oweals/openssl.git
21 years agoUpdate PRNG entry:
Bodo Möller [Tue, 4 Feb 2003 12:28:11 +0000 (12:28 +0000)]
Update PRNG entry:
- OpenSSL version differences
- Sun /dev/urandom patch information

21 years agoOld-style callbacks can be NULL!
Ben Laurie [Sat, 1 Feb 2003 20:58:59 +0000 (20:58 +0000)]
Old-style callbacks can be NULL!

21 years agoFix warning.
Ben Laurie [Sat, 1 Feb 2003 20:55:29 +0000 (20:55 +0000)]
Fix warning.

21 years agoWe can't say in advance what the argument to BIO_socket_ioctl() should be, so
Richard Levitte [Fri, 31 Jan 2003 12:20:35 +0000 (12:20 +0000)]
We can't say in advance what the argument to BIO_socket_ioctl() should be, so
let's make that a void *.  Also, BIO_socket_nbio() should send it an int
argument, not a long.
PR: 457

21 years agoDocument -engine where missing.
Richard Levitte [Thu, 30 Jan 2003 22:02:27 +0000 (22:02 +0000)]
Document -engine where missing.
PR: 424

21 years agoA few small bugs with BIO popping.
Richard Levitte [Thu, 30 Jan 2003 21:49:12 +0000 (21:49 +0000)]
A few small bugs with BIO popping.
PR: 364

21 years agoThe util scripts need to handled no-hw.
Richard Levitte [Thu, 30 Jan 2003 20:03:45 +0000 (20:03 +0000)]
The util scripts need to handled no-hw.
PR: 327

21 years agoThe MASM situation is more difficult than described so far. It is part of VC++ 7.
Richard Levitte [Thu, 30 Jan 2003 19:05:25 +0000 (19:05 +0000)]
The MASM situation is more difficult than described so far.  It is part of VC++ 7.
PR: 327

21 years agoFor VC++7 and up, the file is VSVARS32.BAT.
Richard Levitte [Thu, 30 Jan 2003 19:01:56 +0000 (19:01 +0000)]
For VC++7 and up, the file is VSVARS32.BAT.
PR: 327

21 years agoThe OPENSSL_NO_ENGINE has small problem: it changes certain structures. That's
Richard Levitte [Thu, 30 Jan 2003 18:52:46 +0000 (18:52 +0000)]
The OPENSSL_NO_ENGINE has small problem: it changes certain structures.  That's
bad, so let's not check OPENSSL_NO_ENGINE in those places.  Fortunately, all
the header files where the problem existed include ossl_typ.h, which makes
a 'forward declaration' of the ENGINE type.

21 years agomake update
Richard Levitte [Thu, 30 Jan 2003 17:53:02 +0000 (17:53 +0000)]
make update

21 years agoAdd the possibility to build without the ENGINE framework.
Richard Levitte [Thu, 30 Jan 2003 17:39:26 +0000 (17:39 +0000)]
Add the possibility to build without the ENGINE framework.
PR: 287

21 years agoSummarise the last couple of commits.
Geoff Thorpe [Thu, 30 Jan 2003 15:52:40 +0000 (15:52 +0000)]
Summarise the last couple of commits.

21 years agoThis glues the GMP wrapper ENGINE into OpenSSL if it is being built (ie. if
Geoff Thorpe [Thu, 30 Jan 2003 15:49:03 +0000 (15:49 +0000)]
This glues the GMP wrapper ENGINE into OpenSSL if it is being built (ie. if
the OPENSSL_USE_GMP symbol is defined). Also, I've re-ordered the listing
of other builtin ENGINEs to be alphabetical (though "dynamic" will still
come first).

21 years agoCommit a slightly modified version of an old experiment to do RSA private
Geoff Thorpe [Thu, 30 Jan 2003 15:43:07 +0000 (15:43 +0000)]
Commit a slightly modified version of an old experiment to do RSA private
key operations using the GMP library. The default is not to build (or use)
this code unless OPENSSL_USE_GMP is defined (because it will impose header
and linker dependencies that might need specifying too).

21 years ago"openssl engine" will not display ENGINE/DSO load failure errors when
Geoff Thorpe [Thu, 30 Jan 2003 14:58:44 +0000 (14:58 +0000)]
"openssl engine" will not display ENGINE/DSO load failure errors when
testing availability of engines with "-t" - the old behaviour of is
produced by increasing the feature's verbosity with "-tt".

