2 OpenSSL STATUS Last modified at
3 ______________ $Date: 2000/03/24 16:55:32 $
7 o OpenSSL 0.9.5a: Under development...
8 0.9.5a-beta1 is available.
9 sunos-gcc - failed (ssize_t)
10 ultrix-gcc, ultrix-cc - failed (ssize_t)
11 sco5-cc - bc fails in test
12 VC-Win32 (MSVC6SP3, nasm) - failed (fixed)
13 FreeBSD-elf (i686-pc-freebsd3.2)- failed (fixed)
14 HPUX (hpux-parisc-cc w/ +02) - passed
16 solaris-sparcv9-cc - passed
17 0.9.5a-beta2 is available.
19 linux-ppc (egcs 2.91.66) - passed
21 FreeBSD-elf (i586-pc-freebsd3.2)- passed
22 solaris-sparcv9-cc - passed
23 solaris-sparcv9-gcc - passed
25 VMS/Alpha 7.1 w/ DEC C 5.6-003 - passed
26 VMS/Alpha 7.2-1 w/ CPQ C 6.2-003- passed
27 AIX 4.3.3.0 w/ cc - passed
28 AIX 4.3.3.0 w/ gcc (2.7.2.3?) - passed
29 Irix 6.4 w/ cc - passed
30 Irix 6.4 w/ gcc - passed
31 Irix 6.5 w/ cc - passed
32 Irix 6.5 w/ gcc - passed
34 Unixware 7.0.1 w/ native cc - passed
35 Solaris-x86 2.6 w/ gcc 2.7.2.3 - passed
36 Solaris-x86 2.7 w/ gcc 2.7.2.3 - passed
37 True64 Unix w/ gcc 2.7.2.3 - passed
38 Win32 w/ VC++ 5 & NASM 0.98 - passed
40 o OpenSSL 0.9.5: Released on February 28th, 2000
41 o OpenSSL 0.9.4: Released on August 09th, 1999
42 o OpenSSL 0.9.3a: Released on May 29th, 1999
43 o OpenSSL 0.9.3: Released on May 25th, 1999
44 o OpenSSL 0.9.2b: Released on March 22th, 1999
45 o OpenSSL 0.9.1c: Released on December 23th, 1998
51 o CA.pl patch (Damien Miller)
55 o Steve is currently working on (in no particular order):
56 Proper (or at least usable) certificate chain verification.
57 Private key, certificate and CRL API and implementation.
58 Developing and bugfixing PKCS#7 (S/MIME code).
59 Various X509 issues: character sets, certificate request extensions.
60 Documentation for the openssl utility.
64 o non-blocking socket on AIX
65 o $(PERL) in */Makefile.ssl
66 o "Sign the certificate?" - "n" creates empty certificate file
70 o internal_verify doesn't know about X509.v3 (basicConstraints
73 o The Makefile hierarchy and build mechanism is still not a round thing:
75 1. The config vs. Configure scripts
76 It's the same nasty situation as for Apache with APACI vs.
77 src/Configure. It confuses.
78 Suggestion: Merge Configure and config into a single configure
79 script with a Autoconf style interface ;-) and remove
80 Configure and config. Or even let us use GNU Autoconf
81 itself. Then we can avoid a lot of those platform checks
82 which are currently in Configure.
84 o Support for Shared Libraries has to be added at least
85 for the major Unix platforms. The details we can rip from the stuff
86 Ralf has done for the Apache src/Configure script. Ben wants the
87 solution to be really simple.
89 Status: Ralf will look how we can easily incorporate the
90 compiler PIC and linker DSO flags from Apache
91 into the OpenSSL Configure script.
93 Ulf: +1 for using GNU autoconf and libtool (but not automake,
94 which apparently is not flexible enough to generate
98 o The perl/ stuff needs a major overhaul. Currently it's
99 totally obsolete. Either we clean it up and enhance it to be up-to-date
100 with the C code or we also could replace it with the really nice
101 Net::SSLeay package we can find under
102 http://www.neuronio.pt/SSLeay.pm.html. Ralf uses this package for a
103 longer time and it works fine and is a nice Perl module. Best would be
104 to convince the author to work for the OpenSSL project and create a
105 Net::OpenSSL or Crypt::OpenSSL package out of it and maintains it for
108 Status: Ralf thinks we should both contact the author of Net::SSLeay
109 and look how much effort it is to bring Eric's perl/ stuff up
113 o The EVP and ASN1 stuff is a mess. Currently you have one EVP_CIPHER
114 structure for each cipher. This may make sense for things like DES but
115 for variable length ciphers like RC2 and RC4 it is NBG. Need a way to
116 use the EVP interface and set up the cipher parameters. The ASN1 stuff
117 is also foo wrt ciphers whose AlgorithmIdentifier has more than just
118 an IV in it (e.g. RC2, RC5). This also means that EVP_Seal and EVP_Open
119 don't work unless the key length matches the fixed value (some vendors
120 use a key length decided by the size of the RSA encrypted key and expect
123 o ERR_error_string(..., buf) does not know how large buf is,
124 there should be ERR_error_string_n(..., buf, bufsize)