oweals/openssl.git
23 years agoIf OPENSSL_BUILD_SHLIBCRYPTO (for files that end up as libcrypto
Richard Levitte [Sun, 31 Dec 2000 00:23:17 +0000 (00:23 +0000)]
If OPENSSL_BUILD_SHLIBCRYPTO (for files that end up as libcrypto
objects) or OPENSSL_BUILD_SHLIBSSL (for files that end up as libssl
objects) is defined, redefine OPENSSL_EXTERN to be OPENSSL_EXPORT.
This is actually only important on Win32, and can safely be ignored in
all other cases, at least for now.

23 years agoDefine OPENSSL_EXPORT and OPENSSL_IMPORT and give OPENSSL_EXTERN the
Richard Levitte [Sun, 31 Dec 2000 00:15:57 +0000 (00:15 +0000)]
Define OPENSSL_EXPORT and OPENSSL_IMPORT and give OPENSSL_EXTERN the
default value OPENSSL_IMPORT.  Explain the use of all those macros.

23 years agoRemove anything connected to RSAref, since that's gone by now.
Richard Levitte [Sun, 31 Dec 2000 00:06:47 +0000 (00:06 +0000)]
Remove anything connected to RSAref, since that's gone by now.
Add the C macros OPENSSL_BUILD_SHLIBCRYPTO and OPENSSL_BUILD_SHLIBSSL
to the build of the object files as appropriate for each library.

23 years agoUhmm, the keyword TRUE does not exist of course...
Richard Levitte [Sun, 31 Dec 2000 00:00:02 +0000 (00:00 +0000)]
Uhmm, the keyword TRUE does not exist of course...

23 years agoDelete PKCS#12 redundant files.
Dr. Stephen Henson [Sat, 30 Dec 2000 12:26:33 +0000 (12:26 +0000)]
Delete PKCS#12 redundant files.

23 years agoASN1_ITEM versions of ASN1_d2i_{fp, bio} and replacement of
Dr. Stephen Henson [Sat, 30 Dec 2000 02:40:26 +0000 (02:40 +0000)]
ASN1_ITEM versions of ASN1_d2i_{fp, bio} and replacement of
most of the old wrappers. A few of the old versions remain
because they are non standard and the corresponding ASN1
code has not been reimplemented yet.

23 years agoReplace the old ASN1_dup() with ASN1_item_dup() and
Dr. Stephen Henson [Fri, 29 Dec 2000 18:23:55 +0000 (18:23 +0000)]
Replace the old ASN1_dup() with ASN1_item_dup() and
zap some evil function pointers casts along the way...

23 years agoCheck for deselection of KRB5. In fact, skip it completely on VMS for now...
Richard Levitte [Fri, 29 Dec 2000 01:54:14 +0000 (01:54 +0000)]
Check for deselection of KRB5.  In fact, skip it completely on VMS for now...

23 years agoCorrect a typo.
Richard Levitte [Fri, 29 Dec 2000 01:41:37 +0000 (01:41 +0000)]
Correct a typo.

23 years ago"make update" plus a rewrite of both .num files.
Richard Levitte [Fri, 29 Dec 2000 00:19:12 +0000 (00:19 +0000)]
"make update" plus a rewrite of both .num files.

23 years agoEnhancements to mkdef.pl:
Richard Levitte [Fri, 29 Dec 2000 00:05:14 +0000 (00:05 +0000)]
Enhancements to mkdef.pl:

* detect "unknown" algorithms (any C macro starting with NO_ that is
  not explicitely mentioned in mkdef.pl as a known algorithm) and
  report.
* add a number of algorithms that can be deselected.
* look in ssl/kssl.h as well.
* accept multiple whitespace (not just one SPC) in preprocessor lines.

23 years agoASN1_ITEM version of ASN1_dup(). Might want
Dr. Stephen Henson [Thu, 28 Dec 2000 22:41:46 +0000 (22:41 +0000)]
ASN1_ITEM version of ASN1_dup(). Might want
something more efficient later...

23 years agoUpdate VMS build procedures to match the current status.
Richard Levitte [Thu, 28 Dec 2000 22:26:11 +0000 (22:26 +0000)]
Update VMS build procedures to match the current status.

