external/srs
1
0
Fork 0
mirror of https://github.com/ossrs/srs.git synced 2025-03-09 15:49:59 +00:00
Code Issues Releases Wiki Activity

AppleM1: Update openssl to v1.1.1l

Browse source
This commit is contained in:
winlin 2022-08-14 19:05:01 +08:00
parent 1fe12b8e8c
commit b787656eea
990 changed files with 13406 additions and 18710 deletions
Download patch file Download diff file Expand all files Collapse all files

636
trunk/3rdparty/openssl-1.1-fit/crypto/aes/aes_core.c vendored
View file

@ -1,5 +1,5 @@
/*
* Copyright 2002-2016 The OpenSSL Project Authors. All Rights Reserved.
* Copyright 2002-2020 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
@ -41,9 +41,639 @@
#include <stdlib.h>
#include <openssl/crypto.h>
#include <openssl/aes.h>
#include "aes_locl.h"
#include "aes_local.h"
#ifndef AES_ASM
#if defined(OPENSSL_AES_CONST_TIME) && !defined(AES_ASM)
typedef union {
unsigned char b[8];
u32 w[2];
u64 d;
} uni;
/*
* Compute w := (w * x) mod (x^8 + x^4 + x^3 + x^1 + 1)
* Therefore the name "xtime".
*/
static void XtimeWord(u32 *w)
{
u32 a, b;
a = *w;
b = a & 0x80808080u;
a ^= b;
b -= b >> 7;
b &= 0x1B1B1B1Bu;
b ^= a << 1;
*w = b;
}
static void XtimeLong(u64 *w)
{
u64 a, b;
a = *w;
b = a & 0x8080808080808080uLL;
a ^= b;
b -= b >> 7;
b &= 0x1B1B1B1B1B1B1B1BuLL;
b ^= a << 1;
*w = b;
}
/*
* This computes w := S * w ^ -1 + c, where c = {01100011}.
* Instead of using GF(2^8) mod (x^8+x^4+x^3+x+1} we do the inversion
* in GF(GF(GF(2^2)^2)^2) mod (X^2+X+8)
* and GF(GF(2^2)^2) mod (X^2+X+2)
* and GF(2^2) mod (X^2+X+1)
* The first part of the algorithm below transfers the coordinates
* {0x01,0x02,0x04,0x08,0x10,0x20,0x40,0x80} =>
* {1,Y,Y^2,Y^3,Y^4,Y^5,Y^6,Y^7} with Y=0x41:
* {0x01,0x41,0x66,0x6c,0x56,0x9a,0x58,0xc4}
* The last part undoes the coordinate transfer and the final affine
* transformation S:
* b[i] = b[i] + b[(i+4)%8] + b[(i+5)%8] + b[(i+6)%8] + b[(i+7)%8] + c[i]
* in one step.
* The multiplication in GF(2^2^2^2) is done in ordinary coords:
* A = (a0*1 + a1*x^4)
* B = (b0*1 + b1*x^4)
* AB = ((a0*b0 + 8*a1*b1)*1 + (a1*b0 + (a0+a1)*b1)*x^4)
* When A = (a0,a1) is given we want to solve AB = 1:
* (a) 1 = a0*b0 + 8*a1*b1
* (b) 0 = a1*b0 + (a0+a1)*b1
* => multiply (a) by a1 and (b) by a0
* (c) a1 = a1*a0*b0 + (8*a1*a1)*b1
* (d) 0 = a1*a0*b0 + (a0*a0+a1*a0)*b1
* => add (c) + (d)
* (e) a1 = (a0*a0 + a1*a0 + 8*a1*a1)*b1
* => therefore
* b1 = (a0*a0 + a1*a0 + 8*a1*a1)^-1 * a1
* => and adding (a1*b0) to (b) we get
* (f) a1*b0 = (a0+a1)*b1
* => therefore
* b0 = (a0*a0 + a1*a0 + 8*a1*a1)^-1 * (a0+a1)
* Note this formula also works for the case
* (a0+a1)*a0 + 8*a1*a1 = 0
* if the inverse element for 0^-1 is mapped to 0.
* Repeat the same for GF(2^2^2) and GF(2^2).
* We get the following algorithm:
* inv8(a0,a1):
* x0 = a0^a1
* [y0,y1] = mul4([x0,a1],[a0,a1]); (*)
* y1 = mul4(8,y1);
* t = inv4(y0^y1);
* [b0,b1] = mul4([x0,a1],[t,t]); (*)
* return [b0,b1];
* The non-linear multiplies (*) can be done in parallel at no extra cost.
*/
static void SubWord(u32 *w)
{
u32 x, y, a1, a2, a3, a4, a5, a6;
x = *w;
y = ((x & 0xFEFEFEFEu) >> 1) | ((x & 0x01010101u) << 7);
x &= 0xDDDDDDDDu;
x ^= y & 0x57575757u;
y = ((y & 0xFEFEFEFEu) >> 1) | ((y & 0x01010101u) << 7);
x ^= y & 0x1C1C1C1Cu;
y = ((y & 0xFEFEFEFEu) >> 1) | ((y & 0x01010101u) << 7);
x ^= y & 0x4A4A4A4Au;
y = ((y & 0xFEFEFEFEu) >> 1) | ((y & 0x01010101u) << 7);
x ^= y & 0x42424242u;
y = ((y & 0xFEFEFEFEu) >> 1) | ((y & 0x01010101u) << 7);
x ^= y & 0x64646464u;
y = ((y & 0xFEFEFEFEu) >> 1) | ((y & 0x01010101u) << 7);
x ^= y & 0xE0E0E0E0u;
a1 = x;
a1 ^= (x & 0xF0F0F0F0u) >> 4;
a2 = ((x & 0xCCCCCCCCu) >> 2) | ((x & 0x33333333u) << 2);
a3 = x & a1;
a3 ^= (a3 & 0xAAAAAAAAu) >> 1;
a3 ^= (((x << 1) & a1) ^ ((a1 << 1) & x)) & 0xAAAAAAAAu;
a4 = a2 & a1;
a4 ^= (a4 & 0xAAAAAAAAu) >> 1;
a4 ^= (((a2 << 1) & a1) ^ ((a1 << 1) & a2)) & 0xAAAAAAAAu;
a5 = (a3 & 0xCCCCCCCCu) >> 2;
a3 ^= ((a4 << 2) ^ a4) & 0xCCCCCCCCu;
a4 = a5 & 0x22222222u;
a4 |= a4 >> 1;
a4 ^= (a5 << 1) & 0x22222222u;
a3 ^= a4;
a5 = a3 & 0xA0A0A0A0u;
a5 |= a5 >> 1;
a5 ^= (a3 << 1) & 0xA0A0A0A0u;
a4 = a5 & 0xC0C0C0C0u;
a6 = a4 >> 2;
a4 ^= (a5 << 2) & 0xC0C0C0C0u;
a5 = a6 & 0x20202020u;
