xref: /OK3568_Linux_fs/kernel/lib/zstd/huf.h (revision 4882a59341e53eb6f0b4789bf948001014eff981)
1*4882a593Smuzhiyun /*
2*4882a593Smuzhiyun  * Huffman coder, part of New Generation Entropy library
3*4882a593Smuzhiyun  * header file
4*4882a593Smuzhiyun  * Copyright (C) 2013-2016, Yann Collet.
5*4882a593Smuzhiyun  *
6*4882a593Smuzhiyun  * BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
7*4882a593Smuzhiyun  *
8*4882a593Smuzhiyun  * Redistribution and use in source and binary forms, with or without
9*4882a593Smuzhiyun  * modification, are permitted provided that the following conditions are
10*4882a593Smuzhiyun  * met:
11*4882a593Smuzhiyun  *
12*4882a593Smuzhiyun  *   * Redistributions of source code must retain the above copyright
13*4882a593Smuzhiyun  * notice, this list of conditions and the following disclaimer.
14*4882a593Smuzhiyun  *   * Redistributions in binary form must reproduce the above
15*4882a593Smuzhiyun  * copyright notice, this list of conditions and the following disclaimer
16*4882a593Smuzhiyun  * in the documentation and/or other materials provided with the
17*4882a593Smuzhiyun  * distribution.
18*4882a593Smuzhiyun  *
19*4882a593Smuzhiyun  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
20*4882a593Smuzhiyun  * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
21*4882a593Smuzhiyun  * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
22*4882a593Smuzhiyun  * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
23*4882a593Smuzhiyun  * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
24*4882a593Smuzhiyun  * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
25*4882a593Smuzhiyun  * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
26*4882a593Smuzhiyun  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
27*4882a593Smuzhiyun  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
28*4882a593Smuzhiyun  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
29*4882a593Smuzhiyun  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
30*4882a593Smuzhiyun  *
31*4882a593Smuzhiyun  * This program is free software; you can redistribute it and/or modify it under
32*4882a593Smuzhiyun  * the terms of the GNU General Public License version 2 as published by the
33*4882a593Smuzhiyun  * Free Software Foundation. This program is dual-licensed; you may select
34*4882a593Smuzhiyun  * either version 2 of the GNU General Public License ("GPL") or BSD license
35*4882a593Smuzhiyun  * ("BSD").
36*4882a593Smuzhiyun  *
37*4882a593Smuzhiyun  * You can contact the author at :
38*4882a593Smuzhiyun  * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
39*4882a593Smuzhiyun  */
40*4882a593Smuzhiyun #ifndef HUF_H_298734234
41*4882a593Smuzhiyun #define HUF_H_298734234
42*4882a593Smuzhiyun 
43*4882a593Smuzhiyun /* *** Dependencies *** */
44*4882a593Smuzhiyun #include <linux/types.h> /* size_t */
45*4882a593Smuzhiyun 
46*4882a593Smuzhiyun /* ***   Tool functions *** */
47*4882a593Smuzhiyun #define HUF_BLOCKSIZE_MAX (128 * 1024) /**< maximum input size for a single block compressed with HUF_compress */
48*4882a593Smuzhiyun size_t HUF_compressBound(size_t size); /**< maximum compressed size (worst case) */
49*4882a593Smuzhiyun 
50*4882a593Smuzhiyun /* Error Management */
51*4882a593Smuzhiyun unsigned HUF_isError(size_t code); /**< tells if a return value is an error code */
52*4882a593Smuzhiyun 
53*4882a593Smuzhiyun /* ***   Advanced function   *** */
