86 asm, much faster :-).
0092     - Started sha1 in x86 asm, needs more work.
0093     - Quite a few speedups in the BN stuff.  RSA public operation
0094       has been made faster by caching the BN_MONT_CTX structure.
0095       The calulating of the Ai where A*Ai === 1 mod m was rather
0096       expensive.  Basically a 40-50% speedup on public operations.
0097       The RSA speedup is now 15% on pentiums and %20 on pentium
0098       pro.
0100 30-Sep-97
0101     - After doing some profiling, I added x86 adm for bn_add_words(),
0102       which just adds 2 arrays of longs together.  A %10 speedup
0103       for 512 and 1024 bit RSA on the pentium pro.
0105 29-Sep-97
0106     - Converted the x86 bignum assembler to us the perl scripts
0107       for generation.
0109 23-Sep-97
0110     - If SSL_set_session() is passed a NULL session, it now clears the
0111       current session-id.
0113 22-Sep-97
0114     - Added a '-ss_cert file' to apps/ca.c.  This will sign selfsigned
0115       certificates.
0116     - Bug in crypto/evp/encode.c where by decoding of 65 base64
0117       encoded lines, one line at a time (via a memory BIO) would report
0118       EOF after the first line was decoded.
0119     - Fix in X509_find_by_issuer_and_serial() from
0120       Dr Stephen Henson <>
0122 19-Sep-97
0123     - NO_FP_API and NO_STDIO added.
0124     - Put in sh config command.  It auto runs Configure with the correct
0125       parameters.
0127 18-Sep-97
0128     - Fix x509.c so if a DSA cert has different parameters to its parent,
0129       they are left in place.  Not tested yet.
0131 16-Sep-97
0132     - ssl_create_cipher_list() had some bugs, fixes from
0133       Patrick Eisenacher <>
0134     - Fixed a bug in the Base64 BIO, where it would return 1 instead
0135       of -1 when end of input was encountered but should retry.
0136       Basically a Base64/Memory BIO interaction problem.
0137     - Added a HMAC set of functions in preporarion for TLS work.
0139 15-Sep-97
0140     - Top level makefile tweak - Cameron Simpson <>
0141     - Prime generation spead up %25 (512 bit prime, pentium pro linux)
0142       by using montgomery multiplication in the prime number test.
0144 11-Sep-97
0145     - Ugly bug in ssl3_write_bytes().  Basically if application land
0146       does a SSL_write(ssl,buf,len) where len > 16k, the SSLv3 write code
0147       did not check the size and tried to copy the entire buffer.
0148       This would tend to cause memory overwrites since SSLv3 has
0149       a maximum packet size of 16k.  If your program uses
0150       buffers <= 16k, you would probably never see this problem.
0151     - Fixed a few errors that were cause by malloc() not returning
0152       0 initialised memory..
0153     - SSL_OP_NETSCAPE_CA_DN_BUG was being switched on when using
0154       SSL_CTX_set_options(ssl_ctx,SSL_OP_ALL); which was a bad thing
0155       since this flags stops SSLeay being able to handle client
0156       cert requests correctly.
0158 08-Sep-97
0159     - SSL_SESS_CACHE_NO_INTERNAL_LOOKUP option added.  When switched
0160       on, the SSL server routines will not use a SSL_SESSION that is
0161       held in it's cache.  This in intended to be used with the session-id
0162       callbacks so that while the session-ids are still stored in the
0163       cache, the decision to use them and how to look them up can be
0164       done by the callbacks.  The are the 'new', 'get' and 'remove'
0165       callbacks.  This can be used to determine the session-id
0166       to use depending on information like which port/host the connection
0167       is coming from.  Since the are also SSL_SESSION_set_app_data() and
0168       SSL_SESSION_get_app_data() functions, the application can hold
0169       information against the session-id as well.
0171 03-Sep-97
0172     - Added lookup of CRLs to the by_dir method,
0173       X509_load_crl_file() also added.  Basically it means you can
0174       lookup CRLs via the same system used to lookup certificates.
0175     - Changed things so that the X509_NAME structure can contain
0176       ASN.1 BIT_STRINGS which is required for the unique
0177       identifier OID.
0178     - Fixed some problems with the auto flushing of the session-id
0179       cache.  It was not occuring on the server side.
0181 02-Sep-97
0182     - Added SSL_CTX_sess_cache_size(SSL_CTX *ctx,unsigned long size)
0183       which is the maximum number of entries allowed in the
0184       session-id cache.  This is enforced with a simple FIFO list.
0185       The default size is 20*1024 entries which is rather large :-).
0186       The Timeout code is still always operating.
0188 01-Sep-97
0189     - Added an argument to all the 'generate private key/prime`
0190       callbacks.  It is the last parameter so this should not
0191       break existing code but it is needed for C++.
0192     - Added the BIO_FLAGS_BASE64_NO_NL flag for the BIO_f_base64()
0193       BIO.  This lets the BIO read and write base64 encoded data
0194       without inserting or looking for '\n' characters.  The '-A'
0195       flag turns this on when using apps/enc.c.