a5 |= a5 >> 1;
a5 ^= (a6 << 1) & 0x20202020u;
a4 |= a5;
a3 ^= a4 >> 4;
a3 &= 0x0F0F0F0Fu;
a2 = a3;
a2 ^= (a3 & 0x0C0C0C0Cu) >> 2;
a4 = a3 & a2;
a4 ^= (a4 & 0x0A0A0A0A0Au) >> 1;
a4 ^= (((a3 << 1) & a2) ^ ((a2 << 1) & a3)) & 0x0A0A0A0Au;
a5 = a4 & 0x08080808u;
a5 |= a5 >> 1;
a5 ^= (a4 << 1) & 0x08080808u;
a4 ^= a5 >> 2;
a4 &= 0x03030303u;
a4 ^= (a4 & 0x02020202u) >> 1;
a4 |= a4 << 2;
a3 = a2 & a4;
a3 ^= (a3 & 0x0A0A0A0Au) >> 1;
a3 ^= (((a2 << 1) & a4) ^ ((a4 << 1) & a2)) & 0x0A0A0A0Au;
a3 |= a3 << 4;
a2 = ((a1 & 0xCCCCCCCCu) >> 2) | ((a1 & 0x33333333u) << 2);
x = a1 & a3;
x ^= (x & 0xAAAAAAAAu) >> 1;
x ^= (((a1 << 1) & a3) ^ ((a3 << 1) & a1)) & 0xAAAAAAAAu;
a4 = a2 & a3;
a4 ^= (a4 & 0xAAAAAAAAu) >> 1;
a4 ^= (((a2 << 1) & a3) ^ ((a3 << 1) & a2)) & 0xAAAAAAAAu;
a5 = (x & 0xCCCCCCCCu) >> 2;
x ^= ((a4 << 2) ^ a4) & 0xCCCCCCCCu;
a4 = a5 & 0x22222222u;
a4 |= a4 >> 1;
a4 ^= (a5 << 1) & 0x22222222u;
x ^= a4;
y = ((x & 0xFEFEFEFEu) >> 1) | ((x & 0x01010101u) << 7);
x &= 0x39393939u;
x ^= y & 0x3F3F3F3Fu;
y = ((y & 0xFCFCFCFCu) >> 2) | ((y & 0x03030303u) << 6);
x ^= y & 0x97979797u;
y = ((y & 0xFEFEFEFEu) >> 1) | ((y & 0x01010101u) << 7);
x ^= y & 0x9B9B9B9Bu;
y = ((y & 0xFEFEFEFEu) >> 1) | ((y & 0x01010101u) << 7);
x ^= y & 0x3C3C3C3Cu;
y = ((y & 0xFEFEFEFEu) >> 1) | ((y & 0x01010101u) << 7);
x ^= y & 0xDDDDDDDDu;
y = ((y & 0xFEFEFEFEu) >> 1) | ((y & 0x01010101u) << 7);
x ^= y & 0x72727272u;
x ^= 0x63636363u;
*w = x;
}
static void SubLong(u64 *w)
{
u64 x, y, a1, a2, a3, a4, a5, a6;
x = *w;
y = ((x & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((x & 0x0101010101010101uLL) << 7);
x &= 0xDDDDDDDDDDDDDDDDuLL;
x ^= y & 0x5757575757575757uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x1C1C1C1C1C1C1C1CuLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x4A4A4A4A4A4A4A4AuLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x4242424242424242uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x6464646464646464uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0xE0E0E0E0E0E0E0E0uLL;
a1 = x;
a1 ^= (x & 0xF0F0F0F0F0F0F0F0uLL) >> 4;
a2 = ((x & 0xCCCCCCCCCCCCCCCCuLL) >> 2) | ((x & 0x3333333333333333uLL) << 2);
a3 = x & a1;
a3 ^= (a3 & 0xAAAAAAAAAAAAAAAAuLL) >> 1;
a3 ^= (((x << 1) & a1) ^ ((a1 << 1) & x)) & 0xAAAAAAAAAAAAAAAAuLL;
a4 = a2 & a1;
a4 ^= (a4 & 0xAAAAAAAAAAAAAAAAuLL) >> 1;
a4 ^= (((a2 << 1) & a1) ^ ((a1 << 1) & a2)) & 0xAAAAAAAAAAAAAAAAuLL;
a5 = (a3 & 0xCCCCCCCCCCCCCCCCuLL) >> 2;
a3 ^= ((a4 << 2) ^ a4) & 0xCCCCCCCCCCCCCCCCuLL;
a4 = a5 & 0x2222222222222222uLL;
a4 |= a4 >> 1;
a4 ^= (a5 << 1) & 0x2222222222222222uLL;
a3 ^= a4;
a5 = a3 & 0xA0A0A0A0A0A0A0A0uLL;
a5 |= a5 >> 1;
a5 ^= (a3 << 1) & 0xA0A0A0A0A0A0A0A0uLL;
a4 = a5 & 0xC0C0C0C0C0C0C0C0uLL;
a6 = a4 >> 2;
a4 ^= (a5 << 2) & 0xC0C0C0C0C0C0C0C0uLL;
a5 = a6 & 0x2020202020202020uLL;
a5 |= a5 >> 1;
a5 ^= (a6 << 1) & 0x2020202020202020uLL;
a4 |= a5;
a3 ^= a4 >> 4;
a3 &= 0x0F0F0F0F0F0F0F0FuLL;
a2 = a3;
a2 ^= (a3 & 0x0C0C0C0C0C0C0C0CuLL) >> 2;
a4 = a3 & a2;
a4 ^= (a4 & 0x0A0A0A0A0A0A0A0AuLL) >> 1;
a4 ^= (((a3 << 1) & a2) ^ ((a2 << 1) & a3)) & 0x0A0A0A0A0A0A0A0AuLL;
a5 = a4 & 0x0808080808080808uLL;
a5 |= a5 >> 1;
a5 ^= (a4 << 1) & 0x0808080808080808uLL;
a4 ^= a5 >> 2;
a4 &= 0x0303030303030303uLL;
a4 ^= (a4 & 0x0202020202020202uLL) >> 1;
a4 |= a4 << 2;
a3 = a2 & a4;
a3 ^= (a3 & 0x0A0A0A0A0A0A0A0AuLL) >> 1;
a3 ^= (((a2 << 1) & a4) ^ ((a4 << 1) & a2)) & 0x0A0A0A0A0A0A0A0AuLL;
a3 |= a3 << 4;
a2 = ((a1 & 0xCCCCCCCCCCCCCCCCuLL) >> 2) | ((a1 & 0x3333333333333333uLL) << 2);
x = a1 & a3;
x ^= (x & 0xAAAAAAAAAAAAAAAAuLL) >> 1;
x ^= (((a1 << 1) & a3) ^ ((a3 << 1) & a1)) & 0xAAAAAAAAAAAAAAAAuLL;
a4 = a2 & a3;
a4 ^= (a4 & 0xAAAAAAAAAAAAAAAAuLL) >> 1;
a4 ^= (((a2 << 1) & a3) ^ ((a3 << 1) & a2)) & 0xAAAAAAAAAAAAAAAAuLL;
a5 = (x & 0xCCCCCCCCCCCCCCCCuLL) >> 2;
x ^= ((a4 << 2) ^ a4) & 0xCCCCCCCCCCCCCCCCuLL;
a4 = a5 & 0x2222222222222222uLL;
a4 |= a4 >> 1;
a4 ^= (a5 << 1) & 0x2222222222222222uLL;
x ^= a4;
y = ((x & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((x & 0x0101010101010101uLL) << 7);
x &= 0x3939393939393939uLL;
x ^= y & 0x3F3F3F3F3F3F3F3FuLL;
y = ((y & 0xFCFCFCFCFCFCFCFCuLL) >> 2) | ((y & 0x0303030303030303uLL) << 6);
