openssl.git
19 years agocorrections
Ulf Möller [Sun, 27 Feb 2000 18:35:29 +0000 (18:35 +0000)]
corrections

19 years agoWorks with gcc 2.8.0 as well.
Richard Levitte [Sun, 27 Feb 2000 17:56:54 +0000 (17:56 +0000)]
Works with gcc 2.8.0 as well.

19 years agoNo, the VAX is not a 64 bit architecture.
Richard Levitte [Sun, 27 Feb 2000 17:43:39 +0000 (17:43 +0000)]
No, the VAX is not a 64 bit architecture.

19 years agoUse standard header file string.h for memset prototype (where
Bodo Möller [Sun, 27 Feb 2000 17:37:41 +0000 (17:37 +0000)]
Use standard header file string.h for memset prototype (where
"standard" refers to the C language, probably there's also some
standard that defines memory.h).

19 years agoDon't call BN_rand with zero bits in bntest.c
Dr. Stephen Henson [Sun, 27 Feb 2000 17:34:30 +0000 (17:34 +0000)]
Don't call BN_rand with zero bits in bntest.c

19 years agoMake sure to catch UnixWare 7, even if the version is 7.1.1 or
Richard Levitte [Sun, 27 Feb 2000 17:23:25 +0000 (17:23 +0000)]
Make sure to catch UnixWare 7, even if the version is 7.1.1 or
something like that.

19 years agoUpdate config for the unixware targets by looking at what was done in
Richard Levitte [Sun, 27 Feb 2000 17:17:43 +0000 (17:17 +0000)]
Update config for the unixware targets by looking at what was done in
Apache's GuessOS (from Apache 1.3.1).

19 years agoDeclare memset.
Ben Laurie [Sun, 27 Feb 2000 17:01:20 +0000 (17:01 +0000)]
Declare memset.

19 years agoTypo.
Ben Laurie [Sun, 27 Feb 2000 16:56:48 +0000 (16:56 +0000)]
Typo.

19 years agoAIX and HP-UX are passing the tests.
Richard Levitte [Sun, 27 Feb 2000 16:42:55 +0000 (16:42 +0000)]
AIX and HP-UX are passing the tests.

Warning about old gcc, though (and I recall getting in trouble with
gcc 2.8.0.  I'll try that again).

19 years agoOpenSSL doesn't compile well on OpenVMS/VAX, it seems.
Richard Levitte [Sun, 27 Feb 2000 16:11:06 +0000 (16:11 +0000)]
OpenSSL doesn't compile well on OpenVMS/VAX, it seems.

I'm looking into it, it's probably a simple script error.

19 years agoOne test passed for VMS.
Richard Levitte [Sun, 27 Feb 2000 15:58:09 +0000 (15:58 +0000)]
One test passed for VMS.

19 years agoA new beta has been released, and one test is documented.
Richard Levitte [Sun, 27 Feb 2000 14:44:43 +0000 (14:44 +0000)]
A new beta has been released, and one test is documented.

19 years agoOpenVMS often has problems with files with more than on period in it.
Richard Levitte [Sun, 27 Feb 2000 14:03:58 +0000 (14:03 +0000)]
OpenVMS often has problems with files with more than on period in it.
Here's instructions on what to do if you get into trouble because of
that.

19 years agoremove obsolete BN_CTX info
Ulf Möller [Sun, 27 Feb 2000 12:30:07 +0000 (12:30 +0000)]
remove obsolete BN_CTX info

19 years agoAdd a comment.
Bodo Möller [Sun, 27 Feb 2000 11:48:32 +0000 (11:48 +0000)]
Add a comment.

19 years agoFor lack of a better name, this is now called 0.9.5beta3-dev until the
Richard Levitte [Sun, 27 Feb 2000 11:12:58 +0000 (11:12 +0000)]
For lack of a better name, this is now called 0.9.5beta3-dev until the
release.

