21 years ago - Made sure some changed behavior is documented in CHANGES.
Richard Levitte [Sat, 18 Dec 1999 01:14:39 +0000 (01:14 +0000)]
 - Made sure some changed behavior is documented in CHANGES.
 - Moved the handling of compile-time defaults from crypto.h to
   mem_dbg.c, since it doesn't make sense for the library users to try
   to affect this without recompiling libcrypto.
   and constant definitions.
 - Aesthetic correction.

21 years agoPoint out that openssl-bugs is public.
Bodo Möller [Sat, 18 Dec 1999 01:13:30 +0000 (01:13 +0000)]
Point out that openssl-bugs is public.

21 years agoAdd a comment.
Bodo Möller [Sat, 18 Dec 1999 00:30:32 +0000 (00:30 +0000)]
Add a comment.

21 years agoAdd missing semicolon to make compiler happy, and switch back
Bodo Möller [Sat, 18 Dec 1999 00:28:21 +0000 (00:28 +0000)]
Add missing semicolon to make compiler happy, and switch back
from MemCheck_start() to CRYPTO_mem_ctrl(CRYPTO_MEM_CHECK_ON)
because that is what applications should use
(MemCheck_start/stop never really worked for applications
unless CRYPTO_MDEBUG was defined both when compiling the library
and when compiling the application, so probably we should
get rid of it).

21 years agoClear out license confusion.
Richard Levitte [Fri, 17 Dec 1999 16:49:23 +0000 (16:49 +0000)]
Clear out license confusion.

21 years agoRebuild of the OpenSSL memory allocation and deallocation routines.
Richard Levitte [Fri, 17 Dec 1999 12:56:24 +0000 (12:56 +0000)]
Rebuild of the OpenSSL memory allocation and deallocation routines.
With this change, the following is provided and present at all times
(meaning CRYPTO_MDEBUG is no longer required to get this functionality):

  - hooks to provide your own allocation and deallocation routines.
    They have to have the same interface as malloc(), realloc() and
    free().  They are registered by calling CRYPTO_set_mem_functions()
    with the function pointers.

  - hooks to provide your own memory debugging routines.  The have to
    have the same interface as as the CRYPTO_dbg_*() routines.  They
    are registered by calling CRYPTO_set_mem_debug_functions() with
    the function pointers.

I moved everything that was already built into OpenSSL and did memory
debugging to a separate file (mem_dbg.c), to make it clear what is

With this, the relevance of the CRYPTO_MDEBUG has changed.  The only
thing in crypto/crypto.h that it affects is the definition of the
MemCheck_start and MemCheck_stop macros.

21 years agoUse MemCheck_start() instead of CRYPTO_mem_ctrl(), and generate a small leak to test...
Richard Levitte [Fri, 17 Dec 1999 12:50:06 +0000 (12:50 +0000)]
Use MemCheck_start() instead of CRYPTO_mem_ctrl(), and generate a small leak to test (conditional)

21 years agoSynchronise VMS scripts with Unix Makefiles
Richard Levitte [Thu, 16 Dec 1999 19:57:50 +0000 (19:57 +0000)]
Synchronise VMS scripts with Unix Makefiles

21 years agoUpdate contact information (openssl-bugs, openssl-security).
Ulf Möller [Thu, 16 Dec 1999 15:10:29 +0000 (15:10 +0000)]
Update contact information (openssl-bugs, openssl-security).

21 years agoDelete an unused variable and make the PKCS#12 keygen debugging code work
Dr. Stephen Henson [Wed, 15 Dec 1999 02:36:48 +0000 (02:36 +0000)]
Delete an unused variable and make the PKCS#12 keygen debugging code work

21 years agoFix the S/MIME code to use canonical MIME format for
Dr. Stephen Henson [Wed, 15 Dec 1999 01:26:17 +0000 (01:26 +0000)]
Fix the S/MIME code to use canonical MIME format for
encrypted mail. Also update the smime docs.

