Use a tuklib module for integer handling.
This replaces bswap.h and integer.h. The tuklib module uses <byteswap.h> on GNU, <sys/endian.h> on *BSDs and <sys/byteorder.h> on Solaris, which may contain optimized code like inline assembly.
This commit is contained in:
Lasse Collin
parent
29fd321033
commit
ebfb2c5e1f
54
configure.ac
54
configure.ac
|
|
@ -315,35 +315,6 @@ AM_CONDITIONAL(COND_ASM_X86, test "x$enable_assembler" = xx86)
|
|||
AM_CONDITIONAL(COND_ASM_X86_64, test "x$enable_assembler" = xx86_64)
|
||||
|
||||
|
||||
################################
|
||||
# Fast unaligned memory access #
|
||||
################################
|
||||
|
||||
AC_MSG_CHECKING([if unaligned memory access should be used])
|
||||
AC_ARG_ENABLE([unaligned-access], AC_HELP_STRING([--enable-unaligned-access],
|
||||
[Enable if the system supports *fast* unaligned memory access
|
||||
with 16-bit and 32-bit integers. By default, this is enabled
|
||||
only on x86, x86_64, and big endian PowerPC.]),
|
||||
[], [enable_unaligned_access=auto])
|
||||
if test "x$enable_unaligned_access" = xauto ; then
|
||||
case $host_cpu in
|
||||
i?86|x86_64|powerpc|powerpc64)
|
||||
enable_unaligned_access=yes
|
||||
;;
|
||||
*)
|
||||
enable_unaligned_access=no
|
||||
;;
|
||||
esac
|
||||
fi
|
||||
if test "x$enable_unaligned_access" = xyes ; then
|
||||
AC_DEFINE([HAVE_FAST_UNALIGNED_ACCESS], [1], [Define to 1 if
|
||||
the system supports fast unaligned memory access.])
|
||||
AC_MSG_RESULT([yes])
|
||||
else
|
||||
AC_MSG_RESULT([no])
|
||||
fi
|
||||
|
||||
|
||||
#####################
|
||||
# Size optimization #
|
||||
#####################
|
||||
|
|
@ -508,30 +479,6 @@ AC_CHECK_HEADERS([fcntl.h limits.h sys/time.h],
|
|||
[],
|
||||
[AC_MSG_ERROR([Required header file(s) are missing.])])
|
||||
|
||||
# If any of these headers are missing, things should still work correctly:
|
||||
AC_CHECK_HEADERS([byteswap.h])
|
||||
|
||||
# Even if we have byteswap.h, we may lack the specific macros/functions.
|
||||
if test x$ac_cv_header_byteswap_h = xyes ; then
|
||||
m4_foreach([FUNC], [bswap_16,bswap_32,bswap_64], [
|
||||
AC_MSG_CHECKING([if FUNC is available])
|
||||
AC_LINK_IFELSE([AC_LANG_SOURCE([
|
||||
#include <byteswap.h>
|
||||
int
|
||||
main(void)
|
||||
{
|
||||
FUNC[](42);
|
||||
return 0;
|
||||
}
|
||||
])], [
|
||||
AC_DEFINE(HAVE_[]m4_toupper(FUNC), [1],
|
||||
[Define to 1 if] FUNC [is available.])
|
||||
AC_MSG_RESULT([yes])
|
||||
], [AC_MSG_RESULT([no])])
|
||||
|
||||
])dnl
|
||||
fi
|
||||
|
||||
|
||||
###############################################################################
|
||||
# Checks for typedefs, structures, and compiler characteristics.
|
||||
|
|
@ -578,6 +525,7 @@ gl_GETOPT
|
|||
AC_CHECK_FUNCS([futimens futimes futimesat utimes utime], [break])
|
||||
|
||||
TUKLIB_PROGNAME
|
||||
TUKLIB_INTEGER
|
||||
TUKLIB_PHYSMEM
|
||||
TUKLIB_CPUCORES
|
||||
|
||||
|
|
|
|||
|
|
@ -0,0 +1,74 @@
|
|||
#
|
||||
# SYNOPSIS
|
||||
#
|
||||
# TUKLIB_INTEGER
|
||||
#
|
||||
# DESCRIPTION
|
||||
#
|
||||
# Checks for tuklib_integer.h:
|
||||
# - Endianness
|
||||
# - Does operating system provide byte swapping macros
|
||||
# - Does the hardware support fast unaligned access to 16-bit
|
||||
# and 32-bit integers
|
||||
#
|
||||
# COPYING
|
||||
#
|
||||
# Author: Lasse Collin
|
||||
#
|
||||
# This file has been put into the public domain.
|
||||
# You can do whatever you want with this file.
|
||||
#
|
||||
|
||||
AC_DEFUN_ONCE([TUKLIB_INTEGER], [
|
||||
AC_REQUIRE([TUKLIB_COMMON])
|
||||
AC_REQUIRE([AC_C_BIGENDIAN])
|
||||
AC_CHECK_HEADERS([byteswap.h sys/endian.h sys/byteorder.h], [break])
|
||||
|
||||
# Even if we have byteswap.h, we may lack the specific macros/functions.
|
||||
if test x$ac_cv_header_byteswap_h = xyes ; then
|
||||
m4_foreach([FUNC], [bswap_16,bswap_32,bswap_64], [
|
||||
AC_MSG_CHECKING([if FUNC is available])
|
||||
AC_LINK_IFELSE([AC_LANG_SOURCE([
|
||||
#include <byteswap.h>
|
||||
int
|
||||
main(void)
|
||||
{
|
||||
FUNC[](42);
|
||||
return 0;
|
||||
}
|
||||
])], [
|
||||
AC_DEFINE(HAVE_[]m4_toupper(FUNC), [1],
|
||||
[Define to 1 if] FUNC [is available.])
