mirror of
https://github.com/wolfSSL/wolfssl.git
synced 2026-01-28 20:59:57 +01:00
fa98477f22
For byte order reverse and rotation we have builtins in the CC-RX compiler. Especially when rotating registers with fixed amounts this can be compiled efficiently into opcodes with embedded values for shift (without needing other registers).
525 lines
14 KiB
C
525 lines
14 KiB
C
/* misc.c
|
|
*
|
|
* Copyright (C) 2006-2021 wolfSSL Inc.
|
|
*
|
|
* This file is part of wolfSSL.
|
|
*
|
|
* wolfSSL is free software; you can redistribute it and/or modify
|
|
* it under the terms of the GNU General Public License as published by
|
|
* the Free Software Foundation; either version 2 of the License, or
|
|
* (at your option) any later version.
|
|
*
|
|
* wolfSSL is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
* GNU General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU General Public License
|
|
* along with this program; if not, write to the Free Software
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1335, USA
|
|
*/
|
|
/*
|
|
|
|
DESCRIPTION
|
|
This module implements the arithmetic-shift right, left, byte swapping, XOR,
|
|
masking and clearing memory logic.
|
|
|
|
*/
|
|
#ifdef HAVE_CONFIG_H
|
|
#include <config.h>
|
|
#endif
|
|
|
|
#include <wolfssl/wolfcrypt/settings.h>
|
|
|
|
#ifndef WOLF_CRYPT_MISC_C
|
|
#define WOLF_CRYPT_MISC_C
|
|
|
|
#include <wolfssl/wolfcrypt/misc.h>
|
|
|
|
/* inlining these functions is a huge speed increase and a small size decrease,
|
|
because the functions are smaller than function call setup/cleanup, e.g.,
|
|
md5 benchmark is twice as fast with inline. If you don't want it, then
|
|
define NO_INLINE and compile this file into wolfssl, otherwise it's used as
|
|
a source header
|
|
*/
|
|
|
|
#ifdef NO_INLINE
|
|
#define WC_STATIC
|
|
#else
|
|
#define WC_STATIC static
|
|
#endif
|
|
|
|
/* Check for if compiling misc.c when not needed. */
|
|
#if !defined(WOLFSSL_MISC_INCLUDED) && !defined(NO_INLINE)
|
|
#ifndef WOLFSSL_IGNORE_FILE_WARN
|
|
#warning misc.c does not need to be compiled when using inline (NO_INLINE not defined)
|
|
#endif
|
|
|
|
#else
|
|
|
|
|
|
#if defined(__ICCARM__)
|
|
#include <intrinsics.h>
|
|
#endif
|
|
|
|
|
|
#ifdef INTEL_INTRINSICS
|
|
|
|
#include <stdlib.h> /* get intrinsic definitions */
|
|
|
|
/* for non visual studio probably need no long version, 32 bit only
|
|
* i.e., _rotl and _rotr */
|
|
#pragma intrinsic(_lrotl, _lrotr)
|
|
|
|
WC_STATIC WC_INLINE word32 rotlFixed(word32 x, word32 y)
|
|
{
|
|
return y ? _lrotl(x, y) : x;
|
|
}
|
|
|
|
