18 years agoPR: 938
Dr. Stephen Henson [Sun, 14 Nov 2004 13:55:16 +0000 (13:55 +0000)]
PR: 938


18 years agoZap obsolete der_chop script.
Dr. Stephen Henson [Sun, 14 Nov 2004 00:08:36 +0000 (00:08 +0000)]
Zap obsolete der_chop script.

18 years agoPR: 969
Dr. Stephen Henson [Sat, 13 Nov 2004 13:38:34 +0000 (13:38 +0000)]
PR: 969
Submitted by: David Holmes <>

18 years agoFix x509.c so it creates serial number file again if no
Dr. Stephen Henson [Sat, 13 Nov 2004 13:26:06 +0000 (13:26 +0000)]
Fix x509.c so it creates serial number file again if no
serial number is supplied on command line.

18 years agoCut'n'paste mistake. All tested OK now...
Richard Levitte [Thu, 11 Nov 2004 19:36:08 +0000 (19:36 +0000)]
Cut'n'paste mistake.  All tested OK now...

18 years agoWhoops, syntactic mistake...
Richard Levitte [Thu, 11 Nov 2004 18:58:01 +0000 (18:58 +0000)]
Whoops, syntactic mistake...

18 years agoSome find it confusing that environment variables are set when shared
Richard Levitte [Thu, 11 Nov 2004 18:18:43 +0000 (18:18 +0000)]
Some find it confusing that environment variables are set when shared
libraries aren't built or used.  I can see the point, so I'm
reorganising a little for clarity.

18 years agoUse the default_md config file value when signing CRLs.
Dr. Stephen Henson [Thu, 11 Nov 2004 13:47:06 +0000 (13:47 +0000)]
Use the default_md config file value when signing CRLs.


18 years agoDon't return an error with crl -noout.
Dr. Stephen Henson [Thu, 11 Nov 2004 02:13:08 +0000 (02:13 +0000)]
Don't return an error with crl -noout.

Sumbmitted by: Michael Konietzka <>

18 years agoAs was shown by Marc Bevand reordering of couple of load operations
Andy Polyakov [Tue, 9 Nov 2004 17:23:26 +0000 (17:23 +0000)]
As was shown by Marc Bevand reordering of couple of load operations
results in even higher performance gain of 3.3x:-) At least on

18 years agoMake sure LD_PRELOAD is only set when we build shared libraries (and
Richard Levitte [Fri, 5 Nov 2004 09:12:10 +0000 (09:12 +0000)]
Make sure LD_PRELOAD is only set when we build shared libraries (and
therefore link with them).  Add LD_PRELOAD setting code where it was
still missing.

PR: 966

18 years agoDon't use $(EXHEADER) directly in for loops, as most shells will break
Richard Levitte [Tue, 2 Nov 2004 23:55:01 +0000 (23:55 +0000)]
Don't use $(EXHEADER) directly in for loops, as most shells will break
if $(EXHEADER) is empty.

Notified by many, solution suggested by Carson Gaspar <>

18 years agoBecause -rpath/-R may have been used, our settings of LD_LIBRARY_PATH
Richard Levitte [Tue, 2 Nov 2004 01:13:04 +0000 (01:13 +0000)]
Because -rpath/-R may have been used, our settings of LD_LIBRARY_PATH
and friends may be entirely useless.  In such a case, LD_PRELOAD is
the answer, at least on platforms using LD_LIBRARY_PATH.  There might
be other variables to set on other platforms, please fill us in...

For now, we only do this with the tests, so they won't fail for silly
reasons like getting dynamically linked to older installed libraries
rather than the newly built ones...

PR: 960

18 years agoMake sure memmove() is defined, even on SunOS 4.1.4.
Richard Levitte [Mon, 1 Nov 2004 07:58:38 +0000 (07:58 +0000)]
Make sure memmove() is defined, even on SunOS 4.1.4.
PR: 963

18 years agoUpdate NEWS
Dr. Stephen Henson [Mon, 25 Oct 2004 17:11:19 +0000 (17:11 +0000)]
Update NEWS

18 years agoUpdate FAQ.
Dr. Stephen Henson [Mon, 25 Oct 2004 12:36:33 +0000 (12:36 +0000)]
Update FAQ.