19 years agoChange version string to reflect the release of beta 2. OpenSSL_0_9_5beta2
Richard Levitte [Sun, 27 Feb 2000 11:07:23 +0000 (11:07 +0000)]
Change version string to reflect the release of beta 2.

19 years agoStop logging all the files that are copied all over the place...
Richard Levitte [Sun, 27 Feb 2000 11:04:09 +0000 (11:04 +0000)]
Stop logging all the files that are copied all over the place...

19 years agoNew logical names to skip algorithms are now supported.
Richard Levitte [Sun, 27 Feb 2000 10:41:31 +0000 (10:41 +0000)]
New logical names to skip algorithms are now supported.

19 years agoNew logical names to skip algorithms are now supported.
Richard Levitte [Sun, 27 Feb 2000 10:39:41 +0000 (10:39 +0000)]
New logical names to skip algorithms are now supported.
Also, on Alpha, vms.mar is no longer used (it produced a lot of
confusing messages) for BN.  Instead, the assembler part of bn_lcl.h
is used.

19 years agoNew "target": CONFIG. This will build the opensslconf.h file from
Richard Levitte [Sun, 27 Feb 2000 10:35:15 +0000 (10:35 +0000)]
New "target": CONFIG.  This will build the opensslconf.h file from
what is known about VAX and Alpha running VMS, and from the
opensslconf.h.in (in VMS often named OPENSSLCONF.H_IN) file.

19 years agoAdd and change docs for the changes that have been made.
Richard Levitte [Sun, 27 Feb 2000 10:26:35 +0000 (10:26 +0000)]
Add and change docs for the changes that have been made.

19 years agoput function names in the title.
Ulf Möller [Sun, 27 Feb 2000 02:38:41 +0000 (02:38 +0000)]
put function names in the title.

19 years agoStatement that it fails only on 32-bit architectures isn't true.
Andy Polyakov [Sun, 27 Feb 2000 02:34:37 +0000 (02:34 +0000)]
Statement that it fails only on 32-bit architectures isn't true.

19 years agoOverly long lines look ugly in the DOS editor. :)
Ulf Möller [Sun, 27 Feb 2000 02:19:48 +0000 (02:19 +0000)]
Overly long lines look ugly in the DOS editor. :)

19 years agoBug fix!
Ulf Möller [Sun, 27 Feb 2000 02:05:39 +0000 (02:05 +0000)]
Bug fix!

19 years agoFix so Win32 compiles again...
Dr. Stephen Henson [Sun, 27 Feb 2000 01:39:41 +0000 (01:39 +0000)]
Fix so Win32 compiles again...

19 years agoFix so Win32 assembly language works with MASM.
Dr. Stephen Henson [Sun, 27 Feb 2000 01:15:25 +0000 (01:15 +0000)]
Fix so Win32 assembly language works with MASM.

Add info about where to get MASM.

19 years ago*** empty log message ***
Ulf Möller [Sun, 27 Feb 2000 00:00:56 +0000 (00:00 +0000)]
*** empty log message ***

19 years agoworkaround no longer needed
Ulf Möller [Sat, 26 Feb 2000 22:37:43 +0000 (22:37 +0000)]
workaround no longer needed

19 years agoremove
Ulf Möller [Sat, 26 Feb 2000 22:37:34 +0000 (22:37 +0000)]
remove

19 years agoReorganize bn_mul.c (no bugfix yet), remove obsolete files in BN library.
Ulf Möller [Sat, 26 Feb 2000 22:16:47 +0000 (22:16 +0000)]
Reorganize bn_mul.c (no bugfix yet), remove obsolete files in BN library.

19 years agominor clarification
Ulf Möller [Sat, 26 Feb 2000 22:06:02 +0000 (22:06 +0000)]
minor clarification

19 years agoMake ASN1 types real typedefs.
Dr. Stephen Henson [Sat, 26 Feb 2000 19:25:31 +0000 (19:25 +0000)]
Make ASN1 types real typedefs.