0196     - RSA_NO_PADDING added to help BSAFE functionality.  This is a
0197       very dangerous thing to use, since RSA private key
0198       operations without random padding bytes (as PKCS#1 adds) can
0199       be attacked such that the private key can be revealed.
0200     - ASN.1 bug and rc2-40-cbc and rc4-40 added by
0201       Dr Stephen Henson <>
0203 31-Aug-97 (stuff added while I was away)    
0204     - Linux pthreads by Tim Hudson (
0205     - RSA_flags() added allowing bypass of pub/priv match check
0206       in ssl/ssl_rsa.c - Tim Hudson.
0207     - A few minor bugs.
0209 SSLeay 0.8.1 released.
0211 19-Jul-97
0212     - Server side initated dynamic renegotiation is broken.  I will fix
0213       it when I get back from holidays.
0215 15-Jul-97
0216     - Quite a few small changes.
0217     - INVALID_SOCKET usage cleanups from Alex Kiernan <>
0219 09-Jul-97
0220     - Added 2 new values to the SSL info callback.
0221       SSL_CB_START which is passed when the SSL protocol is started
0222       and SSL_CB_DONE when it has finished sucsessfully.
0224 08-Jul-97
0225     - Fixed a few bugs problems in apps/req.c and crypto/asn1/x_pkey.c
0226       that related to DSA public/private keys.
0227     - Added all the relevent PEM and normal IO functions to support
0228       reading and writing RSAPublic keys.
0229     - Changed makefiles to use ${AR} instead of 'ar r'
0231 07-Jul-97
0232     - Error in ERR_remove_state() that would leave a dangling reference
0233       to a free()ed location - thanks to Alex Kiernan <>
0234     - s_client now prints the X509_NAMEs passed from the server
0235       when requesting a client cert.
0236     - Added a ssl->type, which is one of SSL_ST_CONNECT or
0237       SSL_ST_ACCEPT.  I had to add it so I could tell if I was
0238       a connect or an accept after the handshake had finished.
0239     - SSL_get_client_CA_list(SSL *s) now returns the CA names
0240       passed by the server if called by a client side SSL.
0242 05-Jul-97
0243     - Bug in X509_NAME_get_text_by_OBJ(), looking starting at index
0244       0, not -1 :-(  Fix from Tim Hudson (
0246 04-Jul-97
0247     - Fixed some things in X509_NAME_add_entry(), thanks to
0248       Matthew Donald <>.
0249     - I had a look at the cipher section and though that it was a
0250       bit confused, so I've changed it.
0251     - I was not setting up the RC4-64-MD5 cipher correctly.  It is
0252       a MS special that appears in exported MS Money.
0253     - Error in all my DH ciphers.  Section of the SSLv3
0254       spec.  I was missing the two byte length header for the
0255       ClientDiffieHellmanPublic value.  This is a packet sent from
0256       the client to the server.  The SSL_OP_SSLEAY_080_CLIENT_DH_BUG
0257       option will enable SSLeay server side SSLv3 accept either
0258       the correct or my 080 packet format.
0259     - Fixed a few typos in crypto/
0261 02-Jul-97
0262     - Alias mapping for EVP_get_(digest|cipher)byname is now
0263       performed before a lookup for actual cipher.  This means
0264       that an alias can be used to 're-direct' a cipher or a
0265       digest.
0266     - ASN1_read_bio() had a bug that only showed up when using a
0267       memory BIO.  When EOF is reached in the memory BIO, it is
0268       reported as a -1 with BIO_should_retry() set to true.
0270 01-Jul-97
0271     - Fixed an error in X509_verify_cert() caused by my
0272       miss-understanding how 'do { contine } while(0);' works.
0273       Thanks to Emil Sit <> for educating me :-)
0275 30-Jun-97
0276     - Base64 decoding error.  If the last data line did not end with
0277       a '=', sometimes extra data would be returned.
0278     - Another 'cut and paste' bug in x509.c related to setting up the
0279       STDout BIO.
0281 27-Jun-97
0282     - apps/ciphers.c was not printing due to an editing error.
0283     - Alex Kiernan <> send in a nice fix for
0284       a library build error in util/
0286 26-Jun-97
0287     - Still did not have the auto 'experimental' code removal
0288       script correct.
0289     - A few header tweaks for Watcom 11.0 under Win32 from
0290       Rolf Lindemann <>
0291     - 0 length OCTET_STRING bug in asn1_parse
0292     - A minor fix with an non-existent function in the MS .def files.
0293     - A few changes to the PKCS7 stuff.
0295 25-Jun-97
0296     SSLeay 0.8.0 finally it gets released.
0298 24-Jun-97
0299     Added a SSL_OP_EPHEMERAL_RSA option which causes all SSLv3 RSA keys to
0300     use a temporary RSA key.  This is experimental and needs some more work.
0301     Fixed a few Win16 build problems.
0303 23-Jun-97
0304     SSLv3 bug. I was not doing the 'lookup' of the CERT structure
0305     correctly. I was taking the SSL->ctx->default_cert when I should
0306     have been using SSL->cert. The bug was in ssl/s3_srvr.c
0308 20-Jun-97
0309     X509_ATTRIBUTES were being encoded wrongly by apps/reg.c and the
0310     rest of the library. Even though I had the code required to do
0311     it correctly, apps/req.c was doing the wrong thing.  I have fixed
0312     and tested everything.
