2 OpenSSL STATUS Last modified at
3 ______________ $Date: 1999/04/16 11:32:33 $
7 o OpenSSL 0.9.3: Under development...
8 Proposed freeze date: Mon May 8th, 1999
9 Proposed release date: Mon May 17th, 1999
10 o OpenSSL 0.9.2b: Released on March 22th, 1999
11 o OpenSSL 0.9.1c: Released on December 23th, 1998
15 o Compilation warnings: ctype-related int vs. char
16 o Compilation error: "unsigned long*" and "int*" under AIX
17 o Undefined BN symbols in assembler stuff on Solaris
18 (see posting "openssl-SNAP-19990308-2130 on Solaris-2.6 SC4" in openssl-dev)
19 o BN assembler code in r3000.s, mips*.s and pa-risc*.s is broken
20 o BSD/OS: assembler functions must not have leading underscores
21 o Niels Poppe reports that RSA fails with alpha.s
22 o exptest fails with irix64-*
23 (Don Badrak <dbadrak@geo.census.gov>: "Re: Problems to compile openssl
24 on IRIX 6.2", openssl-users)
25 o BN_add test fails on Caldera OpenLinux 1.3
26 (Marc Christensen <Marc.Christensen@m.cc.utah.edu>
27 "Compiles but fails big number test?", openssl-users)
28 o make test fails on AIX
29 (jvant@bsumail.idbsu.edu on openssl-users)
33 o OCSP (titchenert@certco.com)
34 o unsigned vs. non-unsigned (levitte@stacken.kth.se)
35 o Install prefix for packagers (dharris@drh.net)
36 o getenv in ca.c and x509_def.c (jaltman@watsun.cc.columbia.edu)
37 o linux dynamic libs (colin@field.medicine.adelaide.edu.au)
38 o MingW support (niklas@canit.se)
39 o SPARC v8 assembler implementations of bn_asm.c
40 (appro@fy.chalmers.se)
44 o Steve is currently working on (in no particular order):
45 Proper (or at least usable) certificate chain verification.
46 Documentation on X509 V3 extension code.
47 PKCS#12 code cleanup and enhancement.
49 o Mark is currently working on:
50 Folding in any changes that are in the C2Net code base that were
51 not in the original SSLeay-0.9.1.b release. Plus other minor
54 o Ralf is currently working on:
55 1. Support for SSL_set_default_verify_paths(),
56 SSL_load_verify_locations(), SSL_get_cert_store() and
57 SSL_set_cert_store() functions which work like their existing
58 SSL_CTX_xxx() variants but on a per connection basis. That's needed
59 to let us provide full-featured per-URL client verification in
60 mod_ssl or Apache-SSL.
61 => It still dumps core, so I suspend this and investigate
62 again for OpenSSL 0.9.3.
63 2. The perl/ stuff to make it really work the first time ;-)
64 => I'll investigate a few more hours for OpenSSL 0.9.2
65 3. The new documentation set in POD format under doc/
66 => I'll investigate a few more hours for OpenSSL 0.9.2
67 4. More cleanups to get rid of obsolete/old/ugly files in the
68 source tree which are not really needed.
69 => Done all which were possible with my personal knowledge
71 o Ben is currently working on:
72 1. Function Prototype Thought Police issues.
73 2. Integrated documentation.
74 3. New TLS Ciphersuites.
75 4. Anything else that takes his fancy.
80 o salzr@certco.com (Rich Salz): Bug in X509_name_print
81 <29E0A6D39ABED111A36000A0C99609CA2C2BA4@macertco-srv1.ma.certco.com>
82 o Jean-Hugues ROYER <jhroyer@joher.com>: rsa_oaep.c with Watcom C
83 o $(PERL) in */Makefile.ssl
84 o "Sign the certificate?" - "n" creates empty certificate file
85 o dubious declaration of crypt() in des.h
89 o The Makefile hierarchy and build mechanism is still not a round thing:
91 1. The config vs. Configure scripts
92 It's the same nasty situation as for Apache with APACI vs.
93 src/Configure. It confuses.
94 Suggestion: Merge Configure and config into a single configure
95 script with a Autoconf style interface ;-) and remove
96 Configure and config. Or even let us use GNU Autoconf
97 itself. Then we can avoid a lot of those platform checks
98 which are currently in Configure.