Rebuild error files.

19 years agoBN_div_recp fix. I've ran divtest for 10 mins and it didn't exhibit a
Andy Polyakov [Sat, 26 Feb 2000 16:20:05 +0000 (16:20 +0000)]
BN_div_recp fix. I've ran divtest for 10 mins and it didn't exhibit a
single fault:-) Needless to mention that bnbug.c posted couple of days
ago passes as well...

19 years agoA small script to compile mttest.c on VMS as well.
Richard Levitte [Sat, 26 Feb 2000 11:17:33 +0000 (11:17 +0000)]
A small script to compile mttest.c on VMS as well.

19 years agoMake sure there's some entropy, and log a few more errors.
Richard Levitte [Sat, 26 Feb 2000 11:02:31 +0000 (11:02 +0000)]
Make sure there's some entropy, and log a few more errors.

19 years agoMore get0 et al. changes. Also provide fgrep targets in CHANGES
Bodo Möller [Sat, 26 Feb 2000 08:36:46 +0000 (08:36 +0000)]
More get0 et al. changes.  Also provide fgrep targets in CHANGES
where the new functions are mentioned.

19 years agomake update
Richard Levitte [Sat, 26 Feb 2000 04:48:22 +0000 (04:48 +0000)]
make update

19 years agoMake sure that all test files are gone before starting the tests, or
Richard Levitte [Sat, 26 Feb 2000 03:59:46 +0000 (03:59 +0000)]
Make sure that all test files are gone before starting the tests, or
backup will complain about some version not existing.  Perhaps.

19 years agoMove to using the same perl code as Makefile.ssl
Richard Levitte [Sat, 26 Feb 2000 03:58:36 +0000 (03:58 +0000)]
Move to using the same perl code as Makefile.ssl

19 years agoJust as in Unix, make sure to generate some kind of "random data".
Richard Levitte [Sat, 26 Feb 2000 03:57:35 +0000 (03:57 +0000)]
Just as in Unix, make sure to generate some kind of "random data".

19 years agoTypo corrected.
Richard Levitte [Sat, 26 Feb 2000 03:56:09 +0000 (03:56 +0000)]
Typo corrected.

19 years agoDon't include sys/types.h if NO_SYS_TYPES_H is defined.
Richard Levitte [Sat, 26 Feb 2000 03:55:15 +0000 (03:55 +0000)]
Don't include sys/types.h if NO_SYS_TYPES_H is defined.

19 years agoThe OpenVMS library is most definitely not built for anything but
Richard Levitte [Sat, 26 Feb 2000 03:53:58 +0000 (03:53 +0000)]
The OpenVMS library is most definitely not built for anything but
files, unless it's all in unixly syntax.  We can't guarantee that
right now, so let's skip the whole test suit.  There are other places
(like the open()) where errors are detected anyway.

19 years agoAdd support for Unixware 7. Thanks to Ron Record <rr@sco.com> for the
Richard Levitte [Sat, 26 Feb 2000 03:51:55 +0000 (03:51 +0000)]
Add support for Unixware 7.  Thanks to Ron Record <rr@sco.com> for the
patch.

19 years agoMake clear which naming convention is meant.
Ulf Möller [Sat, 26 Feb 2000 02:24:16 +0000 (02:24 +0000)]
Make clear which naming convention is meant.

19 years agoFix shadow warning.
Dr. Stephen Henson [Sat, 26 Feb 2000 02:16:36 +0000 (02:16 +0000)]
Fix shadow warning.

19 years agoftruncate() and fileno() are not supported on all versions of OpenVMS
Richard Levitte [Sat, 26 Feb 2000 01:55:56 +0000 (01:55 +0000)]
ftruncate() and fileno() are not supported on all versions of OpenVMS
(they don't really exist before version 7), so that solution was toast.
Instead, let's do it the way it's done on Unix, but then remove older
versions of the file.