54*4882a593Smuzhiyun 
55*4882a593Smuzhiyun /** HUF_compress4X_wksp() :
56*4882a593Smuzhiyun *   Same as HUF_compress2(), but uses externally allocated `workSpace`, which must be a table of >= 1024 unsigned */
57*4882a593Smuzhiyun size_t HUF_compress4X_wksp(void *dst, size_t dstSize, const void *src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void *workSpace,
58*4882a593Smuzhiyun 			   size_t wkspSize); /**< `workSpace` must be a table of at least HUF_COMPRESS_WORKSPACE_SIZE_U32 unsigned */
59*4882a593Smuzhiyun 
60*4882a593Smuzhiyun /* *** Dependencies *** */
61*4882a593Smuzhiyun #include "mem.h" /* U32 */
62*4882a593Smuzhiyun 
63*4882a593Smuzhiyun /* *** Constants *** */
64*4882a593Smuzhiyun #define HUF_TABLELOG_MAX 12     /* max configured tableLog (for static allocation); can be modified up to HUF_ABSOLUTEMAX_TABLELOG */
65*4882a593Smuzhiyun #define HUF_TABLELOG_DEFAULT 11 /* tableLog by default, when not specified */
66*4882a593Smuzhiyun #define HUF_SYMBOLVALUE_MAX 255
67*4882a593Smuzhiyun 
68*4882a593Smuzhiyun #define HUF_TABLELOG_ABSOLUTEMAX 15 /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */
69*4882a593Smuzhiyun #if (HUF_TABLELOG_MAX > HUF_TABLELOG_ABSOLUTEMAX)
70*4882a593Smuzhiyun #error "HUF_TABLELOG_MAX is too large !"
71*4882a593Smuzhiyun #endif
72*4882a593Smuzhiyun 
73*4882a593Smuzhiyun /* ****************************************
74*4882a593Smuzhiyun *  Static allocation
75*4882a593Smuzhiyun ******************************************/
76*4882a593Smuzhiyun /* HUF buffer bounds */
77*4882a593Smuzhiyun #define HUF_CTABLEBOUND 129
78*4882a593Smuzhiyun #define HUF_BLOCKBOUND(size) (size + (size >> 8) + 8)			 /* only true if incompressible pre-filtered with fast heuristic */
79*4882a593Smuzhiyun #define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size)) /* Macro version, useful for static allocation */
80*4882a593Smuzhiyun 
81*4882a593Smuzhiyun /* static allocation of HUF's Compression Table */
82*4882a593Smuzhiyun #define HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue) \
83*4882a593Smuzhiyun 	U32 name##hb[maxSymbolValue + 1];              \
84*4882a593Smuzhiyun 	void *name##hv = &(name##hb);                  \
85*4882a593Smuzhiyun 	HUF_CElt *name = (HUF_CElt *)(name##hv) /* no final ; */
86*4882a593Smuzhiyun 
87*4882a593Smuzhiyun /* static allocation of HUF's DTable */
88*4882a593Smuzhiyun typedef U32 HUF_DTable;
89*4882a593Smuzhiyun #define HUF_DTABLE_SIZE(maxTableLog) (1 + (1 << (maxTableLog)))
90*4882a593Smuzhiyun #define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = {((U32)((maxTableLog)-1) * 0x01000001)}
91*4882a593Smuzhiyun #define HUF_CREATE_STATIC_DTABLEX4(DTable, maxTableLog) HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = {((U32)(maxTableLog)*0x01000001)}
92*4882a593Smuzhiyun 
93*4882a593Smuzhiyun /* The workspace must have alignment at least 4 and be at least this large */
94*4882a593Smuzhiyun #define HUF_COMPRESS_WORKSPACE_SIZE (6 << 10)
95*4882a593Smuzhiyun #define HUF_COMPRESS_WORKSPACE_SIZE_U32 (HUF_COMPRESS_WORKSPACE_SIZE / sizeof(U32))
96*4882a593Smuzhiyun 
97*4882a593Smuzhiyun /* The workspace must have alignment at least 4 and be at least this large */
98*4882a593Smuzhiyun #define HUF_DECOMPRESS_WORKSPACE_SIZE (3 << 10)
99*4882a593Smuzhiyun #define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))
100*4882a593Smuzhiyun 
101*4882a593Smuzhiyun /* ****************************************
