3 After quite some time (3 months), the new release. I have been very busy
4 for the last few months and so this is mostly bug fixes and improvments.
8 - assember for x86 DES. For all those gcc based systems, this is a big
9 improvement. From 117,000 DES operation a second on a pentium 100,
10 I now get 191,000. I have also reworked the C version so it
11 now gives 148,000 DESs per second.
12 - As mentioned above, the inner DES macros now have some more variant that
13 sometimes help, sometimes hinder performance. There are now 3 options
14 DES_PTR (ptr vs array lookup), DES_UNROLL (full vs partial loop unrolling)
15 and DES_RISC (a more register intensive version of the inner macro).
16 The crypto/des/des_opts.c program, when compiled and run, will give
17 an indication of the correct options to use.
18 - The BIO stuff has been improved. Read doc/bio.doc. There are now
19 modules for encryption and base64 encoding and a BIO_printf() function.
20 - The CA program will accept simple one line X509v3 extensions in the
21 ssleay.cnf file. Have a look at the example. Currently this just
22 puts the text into the certificate as an OCTET_STRING so currently
23 the more advanced X509v3 data types are not handled but this is enough
24 for the netscape extensions.
25 - There is the start of a nicer higher level interface to the X509
27 - Quite a lot of bug fixes.
28 - CRYPTO_malloc_init() (or CRYPTO_set_mem_functions()) can be used
29 to define the malloc(), free() and realloc() routines to use
30 (look in crypto/crypto.h). This is mostly needed for Windows NT/95 when
31 using DLLs and mixing CRT libraries.
33 In general, read the 'VERSION' file for changes and be aware that some of
34 the new stuff may not have been tested quite enough yet, so don't just plonk
35 in SSLeay 0.6.5 when 0.6.4 used to work and expect nothing to break.
37 SSLeay 0.6.4 30/08/96 eay
39 I've just finished some test builds on Windows NT, Windows 3.1, Solaris 2.3,
40 Solaris 2.5, Linux, IRIX, HPUX 10 and everthing seems to work :-).
42 The main changes in this release
44 - Thread safe. have a read of doc/threads.doc and play in the mt directory.
45 For anyone using 0.6.3 with threads, I found 2 major errors so consider
46 moving to 0.6.4. I have a test program that builds under NT and
48 - The get session-id callback has changed. Have a read of doc/callback.doc.
49 - The X509_cert_verify callback (the SSL_verify callback) now
50 has another argument. Have a read of doc/callback.doc
51 - 'ca -preserve', sign without re-ordering the DN. Not tested much.
53 - Compile time memory leak detection can now be built into SSLeay.
55 - CONF routines now understand '\', '\n', '\r' etc. What this means is that
56 the SPKAC object mentioned in doc/ns-ca.doc can be on multiple lines.
57 - 'ssleay ciphers' added, lists the default cipher list for SSLeay.
58 - RC2 key setup is now compatable with Netscape.
59 - Modifed server side of SSL implementation, big performance difference when
60 using session-id reuse.
64 Bug fixes and the addition of some nice stuff to the 'ca' program.
65 Have a read of doc/ns-ca.doc for how hit has been modified so
66 it can be driven from a CGI script. The CGI script is not provided,
67 but that is just being left as an excersize for the reader :-).
71 This is most bug fixes and functionality improvements.
74 - More thread debugging patches, the thread stuff is still being
75 tested, but for those keep to play with stuff, have a look in
76 crypto/cryptlib.c. The application needs to define 1 (or optionaly
77 a second) callback that is used to implement locking. Compiling
78 with LOCK_DEBUG spits out lots of locking crud :-).
79 This is what I'm currently working on.
80 - SSL_CTX_set_default_passwd_cb() can be used to define the callback
81 function used in the SSL*_file() functions used to load keys. I was
82 always of the opinion that people should call
83 PEM_read_RSAPrivateKey() and pass the callback they want to use, but
84 it appears they just want to use the SSL_*_file() function() :-(.
