oweals/openssl.git
22 years agoFix crahses and leaks in pkcs12 utility -chain option
Dr. Stephen Henson [Thu, 22 Aug 2002 21:54:24 +0000 (21:54 +0000)]
Fix crahses and leaks in pkcs12 utility -chain option

22 years agoTypo, I assigned the variable ld instead of my_ld...
Richard Levitte [Sun, 18 Aug 2002 21:33:02 +0000 (21:33 +0000)]
Typo, I assigned the variable ld instead of my_ld...

22 years agoSo, I discovered that if you have your $PATH set so a ld different from
Richard Levitte [Sat, 17 Aug 2002 13:46:52 +0000 (13:46 +0000)]
So, I discovered that if you have your $PATH set so a ld different from
GNU ld comes first, checking the usage of collect2 gives that instead of
GNU ld, even if GNU ld would be the one that would get used if we link using
gcc.  It's much better, apparently, to ask gcc directly what the path to
GNU ld is (provided it's there at all and gcc knows about it), and ask
the result if it's a GNU or not.  The bonus is that our GNU ld detection
mechanism got shorter and easier to understand...

22 years agoReorder cleanup sequence in SSL_CTX_free() to leave ex_data for remove_cb().
Lutz Jänicke [Fri, 16 Aug 2002 17:02:30 +0000 (17:02 +0000)]
Reorder cleanup sequence in SSL_CTX_free() to leave ex_data for remove_cb().
Submitted by:
Reviewed by:
PR: 212

22 years agoFix wrong URI.
Lutz Jänicke [Fri, 16 Aug 2002 16:44:54 +0000 (16:44 +0000)]
Fix wrong URI.
Submitted by: Mike Castle <dalgoda@ix.netcom.com>
Reviewed by:
PR: 200

22 years agoisalist was less trustable than I thought (or rather, one can trust it to
Richard Levitte [Fri, 16 Aug 2002 09:41:35 +0000 (09:41 +0000)]
isalist was less trustable than I thought (or rather, one can trust it to
come up with all kinds of names we don't have in our targets).
Besides, our sparcv9 targets currently generate sparcv8 code, I'm told.

22 years agoAdd -lz to the ld flags when the user has chosen to link in zlib *statically*.
Richard Levitte [Fri, 16 Aug 2002 06:15:58 +0000 (06:15 +0000)]
Add -lz to the ld flags when the user has chosen to link in zlib *statically*.
Notified by Doug Kaufman <dkaufman@rahul.net>

22 years agoFix block_size field for CFB and OFB modes: it should be 1.
Dr. Stephen Henson [Fri, 16 Aug 2002 01:38:34 +0000 (01:38 +0000)]
Fix block_size field for CFB and OFB modes: it should be 1.

22 years agoq may be used uninitialised
Richard Levitte [Thu, 15 Aug 2002 22:06:14 +0000 (22:06 +0000)]
q may be used uninitialised

22 years agouse correct function code in error message
Bodo Möller [Thu, 15 Aug 2002 16:17:02 +0000 (16:17 +0000)]
use correct function code in error message

22 years agoIt seems like sun4u doesn't always have a sparcv9 inside. Trust isalist.
Richard Levitte [Thu, 15 Aug 2002 15:21:00 +0000 (15:21 +0000)]
It seems like sun4u doesn't always have a sparcv9 inside.  Trust isalist.
PR: 220

22 years agofix warnings (CHARSET_EBCDIC)
Bodo Möller [Thu, 15 Aug 2002 14:52:22 +0000 (14:52 +0000)]
fix warnings (CHARSET_EBCDIC)

Submitted by: Lorinczy Zsigmond <lzsiga@mail.ahiv.hu>

22 years agofix manpage
Bodo Möller [Thu, 15 Aug 2002 14:22:49 +0000 (14:22 +0000)]
fix manpage

22 years agoI think that's the last forgotten compilation module.
Richard Levitte [Thu, 15 Aug 2002 11:48:58 +0000 (11:48 +0000)]
I think that's the last forgotten compilation module.