21 years agoSolaris x86 assembler problem is already addressed in ./config
Ulf Möller [Tue, 14 Dec 1999 15:28:10 +0000 (15:28 +0000)]
Solaris x86 assembler problem is already addressed in ./config
(bug reports keep coming in because that was still missing in 0.9.4)

21 years agoVarious S/MIME fixes. Fix for memory leak, recipient list bug
Dr. Stephen Henson [Tue, 14 Dec 1999 02:44:27 +0000 (02:44 +0000)]
Various S/MIME fixes. Fix for memory leak, recipient list bug
and not excluding parameters with DSA keys.

21 years agoFix a typo in a_enum.c.
Dr. Stephen Henson [Mon, 13 Dec 1999 13:14:14 +0000 (13:14 +0000)]
Fix a typo in a_enum.c.

21 years agoVarious S/MIME fixes.
Dr. Stephen Henson [Sat, 11 Dec 1999 20:04:06 +0000 (20:04 +0000)]
Various S/MIME fixes.

21 years agoFix for crashing INTEGERs, ENUMERATEDs and OBJECT IDENTIFIERs.
Dr. Stephen Henson [Fri, 10 Dec 1999 13:46:48 +0000 (13:46 +0000)]

Also fix a memory leak in PKCS#7 routines.

21 years agoMake the PKCS#7 S/MIME functions check for passed NULL pointers.
Dr. Stephen Henson [Thu, 9 Dec 1999 01:31:32 +0000 (01:31 +0000)]
Make the PKCS#7 S/MIME functions check for passed NULL pointers.

Fix the usage message of smime utility and sanitise the return

Add some documentation.

21 years agoDon't use inline assembler on x86 Solaris (would need a different syntax).
Ulf Möller [Wed, 8 Dec 1999 22:55:06 +0000 (22:55 +0000)]
Don't use inline assembler on x86 Solaris (would need a different syntax).

21 years agoUseless files deleted -- they were just copies of files of the same name
Bodo Möller [Wed, 8 Dec 1999 18:30:39 +0000 (18:30 +0000)]
Useless files deleted -- they were just copies of files of the same name
in the apps/ directory (which were recently changed).

21 years agoAdd i2d_ASN1_PRINTABLESTRING() function, and do 'make update'
Dr. Stephen Henson [Wed, 8 Dec 1999 00:56:15 +0000 (00:56 +0000)]
Add i2d_ASN1_PRINTABLESTRING() function, and do 'make update'

21 years agoModify S/MIME application so the -signer option writes the signer(s)
Dr. Stephen Henson [Tue, 7 Dec 1999 02:35:52 +0000 (02:35 +0000)]
Modify S/MIME application so the -signer option writes the signer(s)
to a file if we are verifying.

21 years agoMerge in my S/MIME library and utility.
Dr. Stephen Henson [Sun, 5 Dec 1999 00:40:59 +0000 (00:40 +0000)]
Merge in my S/MIME library and utility.

21 years agoCORE SDI proposed patch doesn't make any sense. Undo.
Ulf Möller [Sat, 4 Dec 1999 01:19:23 +0000 (01:19 +0000)]
CORE SDI proposed patch doesn't make any sense. Undo.

21 years agoOops!
Ulf Möller [Sat, 4 Dec 1999 00:13:37 +0000 (00:13 +0000)]

21 years agoCircumvent an exploitable buffer overrun error in RSA Security's RSAREF
Ulf Möller [Fri, 3 Dec 1999 23:56:08 +0000 (23:56 +0000)]
Circumvent an exploitable buffer overrun error in RSA Security's RSAREF
library. See: http://www.CORE-SDI.COM/english/ssh/index.html

Submitted by:
Reviewed by:

21 years agoUse des_set_key_unchecked, not des_set_key.
Bodo Möller [Fri, 3 Dec 1999 20:26:20 +0000 (20:26 +0000)]
Use des_set_key_unchecked, not des_set_key.