21 years agoSmall typo, OENSSL should really be spelled OPENSSL.
Richard Levitte [Thu, 30 Jan 2003 11:08:44 +0000 (11:08 +0000)]
Small typo, OENSSL should really be spelled OPENSSL.
PR: 476

21 years agoFix a memory leak in SSL.
Richard Levitte [Thu, 30 Jan 2003 11:00:34 +0000 (11:00 +0000)]
Fix a memory leak in SSL.
PR: 477

21 years agocert_sk isn't always allocated, so freeing it may cause a crash.
Richard Levitte [Thu, 30 Jan 2003 10:27:43 +0000 (10:27 +0000)]
cert_sk isn't always allocated, so freeing it may cause a crash.
PR: 481

21 years agoDVCS (see RFC 3029) was missing among the possible purposes.
Richard Levitte [Wed, 29 Jan 2003 15:06:35 +0000 (15:06 +0000)]
DVCS (see RFC 3029) was missing among the possible purposes.
Notified privately to me by Peter Sylvester <Peter.Sylvester@EdelWeb.fr>,
one of the authors of said RFC

21 years agosimplify
Bodo Möller [Tue, 28 Jan 2003 13:08:21 +0000 (13:08 +0000)]
simplify

Submitted by: Nils Larsch

21 years agoCorrect EVP_SealInit() documentation, iv is an output
Dr. Stephen Henson [Sun, 26 Jan 2003 13:38:56 +0000 (13:38 +0000)]
Correct EVP_SealInit() documentation, iv is an output
parameter.

21 years agocofactor is optional in parameter encodings
Bodo Möller [Sat, 25 Jan 2003 15:28:49 +0000 (15:28 +0000)]
cofactor is optional in parameter encodings

Submitted by: Nils Larsch

21 years agoconsistency
Bodo Möller [Fri, 24 Jan 2003 22:28:32 +0000 (22:28 +0000)]
consistency

21 years agoFor ecdsa-with-SHA1, as for id-dsa-with-sha1, omit 'parameters'
Bodo Möller [Fri, 24 Jan 2003 21:43:08 +0000 (21:43 +0000)]
For ecdsa-with-SHA1, as for id-dsa-with-sha1, omit 'parameters'
in AlgorithmIdentifier

Submitted by: Nils Larsch

21 years agoProvide "dummy" &main::picmeup even in Windows perlasm modules.
Andy Polyakov [Fri, 24 Jan 2003 09:39:31 +0000 (09:39 +0000)]
Provide "dummy" &main::picmeup even in Windows perlasm modules.

21 years agoCheck return value of gmtime() and add error codes
Dr. Stephen Henson [Fri, 24 Jan 2003 01:12:01 +0000 (01:12 +0000)]
Check return value of gmtime() and add error codes
where it fails in ASN1_TIME_set().

Edit asn1.h so the new error code is the same in 0.9.7
and 0.9.8, rebuild new error codes.

Clear error queue in req.c if *_min or *_max is absent.

21 years agoWorkaround for lame compiler bug introduced in "CPU pack" for MSVC6SP5.
Andy Polyakov [Thu, 23 Jan 2003 10:05:39 +0000 (10:05 +0000)]
Workaround for lame compiler bug introduced in "CPU pack" for MSVC6SP5.

21 years agoThe patch speaks for itself.
Andy Polyakov [Thu, 23 Jan 2003 09:52:34 +0000 (09:52 +0000)]
The patch speaks for itself.

21 years agoMissing 0 broke FreeBSD build.
Richard Levitte [Thu, 23 Jan 2003 08:10:04 +0000 (08:10 +0000)]
Missing 0 broke FreeBSD build.
PR: 470

21 years agofix warnings
Bodo Möller [Tue, 21 Jan 2003 09:53:14 +0000 (09:53 +0000)]
fix warnings

Submitted by: Nils Larsch

21 years agoSuggestion was to change ${MACHINE} to i586 in lines in question. Well,
Andy Polyakov [Sun, 19 Jan 2003 21:47:06 +0000 (21:47 +0000)]
Suggestion was to change ${MACHINE} to i586 in lines in question. Well,
"whatever" doesn't the same (avoids 386 being passed to ./Configure),
consistent with other elder SCO targets and denotes that we probably
shouldn't care much about every out-of-date platform.