22 years agoMissing =back.
Richard Levitte [Thu, 15 Aug 2002 10:59:59 +0000 (10:59 +0000)]
Missing =back.
Part of PR 196

22 years agoAdd a FAQ entry for yet another bc failure.
Richard Levitte [Thu, 15 Aug 2002 10:50:42 +0000 (10:50 +0000)]
Add a FAQ entry for yet another bc failure.
PR: 199

22 years agoYet a couple of modules forgotten. These weren't important for
Richard Levitte [Thu, 15 Aug 2002 09:40:08 +0000 (09:40 +0000)]
Yet a couple of modules forgotten.  These weren't important for
OpenSSL itself, since they aren't used there (yet).  It became quite
visible qhen building a shared library, however...

22 years agoSometimes, the value of the variable containing the compiler call can
Richard Levitte [Thu, 15 Aug 2002 08:29:26 +0000 (08:29 +0000)]
Sometimes, the value of the variable containing the compiler call can
become rather large.  This becomes a problem when the default 1024
character large buffer that WRITE uses isn't enough.  WRITE/SYMBOL
uses a 2048 byte large buffer instead.

22 years agochange how pod2man is handled: explicitly invoke '$PERL' only when
Bodo Möller [Wed, 14 Aug 2002 14:06:11 +0000 (14:06 +0000)]
change how pod2man is handled: explicitly invoke '$PERL' only when
needed, call 'pod2man' directly if this works

PR: 217

22 years agoInstead of returning errors when certain flags are unusable, just ignore them.
Richard Levitte [Wed, 14 Aug 2002 12:18:36 +0000 (12:18 +0000)]
Instead of returning errors when certain flags are unusable, just ignore them.
That will make the test go through even if DH (or in some cases ECDH) aren't
built into OpenSSL.
PR: 216, part 2

22 years agoMake sure that the test input file comes in the same record format as the
Richard Levitte [Wed, 14 Aug 2002 11:09:03 +0000 (11:09 +0000)]
Make sure that the test input file comes in the same record format as the
typical output from a program in C.
PR: 222

22 years agoFix typo in OBJ_txt2obj which incorrectly passed the content
Dr. Stephen Henson [Wed, 14 Aug 2002 00:50:35 +0000 (00:50 +0000)]
Fix typo in OBJ_txt2obj which incorrectly passed the content
length, instead of the encoding length to d2i_ASN1_OBJECT.

This wasn't visible before becuse ASN1_get_object() used
to read past the length of the supplied buffer.

22 years agoSome files deserve to be ignored
Richard Levitte [Tue, 13 Aug 2002 22:41:39 +0000 (22:41 +0000)]
Some files deserve to be ignored

22 years agoThis commit was manufactured by cvs2svn to create branch
cvs2svn [Tue, 13 Aug 2002 22:41:21 +0000 (22:41 +0000)]
This commit was manufactured by cvs2svn to create branch
'OpenSSL_0_9_7-stable'.

22 years agoSome files deserve to be ignored
Richard Levitte [Tue, 13 Aug 2002 22:41:18 +0000 (22:41 +0000)]
Some files deserve to be ignored

22 years agoMerge in demo engines from 0.9.7-stable.
Richard Levitte [Tue, 13 Aug 2002 12:30:27 +0000 (12:30 +0000)]
Merge in demo engines from 0.9.7-stable.

22 years agoOK, I've amused myself with making sure the engines that have been
Richard Levitte [Tue, 13 Aug 2002 12:26:41 +0000 (12:26 +0000)]
OK, I've amused myself with making sure the engines that have been
contributed TO WORK WITH 0.9.7 can be built as dynamically loadable
libraries.

For now, they're not included in crypto/engine/ since 0.9.7 is in
feature freeze.  Further discussion might change that, but don't hold
your breath.

22 years agoExport text symbols as well (AIX experts might need to correct me
Richard Levitte [Tue, 13 Aug 2002 12:19:55 +0000 (12:19 +0000)]
Export text symbols as well (AIX experts might need to correct me
here).

22 years agoComma forgotten.
Richard Levitte [Tue, 13 Aug 2002 11:49:47 +0000 (11:49 +0000)]
Comma forgotten.

22 years agoOh, ec2_smpt.c is #included by ec2_smpl.c!
Richard Levitte [Tue, 13 Aug 2002 11:47:26 +0000 (11:47 +0000)]
Oh, ec2_smpt.c is #included by ec2_smpl.c!

