openssl.git
24 years ago"make update"
Richard Levitte [Sun, 9 Apr 2000 12:52:40 +0000 (12:52 +0000)]
"make update"

24 years agoI've always wanted to make the CONF library more adaptable. Here's
Richard Levitte [Sun, 9 Apr 2000 12:04:35 +0000 (12:04 +0000)]
I've always wanted to make the CONF library more adaptable.  Here's
the result.

I have retained the old behavior of the CONF_* functions, and have
added a more "object oriented" interface through NCONF_* functions
(New CONF, you see :-)), working the same way as, for example, the
BIO interface.  Really, the CONF_* are rewritten so they use the
NCONF_* functions internally.

In addition to that, I've split the old conf.c code into two files,
conf_def.c and conf_api.c.  conf_def.c contains the default config
object that reads a configuration file the standard OpenSSL way, as
well as configuration file with Win32 registry file syntax (I'm not
sure I got that one right).  conf_api.c provides an API to build other
configuration file readers around (can you see a configuraion file in
XML?  I can :-)).

Finally, I've changed the name conf_lcl.h to conf_def.h, since it's
made specifically for that "class" and none others.

24 years ago"make update" for DSO additions.
Bodo Möller [Thu, 6 Apr 2000 22:33:39 +0000 (22:33 +0000)]
"make update" for DSO additions.

24 years agoIn theory, TLS v1 ciphersuites are not the same as SSL v3 ciphersuites
Bodo Möller [Thu, 6 Apr 2000 22:33:14 +0000 (22:33 +0000)]
In theory, TLS v1 ciphersuites are not the same as SSL v3 ciphersuites

24 years agoClarifications.
Bodo Möller [Thu, 6 Apr 2000 22:30:57 +0000 (22:30 +0000)]
Clarifications.

24 years agoAvoid memory leak.
Bodo Möller [Thu, 6 Apr 2000 22:25:49 +0000 (22:25 +0000)]
Avoid memory leak.

24 years agoConstification, and a silly mistake in the comments.
Geoff Thorpe [Thu, 6 Apr 2000 14:54:54 +0000 (14:54 +0000)]
Constification, and a silly mistake in the comments.

24 years agoI forgot about $openssl_other_defines ... should probably do this
Geoff Thorpe [Thu, 6 Apr 2000 11:56:45 +0000 (11:56 +0000)]
I forgot about $openssl_other_defines ... should probably do this
for consistency. Not sure though whether HAVE_DLFCN_H should be
included too? If we go the autoconf route then this probably wouldn't
be included.

24 years agoThis helps make the DSO stuff more portable;
Geoff Thorpe [Thu, 6 Apr 2000 07:09:45 +0000 (07:09 +0000)]
This helps make the DSO stuff more portable;
* "no-dso" option available in Configure so that all DSO methods will
  return NULL, overriding any support the platform might otherwise
  have built.
* dlfcn_no_h config string now available rather than just dlfcn. This
  is for platforms that have dlfcn.h functions but do not have (or
  need) the dlfcn.h header file.

24 years agoBetter make a note of what's going on ... :-)
Geoff Thorpe [Tue, 4 Apr 2000 22:49:27 +0000 (22:49 +0000)]
Better make a note of what's going on ... :-)

24 years agoThis commit ties the new DSO code (crypto/dso/) into the build for a
Geoff Thorpe [Tue, 4 Apr 2000 22:32:19 +0000 (22:32 +0000)]
This commit ties the new DSO code (crypto/dso/) into the build for a
variety of platforms. A few are missing, and they will be added in
eventually, but as this is new stuff, it was better to not break lots of
platforms in one go that we can't easily test. The changes to "Configure"
should illustrate how to add support to other systems if you feel like
having a go.

NB: I'll add something shortly to allow you to add "dlfcn.h" support on
those platforms that don't have (or need) a dlfcn.h header file. (The
symbol for Configure will probably by "dlfcn_no_h").

Thanks to Richard Levitte, who is responsible for the dso_dl.c support,
understanding the trickier aspects of the build process, and giving great
feedback on everything else.