|
||||
AC_MSG_RESULT([yes])
|
||||
], [AC_MSG_RESULT([no])])
|
||||
|
||||
])dnl
|
||||
fi
|
||||
|
||||
AC_MSG_CHECKING([if unaligned memory access should be used])
|
||||
AC_ARG_ENABLE([unaligned-access], AC_HELP_STRING([--enable-unaligned-access],
|
||||
[Enable if the system supports *fast* unaligned memory access
|
||||
with 16-bit and 32-bit integers. By default, this is enabled
|
||||
only on x86, x86_64, and big endian PowerPC.]),
|
||||
[], [enable_unaligned_access=auto])
|
||||
if test "x$enable_unaligned_access" = xauto ; then
|
||||
# TODO: There may be other architectures, on which unaligned access
|
||||
# is OK.
|
||||
case $host_cpu in
|
||||
i?86|x86_64|powerpc|powerpc64)
|
||||
enable_unaligned_access=yes
|
||||
;;
|
||||
*)
|
||||
enable_unaligned_access=no
|
||||
;;
|
||||
esac
|
||||
fi
|
||||
if test "x$enable_unaligned_access" = xyes ; then
|
||||
AC_DEFINE([TUKLIB_FAST_UNALIGNED_ACCESS], [1], [Define to 1 if
|
||||
the system supports fast unaligned access to 16-bit and
|
||||
32-bit integers.])
|
||||
AC_MSG_RESULT([yes])
|
||||
else
|
||||
AC_MSG_RESULT([no])
|
||||
fi
|
||||
])dnl
|
||||
|
|
@ -1,52 +0,0 @@
|
|||
///////////////////////////////////////////////////////////////////////////////
|
||||
//
|
||||
/// \file bswap.h
|
||||
/// \brief Byte swapping
|
||||
//
|
||||
// Author: Lasse Collin
|
||||
//
|
||||
// This file has been put into the public domain.
|
||||
// You can do whatever you want with this file.
|
||||
//
|
||||
///////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
#ifndef LZMA_BSWAP_H
|
||||
#define LZMA_BSWAP_H
|
||||
|
||||
// NOTE: We assume that config.h is already #included.
|
||||
|
||||
// At least glibc has byteswap.h which contains inline assembly code for
|
||||
// byteswapping. Some systems have byteswap.h but lack one or more of the
|
||||
// bswap_xx macros/functions, which is why we check them separately even
|
||||
// if byteswap.h is available.
|
||||
|
||||
#ifdef HAVE_BYTESWAP_H
|
||||
# include <byteswap.h>
|
||||
#endif
|
||||
|
||||
#ifndef HAVE_BSWAP_16
|
||||
# define bswap_16(num) \
|
||||
(((num) << 8) | ((num) >> 8))
|
||||
#endif
|
||||
|
||||
#ifndef HAVE_BSWAP_32
|
||||
# define bswap_32(num) \
|
||||
( (((num) << 24) ) \
|
||||
| (((num) << 8) & UINT32_C(0x00FF0000)) \
|
||||
| (((num) >> 8) & UINT32_C(0x0000FF00)) \
|
||||
| (((num) >> 24) ) )
|
||||
#endif
|
||||
|
||||
#ifndef HAVE_BSWAP_64
|
||||
# define bswap_64(num) \
|
||||
( (((num) << 56) ) \
|
||||
| (((num) << 40) & UINT64_C(0x00FF000000000000)) \
|
||||
| (((num) << 24) & UINT64_C(0x0000FF0000000000)) \
|
||||
| (((num) << 8) & UINT64_C(0x000000FF00000000)) \
|
||||
| (((num) >> 8) & UINT64_C(0x00000000FF000000)) \
|
||||
| (((num) >> 24) & UINT64_C(0x0000000000FF0000)) \
|
||||
| (((num) >> 40) & UINT64_C(0x000000000000FF00)) \
|
||||
| (((num) >> 56) ) )
|
||||
#endif
|
||||
|
||||
#endif
|
||||
|
|
@ -1,170 +0,0 @@
|
|||
///////////////////////////////////////////////////////////////////////////////
|
||||
//
|
||||
/// \file integer.h
|
||||
/// \brief Reading and writing integers from and to buffers
|
||||
//
|
||||
// Author: Lasse Collin
|
||||
//
|
||||
// This file has been put into the public domain.
|
||||
// You can do whatever you want with this file.
|
||||
//
|
||||
///////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
#ifndef LZMA_INTEGER_H
|
||||
#define LZMA_INTEGER_H
|
||||
|
||||
// On big endian, we need byte swapping. These macros may be used outside
|
||||
// this file, so don't put these inside HAVE_FAST_UNALIGNED_ACCESS.
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
# include "bswap.h"
|
||||
# define integer_le_16(n) bswap_16(n)
|
||||
# define integer_le_32(n) bswap_32(n)
|
||||
# define integer_le_64(n) bswap_64(n)
|
||||
#else
|
||||
# define integer_le_16(n) (n)
|
||||
# define integer_le_32(n) (n)
|
||||
# define integer_le_64(n) (n)
|
||||
#endif
|
||||
|
||||
|
||||
// I'm aware of AC_CHECK_ALIGNED_ACCESS_REQUIRED from Autoconf archive, but
|
||||
// it's not useful here. We don't care if unaligned access is supported,
|
||||
// we care if it is fast. Some systems can emulate unaligned access in
|
||||
// software, which is horribly slow; we want to use byte-by-byte access on
|
||||
// such systems but the Autoconf test would detect such a system as
|
||||
// supporting unaligned access.
|
||||
//
|
||||
// NOTE: HAVE_FAST_UNALIGNED_ACCESS indicates only support for 16-bit and
|
||||
// 32-bit integer loads and stores. 64-bit integers may or may not work.
|
||||
// That's why 64-bit functions are commented out.
|
||||
//
|
||||
// TODO: Big endian PowerPC supports byte swapping load and store instructions
|
||||
// that also allow unaligned access. Inline assembler could be OK for that.
|
||||
//
|
||||
// Performance of these functions isn't that important until LZMA3, but it
|
||||
// doesn't hurt to have these ready already.