21 years agoThis is an *initial* tune-up. This update puts Itanium2 back on par with
Andy Polyakov [Sun, 19 Jan 2003 21:29:59 +0000 (21:29 +0000)]
This is an *initial* tune-up. This update puts Itanium2 back on par with
Itanium. I mean if overall performance improvement over C version was X
for Itanium, it's X even for Itanium2.

21 years agoOops! Missed closing quote... Didn't have time to verify before a snapshot
Andy Polyakov [Sun, 19 Jan 2003 11:39:19 +0000 (11:39 +0000)]
Oops! Missed closing quote... Didn't have time to verify before a snapshot
was cut...

21 years agoSCO target missed .so suffix.
Andy Polyakov [Sat, 18 Jan 2003 21:57:30 +0000 (21:57 +0000)]
SCO target missed .so suffix.

21 years ago-lresolv is not present on SCO Unix, RT#460.
Andy Polyakov [Sat, 18 Jan 2003 18:12:23 +0000 (18:12 +0000)]
-lresolv is not present on SCO Unix, RT#460.

21 years agoCaldera/SCO targets erroneously limit themselves to 386. See RT#464.
Andy Polyakov [Sat, 18 Jan 2003 15:17:26 +0000 (15:17 +0000)]
Caldera/SCO targets erroneously limit themselves to 386. See RT#464.

21 years agoFix for AIX shared build, see RT#463.
Andy Polyakov [Sat, 18 Jan 2003 15:13:03 +0000 (15:13 +0000)]
Fix for AIX shared build, see RT#463.

21 years agoEVP_DecryptInit() should call EVP_CipherInit() not EVP_CipherInit_ex().
Dr. Stephen Henson [Fri, 17 Jan 2003 00:48:47 +0000 (00:48 +0000)]
EVP_DecryptInit() should call EVP_CipherInit() not EVP_CipherInit_ex().

21 years agoIngore the correct flag file.
Richard Levitte [Thu, 16 Jan 2003 21:36:17 +0000 (21:36 +0000)]
Ingore the correct flag file.

21 years agoA few more files to ignore
Richard Levitte [Thu, 16 Jan 2003 21:32:56 +0000 (21:32 +0000)]
A few more files to ignore

21 years agomake update
Richard Levitte [Thu, 16 Jan 2003 21:20:30 +0000 (21:20 +0000)]
make update

21 years agoFreeBSD has /dev/crypto as well.
Richard Levitte [Thu, 16 Jan 2003 18:29:30 +0000 (18:29 +0000)]
FreeBSD has /dev/crypto as well.
PR: 462

21 years agoAdd some debugging output.
Richard Levitte [Thu, 16 Jan 2003 17:28:46 +0000 (17:28 +0000)]
Add some debugging output.

21 years agoArmor against systems without ranlib...
Lutz Jänicke [Thu, 16 Jan 2003 17:22:30 +0000 (17:22 +0000)]
Armor against systems without ranlib...
Submitted by: Thierry Lelegard <thierry.lelegard@canal-plus.fr>

PR: 461

21 years agoavoid potential confusion about curves (prime192v1 and prime256v1 are
Bodo Möller [Thu, 16 Jan 2003 16:05:23 +0000 (16:05 +0000)]
avoid potential confusion about curves (prime192v1 and prime256v1 are
also known as secp192r1 and secp256r1, respectively)

Submitted by: Nils Larsch, Bodo Moeller

21 years agomake update
Richard Levitte [Thu, 16 Jan 2003 13:01:36 +0000 (13:01 +0000)]
make update

21 years agoncr-scde target needs -lc89 for strcasecmp() and ftime() (Tim Rice, Martin
Lutz Jänicke [Thu, 16 Jan 2003 07:54:52 +0000 (07:54 +0000)]
ncr-scde target needs -lc89 for strcasecmp() and ftime() (Tim Rice, Martin
Megele).