21 years agoAdd functions des_set_key_checked, des_set_key_unchecked.
Bodo Möller [Fri, 3 Dec 1999 20:24:21 +0000 (20:24 +0000)]
Add functions des_set_key_checked, des_set_key_unchecked.
Never use des_set_key (it depends on the global variable des_check_key),
but usually des_set_key_unchecked.
Only destest.c bothered to look at the return values of des_set_key,
but it did not set des_check_key -- if it had done so,
most checks would have failed because of wrong parity and
because of weak keys.

21 years agoNew function PKC12_newpass()
Dr. Stephen Henson [Fri, 3 Dec 1999 03:46:18 +0000 (03:46 +0000)]
New function PKC12_newpass()

21 years agoFix a bug in the modified purpose code: it wasn't updated to use the
Dr. Stephen Henson [Fri, 3 Dec 1999 00:53:48 +0000 (00:53 +0000)]
Fix a bug in the modified purpose code: it wasn't updated to use the
new purpose getting function.

Update the ca-cert.pem and pca-cert.pem "CA" certificates so they
really are CA certificate: that is they have the appropriate extensions.

21 years agoChange the trust and purpose code so it doesn't need init
Dr. Stephen Henson [Thu, 2 Dec 1999 02:33:56 +0000 (02:33 +0000)]
Change the trust and purpose code so it doesn't need init
either and has a static and dynamic mix.

21 years agoModify the X509 V3 extension lookup code.
Dr. Stephen Henson [Wed, 1 Dec 1999 01:49:46 +0000 (01:49 +0000)]
Modify the X509 V3 extension lookup code.

21 years agoMake salting the default. Fail gracefully if the input is not salted.
Ben Laurie [Tue, 30 Nov 1999 20:15:19 +0000 (20:15 +0000)]
Make salting the default. Fail gracefully if the input is not salted.

21 years agoDocument the extension tests performed by the -purpose test
Dr. Stephen Henson [Tue, 30 Nov 1999 14:39:58 +0000 (14:39 +0000)]
Document the extension tests performed by the -purpose test
in the x509 utility.

21 years agoDocument all possible errors (and some impossible) from the verify program.
Dr. Stephen Henson [Tue, 30 Nov 1999 02:28:42 +0000 (02:28 +0000)]
Document all possible errors (and some impossible) from the verify program.

21 years agoRemainder of SSL purpose and trust code: trust and purpose setting in
Dr. Stephen Henson [Mon, 29 Nov 1999 22:35:00 +0000 (22:35 +0000)]
Remainder of SSL purpose and trust code: trust and purpose setting in
SSL_CTX and SSL, functions to set them and defaults if no values set.

21 years agoAdd part of chain verify SSL support code: not complete or doing anything
Dr. Stephen Henson [Mon, 29 Nov 1999 01:09:25 +0000 (01:09 +0000)]
Add part of chain verify SSL support code: not complete or doing anything

Add a function X509_STORE_CTX_purpose_inherit() which implements the logic
of "inheriting" purpose and trust from a parent structure and using a default:
this will be used in the SSL code and possibly future S/MIME.

Partial documentation of the 'verify' utility. Still need to document how all
the extension checking works and the various error messages.

21 years agoAdd trust setting support to the verify code. It now checks the
Dr. Stephen Henson [Sat, 27 Nov 1999 19:43:10 +0000 (19:43 +0000)]
Add trust setting support to the verify code. It now checks the
trust settings of the root CA.

After a few fixes it seems to work OK.

Still need to add support to SSL and S/MIME code though.

21 years agoAdd compilation of x509_trs
Richard Levitte [Sat, 27 Nov 1999 15:26:48 +0000 (15:26 +0000)]
Add compilation of x509_trs

21 years agoOops! Commit died on me :-(
Dr. Stephen Henson [Sat, 27 Nov 1999 01:18:39 +0000 (01:18 +0000)]
Oops! Commit died on me :-(

21 years agoInitial trust code: allow setting of trust checking functions
Dr. Stephen Henson [Sat, 27 Nov 1999 01:14:04 +0000 (01:14 +0000)]
Initial trust code: allow setting of trust checking functions
in a table. Doesn't do too much yet.