That new mechanism *may* fail for some unixly formated file spec,
although I wouldn't worry too much about it.

19 years agoRename functions for new convention.
Dr. Stephen Henson [Sat, 26 Feb 2000 01:55:33 +0000 (01:55 +0000)]
Rename functions for new convention.

19 years agoLet's avoid compiler warnings over types.
Richard Levitte [Sat, 26 Feb 2000 01:45:17 +0000 (01:45 +0000)]
Let's avoid compiler warnings over types.

Actually, this should be resolved by using size_t a bit more, but
that'll be for later.

19 years agoCompile rand_egd.c on VMS as well.
Richard Levitte [Sat, 26 Feb 2000 00:29:33 +0000 (00:29 +0000)]
Compile rand_egd.c on VMS as well.

19 years agoFor safety, check __VMS as well.
Richard Levitte [Sat, 26 Feb 2000 00:26:17 +0000 (00:26 +0000)]
For safety, check __VMS as well.

19 years agoCorrect small typo
Richard Levitte [Sat, 26 Feb 2000 00:18:48 +0000 (00:18 +0000)]
Correct small typo

19 years agoSince a stack with quite long name is declared here, vms_idhacks.h is
Richard Levitte [Sat, 26 Feb 2000 00:01:16 +0000 (00:01 +0000)]
Since a stack with quite long name is declared here, vms_idhacks.h is
also needed to get around C compilers on VMS that set the symbol limit
to 31 characters.  Bot the macros VMS and __VMS are checked, since
there's no real way to know if e_os.h has been included yet.

19 years agoSmall error fix.
Richard Levitte [Fri, 25 Feb 2000 23:33:16 +0000 (23:33 +0000)]
Small error fix.

19 years agoAdjust all the old scripts to deal with the new location.
Richard Levitte [Fri, 25 Feb 2000 23:18:27 +0000 (23:18 +0000)]
Adjust all the old scripts to deal with the new location.

19 years agoNew script to compile on systems that already have pthreads in the
Richard Levitte [Fri, 25 Feb 2000 23:17:46 +0000 (23:17 +0000)]
New script to compile on systems that already have pthreads in the
system.

19 years agoUpdate the two threads modules to conform to our recommendations (use
Richard Levitte [Fri, 25 Feb 2000 23:16:56 +0000 (23:16 +0000)]
Update the two threads modules to conform to our recommendations (use
CRYPTO_num_locks() instead of CRYPTO_NUM_LOCKS!), and correct all the
inconsistencies with the rest of OpenSSL.

At least, this compiles nicely on Linux using PTHREADS.  I've done no
other tests so far.

19 years agoSome time in history, SSL_CTX_sessions() disappeared. It is now
Richard Levitte [Fri, 25 Feb 2000 22:50:28 +0000 (22:50 +0000)]
Some time in history, SSL_CTX_sessions() disappeared.  It is now
restored, but not as a macro this time...

19 years agoflush output.
Ulf Möller [Fri, 25 Feb 2000 22:08:37 +0000 (22:08 +0000)]
flush output.
Looks like it fails when b is a power of 2, but I never get incorrect
results.

19 years agoAnother test passed
Richard Levitte [Fri, 25 Feb 2000 20:46:09 +0000 (20:46 +0000)]
Another test passed

19 years agoEnhance consistency by using BIO_flush() instead of fflush().
Richard Levitte [Fri, 25 Feb 2000 20:40:12 +0000 (20:40 +0000)]
Enhance consistency by using BIO_flush() instead of fflush().

19 years agoChanges to synchronise with Unix.
Richard Levitte [Fri, 25 Feb 2000 20:37:46 +0000 (20:37 +0000)]
Changes to synchronise with Unix.
(actually, much more is needed, like a real config script)

19 years agoTest the division functions.
Ulf Möller [Fri, 25 Feb 2000 20:28:54 +0000 (20:28 +0000)]
Test the division functions.

Apparently BN_div_recp reports an error for small divisors
(1,2,4,8,40).

