diff options
Diffstat (limited to 'linux/lz4defs.h')
| -rw-r--r-- | linux/lz4defs.h | 347 |
1 files changed, 197 insertions, 150 deletions
diff --git a/linux/lz4defs.h b/linux/lz4defs.h index 586b217f..42045272 100644 --- a/linux/lz4defs.h +++ b/linux/lz4defs.h @@ -1,181 +1,228 @@ +#ifndef __LZ4DEFS_H__ +#define __LZ4DEFS_H__ + /* - * lz4defs.h -- architecture specific defines - * - * Copyright (C) 2013, LG Electronics, Kyungsik Lee <kyungsik.lee@lge.com> + * lz4defs.h -- common and architecture specific defines for the kernel usage + + * LZ4 - Fast LZ compression algorithm + * Copyright (C) 2011-2016, Yann Collet. + * BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php) + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are + * met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following disclaimer + * in the documentation and/or other materials provided with the + * distribution. + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * You can contact the author at : + * - LZ4 homepage : http://www.lz4.org + * - LZ4 source repository : https://github.com/lz4/lz4 * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. + * Changed for kernel usage by: + * Sven Schmidt <4sschmid@informatik.uni-hamburg.de> */ -/* - * Detects 64 bits mode - */ -#if __SIZEOF_POINTER__ == 8 +#include <asm/unaligned.h> +#include <linux/bitops.h> +#include <linux/string.h> /* memset, memcpy */ + +#define FORCE_INLINE __always_inline + +/*-************************************ + * Basic Types + **************************************/ +#include <linux/types.h> + +typedef uint8_t BYTE; +typedef uint16_t U16; +typedef uint32_t U32; +typedef int32_t S32; +typedef uint64_t U64; +typedef uintptr_t uptrval; + +/*-************************************ + * Architecture specifics + **************************************/ +#if defined(CONFIG_64BIT) #define LZ4_ARCH64 1 #else #define LZ4_ARCH64 0 #endif -#include <linux/string.h> -#include <asm/unaligned.h> +#if defined(__LITTLE_ENDIAN) +#define LZ4_LITTLE_ENDIAN 1 +#else +#define LZ4_LITTLE_ENDIAN 0 +#endif -#define A32(_p) get_unaligned((u32 *) (_p)) -#define A16(_p) get_unaligned((u16 *) (_p)) - -#define GET_LE16_ADVANCE(_src) \ -({ \ - u16 _r = get_unaligned_le16(_src); \ - (_src) += 2; \ - _r; \ -}) - -#define PUT_LE16_ADVANCE(_dst, _v) \ -do { \ - put_unaligned_le16((_v), (_dst)); \ - (_dst) += 2; \ -} while (0) - -#define LENGTH_LONG 15 -#define COPYLENGTH 8 -#define ML_BITS 4 -#define ML_MASK ((1U << ML_BITS) - 1) -#define RUN_BITS (8 - ML_BITS) -#define RUN_MASK ((1U << RUN_BITS) - 1) -#define MEMORY_USAGE 14 -#define MINMATCH 4 -#define SKIPSTRENGTH 6 -#define LASTLITERALS 5 -#define MFLIMIT (COPYLENGTH + MINMATCH) -#define MINLENGTH (MFLIMIT + 1) -#define MAXD_LOG 16 -#define MAXD (1 << MAXD_LOG) -#define MAXD_MASK (u32)(MAXD - 1) -#define MAX_DISTANCE (MAXD - 1) -#define HASH_LOG (MAXD_LOG - 1) -#define HASHTABLESIZE (1 << HASH_LOG) -#define MAX_NB_ATTEMPTS 256 -#define OPTIMAL_ML (int)((ML_MASK-1)+MINMATCH) -#define LZ4_64KLIMIT ((1<<16) + (MFLIMIT - 1)) - -#define __HASH_VALUE(p, bits) \ - (((A32(p)) * 2654435761U) >> (32 - (bits))) - -#define HASH_VALUE(p) __HASH_VALUE(p, HASH_LOG) - -#define MEMCPY_ADVANCE(_dst, _src, length) \ -do { \ - typeof(length) _length = (length); \ - memcpy(_dst, _src, _length); \ - _src += _length; \ - _dst += _length; \ -} while (0) - -#define MEMCPY_ADVANCE_BYTES(_dst, _src, _length) \ -do { \ - const u8 *_end = (_src) + (_length); \ - while ((_src) < _end) \ - *_dst++ = *_src++; \ -} while (0) - -#define STEPSIZE __SIZEOF_LONG__ - -#define LZ4_COPYPACKET(_src, _dst) \ -do { \ - MEMCPY_ADVANCE(_dst, _src, STEPSIZE); \ - MEMCPY_ADVANCE(_dst, _src, COPYLENGTH - STEPSIZE);\ -} while (0) +/*-************************************ + * Constants + **************************************/ +#define MINMATCH 4 -/* - * Equivalent to MEMCPY_ADVANCE - except may overrun @_dst and @_src by - * COPYLENGTH: - * - * Note: src and dst may overlap (with src < dst) - we must do the copy in - * STEPSIZE chunks for correctness - * - * Note also: length may be negative - we must not call memcpy if length is - * negative, but still adjust dst and src by length - */ -#define MEMCPY_ADVANCE_CHUNKED(_dst, _src, _length) \ -do { \ - u8 *_end = (_dst) + (_length); \ - while ((_dst) < _end) \ - LZ4_COPYPACKET(_src, _dst); \ - _src -= (_dst) - _end; \ - _dst = _end; \ -} while (0) - -#define MEMCPY_ADVANCE_CHUNKED_NOFIXUP(_dst, _src, _end)\ -do { \ - while ((_dst) < (_end)) \ - LZ4_COPYPACKET((_src), (_dst)); \ -} while (0) - -struct lz4_hashtable { +#define WILDCOPYLENGTH 8 +#define LASTLITERALS 5 +#define MFLIMIT (WILDCOPYLENGTH + MINMATCH) + +/* Increase this value ==> compression run slower on incompressible data */ +#define LZ4_SKIPTRIGGER 6 + +#define HASH_UNIT sizeof(size_t) + +#define KB (1 << 10) +#define MB (1 << 20) +#define GB (1U << 30) + +#define MAXD_LOG 16 +#define MAX_DISTANCE ((1 << MAXD_LOG) - 1) +#define STEPSIZE sizeof(size_t) + +#define ML_BITS 4 +#define ML_MASK ((1U << ML_BITS) - 1) +#define RUN_BITS (8 - ML_BITS) +#define RUN_MASK ((1U << RUN_BITS) - 1) + +/*-************************************ + * Reading and writing into memory + **************************************/ +static FORCE_INLINE U16 LZ4_read16(const void *ptr) +{ + return get_unaligned((const U16 *)ptr); +} + +static FORCE_INLINE U32 LZ4_read32(const void *ptr) +{ + return get_unaligned((const U32 *)ptr); +} + +static FORCE_INLINE size_t LZ4_read_ARCH(const void *ptr) +{ + return get_unaligned((const size_t *)ptr); +} + +static FORCE_INLINE void LZ4_write16(void *memPtr, U16 value) +{ + put_unaligned(value, (U16 *)memPtr); +} + +static FORCE_INLINE void LZ4_write32(void *memPtr, U32 value) +{ + put_unaligned(value, (U32 *)memPtr); +} + +static FORCE_INLINE U16 LZ4_readLE16(const void *memPtr) +{ + return get_unaligned_le16(memPtr); +} + +static FORCE_INLINE void LZ4_writeLE16(void *memPtr, U16 value) +{ + return put_unaligned_le16(value, memPtr); +} + +static FORCE_INLINE void LZ4_copy8(void *dst, const void *src) +{ #if LZ4_ARCH64 - const u8 * const base; - u32 *table; + U64 