22 years agoDo not include openssl/ripemd.h when the RIPEMD algorithm has been deselected.
Richard Levitte [Mon, 12 Aug 2002 21:22:05 +0000 (21:22 +0000)]
Do not include openssl/ripemd.h when the RIPEMD algorithm has been deselected.
PR: 216, point 1

22 years agoDo not include openssl/ripemd.h when the RIPEMD algorithm has been deselected.
Richard Levitte [Mon, 12 Aug 2002 21:22:00 +0000 (21:22 +0000)]
Do not include openssl/ripemd.h when the RIPEMD algorithm has been deselected.
PR: 216, point 1

22 years agoDo not use the word 'modulus', which is a class template name in VC++ 6.0/SP5.
Richard Levitte [Mon, 12 Aug 2002 21:20:30 +0000 (21:20 +0000)]
Do not use the word 'modulus', which is a class template name in VC++ 6.0/SP5.
PR: 216, point 3

22 years agoDo not use the word 'modulus', which is a class template name in VC++ 6.0/SP5.
Richard Levitte [Mon, 12 Aug 2002 21:20:25 +0000 (21:20 +0000)]
Do not use the word 'modulus', which is a class template name in VC++ 6.0/SP5.
PR: 216, point 3

22 years agoScripts for testing ECC ciphersuites.
Bodo Möller [Mon, 12 Aug 2002 15:18:48 +0000 (15:18 +0000)]
Scripts for testing ECC ciphersuites.

Submitted by: Sun Microsystems Labs

22 years agoA few files in the ENGINE and EVP sections forgotten.
Richard Levitte [Mon, 12 Aug 2002 13:31:57 +0000 (13:31 +0000)]
A few files in the ENGINE and EVP sections forgotten.

22 years agofix previous commit (there's no SSLEAY_VERSION_TEXT)
Bodo Möller [Mon, 12 Aug 2002 11:21:02 +0000 (11:21 +0000)]
fix previous commit (there's no SSLEAY_VERSION_TEXT)

22 years agoremove comment
Bodo Möller [Mon, 12 Aug 2002 08:54:40 +0000 (08:54 +0000)]
remove comment

Submitted by: Douglas Stebila

22 years agoremove debug messages
Bodo Möller [Mon, 12 Aug 2002 08:52:23 +0000 (08:52 +0000)]
remove debug messages

Submitted by: Douglas Stebila

22 years agofix comment
Bodo Möller [Mon, 12 Aug 2002 08:51:30 +0000 (08:51 +0000)]
fix comment

Submitted by: Douglas Stebila

22 years agoget rid of EVP_PKEY_ECDSA (now we have EVP_PKEY_EC instead)
Bodo Möller [Mon, 12 Aug 2002 08:47:41 +0000 (08:47 +0000)]
get rid of EVP_PKEY_ECDSA (now we have EVP_PKEY_EC instead)

Submitted by: Nils Larsch

22 years agoadd 0.9.6g information
Bodo Möller [Mon, 12 Aug 2002 08:45:00 +0000 (08:45 +0000)]
add 0.9.6g information

22 years agoadd 0.9.6g information
Bodo Möller [Mon, 12 Aug 2002 08:43:32 +0000 (08:43 +0000)]
add 0.9.6g information

22 years agoUpdate with the status for 0.9.6g.
Richard Levitte [Mon, 12 Aug 2002 06:55:02 +0000 (06:55 +0000)]
Update with the status for 0.9.6g.

22 years agoUpdate with the status for 0.9.6g.
Richard Levitte [Mon, 12 Aug 2002 06:54:54 +0000 (06:54 +0000)]
Update with the status for 0.9.6g.

22 years agoUpdate with the status for 0.9.6g.
Richard Levitte [Mon, 12 Aug 2002 06:06:27 +0000 (06:06 +0000)]
Update with the status for 0.9.6g.

22 years agoIn case of shared libraries, we might run one version of the
Richard Levitte [Sun, 11 Aug 2002 21:48:44 +0000 (21:48 +0000)]
In case of shared libraries, we might run one version of the
application with a different version of the library.  Detect if there
is a difference of versions, and print both versions in that case.
This might prove to be a good enough debugging tool in case of doubt.