WC_STATIC WC_INLINE word32 rotrFixed(word32 x, word32 y)
|
|
{
|
|
return y ? _lrotr(x, y) : x;
|
|
}
|
|
|
|
#elif defined(__CCRX__)
|
|
|
|
#include <builtin.h> /* get intrinsic definitions */
|
|
|
|
#if !defined(NO_INLINE)
|
|
|
|
#define rotlFixed(x, y) _builtin_rotl(x, y)
|
|
|
|
#define rotrFixed(x, y) _builtin_rotr(x, y)
|
|
|
|
#else /* create real function */
|
|
|
|
WC_STATIC WC_INLINE word32 rotlFixed(word32 x, word32 y)
|
|
{
|
|
return _builtin_rotl(x, y);
|
|
}
|
|
|
|
WC_STATIC WC_INLINE word32 rotrFixed(word32 x, word32 y)
|
|
{
|
|
return _builtin_rotr(x, y);
|
|
}
|
|
|
|
#endif
|
|
|
|
#else /* generic */
|
|
/* This routine performs a left circular arithmetic shift of <x> by <y> value. */
|
|
|
|
WC_STATIC WC_INLINE word32 rotlFixed(word32 x, word32 y)
|
|
{
|
|
return (x << y) | (x >> (sizeof(y) * 8 - y));
|
|
}
|
|
|
|
/* This routine performs a right circular arithmetic shift of <x> by <y> value. */
|
|
WC_STATIC WC_INLINE word32 rotrFixed(word32 x, word32 y)
|
|
{
|
|
return (x >> y) | (x << (sizeof(y) * 8 - y));
|
|
}
|
|
|
|
#endif
|
|
|
|
#ifdef WC_RC2
|
|
|
|
/* This routine performs a left circular arithmetic shift of <x> by <y> value */
|
|
WC_STATIC WC_INLINE word16 rotlFixed16(word16 x, word16 y)
|
|
{
|
|
return (x << y) | (x >> (sizeof(y) * 8 - y));
|
|
}
|
|
|
|
|
|
/* This routine performs a right circular arithmetic shift of <x> by <y> value */
|
|
WC_STATIC WC_INLINE word16 rotrFixed16(word16 x, word16 y)
|
|
{
|
|
return (x >> y) | (x << (sizeof(y) * 8 - y));
|
|
}
|
|
|
|
#endif /* WC_RC2 */
|
|
|
|
/* This routine performs a byte swap of 32-bit word value. */
|
|
#if defined(__CCRX__) && !defined(NO_INLINE) // shortest version for CC-RX
|
|
#define ByteReverseWord32(value) _builtin_revl(value)
|
|
#else
|
|
WC_STATIC WC_INLINE word32 ByteReverseWord32(word32 value)
|
|
{
|
|
#ifdef PPC_INTRINSICS
|
|
/* PPC: load reverse indexed instruction */
|
|
return (word32)__lwbrx(&value,0);
|
|
#elif defined(__ICCARM__)
|
|
return (word32)__REV(value);
|
|
#elif defined(KEIL_INTRINSICS)
|
|
return (word32)__rev(value);
|
|
#elif defined(__CCRX__)
|
|
return (word32)_builtin_revl(value);
|
|
#elif defined(WOLF_ALLOW_BUILTIN) && \
|
|
defined(__GNUC_PREREQ) && __GNUC_PREREQ(4, 3)
|
|
return (word32)__builtin_bswap32(value);
|
|
#elif defined(WOLFSSL_BYTESWAP32_ASM) && defined(__GNUC__) && \
|
|
defined(__aarch64__)
|
|
__asm__ volatile (
|
|
"REV32 %0, %0 \n"
|
|
: "+r" (value)
|
|
:
|
|
);
|
|
return value;
|
|
#elif defined(WOLFSSL_BYTESWAP32_ASM) && defined(__GNUC__) && \
|
|
(defined(__thumb__) || defined(__arm__))
|
|