0314     Missing a few #ifdef FIONBIO sections in crypto/bio/bss_acpt.c.
0316 19-Jun-97
0317     Fixed a bug in the SSLv2 server side first packet handling. When
0318     using the non-blocking test BIO, the ssl->s2->first_packet flag
0319     was being reset when a would-block failure occurred when reading
0320     the first 5 bytes of the first packet. This caused the checking
0321     logic to run at the wrong time and cause an error.
0323     Fixed a problem with specifying cipher. If RC4-MD5 were used,
0324     only the SSLv3 version would be picked up.  Now this will pick
0325     up both SSLv2 and SSLv3 versions. This required changing the
0326     SSL_CIPHER->mask values so that they only mask the ciphers,
0327     digests, authentication, export type and key-exchange algorithms.
0329     I found that when a SSLv23 session is established, a reused
0330     session, of type SSLv3 was attempting to write the SSLv2 
0331     ciphers, which were invalid. The SSL_METHOD->put_cipher_by_char 
0332     method has been modified so it will only write out cipher which
0333     that method knows about.  
0336  Changes between 0.8.0 and 0.8.1
0338   *) Mostly bug fixes. 
0339      There is an Ephemeral DH cipher problem which is fixed.
0341  SSLeay 0.8.0
0343 This version of SSLeay has quite a lot of things different from the
0344 previous version.
0346 Basically check all callback parameters, I will be producing documentation
0347 about how to use things in th future.  Currently I'm just getting 080 out
0348 the door.  Please not that there are several ways to do everything, and
0349 most of the applications in the apps directory are hybrids, some using old
0350 methods and some using new methods.
0352 Have a look in demos/bio for some very simple programs and
0353 apps/s_client.c and apps/s_server.c for some more advanced versions.
0354 Notes are definitly needed but they are a week or so away.
0356 Anyway, some quick nots from Tim Hudson (
0357 ---
0358 Quick porting notes for moving from SSLeay-0.6.x to SSLeay-0.8.x to
0359 get those people that want to move to using the new code base off to
0360 a quick start.
0362 Note that Eric has tidied up a lot of the areas of the API that were
0363 less than desirable and renamed quite a few things (as he had to break
0364 the API in lots of places anyrate). There are a whole pile of additional
0365 functions for making dealing with (and creating) certificates a lot
0366 cleaner.
0368 01-Jul-97
0369 Tim Hudson
0372 ---8<---
0374 To maintain code that uses both SSLeay-0.6.x and SSLeay-0.8.x you could
0375 use something like the following (assuming you #include "crypto.h" which
0376 is something that you really should be doing).
0378 #if SSLEAY_VERSION_NUMBER >= 0x0800
0379 #define SSLEAY8
0380 #endif
0382 buffer.h -> splits into buffer.h and bio.h so you need to include bio.h
0383             too if you are working with BIO internal stuff (as distinct
0384         from simply using the interface in an opaque manner)
0386 #include "bio.h"    - required along with "buffer.h" if you write
0387               your own BIO routines as the buffer and bio
0388               stuff that was intermixed has been separated
0389               out 
0391 envelope.h -> evp.h  (which should have been done ages ago)
0393 Initialisation ... don't forget these or you end up with code that
0394 is missing the bits required to do useful things (like ciphers):
0396 SSLeay_add_ssl_algorithms()
0397 (probably also want SSL_load_error_strings() too but you should have
0398  already had that call in place)
0400 SSL_CTX_new()   - requires an extra method parameter
0401               SSL_CTX_new(SSLv23_method()) 
0402               SSL_CTX_new(SSLv2_method()) 
0403               SSL_CTX_new(SSLv3_method()) 
0405           OR to only have the server or the client code
0406               SSL_CTX_new(SSLv23_server_method()) 
0407               SSL_CTX_new(SSLv2_server_method()) 
0408               SSL_CTX_new(SSLv3_server_method()) 
0409           or  
0410               SSL_CTX_new(SSLv23_client_method()) 
0411               SSL_CTX_new(SSLv2_client_method()) 
0412               SSL_CTX_new(SSLv3_client_method()) 
0414 SSL_set_default_verify_paths() ... renamed to the more appropriate
0415 SSL_CTX_set_default_verify_paths()
0417 If you want to use client certificates then you have to add in a bit
0418 of extra stuff in that a SSLv3 server sends a list of those CAs that
0419 it will accept certificates from ... so you have to provide a list to
0420 SSLeay otherwise certain browsers will not send client certs.
0422 SSL_CTX_set_client_CA_list(ctx,SSL_load_client_CA_file(s_cert_file));
0425 X509_NAME_oneline(X)    -> X509_NAME_oneline(X,NULL,0)  
0426                or provide a buffer and size to copy the
0427                result into
0429 X509_add_cert ->  X509_STORE_add_cert (and you might want to read the
0430           notes on X509_NAME structure changes too)
0434 =================
0436 The codes have all be renamed from VERIFY_ERR_* to X509_V_ERR_* to
0437 more accurately reflect things.