22 years agoOne more file to compile on VMS as well.
Richard Levitte [Sun, 11 Aug 2002 12:27:28 +0000 (12:27 +0000)]
One more file to compile on VMS as well.

22 years agoRemove clashes between symbols that have the same name except for
Richard Levitte [Sun, 11 Aug 2002 12:23:08 +0000 (12:23 +0000)]
Remove clashes between symbols that have the same name except for
casing.

22 years agoMore long symbols to shorten.
Richard Levitte [Sun, 11 Aug 2002 11:50:32 +0000 (11:50 +0000)]
More long symbols to shorten.

22 years agomem* functions are declared in string.h.
Richard Levitte [Sat, 10 Aug 2002 01:36:14 +0000 (01:36 +0000)]
mem* functions are declared in string.h.

22 years agoSynchronise tests with Unix.
Richard Levitte [Sat, 10 Aug 2002 01:35:10 +0000 (01:35 +0000)]
Synchronise tests with Unix.

22 years agoA new header.
Richard Levitte [Fri, 9 Aug 2002 22:34:57 +0000 (22:34 +0000)]
A new header.

22 years agoSynchronise with Unix.
Richard Levitte [Fri, 9 Aug 2002 22:32:32 +0000 (22:32 +0000)]
Synchronise with Unix.
(I expect the next run will generate lots of errors on VMS :-)).

22 years agoMore long names to shorten.
Richard Levitte [Fri, 9 Aug 2002 22:23:33 +0000 (22:23 +0000)]
More long names to shorten.

22 years ago.cvsignore for crypto/ecdh
Bodo Möller [Fri, 9 Aug 2002 12:17:03 +0000 (12:17 +0000)]
.cvsignore for crypto/ecdh

22 years agomake update
Bodo Möller [Fri, 9 Aug 2002 12:16:15 +0000 (12:16 +0000)]
make update

22 years agofix warnings
Bodo Möller [Fri, 9 Aug 2002 11:58:28 +0000 (11:58 +0000)]
fix warnings

22 years agoadd field type to text output
Bodo Möller [Fri, 9 Aug 2002 10:44:44 +0000 (10:44 +0000)]
add field type to text output
don't print seed value as a number (leading zeros must not be removed)

Submitted by: Nils Larsch

22 years agouse 0, not NULL
Bodo Möller [Fri, 9 Aug 2002 09:39:53 +0000 (09:39 +0000)]
use 0, not NULL

Submitted by: Nils Larsch

22 years agoECDH engine support
Bodo Möller [Fri, 9 Aug 2002 09:39:11 +0000 (09:39 +0000)]
ECDH engine support

Submitted by: Douglas Stebila

22 years agoECC ciphersuite support
Bodo Möller [Fri, 9 Aug 2002 08:56:08 +0000 (08:56 +0000)]
ECC ciphersuite support

Submitted by: Douglas Stebila <douglas.stebila@sun.com>
(Authors: Vipul Gupta and Sumit Gupta, Sun Microsystems Laboratories)

22 years agoWhen we want to give a -f argument to $(MAKE), we'd better make sure the
Richard Levitte [Fri, 9 Aug 2002 08:50:30 +0000 (08:50 +0000)]
When we want to give a -f argument to $(MAKE), we'd better make sure the
variable doesn't already contain a -f argument.
PR: 203, part 4

22 years agoWhen we want to give a -f argument to $(MAKE), we'd better make sure the
Richard Levitte [Fri, 9 Aug 2002 08:49:46 +0000 (08:49 +0000)]
When we want to give a -f argument to $(MAKE), we'd better make sure the
variable doesn't already contain a -f argument.
PR: 203, part 4

22 years agoAdd ECDH support.
Bodo Möller [Fri, 9 Aug 2002 08:43:04 +0000 (08:43 +0000)]
Add ECDH support.

