openssl.git
23 years agoUpdate BIO_s_connect().
Dr. Stephen Henson [Fri, 15 Sep 2000 00:28:47 +0000 (00:28 +0000)]
Update BIO_s_connect().

23 years agoSome small clarifications.
Bodo Möller [Thu, 14 Sep 2000 22:09:55 +0000 (22:09 +0000)]
Some small clarifications.

23 years agoIn the name section, all the functions described shoud be enumerated.
Richard Levitte [Thu, 14 Sep 2000 21:23:28 +0000 (21:23 +0000)]
In the name section, all the functions described shoud be enumerated.
This will also make it much simpler to generate softlinks name like
each function to man-pages containing the info.

23 years agoUpdate the info on version numbering
Richard Levitte [Thu, 14 Sep 2000 21:22:19 +0000 (21:22 +0000)]
Update the info on version numbering

23 years agoBIO_seed() and BIO_tell() were documented in two other documents,
Richard Levitte [Thu, 14 Sep 2000 20:24:56 +0000 (20:24 +0000)]
BIO_seed() and BIO_tell() were documented in two other documents,
which is redundant.  They are now in their own document.

Also, in the name section, all the functions described shoud be
enumerated.  This will also make it much simpler to generate softlinks
name like each function to man-pages containing the info.

23 years agoBIO_seed() and BIO_tell() were documented in two other documents,
Richard Levitte [Thu, 14 Sep 2000 20:23:17 +0000 (20:23 +0000)]
BIO_seed() and BIO_tell() were documented in two other documents,
which is redundant.  Instead, move them to their own page.

23 years agoItems without a =over and a =back are ignored.
Richard Levitte [Thu, 14 Sep 2000 20:22:14 +0000 (20:22 +0000)]
Items without a =over and a =back are ignored.

23 years agoBIO_s_bio() manual page detailing BIO pair.
Dr. Stephen Henson [Thu, 14 Sep 2000 18:55:39 +0000 (18:55 +0000)]
BIO_s_bio() manual page detailing BIO pair.

This combines several manual pages provided by
Lutz Jaenicke <Lutz.Jaenicke@aet.TU-Cottbus.DE>,
various comments by Bodo to the lists and a bit
of source examination by me.

23 years agoBug fix: Montgomery multiplication could produce results with the wrong
Ulf Möller [Thu, 14 Sep 2000 18:37:53 +0000 (18:37 +0000)]
Bug fix: Montgomery multiplication could produce results with the wrong
sign.

23 years agoRemove engine stuff that was erroneously put in the main trunk.
Richard Levitte [Thu, 14 Sep 2000 18:16:25 +0000 (18:16 +0000)]
Remove engine stuff that was erroneously put in the main trunk.

23 years agoAdd Damien Miller's RPM specification file with a few modifications.
Richard Levitte [Thu, 14 Sep 2000 15:28:44 +0000 (15:28 +0000)]
Add Damien Miller's RPM specification file with a few modifications.

23 years agoRemove indentation in the NAME section. There's really no need to
Richard Levitte [Thu, 14 Sep 2000 13:17:55 +0000 (13:17 +0000)]
Remove indentation in the NAME section.  There's really no need to
indent there, especially since the pod2* scripts will regard that as
preformated text.  In one case, indent a code section one step.

23 years agoAdd a number of documentation files, mostly for SSL routines, but also
Richard Levitte [Thu, 14 Sep 2000 13:11:56 +0000 (13:11 +0000)]
Add a number of documentation files, mostly for SSL routines, but also
for a few BIO routines.
Submitted by Lutz Jaenicke <Lutz.Jaenicke@aet.TU-Cottbus.DE>

23 years agoAdd a configuration for Sony News 4.
Richard Levitte [Thu, 14 Sep 2000 12:48:48 +0000 (12:48 +0000)]
Add a configuration for Sony News 4.
Submitted by NAKAJI Hiroyuki <nakaji@tutrp.tut.ac.jp>

23 years agoInitial connect BIO docs.
Dr. Stephen Henson [Thu, 14 Sep 2000 12:44:34 +0000 (12:44 +0000)]
Initial connect BIO docs.