Make the -<digestname> options in 'x509' affect all relevant

Change the name of the 'notrust' options to 'reject' as this
causes less confusion and is a better description of the

A few constification changes.

21 years agoNew options to the -verify program which can be used for chain verification.
Dr. Stephen Henson [Fri, 26 Nov 1999 00:27:07 +0000 (00:27 +0000)]
New options to the -verify program which can be used for chain verification.
Extend the X509_PURPOSE structure to include shortnames for purposed and default
trust ids.

Still need some extendable trust checking code and integration with the SSL and
S/MIME code.

21 years agoInitial chain verify code: not tested probably not working
Dr. Stephen Henson [Wed, 24 Nov 1999 01:31:49 +0000 (01:31 +0000)]
Initial chain verify code: not tested probably not working
at present. However nothing enables it yet so this doesn't
matter :-)

21 years agoSupport for authority information access extension.
Dr. Stephen Henson [Tue, 23 Nov 1999 18:50:28 +0000 (18:50 +0000)]
Support for authority information access extension.

Fix so EVP_PKEY_rset_*() check return codes.

21 years agoTransparent support for PKCS#8 private keys in RSA/DSA.
Dr. Stephen Henson [Sun, 21 Nov 1999 22:28:31 +0000 (22:28 +0000)]
Transparent support for PKCS#8 private keys in RSA/DSA.

New universal public key format.

Fix CRL+cert load problem in by_file.c

Make verify report errors when loading files or dirs

21 years agoSupport for otherName in GeneralName.
Dr. Stephen Henson [Fri, 19 Nov 1999 02:19:58 +0000 (02:19 +0000)]
Support for otherName in GeneralName.

21 years agoUpdate dependencies.
Ben Laurie [Thu, 18 Nov 1999 14:32:54 +0000 (14:32 +0000)]
Update dependencies.

21 years agoFix warning.
Ben Laurie [Thu, 18 Nov 1999 14:10:53 +0000 (14:10 +0000)]
Fix warning.

21 years agoRestore traditional SSL_get_session behaviour so that s_client and s_server
Bodo Möller [Wed, 17 Nov 1999 21:36:13 +0000 (21:36 +0000)]
Restore traditional SSL_get_session behaviour so that s_client and s_server
don't leak tons of memory.

21 years agoMissing #ifdef NO_DES
Ulf Möller [Wed, 17 Nov 1999 13:03:29 +0000 (13:03 +0000)]
Missing #ifdef NO_DES

21 years agoModify verify code to handle self signed certificates.
Dr. Stephen Henson [Wed, 17 Nov 1999 01:20:29 +0000 (01:20 +0000)]
Modify verify code to handle self signed certificates.

21 years agoStore verify_result with sessions to avoid potential security hole.
Bodo Möller [Tue, 16 Nov 1999 23:15:41 +0000 (23:15 +0000)]
Store verify_result with sessions to avoid potential security hole.

21 years agoFix for a bug in PKCS#7 code and non-detached data.
Dr. Stephen Henson [Tue, 16 Nov 1999 14:54:50 +0000 (14:54 +0000)]
Fix for a bug in PKCS#7 code and non-detached data.

Remove rc4-64 from ciphers since it doesn't exist...

21 years agoClarify docs.
Dr. Stephen Henson [Tue, 16 Nov 1999 02:51:41 +0000 (02:51 +0000)]
Clarify docs.

21 years agoAdd a salt to the key derivation using the 'enc' program.
Dr. Stephen Henson [Tue, 16 Nov 1999 02:49:25 +0000 (02:49 +0000)]
Add a salt to the key derivation using the 'enc' program.

21 years agoNew function X509_cmp().
Dr. Stephen Henson [Tue, 16 Nov 1999 00:56:03 +0000 (00:56 +0000)]
New function X509_cmp().

21 years agoThis corrects the reference count handling in SSL_get_session.
Mark J. Cox [Mon, 15 Nov 1999 16:31:31 +0000 (16:31 +0000)]
This corrects the reference count handling in SSL_get_session.
Previously, the returned SSL_SESSION didn't have its reference count
incremented so the SSL_SESSION could be freed at any time causing
seg-faults if the pointer was subsequently used. Code that uses
SSL_get_session must now make a corresponding SSL_SESSION_free() call when
it is done to avoid memory leaks (or blocked up session caches).