Additional changes:
 - use EC_GROUP_get_degree() in apps/req.c
 - add ECDSA and ECDH to apps/speed.c
 - adds support for EC curves over binary fields to ECDSA
 - new function EC_KEY_up_ref() in crypto/ec/ec_key.c
 - reorganize crypto/ecdsa/ecdsatest.c
 - add engine support for ECDH
 - fix a few bugs in ECDSA engine support

Submitted by: Douglas Stebila <douglas.stebila@sun.com>

22 years agoParse version numbers prefixed with text (egcs does that, even with
Richard Levitte [Fri, 9 Aug 2002 07:32:31 +0000 (07:32 +0000)]
Parse version numbers prefixed with text (egcs does that, even with
-dumpversion).
PR: 203, part 1

22 years agoParse version numbers prefixed with text (egcs does that, even with
Richard Levitte [Fri, 9 Aug 2002 07:32:24 +0000 (07:32 +0000)]
Parse version numbers prefixed with text (egcs does that, even with
-dumpversion).
PR: 203, part 1

22 years ago0.9.6f is released
Richard Levitte [Thu, 8 Aug 2002 22:56:05 +0000 (22:56 +0000)]
0.9.6f is released

22 years ago0.9.6f is released
Richard Levitte [Thu, 8 Aug 2002 22:55:28 +0000 (22:55 +0000)]
0.9.6f is released

22 years agouse a generic EC_KEY structure (EC keys are not ECDSA specific)
Bodo Möller [Wed, 7 Aug 2002 10:49:54 +0000 (10:49 +0000)]
use a generic EC_KEY structure (EC keys are not ECDSA specific)

Submitted by: Nils Larsch

22 years agoavoid SIGSEGV
Bodo Möller [Wed, 7 Aug 2002 07:53:47 +0000 (07:53 +0000)]
avoid SIGSEGV

Submitted by: Nils Larsch, Douglas Stebila

22 years agoThese are updates/fixes to DH/DSA/RAND docs based on the fixes to the RSA
Geoff Thorpe [Mon, 5 Aug 2002 16:28:59 +0000 (16:28 +0000)]
These are updates/fixes to DH/DSA/RAND docs based on the fixes to the RSA
docs. There were a couple of other places where the docs were not
synchronised with the API that are now fixed. One or two still remain to be
fixed though ...

22 years agoThese are updates/fixes to DH/DSA/RAND docs based on the fixes to the RSA
Geoff Thorpe [Mon, 5 Aug 2002 16:27:01 +0000 (16:27 +0000)]
These are updates/fixes to DH/DSA/RAND docs based on the fixes to the RSA
docs. There were a couple of other places (including RSA) where the docs
were not quite synchronised with the API that are now fixed. One or two
still remain to be fixed though ...

22 years agotypo fix
Geoff Thorpe [Mon, 5 Aug 2002 02:55:27 +0000 (02:55 +0000)]
typo fix

22 years agotypo fix
Geoff Thorpe [Mon, 5 Aug 2002 02:54:57 +0000 (02:54 +0000)]
typo fix

22 years agoVarious parts of the RSA documentation were inaccurate and out of date and
Geoff Thorpe [Sun, 4 Aug 2002 21:10:05 +0000 (21:10 +0000)]
Various parts of the RSA documentation were inaccurate and out of date and
this fixes those that I'm currently aware of. In particular, the ENGINE
interference in the RSA API has hopefully been clarified. This still needs
to be done for other areas of the API ...

22 years agoVarious parts of the RSA documentation were inaccurate and out of date and
Geoff Thorpe [Sun, 4 Aug 2002 21:08:36 +0000 (21:08 +0000)]
Various parts of the RSA documentation were inaccurate and out of date and
this fixes those that I'm currently aware of. In particular, the ENGINE
interference in the RSA API has hopefully been clarified. This still needs
to be done for other areas of the API ...

22 years agoA single monolithic man page for the ENGINE stuff. This is a rough
Geoff Thorpe [Sun, 4 Aug 2002 21:03:04 +0000 (21:03 +0000)]
A single monolithic man page for the ENGINE stuff. This is a rough
first-cut but provides better documentation than having nothing on the
ENGINE API.

22 years agoThis commit was manufactured by cvs2svn to create branch
cvs2svn [Sun, 4 Aug 2002 20:57:20 +0000 (20:57 +0000)]
This commit was manufactured by cvs2svn to create branch
'OpenSSL_0_9_7-stable'.

22 years agoA single monolithic man page for the ENGINE stuff. This is a rough
Geoff Thorpe [Sun, 4 Aug 2002 20:57:19 +0000 (20:57 +0000)]
A single monolithic man page for the ENGINE stuff. This is a rough
first-cut but provides better documentation than having nothing on the
ENGINE API.