23 years agoRemove indentation in the NAME section. There's really no need to
Richard Levitte [Thu, 14 Sep 2000 12:14:41 +0000 (12:14 +0000)]
Remove indentation in the NAME section.  There's really no need to
indent there, especially since the pod2* scripts will regard that as
preformated text.  In one case, indent a code section one step.

23 years agorsa_num2 is no longer used, so remove it.
Richard Levitte [Thu, 14 Sep 2000 11:09:03 +0000 (11:09 +0000)]
rsa_num2 is no longer used, so remove it.

23 years agoJeffrey Altman convinced me this patch was really needed, or there is
Richard Levitte [Thu, 14 Sep 2000 09:19:19 +0000 (09:19 +0000)]
Jeffrey Altman convinced me this patch was really needed, or there is
no way to make sure GetCursorInfo will give us a valid answer.

23 years agoDocs for socket BIO.
Dr. Stephen Henson [Thu, 14 Sep 2000 00:05:50 +0000 (00:05 +0000)]
Docs for socket BIO.

23 years ago'make update'
Bodo Möller [Wed, 13 Sep 2000 22:04:09 +0000 (22:04 +0000)]
'make update'

23 years agoBIO_s_fd() manual page.
Dr. Stephen Henson [Wed, 13 Sep 2000 17:27:42 +0000 (17:27 +0000)]
BIO_s_fd() manual page.

23 years agoPoint to Peter Gutmann's revised paper.
Ulf Möller [Wed, 13 Sep 2000 14:24:07 +0000 (14:24 +0000)]
Point to Peter Gutmann's revised paper.
The copy at www.usenix.org is the old version.

23 years agoOne more passed test
Richard Levitte [Wed, 13 Sep 2000 12:14:39 +0000 (12:14 +0000)]
One more passed test

23 years agoUpdate info on what has been fixed
Richard Levitte [Wed, 13 Sep 2000 11:35:54 +0000 (11:35 +0000)]
Update info on what has been fixed

23 years agoUpdate info on what has been fixed, and switch format for failure data
Richard Levitte [Wed, 13 Sep 2000 11:33:09 +0000 (11:33 +0000)]
Update info on what has been fixed, and switch format for failure data

23 years agoMake sure that Configure will defined DSO_WIN32 for the Win32 targets. I feel a...
Richard Levitte [Wed, 13 Sep 2000 11:29:15 +0000 (11:29 +0000)]
Make sure that Configure will defined DSO_WIN32 for the Win32 targets.  I feel a bit unsure if this should really be done for Mingw32 and CygWin32

23 years ago3 changes:
Richard Levitte [Wed, 13 Sep 2000 07:17:35 +0000 (07:17 +0000)]
3 changes:
- Make sure PCURSORINFO is defined even on systems that do not provide it.
- Change the reference to Peter Gutmann's paper.
- Make sure we don't walk the whole heap lists for performance reasons.
  Jeffrey Altman suggests following Peter Gutmann's advice to keep it
  to 50 heap entries per heap list.

23 years agoThe other log message should have read "Note the DSA change".
Ulf Möller [Wed, 13 Sep 2000 02:01:35 +0000 (02:01 +0000)]
The other log message should have read "Note the DSA change".

23 years agoMore Windows failures reported
Ulf Möller [Wed, 13 Sep 2000 01:50:24 +0000 (01:50 +0000)]
More Windows failures reported

23 years agoNot the DSA change.
Ulf Möller [Wed, 13 Sep 2000 01:48:05 +0000 (01:48 +0000)]
Not the DSA change.

23 years agoDon't set the two top bits to one when generating a random number < q.:wq
Ulf Möller [Wed, 13 Sep 2000 01:45:54 +0000 (01:45 +0000)]
Don't set the two top bits to one when generating a random number < q.:wq

23 years agoClarify some of the I/O issues.
Dr. Stephen Henson [Wed, 13 Sep 2000 00:20:24 +0000 (00:20 +0000)]
Clarify some of the I/O issues.

Add case of using select() and blocking I/O with
BIOs and why you shouldn't (thanks Bodo!).

23 years agoA few more systems reported successfull.
Richard Levitte [Tue, 12 Sep 2000 22:19:41 +0000 (22:19 +0000)]
A few more systems reported successfull.