__asm__ volatile (
|
|
"REV %0, %0 \n"
|
|
: "+r" (value)
|
|
:
|
|
);
|
|
return value;
|
|
#elif defined(FAST_ROTATE)
|
|
/* 5 instructions with rotate instruction, 9 without */
|
|
return (rotrFixed(value, 8U) & 0xff00ff00) |
|
|
(rotlFixed(value, 8U) & 0x00ff00ff);
|
|
#else
|
|
/* 6 instructions with rotate instruction, 8 without */
|
|
value = ((value & 0xFF00FF00) >> 8) | ((value & 0x00FF00FF) << 8);
|
|
return rotlFixed(value, 16U);
|
|
#endif
|
|
}
|
|
#endif /* __CCRX__ */
|
|
/* This routine performs a byte swap of words array of a given count. */
|
|
WC_STATIC WC_INLINE void ByteReverseWords(word32* out, const word32* in,
|
|
word32 byteCount)
|
|
{
|
|
word32 count = byteCount/(word32)sizeof(word32), i;
|
|
|
|
for (i = 0; i < count; i++)
|
|
out[i] = ByteReverseWord32(in[i]);
|
|
|
|
}
|
|
|
|
#if defined(WORD64_AVAILABLE) && !defined(WOLFSSL_NO_WORD64_OPS)
|
|
|
|
|
|
WC_STATIC WC_INLINE word64 rotlFixed64(word64 x, word64 y)
|
|
{
|
|
return (x << y) | (x >> (sizeof(y) * 8 - y));
|
|
}
|
|
|
|
|
|
WC_STATIC WC_INLINE word64 rotrFixed64(word64 x, word64 y)
|
|
{
|
|
return (x >> y) | (x << (sizeof(y) * 8 - y));
|
|
}
|
|
|
|
|
|
WC_STATIC WC_INLINE word64 ByteReverseWord64(word64 value)
|
|
{
|
|
#if defined(WOLF_ALLOW_BUILTIN) && defined(__GNUC_PREREQ) && __GNUC_PREREQ(4, 3)
|
|
return (word64)__builtin_bswap64(value);
|
|
#elif defined(WOLFCRYPT_SLOW_WORD64)
|
|
return (word64)((word64)ByteReverseWord32((word32) value)) << 32 |
|
|
(word64)ByteReverseWord32((word32)(value >> 32));
|
|
#else
|
|
value = ((value & W64LIT(0xFF00FF00FF00FF00)) >> 8) |
|
|
((value & W64LIT(0x00FF00FF00FF00FF)) << 8);
|
|
value = ((value & W64LIT(0xFFFF0000FFFF0000)) >> 16) |
|
|
((value & W64LIT(0x0000FFFF0000FFFF)) << 16);
|
|
return rotlFixed64(value, 32U);
|
|
#endif
|
|
}
|
|
|
|
|
|
WC_STATIC WC_INLINE void ByteReverseWords64(word64* out, const word64* in,
|
|
word32 byteCount)
|
|
{
|
|
word32 count = byteCount/(word32)sizeof(word64), i;
|
|
|
|
for (i = 0; i < count; i++)
|
|
out[i] = ByteReverseWord64(in[i]);
|
|
|
|
}
|
|
|
|
#endif /* WORD64_AVAILABLE && !WOLFSSL_NO_WORD64_OPS */
|
|
|
|
#ifndef WOLFSSL_NO_XOR_OPS
|
|
/* This routine performs a bitwise XOR operation of <*r> and <*a> for <n> number
|
|
of wolfssl_words, placing the result in <*r>. */
|
|
WC_STATIC WC_INLINE void XorWordsOut(wolfssl_word* r, const wolfssl_word* a,
|
|
const wolfssl_word* b, word32 n)
|
|
{
|
|
word32 i;
|
|
|
|
for (i = 0; i < n; i++) r[i] = a[i] ^ b[i];
|
|
}
|
|
|
|
/* This routine performs a bitwise XOR operation of <*buf> and <*mask> of n
|
|