Submitted By: Geoff Thorpe <>

21 years ago'req' fixes. Reinstate length check one request fields.
Dr. Stephen Henson [Sun, 14 Nov 1999 23:10:50 +0000 (23:10 +0000)]
'req' fixes. Reinstate length check one request fields.
Fix to stop null being added to attributes.
Modify X509_LOOKUP, X509_INFO to handle auxiliary info.

21 years agoAdd some examples to the enc man page.
Dr. Stephen Henson [Sun, 14 Nov 1999 13:34:34 +0000 (13:34 +0000)]
Add some examples to the enc man page.

21 years agoAdd support for the 40 and 64 bit RC2 and RC4 ciphers in 'enc'
Dr. Stephen Henson [Sun, 14 Nov 1999 03:23:17 +0000 (03:23 +0000)]
Add support for the 40 and 64 bit RC2 and RC4 ciphers in 'enc'
add documentation for 'enc'.

21 years agoAdd info about the header and footer lines used in PEM formats
Dr. Stephen Henson [Sat, 13 Nov 1999 21:58:39 +0000 (21:58 +0000)]
Add info about the header and footer lines used in PEM formats
and add an nseq manpage.

21 years agoCorrect x509 manpaghe and add a crl manpage
Dr. Stephen Henson [Sat, 13 Nov 1999 21:28:01 +0000 (21:28 +0000)]
Correct x509 manpaghe and add a crl manpage

21 years agoThe info removal code was overcomplicated, and error-prone (references being wrongly...
Richard Levitte [Fri, 12 Nov 1999 21:51:24 +0000 (21:51 +0000)]
The info removal code was overcomplicated, and error-prone (references being wrongly decreased).  Fixed.

21 years agoAvoid deadlock.
Bodo Möller [Fri, 12 Nov 1999 16:20:30 +0000 (16:20 +0000)]
Avoid deadlock.

21 years agoAdd an spkac manual page and fix the pkcs7 manpage.
Dr. Stephen Henson [Fri, 12 Nov 1999 14:04:41 +0000 (14:04 +0000)]
Add an spkac manual page and fix the pkcs7 manpage.

21 years agoAvoid some silly compiler warnings, and add the change log I forgot :-)
Richard Levitte [Fri, 12 Nov 1999 03:12:46 +0000 (03:12 +0000)]
Avoid some silly compiler warnings, and add the change log I forgot :-)

21 years agoTwo changes have been made:
Richard Levitte [Fri, 12 Nov 1999 02:51:24 +0000 (02:51 +0000)]
Two changes have been made:

  1. Added code to the memory leak detecting code to give the user the
     possibility to add information, thereby forming a traceback.

  2. Make the memory leak detecting code multithread-safe.

The idea is that we're actually dealing with two separate critical
sections, one containing the hash tables with the information, the
other containing the current memory checking mode.  Those should not
be handled with the same lock, especially since their handling overlap.
Hence, the added second lock.

21 years agoSome crypto applications are now being built on Unix, so they should on VMS as well...
Richard Levitte [Fri, 12 Nov 1999 02:21:49 +0000 (02:21 +0000)]
Some crypto applications are now being built on Unix, so they should on VMS as well.  Not by default, however.

21 years agoIt's possible that considering the configuration file as a binary file
Richard Levitte [Fri, 12 Nov 1999 02:19:05 +0000 (02:19 +0000)]
It's possible that considering the configuration file as a binary file
works on Unix and MS-DOS/Windows.  It does not under VMS, so open it
as text.

21 years agoAvoid silly compiler warnings about functions not being declared and an int missing.
Richard Levitte [Fri, 12 Nov 1999 02:10:23 +0000 (02:10 +0000)]
Avoid silly compiler warnings about functions not being declared and an int missing.