102*4882a593Smuzhiyun *  Advanced decompression functions
103*4882a593Smuzhiyun ******************************************/
104*4882a593Smuzhiyun size_t HUF_decompress4X_DCtx_wksp(HUF_DTable *dctx, void *dst, size_t dstSize, const void *cSrc, size_t cSrcSize, void *workspace, size_t workspaceSize); /**< decodes RLE and uncompressed */
105*4882a593Smuzhiyun size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable *dctx, void *dst, size_t dstSize, const void *cSrc, size_t cSrcSize, void *workspace,
106*4882a593Smuzhiyun 				size_t workspaceSize);							       /**< considers RLE and uncompressed as errors */
107*4882a593Smuzhiyun size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable *dctx, void *dst, size_t dstSize, const void *cSrc, size_t cSrcSize, void *workspace,
108*4882a593Smuzhiyun 				   size_t workspaceSize); /**< single-symbol decoder */
109*4882a593Smuzhiyun size_t HUF_decompress4X4_DCtx_wksp(HUF_DTable *dctx, void *dst, size_t dstSize, const void *cSrc, size_t cSrcSize, void *workspace,
110*4882a593Smuzhiyun 				   size_t workspaceSize); /**< double-symbols decoder */
111*4882a593Smuzhiyun 
112*4882a593Smuzhiyun /* ****************************************
113*4882a593Smuzhiyun *  HUF detailed API
114*4882a593Smuzhiyun ******************************************/
115*4882a593Smuzhiyun /*!
116*4882a593Smuzhiyun HUF_compress() does the following:
117*4882a593Smuzhiyun 1. count symbol occurrence from source[] into table count[] using FSE_count()
118*4882a593Smuzhiyun 2. (optional) refine tableLog using HUF_optimalTableLog()
119*4882a593Smuzhiyun 3. build Huffman table from count using HUF_buildCTable()
120*4882a593Smuzhiyun 4. save Huffman table to memory buffer using HUF_writeCTable_wksp()
121*4882a593Smuzhiyun 5. encode the data stream using HUF_compress4X_usingCTable()
122*4882a593Smuzhiyun 
123*4882a593Smuzhiyun The following API allows targeting specific sub-functions for advanced tasks.
124*4882a593Smuzhiyun For example, it's possible to compress several blocks using the same 'CTable',
125*4882a593Smuzhiyun or to save and regenerate 'CTable' using external methods.
126*4882a593Smuzhiyun */
127*4882a593Smuzhiyun /* FSE_count() : find it within "fse.h" */
128*4882a593Smuzhiyun unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue);
129*4882a593Smuzhiyun typedef struct HUF_CElt_s HUF_CElt; /* incomplete type */
130*4882a593Smuzhiyun size_t HUF_writeCTable_wksp(void *dst, size_t maxDstSize, const HUF_CElt *CTable, unsigned maxSymbolValue, unsigned huffLog, void *workspace, size_t workspaceSize);
131*4882a593Smuzhiyun size_t HUF_compress4X_usingCTable(void *dst, size_t dstSize, const void *src, size_t srcSize, const HUF_CElt *CTable);
132*4882a593Smuzhiyun 
133*4882a593Smuzhiyun typedef enum {
134*4882a593Smuzhiyun 	HUF_repeat_none,  /**< Cannot use the previous table */
135*4882a593Smuzhiyun 	HUF_repeat_check, /**< Can use the previous table but it must be checked. Note : The previous table must have been constructed by HUF_compress{1,
136*4882a593Smuzhiyun 			     4}X_repeat */
137*4882a593Smuzhiyun 	HUF_repeat_valid  /**< Can use the previous table and it is asumed to be valid */
138*4882a593Smuzhiyun } HUF_repeat;
139*4882a593Smuzhiyun /** HUF_compress4X_repeat() :
140*4882a593Smuzhiyun *   Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
141*4882a593Smuzhiyun *   If it uses hufTable it does not modify hufTable or repeat.