23 years agoNew SSL API mode 'SSL_MODE_AUTO_RETRY', which disables the default
Bodo Möller [Tue, 12 Sep 2000 20:28:30 +0000 (20:28 +0000)]
New SSL API mode 'SSL_MODE_AUTO_RETRY', which disables the default
behaviour that SSL_read may result in SSL_ERROR_WANT_READ.

23 years agobug: RAND_poll().
Ulf Möller [Tue, 12 Sep 2000 16:40:59 +0000 (16:40 +0000)]
bug: RAND_poll().

23 years agoHolger Reif reports a few more Solaris successes.
Richard Levitte [Tue, 12 Sep 2000 15:46:04 +0000 (15:46 +0000)]
Holger Reif reports a few more Solaris successes.

23 years agoNote the failure on Win32
Richard Levitte [Tue, 12 Sep 2000 10:07:19 +0000 (10:07 +0000)]
Note the failure on Win32

23 years agoFreeBSD and solaris with gcc passed
Richard Levitte [Tue, 12 Sep 2000 10:05:11 +0000 (10:05 +0000)]
FreeBSD and solaris with gcc passed

23 years agoBetter error checking for RSA and DSA signature and verification speed
Richard Levitte [Tue, 12 Sep 2000 08:12:52 +0000 (08:12 +0000)]
Better error checking for RSA and DSA signature and verification speed
tests.  This was required to not get mysterious errors when they
wouldn't quite want to work.

23 years agoLinux in Sparc v7 passed
Richard Levitte [Tue, 12 Sep 2000 06:49:03 +0000 (06:49 +0000)]
Linux in Sparc v7 passed

23 years agoSCO 5.0.5 with both gcc and cc passed
Richard Levitte [Tue, 12 Sep 2000 06:44:52 +0000 (06:44 +0000)]
SCO 5.0.5 with both gcc and cc passed

23 years agoMore BIO docs.
Dr. Stephen Henson [Tue, 12 Sep 2000 01:56:56 +0000 (01:56 +0000)]
More BIO docs.

23 years agoDSA_verify() and DSA_sign() might return -1...
Richard Levitte [Mon, 11 Sep 2000 22:21:38 +0000 (22:21 +0000)]
DSA_verify() and DSA_sign() might return -1...

23 years agoA couple more HP-UX targets tested.
Richard Levitte [Mon, 11 Sep 2000 22:02:07 +0000 (22:02 +0000)]
A couple more HP-UX targets tested.

23 years agodebug-linux-elf and debug-linux-elf-efence need to be linked with
Richard Levitte [Mon, 11 Sep 2000 21:57:27 +0000 (21:57 +0000)]
debug-linux-elf and debug-linux-elf-efence need to be linked with
libdl just as linux-elf...

23 years agoFailure on Solaris when using the CSwift card.
Richard Levitte [Mon, 11 Sep 2000 20:32:35 +0000 (20:32 +0000)]
Failure on Solaris when using the CSwift card.

23 years agomkdef.pl still needed better logic. Also, the semantics of the
Richard Levitte [Mon, 11 Sep 2000 20:04:58 +0000 (20:04 +0000)]
mkdef.pl still needed better logic.  Also, the semantics of the
platforms list is clarified (it's however not quite followed in the
RSAREF case...).

RSAREF is also checked now.

23 years agoDocument an old change.
Ben Laurie [Mon, 11 Sep 2000 17:58:09 +0000 (17:58 +0000)]
Document an old change.

23 years agomkdef.pl has erroneous conditions to check if a symbol is excluded
Richard Levitte [Mon, 11 Sep 2000 17:31:05 +0000 (17:31 +0000)]
mkdef.pl has erroneous conditions to check if a symbol is excluded
from the given target.  Fixed, I hope.

23 years agolinux-elf passed
Richard Levitte [Mon, 11 Sep 2000 17:10:21 +0000 (17:10 +0000)]
linux-elf passed

23 years agoOpenBSD doesn't support timeb.
Richard Levitte [Mon, 11 Sep 2000 16:46:35 +0000 (16:46 +0000)]
OpenBSD doesn't support timeb.