18 years agoFix race condition when SSL ciphers are initialized.
Dr. Stephen Henson [Mon, 25 Oct 2004 11:14:16 +0000 (11:14 +0000)]
Fix race condition when SSL ciphers are initialized.

18 years agoUpdate ECDSA and ECDH for OPENSSL_NO_ENGINE.
Geoff Thorpe [Thu, 21 Oct 2004 00:06:14 +0000 (00:06 +0000)]

Reported by: Maxim Masiutin
Submitted by: Nils Larsch

18 years agoBecause libraries on Windows lack useful version information, the zlib
Richard Levitte [Thu, 14 Oct 2004 05:48:59 +0000 (05:48 +0000)]
Because libraries on Windows lack useful version information, the zlib
guys had to change the name to differentiate with older versions when
a backward incompatibility came up.  Of course, we need to adapt.
This change simply tries to load the library through the newer name
(ZLIB1) first, and if that fails, it tries the good old ZLIB.

18 years agoOops!
Dr. Stephen Henson [Mon, 4 Oct 2004 17:28:31 +0000 (17:28 +0000)]

18 years agoFix race condition when CRL checking is enabled.
Dr. Stephen Henson [Mon, 4 Oct 2004 16:30:12 +0000 (16:30 +0000)]
Fix race condition when CRL checking is enabled.

18 years agoUpdate debug-steve
Dr. Stephen Henson [Fri, 1 Oct 2004 11:35:38 +0000 (11:35 +0000)]
Update debug-steve

18 years agoDon't use C++ reserved work "explicit".
Dr. Stephen Henson [Fri, 1 Oct 2004 11:21:53 +0000 (11:21 +0000)]
Don't use C++ reserved work "explicit".

18 years agoFix Solaris 10_x86 shared build. -Bsymbolic is required to avoid
Andy Polyakov [Tue, 28 Sep 2004 20:45:10 +0000 (20:45 +0000)]
Fix Solaris 10_x86 shared build. -Bsymbolic is required to avoid
"remaining relocations" in assembler modules. The latter seems to
be new behaviour, elder as/ld managed to resolve this relocations
as internal. It's possible to address this problem differently,
but I settle for -Bsymbolic...
PR: 546

18 years agousr/doc has recently changed to usr/share/doc on Cygwin.
Richard Levitte [Tue, 28 Sep 2004 13:10:16 +0000 (13:10 +0000)]
usr/doc has recently changed to usr/share/doc on Cygwin.
Notified by Corinna Vinschen <>

18 years agoMove the declaration of alloca() so it's ony declared when really
Richard Levitte [Mon, 27 Sep 2004 21:59:44 +0000 (21:59 +0000)]
Move the declaration of alloca() so it's ony declared when really

18 years agoSHA1 asm Pentium tune-up. Performance loss is not as bad anymore.
Andy Polyakov [Mon, 27 Sep 2004 09:37:03 +0000 (09:37 +0000)]
SHA1 asm Pentium tune-up. Performance loss is not as bad anymore.

18 years agosha256_block advances the input pointer double as fast sometimes. Fix the
Andy Polyakov [Mon, 27 Sep 2004 09:35:59 +0000 (09:35 +0000)]
sha256_block advances the input pointer double as fast sometimes. Fix the
bug and test that it's actually gone.
PR: 950

18 years agoNils Larsch reported that this include is required. Strange that this had
Geoff Thorpe [Fri, 24 Sep 2004 23:37:52 +0000 (23:37 +0000)]
Nils Larsch reported that this include is required. Strange that this had
gone unnoticed ...