x ^= y & 0x9797979797979797uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x9B9B9B9B9B9B9B9BuLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x3C3C3C3C3C3C3C3CuLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0xDDDDDDDDDDDDDDDDuLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x7272727272727272uLL;
x ^= 0x6363636363636363uLL;
*w = x;
}
/*
* This computes w := (S^-1 * (w + c))^-1
*/
static void InvSubLong(u64 *w)
{
u64 x, y, a1, a2, a3, a4, a5, a6;
x = *w;
x ^= 0x6363636363636363uLL;
y = ((x & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((x & 0x0101010101010101uLL) << 7);
x &= 0xFDFDFDFDFDFDFDFDuLL;
x ^= y & 0x5E5E5E5E5E5E5E5EuLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0xF3F3F3F3F3F3F3F3uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0xF5F5F5F5F5F5F5F5uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x7878787878787878uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x7777777777777777uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x1515151515151515uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0xA5A5A5A5A5A5A5A5uLL;
a1 = x;
a1 ^= (x & 0xF0F0F0F0F0F0F0F0uLL) >> 4;
a2 = ((x & 0xCCCCCCCCCCCCCCCCuLL) >> 2) | ((x & 0x3333333333333333uLL) << 2);
a3 = x & a1;
a3 ^= (a3 & 0xAAAAAAAAAAAAAAAAuLL) >> 1;
a3 ^= (((x << 1) & a1) ^ ((a1 << 1) & x)) & 0xAAAAAAAAAAAAAAAAuLL;
a4 = a2 & a1;
a4 ^= (a4 & 0xAAAAAAAAAAAAAAAAuLL) >> 1;
a4 ^= (((a2 << 1) & a1) ^ ((a1 << 1) & a2)) & 0xAAAAAAAAAAAAAAAAuLL;
a5 = (a3 & 0xCCCCCCCCCCCCCCCCuLL) >> 2;
a3 ^= ((a4 << 2) ^ a4) & 0xCCCCCCCCCCCCCCCCuLL;
a4 = a5 & 0x2222222222222222uLL;
a4 |= a4 >> 1;
a4 ^= (a5 << 1) & 0x2222222222222222uLL;
a3 ^= a4;
a5 = a3 & 0xA0A0A0A0A0A0A0A0uLL;
a5 |= a5 >> 1;
a5 ^= (a3 << 1) & 0xA0A0A0A0A0A0A0A0uLL;
a4 = a5 & 0xC0C0C0C0C0C0C0C0uLL;
a6 = a4 >> 2;
a4 ^= (a5 << 2) & 0xC0C0C0C0C0C0C0C0uLL;
a5 = a6 & 0x2020202020202020uLL;
a5 |= a5 >> 1;
a5 ^= (a6 << 1) & 0x2020202020202020uLL;
a4 |= a5;
a3 ^= a4 >> 4;
a3 &= 0x0F0F0F0F0F0F0F0FuLL;
a2 = a3;
a2 ^= (a3 & 0x0C0C0C0C0C0C0C0CuLL) >> 2;
a4 = a3 & a2;
a4 ^= (a4 & 0x0A0A0A0A0A0A0A0AuLL) >> 1;
a4 ^= (((a3 << 1) & a2) ^ ((a2 << 1) & a3)) & 0x0A0A0A0A0A0A0A0AuLL;
a5 = a4 & 0x0808080808080808uLL;
a5 |= a5 >> 1;
a5 ^= (a4 << 1) & 0x0808080808080808uLL;
a4 ^= a5 >> 2;
a4 &= 0x0303030303030303uLL;
a4 ^= (a4 & 0x0202020202020202uLL) >> 1;
a4 |= a4 << 2;
a3 = a2 & a4;
a3 ^= (a3 & 0x0A0A0A0A0A0A0A0AuLL) >> 1;
a3 ^= (((a2 << 1) & a4) ^ ((a4 << 1) & a2)) & 0x0A0A0A0A0A0A0A0AuLL;
a3 |= a3 << 4;
a2 = ((a1 & 0xCCCCCCCCCCCCCCCCuLL) >> 2) | ((a1 & 0x3333333333333333uLL) << 2);
x = a1 & a3;
x ^= (x & 0xAAAAAAAAAAAAAAAAuLL) >> 1;
x ^= (((a1 << 1) & a3) ^ ((a3 << 1) & a1)) & 0xAAAAAAAAAAAAAAAAuLL;
a4 = a2 & a3;
a4 ^= (a4 & 0xAAAAAAAAAAAAAAAAuLL) >> 1;
a4 ^= (((a2 << 1) & a3) ^ ((a3 << 1) & a2)) & 0xAAAAAAAAAAAAAAAAuLL;
a5 = (x & 0xCCCCCCCCCCCCCCCCuLL) >> 2;
x ^= ((a4 << 2) ^ a4) & 0xCCCCCCCCCCCCCCCCuLL;
a4 = a5 & 0x2222222222222222uLL;
a4 |= a4 >> 1;
a4 ^= (a5 << 1) & 0x2222222222222222uLL;
x ^= a4;
y = ((x & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((x & 0x0101010101010101uLL) << 7);
x &= 0xB5B5B5B5B5B5B5B5uLL;
x ^= y & 0x4040404040404040uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x8080808080808080uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x1616161616161616uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0xEBEBEBEBEBEBEBEBuLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x9797979797979797uLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0xFBFBFBFBFBFBFBFBuLL;
y = ((y & 0xFEFEFEFEFEFEFEFEuLL) >> 1) | ((y & 0x0101010101010101uLL) << 7);
x ^= y & 0x7D7D7D7D7D7D7D7DuLL;
*w = x;
}
static void ShiftRows(u64 *state)
{
unsigned char s[4];
unsigned char *s0;
int r;
s0 = (unsigned char *)state;
for (r = 0; r < 4; r++) {
s[0] = s0[0*4 + r];
s[1] = s0[1*4 + r];
s[2] = s0[2*4 + r];
s[3] = s0[3*4 + r];
s0[0*4 + r] = s[(r+0) % 4];
s0[1*4 + r] = s[(r+1) % 4];
s0[2*4 + r] = s[(r+2) % 4];
s0[3*4 + r] = s[(r+3) % 4];
}
}
static void InvShiftRows(u64 *state)
{
unsigned char s[4];
unsigned char *s0;
int r;
s0 = (unsigned char *)state;
for (r = 0; r < 4; r++) {
s[0] = s0[0*4 + r];
s[1] = s0[1*4 + r];
s[2] = s0[2*4 + r];