|
||||
#ifdef HAVE_FAST_UNALIGNED_ACCESS
|
||||
|
||||
static inline uint16_t
|
||||
integer_read_16(const uint8_t buf[static 2])
|
||||
{
|
||||
uint16_t ret = *(const uint16_t *)(buf);
|
||||
return integer_le_16(ret);
|
||||
}
|
||||
|
||||
|
||||
static inline uint32_t
|
||||
integer_read_32(const uint8_t buf[static 4])
|
||||
{
|
||||
uint32_t ret = *(const uint32_t *)(buf);
|
||||
return integer_le_32(ret);
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
static inline uint64_t
|
||||
integer_read_64(const uint8_t buf[static 8])
|
||||
{
|
||||
uint64_t ret = *(const uint64_t *)(buf);
|
||||
return integer_le_64(ret);
|
||||
}
|
||||
*/
|
||||
|
||||
|
||||
static inline void
|
||||
integer_write_16(uint8_t buf[static 2], uint16_t num)
|
||||
{
|
||||
*(uint16_t *)(buf) = integer_le_16(num);
|
||||
}
|
||||
|
||||
|
||||
static inline void
|
||||
integer_write_32(uint8_t buf[static 4], uint32_t num)
|
||||
{
|
||||
*(uint32_t *)(buf) = integer_le_32(num);
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
static inline void
|
||||
integer_write_64(uint8_t buf[static 8], uint64_t num)
|
||||
{
|
||||
*(uint64_t *)(buf) = integer_le_64(num);
|
||||
}
|
||||
*/
|
||||
|
||||
|
||||
#else
|
||||
|
||||
static inline uint16_t
|
||||
integer_read_16(const uint8_t buf[static 2])
|
||||
{
|
||||
uint16_t ret = buf[0] | (buf[1] << 8);
|
||||
return ret;
|
||||
}
|
||||
|
||||
|
||||
static inline uint32_t
|
||||
integer_read_32(const uint8_t buf[static 4])
|
||||
{
|
||||
uint32_t ret = buf[0];
|
||||
ret |= (uint32_t)(buf[1]) << 8;
|
||||
ret |= (uint32_t)(buf[2]) << 16;
|
||||
ret |= (uint32_t)(buf[3]) << 24;
|
||||
return ret;
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
static inline uint64_t
|
||||
integer_read_64(const uint8_t buf[static 8])
|
||||
{
|
||||
uint64_t ret = buf[0];
|
||||
ret |= (uint64_t)(buf[1]) << 8;
|
||||
ret |= (uint64_t)(buf[2]) << 16;
|
||||
ret |= (uint64_t)(buf[3]) << 24;
|
||||
ret |= (uint64_t)(buf[4]) << 32;
|
||||
ret |= (uint64_t)(buf[5]) << 40;
|
||||
ret |= (uint64_t)(buf[6]) << 48;
|
||||
ret |= (uint64_t)(buf[7]) << 56;
|
||||
return ret;
|
||||
}
|
||||
*/
|
||||
|
||||
|
||||
static inline void
|
||||
integer_write_16(uint8_t buf[static 2], uint16_t num)
|
||||
{
|
||||
buf[0] = (uint8_t)(num);
|
||||
buf[1] = (uint8_t)(num >> 8);
|
||||
}
|
||||
|
||||
|
||||
static inline void
|
||||
integer_write_32(uint8_t buf[static 4], uint32_t num)
|
||||
{
|
||||
buf[0] = (uint8_t)(num);
|
||||
buf[1] = (uint8_t)(num >> 8);
|
||||
buf[2] = (uint8_t)(num >> 16);
|
||||
buf[3] = (uint8_t)(num >> 24);
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
static inline void
|
||||
integer_write_64(uint8_t buf[static 8], uint64_t num)
|
||||
{
|
||||
buf[0] = (uint8_t)(num);
|
||||
buf[1] = (uint8_t)(num >> 8);
|
||||
buf[2] = (uint8_t)(num >> 16);
|
||||
buf[3] = (uint8_t)(num >> 24);
|
||||
buf[4] = (uint8_t)(num >> 32);
|
||||
buf[5] = (uint8_t)(num >> 40);
|
||||
buf[6] = (uint8_t)(num >> 48);
|
||||
buf[7] = (uint8_t)(num >> 56);
|
||||
}
|
||||
*/
|
||||
|
||||
#endif
|
||||
|
||||
#endif
|
||||
|
|
@ -1 +1,7 @@
|
|||
#include "sysdefs.h"
|
||||
#ifdef HAVE_CONFIG_H
|
||||
# include "sysdefs.h"
|
||||
#else
|
||||
# include <stddef.h>
|
||||
# include <inttypes.h>
|
||||
# include <limits.h>
|
||||
#endif
|
||||
|
|
|
|||
|
|
@ -0,0 +1,350 @@
|
|||
///////////////////////////////////////////////////////////////////////////////
|
||||
//
|
||||
/// \file tuklib_integer.h
|
||||
/// \brief Byte swapping and endianness related macros and functions
|
||||
///
|
||||
/// This file provides macros or functions to do basic endianness related
|
||||
/// integer operations (XX = 16, 32, or 64; Y = b or l):
|
||||
/// - Byte swapping: bswapXX(num)
|
||||
/// - Byte order conversions to/from native: convXXYe(num)
|
||||
/// - Aligned reads: readXXYe(ptr)
|
||||
/// - Aligned writes: writeXXYe(ptr, num)
|
||||
/// - Unaligned reads (16/32-bit only): unaligned_readXXYe(ptr)
|
||||
/// - Unaligned writes (16/32-bit only): unaligned_writeXXYe(ptr, num)
|
||||
///
|
||||
/// Since they can macros, the arguments should have no side effects since
|
||||
/// they may be evaluated more than once.
|
||||
///
|
||||
/// \todo PowerPC and possibly some other architectures support
|
||||
/// byte swapping load and store instructions. This file
|
||||
/// doesn't take advantage of those instructions.
|
||||
//
|
||||
// Author: Lasse Collin
|
||||
//
|
||||
// This file has been put into the public domain.
|
||||
// You can do whatever you want with this file.