0439 The verification callback args are now packaged differently so that
0440 extra fields for verification can be added easily in future without
0441 having to break things by adding extra parameters each release :-)
0443 X509_cert_verify_error_string -> X509_verify_cert_error_string
0447 =============
0449 Eric has fixed things so that extra flags can be introduced in
0450 the BIO layer in future without having to play with all the BIO
0451 modules by adding in some macros.
0453 The ugly stuff using 
0454     b->flags ~= (BIO_FLAGS_RW|BIO_FLAGS_SHOULD_RETRY)
0455 becomes
0456     BIO_clear_retry_flags(b)
0459 becomes
0460     BIO_set_retry_read(b)
0462 Also ... BIO_get_retry_flags(b), BIO_set_flags(b)
0467 ============
0469 X509_NAME has been altered so that it isn't just a STACK ... the STACK
0470 is now in the "entries" field ... and there are a pile of nice functions
0471 for getting at the details in a much cleaner manner.
0473 SSL_CTX has been altered ... "cert" is no longer a direct member of this
0474 structure ... things are now down under "cert_store" (see x509_vfy.h) and
0475 things are no longer in a CERTIFICATE_CTX but instead in a X509_STORE.
0476 If your code "knows" about this level of detail then it will need some 
0477 surgery.
0479 If you depending on the incorrect spelling of a number of the error codes
0480 then you will have to change your code as these have been fixed.
0482 ENV_CIPHER "type" got renamed to "nid" and as that is what it actually
0483 has been all along so this makes things clearer.
0484 ify_cert_error_string(ctx->error));
0487             and SSL_R_REUSE_CIPHER_LIST_NOT_ZERO
0491  Changes between 0.7.x and 0.8.0
0493   *) There have been lots of changes, mostly the addition of SSLv3.
0494      There have been many additions from people and amongst
0495      others, C2Net has assisted greatly.
0497  Changes between 0.7.x and 0.7.x
0499   *) Internal development version only
0501 SSLeay 0.6.6 13-Jan-1997
0503 The main additions are
0505 - assember for x86 DES improvments.
0506   From 191,000 per second on a pentium 100, I now get 281,000.  The inner
0507   loop and the IP/FP modifications are from
0508   Svend Olaf Mikkelsen <>.  Many thanks for his
0509   contribution.
0510 - The 'DES macros' introduced in 0.6.5 now have 3 types.
0511   DES_PTR1, DES_PTR2 and 'normal'.  As per before, des_opts reports which
0512   is best and there is a summery of mine in crypto/des/options.txt
0513 - A few bug fixes.
0514 - Added blowfish.  It is not used by SSL but all the other stuff that
0515   deals with ciphers can use it in either ecb, cbc, cfb64 or ofb64 modes.
0516   There are 3 options for optimising Blowfish.  BF_PTR, BF_PTR2 and 'normal'.
0517   BF_PTR2 is pentium/x86 specific.  The correct option is setup in
0518   the 'Configure' script.
0519 - There is now a 'get client certificate' callback which can be
0520   'non-blocking'.  If more details are required, let me know.  It will
0521   documented more in SSLv3 when I finish it.
0522 - Bug fixes from 0.6.5 including the infamous 'ca' bug.  The 'make test'
0523   now tests the ca program.
0524 - Lots of little things modified and tweaked.
0526  SSLeay 0.6.5
0528 After quite some time (3 months), the new release.  I have been very busy
0529 for the last few months and so this is mostly bug fixes and improvments.
0531 The main additions are
0533 - assember for x86 DES.  For all those gcc based systems, this is a big
0534   improvement.  From 117,000 DES operation a second on a pentium 100,
0535   I now get 191,000.  I have also reworked the C version so it
0536   now gives 148,000 DESs per second.  
0537 - As mentioned above, the inner DES macros now have some more variant that
0538   sometimes help, sometimes hinder performance.  There are now 3 options
0539   DES_PTR (ptr vs array lookup), DES_UNROLL (full vs partial loop unrolling)
0540   and DES_RISC (a more register intensive version of the inner macro).
0541   The crypto/des/des_opts.c program, when compiled and run, will give
0542   an indication of the correct options to use.
0543 - The BIO stuff has been improved.  Read doc/bio.doc.  There are now
0544   modules for encryption and base64 encoding and a BIO_printf() function.
0545 - The CA program will accept simple one line X509v3 extensions in the
0546   ssleay.cnf file.  Have a look at the example.  Currently this just
0547   puts the text into the certificate as an OCTET_STRING so currently
0548   the more advanced X509v3 data types are not handled but this is enough
0549   for the netscape extensions.
0550 - There is the start of a nicer higher level interface to the X509
0551   strucutre.
0552 - Quite a lot of bug fixes.
0553 - CRYPTO_malloc_init()  (or CRYPTO_set_mem_functions()) can be used
0554   to define the malloc(), free() and realloc() routines to use
0555   (look in crypto/crypto.h).  This is mostly needed for Windows NT/95 when
0556   using DLLs and mixing CRT libraries.
