Submitted by: Nils Larsch <nla@trustcenter.de>

0.9.7 now lives in the branch OpenSSL_0_9_7-stable.

Submitted by: Nils Larsch <nla@trustcenter.de>

CygWin 1.3.x, which includes thread and shared library support.

Submitted by Corinna Vinschen <vinschen@redhat.com> and modified a
little bit.

weakness in SSL/TLS

extensions to link together, there's no point looping at all.

0.9.7.

CRYPTO_LOCK_RAND...

lets change our prefix to AEPHK_R_.  Otherwise, we get very mysterious
errors because we happen to redefine AEP_R_OK and AEP_R_GENERAL_ERROR.

that were never part of the engine framework.

The aep and sureware implementations are taken directly from 0.9.6c
[engine] and have been modified to fit the newer engine framework and
to be possible to build shared libraries of.

The aep implementation has gone through quite a bunch of tests and is
cleaned up (there were some misunderstandings in it about how to use
locks).

The sureware hasn't been tested at all in this incarnation and is
basically a quick hack to get it to compile properly.

make them ubsec-specific in the ubsec implementation.

automatically.

functions in ui_compat.  This gave reason to rework that part more
thoroughly, so here are the changes made:

1. Add DES_read_password() and DES_read_2passwords() with the same
   functionality as the corresponding old des_ functions, as a
   convenience to the users.
2. Add UI_UTIL_read_pw_string() and UI_UTIL_read_pw() with the
   functionality from des_read_pw_string() and des_read_pw(), again as
   a concenience to the users.
3. Rename des_read_password(), des_read_2passwords(),
   des_read_pw_string() and des_read_pw() by changing des_ to
   _ossl_old_des_, and add the usual mapping macros.
4. Move the implementation of des_read_password() and
   des_read_2passwords() to the des directory, since they are tightly
   tied to DES anyway.

This change was inspired by a patch from Assar Westerlund <assar@sics.se>:

There are some functions that didn't get the kick-away-old-des-and-
replace-des-with-DES action.  Here's a patch that adds DES_ and des_
(in des_old.h) versions of des_read_pw_string et al.  This patch
includes some of the first des_old.h semi-colon macro fixes that I've
already sent.

The following patch makes sure that string2key does not use weak DES
keys (then making them non-weak by xor:ing with 0xF0).

This patch makes the macros in des_old.h actually pretend to be
functions.

There's no reason not to define _ossl_old_crypt when using
PERL5/FreeBSD/darwin/Next, since it makes using crypt and including
des.h break.  Here's a trivial patch.

This patch fixes some of the typos used in macro names in des_old.h
and the number of arguments for some of them.

wincrypt.h is included.

* When linking against shared libraries, the absolute path is remembered.
  - When linking against -L.., '..' is remembered inside the executable,
    so it will fail after "make install" or when not called from inside the
    "apps/" subdirectory of the build tree.
  - When using the "+cdp" option of "ld", the ".." information can be
    exchanged against $(INSTALL_TOP)/lib. In this case the executable
    will however refuse to work before "make install" has been called.
    This makes testing the 'openssl' executable a problem.
* Solution 1:
  Relink the "openssl" executable, when "make install" is called.
  This would however require significant changes to the toplevel Makefile
  and the apps/ Makefile.
* Solution 2:
  Statically link against libssl and libcrypto, so that the "openssl"
  executable is no longer dependant on the openssl shared libraries.

Select option 2 for HP-UX 32bit, as this requires the smallest change.

[See
     Message-ID: <3BB07999.30432AD2@celocom.com>
     Date: Tue, 25 Sep 2001 13:33:29 +0100
     From: Dr S N Henson <drh@celocom.com>
     To: openssl-dev@openssl.org
     Subject: Re: Error in v3_purp.c
]

alpha-cc, alpha-cc-rpath, alpha-gcc, alpha164-cc and alphaold-cc.

Therefore, I've added a sanity checker.

Note that it can be combined with almost any other argument (the other
arguments will be completely ignored), with "reconf" as the blatant
exception, since it also has the behavior of ignoring all following
command line arguments.  If --test-sanity and reconf are both used on
the command line, the first one wins.