oweals/openssl.git
24 years agoFix signed/unsigned warnings.
Ben Laurie [Wed, 16 Feb 2000 12:09:17 +0000 (12:09 +0000)]
Fix signed/unsigned warnings.

24 years agoInstall manpages below OPENSSLDIR (I think it was meant to be this way?).
Ulf Möller [Tue, 15 Feb 2000 18:34:46 +0000 (18:34 +0000)]
Install manpages below OPENSSLDIR (I think it was meant to be this way?).
New variable for man directory.

24 years agoCreate the man directories where the manpages will be put.
Ulf Möller [Tue, 15 Feb 2000 18:21:33 +0000 (18:21 +0000)]
Create the man directories where the manpages will be put.

24 years agoMove initial key to read-only segment.
Andy Polyakov [Tue, 15 Feb 2000 17:43:12 +0000 (17:43 +0000)]
Move initial key to read-only segment.

24 years agoMove CAST_S_tables to read-only segment.
Andy Polyakov [Tue, 15 Feb 2000 17:37:44 +0000 (17:37 +0000)]
Move CAST_S_tables to read-only segment.

24 years agoAvoid GNU C assembler templates under Solaris x86.
Andy Polyakov [Tue, 15 Feb 2000 17:31:12 +0000 (17:31 +0000)]
Avoid GNU C assembler templates under Solaris x86.

24 years agoNew NO_INLINE_ASM macro. Primary target for the moment is Solaris x86
Andy Polyakov [Tue, 15 Feb 2000 17:20:52 +0000 (17:20 +0000)]
New NO_INLINE_ASM macro. Primary target for the moment is Solaris x86
which can't stand GNU C assembler templates.

24 years agoFix for Netscape "hang" bug.
Dr. Stephen Henson [Tue, 15 Feb 2000 14:19:44 +0000 (14:19 +0000)]
Fix for Netscape "hang" bug.

24 years agotest_mont was exercising 100-bit multiplication modulus X*I-bit, where
Andy Polyakov [Tue, 15 Feb 2000 13:50:02 +0000 (13:50 +0000)]
test_mont was exercising 100-bit multiplication modulus X*I-bit, where
X is 5120 on 32-bit and 151552 on 64-bit architectures and I varies
from 0 to 4. As result the test was *unreasonably* slow and virtually
impossible to complete on 64-bit architectures (e.g. IRIX bc couldn't
even swallow such long lines).

24 years agoRemove the access() call altogether for VMS, since it doesn't quite
Richard Levitte [Tue, 15 Feb 2000 09:44:54 +0000 (09:44 +0000)]
Remove the access() call altogether for VMS, since it doesn't quite
work for directory specifications (this will be reported as a bug to
DEC^H^H^HCompaq).  It could as well be removed for all others as well,
since stat() and open() will return appropriate errors as well, but I
leave that to someone else to decide.

24 years agoTime ran away...
Richard Levitte [Mon, 14 Feb 2000 16:58:40 +0000 (16:58 +0000)]
Time ran away...

24 years agoAdd a couple of forgotten $(PERL), and make the code to run pod2html a
Richard Levitte [Mon, 14 Feb 2000 16:55:23 +0000 (16:55 +0000)]
Add a couple of forgotten $(PERL), and make the code to run pod2html a
tad more readable.

24 years agoModernise 'selfsign.c' to use new X509_NAME code
Dr. Stephen Henson [Sun, 13 Feb 2000 00:28:26 +0000 (00:28 +0000)]
Modernise 'selfsign.c' to use new X509_NAME code
and add example of extension aliasing. Also fix
the extension aliasing because it didn't work :-)

24 years agoHP-UX tune-up: new unified configs, HP C compiler bug workaround.
Andy Polyakov [Sat, 12 Feb 2000 23:33:01 +0000 (23:33 +0000)]
HP-UX tune-up: new unified configs, HP C compiler bug workaround.