s[3] = s0[3*4 + r];
s0[0*4 + r] = s[(4-r) % 4];
s0[1*4 + r] = s[(5-r) % 4];
s0[2*4 + r] = s[(6-r) % 4];
s0[3*4 + r] = s[(7-r) % 4];
}
}
static void MixColumns(u64 *state)
{
uni s1;
uni s;
int c;
for (c = 0; c < 2; c++) {
s1.d = state[c];
s.d = s1.d;
s.d ^= ((s.d & 0xFFFF0000FFFF0000uLL) >> 16)
| ((s.d & 0x0000FFFF0000FFFFuLL) << 16);
s.d ^= ((s.d & 0xFF00FF00FF00FF00uLL) >> 8)
| ((s.d & 0x00FF00FF00FF00FFuLL) << 8);
s.d ^= s1.d;
XtimeLong(&s1.d);
s.d ^= s1.d;
s.b[0] ^= s1.b[1];
s.b[1] ^= s1.b[2];
s.b[2] ^= s1.b[3];
s.b[3] ^= s1.b[0];
s.b[4] ^= s1.b[5];
s.b[5] ^= s1.b[6];
s.b[6] ^= s1.b[7];
s.b[7] ^= s1.b[4];
state[c] = s.d;
}
}
static void InvMixColumns(u64 *state)
{
uni s1;
uni s;
int c;
for (c = 0; c < 2; c++) {
s1.d = state[c];
s.d = s1.d;
s.d ^= ((s.d & 0xFFFF0000FFFF0000uLL) >> 16)
| ((s.d & 0x0000FFFF0000FFFFuLL) << 16);
s.d ^= ((s.d & 0xFF00FF00FF00FF00uLL) >> 8)
| ((s.d & 0x00FF00FF00FF00FFuLL) << 8);
s.d ^= s1.d;
XtimeLong(&s1.d);
s.d ^= s1.d;
s.b[0] ^= s1.b[1];
s.b[1] ^= s1.b[2];
s.b[2] ^= s1.b[3];
s.b[3] ^= s1.b[0];
s.b[4] ^= s1.b[5];
s.b[5] ^= s1.b[6];
s.b[6] ^= s1.b[7];
s.b[7] ^= s1.b[4];
XtimeLong(&s1.d);
s1.d ^= ((s1.d & 0xFFFF0000FFFF0000uLL) >> 16)
| ((s1.d & 0x0000FFFF0000FFFFuLL) << 16);
s.d ^= s1.d;
XtimeLong(&s1.d);
s1.d ^= ((s1.d & 0xFF00FF00FF00FF00uLL) >> 8)
| ((s1.d & 0x00FF00FF00FF00FFuLL) << 8);
s.d ^= s1.d;
state[c] = s.d;
}
}
static void AddRoundKey(u64 *state, const u64 *w)
{
state[0] ^= w[0];
state[1] ^= w[1];
}
static void Cipher(const unsigned char *in, unsigned char *out,
const u64 *w, int nr)
{
u64 state[2];
int i;
memcpy(state, in, 16);
AddRoundKey(state, w);
for (i = 1; i < nr; i++) {
SubLong(&state[0]);
SubLong(&state[1]);
ShiftRows(state);
MixColumns(state);
AddRoundKey(state, w + i*2);
}
SubLong(&state[0]);
SubLong(&state[1]);
ShiftRows(state);
AddRoundKey(state, w + nr*2);
memcpy(out, state, 16);
}
static void InvCipher(const unsigned char *in, unsigned char *out,
const u64 *w, int nr)
{
u64 state[2];
int i;
memcpy(state, in, 16);
AddRoundKey(state, w + nr*2);
for (i = nr - 1; i > 0; i--) {
InvShiftRows(state);
InvSubLong(&state[0]);
InvSubLong(&state[1]);
AddRoundKey(state, w + i*2);
InvMixColumns(state);
}
InvShiftRows(state);
InvSubLong(&state[0]);
InvSubLong(&state[1]);
AddRoundKey(state, w);
memcpy(out, state, 16);
}
static void RotWord(u32 *x)
{
unsigned char *w0;
unsigned char tmp;
w0 = (unsigned char *)x;
tmp = w0[0];
w0[0] = w0[1];
w0[1] = w0[2];
w0[2] = w0[3];
w0[3] = tmp;
}
static void KeyExpansion(const unsigned char *key, u64 *w,
int nr, int nk)
{
u32 rcon;
uni prev;
u32 temp;
int i, n;
memcpy(w, key, nk*4);
memcpy(&rcon, "\1\0\0\0", 4);
n = nk/2;
prev.d = w[n-1];
for (i = n; i < (nr+1)*2; i++) {
temp = prev.w[1];
if (i % n == 0) {
RotWord(&temp);
SubWord(&temp);
temp ^= rcon;
XtimeWord(&rcon);
} else if (nk > 6 && i % n == 2) {
SubWord(&temp);
}
prev.d = w[i-n];
prev.w[0] ^= temp;
prev.w[1] ^= prev.w[0];
w[i] = prev.d;
}
}
/**
* Expand the cipher key into the encryption key schedule.
*/
int AES_set_encrypt_key(const unsigned char *userKey, const int bits,
AES_KEY *key)
{
u64 *rk;
if (!userKey || !key)
return -1;
if (bits != 128 && bits != 192 && bits != 256)
return -2;
rk = (u64*)key->rd_key;
if (bits == 128)
key->rounds = 10;
else if (bits == 192)
key->rounds = 12;
else
key->rounds = 14;
KeyExpansion(userKey, rk, key->rounds, bits/32);
return 0;
}
/**
* Expand the cipher key into the decryption key schedule.
*/
int AES_set_decrypt_key(const unsigned char *userKey, const int bits,
AES_KEY *key)
{
return AES_set_encrypt_key(userKey, bits, key);
}
/*
* Encrypt a single block
* in and out can overlap
*/
void AES_encrypt(const unsigned char *in, unsigned char *out,
const AES_KEY *key)
{
const u64 *rk;
assert(in && out && key);
rk = (u64*)key->rd_key;
Cipher(in, out, rk, key->rounds);
}
/*
* Decrypt a single block
* in and out can overlap
*/
void AES_decrypt(const unsigned char *in, unsigned char *out,
const AES_KEY *key)
{
const u64 *rk;
assert(in && out && key);
rk = (u64*)key->rd_key;
InvCipher(in, out, rk, key->rounds);
}
#elif !defined(AES_ASM)
/*-
Te0[x] = S [x].[02, 01, 01, 03];
Te1[x] = S [x].[03, 02, 01, 01];