18 years agoImport changed files from LPlib. The changes are logged as follows
Richard Levitte [Thu, 23 Sep 2004 22:11:39 +0000 (22:11 +0000)]
Import changed files from LPlib.  The changes are logged as follows
for LPdir_unix.c in LPlib.  For the other files, only the last log
entry applies.

revision 1.11
date: 2004/09/23 22:07:22;  author: _cvs_levitte;  state: Exp;  lines: +20 -6
Define my own macro LP_ENTRY_SIZE to express the size of my own
buffering of directory entries, and make it depend on whichever comes
first of PATH_MAX and NAME_MAX.  As a fallback, make sure it's set to
255 if neither PATH_MAX or NAME_MAX were defined.  Also, if the size
given from PATH_MAX or NAME_MAX is less than 255, force LP_ENTRY_SIZE
to be 255.

It makes no harm whatsoever if LP_ENTRY_SIZE is larger than the
maximum local path name limit.  It does make a lot of harm if
LP_ENTRY_SIZE is smaller.  255 seemed like a fairly acceptable default
when nothing else is available.
revision 1.10
date: 2004/08/26 13:36:05;  author: _cvs_levitte;  state: Exp;  lines: +13 -13
License correction.  I am not REGENTS, just a COPYRIGHT HOLDER.

18 years agoRemove distracting comments and code. Thanks to Nils for picking up on the
Geoff Thorpe [Sun, 19 Sep 2004 04:55:15 +0000 (04:55 +0000)]
Remove distracting comments and code. Thanks to Nils for picking up on the
outstanding ticket.

PR: 926

18 years agoTwo TODO comments taken care of. Nils pointed out that one of them had already
Geoff Thorpe [Sun, 19 Sep 2004 04:43:46 +0000 (04:43 +0000)]
Two TODO comments taken care of. Nils pointed out that one of them had already
been done, and took care of the other one (which hadn't).

Submitted by: Nils Larsch
Reviewed by: Geoff Thorpe

18 years agoMake -Werror happy again.
Geoff Thorpe [Sat, 18 Sep 2004 01:32:32 +0000 (01:32 +0000)]
Make -Werror happy again.

18 years agoCheck ASN1_TYPE structure type is a SEQUENCE in PKCS7_get_smimecap().
Dr. Stephen Henson [Wed, 15 Sep 2004 23:47:25 +0000 (23:47 +0000)]
Check ASN1_TYPE structure type is a SEQUENCE in PKCS7_get_smimecap().

18 years agoChange values of MBSTRING_* to the form MBSTRING_FLAG|nbyte as assumed
Dr. Stephen Henson [Mon, 13 Sep 2004 22:33:56 +0000 (22:33 +0000)]
Change values of MBSTRING_* to the form MBSTRING_FLAG|nbyte as assumed
in ASN1_STRING_to_UTF8().

18 years ago- There's no more need for the snprintf macro.
Richard Levitte [Mon, 13 Sep 2004 09:15:06 +0000 (09:15 +0000)]
- There's no more need for the snprintf macro.
- Move the inclusion of malloc.h until after all other includes, so we
  can do proper tests of system macros.
- Make sure the correct header file is included to get the builtin
  "alloca" under VMS, and define a macro to map the symbol 'alloca' to

18 years agoSynchronise with Unix build.
Richard Levitte [Sun, 12 Sep 2004 13:02:04 +0000 (13:02 +0000)]
Synchronise with Unix build.

18 years agoWhen looking for request extensions in a certificate look first
Dr. Stephen Henson [Fri, 10 Sep 2004 20:20:54 +0000 (20:20 +0000)]
When looking for request extensions in a certificate look first
for the PKCS#9 OID then the non standard MS OID.

18 years agomake update
Richard Levitte [Fri, 10 Sep 2004 10:30:33 +0000 (10:30 +0000)]
make update

18 years agoMake VIA Padlock engine more platform friendly and eliminate compiler
Andy Polyakov [Thu, 9 Sep 2004 14:54:12 +0000 (14:54 +0000)]
Make VIA Padlock engine more platform friendly and eliminate compiler

Submitted by: Doug Kaufman <>

18 years agox86 assembler updates: more instructions, new OPENSSL_instrument_halt
Andy Polyakov [Thu, 9 Sep 2004 14:50:32 +0000 (14:50 +0000)]
x86 assembler updates: more instructions, new OPENSSL_instrument_halt
[for DJGPP]...