24 years agoMake pkcs12 and smime applications seed random number
Dr. Stephen Henson [Sat, 12 Feb 2000 03:03:04 +0000 (03:03 +0000)]
Make pkcs12 and smime applications seed random number
generator (otherwise they don't work) and add -rand
option. Update docs.

24 years agoA hack to make sure access() will give us the correct answer about the
Richard Levitte [Fri, 11 Feb 2000 18:12:47 +0000 (18:12 +0000)]
A hack to make sure access() will give us the correct answer about the
accessability of an "empty" directory.  Thsi *is* weird, and a better
solution will be provided in apps/ca.c, when I get time to hack at it.

24 years agoCorrections.
Bodo Möller [Fri, 11 Feb 2000 17:18:50 +0000 (17:18 +0000)]
Corrections.

24 years agoUpdate.
Bodo Möller [Fri, 11 Feb 2000 16:31:04 +0000 (16:31 +0000)]
Update.

24 years agoImplement MD5-based "apr1" password hash.
Bodo Möller [Fri, 11 Feb 2000 16:25:44 +0000 (16:25 +0000)]
Implement MD5-based "apr1" password hash.

24 years agoCorrect time in seconds instead of minutes.
Ben Laurie [Fri, 11 Feb 2000 16:08:40 +0000 (16:08 +0000)]
Correct time in seconds instead of minutes.

24 years agoFix shadow.
Ben Laurie [Fri, 11 Feb 2000 13:11:18 +0000 (13:11 +0000)]
Fix shadow.

24 years agoAdd compilation of the new passwd utility.
Richard Levitte [Fri, 11 Feb 2000 11:21:50 +0000 (11:21 +0000)]
Add compilation of the new passwd utility.

24 years agoAdd references to the new passwd utility.
Richard Levitte [Fri, 11 Feb 2000 11:21:01 +0000 (11:21 +0000)]
Add references to the new passwd utility.

24 years agoMake gcc 2.95.2 happy again, even under ``-Wall -Wshadow -Wpointer-arith -Wcast-align
Ralf S. Engelschall [Fri, 11 Feb 2000 09:47:18 +0000 (09:47 +0000)]
Make gcc 2.95.2 happy again, even under ``-Wall -Wshadow -Wpointer-arith -Wcast-align
-Wmissing-prototypes -Wmissing-declarations -Wnested-externs -Winline''.

24 years ago'passwd' tool.
Bodo Möller [Thu, 10 Feb 2000 21:50:52 +0000 (21:50 +0000)]
'passwd' tool.

24 years ago16 * 8 = 128.
Bodo Möller [Thu, 10 Feb 2000 21:47:06 +0000 (21:47 +0000)]
16 * 8 = 128.

24 years agoMove down the attributions
Richard Levitte [Tue, 8 Feb 2000 23:40:20 +0000 (23:40 +0000)]
Move down the attributions

24 years agoNote about des_ncbc_encrypt.
Bodo Möller [Tue, 8 Feb 2000 21:50:52 +0000 (21:50 +0000)]
Note about des_ncbc_encrypt.

24 years agoConstify
Richard Levitte [Tue, 8 Feb 2000 14:19:14 +0000 (14:19 +0000)]
Constify

24 years agoUpdate docs.
Dr. Stephen Henson [Tue, 8 Feb 2000 13:37:08 +0000 (13:37 +0000)]
Update docs.

24 years agoAdd command line password options to the reamining utilities,
Dr. Stephen Henson [Tue, 8 Feb 2000 01:34:59 +0000 (01:34 +0000)]
Add command line password options to the reamining utilities,
amend docs.