0558 In general, read the 'VERSION' file for changes and be aware that some of
0559 the new stuff may not have been tested quite enough yet, so don't just plonk
0560 in SSLeay 0.6.5 when 0.6.4 used to work and expect nothing to break.
0562 SSLeay 0.6.4 30/08/96 eay
0564 I've just finished some test builds on Windows NT, Windows 3.1, Solaris 2.3,
0565 Solaris 2.5, Linux, IRIX, HPUX 10 and everthing seems to work :-).
0567 The main changes in this release
0569 - Thread safe.  have a read of doc/threads.doc and play in the mt directory.
0570   For anyone using 0.6.3 with threads, I found 2 major errors so consider
0571   moving to 0.6.4.  I have a test program that builds under NT and
0572   solaris.
0573 - The get session-id callback has changed.  Have a read of doc/callback.doc.
0574 - The X509_cert_verify callback (the SSL_verify callback) now
0575   has another argument.  Have a read of doc/callback.doc
0576 - 'ca -preserve', sign without re-ordering the DN.  Not tested much.
0577 - VMS support.
0578 - Compile time memory leak detection can now be built into SSLeay.
0579   Read doc/memory.doc
0580 - CONF routines now understand '\', '\n', '\r' etc.  What this means is that
0581   the  SPKAC object mentioned in doc/ns-ca.doc can be on multiple lines.
0582 - 'ssleay ciphers' added, lists the default cipher list for SSLeay.
0583 - RC2 key setup is now compatable with Netscape.
0584 - Modifed server side of SSL implementation, big performance difference when
0585       using session-id reuse.
0587 0.6.3
0589 Bug fixes and the addition of some nice stuff to the 'ca' program.
0590 Have a read of doc/ns-ca.doc for how hit has been modified so
0591 it can be driven from a CGI script.  The CGI script is not provided,
0592 but that is just being left as an excersize for the reader :-).
0594 0.6.2
0596 This is most bug fixes and functionality improvements.
0598 Additions are
0599 - More thread debugging patches, the thread stuff is still being
0600   tested, but for those keep to play with stuff, have a look in
0601   crypto/cryptlib.c.  The application needs to define 1 (or optionaly
0602   a second) callback that is used to implement locking.  Compiling
0603   with LOCK_DEBUG spits out lots of locking crud :-).
0604   This is what I'm currently working on.
0605 - SSL_CTX_set_default_passwd_cb() can be used to define the callback
0606   function used in the SSL*_file() functions used to load keys.  I was
0607   always of the opinion that people should call
0608   PEM_read_RSAPrivateKey() and pass the callback they want to use, but
0609   it appears they just want to use the SSL_*_file() function() :-(.
0610 - 'enc' now has a -kfile so a key can be read from a file.  This is
0611   mostly used so that the passwd does not appear when using 'ps',
0612   which appears imposible to stop under solaris.
0613 - X509v3 certificates now work correctly.  I even have more examples
0614   in my tests :-).  There is now a X509_EXTENSION type that is used in
0615   X509v3 certificates and CRLv2.
0616 - Fixed that signature type error :-(
0617 - Fixed quite a few potential memory leaks and problems when reusing
0618   X509, CRL and REQ structures.
0619 - EVP_set_pw_prompt() now sets the library wide default password
0620   prompt.
0621 - The 'pkcs7' command will now, given the -print_certs flag, output in
0622   pem format, all certificates and CRL contained within.  This is more
0623   of a pre-emtive thing for the new verisign distribution method.  I
0624   should also note, that this also gives and example in code, of how
0625   to do this :-), or for that matter, what is involved in going the
0626   other way (list of certs and crl -> pkcs7).
0627 - Added RSA's DESX to the DES library.  It is also available via the
0628   EVP_desx_cbc() method and via 'enc desx'. 
0630 SSLeay 0.6.1
0632 The main functional changes since 0.6.0 are as follows
0633 - Bad news, the Microsoft 060 DLL's are not compatable, but the good news is
0634   that from now on, I'll keep the .def numbers the same so they will be.
0635 - RSA private key operations are about 2 times faster that 0.6.0
0636 - The SSL_CTX now has more fields so default values can be put against
0637   it.  When an SSL structure is created, these default values are used
0638   but can be overwritten.  There are defaults for cipher, certificate,
0639   private key, verify mode and callback.  This means SSL session
0640   creation can now be
0641   ssl=SSL_new()
0642   SSL_set_fd(ssl,sock);
0643   SSL_accept(ssl)
0644   ....
0645   All the other uglyness with having to keep a global copy of the
0646   private key and certificate/verify mode in the server is now gone.
0647 - ssl/ssltest.c - one process talking SSL to its self for testing.
0648 - Storage of Session-id's can be controled via a session_cache_mode
0649   flag.  There is also now an automatic default flushing of 
0650   old session-id's.
0651 - The X509_cert_verify() function now has another parameter, this
0652   should not effect most people but it now means that the reason for
0653   the failure to verify is now available via SSL_get_verify_result(ssl).