23 years agoAdd NO_ASN1_OLD to remove some old style functions:
Dr. Stephen Henson [Thu, 28 Dec 2000 22:24:50 +0000 (22:24 +0000)]
Add NO_ASN1_OLD to remove some old style functions:
currently OpenSSL itself wont compile with this set
because some old style stuff remains.

Change old functions X509_sign(), X509_verify() etc
to use new item based functions.

Replace OCSP function declarations with DECLARE macros.

23 years agoASN1_ITEM versions of sign, verify, pack and unpack.
Dr. Stephen Henson [Thu, 28 Dec 2000 19:18:48 +0000 (19:18 +0000)]
ASN1_ITEM versions of sign, verify, pack and unpack.

The old function pointer versions will eventually go
away.

23 years agoNew OCSP extension functions.
Dr. Stephen Henson [Thu, 28 Dec 2000 01:05:05 +0000 (01:05 +0000)]
New OCSP extension functions.

23 years agoGet rid of unused error code.
Bodo Möller [Wed, 27 Dec 2000 23:42:20 +0000 (23:42 +0000)]
Get rid of unused error code.

23 years agoAvoid compiler warnings in hw_ubsec.c: unused static
Dr. Stephen Henson [Wed, 27 Dec 2000 19:20:14 +0000 (19:20 +0000)]
Avoid compiler warnings in hw_ubsec.c: unused static
functions and signed/unsigned mismatch.

This will of course change if some of the unused functions
suddenly get used...

23 years agoX509V3_add_i2d() needs to be able to allocate a
Dr. Stephen Henson [Wed, 27 Dec 2000 13:42:43 +0000 (13:42 +0000)]
X509V3_add_i2d() needs to be able to allocate a
STACK_OF(X509_EXTENSION) so it should be passed
STACK_OF(X509_EXTENSION) ** in the first argument.

Modify wrappers appropriately.

23 years agoFinish SSL_peek/SSL_pending fixes.
Bodo Möller [Tue, 26 Dec 2000 12:07:23 +0000 (12:07 +0000)]
Finish SSL_peek/SSL_pending fixes.

23 years agoFix SSL_peek and SSL_pending.
Bodo Möller [Mon, 25 Dec 2000 18:40:46 +0000 (18:40 +0000)]
Fix SSL_peek and SSL_pending.

23 years agoNew function X509V3_add_i2d() this is used for
Dr. Stephen Henson [Sun, 24 Dec 2000 18:02:33 +0000 (18:02 +0000)]
New function X509V3_add_i2d() this is used for
encoding, replacing and deleting extensions.

Fix X509V3_get_d2i() so it uses takes note of
new critical behaviour.

23 years agoRe-order the options in tunala and add command switches like s_server for
Geoff Thorpe [Thu, 21 Dec 2000 02:49:13 +0000 (02:49 +0000)]
Re-order the options in tunala and add command switches like s_server for
disabling different SSL/TLS protocol versions.

23 years agoVarious Win32 related fixes. Doesn't compile yet on
Dr. Stephen Henson [Thu, 21 Dec 2000 01:38:55 +0000 (01:38 +0000)]
Various Win32 related fixes. Doesn't compile yet on
Win32 but it is getting there...

Update mkdef.pl to handle ASN1_ANY and fix headers.

Stop various VC++ warnings.

Include some fixes from "Peter 'Luna' Runestig"
<peter@runestig.com>

Remove external declaration for des_set_weak_key_flag:
it doesn't exist.

23 years agoThis adds support to 'tunala' for supplying DH parameters (without which it
Geoff Thorpe [Wed, 20 Dec 2000 22:14:23 +0000 (22:14 +0000)]
This adds support to 'tunala' for supplying DH parameters (without which it
will not support EDH cipher suites). The parameters can either be loaded
from a file (via "-dh_file"), generated by the application on start-up
("-dh_special generate"), or be standard DH parameters (as used in
s_server, etc).

23 years agoSome minor changes to the "tunala" demo.
Geoff Thorpe [Wed, 20 Dec 2000 19:30:19 +0000 (19:30 +0000)]
Some minor changes to the "tunala" demo.