|
||||
//
|
||||
///////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
#ifndef TUKLIB_INTEGER_H
|
||||
#define TUKLIB_INTEGER_H
|
||||
|
||||
#include "tuklib_common.h"
|
||||
|
||||
|
||||
////////////////////////////////////////
|
||||
// Operating system specific features //
|
||||
////////////////////////////////////////
|
||||
|
||||
#if defined(HAVE_BYTESWAP_H)
|
||||
// glibc, uClibc, dietlibc
|
||||
# include <byteswap.h>
|
||||
# ifdef HAVE_BSWAP_16
|
||||
# define bswap16(num) bswap_16(num)
|
||||
# endif
|
||||
# ifdef HAVE_BSWAP_32
|
||||
# define bswap32(num) bswap_32(num)
|
||||
# endif
|
||||
# ifdef HAVE_BSWAP_64
|
||||
# define bswap64(num) bswap_64(num)
|
||||
# endif
|
||||
|
||||
#elif defined(HAVE_SYS_ENDIAN_H)
|
||||
// *BSDs and Darwin
|
||||
# include <sys/endian.h>
|
||||
|
||||
#elif defined(HAVE_SYS_BYTEORDER_H)
|
||||
// Solaris
|
||||
# include <sys/byteorder.h>
|
||||
# ifdef BSWAP_16
|
||||
# define bswap16(num) BSWAP_16(num)
|
||||
# endif
|
||||
# ifdef BSWAP_32
|
||||
# define bswap32(num) BSWAP_32(num)
|
||||
# endif
|
||||
# ifdef BSWAP_64
|
||||
# define bswap64(num) BSWAP_64(num)
|
||||
# endif
|
||||
# ifdef BE_16
|
||||
# define conv16be(num) BE_16(num)
|
||||
# endif
|
||||
# ifdef BE_32
|
||||
# define conv32be(num) BE_32(num)
|
||||
# endif
|
||||
# ifdef BE_64
|
||||
# define conv64be(num) BE_64(num)
|
||||
# endif
|
||||
# ifdef LE_16
|
||||
# define conv16le(num) LE_16(num)
|
||||
# endif
|
||||
# ifdef LE_32
|
||||
# define conv32le(num) LE_32(num)
|
||||
# endif
|
||||
# ifdef LE_64
|
||||
# define conv64le(num) LE_64(num)
|
||||
# endif
|
||||
#endif
|
||||
|
||||
|
||||
///////////////////
|
||||
// Byte swapping //
|
||||
///////////////////
|
||||
|
||||
#ifndef bswap16
|
||||
# define bswap16(num) \
|
||||
(((uint16_t)(num) << 8) | ((uint16_t)(num) >> 8))
|
||||
#endif
|
||||
|
||||
#ifndef bswap32
|
||||
# define bswap32(num) \
|
||||
( (((uint32_t)(num) << 24) ) \
|
||||
| (((uint32_t)(num) << 8) & UINT32_C(0x00FF0000)) \
|
||||
| (((uint32_t)(num) >> 8) & UINT32_C(0x0000FF00)) \
|
||||
| (((uint32_t)(num) >> 24) ) )
|
||||
#endif
|
||||
|
||||
#ifndef bswap64
|
||||
# define bswap64(num) \
|
||||
( (((uint64_t)(num) << 56) ) \
|
||||
| (((uint64_t)(num) << 40) & UINT64_C(0x00FF000000000000)) \
|
||||
| (((uint64_t)(num) << 24) & UINT64_C(0x0000FF0000000000)) \
|
||||
| (((uint64_t)(num) << 8) & UINT64_C(0x000000FF00000000)) \
|
||||
| (((uint64_t)(num) >> 8) & UINT64_C(0x00000000FF000000)) \
|
||||
| (((uint64_t)(num) >> 24) & UINT64_C(0x0000000000FF0000)) \
|
||||
| (((uint64_t)(num) >> 40) & UINT64_C(0x000000000000FF00)) \
|
||||
| (((uint64_t)(num) >> 56) ) )
|
||||
#endif
|
||||
|
||||
// Define conversion macros using the basic byte swapping macros.
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
# ifndef conv16be
|
||||
# define conv16be(num) ((uint16_t)(num))
|
||||
# endif
|
||||
# ifndef conv32be
|
||||
# define conv32be(num) ((uint32_t)(num))
|
||||
# endif
|
||||
# ifndef conv64be
|
||||
# define conv64be(num) ((uint64_t)(num))
|
||||
# endif
|
||||
# ifndef conv16le
|
||||
# define conv16le(num) bswap16(num)
|
||||
# endif
|
||||
# ifndef conv32le
|
||||
# define conv32le(num) bswap32(num)
|
||||
# endif
|
||||
# ifndef conv64le
|
||||
# define conv64le(num) bswap64(num)
|
||||
# endif
|
||||
#else
|
||||
# ifndef conv16be
|
||||
# define conv16be(num) bswap16(num)
|
||||
# endif
|
||||
# ifndef conv32be
|
||||
# define conv32be(num) bswap32(num)
|
||||
# endif
|
||||
# ifndef conv64be
|
||||
# define conv64be(num) bswap64(num)
|
||||
# endif
|
||||
# ifndef conv16le
|
||||
# define conv16le(num) ((uint16_t)(num))
|
||||
# endif
|
||||
# ifndef conv32le
|
||||
# define conv32le(num) ((uint32_t)(num))
|
||||
# endif
|
||||
# ifndef conv64le
|
||||
# define conv64le(num) ((uint64_t)(num))
|
||||
# endif
|
||||
#endif
|
||||
|
||||
|
||||
//////////////////////////////
|
||||
// Aligned reads and writes //
|
||||
//////////////////////////////
|
||||
|
||||
static inline uint16_t
|
||||
read16be(const uint8_t *buf)
|
||||
{
|
||||
uint16_t num = *(const uint16_t *)buf;
|
||||
return conv16be(num);
|
||||
}
|
||||
|
||||
|
||||
static inline uint16_t
|
||||
read16le(const uint8_t *buf)
|
||||
{
|
||||
uint16_t num = *(const uint16_t *)buf;
|
||||
return conv16le(num);
|
||||
}
|
||||
|
||||
|
||||
static inline uint32_t
|
||||
read32be(const uint8_t *buf)
|
||||
{
|
||||
uint32_t num = *(const uint32_t *)buf;
|
||||
return conv32be(num);
|
||||
}
|
||||
|
||||
|
||||
static inline uint32_t
|
||||
read32le(const uint8_t *buf)
|
||||
{
|
||||
uint32_t num = *(const uint32_t *)buf;
|
||||
return conv32le(num);
|
||||
}
|
||||
|
||||
|
||||
static inline uint64_t
|
||||
read64be(const uint8_t *buf)
|
||||
{
|
||||
uint64_t num = *(const uint64_t *)buf;
|
||||
return conv64be(num);
|
||||
}
|
||||
|
||||
|
||||
static inline uint64_t
|
||||
read64le(const uint8_t *buf)
|
||||
{
|
||||
uint64_t num = *(const uint64_t *)buf;
|
||||
return conv64le(num);
|
||||
}
|
||||
|
||||
|
||||
// NOTE: Possible byte swapping must be done in a macro to allow GCC
|
||||
// to optimize byte swapping of constants when using glibc's or *BSD's
|
||||
// byte swapping macros. The actual write is done in an inline function
|
||||
// to make type checking of the buf pointer possible similarly to readXXYe()
|
||||
// functions. This also seems to silence a probably bogus GCC warning about
|
||||
// strict aliasing when buf points to the beginning of an uint8_t array.