0654   You don't have to use a global variable.
0655 - SSL_get_app_data() and SSL_set_app_data() can be used to keep some
0656   application data against the SSL structure.  It is upto the application
0657   to free the data.  I don't use it, but it is available.
0658 - SSL_CTX_set_cert_verify_callback() can be used to specify a
0659   verify callback function that completly replaces my certificate
0660   verification code.  Xcert should be able to use this :-).
0661   The callback is of the form int app_verify_callback(arg,ssl,cert).
0662   This needs to be documented more.
0663 - I have started playing with shared library builds, have a look in
0664   the shlib directory.  It is very simple.  If you need a numbered
0665   list of functions, have a look at misc/crypto.num and misc/ssl.num.
0666 - There is some stuff to do locking to make the library thread safe.
0667   I have only started this stuff and have not finished.  If anyone is
0668   keen to do so, please send me the patches when finished.
0670 So I have finally made most of the additions to the SSL interface that
0671 I thought were needed.
0673 There will probably be a pause before I make any non-bug/documentation
0674 related changes to SSLeay since I'm feeling like a bit of a break.
0676 eric - 12 Jul 1996
0677 I saw recently a comment by some-one that we now seem to be entering
0678 the age of perpetual Beta software.
0679 Pioneered by packages like linux but refined to an art form by
0680 netscape.
0682 I too wish to join this trend with the anouncement of SSLeay 0.6.0 :-).
0684 There are quite a large number of sections that are 'works in
0685 progress' in this package.  I will also list the major changes and
0686 what files you should read.
0688 BIO - this is the new IO structure being used everywhere in SSLeay.  I
0689 started out developing this because of microsoft, I wanted a mechanism
0690 to callback to the application for all IO, so Windows 3.1 DLL
0691 perversion could be hidden from me and the 15 different ways to write
0692 to a file under NT would also not be dictated by me at library build
0693 time.  What the 'package' is is an API for a data structure containing
0694 functions.  IO interfaces can be written to conform to the
0695 specification.  This in not intended to hide the underlying data type
0696 from the application, but to hide it from SSLeay :-).
0697 I have only really finished testing the FILE * and socket/fd modules.
0698 There are also 'filter' BIO's.  Currently I have only implemented
0699 message digests, and it is in use in the dgst application.  This
0700 functionality will allow base64/encrypto/buffering modules to be
0701 'push' into a BIO without it affecting the semantics.  I'm also
0702 working on an SSL BIO which will hide the SSL_accept()/SLL_connet()
0703 from an event loop which uses the interface.
0704 It is also possible to 'attach' callbacks to a BIO so they get called
0705 before and after each operation, alowing extensive debug output
0706 to be generated (try running dgst with -d).
0708 Unfortunaly in the conversion from 0.5.x to 0.6.0, quite a few
0709 functions that used to take FILE *, now take BIO *.
0710 The wrappers are easy to write
0712 function_fp(fp,x)
0713 FILE *fp;
0714     {
0715     BIO *b;
0716     int ret;
0718     if ((b=BIO_new(BIO_s_file())) == NULL) error.....
0719     BIO_set_fp(b,fp,BIO_NOCLOSE);
0720     ret=function_bio(b,x);
0721     BIO_free(b);
0722     return(ret);
0723     }
0724 Remember, there are no functions that take FILE * in SSLeay when
0725 compiled for Windows 3.1 DLL's.
0727 --
0728 I have added a general EVP_PKEY type that can hold a public/private
0729 key.  This is now what is used by the EVP_ functions and is passed
0730 around internally.  I still have not done the PKCS#8 stuff, but
0731 X509_PKEY is defined and waiting :-)
0733 --
0734 For a full function name listings, have a look at ms/crypt32.def and
0735 ms/ssl32.def.  These are auto-generated but are complete.
0736 Things like ASN1_INTEGER_get() have been added and are in here if you
0737 look.  I have renamed a few things, again, have a look through the
0738 function list and you will probably find what you are after.  I intend
0739 to at least put a one line descrition for each one.....
0741 --
0742 Microsoft - thats what this release is about, read the MICROSOFT file.
0744 --
0745 Multi-threading support.  I have started hunting through the code and
0746 flaging where things need to be done.  In a state of work but high on
0747 the list.
0749 --
0750 For random numbers, edit e_os.h and set DEVRANDOM (it's near the top)
0751 be be you random data device, otherwise 'RFILE' in e_os.h
0752 will be used, in your home directory.  It will be updated
0753 periodically.  The environment variable RANDFILE will override this
0754 choice and read/write to that file instead.  DEVRANDOM is used in
0755 conjunction to the RFILE/RANDFILE.  If you wish to 'seed' the random
0756 number generator, pick on one of these files.
0758 --
0760 The list of things to read and do
0762 dgst -d
0763 s_client -state (this uses a callback placed in the SSL state loop and
0764         will be used else-where to help debug/monitor what
0765         is happening.)
0767 doc/why.doc
0768 doc/bio.doc <- hmmm, needs lots of work.
0769 doc/bss_file.doc <- one that is working :-)
0770 doc/session.doc <- it has changed
0771 doc/speed.doc
0772  also play with ssleay version -a.  I have now added a SSLeay()
0773  function that returns a version number, eg 0600 for this release
0774  which is primarily to be used to check DLL version against the
0775  application.