a = get_unaligned((const U64 *)src); + + put_unaligned(a, (U64 *)dst); #else - const int base; - const u8 *table; -#endif -}; + U32 a = get_unaligned((const U32 *)src); + U32 b = get_unaligned((const U32 *)src + 1); -#if LZ4_ARCH64 -#define HTYPE u32 -#else /* 32-bit */ -#define HTYPE const u8* + put_unaligned(a, (U32 *)dst); + put_unaligned(b, (U32 *)dst + 1); #endif +} -#ifdef __BIG_ENDIAN -#define LZ4_NBCOMMONBYTES(val) (__builtin_clzl(val) >> 3) +/* + * customized variant of memcpy, + * which can overwrite up to 7 bytes beyond dstEnd + */ +static FORCE_INLINE void LZ4_wildCopy(void *dstPtr, + const void *srcPtr, void *dstEnd) +{ + BYTE *d = (BYTE *)dstPtr; + const BYTE *s = (const BYTE *)srcPtr; + BYTE *const e = (BYTE *)dstEnd; + + do { + LZ4_copy8(d, s); + d += 8; + s += 8; + } while (d < e); +} + +static FORCE_INLINE unsigned int LZ4_NbCommonBytes(register size_t val) +{ +#if LZ4_LITTLE_ENDIAN + return __ffs(val) >> 3; #else -#define LZ4_NBCOMMONBYTES(val) (__builtin_ctzl(val) >> 3) + return (BITS_PER_LONG - 1 - __fls(val)) >> 3; #endif +} -static inline unsigned common_length(const u8 *l, const u8 *r, - const u8 *const l_end) +static FORCE_INLINE unsigned int LZ4_count( + const BYTE *pIn, + const BYTE *pMatch, + const BYTE *pInLimit) { - const u8 *l_start = l; + const BYTE *const pStart = pIn; - while (likely(l <= l_end - sizeof(long))) { - unsigned long diff = - get_unaligned((unsigned long *) l) ^ - get_unaligned((unsigned long *) r); + while (likely(pIn < pInLimit - (STEPSIZE - 1))) { + size_t const diff = LZ4_read_ARCH(pMatch) ^ LZ4_read_ARCH(pIn); - if (diff) - return l + LZ4_NBCOMMONBYTES(diff) - l_start; + if (!diff) { + pIn += STEPSIZE; + pMatch += STEPSIZE; + continue; + } - l += sizeof(long); - r += sizeof(long); + pIn += LZ4_NbCommonBytes(diff); + + return (unsigned int)(pIn - pStart); } + #if LZ4_ARCH64 - if (l <= l_end - 4 && A32(r) == A32(l)) { - l += 4; - r += 4; + if ((pIn < (pInLimit - 3)) + && (LZ4_read32(pMatch) == LZ4_read32(pIn))) { + pIn += 4; + pMatch += 4; } #endif - if (l <= l_end - 2 && A16(r) == A16(l)) { - l += 2; - r += 2; - } - if (l <= l_end - 1 && *r == *l) { - l++; - r++; + + if ((pIn < (pInLimit - 1)) + && (LZ4_read16(pMatch) == LZ4_read16(pIn))) { + pIn += 2; + pMatch += 2; } - return l - l_start; -} + if ((pIn < pInLimit) && (*pMatch == *pIn)) + pIn++; -static inline unsigned encode_length(u8 **op, unsigned length) -{ - if (length >= LENGTH_LONG) { - length -= LENGTH_LONG; - - for (; length > 254 ; length -= 255) - *(*op)++ = 255; - *(*op)++ = length; - return LENGTH_LONG; - } else - return length; + return (unsigned int)(pIn - pStart); } + +typedef enum { noLimit = 0, limitedOutput = 1 } limitedOutput_directive; +typedef enum { byPtr, byU32, byU16 } tableType_t; + +typedef enum { noDict = 0, withPrefix64k, usingExtDict } dict_directive; +typedef enum { noDictIssue = 0, dictSmall } dictIssue_directive; + +typedef enum { endOnOutputSize = 0, endOnInputSize = 1 } endCondition_directive; +typedef enum { full = 0, partial = 1 } earlyEnd_directive; + +#endif |