abc/GmSSL
1
0
Fork 0
mirror of https://github.com/guanzhi/GmSSL.git synced 2026-08-08 23:13:38 +08:00
Code Issues Packages Projects Releases Wiki Activity

Revert "quantum init"

Browse Source 
This reverts commit 53af3b51ae.
This commit is contained in:
Simon
2018-01-04 14:03:27 +08:00
parent 53af3b51ae
commit 38395c5c80
2361 changed files with 144430 additions and 387427 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
crypto/des/asm/crypt586.pl
View File

@@ -1,15 +1,21 @@
#!/usr/local/bin/perl
#! /usr/bin/env perl
# Copyright 1995-2016 The OpenSSL Project Authors. All Rights Reserved.
#
# Licensed under the OpenSSL license (the "License"). You may not use
# this file except in compliance with the License. You can obtain a copy
# in the file LICENSE in the source distribution or at
# https://www.openssl.org/source/license.html
# The inner loop instruction sequence and the IP/FP modifications are from
# Svend Olaf Mikkelsen <svolaf@inet.uni-c.dk>
# I've added the stuff needed for crypt() but I've not worried about making
# things perfect.
#
$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
push(@INC,"${dir}","${dir}../../perlasm");
require "x86asm.pl";
$output=pop;
open STDOUT,">$output";
&asm_init($ARGV[0],"crypt586.pl");
$L="edi";
@@ -19,6 +25,8 @@ $R="esi";
&fcrypt_body("fcrypt_body");
&asm_finish();
close STDOUT;
sub fcrypt_body
{
local($name,$do_ip)=@_;

14
crypto/des/asm/des-586.pl
View File

@@ -1,8 +1,13 @@
#!/usr/local/bin/perl
#! /usr/bin/env perl
# Copyright 1995-2016 The OpenSSL Project Authors. All Rights Reserved.
#
# Licensed under the OpenSSL license (the "License"). You may not use
# this file except in compliance with the License. You can obtain a copy
# in the file LICENSE in the source distribution or at
# https://www.openssl.org/source/license.html
# The inner loop instruction sequence and the IP/FP modifications are from
# Svend Olaf Mikkelsen <svolaf@inet.uni-c.dk>
#
$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
push(@INC,"${dir}","${dir}../../perlasm");
@@ -15,6 +20,9 @@ require "desboth.pl";
# format.
#
$output=pop;
open STDOUT,">$output";
&asm_init($ARGV[0],"des-586.pl");
$L="edi";
@@ -39,6 +47,8 @@ $small_footprint=1 if (grep(/\-DOPENSSL_SMALL_FOOTPRINT/,@ARGV));
&asm_finish();
close STDOUT;
sub DES_encrypt_internal()
{
&function_begin_B("_x86_DES_encrypt");

