Commit graph

77 commits

Author SHA1 Message Date
Bodo Möller
812d8a176c Unobtrusive backport of 32-bit x86 Montgomery improvements from 0.9.9-dev:
you need to use "enable-montasm" to see a difference.  (Huge speed
advantage, but BN_MONT_CTX is not binary compatible, so this can't be
enabled by default in the 0.9.8 branch.)

The CHANGES entry also covers the 64-bit x86 backport in November 2007
by appro.
2008-05-01 23:11:34 +00:00
Andy Polyakov
2923e91a98 Allow 32-bit perl to generate x86_64 assembler. 2008-02-13 20:01:48 +00:00
Andy Polyakov
4f466f8e81 x86_64-xlate.pl update from HEAD. 2008-02-11 13:07:11 +00:00
Dr. Stephen Henson
3b0e61a812 Netware support.
Submitted by: Guenter Knauf <eflash@gmx.net>
2008-01-03 22:53:06 +00:00
Andy Polyakov
095db72024 x86cpuid.pl update [from HEAD]. 2007-11-11 19:44:42 +00:00
Andy Polyakov
6f57311da0 Add SHA x86_64 assembler [from HEAD]. 2007-11-11 13:56:47 +00:00
Andy Polyakov
5f761514e1 Make it possible for older masm to compile sse2 modules.
PR: 1592
2007-10-21 14:15:40 +00:00
Andy Polyakov
a8098740c6 Relax ISA detection. 2007-07-19 10:45:03 +00:00
Andy Polyakov
e3af0d041e Fix masm type-casting problem in SSE2 code. 2007-07-18 20:25:17 +00:00
Dr. Stephen Henson
269d2575cd Fix from HEAD. 2006-02-04 01:26:49 +00:00
Andy Polyakov
05decf3638 Perlasm update from HEAD addressing build problems on non-ELF platforms
after http://cvs.openssl.org/chngview?cn=14661.
2005-12-06 18:47:18 +00:00
Andy Polyakov
cb59297438 Proper solution to nasm compilation problems in Borland context. 2005-09-20 06:15:18 +00:00
Andy Polyakov
b44e2d3416 x86nasm.pl update from HEAD. 2005-06-28 11:30:12 +00:00
Andy Polyakov
8d56c8b51c Solaris x86_64 /usr/ccs/bin/as support. 2005-06-20 14:50:26 +00:00
Andy Polyakov
e5cd536894 Some assemblers are too rudimentary to understand dynamic labels. 2005-05-25 21:37:18 +00:00
Andy Polyakov
225f4daf15 Still SEGV trouble in .init segment under Solaris x86... 2005-05-21 17:51:12 +00:00
Andy Polyakov
e19e549041 Comply with optimization manual (no data should share cache-line with code). 2005-05-09 21:48:01 +00:00
Andy Polyakov
d7561ac576 Allow for 64-bit cache-line alignments in code segment. 2005-05-09 21:27:40 +00:00
Andy Polyakov
5d0d60e2f5 x86_64 assembler translator update. 2005-05-07 08:13:51 +00:00
Andy Polyakov
70cf309517 x86_64 assembler translator update. 2005-05-04 08:42:47 +00:00
Andy Polyakov
4b45051902 x86_64 assembler translator update. 2005-05-03 15:35:14 +00:00
Andy Polyakov
3d5fd31280 Avoid L1 cache aliasing even between key and S-boxes. 2005-04-24 21:09:20 +00:00
Andy Polyakov
1cfd258ed6 Throw in x86_64 AT&T to MASM assembler converter to facilitate development
of dual-ABI Unix/Win64 modules.
2005-04-17 21:05:57 +00:00
Andy Polyakov
e5dbccc182 Solaris x86 linker erroneously pads .init segment with zeros instead of
nops, which causes SEGV at startup. So I don't align anymore.
2005-04-04 17:07:16 +00:00
Andy Polyakov
da30c74a27 Remove unused assembler modules. 2005-02-06 13:43:02 +00:00
Andy Polyakov
addb6e16a8 Throw in AES CBC assembler, up to +40% on aes-128-cbc benchmark. 2005-01-18 01:04:41 +00:00
Andy Polyakov
556b8f3f77 Engage AES x86 assembler module for COFF and a.out targets. 2004-12-26 10:58:39 +00:00
Andy Polyakov
d1df5b4339 x86 perlasm update to accomodate aes-586.pl. 2004-12-23 21:43:25 +00:00