21 years agoSome new names in asn1.h are longer than 31 chars, which disturbs the VMS C compilers...
Richard Levitte [Fri, 12 Nov 1999 02:04:30 +0000 (02:04 +0000)]
Some new names in asn1.h are longer than 31 chars, which disturbs the VMS C compilers...

21 years agoAdjust to changes in apps/openssl.cnf
Richard Levitte [Fri, 12 Nov 1999 01:59:47 +0000 (01:59 +0000)]
Adjust to changes in apps/openssl.cnf

21 years agoAdjust to changes in apps/Makefile.ssl
Richard Levitte [Fri, 12 Nov 1999 01:52:59 +0000 (01:52 +0000)]
Adjust to changes in apps/Makefile.ssl

21 years agoDIFFERENCE doesn't handle long (>255 chars) lines well. Use BACKUP instead. No...
Richard Levitte [Fri, 12 Nov 1999 01:46:50 +0000 (01:46 +0000)]
DIFFERENCE doesn't handle long (>255 chars) lines well.  Use BACKUP instead.  No, I'm not joking.

21 years agoadjust to changes in test/Makefile.ssl
Richard Levitte [Fri, 12 Nov 1999 01:45:04 +0000 (01:45 +0000)]
adjust to changes in test/Makefile.ssl

21 years agoadjust to changes in test/testssl
Richard Levitte [Fri, 12 Nov 1999 01:43:55 +0000 (01:43 +0000)]
adjust to changes in test/testssl

21 years agoMake sure installed files are world readable
Richard Levitte [Fri, 12 Nov 1999 01:42:59 +0000 (01:42 +0000)]
Make sure installed files are world readable

21 years agoMerge some common functionality in the apps, delete
Dr. Stephen Henson [Fri, 12 Nov 1999 01:42:25 +0000 (01:42 +0000)]
Merge some common functionality in the apps, delete
the encryption option in the pkcs7 utility (they never
did anything) and add a couple more options to pkcs7.

21 years agoOops forgot the S/MIME v3 RFC.
Dr. Stephen Henson [Fri, 12 Nov 1999 01:07:33 +0000 (01:07 +0000)]
Oops forgot the S/MIME v3 RFC.

21 years agoMore docs and corrections/updates
Dr. Stephen Henson [Fri, 12 Nov 1999 01:04:39 +0000 (01:04 +0000)]
More docs and corrections/updates

21 years agoAdd password command line options to some utils. Fix and update man
Dr. Stephen Henson [Thu, 11 Nov 1999 18:41:31 +0000 (18:41 +0000)]
Add password command line options to some utils. Fix and update man

21 years agoFix a couple of outstanding issues: update STATUS file, fix NO_FP_API problems.
Dr. Stephen Henson [Thu, 11 Nov 1999 13:58:41 +0000 (13:58 +0000)]
Fix a couple of outstanding issues: update STATUS file, fix NO_FP_API problems.

Update docs, change 'ca' to use the new callback parameter. Now moved key_callback
into app.c because some other utilities will use it soon.

21 years agoOops. The pkcs8 man page wasn't finished: this is an updated version
Dr. Stephen Henson [Thu, 11 Nov 1999 00:48:39 +0000 (00:48 +0000)]
Oops. The pkcs8 man page wasn't finished: this is an updated version

21 years agoVery preliminary POD format documentation for some
Dr. Stephen Henson [Wed, 10 Nov 1999 02:52:17 +0000 (02:52 +0000)]
Very preliminary POD format documentation for some
of the openssl utility commands...

21 years agoUndo silly change.
Bodo Möller [Tue, 9 Nov 1999 16:41:52 +0000 (16:41 +0000)]
Undo silly change.

21 years agoFix (spurious) warnings.
Ben Laurie [Tue, 9 Nov 1999 12:09:24 +0000 (12:09 +0000)]
Fix (spurious) warnings.

21 years agoAvoid some warnings.
Bodo Möller [Tue, 9 Nov 1999 10:00:15 +0000 (10:00 +0000)]
Avoid some warnings.