* Seal off some buffer functions so that only the higher-level IO functions
  are exposed.

* Using the above change to buffer, add support to tunala for displaying
  traffic totals when a tunnel closes. Useful in debugging and analysis -
  you get to see the total encrypted traffic versus the total tunneled
  traffic. This shows not only how much expansion your data suffers from
  SSL (a lot if you send/receive a few bytes at a time), but also the
  overhead of SSL handshaking relative to the payload sent through the
  tunnel. This is controlled by the "-out_totals" switch to tunala.

* Fix and tweak some bits in the README.

Eg. sample output of "-out_totals" from a tunnel client when tunneling a brief
"telnet" session.

Tunnel closing, traffic stats follow
    SSL (network) traffic to/from server;     7305 bytes in,     3475 bytes out
    tunnelled data to/from server;            4295 bytes in,      186 bytes out

23 years agobranches have been merged.
Ulf Möller [Wed, 20 Dec 2000 15:42:01 +0000 (15:42 +0000)]
branches have been merged.

23 years agoStop build when an error occurs.
Ulf Möller [Wed, 20 Dec 2000 15:23:31 +0000 (15:23 +0000)]
Stop build when an error occurs.
"Peter 'Luna' Runestig" <peter+openssl-dev@runestig.com>

23 years agoDon't access non-existing element buf[256], use buf[255] instead.
Bodo Möller [Wed, 20 Dec 2000 10:11:06 +0000 (10:11 +0000)]
Don't access non-existing element buf[256], use buf[255] instead.

Submitted by: draslar <draslar@elray.ch>

23 years agoSplit a CHANGES entry so that one of the halves matches the
Bodo Möller [Wed, 20 Dec 2000 10:09:08 +0000 (10:09 +0000)]
Split a CHANGES entry so that one of the halves matches the
corresponding new entry in the OpenSSL_0_9_6-stable branch.

23 years agoFixes to OCSP print code.
Dr. Stephen Henson [Wed, 20 Dec 2000 00:46:44 +0000 (00:46 +0000)]
Fixes to OCSP print code.

Don't try to print request certificates if signature is not present.

Remove unnecessary test for certificates being NULL.

Fix typos in printed output.

Tidy up output.

Fix for typo in OCSP_SERVICELOC ASN1 template.

Also give a bit more info in CHANGES about the ASN1 revision.

23 years agofix indentation
Bodo Möller [Tue, 19 Dec 2000 12:39:45 +0000 (12:39 +0000)]
fix indentation

23 years agoDon't hold CRYPTO_LOCK_RSA during time-consuming operations.
Bodo Möller [Tue, 19 Dec 2000 12:31:41 +0000 (12:31 +0000)]
Don't hold CRYPTO_LOCK_RSA during time-consuming operations.

23 years agoAdd a comment (intended change)
Bodo Möller [Mon, 18 Dec 2000 16:39:00 +0000 (16:39 +0000)]
Add a comment (intended change)

23 years agoImport s2_pkt.c wbuf fixes from OpenSSL_0_9_6-stable branch.
Bodo Möller [Mon, 18 Dec 2000 11:35:32 +0000 (11:35 +0000)]
Import s2_pkt.c wbuf fixes from OpenSSL_0_9_6-stable branch.

23 years agoComment correction.
Bodo Möller [Mon, 18 Dec 2000 09:18:22 +0000 (09:18 +0000)]
Comment correction.

23 years agoObtain lock CRYPTO_LOCK_RSA before creating BN_MONT_CTX
Bodo Möller [Mon, 18 Dec 2000 09:00:48 +0000 (09:00 +0000)]
Obtain lock CRYPTO_LOCK_RSA before creating BN_MONT_CTX
structures and setting rsa->_method_mod_{n,p,q}.

Submitted by: "Reddie, Steven" <Steven.Reddie@ca.com>

23 years agoundo previous change: '-prexit' is already available in current versions of s_client
Bodo Möller [Mon, 18 Dec 2000 08:18:07 +0000 (08:18 +0000)]
undo previous change: '-prexit' is already available in current versions of s_client

23 years agoRedo OCSP response printing. Remove duplicate or
Dr. Stephen Henson [Sun, 17 Dec 2000 14:09:43 +0000 (14:09 +0000)]
Redo OCSP response printing. Remove duplicate or
obsolete code. Delete some redundant files.