PR: 450

21 years agoFix possible NULL dereferencial.
Richard Levitte [Thu, 16 Jan 2003 06:00:55 +0000 (06:00 +0000)]
Fix possible NULL dereferencial.
Notified by Verdon Walker <VWalker@novell.com>

21 years agoAdd verbosity
Richard Levitte [Wed, 15 Jan 2003 17:23:16 +0000 (17:23 +0000)]
Add verbosity

21 years agoFix initialization sequence to prevent freeing of unitialized objects.
Lutz Jänicke [Wed, 15 Jan 2003 14:54:59 +0000 (14:54 +0000)]
Fix initialization sequence to prevent freeing of unitialized objects.
Submitted by: Nils Larsch <nla@trustcenter.de>

PR: 459

21 years agoupdate error library for EC... changes
Bodo Möller [Wed, 15 Jan 2003 11:47:28 +0000 (11:47 +0000)]
update error library for EC... changes

Submitted by: Nils Larsch

21 years agoReally fix SSLv2 session ID handling
Lutz Jänicke [Wed, 15 Jan 2003 09:51:22 +0000 (09:51 +0000)]
Really fix SSLv2 session ID handling

PR: 377

21 years agoAs with RSA, which was modified recently, this change makes it possible to
Geoff Thorpe [Wed, 15 Jan 2003 02:01:55 +0000 (02:01 +0000)]
As with RSA, which was modified recently, this change makes it possible to
override key-generation implementations by placing handlers in the methods
for DSA and DH. Also, parameter generation for DSA and DH is possible by
another new handler for each method.

21 years agoSet EXPORT_VAR_AS_FN for BC-32 to work around a compiler bug,
Dr. Stephen Henson [Tue, 14 Jan 2003 20:54:18 +0000 (20:54 +0000)]
Set EXPORT_VAR_AS_FN for BC-32 to work around a compiler bug,

21 years agoExtend the HOWTO on creating certificates, and add a HOWTO in creating keys.
Richard Levitte [Tue, 14 Jan 2003 15:42:16 +0000 (15:42 +0000)]
Extend the HOWTO on creating certificates, and add a HOWTO in creating keys.
PR: 422

21 years agoDocument hpux-parisc2-cc problems, probably due to optimizer bug.
Lutz Jänicke [Tue, 14 Jan 2003 13:57:06 +0000 (13:57 +0000)]
Document hpux-parisc2-cc problems, probably due to optimizer bug.

PR: 426

21 years agoCorrect an example that has a few typos.
Richard Levitte [Tue, 14 Jan 2003 13:56:38 +0000 (13:56 +0000)]
Correct an example that has a few typos.
PR: 458

21 years agoMerge from 0.9.7-stable.
Richard Levitte [Mon, 13 Jan 2003 17:16:25 +0000 (17:16 +0000)]
Merge from 0.9.7-stable.

21 years agoMinor FAQ update
Andy Polyakov [Mon, 13 Jan 2003 16:34:05 +0000 (16:34 +0000)]
Minor FAQ update

21 years ago"!Cname surname" has now become redundant ...
Bodo Möller [Mon, 13 Jan 2003 15:57:33 +0000 (15:57 +0000)]
"!Cname surname" has now become redundant ...

21 years agoundo part of a recent change: it's "surname", not "surName"
Bodo Möller [Mon, 13 Jan 2003 15:52:04 +0000 (15:52 +0000)]
undo part of a recent change: it's "surname", not "surName"
(see X.520 aka ISO/IEC 9594-6)

21 years agoMake sure not to declare a clashing read() for DJGPP.
Richard Levitte [Mon, 13 Jan 2003 15:35:22 +0000 (15:35 +0000)]
Make sure not to declare a clashing read() for DJGPP.
PR: 440

21 years agoDJGPP doesn't have DLLs, so skip adding to %PATH% in that environment.
Richard Levitte [Mon, 13 Jan 2003 15:16:40 +0000 (15:16 +0000)]
DJGPP doesn't have DLLs, so skip adding to %PATH% in that environment.
PR: 453

21 years agoadd something to the '$no_shared_warn' text
Bodo Möller [Mon, 13 Jan 2003 14:58:49 +0000 (14:58 +0000)]
add something to the '$no_shared_warn' text

21 years agoCorrect a misleading comment.
Richard Levitte [Mon, 13 Jan 2003 14:53:43 +0000 (14:53 +0000)]
Correct a misleading comment.
PR: 456

21 years agoRevert: the names of the cygwin distribution will not be named openssl2-*.
Richard Levitte [Mon, 13 Jan 2003 14:47:52 +0000 (14:47 +0000)]
Revert: the names of the cygwin distribution will not be named openssl2-*.
Requested by Corinna Vinschen <vinschen@redhat.com>

21 years agoUI_UTIL_read_pw() misinterpreted the values returned from UI functions.
Richard Levitte [Mon, 13 Jan 2003 14:17:43 +0000 (14:17 +0000)]
UI_UTIL_read_pw() misinterpreted the values returned from UI functions.
PR: 456

21 years agodocument BN_GENCB API by adding an example
Bodo Möller [Mon, 13 Jan 2003 13:44:20 +0000 (13:44 +0000)]
document BN_GENCB API by adding an example

21 years agoTypo.
Bodo Möller [Mon, 13 Jan 2003 13:18:22 +0000 (13:18 +0000)]
Typo.