21 years agoFix to the -revoke option in ca. It was leaking memory, crashing and just
Dr. Stephen Henson [Mon, 8 Nov 1999 13:58:08 +0000 (13:58 +0000)]
Fix to the -revoke option in ca. It was leaking memory, crashing and just
plain not working :-(

Also fix some memory leaks in the new X509_NAME code.

Fix so new app_rand code doesn't crash 'x509' and move #include so it compiles
under Win32.

21 years agoAllow additional information to be attached to a
Dr. Stephen Henson [Thu, 4 Nov 1999 00:45:35 +0000 (00:45 +0000)]
Allow additional information to be attached to a
certificate: currently this includes trust settings
and a "friendly name".

21 years agoFix assembler for Alpha (tested only on DEC OSF not Linux or *BSD). The
Mark J. Cox [Wed, 3 Nov 1999 14:10:10 +0000 (14:10 +0000)]
Fix assembler for Alpha (tested only on DEC OSF not Linux or *BSD).  The
problem was that one of the replacement routines had not been working since
SSLeay releases.  For now the offending routine has been replaced with
non-optimised assembler.  Even so, this now gives around 95% performance
improvement for 1024 bit RSA signs.

21 years ago*** empty log message ***
Ulf Möller [Sat, 30 Oct 1999 19:09:05 +0000 (19:09 +0000)]
*** empty log message ***

21 years agoFix to PKCS#7 routines so it can decrypt some oddball RC2 handling.
Dr. Stephen Henson [Fri, 29 Oct 1999 13:06:25 +0000 (13:06 +0000)]
Fix to PKCS#7 routines so it can decrypt some oddball RC2 handling.

21 years agoContinued multibyte character support.
Dr. Stephen Henson [Wed, 27 Oct 1999 00:15:11 +0000 (00:15 +0000)]
Continued multibyte character support.

Add a bunch of functions to simplify the creation of X509_NAME structures.

Change the X509_NAME_entry_add stuff in req/ca so it no longer uses
X509_NAME_entry_count(): passing -1 has the same effect.

21 years agoAlways hash the pid in the first iteration in ssleay_rand_bytes,
Bodo Möller [Tue, 26 Oct 1999 16:26:48 +0000 (16:26 +0000)]
Always hash the pid in the first iteration in ssleay_rand_bytes,
don't try to detect fork()s by looking at getpid().
The reason is that threads sharing the same memory can have different
PIDs; it's inefficient to run RAND_seed each time a different thread
calls RAND_bytes.

21 years agoMake md_rand.c more robust.
Bodo Möller [Tue, 26 Oct 1999 14:49:12 +0000 (14:49 +0000)]
Make md_rand.c more robust.

21 years agoWarn about RANDFILE being overwritten.
Bodo Möller [Tue, 26 Oct 1999 11:27:42 +0000 (11:27 +0000)]
Warn about RANDFILE being overwritten.

21 years agoDon't be overly paranoid.
Bodo Möller [Tue, 26 Oct 1999 11:19:42 +0000 (11:19 +0000)]
Don't be overly paranoid.

21 years agoNew file app_rand.c with some functionality used in various openssl
Bodo Möller [Tue, 26 Oct 1999 01:59:11 +0000 (01:59 +0000)]
New file app_rand.c with some functionality used in various openssl

21 years agoVarious randomness handling bugfixes and improvements --
Bodo Möller [Tue, 26 Oct 1999 01:56:29 +0000 (01:56 +0000)]
Various randomness handling bugfixes and improvements --
some utilities that should have used RANDFILE did not,
and -rand handling was broken except in genrsa.

21 years agoReport an error from X509_STORE_load_locations
Bodo Möller [Tue, 26 Oct 1999 01:52:16 +0000 (01:52 +0000)]
Report an error from X509_STORE_load_locations
when X509_LOOKUP_load_file or X509_LOOKUP_add_dir failed.

21 years agoUpdate Borland C++ builder support.
Bodo Möller [Mon, 25 Oct 1999 21:38:43 +0000 (21:38 +0000)]
Update Borland C++ builder support.

Submitted by: Janez Jere <>