[Don't use this stuff if you're easily offended by changes to the
interface or behaviour - it's still work in progress.]

PR:

24 years agoThis is a set of startup code for the DSO support, it's not yet linked into
Geoff Thorpe [Tue, 4 Apr 2000 21:57:11 +0000 (21:57 +0000)]
This is a set of startup code for the DSO support, it's not yet linked into
the build process (an upcoming commit no doubt), and is very much *new*
code - what that means is that it compiles ok - usually. It certainly
doesn't mean it runs well or even properly yet. Please don't muck round
with this unless you're looking to help out and hunt bugs. :-)

Currently this code doesn't have any support for controlling the "load"
behaviour (eg. paths, filename translations, etc). That'll be handled
using DSO_ctrl() and various flags, once we work out a sensible set of
flags.

24 years agoExtend entry on ERR_print_errors.
Bodo Möller [Tue, 4 Apr 2000 21:31:14 +0000 (21:31 +0000)]
Extend entry on ERR_print_errors.

24 years agoRelease done, change status.
Richard Levitte [Sat, 1 Apr 2000 12:32:10 +0000 (12:32 +0000)]
Release done, change status.

24 years agoTagging has now been done, update to the next version (it's not quite
Richard Levitte [Sat, 1 Apr 2000 11:24:27 +0000 (11:24 +0000)]
Tagging has now been done, update to the next version (it's not quite
as important to keep a low profile here :-))

24 years agoTagging has now been done, update to the next possible version (I keep
Richard Levitte [Sat, 1 Apr 2000 11:21:14 +0000 (11:21 +0000)]
Tagging has now been done, update to the next possible version (I keep
a low profile, so we don't get discontinuity in the numbering...)

24 years agoVersion and name changes, and a last minute changelog OpenSSL_0_9_5a
Richard Levitte [Sat, 1 Apr 2000 11:15:15 +0000 (11:15 +0000)]
Version and name changes, and a last minute changelog

24 years agoBuilding version 0.9.5a
Richard Levitte [Sat, 1 Apr 2000 11:03:41 +0000 (11:03 +0000)]
Building version 0.9.5a

24 years ago_lrotl and _lrotr are defined in MSVC but nowhere else that we know
Richard Levitte [Sat, 1 Apr 2000 10:39:33 +0000 (10:39 +0000)]
_lrotl and _lrotr are defined in MSVC but nowhere else that we know
off.

24 years agoEntry for ssleay_rand_status locking fix.
Bodo Möller [Thu, 30 Mar 2000 08:12:35 +0000 (08:12 +0000)]
Entry for ssleay_rand_status locking fix.

24 years ago'entropy >= ENTROPY_NEEDED' should be evaluated while the
Bodo Möller [Thu, 30 Mar 2000 06:11:54 +0000 (06:11 +0000)]
'entropy >= ENTROPY_NEEDED' should be evaluated while the
variables are locked.

24 years agoAdd a couple of FAQs.
Dr. Stephen Henson [Thu, 30 Mar 2000 02:00:25 +0000 (02:00 +0000)]
Add a couple of FAQs.

24 years agoSince ssleay_rand_initialize() unlocks then locks CRYPTO_LOCK_RAND,
Richard Levitte [Wed, 29 Mar 2000 17:25:52 +0000 (17:25 +0000)]
Since ssleay_rand_initialize() unlocks then locks CRYPTO_LOCK_RAND,
it's a good thing if ssleay_rand_status() would do the corresponding
lock and unlock as everyone else...

24 years agoAdd pointer to EGD manpage.
Ulf Möller [Tue, 28 Mar 2000 19:45:06 +0000 (19:45 +0000)]
Add pointer to EGD manpage.
Update the "randomness" section for the upcoming 0.9.5a release.