|
||||
|
||||
#define write16be(buf, num) write16ne((buf), conv16be(num))
|
||||
#define write16le(buf, num) write16ne((buf), conv16le(num))
|
||||
#define write32be(buf, num) write32ne((buf), conv32be(num))
|
||||
#define write32le(buf, num) write32ne((buf), conv32le(num))
|
||||
#define write64be(buf, num) write64ne((buf), conv64be(num))
|
||||
#define write64le(buf, num) write64ne((buf), conv64le(num))
|
||||
|
||||
|
||||
static inline void
|
||||
write16ne(uint8_t *buf, uint16_t num)
|
||||
{
|
||||
*(uint16_t *)buf = num;
|
||||
return;
|
||||
}
|
||||
|
||||
|
||||
static inline void
|
||||
write32ne(uint8_t *buf, uint32_t num)
|
||||
{
|
||||
*(uint32_t *)buf = num;
|
||||
return;
|
||||
}
|
||||
|
||||
|
||||
static inline void
|
||||
write64ne(uint8_t *buf, uint64_t num)
|
||||
{
|
||||
*(uint64_t *)buf = num;
|
||||
return;
|
||||
}
|
||||
|
||||
|
||||
////////////////////////////////
|
||||
// Unaligned reads and writes //
|
||||
////////////////////////////////
|
||||
|
||||
// NOTE: TUKLIB_FAST_UNALIGNED_ACCESS indicates only support for 16-bit and
|
||||
// 32-bit unaligned integer loads and stores. It's possible that 64-bit
|
||||
// unaligned access doesn't work or is slower than byte-by-byte access.
|
||||
// Since unaligned 64-bit is probably not needed as often as 16-bit or
|
||||
// 32-bit, we simply don't support 64-bit unaligned access for now.
|
||||
#ifdef TUKLIB_FAST_UNALIGNED_ACCESS
|
||||
# define unaligned_read16be read16be
|
||||
# define unaligned_read16le read16le
|
||||
# define unaligned_read32be read32be
|
||||
# define unaligned_read32le read32le
|
||||
# define unaligned_write16be write16be
|
||||
# define unaligned_write16le write16le
|
||||
# define unaligned_write32be write32be
|
||||
# define unaligned_write32le write32le
|
||||
|
||||
#else
|
||||
|
||||
static inline uint16_t
|
||||
unaligned_read16be(const uint8_t *buf)
|
||||
{
|
||||
uint16_t num = ((uint16_t)buf[0] << 8) | buf[1];
|
||||
return num;
|
||||
}
|
||||
|
||||
|
||||
static inline uint16_t
|
||||
unaligned_read16le(const uint8_t *buf)
|
||||
{
|
||||
uint16_t num = ((uint32_t)buf[0]) | ((uint16_t)buf[1] << 8);
|
||||
return num;
|
||||
}
|
||||
|
||||
|
||||
static inline uint32_t
|
||||
unaligned_read32be(const uint8_t *buf)
|
||||
{
|
||||
uint32_t num = (uint32_t)buf[0] << 24;
|
||||
num |= (uint32_t)buf[1] << 16;
|
||||
num |= (uint32_t)buf[2] << 8;
|
||||
num |= (uint32_t)buf[3];
|
||||
return num;
|
||||
}
|
||||
|
||||
|
||||
static inline uint32_t
|
||||
unaligned_read32le(const uint8_t *buf)
|
||||
{
|
||||
uint32_t num = (uint32_t)buf[0];
|
||||
num |= (uint32_t)buf[1] << 8;
|
||||
num |= (uint32_t)buf[2] << 16;
|
||||
num |= (uint32_t)buf[3] << 24;
|
||||
return num;
|
||||
}
|
||||
|
||||
|
||||
static inline void
|
||||
unaligned_write16be(uint8_t *buf, uint16_t num)
|
||||
{
|
||||
buf[0] = num >> 8;
|
||||
buf[1] = num;
|
||||
return;
|
||||
}
|
||||
|
||||
|
||||
static inline void
|
||||
unaligned_write16le(uint8_t *buf, uint16_t num)
|
||||
{
|
||||
buf[0] = num;
|
||||
buf[1] = num >> 8;
|
||||
return;
|
||||
}
|
||||
|
||||
|
||||
static inline void
|
||||
unaligned_write32be(uint8_t *buf, uint32_t num)
|
||||
{
|
||||
buf[0] = num >> 24;
|
||||
buf[1] = num >> 16;
|
||||
buf[2] = num >> 8;
|
||||
buf[3] = num;
|
||||
return;
|
||||
}
|
||||
|
||||
|
||||
static inline void
|
||||
unaligned_write32le(uint8_t *buf, uint32_t num)
|
||||
{
|
||||
buf[0] = num;
|
||||
buf[1] = num >> 8;
|
||||
buf[2] = num >> 16;
|
||||
buf[3] = num >> 24;
|
||||
return;
|
||||
}
|
||||
|
||||
#endif
|
||||
#endif
|
||||
|
|
@ -150,13 +150,13 @@ lzma_check_finish(lzma_check_state *check, lzma_check type)
|
|||
switch (type) {
|
||||
#ifdef HAVE_CHECK_CRC32
|
||||
case LZMA_CHECK_CRC32:
|
||||
check->buffer.u32[0] = integer_le_32(check->state.crc32);
|
||||