counts, placing the result in <*buf>. */
|
|
|
|
WC_STATIC WC_INLINE void xorbufout(void*out, const void* buf, const void* mask,
|
|
word32 count)
|
|
{
|
|
if (((wolfssl_word)out | (wolfssl_word)buf | (wolfssl_word)mask | count) % \
|
|
WOLFSSL_WORD_SIZE == 0)
|
|
XorWordsOut( (wolfssl_word*)out, (wolfssl_word*)buf,
|
|
(const wolfssl_word*)mask, count / WOLFSSL_WORD_SIZE);
|
|
else {
|
|
word32 i;
|
|
byte* o = (byte*)out;
|
|
byte* b = (byte*)buf;
|
|
const byte* m = (const byte*)mask;
|
|
|
|
for (i = 0; i < count; i++) o[i] = b[i] ^ m[i];
|
|
}
|
|
}
|
|
|
|
/* This routine performs a bitwise XOR operation of <*r> and <*a> for <n> number
|
|
of wolfssl_words, placing the result in <*r>. */
|
|
WC_STATIC WC_INLINE void XorWords(wolfssl_word* r, const wolfssl_word* a, word32 n)
|
|
{
|
|
word32 i;
|
|
|
|
for (i = 0; i < n; i++) r[i] ^= a[i];
|
|
}
|
|
|
|
/* This routine performs a bitwise XOR operation of <*buf> and <*mask> of n
|
|
counts, placing the result in <*buf>. */
|
|
|
|
WC_STATIC WC_INLINE void xorbuf(void* buf, const void* mask, word32 count)
|
|
{
|
|
if (((wolfssl_word)buf | (wolfssl_word)mask | count) % WOLFSSL_WORD_SIZE == 0)
|
|
XorWords( (wolfssl_word*)buf,
|
|
(const wolfssl_word*)mask, count / WOLFSSL_WORD_SIZE);
|
|
else {
|
|
word32 i;
|
|
byte* b = (byte*)buf;
|
|
const byte* m = (const byte*)mask;
|
|
|
|
for (i = 0; i < count; i++) b[i] ^= m[i];
|
|
}
|
|
}
|
|
#endif
|
|
|
|
#ifndef WOLFSSL_NO_FORCE_ZERO
|
|
/* This routine fills the first len bytes of the memory area pointed by mem
|
|
with zeros. It ensures compiler optimizations doesn't skip it */
|
|
WC_STATIC WC_INLINE void ForceZero(const void* mem, word32 len)
|
|
{
|
|
volatile byte* z = (volatile byte*)mem;
|
|
|
|
#if (defined(WOLFSSL_X86_64_BUILD) || defined(WOLFSSL_AARCH64_BUILD)) \
|
|
&& defined(WORD64_AVAILABLE)
|
|
volatile word64* w;
|
|
#ifndef WOLFSSL_UNALIGNED_64BIT_ACCESS
|
|
word32 l = (sizeof(word64) - ((size_t)z & (sizeof(word64)-1))) &
|
|
(sizeof(word64)-1);
|
|
|
|
if (len < l) l = len;
|
|
len -= l;
|
|
while (l--) *z++ = 0;
|
|
#endif
|
|
for (w = (volatile word64*)z; len >= sizeof(*w); len -= sizeof(*w))
|
|
*w++ = 0;
|
|
z = (volatile byte*)w;
|
|
#endif
|
|
|
|
while (len--) *z++ = 0;
|
|
}
|
|
#endif
|
|
|
|
|
|
#ifndef WOLFSSL_NO_CONST_CMP
|
|
/* check all length bytes for equality, return 0 on success */
|
|
WC_STATIC WC_INLINE int ConstantCompare(const byte* a, const byte* b, int length)
|
|
{
|
|
int i;
|
|
int compareSum = 0;
|
|
|
|
for (i = 0; i < length; i++) {
|
|
compareSum |= a[i] ^ b[i];
|
|
}
|
|
|
|
return compareSum;
|
|