24 years agoAdd a default banner. Contributed by Joon Radley <jradley@vps.co.za>
Richard Levitte [Tue, 28 Mar 2000 18:56:04 +0000 (18:56 +0000)]
Add a default banner.  Contributed by Joon Radley <jradley@vps.co.za>

24 years agoDon't force the use of GNU make with Borland C++ Builder. Contributed by Joon Radley...
Richard Levitte [Tue, 28 Mar 2000 18:54:27 +0000 (18:54 +0000)]
Don't force the use of GNU make with Borland C++ Builder.  Contributed by Joon Radley <jradley@vps.co.za>

24 years agoFix typo in -clrext option, but add a compatibility hack because
Bodo Möller [Mon, 27 Mar 2000 18:10:08 +0000 (18:10 +0000)]
Fix typo in -clrext option, but add a compatibility hack because
0.9.5a should not break anything that works in 0.9.5.

24 years agoComments for SSL_get_peer_cert_chain inconsistency.
Bodo Möller [Mon, 27 Mar 2000 18:07:45 +0000 (18:07 +0000)]
Comments for SSL_get_peer_cert_chain inconsistency.

24 years agoRemoved, especially since it conflicts with des.pod on
Richard Levitte [Mon, 27 Mar 2000 12:34:58 +0000 (12:34 +0000)]
Removed, especially since it conflicts with des.pod on
case-insensitive file systems.

24 years agoFix for HMAC.
Dr. Stephen Henson [Mon, 27 Mar 2000 00:53:27 +0000 (00:53 +0000)]
Fix for HMAC.

24 years agoNeXT workaround.
Bodo Möller [Sun, 26 Mar 2000 12:27:30 +0000 (12:27 +0000)]
NeXT workaround.

24 years agoBeautifying code.
Richard Levitte [Sat, 25 Mar 2000 21:17:51 +0000 (21:17 +0000)]
Beautifying code.

24 years agoList "no-..." option first because it's the most frequently needed one.
Bodo Möller [Sat, 25 Mar 2000 19:02:09 +0000 (19:02 +0000)]
List "no-..." option first because it's the most frequently needed one.

24 years agoVi, again
Richard Levitte [Sat, 25 Mar 2000 10:44:28 +0000 (10:44 +0000)]
Vi, again

24 years agoInfo on proposed release date
Richard Levitte [Sat, 25 Mar 2000 10:28:35 +0000 (10:28 +0000)]
Info on proposed release date

24 years agoAnother VAX environment passed
Richard Levitte [Sat, 25 Mar 2000 10:23:46 +0000 (10:23 +0000)]
Another VAX environment passed

24 years agoVI gave me 4-space tabs... Corrected
Richard Levitte [Sat, 25 Mar 2000 10:14:15 +0000 (10:14 +0000)]
VI gave me 4-space tabs...  Corrected

24 years agoOne VAX environment passed
Richard Levitte [Sat, 25 Mar 2000 09:28:54 +0000 (09:28 +0000)]
One VAX environment passed

24 years agoUpdate docs.
Dr. Stephen Henson [Sat, 25 Mar 2000 02:38:28 +0000 (02:38 +0000)]
Update docs.

24 years agoExplain configuration options more completely.
Bodo Möller [Sat, 25 Mar 2000 01:49:07 +0000 (01:49 +0000)]
Explain configuration options more completely.

24 years agoUpdate for new hpux-parisc-cc-o4 entry.
Bodo Möller [Fri, 24 Mar 2000 19:46:22 +0000 (19:46 +0000)]
Update for new hpux-parisc-cc-o4 entry.

24 years agoanother one done.
Geoff Thorpe [Fri, 24 Mar 2000 16:55:32 +0000 (16:55 +0000)]
another one done.

24 years agoMore info in the Win32 test
Richard Levitte [Fri, 24 Mar 2000 16:10:33 +0000 (16:10 +0000)]
More info in the Win32 test

24 years agoA couple of corrections concerning HPUX 10 and shared libraries.
Richard Levitte [Fri, 24 Mar 2000 16:06:22 +0000 (16:06 +0000)]
A couple of corrections concerning HPUX 10 and shared libraries.
Contributed by Lutz Jaenicke.

24 years agoGeoff Thorpe reports a whole bunch of tests passing.
Richard Levitte [Fri, 24 Mar 2000 15:19:07 +0000 (15:19 +0000)]
Geoff Thorpe reports a whole bunch of tests passing.