check->buffer.u32[0] = conv32le(check->state.crc32);
|
||||
break;
|
||||
#endif
|
||||
|
||||
#ifdef HAVE_CHECK_CRC64
|
||||
case LZMA_CHECK_CRC64:
|
||||
check->buffer.u64[0] = integer_le_64(check->state.crc64);
|
||||
check->buffer.u64[0] = conv64le(check->state.crc64);
|
||||
break;
|
||||
#endif
|
||||
|
||||
|
|
|
|||
|
|
@ -29,7 +29,7 @@ lzma_crc32(const uint8_t *buf, size_t size, uint32_t crc)
|
|||
crc = ~crc;
|
||||
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
crc = bswap_32(crc);
|
||||
crc = bswap32(crc);
|
||||
#endif
|
||||
|
||||
if (size > 8) {
|
||||
|
|
@ -75,7 +75,7 @@ lzma_crc32(const uint8_t *buf, size_t size, uint32_t crc)
|
|||
crc = lzma_crc32_table[0][*buf++ ^ A(crc)] ^ S8(crc);
|
||||
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
crc = bswap_32(crc);
|
||||
crc = bswap32(crc);
|
||||
#endif
|
||||
|
||||
return ~crc;
|
||||
|
|
|
|||
|
|
@ -14,12 +14,8 @@
|
|||
//
|
||||
///////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
#include <inttypes.h>
|
||||
#include <stdio.h>
|
||||
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
# include "../../common/bswap.h"
|
||||
#endif
|
||||
#include "../../common/tuklib_integer.h"
|
||||
|
||||
|
||||
static uint32_t crc32_table[8][256];
|
||||
|
|
@ -48,7 +44,7 @@ init_crc32_table(void)
|
|||
#ifdef WORDS_BIGENDIAN
|
||||
for (size_t s = 0; s < 8; ++s)
|
||||
for (size_t b = 0; b < 256; ++b)
|
||||
crc32_table[s][b] = bswap_32(crc32_table[s][b]);
|
||||
crc32_table[s][b] = bswap32(crc32_table[s][b]);
|
||||
#endif
|
||||
|
||||
return;
|
||||
|
|
|
|||
|
|
@ -32,7 +32,7 @@ lzma_crc64(const uint8_t *buf, size_t size, uint64_t crc)
|
|||
crc = ~crc;
|
||||
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
crc = bswap_64(crc);
|
||||
crc = bswap64(crc);
|
||||
#endif
|
||||
|
||||
if (size > 4) {
|
||||
|
|
@ -64,7 +64,7 @@ lzma_crc64(const uint8_t *buf, size_t size, uint64_t crc)
|
|||
crc = lzma_crc64_table[0][*buf++ ^ A1(crc)] ^ S8(crc);
|
||||
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
crc = bswap_64(crc);
|
||||
crc = bswap64(crc);
|
||||
#endif
|
||||
|
||||
return ~crc;
|
||||
|
|
|
|||
|
|
@ -13,12 +13,8 @@
|
|||
//
|
||||
///////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
#include <inttypes.h>
|
||||
#include <stdio.h>
|
||||
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
# include "../../common/bswap.h"
|
||||
#endif
|
||||
#include "../../common/tuklib_integer.h"
|
||||
|
||||
|
||||
static uint64_t crc64_table[4][256];
|
||||
|
|
@ -47,7 +43,7 @@ init_crc64_table(void)
|
|||
#ifdef WORDS_BIGENDIAN
|
||||
for (size_t s = 0; s < 4; ++s)
|
||||
for (size_t b = 0; b < 256; ++b)
|
||||
crc64_table[s][b] = bswap_64(crc64_table[s][b]);
|
||||
crc64_table[s][b] = bswap64(crc64_table[s][b]);
|
||||
#endif
|
||||
|
||||
return;
|
||||
|
|
|
|||
|
|
@ -11,8 +11,6 @@
|
|||
///////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
# include "../../common/bswap.h"
|
||||
|
||||
# define A(x) ((x) >> 24)
|
||||
# define B(x) (((x) >> 16) & 0xFF)
|
||||
# define C(x) (((x) >> 8) & 0xFF)
|
||||
|
|
|
|||
|
|
@ -29,10 +29,6 @@
|
|||
|
||||
#include "check.h"
|
||||
|
||||
#ifndef WORDS_BIGENDIAN
|
||||
# include "../../common/bswap.h"
|
||||
#endif
|
||||
|
||||
// At least on x86, GCC is able to optimize this to a rotate instruction.
|
||||
#define rotr_32(num, amount) ((num) >> (amount) | (num) << (32 - (amount)))
|
||||
|
||||
|
|
@ -123,7 +119,7 @@ process(lzma_check_state *check)
|
|||
uint32_t data[16];
|
||||
|
||||
for (size_t i = 0; i < 16; ++i)
|
||||
data[i] = bswap_32(check->buffer.u32[i]);
|
||||
data[i] = bswap32(check->buffer.u32[i]);
|
||||
|
||||
transform(check->state.sha256.state, data);
|
||||
#endif
|
||||
|
|
@ -194,20 +190,12 @@ lzma_sha256_finish(lzma_check_state *check)
|
|||
// Convert the message size from bytes to bits.