24 years agodes_modes is a section 7 manual, so let's make sure that's where it ends up
Richard Levitte [Mon, 7 Feb 2000 18:25:26 +0000 (18:25 +0000)]
des_modes is a section 7 manual, so let's make sure that's where it ends up

24 years agoFirst try at documenting the DES (and other algorithms) modes
Richard Levitte [Mon, 7 Feb 2000 18:15:06 +0000 (18:15 +0000)]
First try at documenting the DES (and other algorithms) modes

24 years agoRename the X509V3_*_d2i functions to X509_get_ext_d2i() etc.
Dr. Stephen Henson [Mon, 7 Feb 2000 01:17:22 +0000 (01:17 +0000)]
Rename the X509V3_*_d2i functions to X509_get_ext_d2i() etc.
This better reflects their behaviour.

24 years ago"print" is GNU bc specific.
Ulf Möller [Sun, 6 Feb 2000 23:33:06 +0000 (23:33 +0000)]
"print" is GNU bc specific.

24 years agoRefer to EVP_DigestInit() in the hash function descriptions.
Ulf Möller [Sun, 6 Feb 2000 23:26:31 +0000 (23:26 +0000)]
Refer to EVP_DigestInit() in the hash function descriptions.

24 years agoA proposed freeze and release time has come up.
Richard Levitte [Sun, 6 Feb 2000 22:01:10 +0000 (22:01 +0000)]
A proposed freeze and release time has come up.

24 years agoDocument EVP digest operations.
Dr. Stephen Henson [Sun, 6 Feb 2000 17:44:54 +0000 (17:44 +0000)]
Document EVP digest operations.

24 years agoBN bug fixes
Ulf Möller [Sun, 6 Feb 2000 16:35:59 +0000 (16:35 +0000)]
BN bug fixes

24 years agoput missing line back in.
Ulf Möller [Sun, 6 Feb 2000 16:35:28 +0000 (16:35 +0000)]
put missing line back in.

24 years agoImprove bntest slightly, and fix another bug in the BN library.
Ulf Möller [Sun, 6 Feb 2000 15:56:59 +0000 (15:56 +0000)]
Improve bntest slightly, and fix another bug in the BN library.

24 years agoSupport for MacOS X (Rhapsody) is added. Also get rid of volatile
Andy Polyakov [Sun, 6 Feb 2000 11:15:20 +0000 (11:15 +0000)]
Support for MacOS X (Rhapsody) is added. Also get rid of volatile
qualifier in asm definitions as it prevents compiler from moving
the instruction(s) during optimization pass.

24 years agoTypo corrected...
Richard Levitte [Sun, 6 Feb 2000 10:07:32 +0000 (10:07 +0000)]
Typo corrected...

24 years agoMemory leak.
Bodo Möller [Sun, 6 Feb 2000 02:48:53 +0000 (02:48 +0000)]
Memory leak.

24 years agoBN_div bugfix. The q-- loop should not be entered in the n0==d0 case.
Ulf Möller [Sun, 6 Feb 2000 00:25:39 +0000 (00:25 +0000)]
BN_div bugfix. The q-- loop should not be entered in the n0==d0 case.

24 years agoCorrection: openssl.c must get the long version of the apps_startup()
Bodo Möller [Sat, 5 Feb 2000 21:39:20 +0000 (21:39 +0000)]
Correction: openssl.c must get the long version of the apps_startup()
macro

24 years agoCosmetic changes.
Bodo Möller [Sat, 5 Feb 2000 21:28:09 +0000 (21:28 +0000)]
Cosmetic changes.

24 years agoAdd support for some broken PKCS#8 formats.
Dr. Stephen Henson [Sat, 5 Feb 2000 21:07:56 +0000 (21:07 +0000)]
Add support for some broken PKCS#8 formats.

24 years agoCommit patch to bn.h that CVS decided to throw away during 'cvs update',
Bodo Möller [Sat, 5 Feb 2000 20:39:26 +0000 (20:39 +0000)]
Commit patch to bn.h that CVS decided to throw away during 'cvs update',
and initialize too_many because memset(..., 0, ...) is not used here.

24 years agoGenerate just one error code if iterated SSL_CTX_get() fails.
Bodo Möller [Sat, 5 Feb 2000 19:29:00 +0000 (19:29 +0000)]
Generate just one error code if iterated SSL_CTX_get() fails.
Avoid enabled 'assert()' in production library.