0776 util/*  Quite a few will not interest people, but some may, like
0778 util/
0780 try
0781 cc -Iinclude -Icrypto -c crypto/crypto.c
0782 cc -Iinclude -Issl -c ssl/ssl.c
0783 You have just built the SSLeay libraries as 2 object files :-)
0785 Have a general rummage around in the bin stall directory and look at
0786 what is in there, like and c_rehash
0788 There are lots more things but it is 12:30am on a Friday night and I'm
0789 heading home :-).
0791 eric 22-Jun-1996
0792 This version has quite a few major bug fixes and improvements.  It DOES NOT
0793 do SSLv3 yet.
0795 The main things changed
0796 - A Few days ago I added the s_mult application to ssleay which is
0797   a demo of an SSL server running in an event loop type thing.
0798   It supports non-blocking IO, I have finally gotten it right, SSL_accept()
0799   can operate in non-blocking IO mode, look at the code to see how :-).
0800   Have a read of doc/s_mult as well.  This program leaks memory and
0801   file descriptors everywhere but I have not cleaned it up yet.
0802   This is a demo of how to do non-blocking IO.
0803 - The SSL session management has been 'worked over' and there is now
0804   quite an expansive set of functions to manipulate them.  Have a read of
0805   doc/session.doc for some-things I quickly whipped up about how it now works.
0806   This assume you know the SSLv2 protocol :-)
0807 - I can now read/write the netscape certificate format, use the
0808   -inform/-outform  'net' options to the x509 command.  I have not put support
0809   for this type in the other demo programs, but it would be easy to add.
0810 - asn1parse and 'enc' have been modified so that when reading base64
0811   encoded files (pem format), they do not require '-----BEGIN' header lines.
0812   The 'enc' program had a buffering bug fixed, it can be used as a general
0813   base64 -> binary -> base64 filter by doing 'enc -a -e' and 'enc -a -d'
0814   respecivly.  Leaving out the '-a' flag in this case makes the 'enc' command
0815   into a form of 'cat'.
0816 - The 'x509' and 'req' programs have been fixed and modified a little so
0817   that they generate self-signed certificates correctly.  The test
0818   script actually generates a 'CA' certificate and then 'signs' a
0819   'user' certificate.  Have a look at this shell script (test/sstest)
0820   to see how things work, it tests most possible combinations of what can
0821   be done.
0822 - The 'SSL_set_pref_cipher()' function has been 'fixed' and the prefered name
0823   of SSL_set_cipher_list() is now the correct API (stops confusion :-).
0824   If this function is used in the client, only the specified ciphers can
0825   be used, with preference given to the order the ciphers were listed.
0826   For the server, if this is used, only the specified ciphers will be used
0827   to accept connections.  If this 'option' is not used, a default set of
0828   ciphers will be used.  The SSL_CTX_set_cipher_list(SSL_CTX *ctx) sets this
0829   list for all ciphers started against the SSL_CTX.  So the order is
0830   SSL cipher_list, if not present, SSL_CTX cipher list, if not
0831   present, then the library default.
0832   What this means is that normally ciphers like
0833   NULL-MD5 will never be used.  The only way this cipher can be used
0834   for both ends to specify to use it.
0835   To enable or disable ciphers in the library at build time, modify the
0836   first field for the cipher in the ssl_ciphers array in ssl/ssl_lib.c.
0837   This file also contains the 'pref_cipher' list which is the default
0838   cipher preference order.
0839 - I'm not currently sure if the 'rsa -inform net' and the 'rsa -outform net'
0840   options work.  They should, and they enable loading and writing the
0841   netscape rsa private key format.  I will be re-working this section of
0842   SSLeay for the next version.  What is currently in place is a quick and
0843   dirty hack.
0844 - I've re-written parts of the bignum library.  This gives speedups
0845   for all platforms.  I now provide assembler for use under Windows NT.
0846   I have not tested the Windows 3.1 assembler but it is quite simple code.
0847   This gives RSAprivate_key operation encryption times of 0.047s (512bit key)
0848   and 0.230s (1024bit key) on a pentium 100 which I consider reasonable.
0849   Basically the times available under linux/solaris x86 can be achieve under
0850   Windows NT.  I still don't know how these times compare to RSA's BSAFE
0851   library but I have been emailing with people and with their help, I should
0852   be able to get my library's quite a bit faster still (more algorithm changes).
0853   The object file crypto/bn/asm/x86-32.obj should be used when linking
0854   under NT.
0855 - 'make' in the top directory will generate a single makefile
0856   called ''  This makefile contains no perl references and
0857   will build the SSLeay library into the 'tmp' and 'out' directories.
0858   util/ > is how this makefile is
0859   generated.  The command take several option to generate the
0860   makefile for use with cc, gcc, Visual C++ and Borland C++.  This is
0861   still under development.  I have only build .lib's for NT and MSDOS
0862   I will be working on this more.  I still need to play with the
0863   correct compiler setups for these compilers and add some more stuff but
0864   basically if you just want to compile the library
0865   on a 'non-unix' platform, this is a very very good file to start with :-).