23 years agoDon't include e_os.h before the system headers
Richard Levitte [Mon, 11 Sep 2000 16:37:36 +0000 (16:37 +0000)]
Don't include e_os.h before the system headers

23 years agoReports for OpenBSD 2.7 and HP-UX 10.20
Richard Levitte [Mon, 11 Sep 2000 16:36:14 +0000 (16:36 +0000)]
Reports for OpenBSD 2.7 and HP-UX 10.20

23 years agoLast minute update, in time to make it to 0.9.6-beta1
Richard Levitte [Mon, 11 Sep 2000 13:06:48 +0000 (13:06 +0000)]
Last minute update, in time to make it to 0.9.6-beta1

23 years agoTime to release a beta. Change the status accordingly.
Richard Levitte [Mon, 11 Sep 2000 12:39:43 +0000 (12:39 +0000)]
Time to release a beta.  Change the status accordingly.

23 years agoTime to release a beta. Change the version numbers and dates
Richard Levitte [Mon, 11 Sep 2000 12:31:36 +0000 (12:31 +0000)]
Time to release a beta.  Change the version numbers and dates
accordingly.

23 years agoI started with a make update, but a rewrite was actually needed.
Richard Levitte [Mon, 11 Sep 2000 11:45:02 +0000 (11:45 +0000)]
I started with a make update, but a rewrite was actually needed.
Perhaps we should make rewrites the default thing to do?

23 years agoA cast is needed or Borland C will complain.
Richard Levitte [Mon, 11 Sep 2000 11:43:35 +0000 (11:43 +0000)]
A cast is needed or Borland C will complain.

23 years agoDocs for cipher and base64 BIOs.
Dr. Stephen Henson [Mon, 11 Sep 2000 01:04:09 +0000 (01:04 +0000)]
Docs for cipher and base64 BIOs.

23 years agoMore new BIO docs, correct some old ones.
Dr. Stephen Henson [Sun, 10 Sep 2000 17:36:15 +0000 (17:36 +0000)]
More new BIO docs, correct some old ones.

23 years agoMarin Kraemer <Martin.Kraemer@MchP.Siemens.De> sent us patches to make
Richard Levitte [Sun, 10 Sep 2000 14:45:19 +0000 (14:45 +0000)]
Marin Kraemer <Martin.Kraemer@MchP.Siemens.De> sent us patches to make
the OpenSSL commands x50 and req work better on a EBCDIC system.

23 years agoMore preliminary BIO docs...
Dr. Stephen Henson [Sun, 10 Sep 2000 01:52:26 +0000 (01:52 +0000)]
More preliminary BIO docs...

Incomplete and possibly inaccurate. Hope somone is
checking these :-)

23 years agoMore VMS synchronisation
Richard Levitte [Sat, 9 Sep 2000 18:05:27 +0000 (18:05 +0000)]
More VMS synchronisation

23 years agoFix some CygWin problems.
Ulf Möller [Sat, 9 Sep 2000 04:45:18 +0000 (04:45 +0000)]
Fix some CygWin problems.

23 years agoReally add BIO_read this time...
Dr. Stephen Henson [Sat, 9 Sep 2000 01:01:35 +0000 (01:01 +0000)]
Really add BIO_read this time...

23 years agoAdd BIO_read() (etc.) docs.
Dr. Stephen Henson [Sat, 9 Sep 2000 00:59:37 +0000 (00:59 +0000)]
Add BIO_read() (etc.) docs.

Add an ASN1 FAQ because I'm sick of answering it :-)

23 years agoClarify how one should behave when make fails. The fault is not
Richard Levitte [Fri, 8 Sep 2000 22:19:27 +0000 (22:19 +0000)]
Clarify how one should behave when make fails.  The fault is not
necessarely ours.

23 years agoSynchronise the VMS build with the Unix one.
Richard Levitte [Fri, 8 Sep 2000 20:25:49 +0000 (20:25 +0000)]
Synchronise the VMS build with the Unix one.

23 years agoTwo places where I forgot to change vms_idhacks to symhacks.
Richard Levitte [Fri, 8 Sep 2000 06:28:09 +0000 (06:28 +0000)]
Two places where I forgot to change vms_idhacks to symhacks.

23 years agoUpdate verify docs.
Dr. Stephen Henson [Fri, 8 Sep 2000 00:53:58 +0000 (00:53 +0000)]
Update verify docs.

New option to verify program to print out diagnostics.