24 years agoAnother VMS/Alpha with a much more evil compiler passed as well...
Richard Levitte [Fri, 24 Mar 2000 15:08:04 +0000 (15:08 +0000)]
Another VMS/Alpha with a much more evil compiler passed as well...

24 years agoVMS on Alpha passed
Richard Levitte [Fri, 24 Mar 2000 12:40:28 +0000 (12:40 +0000)]
VMS on Alpha passed

24 years agoAdd a few more FAQs.
Dr. Stephen Henson [Fri, 24 Mar 2000 00:47:25 +0000 (00:47 +0000)]
Add a few more FAQs.

24 years agoTest results.
Ulf Möller [Thu, 23 Mar 2000 23:19:12 +0000 (23:19 +0000)]
Test results.

24 years agoSurround the diff with the new snippage delimiters
Richard Levitte [Thu, 23 Mar 2000 22:43:20 +0000 (22:43 +0000)]
Surround the diff with the new snippage delimiters

24 years agoUpdate status
Richard Levitte [Thu, 23 Mar 2000 22:01:25 +0000 (22:01 +0000)]
Update status

24 years agoTagging has been done, update to next probable version...
Richard Levitte [Thu, 23 Mar 2000 21:14:11 +0000 (21:14 +0000)]
Tagging has been done, update to next probable version...

24 years agoI almost forgot... OpenSSL_0_9_5a-beta2
Richard Levitte [Thu, 23 Mar 2000 21:12:50 +0000 (21:12 +0000)]
I almost forgot...

24 years agoTime for version 0.9.5a beta2
Richard Levitte [Thu, 23 Mar 2000 21:07:41 +0000 (21:07 +0000)]
Time for version 0.9.5a beta2

I know it's earlier than announced.  The high amount of problems in
beta1 warants this, however.

24 years agoNew FAQ for the OpenSSH configuration problem.
Richard Levitte [Thu, 23 Mar 2000 19:31:37 +0000 (19:31 +0000)]
New FAQ for the OpenSSH configuration problem.

I know, this isn't really a OpenSSL issue, but we keep getting
questions about it, so we might as well provide the solution...

24 years agoClean up context, even if an error occured.
Richard Levitte [Thu, 23 Mar 2000 18:14:21 +0000 (18:14 +0000)]
Clean up context, even if an error occured.

24 years agossize_t for Ultrix
Ulf Möller [Thu, 23 Mar 2000 17:54:58 +0000 (17:54 +0000)]
ssize_t for Ultrix

24 years agoCorrected. It said before that ivec had to be initialised to zero,
Richard Levitte [Thu, 23 Mar 2000 11:08:49 +0000 (11:08 +0000)]
Corrected.  It said before that ivec had to be initialised to zero,
which isn't true.  What is true, however, is that SSH assumes it is.

24 years ago"make update"
Bodo Möller [Thu, 23 Mar 2000 09:51:23 +0000 (09:51 +0000)]
"make update"

24 years agoA quick hack to have ssize_t work with gcc under SunOS 4 (tested on
Richard Levitte [Thu, 23 Mar 2000 07:32:34 +0000 (07:32 +0000)]
A quick hack to have ssize_t work with gcc under SunOS 4 (tested on
SunOS 4.1.4).

24 years agoUpdate docs and remove old PKCS#7 README file.
Dr. Stephen Henson [Thu, 23 Mar 2000 02:35:47 +0000 (02:35 +0000)]
Update docs and remove old PKCS#7 README file.

24 years agoRAND_event()
Ulf Möller [Wed, 22 Mar 2000 15:30:03 +0000 (15:30 +0000)]
RAND_event()

24 years agoFix a memory leak in PKCS12_parse.
Dr. Stephen Henson [Wed, 22 Mar 2000 13:50:23 +0000 (13:50 +0000)]
Fix a memory leak in PKCS12_parse.
Don't copy private key to X509 etc public key structures.
Fix for warning.