0866   Have a look in the 'microsoft' directory for my current makefiles.
0867   I have not yet modified things to link with sockets under Windows NT.
0868   You guys should be able to do this since this is actually outside of the
0869   SSLeay scope :-).  I will be doing it for myself soon.
0870   util/ takes quite a few options including no-rc, rsaref  and no-sock
0871   to build without RC2/RC4, to require RSAref for linking, and to
0872   build with no socket code.
0874 - Oh yes, the cipher that was reported to be compatible with RSA's RC2 cipher
0875   that was posted to sci.crypt has been added to the library and SSL.
0876   I take the view that if RC2 is going to be included in a standard,
0877   I'll include the cipher to make my package complete.
0878   There are NO_RC2, NO_RC4 and NO_IDEA macros to remove these ciphers
0879   at compile time.  I have not tested this recently but it should all work
0880   and if you are in the USA and don't want RSA threatening to sue you,
0881   you could probably remove the RC4/RC2 code inside these sections.
0882   I may in the future include a perl script that does this code
0883   removal automatically for those in the USA :-).
0884 - I have removed all references to sed in the makefiles.  So basically,
0885   the development environment requires perl and sh.  The build environment
0886   does not (use the makefile).
0887   The Configure script still requires perl, this will probably stay that way
0888   since I have perl for Windows NT :-).
0890 eric (03-May-1996)
0892 PS Have a look in the VERSION file for more details on the changes and
0893    bug fixes.
0894 I have fixed a few bugs, added alpha and x86 assembler and generally cleaned
0895 things up.  This version will be quite stable, mostly because I'm on
0896 holidays until 10-March-1996.  For any problems in the interum, send email
0897 to Tim Hudson <>.
0899 SSLeay 0.5.0
0901 12-12-95
0902 This is going out before it should really be released.
0904 I leave for 11 weeks holidays on the 22-12-95 and so I either sit on
0905 this for 11 weeks or get things out.  It is still going to change a
0906 lot in the next week so if you do grab this version, please test and
0907 give me feed back ASAP, inculuding questions on how to do things with
0908 the library.  This will prompt me to write documentation so I don't
0909 have to answer the same question again :-).
0911 This 'pre' release version is for people who are interested in the
0912 library.  The applications will have to be changed to use
0913 the new version of the SSL interface.  I intend to finish more
0914 documentation before I leave but until then, look at the programs in
0915 the apps directory.  As far as code goes, it is much much nicer than
0916 the old version.
0918 The current library works, has no memory leaks (as far as I can tell)
0919 and is far more bug free that 0.4.5d.  There are no global variable of
0920 consequence (I believe) and I will produce some documentation that
0921 tell where to look for those people that do want to do multi-threaded
0922 stuff.
0924 There should be more documentation.  Have a look in the
0925 doc directory.  I'll be adding more before I leave, it is a start
0926 by mostly documents the crypto library.  Tim Hudson will update
0927 the web page ASAP.  The spelling and grammar are crap but
0928 it is better than nothing :-)
0930 Reasons to start playing with version 0.5.0
0931 - All the programs in the apps directory build into one ssleay binary.
0932 - There is a new version of the 'req' program that generates certificate
0933   requests, there is even documentation for this one :-)
0934 - There is a demo certification authorithy program.  Currently it will
0935   look at the simple database and update it.  It will generate CRL from
0936   the data base.  You need to edit the database by hand to revoke a
0937   certificate, it is my aim to use perl5/Tk but I don't have time to do
0938   this right now.  It will generate the certificates but the management
0939   scripts still need to be written.  This is not a hard task.
0940 - Things have been cleaned up alot.
0941 - Have a look at the enc and dgst programs in the apps directory.
0942 - It supports v3 of x509 certiticates.
0945 Major things missing.
0946 - I have been working on (and thinging about) the distributed x509
0947   hierachy problem.  I have not had time to put my solution in place.
0948   It will have to wait until I come back.
0949 - I have not put in CRL checking in the certificate verification but
0950   it would not be hard to do.  I was waiting until I could generate my
0951   own CRL (which has only been in the last week) and I don't have time
0952   to put it in correctly.
0953 - Montgomery multiplication need to be implemented.  I know the
0954   algorithm, just ran out of time.
0955 - PKCS#7.  I can load and write the DER version.  I need to re-work
0956   things to support BER (if that means nothing, read the ASN1 spec :-).
0957 - Testing of the higher level digital envelope routines.  I have not
0958   played with the *_seal() and *_open() type functions.  They are
0959   written but need testing.  The *_sign() and *_verify() functions are
0960   rock solid. 
0961 - PEM.  Doing this and PKCS#7 have been dependant on the distributed
0962   x509 heirachy problem.  I started implementing my ideas, got
0963   distracted writing a CA program and then ran out of time.  I provide
0964   the functionality of RSAref at least.
0965 - Re work the asm. code for the x86.  I've changed by low level bignum
0966   interface again, so I really need to tweak the x86 stuff.  gcc is
0967   good enough for the other boxes.