NB: This and other manual pages should be updated for the new
BN_GENCB interface.

21 years agotty_in will never be stderr, so it will always be closed, which means stdin
Richard Levitte [Mon, 13 Jan 2003 13:15:16 +0000 (13:15 +0000)]
tty_in will never be stderr, so it will always be closed, which means stdin
might get closed...
Reported by Mark Daniel <Mark.Daniel@wasd.vsm.com.au>

21 years agotypo
Bodo Möller [Mon, 13 Jan 2003 12:52:33 +0000 (12:52 +0000)]
typo

21 years agoOpenBSD FAQ update. Apparently gas from binutils-2.x makes it impossible to
Andy Polyakov [Sun, 12 Jan 2003 21:39:13 +0000 (21:39 +0000)]
OpenBSD FAQ update. Apparently gas from binutils-2.x makes it impossible to
use gcc -fPIC ... on OpenBSD-i386. Alternative solution is provided.

21 years agoJust an extra comment.
Andy Polyakov [Sun, 12 Jan 2003 15:50:59 +0000 (15:50 +0000)]
Just an extra comment.

21 years agoAdd better support for FreeBSD on non-x86 machines.
Richard Levitte [Sun, 12 Jan 2003 04:43:44 +0000 (04:43 +0000)]
Add better support for FreeBSD on non-x86 machines.
Add specific support for FreeBSD on sparc64.
PR: 427

21 years agoIf the user said 'shared' and we haven't included support for shared libraries,
Richard Levitte [Sat, 11 Jan 2003 11:40:39 +0000 (11:40 +0000)]
If the user said 'shared' and we haven't included support for shared libraries,
warn him or her.
Reminder by Andrew Marlow <apm35@student.open.ac.uk>

21 years agoDouble the dollars that are intended for the command line.
Richard Levitte [Fri, 10 Jan 2003 16:32:32 +0000 (16:32 +0000)]
Double the dollars that are intended for the command line.
PR: 423

21 years agoAdd documentation on how to handle the shared libaries.
Richard Levitte [Fri, 10 Jan 2003 16:14:32 +0000 (16:14 +0000)]
Add documentation on how to handle the shared libaries.
PR: 423

21 years agoName the flag file correctly
Richard Levitte [Fri, 10 Jan 2003 16:04:07 +0000 (16:04 +0000)]
Name the flag file correctly

21 years agoCFLAG, not CFLAGS
Richard Levitte [Fri, 10 Jan 2003 11:30:44 +0000 (11:30 +0000)]
CFLAG, not CFLAGS

21 years agoIt's recommended to do 'make clean' after having prepared a new build tree.
Richard Levitte [Fri, 10 Jan 2003 11:00:50 +0000 (11:00 +0000)]
It's recommended to do 'make clean' after having prepared a new build tree.
PR: 437

21 years agoWhen preparing a separate build tree, don't make softlinks to softlinks.
Richard Levitte [Fri, 10 Jan 2003 10:56:14 +0000 (10:56 +0000)]
When preparing a separate build tree, don't make softlinks to softlinks.
Add instructions in INSTALL, for easy access.
PR: 437

21 years agoForce the removal.
Richard Levitte [Fri, 10 Jan 2003 10:54:59 +0000 (10:54 +0000)]
Force the removal.
PR: 437

21 years agoCertain files must be removed before generating them, in case they point
Richard Levitte [Fri, 10 Jan 2003 10:54:18 +0000 (10:54 +0000)]
Certain files must be removed before generating them, in case they point
into a read-only source tree.
PR: 437

21 years agoAvoid unnecessary pollution of object module name table. Cygwin shared
Andy Polyakov [Fri, 10 Jan 2003 10:16:39 +0000 (10:16 +0000)]
Avoid unnecessary pollution of object module name table. Cygwin shared
build workaround (DJGPP swallows it too). One probably should do same
as with ELF calling it COFF, but I'm very short in COFF platforms, so
I just go for easy ad-hoc solution. I'll take care of merge to 0.9.7
later.