18 years agoSynchronise VMS build files with Unixly Makefiles.
Richard Levitte [Wed, 8 Sep 2004 08:13:34 +0000 (08:13 +0000)]
Synchronise VMS build files with Unixly Makefiles.

18 years agoAnother symbol longer than 31 characters...
Richard Levitte [Wed, 8 Sep 2004 08:13:03 +0000 (08:13 +0000)]
Another symbol longer than 31 characters...

18 years agoReformat smime utility.
Dr. Stephen Henson [Tue, 7 Sep 2004 18:38:46 +0000 (18:38 +0000)]
Reformat smime utility.

Add support for policy checking in verify utility.

18 years agoDon't use 'explicit' for variable name.
Dr. Stephen Henson [Tue, 7 Sep 2004 00:31:08 +0000 (00:31 +0000)]
Don't use 'explicit' for variable name.

18 years agoReformat smime.c
Dr. Stephen Henson [Tue, 7 Sep 2004 00:28:17 +0000 (00:28 +0000)]
Reformat smime.c

18 years agoNew X509_VERIFY_PARAM structure and associated functionality.
Dr. Stephen Henson [Mon, 6 Sep 2004 18:43:01 +0000 (18:43 +0000)]
New X509_VERIFY_PARAM structure and associated functionality.

This tidies up verify parameters and adds support for integrated policy

Add support for policy related command line options. Currently only in smime

WARNING: experimental code subject to change.

18 years agoStop compiler warnings.
Dr. Stephen Henson [Mon, 6 Sep 2004 18:37:46 +0000 (18:37 +0000)]
Stop compiler warnings.

18 years agoTABLE OpenBSD-i386 update
Andy Polyakov [Sun, 29 Aug 2004 22:05:02 +0000 (22:05 +0000)]
TABLE OpenBSD-i386 update

18 years agoProper support for OpenBSD-i386 shared build, including assember modules!
Andy Polyakov [Sun, 29 Aug 2004 21:36:37 +0000 (21:36 +0000)]
Proper support for OpenBSD-i386 shared build, including assember modules!
"Proper" means "compiles and passes test." Versioning is broken (I think).

18 years agoOPENSSL_ia32cap final touches. Note that OPENSSL_ia32cap is no longer a
Andy Polyakov [Sun, 29 Aug 2004 16:36:05 +0000 (16:36 +0000)]
OPENSSL_ia32cap final touches. Note that OPENSSL_ia32cap is no longer a
symbol, but a macro expanded as (*(OPENSSL_ia32cap_loc())). The latter
is the only one to be exported to application.

18 years agoFix compiler warnings in crypto/evp/bio_ok.c as pointed out by Geoff.
Andy Polyakov [Sun, 29 Aug 2004 16:19:27 +0000 (16:19 +0000)]
Fix compiler warnings in crypto/evp/bio_ok.c as pointed out by Geoff.

18 years agocrypto/perlasm update primarily to unify Netware modules. Once it's verified
Andy Polyakov [Sun, 29 Aug 2004 16:10:27 +0000 (16:10 +0000)]
crypto/perlasm update primarily to unify Netware modules. Once it's verified
x86* will be deleted. In addition this update implements initseg
on several additional [in addition to ELF] platforms. Functions registered
with initseg are supposed to be called prior main().

18 years agoMinor VIA Padlock engine update: eliminate -Wunused warning when *not*
Andy Polyakov [Tue, 24 Aug 2004 09:01:09 +0000 (09:01 +0000)]
Minor VIA Padlock engine update: eliminate -Wunused warning when *not*
compiling the engine and inline memcpy in performance critical pathes.

18 years agoMake aes_ctr.c 64-bit savvy.
Andy Polyakov [Mon, 23 Aug 2004 22:19:51 +0000 (22:19 +0000)]
Make aes_ctr.c 64-bit savvy.

18 years ago'compatibility', not 'computability' :-)...
Richard Levitte [Wed, 18 Aug 2004 15:48:33 +0000 (15:48 +0000)]
'compatibility', not 'computability' :-)...