85 - 'enc' now has a -kfile so a key can be read from a file. This is
86 mostly used so that the passwd does not appear when using 'ps',
87 which appears imposible to stop under solaris.
88 - X509v3 certificates now work correctly. I even have more examples
89 in my tests :-). There is now a X509_EXTENSION type that is used in
90 X509v3 certificates and CRLv2.
91 - Fixed that signature type error :-(
92 - Fixed quite a few potential memory leaks and problems when reusing
93 X509, CRL and REQ structures.
94 - EVP_set_pw_prompt() now sets the library wide default password
96 - The 'pkcs7' command will now, given the -print_certs flag, output in
97 pem format, all certificates and CRL contained within. This is more
98 of a pre-emtive thing for the new verisign distribution method. I
99 should also note, that this also gives and example in code, of how
100 to do this :-), or for that matter, what is involved in going the
101 other way (list of certs and crl -> pkcs7).
102 - Added RSA's DESX to the DES library. It is also available via the
103 EVP_desx_cbc() method and via 'enc desx'.
107 The main functional changes since 0.6.0 are as follows
108 - Bad news, the Microsoft 060 DLL's are not compatable, but the good news is
109 that from now on, I'll keep the .def numbers the same so they will be.
110 - RSA private key operations are about 2 times faster that 0.6.0
111 - The SSL_CTX now has more fields so default values can be put against
112 it. When an SSL structure is created, these default values are used
113 but can be overwritten. There are defaults for cipher, certificate,
114 private key, verify mode and callback. This means SSL session
117 SSL_set_fd(ssl,sock);
120 All the other uglyness with having to keep a global copy of the
121 private key and certificate/verify mode in the server is now gone.
122 - ssl/ssltest.c - one process talking SSL to its self for testing.
123 - Storage of Session-id's can be controled via a session_cache_mode
124 flag. There is also now an automatic default flushing of
126 - The X509_cert_verify() function now has another parameter, this
127 should not effect most people but it now means that the reason for
128 the failure to verify is now available via SSL_get_verify_result(ssl).
129 You don't have to use a global variable.
130 - SSL_get_app_data() and SSL_set_app_data() can be used to keep some
131 application data against the SSL structure. It is upto the application
132 to free the data. I don't use it, but it is available.
133 - SSL_CTX_set_cert_verify_callback() can be used to specify a
134 verify callback function that completly replaces my certificate
135 verification code. Xcert should be able to use this :-).
136 The callback is of the form int app_verify_callback(arg,ssl,cert).
137 This needs to be documented more.
138 - I have started playing with shared library builds, have a look in
139 the shlib directory. It is very simple. If you need a numbered
140 list of functions, have a look at misc/crypto.num and misc/ssl.num.
141 - There is some stuff to do locking to make the library thread safe.
142 I have only started this stuff and have not finished. If anyone is
143 keen to do so, please send me the patches when finished.
145 So I have finally made most of the additions to the SSL interface that
146 I thought were needed.
148 There will probably be a pause before I make any non-bug/documentation
149 related changes to SSLeay since I'm feeling like a bit of a break.
152 I saw recently a comment by some-one that we now seem to be entering
153 the age of perpetual Beta software.
154 Pioneered by packages like linux but refined to an art form by
157 I too wish to join this trend with the anouncement of SSLeay 0.6.0 :-).
159 There are quite a large number of sections that are 'works in
160 progress' in this package. I will also list the major changes and
161 what files you should read.
163 BIO - this is the new IO structure being used everywhere in SSLeay. I
164 started out developing this because of microsoft, I wanted a mechanism
165 to callback to the application for all IO, so Windows 3.1 DLL
166 perversion could be hidden from me and the 15 different ways to write
167 to a file under NT would also not be dictated by me at library build
168 time. What the 'package' is is an API for a data structure containing
169 functions. IO interfaces can be written to conform to the
170 specification. This in not intended to hide the underlying data type
171 from the application, but to hide it from SSLeay :-).