23 years agoWhen mentioning features that don't exist in current releases of
Bodo Möller [Sun, 17 Dec 2000 00:32:49 +0000 (00:32 +0000)]
When mentioning features that don't exist in current releases of
OpenSSL (such as the new undocumented '-prexit' option to s_client),
the FAQ should point out that they don't: The FAQ is not just part
of the release, it's current version is also published on the web.

23 years agoSimplify preprocessor statements.
Bodo Möller [Sun, 17 Dec 2000 00:30:03 +0000 (00:30 +0000)]
Simplify preprocessor statements.

23 years agoAdd OCSP service locator extension.
Dr. Stephen Henson [Sat, 16 Dec 2000 12:51:58 +0000 (12:51 +0000)]
Add OCSP service locator extension.

23 years agoDo not poll DEVRANDOM if weäre building without an file pointer API.
Richard Levitte [Sat, 16 Dec 2000 11:19:19 +0000 (11:19 +0000)]
Do not poll DEVRANDOM if weäre building without an file pointer API.
Spotted by "David Schwartz" <davids@webmaster.com>.

23 years agoMake sure each FAQ item has an index entry.
Richard Levitte [Sat, 16 Dec 2000 11:14:41 +0000 (11:14 +0000)]
Make sure each FAQ item has an index entry.

23 years agoAdd support for the noCheck OCSP extension. This is
Dr. Stephen Henson [Sat, 16 Dec 2000 01:58:58 +0000 (01:58 +0000)]
Add support for the noCheck OCSP extension. This is
just a NULL and appears in a certificate.

23 years agoMake mkdef.pl parse some ASN1 IMPLEMENT macros.
Dr. Stephen Henson [Sat, 16 Dec 2000 01:19:24 +0000 (01:19 +0000)]
Make mkdef.pl parse some ASN1 IMPLEMENT macros.

Initial support for variables in DEF files.

23 years agoAdd a couple of FAQs.
Dr. Stephen Henson [Sat, 16 Dec 2000 01:07:16 +0000 (01:07 +0000)]
Add a couple of FAQs.

23 years ago"Andrew W. Gray" <agray@iconsinc.com> says /GD is no longer a valid
Ulf Möller [Fri, 15 Dec 2000 20:59:37 +0000 (20:59 +0000)]
"Andrew W. Gray" <agray@iconsinc.com> says /GD is no longer a valid
compiler switch.

23 years agoIf CONF_get_string returns NULL and we want to tolerate this
Bodo Möller [Fri, 15 Dec 2000 16:59:49 +0000 (16:59 +0000)]
If CONF_get_string returns NULL and we want to tolerate this
(e.g., use a default), we have to call ERR_clear_error().

23 years agoLocking issues.
Bodo Möller [Fri, 15 Dec 2000 16:40:35 +0000 (16:40 +0000)]
Locking issues.

23 years agoImplement some standard OCSP extensions in the v3 code. These
Dr. Stephen Henson [Fri, 15 Dec 2000 13:42:00 +0000 (13:42 +0000)]
Implement some standard OCSP extensions in the v3 code. These
are all raw print only extensions at present.

23 years agoThe C version of bn_sub_part_words is needed not only
Bodo Möller [Fri, 15 Dec 2000 10:54:03 +0000 (10:54 +0000)]
The C version of bn_sub_part_words is needed not only
in NO_ASM configurations

23 years agoVery few in the "README" is up-to-date
Bodo Möller [Fri, 15 Dec 2000 10:42:11 +0000 (10:42 +0000)]
Very few in the "README" is up-to-date

23 years agoAdd OCSP nonce extension to supported extensions.
Dr. Stephen Henson [Thu, 14 Dec 2000 23:27:20 +0000 (23:27 +0000)]
Add OCSP nonce extension to supported extensions.

This is a little unusual because it can contain no
structure i.e. the extension OCTET STRING content
octets do not contain a DER encoded structure.