18 years agoOn systems that use case-insensitive symbol names (i.e. they're all
Richard Levitte [Thu, 12 Aug 2004 08:58:55 +0000 (08:58 +0000)]
On systems that use case-insensitive symbol names (i.e. they're all
converted to upper case or something like that), the application-
level bio_dump_cb() has a name clash with the new library function
BIO_dump_cb().  The easiest fix is to rename the function at the
application level.

18 years agoBasically, I wanted to be able to make a dump to a FILE*, and not have
Richard Levitte [Wed, 11 Aug 2004 21:13:57 +0000 (21:13 +0000)]
Basically, I wanted to be able to make a dump to a FILE*, and not have
to bother creating a BIO around it.  So here's a few more functions to
make it possible to make the dump using a printing callback, and to
print to a FILE* (based on the callback variant), done in the same
style as the functions in crypto/err/err_prn.c.

18 years agoStupid casts...
Richard Levitte [Wed, 11 Aug 2004 17:41:15 +0000 (17:41 +0000)]
Stupid casts...

18 years agoUpdate FAQ.
Dr. Stephen Henson [Wed, 11 Aug 2004 17:22:13 +0000 (17:22 +0000)]
Update FAQ.

18 years agoMake ASN1_INTEGER_cmp() work as expected with negative integers.
Dr. Stephen Henson [Tue, 10 Aug 2004 17:40:14 +0000 (17:40 +0000)]
Make ASN1_INTEGER_cmp() work as expected with negative integers.

18 years agoCall setup_engine after autoconfig.
Dr. Stephen Henson [Fri, 6 Aug 2004 12:44:34 +0000 (12:44 +0000)]
Call setup_engine after autoconfig.

18 years agoDon't ignore return value of EVP_DigestInit_ex() in md BIOs and dgst utility.
Dr. Stephen Henson [Thu, 5 Aug 2004 18:09:50 +0000 (18:09 +0000)]
Don't ignore return value of EVP_DigestInit_ex() in md BIOs and dgst utility.

18 years agoMake a note of the new engine.
Geoff Thorpe [Wed, 4 Aug 2004 22:42:29 +0000 (22:42 +0000)]
Make a note of the new engine.

18 years agoPadlock engine update to fix a typo in MSC assembler and to address
Andy Polyakov [Wed, 4 Aug 2004 12:58:26 +0000 (12:58 +0000)]
Padlock engine update to fix a typo in MSC assembler and to address
potential corruption problem if user manages to inter-leave aligined
and misaligned requests [as well as some MSC-specific tweaks].

18 years agoDJGPP has opendir() and friends, according to Gisle Vanem <>.
Richard Levitte [Tue, 3 Aug 2004 19:15:21 +0000 (19:15 +0000)]
DJGPP has opendir() and friends, according to Gisle Vanem <>.

18 years agoIf they ask for 386, keep it as 386 as possible...
Andy Polyakov [Mon, 2 Aug 2004 22:41:19 +0000 (22:41 +0000)]
If they ask for 386, keep it as 386 as possible...

18 years agoAvoid a.out name table pollition.
Andy Polyakov [Mon, 2 Aug 2004 22:02:17 +0000 (22:02 +0000)]
Avoid a.out name table pollition.

18 years agoMinor clean-up to make Microsoft compiler shut up.
Andy Polyakov [Mon, 2 Aug 2004 21:54:40 +0000 (21:54 +0000)]
Minor clean-up to make Microsoft compiler shut up.

18 years agoVIA C3 processor extends IA-32 instruction set with instuctions
Andy Polyakov [Mon, 2 Aug 2004 21:48:11 +0000 (21:48 +0000)]
VIA C3 processor extends IA-32 instruction set with instuctions
performing AES encryption in hardware, as well as one accessing
hardware RNG. As you surely imagine this engine access this
extended instruction set. Well, only AES for the moment, support
for RNG is to be added later on...
PR: 889
Submitted by: Michal Ludvig <>

Obtained from:

18 years agoCygwin fix-up for shared build.
Andy Polyakov [Sun, 1 Aug 2004 21:24:34 +0000 (21:24 +0000)]
Cygwin fix-up for shared build.