24 years agoUse MONT_WORD macro to control if the word-based or the bignum
Ulf Möller [Sat, 5 Feb 2000 18:23:05 +0000 (18:23 +0000)]
Use MONT_WORD macro to control if the word-based or the bignum
algorithm is used.

24 years agoNew functions BN_CTX_start(), BN_CTX_get(), BN_CTX_end() to access
Ulf Möller [Sat, 5 Feb 2000 14:17:32 +0000 (14:17 +0000)]
New functions BN_CTX_start(), BN_CTX_get(), BN_CTX_end() to access
temporary BIGNUMs. BN_CTX still uses a fixed number of BIGNUMs, but
the BN_CTX implementation could now easily be changed.

24 years agomd2 is documented in the md5 page. lets see if this works...
Ulf Möller [Sat, 5 Feb 2000 10:43:03 +0000 (10:43 +0000)]
md2 is documented in the md5 page. lets see if this works...

24 years agoDocument RC4.
Ulf Möller [Sat, 5 Feb 2000 10:41:05 +0000 (10:41 +0000)]
Document RC4.

24 years agoReplace ridiculous libdes PRNG with RAND_bytes. These functions are not
Ulf Möller [Sat, 5 Feb 2000 10:39:54 +0000 (10:39 +0000)]
Replace ridiculous libdes PRNG with RAND_bytes. These functions are not
used anywhere in OpenSSL, but might be used by libdes applications.

24 years agoNew xcbc_ok test vector is required after the parity bits in cbc2_key
Andy Polyakov [Fri, 4 Feb 2000 15:16:22 +0000 (15:16 +0000)]
New xcbc_ok test vector is required after the parity bits in cbc2_key
were fixed up. The catch is that in the DESX test the cbc2_key is used
as whitening key where *all* 64 bits are significant.

24 years agoRename SSLeay_add_all_algorithms() et al to
Dr. Stephen Henson [Fri, 4 Feb 2000 14:01:38 +0000 (14:01 +0000)]
Rename SSLeay_add_all_algorithms() et al to
OpenSSL_add_all_algorithms(). Move these into
separate files so they work properly.

24 years agoReport progress as in dsatest.c when creating a DHE key.
Bodo Möller [Fri, 4 Feb 2000 11:21:18 +0000 (11:21 +0000)]
Report progress as in dsatest.c when creating a DHE key.

24 years agoUse correct, not American spelling.
Bodo Möller [Fri, 4 Feb 2000 00:56:09 +0000 (00:56 +0000)]
Use correct, not American spelling.

24 years agoCorrect spelling as it was done in the source.
Bodo Möller [Fri, 4 Feb 2000 00:42:21 +0000 (00:42 +0000)]
Correct spelling as it was done in the source.

24 years ago*** empty log message ***
Ulf Möller [Thu, 3 Feb 2000 23:38:55 +0000 (23:38 +0000)]
*** empty log message ***

24 years agoa short page for "speed"
Ulf Möller [Thu, 3 Feb 2000 23:23:57 +0000 (23:23 +0000)]
a short page for "speed"

24 years agoispell (and minor modifications)
Ulf Möller [Thu, 3 Feb 2000 23:23:24 +0000 (23:23 +0000)]
ispell (and minor modifications)

24 years agoDocument hash functions.
Ulf Möller [Thu, 3 Feb 2000 18:22:01 +0000 (18:22 +0000)]
Document hash functions.

24 years agoAdd new -notext option to 'ca', -pubkey option to spkac.
Dr. Stephen Henson [Thu, 3 Feb 2000 02:56:48 +0000 (02:56 +0000)]
Add new -notext option to 'ca', -pubkey option to spkac.

Remove some "WTF??" casts from applications.

Fixes to keep VC++ happy and avoid warnings.

Docs tidy.