2
trunk/3rdparty/openssl-1.1-fit/crypto/aes/aes_ecb.c vendored
View file

@ -10,7 +10,7 @@
#include <assert.h>
#include <openssl/aes.h>
#include "aes_locl.h"
#include "aes_local.h"
void AES_ecb_encrypt(const unsigned char *in, unsigned char *out,
const AES_KEY *key, const int enc)

18
trunk/3rdparty/openssl-1.1-fit/crypto/aes/aes_ige.c vendored
View file

@ -1,5 +1,5 @@
/*
* Copyright 2006-2016 The OpenSSL Project Authors. All Rights Reserved.
* Copyright 2006-2020 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
@ -10,12 +10,7 @@
#include "internal/cryptlib.h"
#include <openssl/aes.h>
#include "aes_locl.h"
#define N_WORDS (AES_BLOCK_SIZE / sizeof(unsigned long))
typedef struct {
unsigned long data[N_WORDS];
} aes_block_t;
#include "aes_local.h"
/* XXX: probably some better way to do this */
#if defined(__i386__) || defined(__x86_64__)
@ -24,6 +19,15 @@ typedef struct {
# define UNALIGNED_MEMOPS_ARE_FAST 0
#endif
#define N_WORDS (AES_BLOCK_SIZE / sizeof(unsigned long))
typedef struct {
unsigned long data[N_WORDS];
#if defined(__GNUC__) && UNALIGNED_MEMOPS_ARE_FAST
} aes_block_t __attribute((__aligned__(1)));
#else
} aes_block_t;
#endif
#if UNALIGNED_MEMOPS_ARE_FAST
# define load_block(d, s) (d) = *(const aes_block_t *)(s)
# define store_block(d, s) *(aes_block_t *)(d) = (s)

9
trunk/3rdparty/openssl-1.1-fit/crypto/aes/aes_locl.h → trunk/3rdparty/openssl-1.1-fit/crypto/aes/aes_local.h vendored
View file

@ -1,5 +1,5 @@
/*
* Copyright 2002-2016 The OpenSSL Project Authors. All Rights Reserved.
* Copyright 2002-2020 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
@ -7,8 +7,8 @@
* https://www.openssl.org/source/license.html
*/
#ifndef HEADER_AES_LOCL_H
# define HEADER_AES_LOCL_H
#ifndef OSSL_CRYPTO_AES_LOCAL_H
# define OSSL_CRYPTO_AES_LOCAL_H
# include <openssl/e_os2.h>
# include <stdio.h>
@ -24,6 +24,7 @@
# define PUTU32(ct, st) { (ct)[0] = (u8)((st) >> 24); (ct)[1] = (u8)((st) >> 16); (ct)[2] = (u8)((st) >> 8); (ct)[3] = (u8)(st); }
# endif
typedef unsigned long long u64;
# ifdef AES_LONG
typedef unsigned long u32;
# else
@ -39,4 +40,4 @@ typedef unsigned char u8;
/* This controls loop-unrolling in aes_core.c */
# undef FULL_UNROLL
#endif /* !HEADER_AES_LOCL_H */
#endif /* !OSSL_CRYPTO_AES_LOCAL_H */

2
trunk/3rdparty/openssl-1.1-fit/crypto/aes/aes_misc.c vendored
View file

@ -9,7 +9,7 @@
#include <openssl/opensslv.h>
#include <openssl/aes.h>
#include "aes_locl.h"
#include "aes_local.h"
const char *AES_options(void)
{

2
trunk/3rdparty/openssl-1.1-fit/crypto/aes/aes_x86core.c vendored
View file

@ -46,7 +46,7 @@
#include <stdlib.h>
#include <openssl/aes.h>
#include "aes_locl.h"
#include "aes_local.h"
/*
* These two parameters control which table, 256-byte or 2KB, is

3000
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aes-586.pl vendored
View file

File diff suppressed because it is too large Load diff

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aes-armv4.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2007-2018 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2007-2020 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
@ -1242,4 +1242,4 @@ while(<SELF>) {
close SELF;
print $code;
close STDOUT; # enforce flush
close STDOUT or die "error closing STDOUT: $!"; # enforce flush

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aes-c64xplus.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2012-2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2012-2020 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
@ -1379,4 +1379,4 @@ AES_Td4:
___
print $code;
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aes-mips.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2010-2018 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2010-2020 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
@ -2167,4 +2167,4 @@ foreach (split("\n",$code)) {
print $_,"\n";
}
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aes-parisc.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2009-2018 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2009-2020 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
@ -1035,4 +1035,4 @@ foreach (split("\n",$code)) {
print $_,"\n";
}
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aes-ppc.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2007-2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2007-2020 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
@ -1456,4 +1456,4 @@ ___
$code =~ s/\`([^\`]*)\`/eval $1/gem;
print $code;
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

12
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aes-s390x.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2007-2018 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2007-2020 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
@ -38,14 +38,14 @@
# Implement AES_set_[en|de]crypt_key. Key schedule setup is avoided
# for 128-bit keys, if hardware support is detected.
# Januray 2009.
# January 2009.
#
# Add support for hardware AES192/256 and reschedule instructions to
# minimize/avoid Address Generation Interlock hazard and to favour
# dual-issue z10 pipeline. This gave ~25% improvement on z10 and
# almost 50% on z9. The gain is smaller on z10, because being dual-
# issue z10 makes it impossible to eliminate the interlock condition:
# critial path is not long enough. Yet it spends ~24 cycles per byte
# critical path is not long enough. Yet it spends ~24 cycles per byte
# processed with 128-bit key.
#
# Unlike previous version hardware support detection takes place only
@ -1987,7 +1987,7 @@ $code.=<<___;
.Lxts_enc_done:
stg $sp,$tweak+0($sp) # wipe tweak
stg $sp,$twesk+8($sp)
stg $sp,$tweak+8($sp)
lm${g} %r6,$ra,6*$SIZE_T($sp)
br $ra
.size AES_xts_encrypt,.-AES_xts_encrypt
@ -2267,7 +2267,7 @@ $code.=<<___;
stg $sp,$tweak-16+8($sp)
.Lxts_dec_done:
stg $sp,$tweak+0($sp) # wipe tweak
stg $sp,$twesk+8($sp)
stg $sp,$tweak+8($sp)
lm${g} %r6,$ra,6*$SIZE_T($sp)
br $ra
.size AES_xts_decrypt,.-AES_xts_decrypt
@ -2279,4 +2279,4 @@ ___
$code =~ s/\`([^\`]*)\`/eval $1/gem;
print $code;
close STDOUT; # force flush
close STDOUT or die "error closing STDOUT: $!"; # force flush

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aes-sparcv9.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2005-2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2005-2020 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
@ -1189,4 +1189,4 @@ ___
$code =~ s/fmovs.*$//gm;
print $code;
close STDOUT; # ensure flush
close STDOUT or die "error closing STDOUT: $!"; # ensure flush

2916
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aes-x86_64.pl vendored
View file

File diff suppressed because it is too large Load diff

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aesfx-sparcv9.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2016-2020 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
@ -1267,4 +1267,4 @@ foreach (split("\n",$code)) {
print $_,"\n";
}
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

6
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aesni-mb-x86_64.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2013-2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2013-2020 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
@ -70,7 +70,7 @@ if (!$avx && $win64 && ($flavour =~ /masm/ || $ENV{ASM} =~ /ml64/) &&
$avx = ($1>=10) + ($1>=11);
}
if (!$avx && `$ENV{CC} -v 2>&1` =~ /((?:^clang|LLVM) version|.*based on LLVM) ([3-9]\.[0-9]+)/) {
if (!$avx && `$ENV{CC} -v 2>&1` =~ /((?:clang|LLVM) version|.*based on LLVM) ([0-9]+\.[0-9]+)/) {
$avx = ($2>=3.0) + ($2>3.0);
}
@ -1471,4 +1471,4 @@ $code =~ s/\`([^\`]*)\`/eval($1)/gem;
$code =~ s/\b(aes.*%xmm[0-9]+).*$/aesni($1)/gem;
print $code;
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