24 years agomore failures
Ulf Möller [Tue, 21 Mar 2000 20:42:35 +0000 (20:42 +0000)]
more failures

24 years agoMore failures.
Ulf Möller [Tue, 21 Mar 2000 11:40:33 +0000 (11:40 +0000)]
More failures.

24 years agoUlf M\81öller reports OpenBSD-x86 and solaris-sparcv9-cc passed.
Richard Levitte [Mon, 20 Mar 2000 21:38:25 +0000 (21:38 +0000)]
Ulf M\81öller reports OpenBSD-x86 and solaris-sparcv9-cc passed.

24 years agoOpenBSD complains.
Ulf Möller [Mon, 20 Mar 2000 18:19:38 +0000 (18:19 +0000)]
OpenBSD complains.

24 years agoIn some of the Makefiles CPP was not defined.
Ulf Möller [Mon, 20 Mar 2000 18:16:52 +0000 (18:16 +0000)]
In some of the Makefiles CPP was not defined.

24 years agoJaenick Lutz reports that hpux-parisc-cc with +02 passed.
Richard Levitte [Mon, 20 Mar 2000 16:52:29 +0000 (16:52 +0000)]
Jaenick Lutz reports that hpux-parisc-cc with +02 passed.

24 years agoRef: Message-ID: <20000320070111.A90876@wiz.Sendmail.COM>
Richard Levitte [Mon, 20 Mar 2000 15:56:14 +0000 (15:56 +0000)]
Ref: Message-ID: <20000320070111.A90876@wiz.Sendmail.COM>

24 years agoProblems with the Windows build.
Ulf Möller [Mon, 20 Mar 2000 13:40:16 +0000 (13:40 +0000)]
Problems with the Windows build.

24 years agoDocument pseudo-commands.
Bodo Möller [Mon, 20 Mar 2000 13:39:06 +0000 (13:39 +0000)]
Document pseudo-commands.

24 years agoMention the bug fixes.
Ulf Möller [Mon, 20 Mar 2000 12:17:20 +0000 (12:17 +0000)]
Mention the bug fixes.

24 years agoI just got a not very pleasant report.
Richard Levitte [Mon, 20 Mar 2000 11:29:11 +0000 (11:29 +0000)]
I just got a not very pleasant report.

24 years agoThere are a few new features in 0.9.5a... And I've probably
Richard Levitte [Mon, 20 Mar 2000 09:23:05 +0000 (09:23 +0000)]
There are a few new features in 0.9.5a...  And I've probably
forgotten a few.

24 years agoNew status
Richard Levitte [Mon, 20 Mar 2000 09:22:04 +0000 (09:22 +0000)]
New status

24 years agoTagging done, we move to the next possible.
Richard Levitte [Mon, 20 Mar 2000 07:47:37 +0000 (07:47 +0000)]
Tagging done, we move to the next possible.

24 years agoBeta 1 of 0.9.5a building. OpenSSL_0_9_5a-beta1
Richard Levitte [Mon, 20 Mar 2000 07:36:22 +0000 (07:36 +0000)]
Beta 1 of 0.9.5a building.

24 years agomake update
Richard Levitte [Mon, 20 Mar 2000 07:35:38 +0000 (07:35 +0000)]
make update

24 years agoChange the version text, it's time to release the first beta of 0.9.5a.
Richard Levitte [Mon, 20 Mar 2000 07:22:47 +0000 (07:22 +0000)]
Change the version text, it's time to release the first beta of 0.9.5a.

24 years agoSample application using RAND_event() to collect entropy from mouse
Ulf Möller [Sun, 19 Mar 2000 22:58:12 +0000 (22:58 +0000)]
Sample application using RAND_event() to collect entropy from mouse
movements, keyboard etc. and write it to a seed file.

24 years agoBug fix: RAND_write_file() failed to write to files created by open()
Ulf Möller [Sun, 19 Mar 2000 22:57:07 +0000 (22:57 +0000)]
Bug fix: RAND_write_file() failed to write to files created by open()
on Win32.

24 years agoMake sure an error condition is returned if, for some reason, the file
Richard Levitte [Sun, 19 Mar 2000 19:10:03 +0000 (19:10 +0000)]
Make sure an error condition is returned if, for some reason, the file
couldn't be opened.