I haven't got mismatches so far. If you can, please run the test
program for a few days (nohup divtest >out& or something), and if it
reports a mismatch, post the output.

19 years agoStatus update
Richard Levitte [Fri, 25 Feb 2000 20:21:46 +0000 (20:21 +0000)]
Status update

19 years agoLinux is almost the only one where getpid() == thread-id. IRIX is
Andy Polyakov [Fri, 25 Feb 2000 17:59:42 +0000 (17:59 +0000)]
Linux is almost the only one where getpid() == thread-id. IRIX is
another one, but only if you stick to sproc(2).

19 years agoUse unsigned loop index to make compilers happy
Bodo Möller [Fri, 25 Feb 2000 16:36:07 +0000 (16:36 +0000)]
Use unsigned loop index to make compilers happy

19 years agoThe main() return value is a program's exit code.
Ulf Möller [Fri, 25 Feb 2000 16:04:52 +0000 (16:04 +0000)]
The main() return value is a program's exit code.

19 years agolinks
Ulf Möller [Fri, 25 Feb 2000 16:00:24 +0000 (16:00 +0000)]
links

19 years agoThere're two (incompatible) ways to write multi-threaded programs under
Andy Polyakov [Fri, 25 Feb 2000 15:44:03 +0000 (15:44 +0000)]
There're two (incompatible) ways to write multi-threaded programs under
IRIX, one is to call sproc(2) when every thread does get own pid and
POSIX threads when all the threads share same pid.

19 years agomention RAND_egd()
Ulf Möller [Fri, 25 Feb 2000 15:43:33 +0000 (15:43 +0000)]
mention RAND_egd()

19 years agoThe previous revision should have generated _more_ warnings, not less ...
Bodo Möller [Fri, 25 Feb 2000 15:32:36 +0000 (15:32 +0000)]
The previous revision should have generated _more_ warnings, not less ...
The return value of handshake_func is signed, not unsigned.

19 years agoMore VC++ pickiness. (destest.c doesn't have a "return" and the usual
Geoff Thorpe [Fri, 25 Feb 2000 15:09:04 +0000 (15:09 +0000)]
More VC++ pickiness. (destest.c doesn't have a "return" and the usual
signed/unsigned stuff in s3_pkt.c)

Submitted by:
Reviewed by:
PR:

19 years agoGets around VC++ compiler pickiness. (long != double)
Geoff Thorpe [Fri, 25 Feb 2000 14:50:37 +0000 (14:50 +0000)]
Gets around VC++ compiler pickiness. (long != double)

PR:

19 years agoAdd OpenSSL licen[cs]e.
Bodo Möller [Fri, 25 Feb 2000 14:47:38 +0000 (14:47 +0000)]
Add OpenSSL licen[cs]e.

19 years agomt contained an old copy of mttest.c. remove it and move the other
Ulf Möller [Fri, 25 Feb 2000 14:40:31 +0000 (14:40 +0000)]
mt contained an old copy of mttest.c. remove it and move the other
files to crypto/threads

19 years agoFix off-by-one error :-)
Bodo Möller [Fri, 25 Feb 2000 14:40:11 +0000 (14:40 +0000)]
Fix off-by-one error :-)

19 years agoIn "make clean", delete files created by "make report".
Bodo Möller [Fri, 25 Feb 2000 14:34:55 +0000 (14:34 +0000)]
In "make clean", delete files created by "make report".

19 years agoFix warnings by using unsigned int where appropriate.
Bodo Möller [Fri, 25 Feb 2000 14:27:31 +0000 (14:27 +0000)]
Fix warnings by using unsigned int where appropriate.

19 years agoIgnore files that, well, should be ignored.
Bodo Möller [Fri, 25 Feb 2000 14:23:44 +0000 (14:23 +0000)]
Ignore files that, well, should be ignored.