12
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aesni-sha1-x86_64.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2011-2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2011-2020 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
@ -108,7 +108,7 @@ $avx=1 if (!$avx && $win64 && ($flavour =~ /nasm/ || $ENV{ASM} =~ /nasm/) &&
$avx=1 if (!$avx && $win64 && ($flavour =~ /masm/ || $ENV{ASM} =~ /ml64/) &&
`ml64 2>&1` =~ /Version ([0-9]+)\./ &&
$1>=10);
$avx=1 if (!$avx && `$ENV{CC} -v 2>&1` =~ /((?:^clang|LLVM) version|.*based on LLVM) ([3-9]\.[0-9]+)/ && $2>=3.0);
$avx=1 if (!$avx && `$ENV{CC} -v 2>&1` =~ /((?:clang|LLVM) version|.*based on LLVM) ([0-9]+\.[0-9]+)/ && $2>=3.0);
$shaext=1; ### set to zero if compiling for 1.0.1
@ -133,6 +133,7 @@ $code.=<<___;
.type aesni_cbc_sha1_enc,\@abi-omnipotent
.align 32
aesni_cbc_sha1_enc:
.cfi_startproc
# caller should check for SSSE3 and AES-NI bits
mov OPENSSL_ia32cap_P+0(%rip),%r10d
mov OPENSSL_ia32cap_P+4(%rip),%r11
@ -151,6 +152,7 @@ ___
$code.=<<___;
jmp aesni_cbc_sha1_enc_ssse3
ret
.cfi_endproc
.size aesni_cbc_sha1_enc,.-aesni_cbc_sha1_enc
___
@ -840,6 +842,7 @@ $code.=<<___;
.type aesni256_cbc_sha1_dec,\@abi-omnipotent
.align 32
aesni256_cbc_sha1_dec:
.cfi_startproc
# caller should check for SSSE3 and AES-NI bits
mov OPENSSL_ia32cap_P+0(%rip),%r10d
mov OPENSSL_ia32cap_P+4(%rip),%r11d
@ -854,6 +857,7 @@ ___
$code.=<<___;
jmp aesni256_cbc_sha1_dec_ssse3
ret
.cfi_endproc
.size aesni256_cbc_sha1_dec,.-aesni256_cbc_sha1_dec
.type aesni256_cbc_sha1_dec_ssse3,\@function,6
@ -1760,6 +1764,7 @@ $code.=<<___;
.type aesni_cbc_sha1_enc_shaext,\@function,6
.align 32
aesni_cbc_sha1_enc_shaext:
.cfi_startproc
mov `($win64?56:8)`(%rsp),$inp # load 7th argument
___
$code.=<<___ if ($win64);
@ -1911,6 +1916,7 @@ $code.=<<___ if ($win64);
___
$code.=<<___;
ret
.cfi_endproc
.size aesni_cbc_sha1_enc_shaext,.-aesni_cbc_sha1_enc_shaext
___
}}}
@ -2137,4 +2143,4 @@ foreach (split("\n",$code)) {
print $_,"\n";
}
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

66
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aesni-sha256-x86_64.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2013-2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2013-2020 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
@ -70,7 +70,7 @@ if (!$avx && $win64 && ($flavour =~ /masm/ || $ENV{ASM} =~ /ml64/) &&
$avx = ($1>=10) + ($1>=12);
}
if (!$avx && `$ENV{CC} -v 2>&1` =~ /((?:^clang|LLVM) version|.*based on LLVM) ([3-9]\.[0-9]+)/) {
if (!$avx && `$ENV{CC} -v 2>&1` =~ /((?:clang|LLVM) version|.*based on LLVM) ([0-9]+\.[0-9]+)/) {
$avx = ($2>=3.0) + ($2>3.0);
}
@ -123,6 +123,7 @@ $code=<<___;
.type $func,\@abi-omnipotent
.align 16
$func:
.cfi_startproc
___
if ($avx) {
$code.=<<___;
@ -162,6 +163,7 @@ $code.=<<___;
ud2
.Lprobe:
ret
.cfi_endproc
.size $func,.-$func
.align 64
@ -1084,7 +1086,23 @@ $code.=<<___;
vmovdqa $t0,0x00(%rsp)
xor $a1,$a1
vmovdqa $t1,0x20(%rsp)
___
$code.=<<___ if (!$win64);
# temporarily use %rsi as frame pointer
mov $_rsp,%rsi
.cfi_def_cfa %rsi,8
___
$code.=<<___;
lea -$PUSH8(%rsp),%rsp
___
$code.=<<___ if (!$win64);
# the frame info is at $_rsp, but the stack is moving...
# so a second frame pointer is saved at -8(%rsp)
# that is in the red zone
mov %rsi,-8(%rsp)
.cfi_cfa_expression %rsp-8,deref,+8
___
$code.=<<___;
mov $B,$a3
vmovdqa $t2,0x00(%rsp)
xor $C,$a3 # magic
@ -1106,7 +1124,17 @@ my @X = @_;
my @insns = (&$body,&$body,&$body,&$body); # 96 instructions
my $base = "+2*$PUSH8(%rsp)";
&lea ("%rsp","-$PUSH8(%rsp)") if (($j%2)==0);
if (($j%2)==0) {
&lea ("%rsp","-$PUSH8(%rsp)");
$code.=<<___ if (!$win64);
.cfi_cfa_expression %rsp+`$PUSH8-8`,deref,+8
# copy secondary frame pointer to new location again at -8(%rsp)
pushq $PUSH8-8(%rsp)
.cfi_cfa_expression %rsp,deref,+8
lea 8(%rsp),%rsp
.cfi_cfa_expression %rsp-8,deref,+8
___
}
foreach (Xupdate_256_AVX()) { # 29 instructions
eval;
eval(shift(@insns));
@ -1232,26 +1260,28 @@ $code.=<<___;
jbe .Loop_avx2
lea (%rsp),$Tbl
# temporarily use $Tbl as index to $_rsp
# this avoids the need to save a secondary frame pointer at -8(%rsp)
.cfi_cfa_expression $Tbl+`16*$SZ+7*8`,deref,+8
.Ldone_avx2:
lea ($Tbl),%rsp
mov $_ivp,$ivp
mov $_rsp,%rsi
mov 16*$SZ+4*8($Tbl),$ivp
mov 16*$SZ+7*8($Tbl),%rsi
.cfi_def_cfa %rsi,8
vmovdqu $iv,($ivp) # output IV
vzeroall
___
$code.=<<___ if ($win64);
movaps `$framesz+16*0`(%rsp),%xmm6
movaps `$framesz+16*1`(%rsp),%xmm7
movaps `$framesz+16*2`(%rsp),%xmm8
movaps `$framesz+16*3`(%rsp),%xmm9
movaps `$framesz+16*4`(%rsp),%xmm10
movaps `$framesz+16*5`(%rsp),%xmm11
movaps `$framesz+16*6`(%rsp),%xmm12
movaps `$framesz+16*7`(%rsp),%xmm13
movaps `$framesz+16*8`(%rsp),%xmm14
movaps `$framesz+16*9`(%rsp),%xmm15
movaps `$framesz+16*0`($Tbl),%xmm6
movaps `$framesz+16*1`($Tbl),%xmm7
movaps `$framesz+16*2`($Tbl),%xmm8
movaps `$framesz+16*3`($Tbl),%xmm9
movaps `$framesz+16*4`($Tbl),%xmm10
movaps `$framesz+16*5`($Tbl),%xmm11
movaps `$framesz+16*6`($Tbl),%xmm12
movaps `$framesz+16*7`($Tbl),%xmm13
movaps `$framesz+16*8`($Tbl),%xmm14
movaps `$framesz+16*9`($Tbl),%xmm15
___
$code.=<<___;
mov -48(%rsi),%r15
@ -1339,6 +1369,7 @@ $code.=<<___;
.type ${func}_shaext,\@function,6
.align 32
${func}_shaext:
.cfi_startproc
mov `($win64?56:8)`(%rsp),$inp # load 7th argument
___
$code.=<<___ if ($win64);
@ -1555,6 +1586,7 @@ $code.=<<___ if ($win64);
___
$code.=<<___;
ret
.cfi_endproc
.size ${func}_shaext,.-${func}_shaext
___
}
@ -1767,4 +1799,4 @@ sub rex {
$code =~ s/\`([^\`]*)\`/eval $1/gem;
$code =~ s/\b(sha256[^\s]*)\s+(.*)/sha256op38($1,$2)/gem;
print $code;
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aesni-x86.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2009-2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2009-2020 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
@ -3412,4 +3412,4 @@ my ($l_,$block,$i1,$i3,$i5) = ($rounds_,$key_,$rounds,$len,$out);
&asm_finish();
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