24 years agoSome 'const's for BNs.
Bodo Möller [Thu, 3 Feb 2000 01:26:07 +0000 (01:26 +0000)]
Some 'const's for BNs.

24 years agoInclude OpenSSL license.
Bodo Möller [Wed, 2 Feb 2000 23:30:32 +0000 (23:30 +0000)]
Include OpenSSL license.

24 years agoTolerate negative numbers in BN_is_prime.
Bodo Möller [Wed, 2 Feb 2000 22:18:01 +0000 (22:18 +0000)]
Tolerate negative numbers in BN_is_prime.

24 years agorndsort{Miller, Rabin} primality test.
Bodo Möller [Wed, 2 Feb 2000 21:20:44 +0000 (21:20 +0000)]
rndsort{Miller, Rabin} primality test.

24 years agoSupport for "multiply high" instruction, see BN_UMULT_HIGH comment in
Andy Polyakov [Wed, 2 Feb 2000 16:18:12 +0000 (16:18 +0000)]
Support for "multiply high" instruction, see BN_UMULT_HIGH comment in
crypto/bn/bn_lcl.h for further details. It should be noted that for
the moment of this writing the code was tested only on Alpha. If
compiled with DEC C the C implementation exhibits 12% performance
improvement over the crypto/bn/asm/alpha.s (on EV56 box running
AlphaLinux). GNU C is (unfortunately) 8% behind the assembler
implementation. But it's OpenVMS Alpha users who *may* benefit most
as 'apps/openssl speed rsa' exhibits 6 (six) times performance
improvement over the original VMS bignum implementation. Where "*may*"
means "as soon as code is enabled though #define SIXTY_FOUR_BIT and
crypto/bn/asm/vms.mar is skipped."

24 years agoUpdate docs.
Dr. Stephen Henson [Wed, 2 Feb 2000 01:33:28 +0000 (01:33 +0000)]
Update docs.

24 years agoBug fix: BN_is_prime() would fail with a high probability for small
Ulf Möller [Tue, 1 Feb 2000 23:48:55 +0000 (23:48 +0000)]
Bug fix: BN_is_prime() would fail with a high probability for small
primes (negligible for larger ones).

24 years agomatch the prototype
Ulf Möller [Tue, 1 Feb 2000 23:47:24 +0000 (23:47 +0000)]
match the prototype

24 years agoRemove an =over that never ends
Ulf Möller [Tue, 1 Feb 2000 15:33:15 +0000 (15:33 +0000)]
Remove an =over that never ends

24 years agolink to SSL_get_error(3)
Ulf Möller [Tue, 1 Feb 2000 15:05:11 +0000 (15:05 +0000)]
link to SSL_get_error(3)

24 years agodhgen is gone.
Ulf Möller [Tue, 1 Feb 2000 15:00:37 +0000 (15:00 +0000)]
dhgen is gone.

24 years agoCorrect one link.
Richard Levitte [Tue, 1 Feb 2000 12:03:47 +0000 (12:03 +0000)]
Correct one link.

24 years agoIf n0 == d0, we must alway compute 'rem' "by hand"
Bodo Möller [Tue, 1 Feb 2000 11:10:54 +0000 (11:10 +0000)]
If n0 == d0, we must alway compute 'rem' "by hand"

24 years agoTypo in preprocessor symbol.
Bodo Möller [Tue, 1 Feb 2000 08:48:30 +0000 (08:48 +0000)]
Typo in preprocessor symbol.

24 years agoPointer to important manual page that should be written.
Bodo Möller [Tue, 1 Feb 2000 08:48:05 +0000 (08:48 +0000)]
Pointer to important manual page that should be written.

24 years agoChange log entry completed.
Bodo Möller [Tue, 1 Feb 2000 07:50:42 +0000 (07:50 +0000)]
Change log entry completed.