23 years agoMake a note of the new engine.
Geoff Thorpe [Thu, 14 Dec 2000 21:49:48 +0000 (21:49 +0000)]
Make a note of the new engine.

23 years agoThis is an engine contributed by Broadcom - it is meant to support the
Geoff Thorpe [Thu, 14 Dec 2000 21:41:55 +0000 (21:41 +0000)]
This is an engine contributed by Broadcom - it is meant to support the
BCM5805 and BCM5820 units. So far I've merely taken a skim over the code
and changed a few things from their original contributed source
(de-shadowing variables, removing variables from the header, and
re-constifying some functions to remove warnings). If this gives
compilation problems on any system, please let me know. We will hopefully
know for sure whether this actually functions on a system with the relevant
hardware in a day or two.  :-)

23 years agoNew function X509V3_extensions_print() this removes extension duplication
Dr. Stephen Henson [Thu, 14 Dec 2000 18:42:28 +0000 (18:42 +0000)]
New function X509V3_extensions_print() this removes extension duplication
from the print routines.

Reorganisation of OCSP code: initial print routines in ocsp_prn.c. Doesn't
work fully because OCSP extensions aren't reimplemented yet.

Implement some ASN1 functions needed to compile OCSP code.

23 years agotypo
Bodo Möller [Thu, 14 Dec 2000 17:47:29 +0000 (17:47 +0000)]
typo

23 years agoThe BN_mul bug test apparently is no longer needed
Bodo Möller [Thu, 14 Dec 2000 17:46:36 +0000 (17:46 +0000)]
The BN_mul bug test apparently is no longer needed

23 years agoFirst step towards SSL_peek fix.
Bodo Möller [Thu, 14 Dec 2000 17:36:59 +0000 (17:36 +0000)]
First step towards SSL_peek fix.

23 years agoUpdate.
Bodo Möller [Thu, 14 Dec 2000 17:29:33 +0000 (17:29 +0000)]
Update.

23 years agoNew function X509_signature_print() to remove some duplicate
Dr. Stephen Henson [Thu, 14 Dec 2000 00:53:10 +0000 (00:53 +0000)]
New function X509_signature_print() to remove some duplicate
code from certificate, CRL and request printing routines.

23 years agoChange the PKCS7 structure to use SEQUENCE OF for the
Dr. Stephen Henson [Wed, 13 Dec 2000 23:54:30 +0000 (23:54 +0000)]
Change the PKCS7 structure to use SEQUENCE OF for the
authenticated attributes: this is used to retain the
original encoding and not break signatures.

Support for a SET OF which reorders the STACK when
encoding a structure. This will be used with the
PKCS7 code.

23 years agoc&p error spotted by Martin Forssen
Ulf Möller [Wed, 13 Dec 2000 23:00:33 +0000 (23:00 +0000)]
c&p error spotted by Martin Forssen

23 years agoReplace the old style OCSP ASN1 module.
Dr. Stephen Henson [Wed, 13 Dec 2000 18:21:51 +0000 (18:21 +0000)]
Replace the old style OCSP ASN1 module.

23 years agoConstification of the data of a hash table. This means the callback
Richard Levitte [Wed, 13 Dec 2000 17:15:03 +0000 (17:15 +0000)]
Constification of the data of a hash table.  This means the callback
functions need to be constified, and therefore meant a number of easy
changes a little everywhere.

Now, if someone could explain to me why OBJ_dup() cheats...

23 years agoProblem: bn_mul_normal() misbehaves if the size of b is 0.
Richard Levitte [Wed, 13 Dec 2000 15:29:29 +0000 (15:29 +0000)]
Problem: bn_mul_normal() misbehaves if the size of b is 0.
Solution: multiply a with 0, putting the result in r, and return.

23 years agoIn bn_mul_recursive(), make sure the comba routines are only called
Richard Levitte [Wed, 13 Dec 2000 14:47:06 +0000 (14:47 +0000)]
In bn_mul_recursive(), make sure the comba routines are only called
when both a and b are of the exact right size.  This may change to
something better later.

23 years agoWhen using -pedantic, it's a good thing to define PEDANTIC as well.
Richard Levitte [Wed, 13 Dec 2000 14:04:08 +0000 (14:04 +0000)]
When using -pedantic, it's a good thing to define PEDANTIC as well.
Also, define a second debugging configuration without assembler.