172 I have only really finished testing the FILE * and socket/fd modules.
173 There are also 'filter' BIO's. Currently I have only implemented
174 message digests, and it is in use in the dgst application. This
175 functionality will allow base64/encrypto/buffering modules to be
176 'push' into a BIO without it affecting the semantics. I'm also
177 working on an SSL BIO which will hide the SSL_accept()/SLL_connet()
178 from an event loop which uses the interface.
179 It is also possible to 'attach' callbacks to a BIO so they get called
180 before and after each operation, alowing extensive debug output
181 to be generated (try running dgst with -d).
183 Unfortunaly in the conversion from 0.5.x to 0.6.0, quite a few
184 functions that used to take FILE *, now take BIO *.
185 The wrappers are easy to write
193 if ((b=BIO_new(BIO_s_file())) == NULL) error.....
194 BIO_set_fp(b,fp,BIO_NOCLOSE);
195 ret=function_bio(b,x);
199 Remember, there are no functions that take FILE * in SSLeay when
200 compiled for Windows 3.1 DLL's.
203 I have added a general EVP_PKEY type that can hold a public/private
204 key. This is now what is used by the EVP_ functions and is passed
205 around internally. I still have not done the PKCS#8 stuff, but
206 X509_PKEY is defined and waiting :-)
209 For a full function name listings, have a look at ms/crypt32.def and
210 ms/ssl32.def. These are auto-generated but are complete.
211 Things like ASN1_INTEGER_get() have been added and are in here if you
212 look. I have renamed a few things, again, have a look through the
213 function list and you will probably find what you are after. I intend
214 to at least put a one line descrition for each one.....
217 Microsoft - thats what this release is about, read the MICROSOFT file.
220 Multi-threading support. I have started hunting through the code and
221 flaging where things need to be done. In a state of work but high on
225 For random numbers, edit e_os.h and set DEVRANDOM (it's near the top)
226 be be you random data device, otherwise 'RFILE' in e_os.h
227 will be used, in your home directory. It will be updated
228 periodically. The environment variable RANDFILE will override this
229 choice and read/write to that file instead. DEVRANDOM is used in
230 conjunction to the RFILE/RANDFILE. If you wish to 'seed' the random
231 number generator, pick on one of these files.
235 The list of things to read and do
238 s_client -state (this uses a callback placed in the SSL state loop and
239 will be used else-where to help debug/monitor what
243 doc/bio.doc <- hmmm, needs lots of work.
244 doc/bss_file.doc <- one that is working :-)
245 doc/session.doc <- it has changed
247 also play with ssleay version -a. I have now added a SSLeay()
248 function that returns a version number, eg 0600 for this release
249 which is primarily to be used to check DLL version against the
251 util/* Quite a few will not interest people, but some may, like
256 cc -Iinclude -Icrypto -c crypto/crypto.c
257 cc -Iinclude -Issl -c ssl/ssl.c
258 You have just built the SSLeay libraries as 2 object files :-)
260 Have a general rummage around in the bin stall directory and look at
261 what is in there, like CA.sh and c_rehash
263 There are lots more things but it is 12:30am on a Friday night and I'm
267 This version has quite a few major bug fixes and improvements. It DOES NOT
270 The main things changed
271 - A Few days ago I added the s_mult application to ssleay which is
272 a demo of an SSL server running in an event loop type thing.
273 It supports non-blocking IO, I have finally gotten it right, SSL_accept()
274 can operate in non-blocking IO mode, look at the code to see how :-).
275 Have a read of doc/s_mult as well. This program leaks memory and
276 file descriptors everywhere but I have not cleaned it up yet.
277 This is a demo of how to do non-blocking IO.
278 - The SSL session management has been 'worked over' and there is now
279 quite an expansive set of functions to manipulate them. Have a read of
280 doc/session.doc for some-things I quickly whipped up about how it now works.
281 This assume you know the SSLv2 protocol :-)
282 - I can now read/write the netscape certificate format, use the
283 -inform/-outform 'net' options to the x509 command. I have not put support
284 for this type in the other demo programs, but it would be easy to add.