142*4882a593Smuzhiyun *   If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
143*4882a593Smuzhiyun *   If preferRepeat then the old table will always be used if valid. */
144*4882a593Smuzhiyun size_t HUF_compress4X_repeat(void *dst, size_t dstSize, const void *src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void *workSpace,
145*4882a593Smuzhiyun 			     size_t wkspSize, HUF_CElt *hufTable, HUF_repeat *repeat,
146*4882a593Smuzhiyun 			     int preferRepeat); /**< `workSpace` must be a table of at least HUF_COMPRESS_WORKSPACE_SIZE_U32 unsigned */
147*4882a593Smuzhiyun 
148*4882a593Smuzhiyun /** HUF_buildCTable_wksp() :
149*4882a593Smuzhiyun  *  Same as HUF_buildCTable(), but using externally allocated scratch buffer.
150*4882a593Smuzhiyun  *  `workSpace` must be aligned on 4-bytes boundaries, and be at least as large as a table of 1024 unsigned.
151*4882a593Smuzhiyun  */
152*4882a593Smuzhiyun size_t HUF_buildCTable_wksp(HUF_CElt *tree, const U32 *count, U32 maxSymbolValue, U32 maxNbBits, void *workSpace, size_t wkspSize);
153*4882a593Smuzhiyun 
154*4882a593Smuzhiyun /*! HUF_readStats() :
155*4882a593Smuzhiyun 	Read compact Huffman tree, saved by HUF_writeCTable().
156*4882a593Smuzhiyun 	`huffWeight` is destination buffer.
157*4882a593Smuzhiyun 	@return : size read from `src` , or an error Code .
158*4882a593Smuzhiyun 	Note : Needed by HUF_readCTable() and HUF_readDTableXn() . */
159*4882a593Smuzhiyun size_t HUF_readStats_wksp(BYTE *huffWeight, size_t hwSize, U32 *rankStats, U32 *nbSymbolsPtr, U32 *tableLogPtr, const void *src, size_t srcSize,
160*4882a593Smuzhiyun 			  void *workspace, size_t workspaceSize);
161*4882a593Smuzhiyun 
162*4882a593Smuzhiyun /** HUF_readCTable() :
163*4882a593Smuzhiyun *   Loading a CTable saved with HUF_writeCTable() */
164*4882a593Smuzhiyun size_t HUF_readCTable_wksp(HUF_CElt *CTable, unsigned maxSymbolValue, const void *src, size_t srcSize, void *workspace, size_t workspaceSize);
165*4882a593Smuzhiyun 
166*4882a593Smuzhiyun /*
167*4882a593Smuzhiyun HUF_decompress() does the following:
168*4882a593Smuzhiyun 1. select the decompression algorithm (X2, X4) based on pre-computed heuristics
169*4882a593Smuzhiyun 2. build Huffman table from save, using HUF_readDTableXn()
170*4882a593Smuzhiyun 3. decode 1 or 4 segments in parallel using HUF_decompressSXn_usingDTable
171*4882a593Smuzhiyun */
172*4882a593Smuzhiyun 
173*4882a593Smuzhiyun /** HUF_selectDecoder() :
174*4882a593Smuzhiyun *   Tells which decoder is likely to decode faster,
175*4882a593Smuzhiyun *   based on a set of pre-determined metrics.
176*4882a593Smuzhiyun *   @return : 0==HUF_decompress4X2, 1==HUF_decompress4X4 .