24 years agoundo. I keep confusing my directories. :(
Ulf Möller [Tue, 1 Feb 2000 02:24:10 +0000 (02:24 +0000)]
undo. I keep confusing my directories. :(

24 years agoEBCDIC support.
Ulf Möller [Tue, 1 Feb 2000 02:21:16 +0000 (02:21 +0000)]
EBCDIC support.

Submitted by: Martin Kraemer <martin.kraemer@mch.sni.de>

24 years agoDocument ERR library.
Ulf Möller [Tue, 1 Feb 2000 01:37:00 +0000 (01:37 +0000)]
Document ERR library.

24 years agospelling
Ulf Möller [Tue, 1 Feb 2000 01:35:52 +0000 (01:35 +0000)]
spelling

24 years agoPrint a reassuring message when Configure is done.
Ulf Möller [Tue, 1 Feb 2000 01:35:29 +0000 (01:35 +0000)]
Print a reassuring message when Configure is done.

24 years agoReference for SHA-1.
Bodo Möller [Mon, 31 Jan 2000 09:02:30 +0000 (09:02 +0000)]
Reference for SHA-1.

24 years agoChecked in some junk. Sorry.
Ulf Möller [Sun, 30 Jan 2000 23:46:18 +0000 (23:46 +0000)]
Checked in some junk. Sorry.

24 years agoNote changes.
Ulf Möller [Sun, 30 Jan 2000 23:34:33 +0000 (23:34 +0000)]
Note changes.

24 years agoSeek out and destroy another evil cast.
Ulf Möller [Sun, 30 Jan 2000 23:33:40 +0000 (23:33 +0000)]
Seek out and destroy another evil cast.

24 years agoTypos.
Ulf Möller [Sun, 30 Jan 2000 23:31:01 +0000 (23:31 +0000)]
Typos.

24 years agoSource code cleanups: Use void * rather than char * in lhash,
Ulf Möller [Sun, 30 Jan 2000 22:20:28 +0000 (22:20 +0000)]
Source code cleanups: Use void * rather than char * in lhash,
eliminate some of the -Wcast-qual warnings (debug-ben-strict target)

24 years agoUpdate docs: corrections, turn buffer docs into manpage, fold SHA1
Ulf Möller [Sun, 30 Jan 2000 22:16:47 +0000 (22:16 +0000)]
Update docs: corrections, turn buffer docs into manpage, fold SHA1
pages into one for improved readability, add lhash manpage

24 years agoSynchronise
Richard Levitte [Sun, 30 Jan 2000 13:44:08 +0000 (13:44 +0000)]
Synchronise

24 years agoDocumentation for BN_is_prime_fasttest.
Bodo Möller [Sun, 30 Jan 2000 11:05:39 +0000 (11:05 +0000)]
Documentation for BN_is_prime_fasttest.

24 years agoMake output of "openssl dsaparam 1024" more interesting :-)
Bodo Möller [Sun, 30 Jan 2000 03:32:28 +0000 (03:32 +0000)]
Make output of "openssl dsaparam 1024" more interesting :-)

24 years agoTiny changes to previous patch (the log message was meant to be
Bodo Möller [Sun, 30 Jan 2000 02:40:38 +0000 (02:40 +0000)]
Tiny changes to previous patch (the log message was meant to be
"Make DSA_generate_parameters faster").

24 years agoMake DSA_generate_parameters, and fix a couple of bug
Bodo Möller [Sun, 30 Jan 2000 02:23:03 +0000 (02:23 +0000)]
Make DSA_generate_parameters, and fix a couple of bug
(including another problem in the s3_srvr.c state machine).

24 years agoUpdate docs.
Dr. Stephen Henson [Sun, 30 Jan 2000 01:39:40 +0000 (01:39 +0000)]
Update docs.

24 years agoReimplement so only one synchronous stack is used. The benefit is
Richard Levitte [Sat, 29 Jan 2000 01:22:03 +0000 (01:22 +0000)]
Reimplement so only one synchronous stack is used.  The benefit is
that function pointers are nicely tucker in their structure.