285 - asn1parse and 'enc' have been modified so that when reading base64
286 encoded files (pem format), they do not require '-----BEGIN' header lines.
287 The 'enc' program had a buffering bug fixed, it can be used as a general
288 base64 -> binary -> base64 filter by doing 'enc -a -e' and 'enc -a -d'
289 respecivly. Leaving out the '-a' flag in this case makes the 'enc' command
290 into a form of 'cat'.
291 - The 'x509' and 'req' programs have been fixed and modified a little so
292 that they generate self-signed certificates correctly. The test
293 script actually generates a 'CA' certificate and then 'signs' a
294 'user' certificate. Have a look at this shell script (test/sstest)
295 to see how things work, it tests most possible combinations of what can
297 - The 'SSL_set_pref_cipher()' function has been 'fixed' and the prefered name
298 of SSL_set_cipher_list() is now the correct API (stops confusion :-).
299 If this function is used in the client, only the specified ciphers can
300 be used, with preference given to the order the ciphers were listed.
301 For the server, if this is used, only the specified ciphers will be used
302 to accept connections. If this 'option' is not used, a default set of
303 ciphers will be used. The SSL_CTX_set_cipher_list(SSL_CTX *ctx) sets this
304 list for all ciphers started against the SSL_CTX. So the order is
305 SSL cipher_list, if not present, SSL_CTX cipher list, if not
306 present, then the library default.
307 What this means is that normally ciphers like
308 NULL-MD5 will never be used. The only way this cipher can be used
309 for both ends to specify to use it.
310 To enable or disable ciphers in the library at build time, modify the
311 first field for the cipher in the ssl_ciphers array in ssl/ssl_lib.c.
312 This file also contains the 'pref_cipher' list which is the default
313 cipher preference order.
314 - I'm not currently sure if the 'rsa -inform net' and the 'rsa -outform net'
315 options work. They should, and they enable loading and writing the
316 netscape rsa private key format. I will be re-working this section of
317 SSLeay for the next version. What is currently in place is a quick and
319 - I've re-written parts of the bignum library. This gives speedups
320 for all platforms. I now provide assembler for use under Windows NT.
321 I have not tested the Windows 3.1 assembler but it is quite simple code.
322 This gives RSAprivate_key operation encryption times of 0.047s (512bit key)
323 and 0.230s (1024bit key) on a pentium 100 which I consider reasonable.
324 Basically the times available under linux/solaris x86 can be achieve under
325 Windows NT. I still don't know how these times compare to RSA's BSAFE
326 library but I have been emailing with people and with their help, I should
327 be able to get my library's quite a bit faster still (more algorithm changes).
328 The object file crypto/bn/asm/x86-32.obj should be used when linking
330 - 'make makefile.one' in the top directory will generate a single makefile
331 called 'makefile.one' This makefile contains no perl references and
332 will build the SSLeay library into the 'tmp' and 'out' directories.
333 util/mk1mf.pl >makefile.one is how this makefile is
334 generated. The mk1mf.pl command take several option to generate the
335 makefile for use with cc, gcc, Visual C++ and Borland C++. This is
336 still under development. I have only build .lib's for NT and MSDOS
337 I will be working on this more. I still need to play with the
338 correct compiler setups for these compilers and add some more stuff but
339 basically if you just want to compile the library
340 on a 'non-unix' platform, this is a very very good file to start with :-).
341 Have a look in the 'microsoft' directory for my current makefiles.
342 I have not yet modified things to link with sockets under Windows NT.
343 You guys should be able to do this since this is actually outside of the
344 SSLeay scope :-). I will be doing it for myself soon.
345 util/mk1mf.pl takes quite a few options including no-rc, rsaref and no-sock
346 to build without RC2/RC4, to require RSAref for linking, and to
347 build with no socket code.
349 - Oh yes, the cipher that was reported to be compatible with RSA's RC2 cipher
350 that was posted to sci.crypt has been added to the library and SSL.