23 years agoStop extension creation code core dumping.
Dr. Stephen Henson [Wed, 13 Dec 2000 13:55:03 +0000 (13:55 +0000)]
Stop extension creation code core dumping.

This was caused by no initialising the buffer
to NULL when using the auto allocating version
if i2d.

23 years agoRewrite the extension code to use an ASN1_ITEM structure
Dr. Stephen Henson [Wed, 13 Dec 2000 13:47:33 +0000 (13:47 +0000)]
Rewrite the extension code to use an ASN1_ITEM structure
for its ASN1 operations as well as the old style function
pointers (i2d, d2i, new, free). Change standard extensions
to support this.

Fix a warning in BN_mul(), bn_mul.c about uninitialised 'j'.

23 years agoAdd description of SSL_[CTX_]_check_private_key().
Lutz Jänicke [Tue, 12 Dec 2000 21:06:29 +0000 (21:06 +0000)]
Add description of SSL_[CTX_]_check_private_key().

23 years agoSSL_new() may potentially add a certfificate. Therefore, wen
Richard Levitte [Tue, 12 Dec 2000 08:08:53 +0000 (08:08 +0000)]
SSL_new() may potentially add a certfificate.  Therefore, wen
duplicating the certificate that is in the original SSL, remove the
one that SSL_new() provided, if any.
Spotted by: Mike Zeoli <zeoli@roguewave.com>

23 years agoChange/add comments
Bodo Möller [Mon, 11 Dec 2000 17:28:07 +0000 (17:28 +0000)]
Change/add comments

23 years agoStop on bntest error.
Ulf Möller [Sun, 10 Dec 2000 04:25:57 +0000 (04:25 +0000)]
Stop on bntest error.

23 years ago*** empty log message ***
Ulf Möller [Sun, 10 Dec 2000 04:22:45 +0000 (04:22 +0000)]
*** empty log message ***

23 years agoMake TYPE_RSA the default type instead of just setting it when -new is
Richard Levitte [Sat, 9 Dec 2000 11:11:35 +0000 (11:11 +0000)]
Make TYPE_RSA the default type instead of just setting it when -new is
given.  That also allows the arguments to come in any order (-new
last, for example).

23 years agoUse continuation lines in test/bctest as far as it is possible
Bodo Möller [Sat, 9 Dec 2000 10:07:47 +0000 (10:07 +0000)]
Use continuation lines in test/bctest as far as it is possible
to dermine what the expression should look like.
Apparently CVS does not like lines longer than about 2^10 characters.

23 years agoDon't throw away bctest's error messages.
Bodo Möller [Sat, 9 Dec 2000 10:05:03 +0000 (10:05 +0000)]
Don't throw away bctest's error messages.

23 years agoTest for SCO bc bug
Ulf Möller [Sat, 9 Dec 2000 02:45:59 +0000 (02:45 +0000)]
Test for SCO bc bug

23 years agoNext step in tidying up the LHASH code.
Geoff Thorpe [Fri, 8 Dec 2000 20:02:01 +0000 (20:02 +0000)]
Next step in tidying up the LHASH code.

DECLARE/IMPLEMENT macros now exist to create type (and prototype) safe
wrapper functions that avoid the use of function pointer casting yet retain
type-safety for type-specific callbacks. However, most of the usage within
OpenSSL itself doesn't really require the extra function because the hash
and compare callbacks are internal functions declared only for use by the
hash table. So this change catches all those cases and reimplements the
functions using the base-level LHASH prototypes and does per-variable
casting inside those functions to convert to the appropriate item type.

The exception so far is in ssl_lib.c where the hash and compare callbacks
are not static - they're exposed in ssl.h so their prototypes should not be
changed. In this last case, the IMPLEMENT_LHASH_*** macros have been left
intact.

23 years agobn_part_sub_word prototype.
Ulf Möller [Fri, 8 Dec 2000 19:10:33 +0000 (19:10 +0000)]
bn_part_sub_word prototype.