30
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aesni-x86_64.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2009-2019 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2009-2020 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
@ -59,7 +59,7 @@
# nothing one can do and the result appears optimal. CCM result is
# identical to CBC, because CBC-MAC is essentially CBC encrypt without
# saving output. CCM CTR "stays invisible," because it's neatly
# interleaved wih CBC-MAC. This provides ~30% improvement over
# interleaved with CBC-MAC. This provides ~30% improvement over
# "straightforward" CCM implementation with CTR and CBC-MAC performed
# disjointly. Parallelizable modes practically achieve the theoretical
# limit.
@ -984,6 +984,7 @@ $code.=<<___;
.type aesni_ccm64_encrypt_blocks,\@function,6
.align 16
aesni_ccm64_encrypt_blocks:
.cfi_startproc
___
$code.=<<___ if ($win64);
lea -0x58(%rsp),%rsp
@ -1066,6 +1067,7 @@ $code.=<<___ if ($win64);
___
$code.=<<___;
ret
.cfi_endproc
.size aesni_ccm64_encrypt_blocks,.-aesni_ccm64_encrypt_blocks
___
######################################################################
@ -1074,6 +1076,7 @@ $code.=<<___;
.type aesni_ccm64_decrypt_blocks,\@function,6
.align 16
aesni_ccm64_decrypt_blocks:
.cfi_startproc
___
$code.=<<___ if ($win64);
lea -0x58(%rsp),%rsp
@ -1173,6 +1176,7 @@ $code.=<<___ if ($win64);
___
$code.=<<___;
ret
.cfi_endproc
.size aesni_ccm64_decrypt_blocks,.-aesni_ccm64_decrypt_blocks
___
}
@ -2339,7 +2343,7 @@ $code.=<<___;
movdqu `16*0`($inp),$inout0 # load input
movdqa $rndkey0,$twmask
movdqu `16*1`($inp),$inout1
pxor @tweak[0],$inout0 # intput^=tweak^round[0]
pxor @tweak[0],$inout0 # input^=tweak^round[0]
movdqu `16*2`($inp),$inout2
pxor @tweak[1],$inout1
aesdec $rndkey1,$inout0
@ -3031,6 +3035,7 @@ $code.=<<___;
.type __ocb_encrypt6,\@abi-omnipotent
.align 32
__ocb_encrypt6:
.cfi_startproc
pxor $rndkey0l,@offset[5] # offset_i ^ round[0]
movdqu ($L_p,$i1),@offset[1]
movdqa @offset[0],@offset[2]
@ -3128,11 +3133,13 @@ __ocb_encrypt6:
aesenclast @offset[4],$inout4
aesenclast @offset[5],$inout5
ret
.cfi_endproc
.size __ocb_encrypt6,.-__ocb_encrypt6
.type __ocb_encrypt4,\@abi-omnipotent
.align 32
__ocb_encrypt4:
.cfi_startproc
pxor $rndkey0l,@offset[5] # offset_i ^ round[0]
movdqu ($L_p,$i1),@offset[1]
movdqa @offset[0],@offset[2]
@ -3197,11 +3204,13 @@ __ocb_encrypt4:
aesenclast @offset[2],$inout2
aesenclast @offset[3],$inout3
ret
.cfi_endproc
.size __ocb_encrypt4,.-__ocb_encrypt4
.type __ocb_encrypt1,\@abi-omnipotent
.align 32
__ocb_encrypt1:
.cfi_startproc
pxor @offset[5],$inout5 # offset_i
pxor $rndkey0l,$inout5 # offset_i ^ round[0]
pxor $inout0,$checksum # accumulate checksum
@ -3232,6 +3241,7 @@ __ocb_encrypt1:
aesenclast $inout5,$inout0
ret
.cfi_endproc
.size __ocb_encrypt1,.-__ocb_encrypt1
.globl aesni_ocb_decrypt
@ -3513,6 +3523,7 @@ $code.=<<___;
.type __ocb_decrypt6,\@abi-omnipotent
.align 32
__ocb_decrypt6:
.cfi_startproc
pxor $rndkey0l,@offset[5] # offset_i ^ round[0]
movdqu ($L_p,$i1),@offset[1]
movdqa @offset[0],@offset[2]
@ -3604,11 +3615,13 @@ __ocb_decrypt6:
aesdeclast @offset[4],$inout4
aesdeclast @offset[5],$inout5
ret
.cfi_endproc
.size __ocb_decrypt6,.-__ocb_decrypt6
.type __ocb_decrypt4,\@abi-omnipotent
.align 32
__ocb_decrypt4:
.cfi_startproc
pxor $rndkey0l,@offset[5] # offset_i ^ round[0]
movdqu ($L_p,$i1),@offset[1]
movdqa @offset[0],@offset[2]
@ -3669,11 +3682,13 @@ __ocb_decrypt4:
aesdeclast @offset[2],$inout2
aesdeclast @offset[3],$inout3
ret
.cfi_endproc
.size __ocb_decrypt4,.-__ocb_decrypt4
.type __ocb_decrypt1,\@abi-omnipotent
.align 32
__ocb_decrypt1:
.cfi_startproc
pxor @offset[5],$inout5 # offset_i
pxor $rndkey0l,$inout5 # offset_i ^ round[0]
pxor $inout5,$inout0 # input ^ round[0] ^ offset_i
@ -3703,6 +3718,7 @@ __ocb_decrypt1:
aesdeclast $inout5,$inout0
ret
.cfi_endproc
.size __ocb_decrypt1,.-__ocb_decrypt1
___
} }}
@ -4637,7 +4653,6 @@ __aesni_set_encrypt_key:
add \$8,%rsp
.cfi_adjust_cfa_offset -8
ret
.cfi_endproc
.LSEH_end_set_encrypt_key:
.align 16
@ -4708,6 +4723,7 @@ __aesni_set_encrypt_key:
shufps \$0b10101010,%xmm1,%xmm1 # critical path
xorps %xmm1,%xmm2
ret
.cfi_endproc
.size ${PREFIX}_set_encrypt_key,.-${PREFIX}_set_encrypt_key
.size __aesni_set_encrypt_key,.-__aesni_set_encrypt_key
___
@ -4812,7 +4828,7 @@ ctr_xts_se_handler:
mov 56($disp),%r11 # disp->HandlerData
mov 0(%r11),%r10d # HandlerData[0]
lea (%rsi,%r10),%r10 # prologue lable
lea (%rsi,%r10),%r10 # prologue label
cmp %r10,%rbx # context->Rip<prologue label
jb .Lcommon_seh_tail
@ -4856,7 +4872,7 @@ ocb_se_handler:
mov 56($disp),%r11 # disp->HandlerData
mov 0(%r11),%r10d # HandlerData[0]
lea (%rsi,%r10),%r10 # prologue lable
lea (%rsi,%r10),%r10 # prologue label
cmp %r10,%rbx # context->Rip<prologue label
jb .Lcommon_seh_tail
@ -5138,4 +5154,4 @@ $code =~ s/\bmovbe\s+%eax,\s*([0-9]+)\(%rsp\)/movbe($1)/gem;
print $code;
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