151
crypto/des/asm/des_enc.m4
View File

@@ -1,26 +1,9 @@
! des_enc.m4
! des_enc.S (generated from des_enc.m4)
! Copyright 2000-2016 The OpenSSL Project Authors. All Rights Reserved.
!
! UltraSPARC assembler version of the LibDES/SSLeay/OpenSSL des_enc.c file.
!
! Version 1.0. 32-bit version.
!
! June 8, 2000.
!
! Version 2.0. 32/64-bit, PIC-ification, blended CPU adaptation
! by Andy Polyakov.
!
! January 1, 2003.
!
! Assembler version: Copyright Svend Olaf Mikkelsen.
!
! Original C code: Copyright Eric A. Young.
!
! This code can be freely used by LibDES/SSLeay/OpenSSL users.
!
! The LibDES/SSLeay/OpenSSL copyright notices must be respected.
!
! This version can be redistributed.
! Licensed under the OpenSSL license (the "License"). You may not use
! this file except in compliance with the License. You can obtain a copy
! in the file LICENSE in the source distribution or at
! https://www.openssl.org/source/license.html
!
! To expand the m4 macros: m4 -B 8192 des_enc.m4 > des_enc.S
!
@@ -48,6 +31,10 @@
#include <openssl/opensslconf.h>
#ifdef OPENSSL_FIPSCANISTER
#include <openssl/fipssyms.h>
#endif
#if defined(__SUNPRO_C) && defined(__sparcv9)
# define ABI64 /* They've said -xarch=v9 at command line */
#elif defined(__GNUC__) && defined(__arch64__)
@@ -63,9 +50,6 @@
# define STPTR stx
# define ARG0 128
# define ARGSZ 8
# ifndef OPENSSL_SYSNAME_ULTRASPARC
# define OPENSSL_SYSNAME_ULTRASPARC
# endif
#else
# define FRAME -96
# define BIAS 0
@@ -268,7 +252,7 @@ define(ip_macro, {
! other half (use).
!
! In this version we do two rounds in a loop repeated 7 times
! and two rounds seperately.
! and two rounds separately.
!
! One half has the bits for the sboxes in the following positions:
!
@@ -425,11 +409,7 @@ $4:
xor $2, local1, $2 ! 1 finished
xor $2, local2, $2 ! 3 finished
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bne,pt %icc, $4
#else
bne $4
#endif
and local4, 252, local1 ! sbox 1 next round
! two rounds more:
@@ -787,18 +767,6 @@ define(load_little_endian, {
! first in memory to rightmost in register
#ifdef OPENSSL_SYSNAME_ULTRASPARC
andcc $1, 3, global0
bne,pn %icc, $5
nop
lda [$1] 0x88, $2
add $1, 4, $4
ba,pt %icc, $5a
lda [$4] 0x88, $3
#endif
$5:
ldub [$1+3], $2
@@ -850,19 +818,6 @@ define(load_little_endian_inc, {
! first in memory to rightmost in register
#ifdef OPENSSL_SYSNAME_ULTRASPARC
andcc $1, 3, global0
bne,pn %icc, $5
nop
lda [$1] 0x88, $2
add $1, 4, $1
lda [$1] 0x88, $3
ba,pt %icc, $5a
add $1, 4, $1
#endif
$5:
ldub [$1+3], $2
@@ -985,18 +940,6 @@ define(store_little_endian, {
! rightmost in register to first in memory
#ifdef OPENSSL_SYSNAME_ULTRASPARC
andcc $1, 3, global0
bne,pn %icc, $5
nop
sta $2, [$1] 0x88
add $1, 4, $4
ba,pt %icc, $5a
sta $3, [$4] 0x88
#endif
$5:
and $2, 255, $4
stub $4, [$1+0]
@@ -1191,11 +1134,7 @@ DES_encrypt1:
ld [in0], in5 ! left
cmp in2, 0 ! enc
#ifdef OPENSSL_SYSNAME_ULTRASPARC
be,pn %icc, .encrypt.dec ! enc/dec
#else
be .encrypt.dec
#endif
ld [in0+4], out5 ! right
! parameter 6 1/2 for include encryption/decryption
@@ -1283,11 +1222,7 @@ DES_encrypt2:
! we use our own stackframe
#ifdef OPENSSL_SYSNAME_ULTRASPARC
be,pn %icc, .encrypt2.dec ! decryption
#else
be .encrypt2.dec
#endif
STPTR in0, [%sp+BIAS+ARG0+0*ARGSZ]
ld [in3], out0 ! key 7531 first round
@@ -1463,11 +1398,7 @@ DES_ncbc_encrypt:
cmp in5, 0 ! enc
#ifdef OPENSSL_SYSNAME_ULTRASPARC
be,pn %icc, .ncbc.dec
#else
be .ncbc.dec
#endif
STPTR in4, IVEC
! addr left right temp label
@@ -1475,11 +1406,7 @@ DES_ncbc_encrypt:
addcc in2, -8, in2 ! bytes missing when first block done
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bl,pn %icc, .ncbc.enc.seven.or.less
#else
bl .ncbc.enc.seven.or.less
#endif
mov in3, in4 ! schedule
.ncbc.enc.next.block:
@@ -1503,11 +1430,7 @@ DES_ncbc_encrypt:
rounds_macro(in5, out5, 1, .ncbc.enc.1, in3, in4) ! include encryption ks in3
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bl,pn %icc, .ncbc.enc.next.block_fp
#else
bl .ncbc.enc.next.block_fp
#endif
add in0, 8, in0 ! input address
! If 8 or more bytes are to be encrypted after this block,
@@ -1538,7 +1461,7 @@ DES_ncbc_encrypt:
xor global4, local1, out5 ! iv xor next block
ba .ncbc.enc.next.block_2
add in1, 8, in1 ! output adress
add in1, 8, in1 ! output address
.ncbc.enc.next.block_fp:
@@ -1548,22 +1471,14 @@ DES_ncbc_encrypt:
addcc in2, -8, in2 ! bytes missing when next block done
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bpos,pt %icc, .ncbc.enc.next.block ! also jumps if 0
#else
bpos .ncbc.enc.next.block
#endif
add in1, 8, in1
.ncbc.enc.seven.or.less:
cmp in2, -8
#ifdef OPENSSL_SYSNAME_ULTRASPARC
ble,pt %icc, .ncbc.enc.finish
#else
ble .ncbc.enc.finish
#endif
nop
add in2, 8, local1 ! bytes to load
@@ -1590,11 +1505,7 @@ DES_ncbc_encrypt:
add in3, 120, in3
LDPTR IVEC, local7 ! ivec
#ifdef OPENSSL_SYSNAME_ULTRASPARC
ble,pn %icc, .ncbc.dec.finish
#else
ble .ncbc.dec.finish
#endif
mov in3, in4 ! schedule
STPTR in1, OUTPUT
@@ -1618,11 +1529,7 @@ DES_ncbc_encrypt:
! in2 is compared to 8 in the rounds
xor out5, in0, out4 ! iv xor
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bl,pn %icc, .ncbc.dec.seven.or.less
#else
bl .ncbc.dec.seven.or.less
#endif
xor in5, in1, global4 ! iv xor
! Load ivec next block now, since input and output address might be the same.
@@ -1635,11 +1542,7 @@ DES_ncbc_encrypt:
add local7, 8, local7
addcc in2, -8, in2
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bg,pt %icc, .ncbc.dec.next.block
#else
bg .ncbc.dec.next.block
#endif
STPTR local7, OUTPUT
@@ -1690,11 +1593,7 @@ DES_ede3_cbc_encrypt:
LDPTR [%fp+BIAS+ARG0+6*ARGSZ], local4 ! ivec
cmp local3, 0 ! enc
#ifdef OPENSSL_SYSNAME_ULTRASPARC
be,pn %icc, .ede3.dec
#else
be .ede3.dec
#endif
STPTR in4, KS2
STPTR in5, KS3
@@ -1703,11 +1602,7 @@ DES_ede3_cbc_encrypt:
addcc in2, -8, in2 ! bytes missing after next block
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bl,pn %icc, .ede3.enc.seven.or.less
#else
bl .ede3.enc.seven.or.less
#endif
STPTR in3, KS1
.ede3.enc.next.block:
@@ -1737,11 +1632,7 @@ DES_ede3_cbc_encrypt:
call .des_enc ! ks3 in3 compares in2 to 8
nop
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bl,pn %icc, .ede3.enc.next.block_fp
#else
bl .ede3.enc.next.block_fp
#endif
add in0, 8, in0
! If 8 or more bytes are to be encrypted after this block,
@@ -1783,22 +1674,14 @@ DES_ede3_cbc_encrypt:
addcc in2, -8, in2 ! bytes missing when next block done
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bpos,pt %icc, .ede3.enc.next.block
#else
bpos .ede3.enc.next.block
#endif
add in1, 8, in1
.ede3.enc.seven.or.less:
cmp in2, -8
#ifdef OPENSSL_SYSNAME_ULTRASPARC
ble,pt %icc, .ede3.enc.finish
#else
ble .ede3.enc.finish
#endif
nop
add in2, 8, local1 ! bytes to load
@@ -1826,11 +1709,7 @@ DES_ede3_cbc_encrypt:
STPTR in3, KS1
cmp in2, 0
#ifdef OPENSSL_SYSNAME_ULTRASPARC
ble %icc, .ede3.dec.finish
#else
ble .ede3.dec.finish
#endif
STPTR in5, KS3
LDPTR [%fp+BIAS+ARG0+6*ARGSZ], local7 ! iv
@@ -1859,11 +1738,7 @@ DES_ede3_cbc_encrypt:
! in2 is compared to 8 in the rounds
xor out5, in0, out4
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bl,pn %icc, .ede3.dec.seven.or.less
#else
bl .ede3.dec.seven.or.less
#endif
xor in5, in1, global4
load_little_endian_inc(local5, in0, in1, local3, .LLE10) ! iv next block
@@ -1874,11 +1749,7 @@ DES_ede3_cbc_encrypt:
addcc in2, -8, in2
add local7, 8, local7
#ifdef OPENSSL_SYSNAME_ULTRASPARC
bg,pt %icc, .ede3.dec.next.block
#else
bg .ede3.dec.next.block
#endif
STPTR local7, OUTPUT
.ede3.dec.store.iv:

9
crypto/des/asm/desboth.pl
View File

@@ -1,4 +1,11 @@
#!/usr/local/bin/perl
#! /usr/bin/env perl
# Copyright 1995-2016 The OpenSSL Project Authors. All Rights Reserved.
#
# Licensed under the OpenSSL license (the "License"). You may not use
# this file except in compliance with the License. You can obtain a copy
# in the file LICENSE in the source distribution or at
# https://www.openssl.org/source/license.html
$L="edi";
$R="esi";

24
crypto/des/asm/dest4-sparcv9.pl
View File

@@ -1,4 +1,11 @@
#!/usr/bin/env perl
#! /usr/bin/env perl
# Copyright 2013-2016 The OpenSSL Project Authors. All Rights Reserved.
#
# Licensed under the OpenSSL license (the "License"). You may not use
# this file except in compliance with the License. You can obtain a copy
# in the file LICENSE in the source distribution or at
# https://www.openssl.org/source/license.html
# ====================================================================
# Written by David S. Miller <davem@devemloft.net> and Andy Polyakov
@@ -27,14 +34,17 @@ $0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
push(@INC,"${dir}","${dir}../../perlasm");
require "sparcv9_modes.pl";
&asm_init(@ARGV);
$code.=<<___ if ($::abibits==64);
.register %g2,#scratch
.register %g3,#scratch
___
$output=pop;
open STDOUT,">$output";
$code.=<<___;
#include "sparc_arch.h"
#ifdef __arch64__
.register %g2,#scratch
.register %g3,#scratch
#endif
.text
___