21 years agoMake sure everything that may be freed is allocated or initiated.
Richard Levitte [Fri, 10 Jan 2003 08:59:46 +0000 (08:59 +0000)]
Make sure everything that may be freed is allocated or initiated.
PR: 446

21 years agoTypos corrected.
Richard Levitte [Fri, 10 Jan 2003 08:54:01 +0000 (08:54 +0000)]
Typos corrected.
PR: 445

21 years agoTypo.
Dr. Stephen Henson [Thu, 9 Jan 2003 16:54:21 +0000 (16:54 +0000)]
Typo.

21 years agoThe naming scheme wasn't quite correct for Cygwin
Richard Levitte [Thu, 9 Jan 2003 16:15:46 +0000 (16:15 +0000)]
The naming scheme wasn't quite correct for Cygwin

21 years agoAt least OpenBSD implements PIC in the same way ELF does.
Andy Polyakov [Thu, 9 Jan 2003 16:06:09 +0000 (16:06 +0000)]
At least OpenBSD implements PIC in the same way ELF does.

21 years agogcc wants character constants to be correct. Before this change, the
Richard Levitte [Thu, 9 Jan 2003 13:25:14 +0000 (13:25 +0000)]
gcc wants character constants to be correct.  Before this change, the
following would happen on Solaris:

  m4 -B 8192 asm/des_enc.m4 > asm/des_enc-sparc.S
  gcc -DOPENSSL_SYSNAME_ULTRASPARC -DOPENSSL_NO_STATIC_ENGINE -fPIC -DOPENSSL_PIC -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DOPENSSL_NO_KRB5 -m32 -mcpu=ultrasparc -O3 -fomit-frame-pointer -Wall -DB_ENDIAN -DBN_DIV2W -DMD5_ASM   -c -o asm/des_enc-sparc.o asm/des_enc-sparc.S
  asm/des_enc-sparc.S:2007: unterminated character constant
  asm/des_enc-sparc.S:2008: unterminated character constant

21 years agoNULL tofree when it is freed to avoid double free.
Dr. Stephen Henson [Thu, 9 Jan 2003 13:06:49 +0000 (13:06 +0000)]
NULL tofree when it is freed to avoid double free.

Make sure key is not NULL before freeing it.

21 years agoFinalizing asm support for UnixWare, SCO, OpenUnix... Note that I've
Andy Polyakov [Thu, 9 Jan 2003 08:42:04 +0000 (08:42 +0000)]
Finalizing asm support for UnixWare, SCO, OpenUnix... Note that I've
replaced #if logic around bn_sub_part_words in bn_mul.c. I rely upon
OPENSSL_BN_ASM_PART_WORDS being added by ./Configure script. Would it
still work on non-Unix platforms?

21 years agoThird argument to shl_load() is "long address", not a pointer.
Lutz Jänicke [Wed, 8 Jan 2003 08:27:50 +0000 (08:27 +0000)]
Third argument to shl_load() is "long address", not a pointer.
(Didn't influence functionality, as on HP-UX 32bit the NULL pointer
is a 32bit 0-value and thus is identical to the required 0L.)

PR: 443

21 years agoRSA_METHOD now supports key-generation, but (for now) none of these
Geoff Thorpe [Tue, 7 Jan 2003 05:53:58 +0000 (05:53 +0000)]
RSA_METHOD now supports key-generation, but (for now) none of these
ENGINEs implement it.

21 years agoThis is the first step in allowing RSA_METHODs to implement their own key
Geoff Thorpe [Tue, 7 Jan 2003 05:51:39 +0000 (05:51 +0000)]
This is the first step in allowing RSA_METHODs to implement their own key
generation. This prototype matches the new API function
RSA_generate_key_ex(), though both may be subject to change during
development before 0.9.8.

21 years agoFix leak.
Dr. Stephen Henson [Sat, 4 Jan 2003 18:25:24 +0000 (18:25 +0000)]
Fix leak.

21 years agoFix a typo.
Andy Polyakov [Sat, 4 Jan 2003 15:18:00 +0000 (15:18 +0000)]
Fix a typo.

21 years agoAnother GAS fix-up and some commentary...
Andy Polyakov [Sat, 4 Jan 2003 14:41:09 +0000 (14:41 +0000)]
Another GAS fix-up and some commentary...

21 years agoGAS can't stand stub, which is stb's synonym.
Andy Polyakov [Sat, 4 Jan 2003 12:49:38 +0000 (12:49 +0000)]
GAS can't stand stub, which is stb's synonym.