|
||||
check->state.sha256.size *= 8;
|
||||
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
check->buffer.u64[(64 - 8) / 8] = check->state.sha256.size;
|
||||
#else
|
||||
check->buffer.u64[(64 - 8) / 8] = bswap_64(check->state.sha256.size);
|
||||
#endif
|
||||
check->buffer.u64[(64 - 8) / 8] = conv64be(check->state.sha256.size);
|
||||
|
||||
process(check);
|
||||
|
||||
for (size_t i = 0; i < 8; ++i)
|
||||
#ifdef WORDS_BIGENDIAN
|
||||
check->buffer.u32[i] = check->state.sha256.state[i];
|
||||
#else
|
||||
check->buffer.u32[i] = bswap_32(check->state.sha256.state[i]);
|
||||
#endif
|
||||
check->buffer.u32[i] = conv32be(check->state.sha256.state[i]);
|
||||
|
||||
return;
|
||||
}
|
||||
|
|
|
|||
|
|
@ -116,7 +116,7 @@ alone_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
|
|||
if (d != UINT32_MAX)
|
||||
++d;
|
||||
|
||||
integer_write_32(next->coder->header + 1, d);
|
||||
unaligned_write32le(next->coder->header + 1, d);
|
||||
|
||||
// - Uncompressed size (always unknown and using EOPM)
|
||||
memset(next->coder->header + 1 + 4, 0xFF, 8);
|
||||
|
|
|
|||
|
|
@ -59,7 +59,7 @@ lzma_block_header_decode(lzma_block *block,
|
|||
const size_t in_size = block->header_size - 4;
|
||||
|
||||
// Verify CRC32
|
||||
if (lzma_crc32(in, in_size, 0) != integer_read_32(in + in_size))
|
||||
if (lzma_crc32(in, in_size, 0) != unaligned_read32le(in + in_size))
|
||||
return LZMA_DATA_ERROR;
|
||||
|
||||
// Check for unsupported flags.
|
||||
|
|
|
|||
|
|
@ -126,7 +126,7 @@ lzma_block_header_encode(const lzma_block *block, uint8_t *out)
|
|||
memzero(out + out_pos, out_size - out_pos);
|
||||
|
||||
// CRC32
|
||||
integer_write_32(out + out_size, lzma_crc32(out, out_size, 0));
|
||||
unaligned_write32le(out + out_size, lzma_crc32(out, out_size, 0));
|
||||
|
||||
return LZMA_OK;
|
||||
}
|
||||
|
|
|
|||
|
|
@ -15,7 +15,7 @@
|
|||
|
||||
#include "sysdefs.h"
|
||||
#include "mythread.h"
|
||||
#include "integer.h"
|
||||
#include "tuklib_integer.h"
|
||||
|
||||
#if defined(_WIN32) || defined(__CYGWIN__)
|
||||
# ifdef DLL_EXPORT
|
||||
|
|
|
|||
|
|
@ -38,7 +38,7 @@ lzma_stream_header_decode(lzma_stream_flags *options, const uint8_t *in)
|
|||
// and unsupported files.
|
||||
const uint32_t crc = lzma_crc32(in + sizeof(lzma_header_magic),
|
||||
LZMA_STREAM_FLAGS_SIZE, 0);
|
||||
if (crc != integer_read_32(in + sizeof(lzma_header_magic)
|
||||
if (crc != unaligned_read32le(in + sizeof(lzma_header_magic)
|
||||
+ LZMA_STREAM_FLAGS_SIZE))
|
||||
return LZMA_DATA_ERROR;
|
||||
|
||||
|
|
@ -67,7 +67,7 @@ lzma_stream_footer_decode(lzma_stream_flags *options, const uint8_t *in)
|
|||
// CRC32
|
||||
const uint32_t crc = lzma_crc32(in + sizeof(uint32_t),
|
||||
sizeof(uint32_t) + LZMA_STREAM_FLAGS_SIZE, 0);
|
||||
if (crc != integer_read_32(in))
|
||||
if (crc != unaligned_read32le(in))
|
||||
return LZMA_DATA_ERROR;
|
||||
|
||||
// Stream Flags
|
||||
|
|
@ -75,7 +75,7 @@ lzma_stream_footer_decode(lzma_stream_flags *options, const uint8_t *in)
|
|||
return LZMA_OPTIONS_ERROR;
|
||||
|
||||
// Backward Size
|
||||
options->backward_size = integer_read_32(in + sizeof(uint32_t));
|
||||
options->backward_size = unaligned_read32le(in + sizeof(uint32_t));
|
||||
options->backward_size = (options->backward_size + 1) * 4;
|
||||
|
||||
return LZMA_OK;
|
||||
|
|
|
|||
|
|
@ -46,7 +46,7 @@ lzma_stream_header_encode(const lzma_stream_flags *options, uint8_t *out)
|
|||
const uint32_t crc = lzma_crc32(out + sizeof(lzma_header_magic),
|
||||
LZMA_STREAM_FLAGS_SIZE, 0);
|
||||
|
||||
integer_write_32(out + sizeof(lzma_header_magic)
|
||||
unaligned_write32le(out + sizeof(lzma_header_magic)
|
||||
+ LZMA_STREAM_FLAGS_SIZE, crc);
|
||||
|
||||
return LZMA_OK;
|
||||
|
|
@ -66,7 +66,7 @@ lzma_stream_footer_encode(const lzma_stream_flags *options, uint8_t *out)
|
|||
if (!is_backward_size_valid(options))
|
||||
return LZMA_PROG_ERROR;
|
||||
|
||||
integer_write_32(out + 4, options->backward_size / 4 - 1);
|
||||
unaligned_write32le(out + 4, options->backward_size / 4 - 1);
|
||||
|
||||
// Stream Flags
|
||||
if (stream_flags_encode(options, out + 2 * 4))
|
||||
|
|
@ -76,7 +76,7 @@ lzma_stream_footer_encode(const lzma_stream_flags *options, uint8_t *out)
|
|||
const uint32_t crc = lzma_crc32(
|
||||
out + 4, 4 + LZMA_STREAM_FLAGS_SIZE, 0);
|
||||
|
||||
integer_write_32(out, crc);
|
||||
unaligned_write32le(out, crc);
|
||||
|
||||
// Magic
|
||||
memcpy(out + 2 * 4 + LZMA_STREAM_FLAGS_SIZE,
|
||||
|
|
|
|||
|
|
@ -37,7 +37,7 @@
|
|||
#define FIX_5_HASH_SIZE (HASH_2_SIZE + HASH_3_SIZE + HASH_4_SIZE)
|
||||
|
||||
// Endianness doesn't matter in hash_2_calc() (no effect on the output).