100 2. The xxx.org -> xxx.h generation:
101 It's not obvious for which file xxx.org is the source.
102 Suggestion: Rename xxx.org to xxx.h.in (Autoconf style), this way
103 one sees that xxx.h.in is the input for xxx.h
107 o The installation under "make install" produces a very
108 installation layout: $prefix/certs and $prefix/private dirs. That's
109 not nice. Ralf suggests to move the two certs and private dirs either
110 to $prefix/etc/, $prefix/lib/ or $prefix/share. Alternatively
111 we could also not install the certs at all.
113 Status: Ralf +1 for both not installing the certs at all and
114 moving it to $prefix/etc/. +0 for $prefix/lib/
116 Paul: why is it not nice?
117 Ralf: because it messes up the install dir when
118 $prefix is not a dedicated area like /usr/local/ssl.
119 When we move them to a standard subdir like
120 etc/ lib/ or share/ we don't mess up things
121 when $prefix is /usr or /usr/local, etc.
122 Additionally it makes package vendors life
125 o Support for Shared Libraries has to be added at least
126 for the major Unix platforms. The details we can rip from the stuff
127 Ralf has done for the Apache src/Configure script. Ben wants the
128 solution to be really simple.
130 Status: Ralf will look how we can easily incorporate the
131 compiler PIC and linker DSO flags from Apache
132 into the OpenSSL Configure script.
134 o The perl/ stuff needs a major overhaul. Currently it's
135 totally obsolete. Either we clean it up and enhance it to be up-to-date
136 with the C code or we also could replace it with the really nice
137 Net::SSLeay package we can find under
138 http://www.neuronio.pt/SSLeay.pm.html. Ralf uses this package for a
139 longer time and it works fine and is a nice Perl module. Best would be
140 to convince the author to work for the OpenSSL project and create a
141 Net::OpenSSL or Crypt::OpenSSL package out of it and maintains it for
144 Status: Ralf thinks we should both contact the author of Net::SSLeay
145 and look how much effort it is to bring Eric's perl/ stuff up
149 o The EVP and ASN1 stuff is a mess. Currently you have one EVP_CIPHER
150 structure for each cipher. This may make sense for things like DES but
151 for variable length ciphers like RC2 and RC4 it is NBG. Need a way to
152 use the EVP interface and set up the cipher parameters. The ASN1 stuff
153 is also foo wrt ciphers whose AlgorithmIdentifier has more than just
154 an IV in it (e.g. RC2, RC5). This also means that EVP_Seal and EVP_Open
155 don't work unless the key length matches the fixed value (some vendors
156 use a key length decided by the size of the RSA encrypted key and expect
159 o Properly initialize the PRNG in the absence of /dev/random.
161 o > NO_RSA (ejs@bfd.com)
162 > ./Configure -DNO_IDEA -DNO_RC5 -DNO_RC4 -DNO_RC2 -DNO_RSA -DNO_ERR linux-elf
163 > I tried for a whole day to do this and could not get it to work. Linux
164 > machine, kernel 2.0.36 and 2.2.1, redhat 5.2 latest, gcc and egcs , no
165 > go. I also noticed the even with -DNO_IDEA, _DNO_RC2, etc. the make
166 > still goes into those subdirectories and 'makes'.
168 o The "exported" header files should not use #include "foo.h", but
169 #include <foo.h> or even better #include <ssl/foo.h> or
170 #include <openssl/foo.h> in order to avoid filename clashes.
172 o ERR_error_string(..., buf) does not know how large buf is,
173 there should be ERR_error_string_n(..., buf, bufsize)
179 "How about making the each of the locations compile-time defined. I
180 would like to (for example) put binaries in /usr/bin, configuration
181 data, certs and keys in /etc/openssl/certs and /etc/openssl/keys, etc.
182 This would also be a great boon to binary package makers. The
183 SSLeay-0.9.1b RPM already includes some patches which do some of this.
184 I can forward them if you wish."
186 o Mats Nilsson <mats.nilsson@xware.se>:
187 "Add reference counting to all substructures of X509 etc. For instance,
188 X509_NAME lacks a reference counter, while EVP_PKEY has one. I'm
189 making COM-wrappers for selected parts of SSLeay for a project of ours,
190 and has found this inconsistency in copy semantics annoying."