351 I take the view that if RC2 is going to be included in a standard,
352 I'll include the cipher to make my package complete.
353 There are NO_RC2, NO_RC4 and NO_IDEA macros to remove these ciphers
354 at compile time. I have not tested this recently but it should all work
355 and if you are in the USA and don't want RSA threatening to sue you,
356 you could probably remove the RC4/RC2 code inside these sections.
357 I may in the future include a perl script that does this code
358 removal automatically for those in the USA :-).
359 - I have removed all references to sed in the makefiles. So basically,
360 the development environment requires perl and sh. The build environment
361 does not (use the makefile.one makefile).
362 The Configure script still requires perl, this will probably stay that way
363 since I have perl for Windows NT :-).
367 PS Have a look in the VERSION file for more details on the changes and
369 I have fixed a few bugs, added alpha and x86 assembler and generally cleaned
370 things up. This version will be quite stable, mostly because I'm on
371 holidays until 10-March-1996. For any problems in the interum, send email
372 to Tim Hudson <tjh@mincom.oz.au>.
377 This is going out before it should really be released.
379 I leave for 11 weeks holidays on the 22-12-95 and so I either sit on
380 this for 11 weeks or get things out. It is still going to change a
381 lot in the next week so if you do grab this version, please test and
382 give me feed back ASAP, inculuding questions on how to do things with
383 the library. This will prompt me to write documentation so I don't
384 have to answer the same question again :-).
386 This 'pre' release version is for people who are interested in the
387 library. The applications will have to be changed to use
388 the new version of the SSL interface. I intend to finish more
389 documentation before I leave but until then, look at the programs in
390 the apps directory. As far as code goes, it is much much nicer than
393 The current library works, has no memory leaks (as far as I can tell)
394 and is far more bug free that 0.4.5d. There are no global variable of
395 consequence (I believe) and I will produce some documentation that
396 tell where to look for those people that do want to do multi-threaded
399 There should be more documentation. Have a look in the
400 doc directory. I'll be adding more before I leave, it is a start
401 by mostly documents the crypto library. Tim Hudson will update
402 the web page ASAP. The spelling and grammar are crap but
403 it is better than nothing :-)
405 Reasons to start playing with version 0.5.0
406 - All the programs in the apps directory build into one ssleay binary.
407 - There is a new version of the 'req' program that generates certificate
408 requests, there is even documentation for this one :-)
409 - There is a demo certification authorithy program. Currently it will
410 look at the simple database and update it. It will generate CRL from
411 the data base. You need to edit the database by hand to revoke a
412 certificate, it is my aim to use perl5/Tk but I don't have time to do
413 this right now. It will generate the certificates but the management
414 scripts still need to be written. This is not a hard task.
415 - Things have been cleaned up alot.
416 - Have a look at the enc and dgst programs in the apps directory.
417 - It supports v3 of x509 certiticates.
420 Major things missing.
421 - I have been working on (and thinging about) the distributed x509
422 hierachy problem. I have not had time to put my solution in place.
423 It will have to wait until I come back.
424 - I have not put in CRL checking in the certificate verification but
425 it would not be hard to do. I was waiting until I could generate my
426 own CRL (which has only been in the last week) and I don't have time
427 to put it in correctly.
428 - Montgomery multiplication need to be implemented. I know the
429 algorithm, just ran out of time.
430 - PKCS#7. I can load and write the DER version. I need to re-work
431 things to support BER (if that means nothing, read the ASN1 spec :-).
432 - Testing of the higher level digital envelope routines. I have not
433 played with the *_seal() and *_open() type functions. They are
434 written but need testing. The *_sign() and *_verify() functions are
436 - PEM. Doing this and PKCS#7 have been dependant on the distributed
437 x509 heirachy problem. I started implementing my ideas, got
438 distracted writing a CA program and then ran out of time. I provide
439 the functionality of RSAref at least.
440 - Re work the asm. code for the x86. I've changed by low level bignum
441 interface again, so I really need to tweak the x86 stuff. gcc is
442 good enough for the other boxes.