|
||||
#ifdef HAVE_FAST_UNALIGNED_ACCESS
|
||||
#ifdef TUKLIB_FAST_UNALIGNED_ACCESS
|
||||
# define hash_2_calc() \
|
||||
const uint32_t hash_value = *(const uint16_t *)(cur);
|
||||
#else
|
||||
|
|
|
|||
|
|
@ -1042,7 +1042,7 @@ lzma_lzma_props_decode(void **options, lzma_allocator *allocator,
|
|||
// All dictionary sizes are accepted, including zero. LZ decoder
|
||||
// will automatically use a dictionary at least a few KiB even if
|
||||
// a smaller dictionary is requested.
|
||||
opt->dict_size = integer_read_32(props + 1);
|
||||
opt->dict_size = unaligned_read32le(props + 1);
|
||||
|
||||
opt->preset_dict = NULL;
|
||||
opt->preset_dict_size = 0;
|
||||
|
|
|
|||
|
|
@ -661,7 +661,7 @@ lzma_lzma_props_encode(const void *options, uint8_t *out)
|
|||
if (lzma_lzma_lclppb_encode(opt, out))
|
||||
return LZMA_PROG_ERROR;
|
||||
|
||||
integer_write_32(out + 1, opt->dict_size);
|
||||
unaligned_write32le(out + 1, opt->dict_size);
|
||||
|
||||
return LZMA_OK;
|
||||
}
|
||||
|
|
|
|||
|
|
@ -24,7 +24,7 @@
|
|||
// needed in lzma_lzma_optimum_*() to test if the match is at least
|
||||
// MATCH_LEN_MIN bytes. Unaligned access gives tiny gain so there's no
|
||||
// reason to not use it when it is supported.
|
||||
#ifdef HAVE_FAST_UNALIGNED_ACCESS
|
||||
#ifdef TUKLIB_FAST_UNALIGNED_ACCESS
|
||||
# define not_equal_16(a, b) \
|
||||
(*(const uint16_t *)(a) != *(const uint16_t *)(b))
|
||||
#else
|
||||
|
|
|
|||
|
|
@ -28,7 +28,7 @@ lzma_simple_props_decode(void **options, lzma_allocator *allocator,
|
|||
if (opt == NULL)
|
||||
return LZMA_MEM_ERROR;
|
||||
|
||||
opt->start_offset = integer_read_32(props);
|
||||
opt->start_offset = unaligned_read32le(props);
|
||||
|
||||
// Don't leave an options structure allocated if start_offset is zero.
|
||||
if (opt->start_offset == 0)
|
||||
|
|
|
|||
|
|
@ -32,7 +32,7 @@ lzma_simple_props_encode(const void *options, uint8_t *out)
|
|||
if (opt == NULL || opt->start_offset == 0)
|
||||
return LZMA_OK;
|
||||
|
||||
integer_write_32(out, opt->start_offset);
|
||||
unaligned_write32le(out, opt->start_offset);
|
||||
|
||||
return LZMA_OK;
|
||||
}
|
||||
|
|
|
|||
|
|
@ -211,7 +211,7 @@ test3(void)
|
|||
// Unsupported filter
|
||||
// NOTE: This may need updating when new IDs become supported.
|
||||
buf[2] ^= 0x1F;
|
||||
integer_write_32(buf + known_options.header_size - 4,
|
||||
unaligned_write32le(buf + known_options.header_size - 4,
|
||||
lzma_crc32(buf, known_options.header_size - 4, 0));
|
||||
expect(lzma_block_header_decode(&decoded_options, NULL, buf)
|
||||
== LZMA_OPTIONS_ERROR);
|
||||
|
|
@ -219,7 +219,7 @@ test3(void)
|
|||
|
||||
// Non-nul Padding
|
||||
buf[known_options.header_size - 4 - 1] ^= 1;
|
||||
integer_write_32(buf + known_options.header_size - 4,
|
||||
unaligned_write32le(buf + known_options.header_size - 4,
|
||||
lzma_crc32(buf, known_options.header_size - 4, 0));
|
||||
expect(lzma_block_header_decode(&decoded_options, NULL, buf)
|
||||
== LZMA_OPTIONS_ERROR);
|
||||
|
|
|
|||
|
|
@ -133,13 +133,13 @@ test_decode_invalid(void)
|
|||
|
||||
// Test 2a (valid CRC32)
|
||||
uint32_t crc = lzma_crc32(buffer + 6, 2, 0);
|
||||
integer_write_32(buffer + 8, crc);
|
||||
unaligned_write32le(buffer + 8, crc);
|
||||
succeed(test_header_decoder(LZMA_OK));
|
||||
|
||||
// Test 2b (invalid Stream Flags with valid CRC32)
|
||||
buffer[6] ^= 0x20;
|
||||
crc = lzma_crc32(buffer + 6, 2, 0);
|
||||
integer_write_32(buffer + 8, crc);
|
||||
unaligned_write32le(buffer + 8, crc);
|
||||
succeed(test_header_decoder(LZMA_OPTIONS_ERROR));
|
||||
|
||||
// Test 3 (invalid CRC32)
|
||||
|
|
@ -151,7 +151,7 @@ test_decode_invalid(void)
|
|||
expect(lzma_stream_footer_encode(&known_flags, buffer) == LZMA_OK);
|
||||
buffer[9] ^= 0x40;
|
||||
crc = lzma_crc32(buffer + 4, 6, 0);
|
||||
integer_write_32(buffer, crc);
|
||||
unaligned_write32le(buffer, crc);
|
||||
succeed(test_footer_decoder(LZMA_OPTIONS_ERROR));
|
||||
|
||||
// Test 5 (invalid Magic Bytes)
|
||||
|
|
|
|||
|
|
@ -14,7 +14,7 @@
|
|||
#define LZMA_TESTS_H
|
||||
|
||||
#include "sysdefs.h"
|
||||
#include "integer.h"
|
||||
#include "tuklib_integer.h"
|
||||
#include "lzma.h"
|
||||
|
||||
#include <stdio.h>
|
||||
|
|
|
|||
Loading…
Reference in New Issue