}
|
|
#endif
|
|
|
|
|
|
#ifndef WOLFSSL_HAVE_MIN
|
|
#define WOLFSSL_HAVE_MIN
|
|
#if defined(HAVE_FIPS) && !defined(min) /* so ifdef check passes */
|
|
#define min min
|
|
#endif
|
|
/* returns the smaller of a and b */
|
|
WC_STATIC WC_INLINE word32 min(word32 a, word32 b)
|
|
{
|
|
return a > b ? b : a;
|
|
}
|
|
#endif /* !WOLFSSL_HAVE_MIN */
|
|
|
|
#ifndef WOLFSSL_HAVE_MAX
|
|
#define WOLFSSL_HAVE_MAX
|
|
#if defined(HAVE_FIPS) && !defined(max) /* so ifdef check passes */
|
|
#define max max
|
|
#endif
|
|
WC_STATIC WC_INLINE word32 max(word32 a, word32 b)
|
|
{
|
|
return a > b ? a : b;
|
|
}
|
|
#endif /* !WOLFSSL_HAVE_MAX */
|
|
|
|
#ifndef WOLFSSL_NO_INT_ENCODE
|
|
/* converts a 32 bit integer to 24 bit */
|
|
WC_STATIC WC_INLINE void c32to24(word32 in, word24 out)
|
|
{
|
|
out[0] = (in >> 16) & 0xff;
|
|
out[1] = (in >> 8) & 0xff;
|
|
out[2] = in & 0xff;
|
|
}
|
|
|
|
/* convert 16 bit integer to opaque */
|
|
WC_STATIC WC_INLINE void c16toa(word16 wc_u16, byte* c)
|
|
{
|
|
c[0] = (wc_u16 >> 8) & 0xff;
|
|
c[1] = wc_u16 & 0xff;
|
|
}
|
|
|
|
/* convert 32 bit integer to opaque */
|
|
WC_STATIC WC_INLINE void c32toa(word32 wc_u32, byte* c)
|
|
{
|
|
c[0] = (wc_u32 >> 24) & 0xff;
|
|
c[1] = (wc_u32 >> 16) & 0xff;
|
|
c[2] = (wc_u32 >> 8) & 0xff;
|
|
c[3] = wc_u32 & 0xff;
|
|
}
|
|
#endif
|
|
|
|
#ifndef WOLFSSL_NO_INT_DECODE
|
|
/* convert a 24 bit integer into a 32 bit one */
|
|
WC_STATIC WC_INLINE void c24to32(const word24 wc_u24, word32* wc_u32)
|
|
{
|
|
*wc_u32 = ((word32)wc_u24[0] << 16) | (wc_u24[1] << 8) | wc_u24[2];
|
|
}
|
|
|
|
|
|
/* convert opaque to 24 bit integer */
|
|
WC_STATIC WC_INLINE void ato24(const byte* c, word32* wc_u24)
|
|
{
|
|
*wc_u24 = ((word32)c[0] << 16) | (c[1] << 8) | c[2];
|
|
}
|
|
|
|
/* convert opaque to 16 bit integer */
|
|
WC_STATIC WC_INLINE void ato16(const byte* c, word16* wc_u16)
|
|
{
|
|
*wc_u16 = (word16) ((c[0] << 8) | (c[1]));
|
|
}
|
|
|
|
/* convert opaque to 32 bit integer */
|
|
WC_STATIC WC_INLINE void ato32(const byte* c, word32* wc_u32)
|
|
{
|
|
*wc_u32 = ((word32)c[0] << 24) | ((word32)c[1] << 16) | (c[2] << 8) | c[3];
|
|
}
|
|
|
|
|
|
WC_STATIC WC_INLINE word32 btoi(byte b)
|
|
{
|
|
return (word32)(b - 0x30);
|
|
}
|
|
#endif
|
|
|
|
|
|
#ifndef WOLFSSL_NO_CT_OPS
|
|
/* Constant time - mask set when a > b. */
|
|
WC_STATIC WC_INLINE byte ctMaskGT(int a, int b)
|
|
{
|
|
return (byte)((((word32)a - b - 1) >> 31) - 1);
|
|
}
|
|
|
|
/* Constant time - mask set when a >= b. */
|
|
WC_STATIC WC_INLINE byte ctMaskGTE(int a, int b)
|
|
{
|
|
return (byte)((((word32)a - b ) >> 31) - 1);