19 years agoVC++ problem
Ulf Möller [Fri, 25 Feb 2000 14:19:45 +0000 (14:19 +0000)]
VC++ problem

19 years ago*** empty log message ***
Ulf Möller [Fri, 25 Feb 2000 14:19:17 +0000 (14:19 +0000)]
*** empty log message ***

19 years agoEGD bugfix.
Ulf Möller [Fri, 25 Feb 2000 14:16:43 +0000 (14:16 +0000)]
EGD bugfix.

Submitted by: Lutz Jaenicke <Lutz.Jaenicke@aet.TU-Cottbus.DE>

19 years agoispell
Ulf Möller [Fri, 25 Feb 2000 14:07:51 +0000 (14:07 +0000)]
ispell

19 years agolinux-ppc
Ulf Möller [Fri, 25 Feb 2000 14:06:27 +0000 (14:06 +0000)]
linux-ppc

Submitted by: Franz Sirl <Franz.Sirl-kernel@lauterbach.com>

19 years agosigned/unsigned mismatch (VC++)
Ulf Möller [Fri, 25 Feb 2000 14:03:21 +0000 (14:03 +0000)]
signed/unsigned mismatch (VC++)

Submitted by: Peter 'Luna' Runestig" <peter+openssl-users@runestig.com>

19 years agodifferent snprintf version.
Ulf Möller [Fri, 25 Feb 2000 14:00:24 +0000 (14:00 +0000)]
different snprintf version.

19 years agoUse threads for linux-ppc.
Bodo Möller [Fri, 25 Feb 2000 13:45:30 +0000 (13:45 +0000)]
Use threads for linux-ppc.

19 years agoClarification.
Bodo Möller [Fri, 25 Feb 2000 07:48:02 +0000 (07:48 +0000)]
Clarification.

19 years agohandle entropy estimate correctly
Bodo Möller [Fri, 25 Feb 2000 07:40:53 +0000 (07:40 +0000)]
handle entropy estimate correctly

19 years agoAllow code which calls RSA temp key callback to cope
Dr. Stephen Henson [Fri, 25 Feb 2000 00:23:48 +0000 (00:23 +0000)]
Allow code which calls RSA temp key callback to cope
with a failure.

Fix typos in some error codes.

19 years agodivide the correct number...
Ulf Möller [Fri, 25 Feb 2000 00:14:50 +0000 (00:14 +0000)]
divide the correct number...

19 years agowork around a bug in BN_div_recp or BN_reciprocal
Ulf Möller [Thu, 24 Feb 2000 23:37:15 +0000 (23:37 +0000)]
work around a bug in BN_div_recp or BN_reciprocal

19 years agoBIO_printf() change
Ulf Möller [Thu, 24 Feb 2000 22:57:42 +0000 (22:57 +0000)]
BIO_printf() change

19 years agosome test results.
Ulf Möller [Thu, 24 Feb 2000 22:24:51 +0000 (22:24 +0000)]
some test results.

19 years agoYet another "unixware" spelling.
Ulf Möller [Thu, 24 Feb 2000 22:24:36 +0000 (22:24 +0000)]
Yet another "unixware" spelling.

19 years agoUse public domain snprintf() implementation by Patrick Powell to avoid
Ulf Möller [Thu, 24 Feb 2000 22:23:45 +0000 (22:23 +0000)]
Use public domain snprintf() implementation by Patrick Powell to avoid
potential buffer overrun in BIO_printf().
----------------------------------------------------------------------
crypto/bio/b_print.c CVS:
----------------------------------------------------------------------

19 years agoVersion 0.9.5beta2-dev (so that the next snapshot will not
Bodo Möller [Thu, 24 Feb 2000 20:40:02 +0000 (20:40 +0000)]
Version 0.9.5beta2-dev (so that the next snapshot will not
claim to be 0.9.5beta1).

(Are the version number examples correct -- the same numerical
code for:
 * 0.9.3beta2-dev 0x00903002
 * 0.9.3beta2     0x00903002
?)