23 years agoTwo new PKCS#12 demo programs.
Dr. Stephen Henson [Thu, 7 Sep 2000 23:14:26 +0000 (23:14 +0000)]
Two new PKCS#12 demo programs.

Update PKCS12_parse().

Make the keyid in certificate aux info more usable.

23 years agoUgh, BIO_find_type() cannot be passed a NULL.
Dr. Stephen Henson [Thu, 7 Sep 2000 17:42:25 +0000 (17:42 +0000)]
Ugh, BIO_find_type() cannot be passed a NULL.

Fix doc example, and fix BIO_find_type().

Fix PKCS7_verify(). It was using 'i' for both the
loop variable and the verify return value.

23 years agoAdd docs for BIO_find_type() and friends.
Dr. Stephen Henson [Thu, 7 Sep 2000 13:04:27 +0000 (13:04 +0000)]
Add docs for BIO_find_type() and friends.

Added function BIO_next() otherwise you can't
traverse a chain without accessing BIO internals.

23 years ago'make update'
Richard Levitte [Thu, 7 Sep 2000 08:46:51 +0000 (08:46 +0000)]
'make update'

23 years ago*.num rewitten to include the extra information.
Richard Levitte [Thu, 7 Sep 2000 08:44:13 +0000 (08:44 +0000)]
*.num rewitten to include the extra information.

23 years agoMajor hack of mkdef.pl. There should be no more need to redo the
Richard Levitte [Thu, 7 Sep 2000 08:43:08 +0000 (08:43 +0000)]
Major hack of mkdef.pl.  There should be no more need to redo the
process when some symbols are missing.  Instead, all needed info is
saved in the .num files, including what conditions are needed for a
specific symbol to exist.

This was needed for the work I'm doing with shared libraries under
VMS.

23 years agoIt's not just VMS that needs some symbols to be hacked. Let's
Richard Levitte [Thu, 7 Sep 2000 08:18:05 +0000 (08:18 +0000)]
It's not just VMS that needs some symbols to be hacked.  Let's
centralise those hacks in crypto/symhacks.h and use it everywhere it's
needed.

23 years agoAnother thing I'm working on.
Richard Levitte [Thu, 7 Sep 2000 08:14:46 +0000 (08:14 +0000)]
Another thing I'm working on.

23 years agoclarification (source/sink BIOs are usually *both* source and sink)
Bodo Möller [Thu, 7 Sep 2000 08:07:55 +0000 (08:07 +0000)]
clarification (source/sink BIOs are usually *both* source and sink)

23 years agoChange the printing mahine used by BIO_printf() and friends so it can
Richard Levitte [Thu, 7 Sep 2000 05:50:14 +0000 (05:50 +0000)]
Change the printing mahine used by BIO_printf() and friends so it can
handle an externally provided "static" buffer as well a a dynamic
buffer.  The "static" buffer is filled first, but if overflowed, the
dynamic buffer is used instead, being allocated somewhere i the heap.

This combines the benefits of putting the output in a preallocated
buffer (on the stack, for example) and in a buffer that grows
somewhere in the heap.

23 years agoSome BIO docs, incomplete, more to follow.
Dr. Stephen Henson [Thu, 7 Sep 2000 00:22:31 +0000 (00:22 +0000)]
Some BIO docs, incomplete, more to follow.

Hmmm I didn't realise BIO_pop() did that:
isn't source wonderful?

23 years agoUse name ...-whatever-solaris2 instead of ...-sun-solaris2
Bodo Möller [Wed, 6 Sep 2000 17:09:58 +0000 (17:09 +0000)]
Use name ...-whatever-solaris2 instead of ...-sun-solaris2
(the middle string describes the architecture).

23 years agoGet rid of ASN1_UTCTIME_get, which cannot work with time_t
Bodo Möller [Wed, 6 Sep 2000 15:40:52 +0000 (15:40 +0000)]
Get rid of ASN1_UTCTIME_get, which cannot work with time_t
return type (on platforms where time_t is a 32 bit value).

New function ASN1_UTCTIME_cmp_time_t as a replacement
for use in apps/x509.c.