177*4882a593Smuzhiyun *   Assumption : 0 < cSrcSize < dstSize <= 128 KB */
178*4882a593Smuzhiyun U32 HUF_selectDecoder(size_t dstSize, size_t cSrcSize);
179*4882a593Smuzhiyun 
180*4882a593Smuzhiyun size_t HUF_readDTableX2_wksp(HUF_DTable *DTable, const void *src, size_t srcSize, void *workspace, size_t workspaceSize);
181*4882a593Smuzhiyun size_t HUF_readDTableX4_wksp(HUF_DTable *DTable, const void *src, size_t srcSize, void *workspace, size_t workspaceSize);
182*4882a593Smuzhiyun 
183*4882a593Smuzhiyun size_t HUF_decompress4X_usingDTable(void *dst, size_t maxDstSize, const void *cSrc, size_t cSrcSize, const HUF_DTable *DTable);
184*4882a593Smuzhiyun size_t HUF_decompress4X2_usingDTable(void *dst, size_t maxDstSize, const void *cSrc, size_t cSrcSize, const HUF_DTable *DTable);
185*4882a593Smuzhiyun size_t HUF_decompress4X4_usingDTable(void *dst, size_t maxDstSize, const void *cSrc, size_t cSrcSize, const HUF_DTable *DTable);
186*4882a593Smuzhiyun 
187*4882a593Smuzhiyun /* single stream variants */
188*4882a593Smuzhiyun 
189*4882a593Smuzhiyun size_t HUF_compress1X_wksp(void *dst, size_t dstSize, const void *src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void *workSpace,
190*4882a593Smuzhiyun 			   size_t wkspSize); /**< `workSpace` must be a table of at least HUF_COMPRESS_WORKSPACE_SIZE_U32 unsigned */
191*4882a593Smuzhiyun size_t HUF_compress1X_usingCTable(void *dst, size_t dstSize, const void *src, size_t srcSize, const HUF_CElt *CTable);
192*4882a593Smuzhiyun /** HUF_compress1X_repeat() :
193*4882a593Smuzhiyun *   Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
194*4882a593Smuzhiyun *   If it uses hufTable it does not modify hufTable or repeat.
195*4882a593Smuzhiyun *   If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
196*4882a593Smuzhiyun *   If preferRepeat then the old table will always be used if valid. */
197*4882a593Smuzhiyun size_t HUF_compress1X_repeat(void *dst, size_t dstSize, const void *src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void *workSpace,
198*4882a593Smuzhiyun 			     size_t wkspSize, HUF_CElt *hufTable, HUF_repeat *repeat,
199*4882a593Smuzhiyun 			     int preferRepeat); /**< `workSpace` must be a table of at least HUF_COMPRESS_WORKSPACE_SIZE_U32 unsigned */
200*4882a593Smuzhiyun 
201*4882a593Smuzhiyun size_t HUF_decompress1X_DCtx_wksp(HUF_DTable *dctx, void *dst, size_t dstSize, const void *cSrc, size_t cSrcSize, void *workspace, size_t workspaceSize);
202*4882a593Smuzhiyun size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable *dctx, void *dst, size_t dstSize, const void *cSrc, size_t cSrcSize, void *workspace,
203*4882a593Smuzhiyun 				   size_t workspaceSize); /**< single-symbol decoder */
204*4882a593Smuzhiyun size_t HUF_decompress1X4_DCtx_wksp(HUF_DTable *dctx, void *dst, size_t dstSize, const void *cSrc, size_t cSrcSize, void *workspace,
205*4882a593Smuzhiyun 				   size_t workspaceSize); /**< double-symbols decoder */
206*4882a593Smuzhiyun 
207*4882a593Smuzhiyun size_t HUF_decompress1X_usingDTable(void *dst, size_t maxDstSize, const void *cSrc, size_t cSrcSize,
208*4882a593Smuzhiyun 				    const HUF_DTable *DTable); /**< automatic selection of sing or double symbol decoder, based on DTable */
209*4882a593Smuzhiyun size_t HUF_decompress1X2_usingDTable(void *dst, size_t maxDstSize, const void *cSrc, size_t cSrcSize, const HUF_DTable *DTable);
210*4882a593Smuzhiyun size_t HUF_decompress1X4_usingDTable(void *dst, size_t maxDstSize, const void *cSrc, size_t cSrcSize, const HUF_DTable *DTable);
211*4882a593Smuzhiyun 
212*4882a593Smuzhiyun #endif /* HUF_H_298734234 */
213