23 years agoMerge from the ASN1 branch of new ASN1 code
Dr. Stephen Henson [Fri, 8 Dec 2000 19:09:35 +0000 (19:09 +0000)]
Merge from the ASN1 branch of new ASN1 code
to main trunk.

Lets see if the makes it to openssl-cvs :-)

23 years agoAdd manual pages for certficate/key loading and friends.
Lutz Jänicke [Fri, 8 Dec 2000 14:29:13 +0000 (14:29 +0000)]
Add manual pages for certficate/key loading and friends.

23 years agoPlaceholder for SCO bc bug detection
Bodo Möller [Fri, 8 Dec 2000 08:30:38 +0000 (08:30 +0000)]
Placeholder for SCO bc bug detection

23 years agoFix some things that look like bugs.
Bodo Möller [Thu, 7 Dec 2000 22:06:09 +0000 (22:06 +0000)]
Fix some things that look like bugs.

One problem that looked like a problem in bn_recp.c at first turned
out to be a BN_mul bug.  An example is given in bn_recp.c; finding
the bug responsible for this is left as an exercise.

23 years agoTEST_MUL and TEST_SQR added.
Bodo Möller [Thu, 7 Dec 2000 11:11:10 +0000 (11:11 +0000)]
TEST_MUL and TEST_SQR added.

23 years agoSign-related fixes (and tests).
Bodo Möller [Thu, 7 Dec 2000 08:48:58 +0000 (08:48 +0000)]
Sign-related fixes (and tests).

BN_mod_exp_mont does not work properly yet if modulus m
is negative (we want computations to be carried out
modulo |m|).

23 years agoMove 'q->neg = 0' to those places where it is needed
Bodo Möller [Thu, 7 Dec 2000 07:55:26 +0000 (07:55 +0000)]
Move 'q->neg = 0' to those places where it is needed
(just in cases someone uses a negative modulus)

23 years agoBN_mod_sqrt documentation/comment
Bodo Möller [Wed, 6 Dec 2000 21:33:58 +0000 (21:33 +0000)]
BN_mod_sqrt documentation/comment

23 years agoDon't check for bc at all. We can now run a meaningful test even if
Ulf Möller [Wed, 6 Dec 2000 16:39:09 +0000 (16:39 +0000)]
Don't check for bc at all. We can now run a meaningful test even if
it is missing.

23 years agoremove a comment that shouldn't have been there any more
Ulf Möller [Wed, 6 Dec 2000 16:30:23 +0000 (16:30 +0000)]
remove a comment that shouldn't have been there any more

23 years agoFaster BN_mod_sqrt algorithm for p == 5 (8).
Bodo Möller [Wed, 6 Dec 2000 12:25:33 +0000 (12:25 +0000)]
Faster BN_mod_sqrt algorithm for p == 5 (8).

23 years agoPrinting "verify ..." should not be counted as a test for the
Bodo Möller [Wed, 6 Dec 2000 12:24:27 +0000 (12:24 +0000)]
Printing "verify ..." should not be counted as a test for the
"xxx tests passed" message.

23 years agoDuring the self test, we only want to know what bctest says on
Richard Levitte [Wed, 6 Dec 2000 10:58:13 +0000 (10:58 +0000)]
During the self test, we only want to know what bctest says on
stderr...

23 years agoHave the self test use bctest to check that bc is sane.
Richard Levitte [Wed, 6 Dec 2000 10:54:31 +0000 (10:54 +0000)]
Have the self test use bctest to check that bc is sane.

23 years agoUse bc's "print" feature whenever it is available,
Bodo Möller [Wed, 6 Dec 2000 10:32:55 +0000 (10:32 +0000)]
Use bc's "print" feature whenever it is available,
not just on certain platforms.

23 years agoAdd a comment.
Bodo Möller [Wed, 6 Dec 2000 09:53:29 +0000 (09:53 +0000)]
Add a comment.

23 years agoWorkaround for broken (or missing) bc.
Bodo Möller [Wed, 6 Dec 2000 09:48:57 +0000 (09:48 +0000)]
Workaround for broken (or missing) bc.

23 years agoremove useless instruction
Ulf Möller [Wed, 6 Dec 2000 04:48:50 +0000 (04:48 +0000)]
remove useless instruction