23 years agoSee RSA Security's press release at
Bodo Möller [Wed, 6 Sep 2000 14:55:11 +0000 (14:55 +0000)]
See RSA Security's press release at
http://www.rsasecurity.com/news/pr/000906-1.html (September 6, 2000):
"RSA Security Releases RSA Encryption Algorithm into Public Domain"

23 years agoAnother superfluous pair of parentheses.
Bodo Möller [Wed, 6 Sep 2000 14:34:32 +0000 (14:34 +0000)]
Another superfluous pair of parentheses.

23 years agoAnother round of indentation changes: Position braces consistently,
Bodo Möller [Wed, 6 Sep 2000 14:14:20 +0000 (14:14 +0000)]
Another round of indentation changes: Position braces consistently,
add some whitespace for 'if ()', 'for ()', 'while ()' to distinguish
keywords from function names, and finally remove parens around return
values (why be stingy with whitespace but fill the source code
with an abundance of parentheses that are not needed to structure
expressions for readability?).

23 years agoMore indentation consistency: for (), while (), if (), return ()
Bodo Möller [Wed, 6 Sep 2000 13:31:44 +0000 (13:31 +0000)]
More indentation consistency: for (), while (), if (), return ()
usually get a space between keyword and opening paranthesis
so that they don't look like function calls, where no space is
used.

23 years agoFix X509_STORE_CTX_init. Make indentation more consistent. Dump core less often.
Bodo Möller [Wed, 6 Sep 2000 13:24:10 +0000 (13:24 +0000)]
Fix X509_STORE_CTX_init.  Make indentation more consistent.  Dump core less often.

23 years ago'make update'
Bodo Möller [Wed, 6 Sep 2000 12:34:10 +0000 (12:34 +0000)]
'make update'

23 years agoChanges for QNX: there is no thread support, and the previous
Bodo Möller [Wed, 6 Sep 2000 12:25:58 +0000 (12:25 +0000)]
Changes for QNX: there is no thread support, and the previous
configuration only worked with no-asm.

23 years agotypo
Bodo Möller [Wed, 6 Sep 2000 12:18:24 +0000 (12:18 +0000)]
typo

23 years agoAdd OAEP. Seed the PRNG.
Bodo Möller [Wed, 6 Sep 2000 11:49:43 +0000 (11:49 +0000)]
Add OAEP. Seed the PRNG.

23 years agoupdate
Bodo Möller [Wed, 6 Sep 2000 10:50:33 +0000 (10:50 +0000)]
update

23 years agoAdd rsautl.
Bodo Möller [Wed, 6 Sep 2000 07:58:27 +0000 (07:58 +0000)]
Add rsautl.

23 years agoClarification for SSL_ERROR_ZERO_RETURN
Bodo Möller [Wed, 6 Sep 2000 07:56:03 +0000 (07:56 +0000)]
Clarification for SSL_ERROR_ZERO_RETURN

23 years agoFix typo in rsautl.
Dr. Stephen Henson [Tue, 5 Sep 2000 22:30:38 +0000 (22:30 +0000)]
Fix typo in rsautl.

Add support for settable verify time in X509_verify_cert().

Document rsautl utility.

23 years agoIgnore executable.
Ben Laurie [Tue, 5 Sep 2000 18:56:55 +0000 (18:56 +0000)]
Ignore executable.

23 years agoHandle WANT_READ more correctly (thanks to Bodo).
Ben Laurie [Tue, 5 Sep 2000 18:47:57 +0000 (18:47 +0000)]
Handle WANT_READ more correctly (thanks to Bodo).

23 years ago*BIG* verify code reorganisation.
Dr. Stephen Henson [Tue, 5 Sep 2000 17:53:58 +0000 (17:53 +0000)]
*BIG* verify code reorganisation.

The old code was painfully primitive and couldn't handle
distinct certificates using the same subject name.

The new code performs several tests on a candidate issuer
certificate based on certificate extensions.

It also adds several callbacks to X509_VERIFY_CTX so its
behaviour can be customised.

Unfortunately some hackery was needed to persuade X509_STORE
to tolerate this. This should go away when X509_STORE is
replaced, sometime...

This must have broken something though :-(

23 years agoDistinguish between assertions and conditions that should cause death.
Ben Laurie [Tue, 5 Sep 2000 17:06:45 +0000 (17:06 +0000)]
Distinguish between assertions and conditions that should cause death.