|
|
}
|
|
|
|
/* Constant time - mask set when a >= b. */
|
|
WC_STATIC WC_INLINE int ctMaskIntGTE(int a, int b)
|
|
{
|
|
return (int)((((word32)a - b ) >> 31) - 1);
|
|
}
|
|
|
|
/* Constant time - mask set when a < b. */
|
|
WC_STATIC WC_INLINE byte ctMaskLT(int a, int b)
|
|
{
|
|
return (byte)((((word32)b - a - 1) >> 31) - 1);
|
|
}
|
|
|
|
/* Constant time - mask set when a <= b. */
|
|
WC_STATIC WC_INLINE byte ctMaskLTE(int a, int b)
|
|
{
|
|
return (byte)((((word32)b - a ) >> 31) - 1);
|
|
}
|
|
|
|
/* Constant time - mask set when a == b. */
|
|
WC_STATIC WC_INLINE byte ctMaskEq(int a, int b)
|
|
{
|
|
return (byte)(~ctMaskGT(a, b)) & (byte)(~ctMaskLT(a, b));
|
|
}
|
|
|
|
/* Constant time - sets 16 bit integer mask when a > b */
|
|
WC_STATIC WC_INLINE word16 ctMask16GT(int a, int b)
|
|
{
|
|
return (word16)((((word32)a - b - 1) >> 31) - 1);
|
|
}
|
|
|
|
/* Constant time - sets 16 bit integer mask when a >= b */
|
|
WC_STATIC WC_INLINE word16 ctMask16GTE(int a, int b)
|
|
{
|
|
return (word16)((((word32)a - b ) >> 31) - 1);
|
|
}
|
|
|
|
/* Constant time - sets 16 bit integer mask when a < b. */
|
|
WC_STATIC WC_INLINE word16 ctMask16LT(int a, int b)
|
|
{
|
|
return (word16)((((word32)b - a - 1) >> 31) - 1);
|
|
}
|
|
|
|
/* Constant time - sets 16 bit integer mask when a <= b. */
|
|
WC_STATIC WC_INLINE word16 ctMask16LTE(int a, int b)
|
|
{
|
|
return (word16)((((word32)b - a ) >> 31) - 1);
|
|
}
|
|
|
|
/* Constant time - sets 16 bit integer mask when a == b. */
|
|
WC_STATIC WC_INLINE word16 ctMask16Eq(int a, int b)
|
|
{
|
|
return (word16)(~ctMask16GT(a, b)) & (word16)(~ctMask16LT(a, b));
|
|
}
|
|
|
|
/* Constant time - mask set when a != b. */
|
|
WC_STATIC WC_INLINE byte ctMaskNotEq(int a, int b)
|
|
{
|
|
return (byte)ctMaskGT(a, b) | (byte)ctMaskLT(a, b);
|
|
}
|
|
|
|
/* Constant time - select a when mask is set and b otherwise. */
|
|
WC_STATIC WC_INLINE byte ctMaskSel(byte m, byte a, byte b)
|
|
{
|
|
return (byte)((b & ((byte)~(word32)m)) | (a & m));
|
|
}
|
|
|
|
/* Constant time - select integer a when mask is set and integer b otherwise. */
|
|
WC_STATIC WC_INLINE int ctMaskSelInt(byte m, int a, int b)
|
|
{
|
|
return (b & (~(signed int)(signed char)m)) |
|
|
(a & ( (signed int)(signed char)m));
|
|
}
|
|
|
|
/* Constant time - bit set when a <= b. */
|
|
WC_STATIC WC_INLINE byte ctSetLTE(int a, int b)
|
|
{
|
|
return (byte)(((word32)a - b - 1) >> 31);
|
|
}
|
|
#endif
|
|
|
|
|
|
#undef WC_STATIC
|
|
|
|
#endif /* !WOLFSSL_MISC_INCLUDED && !NO_INLINE */
|
|
|
|
#endif /* WOLF_CRYPT_MISC_C */
|