18 years agoOpenBSD fix-up for new a.out targets. OpenBSD .s.o rule is busted...
Andy Polyakov [Sun, 1 Aug 2004 21:16:26 +0000 (21:16 +0000)]
OpenBSD fix-up for new a.out targets. OpenBSD .s.o rule is busted...

18 years agoClean-up GAS targets: get rid of "cpp" stuff and replace it with "purified"
Andy Polyakov [Sun, 1 Aug 2004 17:33:58 +0000 (17:33 +0000)]
Clean-up GAS targets: get rid of "cpp" stuff and replace it with "purified"
COFF and a.out targets [similar to ELF targets]. You might notice some
rudementary support for shared mingw builds under cygwin. It works (it
produces cryptoeay32.dll and ssleay32.dll with everything exported by
name), but it's primarily for testing/debugging purposes, at least for

18 years agoDeprecate cpp and gaswin targets. New coff fills in for gaswin, but cpp is
Andy Polyakov [Sun, 1 Aug 2004 17:03:50 +0000 (17:03 +0000)]
Deprecate cpp and gaswin targets. New coff fills in for gaswin, but cpp is
going out...

18 years agoDLLEntryPoint is a collective name, not what linker looks for. However,
Andy Polyakov [Sun, 1 Aug 2004 14:27:43 +0000 (14:27 +0000)]
DLLEntryPoint is a collective name, not what linker looks for. However,
if we explicitly intruct the linker to set entry point, then we become
obliged to initialize run-time library. Instead we can pick name run-time
will call and such name is DllMain. Note that this applies to both
"native" Win32 environment and Cygwin:-)

18 years agoWe build the crypto stuff, not the ssl stuff, in this command procedure...
Richard Levitte [Thu, 29 Jul 2004 22:25:59 +0000 (22:25 +0000)]
We build the crypto stuff, not the ssl stuff, in this command procedure...

18 years agoThe compiler may complain about what looks like a double definition of a
Richard Levitte [Tue, 27 Jul 2004 13:58:27 +0000 (13:58 +0000)]
The compiler may complain about what looks like a double definition of a
static variable

18 years agoOops, wrong version...
Dr. Stephen Henson [Tue, 27 Jul 2004 00:19:58 +0000 (00:19 +0000)]
Oops, wrong version...

18 years agoAdd FIPS library name to error routines.
Dr. Stephen Henson [Tue, 27 Jul 2004 00:19:18 +0000 (00:19 +0000)]
Add FIPS library name to error routines.

18 years agoThis is so to say "damage control" for jumbo "cpuid" patch, see
Andy Polyakov [Mon, 26 Jul 2004 22:01:50 +0000 (22:01 +0000)]
This is so to say "damage control" for jumbo "cpuid" patch, see Now all platform should
be operational, while SSE2 code pathes get engaged on ELF platforms

18 years agoAdd framework for yet another assembler module dubbed "cpuid." Idea
Andy Polyakov [Mon, 26 Jul 2004 20:18:55 +0000 (20:18 +0000)]
Add framework for yet another assembler module dubbed "cpuid." Idea
is to have a placeholder to small routines, which can be written only
in assembler. In IA-32 case this includes processor capability
identification and access to Time-Stamp Counter. As discussed earlier
OPENSSL_ia32cap is introduced to control recently added SSE2 code
pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the
code is operational on ELF platforms only. I haven't checked it yet,
but I have all reasons to believe that Windows build should fail to
link too. I'll be looking into it shortly...

18 years agoZero key-length for HMAC is apparently OK.
Andy Polyakov [Sun, 25 Jul 2004 20:24:49 +0000 (20:24 +0000)]
Zero key-length for HMAC is apparently OK.

18 years agoMake bio_ok.c Microsoft compiler savvy.
Andy Polyakov [Sun, 25 Jul 2004 20:13:30 +0000 (20:13 +0000)]
Make bio_ok.c Microsoft compiler savvy.

18 years agoTypos, typos...
Andy Polyakov [Sun, 25 Jul 2004 20:09:56 +0000 (20:09 +0000)]
Typos, typos...