24 years agooops. don't use "entropy" directly.
Ulf Möller [Sun, 19 Mar 2000 17:20:26 +0000 (17:20 +0000)]
oops. don't use "entropy" directly.

24 years agoNew function RAND_event() collects entropy from Windows events.
Ulf Möller [Sun, 19 Mar 2000 17:14:25 +0000 (17:14 +0000)]
New function RAND_event() collects entropy from Windows events.

24 years agoFLAT_INC is not needed; we use -I.. all the time
Ulf Möller [Sun, 19 Mar 2000 12:15:24 +0000 (12:15 +0000)]
FLAT_INC is not needed; we use -I.. all the time

24 years agoIntegrate podd.h sk.h into set_key.c
Ulf Möller [Sun, 19 Mar 2000 12:13:44 +0000 (12:13 +0000)]
Integrate podd.h sk.h into set_key.c

24 years agoChange the notation and coding of the version to be able to contain
Richard Levitte [Sun, 19 Mar 2000 09:35:19 +0000 (09:35 +0000)]
Change the notation and coding of the version to be able to contain
both a patch level and a beta status.  IMHO, it also makes more sense
to have beta status be part of the development status than to have it
be an alternate name for patch levels under special conditions.

24 years agoKeep in sync with Unix code, and prepare for a misfeature fix.
Richard Levitte [Sun, 19 Mar 2000 09:23:25 +0000 (09:23 +0000)]
Keep in sync with Unix code, and prepare for a misfeature fix.

24 years agomake update
Ulf Möller [Sun, 19 Mar 2000 02:18:45 +0000 (02:18 +0000)]
make update

24 years agocleanup.
Ulf Möller [Sun, 19 Mar 2000 02:10:17 +0000 (02:10 +0000)]
cleanup.

24 years agolibdes manpage.
Ulf Möller [Sun, 19 Mar 2000 02:09:37 +0000 (02:09 +0000)]
libdes manpage.

This may still contain a few errors from the old documentation,
but most of it should make sense.

24 years agodes_quad_cksum() byte order bug fix.
Ulf Möller [Sun, 19 Mar 2000 02:06:37 +0000 (02:06 +0000)]
des_quad_cksum() byte order bug fix.
See http://www.pdc.kth.se/kth-krb/

Their solution for CRAY is somewhat awkward.
I'll assume that a "short" is 32 bits on CRAY to avoid the
#ifdef _CRAY
    typedef struct {
        unsigned int a:32;
        unsigned int b:32;
    } XXX;
#else
    typedef DES_LONG XXX;
#endif

24 years agoadd =cut
Ulf Möller [Sat, 18 Mar 2000 22:00:26 +0000 (22:00 +0000)]
add =cut

24 years agoDES in Perl was incomplete and not very useful
Ulf Möller [Sat, 18 Mar 2000 18:39:58 +0000 (18:39 +0000)]
DES in Perl was incomplete and not very useful

24 years agosuperseded by des_modes.pod
Ulf Möller [Sat, 18 Mar 2000 18:37:43 +0000 (18:37 +0000)]
superseded by des_modes.pod

24 years agochange manpages to pod. Contents are not up to date!
Ulf Möller [Sat, 18 Mar 2000 17:55:09 +0000 (17:55 +0000)]
change manpages to pod. Contents are not up to date!

24 years agoEliminate memory leaks in mem_dbg.c.
Bodo Möller [Sat, 18 Mar 2000 15:18:27 +0000 (15:18 +0000)]
Eliminate memory leaks in mem_dbg.c.

24 years agoSmall bugs in the test scripts removed.
Richard Levitte [Sat, 18 Mar 2000 09:11:00 +0000 (09:11 +0000)]
Small bugs in the test scripts removed.

24 years agoe_os.h: don't do double work with status codes.
Richard Levitte [Sat, 18 Mar 2000 09:09:31 +0000 (09:09 +0000)]
e_os.h: don't do double work with status codes.
openssl.c: make damn sure e_os.h knows about OPENSSL_C