Andy Polyakov
0c0788ba0a Solaris x86 perlasm update. 2004-12-10 11:24:42 +00:00
Andy Polyakov
914c2a28c0 perlasm/x86[ms|nasm] update to accomodate updated RC4 assembler module. 2004-11-27 15:14:58 +00:00
Andy Polyakov
376729e130 RC4 tune-up for Intel P4 core, both 32- and 64-bit ones. As it's
apparently impossible to compose blended code with would perform
satisfactory on all x86 and x86_64 cores, an extra RC4_CHAR
code-path is introduced and P4 core is detected at run-time. This
way we keep original performance on non-P4 implementations and
turbo-charge P4 performance by factor of 2.8x (on 32-bit core).
2004-11-21 10:36:25 +00:00
Andy Polyakov
07d488daf6 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
2004-09-28 20:45:10 +00:00
Andy Polyakov
c85c5c408a x86 assembler updates: more instructions, new OPENSSL_instrument_halt
[for DJGPP]...
2004-09-09 14:50:32 +00:00
Andy Polyakov
16760a3089 Proper support for OpenBSD-i386 shared build, including assember modules!
"Proper" means "compiles and passes test." Versioning is broken (I think).
2004-08-29 21:36:37 +00:00
Andy Polyakov
a8c65b400c crypto/perlasm update primarily to unify Netware modules. Once it's verified
x86*_nw.pl 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().
2004-08-29 16:10:27 +00:00
Andy Polyakov
42096e05f7 Avoid a.out name table pollition. 2004-08-02 22:02:17 +00:00
Andy Polyakov
34413fca84 OpenBSD fix-up for new a.out targets. OpenBSD .s.o rule is busted... 2004-08-01 21:16:26 +00:00
Andy Polyakov
8aae01e223 Deprecate cpp and gaswin targets. New coff fills in for gaswin, but cpp is
going out...
2004-08-01 17:03:50 +00:00
Andy Polyakov
14e21f863a 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...
2004-07-26 20:18:55 +00:00
Andy Polyakov
10e7d6d526 Support for IA-32 SSE2 instruction set. 2004-05-06 10:31:09 +00:00
Richard Levitte
4d8743f490 Netware-specific changes,
PR: 780
Submitted by: Verdon Walker <VWalker@novell.com>
Reviewed by: Richard Levitte
2003-11-28 13:10:58 +00:00
Richard Levitte
1be02dd842 Make MD5 assembler code able to handle messages larger than 2GB on 32-bit
systems and above.
PR: 664
2003-09-27 22:14:39 +00:00
Andy Polyakov
02bf9a151a Provide "dummy" &main::picmeup even in Windows perlasm modules. 2003-01-24 09:39:31 +00:00
Andy Polyakov
989f48d8c1 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.
2003-01-10 10:16:39 +00:00
Andy Polyakov
9d1b5614f9 At least OpenBSD implements PIC in the same way ELF does. 2003-01-09 16:06:09 +00:00
Andy Polyakov
2a9fca82d2 Oops! I've toasted Cygwin! Fixed now. 2003-01-03 21:49:49 +00:00
Andy Polyakov
3cc9a89dda Unified targets for ELF assembler modules. Tested on Linux, Solaris and
FreeBSD. Goal is to extend support even to SCO5, UnixWare/OpenUnix...
2003-01-03 17:37:53 +00:00
Andy Polyakov
4553ed276b Make "perl des-586.pl a.out" work, see RT#402 2002-12-17 08:05:49 +00:00
Andy Polyakov
a45051fbfc DES PIC-ification. "Cygwin" companion. Problem was that preprocessor macro
is not expanded if prepended with a $-sign.
2002-12-15 10:06:27 +00:00
Andy Polyakov
717c5cdcc7 As you might have noticed I tried to change for . prefix, because it's
the one to be used to denote local labels in single function scope.
Problem is that SHA uses same label set across functions, therefore I
have to switch back to $ prefix.
2002-12-14 23:14:00 +00:00