8
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aesp8-ppc.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2014-2018 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2014-2020 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
@ -1829,7 +1829,7 @@ Lctr32_enc8x_three:
stvx_u $out1,$x10,$out
stvx_u $out2,$x20,$out
addi $out,$out,0x30
b Lcbc_dec8x_done
b Lctr32_enc8x_done
.align 5
Lctr32_enc8x_two:
@ -1841,7 +1841,7 @@ Lctr32_enc8x_two:
stvx_u $out0,$x00,$out
stvx_u $out1,$x10,$out
addi $out,$out,0x20
b Lcbc_dec8x_done
b Lctr32_enc8x_done
.align 5
Lctr32_enc8x_one:
@ -3804,4 +3804,4 @@ foreach(split("\n",$code)) {
print $_,"\n";
}
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aest4-sparcv9.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2012-2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2012-2020 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
@ -926,4 +926,4 @@ ___
&emit_assembler();
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

42
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/aesv8-armx.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2014-2019 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2014-2020 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
@ -183,7 +183,12 @@ $code.=<<___;
.Loop192:
vtbl.8 $key,{$in1},$mask
vext.8 $tmp,$zero,$in0,#12
#ifdef __ARMEB__
vst1.32 {$in1},[$out],#16
sub $out,$out,#8
#else
vst1.32 {$in1},[$out],#8
#endif
aese $key,$zero
subs $bits,$bits,#1
@ -715,8 +720,11 @@ $code.=<<___;
ldr $rounds,[$key,#240]
ldr $ctr, [$ivp, #12]
#ifdef __ARMEB__
vld1.8 {$dat0},[$ivp]
#else
vld1.32 {$dat0},[$ivp]
#endif
vld1.32 {q8-q9},[$key] // load key schedule...
sub $rounds,$rounds,#4
mov $step,#16
@ -732,17 +740,17 @@ $code.=<<___;
#ifndef __ARMEB__
rev $ctr, $ctr
#endif
vorr $dat1,$dat0,$dat0
add $tctr1, $ctr, #1
vorr $dat2,$dat0,$dat0
add $ctr, $ctr, #2
vorr $ivec,$dat0,$dat0
rev $tctr1, $tctr1
vmov.32 ${dat1}[3],$tctr1
vmov.32 ${ivec}[3],$tctr1
add $ctr, $ctr, #2
vorr $dat1,$ivec,$ivec
b.ls .Lctr32_tail
rev $tctr2, $ctr
vmov.32 ${ivec}[3],$tctr2
sub $len,$len,#3 // bias
vmov.32 ${dat2}[3],$tctr2
vorr $dat2,$ivec,$ivec
b .Loop3x_ctr32
.align 4
@ -769,11 +777,11 @@ $code.=<<___;
aese $dat1,q8
aesmc $tmp1,$dat1
vld1.8 {$in0},[$inp],#16
vorr $dat0,$ivec,$ivec
add $tctr0,$ctr,#1
aese $dat2,q8
aesmc $dat2,$dat2
vld1.8 {$in1},[$inp],#16
vorr $dat1,$ivec,$ivec
rev $tctr0,$tctr0
aese $tmp0,q9
aesmc $tmp0,$tmp0
aese $tmp1,q9
@ -782,8 +790,6 @@ $code.=<<___;
mov $key_,$key
aese $dat2,q9
aesmc $tmp2,$dat2
vorr $dat2,$ivec,$ivec
add $tctr0,$ctr,#1
aese $tmp0,q12
aesmc $tmp0,$tmp0
aese $tmp1,q12
@ -799,20 +805,22 @@ $code.=<<___;
aese $tmp1,q13
aesmc $tmp1,$tmp1
veor $in2,$in2,$rndlast
rev $tctr0,$tctr0
vmov.32 ${ivec}[3], $tctr0
aese $tmp2,q13
aesmc $tmp2,$tmp2
vmov.32 ${dat0}[3], $tctr0
vorr $dat0,$ivec,$ivec
rev $tctr1,$tctr1
aese $tmp0,q14
aesmc $tmp0,$tmp0
vmov.32 ${ivec}[3], $tctr1
rev $tctr2,$ctr
aese $tmp1,q14
aesmc $tmp1,$tmp1
vmov.32 ${dat1}[3], $tctr1
rev $tctr2,$ctr
vorr $dat1,$ivec,$ivec
vmov.32 ${ivec}[3], $tctr2
aese $tmp2,q14
aesmc $tmp2,$tmp2
vmov.32 ${dat2}[3], $tctr2
vorr $dat2,$ivec,$ivec
subs $len,$len,#3
aese $tmp0,q15
aese $tmp1,q15
@ -1008,4 +1016,4 @@ if ($flavour =~ /64/) { ######## 64-bit code
}
}
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/bsaes-armv7.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2012-2018 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2012-2020 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
@ -2488,4 +2488,4 @@ close SELF;
print $code;
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

3239
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/bsaes-x86_64.pl vendored
View file

File diff suppressed because it is too large Load diff

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/vpaes-armv8.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2015-2019 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2015-2020 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
@ -1274,4 +1274,4 @@ ___
} }
print $code;
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/vpaes-ppc.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2013-2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2013-2020 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
@ -1591,4 +1591,4 @@ foreach (split("\n",$code)) {
print $_,"\n";
}
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/vpaes-x86.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2011-2016 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2011-2020 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
@ -913,4 +913,4 @@ $k_dsbo=0x2c0; # decryption sbox final output
&asm_finish();
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";

4
trunk/3rdparty/openssl-1.1-fit/crypto/aes/asm/vpaes-x86_64.pl vendored
View file

@ -1,5 +1,5 @@
#! /usr/bin/env perl
# Copyright 2011-2019 The OpenSSL Project Authors. All Rights Reserved.
# Copyright 2011-2020 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
@ -1238,4 +1238,4 @@ $code =~ s/\`([^\`]*)\`/eval($1)/gem;
print $code;
close STDOUT;
close STDOUT or die "error closing STDOUT: $!";