18 years agoMake bio_ok.c 64-bit savvy.
Andy Polyakov [Sun, 25 Jul 2004 19:37:41 +0000 (19:37 +0000)]
Make bio_ok.c 64-bit savvy.

18 years agoStricter boundary condition check in HMAC_Init_ex.
Andy Polyakov [Sun, 25 Jul 2004 19:25:05 +0000 (19:25 +0000)]
Stricter boundary condition check in HMAC_Init_ex.

18 years agoMinor 64-bit md32_common.h update and minor unsignification of digests.
Andy Polyakov [Sun, 25 Jul 2004 19:10:43 +0000 (19:10 +0000)]
Minor 64-bit md32_common.h update and minor unsignification of digests.

18 years ago'apps/openssl dgst -help' update and minor apps/speed.c update.
Andy Polyakov [Sun, 25 Jul 2004 18:57:35 +0000 (18:57 +0000)]
'apps/openssl dgst -help' update and minor apps/speed.c update.

18 years agoMake SHA-256/-512 optional. Note that no-sha switches off *all* SHA.
Andy Polyakov [Sun, 25 Jul 2004 18:25:24 +0000 (18:25 +0000)]
Make SHA-256/-512 optional. Note that no-sha switches off *all* SHA.

18 years agoSome compilers are just too whiny. Nothing makes Microsoft compiler
Andy Polyakov [Sun, 25 Jul 2004 17:00:56 +0000 (17:00 +0000)]
Some compilers are just too whiny. Nothing makes Microsoft compiler
stop complaining about loss of precision, but explicit cast.

18 years agoSome compilers are just too whiny. DEC C doesn't like long long...
Andy Polyakov [Sun, 25 Jul 2004 16:54:08 +0000 (16:54 +0000)]
Some compilers are just too whiny. DEC C doesn't like long long...

18 years agoTypos and due casts. As for the latter. It's "safe" to cast as below,
Andy Polyakov [Sun, 25 Jul 2004 16:48:28 +0000 (16:48 +0000)]
Typos and due casts. As for the latter. It's "safe" to cast as below,
because "wrong" casts will either be optimized away or never performed.

18 years agoMinor HP-UX make update. IA-64-based HP-UX favor .so extension for shared
Andy Polyakov [Sat, 24 Jul 2004 14:17:32 +0000 (14:17 +0000)]
Minor HP-UX make update. IA-64-based HP-UX favor .so extension for shared
libraries. Old .sl extension works just fine, but it .so which is default.

18 years agoTABLE update.
Andy Polyakov [Fri, 23 Jul 2004 23:29:25 +0000 (23:29 +0000)]
TABLE update.

18 years agoBuild-n-link new IA-64 modules on Linux and HP-UX.
Andy Polyakov [Fri, 23 Jul 2004 23:27:10 +0000 (23:27 +0000)]
Build-n-link new IA-64 modules on Linux and HP-UX.

18 years agoVarious IA-64 assembler fix-ups.
Andy Polyakov [Fri, 23 Jul 2004 22:54:18 +0000 (22:54 +0000)]
Various IA-64 assembler fix-ups.

18 years agoVC-NT was taken away by mistake, putting it back...
Andy Polyakov [Fri, 23 Jul 2004 20:46:41 +0000 (20:46 +0000)]
VC-NT was taken away by mistake, putting it back...

18 years agoWin64 placeholder targets. This is merely naming suggestion. As we know
Andy Polyakov [Fri, 23 Jul 2004 19:18:20 +0000 (19:18 +0000)]
Win64 placeholder targets. This is merely naming suggestion. As we know
Win64 comes in two flavors, IA-64/Itanium and AMD64/Opteron. The
suggestion is to refer to former as WIN64I and latter - WIN64A

18 years agoAdapt for Win64/AMD64 assembler.
Andy Polyakov [Fri, 23 Jul 2004 17:51:17 +0000 (17:51 +0000)]
Adapt for Win64/AMD64 assembler.

18 years agoFrom LPlib:
Richard Levitte [Thu, 22 Jul 2004 18:34:06 +0000 (18:34 +0000)]
From LPlib:

Apparently, the length *including* the NUL byte should be used.

Contributed by Andy Polyakov <>