22 years agoFix "make install_docs" (and thus "make install").
Geoff Thorpe [Sun, 4 Aug 2002 20:40:23 +0000 (20:40 +0000)]
Fix "make install_docs" (and thus "make install").

22 years agoFix "make install_docs" (and thus "make install").
Geoff Thorpe [Sun, 4 Aug 2002 20:38:54 +0000 (20:38 +0000)]
Fix "make install_docs" (and thus "make install").

22 years agooops -- must use EVP_MD_size, not EVP_MD_block_size
Bodo Möller [Sat, 3 Aug 2002 18:50:17 +0000 (18:50 +0000)]
oops -- must use EVP_MD_size, not EVP_MD_block_size

22 years agooops -- must use EVP_MD_size, not EVP_MD_block_size
Bodo Möller [Sat, 3 Aug 2002 18:49:39 +0000 (18:49 +0000)]
oops -- must use EVP_MD_size, not EVP_MD_block_size

22 years agooops, undo previous change (was just for testing)
Bodo Möller [Sat, 3 Aug 2002 18:28:34 +0000 (18:28 +0000)]
oops, undo previous change (was just for testing)

22 years agofix bn_expand2
Bodo Möller [Sat, 3 Aug 2002 18:27:47 +0000 (18:27 +0000)]
fix bn_expand2

22 years agouse bn_wexpand instead of bn_expand2 (the latter is not needed here,
Bodo Möller [Sat, 3 Aug 2002 17:51:29 +0000 (17:51 +0000)]
use bn_wexpand instead of bn_expand2 (the latter is not needed here,
and it does not yet work correctly)

Submitted by: Douglas Stebila

22 years agoFix typo
Dr. Stephen Henson [Fri, 2 Aug 2002 19:03:41 +0000 (19:03 +0000)]
Fix typo

22 years agoFix typo
Dr. Stephen Henson [Fri, 2 Aug 2002 18:58:33 +0000 (18:58 +0000)]
Fix typo

22 years agoFix the ASN1 sanity check: correct header length
Dr. Stephen Henson [Fri, 2 Aug 2002 18:48:55 +0000 (18:48 +0000)]
Fix the ASN1 sanity check: correct header length
calculation and check overflow against LONG_MAX.

22 years agoFix the ASN1 sanity check: correct header length
Dr. Stephen Henson [Fri, 2 Aug 2002 18:42:40 +0000 (18:42 +0000)]
Fix the ASN1 sanity check: correct header length
calculation and check overflow against LONG_MAX.

22 years agodisable Sun divison algorithm by default
Bodo Möller [Fri, 2 Aug 2002 18:26:02 +0000 (18:26 +0000)]
disable Sun divison algorithm by default

22 years agofix bn_expand2
Bodo Möller [Fri, 2 Aug 2002 18:23:55 +0000 (18:23 +0000)]
fix bn_expand2

22 years agooptical changes
Bodo Möller [Fri, 2 Aug 2002 17:25:05 +0000 (17:25 +0000)]
optical changes

22 years agotypo
Bodo Möller [Fri, 2 Aug 2002 15:28:31 +0000 (15:28 +0000)]
typo

22 years agoremove obsolete part of comment
Bodo Möller [Fri, 2 Aug 2002 15:13:10 +0000 (15:13 +0000)]
remove obsolete part of comment

22 years agoremove obsoleted disabled code
Bodo Möller [Fri, 2 Aug 2002 15:07:08 +0000 (15:07 +0000)]
remove obsoleted disabled code

22 years agoLet BN_rand_range() abort with an error after 100 iterations
Bodo Möller [Fri, 2 Aug 2002 15:02:03 +0000 (15:02 +0000)]
Let BN_rand_range() abort with an error after 100 iterations
without success.

22 years agoupdate
Bodo Möller [Fri, 2 Aug 2002 14:58:09 +0000 (14:58 +0000)]
update

22 years agoChange BN_mod_sqrt() so that it verifies that the input value is
Bodo Möller [Fri, 2 Aug 2002 14:57:53 +0000 (14:57 +0000)]
Change BN_mod_sqrt() so that it verifies that the input value is
really the square of the return value.