131
crypto/des/asm/readme
View File

@@ -1,131 +0,0 @@
First up, let me say I don't like writing in assembler. It is not portable,
dependant on the particular CPU architecture release and is generally a pig
to debug and get right. Having said that, the x86 architecture is probably
the most important for speed due to number of boxes and since
it appears to be the worst architecture to to get
good C compilers for. So due to this, I have lowered myself to do
assembler for the inner DES routines in libdes :-).
The file to implement in assembler is des_enc.c. Replace the following
4 functions
des_encrypt1(DES_LONG data[2],des_key_schedule ks, int encrypt);
des_encrypt2(DES_LONG data[2],des_key_schedule ks, int encrypt);
des_encrypt3(DES_LONG data[2],des_key_schedule ks1,ks2,ks3);
des_decrypt3(DES_LONG data[2],des_key_schedule ks1,ks2,ks3);
They encrypt/decrypt the 64 bits held in 'data' using
the 'ks' key schedules. The only difference between the 4 functions is that
des_encrypt2() does not perform IP() or FP() on the data (this is an
optimization for when doing triple DES and des_encrypt3() and des_decrypt3()
perform triple des. The triple DES routines are in here because it does
make a big difference to have them located near the des_encrypt2 function
at link time..
Now as we all know, there are lots of different operating systems running on
x86 boxes, and unfortunately they normally try to make sure their assembler
formating is not the same as the other peoples.
The 4 main formats I know of are
Microsoft Windows 95/Windows NT
Elf Includes Linux and FreeBSD(?).
a.out The older Linux.
Solaris Same as Elf but different comments :-(.
Now I was not overly keen to write 4 different copies of the same code,
so I wrote a few perl routines to output the correct assembler, given
a target assembler type. This code is ugly and is just a hack.
The libraries are x86unix.pl and x86ms.pl.
des586.pl, des686.pl and des-som[23].pl are the programs to actually
generate the assembler.
So to generate elf assembler
perl des-som3.pl elf >dx86-elf.s
For Windows 95/NT
perl des-som2.pl win32 >win32.asm
[ update 4 Jan 1996 ]
I have added another way to do things.
perl des-som3.pl cpp >dx86-cpp.s
generates a file that will be included by dx86unix.cpp when it is compiled.
To build for elf, a.out, solaris, bsdi etc,
cc -E -DELF asm/dx86unix.cpp | as -o asm/dx86-elf.o
cc -E -DSOL asm/dx86unix.cpp | as -o asm/dx86-sol.o
cc -E -DOUT asm/dx86unix.cpp | as -o asm/dx86-out.o
cc -E -DBSDI asm/dx86unix.cpp | as -o asm/dx86bsdi.o
This was done to cut down the number of files in the distribution.
Now the ugly part. I acquired my copy of Intels
"Optimization's For Intel's 32-Bit Processors" and found a few interesting
things. First, the aim of the exersize is to 'extract' one byte at a time
from a word and do an array lookup. This involves getting the byte from
the 4 locations in the word and moving it to a new word and doing the lookup.
The most obvious way to do this is
xor eax, eax # clear word
movb al, cl # get low byte
xor edi DWORD PTR 0x100+des_SP[eax] # xor in word
movb al, ch # get next byte
xor edi DWORD PTR 0x300+des_SP[eax] # xor in word
shr ecx 16
which seems ok. For the pentium, this system appears to be the best.
One has to do instruction interleaving to keep both functional units
operating, but it is basically very efficient.
Now the crunch. When a full register is used after a partial write, eg.
mov al, cl
xor edi, DWORD PTR 0x100+des_SP[eax]
386 - 1 cycle stall
486 - 1 cycle stall
586 - 0 cycle stall
686 - at least 7 cycle stall (page 22 of the above mentioned document).
So the technique that produces the best results on a pentium, according to
the documentation, will produce hideous results on a pentium pro.
To get around this, des686.pl will generate code that is not as fast on
a pentium, should be very good on a pentium pro.
mov eax, ecx # copy word
shr ecx, 8 # line up next byte
and eax, 0fch # mask byte
xor edi DWORD PTR 0x100+des_SP[eax] # xor in array lookup
mov eax, ecx # get word
shr ecx 8 # line up next byte
and eax, 0fch # mask byte
xor edi DWORD PTR 0x300+des_SP[eax] # xor in array lookup
Due to the execution units in the pentium, this actually works quite well.
For a pentium pro it should be very good. This is the type of output
Visual C++ generates.
There is a third option. instead of using
mov al, ch
which is bad on the pentium pro, one may be able to use
movzx eax, ch
which may not incur the partial write penalty. On the pentium,
this instruction takes 4 cycles so is not worth using but on the
pentium pro it appears it may be worth while. I need access to one to
experiment :-).
eric (20 Oct 1996)
22 Nov 1996 - I have asked people to run the 2 different version on pentium
pros and it appears that the intel documentation is wrong. The
mov al,bh is still faster on a pentium pro, so just use the des586.pl
install des686.pl
3 Dec 1996 - I added des_encrypt3/des_decrypt3 because I have moved these
functions into des_enc.c because it does make a massive performance
difference on some boxes to have the functions code located close to
the des_encrypt2() function.
9 Jan 1997 - des-som2.pl is now the correct perl script to use for
pentiums. It contains an inner loop from
Svend Olaf Mikkelsen <svolaf@inet.uni-c.dk> which does raw ecb DES calls at
273,000 per second. He had a previous version at 250,000 and the best
I was able to get was 203,000. The content has not changed, this is all
due to instruction sequencing (and actual instructions choice) which is able
to keep both functional units of the pentium going.
We may have lost the ugly register usage restrictions when x86 went 32 bit
but for the pentium it has been replaced by evil instruction ordering tricks.
13 Jan 1997 - des-som3.pl, more optimizations from Svend Olaf.
raw DES at 281,000 per second on a pentium 100.