xref: /freebsd-src/sys/contrib/openzfs/module/zstd/lib/compress/zstd_ldm.h (revision c03c5b1c80914ec656fbee84539355d1fad68bf9)
1*c03c5b1cSMartin Matuska /*
2*c03c5b1cSMartin Matuska  * Copyright (c) 2016-2020, Yann Collet, Facebook, Inc.
3*c03c5b1cSMartin Matuska  * All rights reserved.
4*c03c5b1cSMartin Matuska  *
5*c03c5b1cSMartin Matuska  * This source code is licensed under both the BSD-style license (found in the
6*c03c5b1cSMartin Matuska  * LICENSE file in the root directory of this source tree) and the GPLv2 (found
7*c03c5b1cSMartin Matuska  * in the COPYING file in the root directory of this source tree).
8*c03c5b1cSMartin Matuska  * You may select, at your option, one of the above-listed licenses.
9*c03c5b1cSMartin Matuska  */
10*c03c5b1cSMartin Matuska 
11*c03c5b1cSMartin Matuska #ifndef ZSTD_LDM_H
12*c03c5b1cSMartin Matuska #define ZSTD_LDM_H
13*c03c5b1cSMartin Matuska 
14*c03c5b1cSMartin Matuska #if defined (__cplusplus)
15*c03c5b1cSMartin Matuska extern "C" {
16*c03c5b1cSMartin Matuska #endif
17*c03c5b1cSMartin Matuska 
18*c03c5b1cSMartin Matuska #include "zstd_compress_internal.h"   /* ldmParams_t, U32 */
19*c03c5b1cSMartin Matuska #include "../zstd.h"   /* ZSTD_CCtx, size_t */
20*c03c5b1cSMartin Matuska 
21*c03c5b1cSMartin Matuska /*-*************************************
22*c03c5b1cSMartin Matuska *  Long distance matching
23*c03c5b1cSMartin Matuska ***************************************/
24*c03c5b1cSMartin Matuska 
25*c03c5b1cSMartin Matuska #define ZSTD_LDM_DEFAULT_WINDOW_LOG ZSTD_WINDOWLOG_LIMIT_DEFAULT
26*c03c5b1cSMartin Matuska 
27*c03c5b1cSMartin Matuska void ZSTD_ldm_fillHashTable(
28*c03c5b1cSMartin Matuska             ldmState_t* state, const BYTE* ip,
29*c03c5b1cSMartin Matuska             const BYTE* iend, ldmParams_t const* params);
30*c03c5b1cSMartin Matuska 
31*c03c5b1cSMartin Matuska /**
32*c03c5b1cSMartin Matuska  * ZSTD_ldm_generateSequences():
33*c03c5b1cSMartin Matuska  *
34*c03c5b1cSMartin Matuska  * Generates the sequences using the long distance match finder.
35*c03c5b1cSMartin Matuska  * Generates long range matching sequences in `sequences`, which parse a prefix
36*c03c5b1cSMartin Matuska  * of the source. `sequences` must be large enough to store every sequence,
37*c03c5b1cSMartin Matuska  * which can be checked with `ZSTD_ldm_getMaxNbSeq()`.
38*c03c5b1cSMartin Matuska  * @returns 0 or an error code.
39*c03c5b1cSMartin Matuska  *
40*c03c5b1cSMartin Matuska  * NOTE: The user must have called ZSTD_window_update() for all of the input
41*c03c5b1cSMartin Matuska  * they have, even if they pass it to ZSTD_ldm_generateSequences() in chunks.
42*c03c5b1cSMartin Matuska  * NOTE: This function returns an error if it runs out of space to store
43*c03c5b1cSMartin Matuska  *       sequences.
44*c03c5b1cSMartin Matuska  */
45*c03c5b1cSMartin Matuska size_t ZSTD_ldm_generateSequences(
46*c03c5b1cSMartin Matuska             ldmState_t* ldms, rawSeqStore_t* sequences,
47*c03c5b1cSMartin Matuska             ldmParams_t const* params, void const* src, size_t srcSize);
48*c03c5b1cSMartin Matuska 
49*c03c5b1cSMartin Matuska /**
50*c03c5b1cSMartin Matuska  * ZSTD_ldm_blockCompress():
51*c03c5b1cSMartin Matuska  *
52*c03c5b1cSMartin Matuska  * Compresses a block using the predefined sequences, along with a secondary
53*c03c5b1cSMartin Matuska  * block compressor. The literals section of every sequence is passed to the
54*c03c5b1cSMartin Matuska  * secondary block compressor, and those sequences are interspersed with the
55*c03c5b1cSMartin Matuska  * predefined sequences. Returns the length of the last literals.
56*c03c5b1cSMartin Matuska  * Updates `rawSeqStore.pos` to indicate how many sequences have been consumed.
57*c03c5b1cSMartin Matuska  * `rawSeqStore.seq` may also be updated to split the last sequence between two
58*c03c5b1cSMartin Matuska  * blocks.
59*c03c5b1cSMartin Matuska  * @return The length of the last literals.
60*c03c5b1cSMartin Matuska  *
61*c03c5b1cSMartin Matuska  * NOTE: The source must be at most the maximum block size, but the predefined
62*c03c5b1cSMartin Matuska  * sequences can be any size, and may be longer than the block. In the case that
63*c03c5b1cSMartin Matuska  * they are longer than the block, the last sequences may need to be split into
64*c03c5b1cSMartin Matuska  * two. We handle that case correctly, and update `rawSeqStore` appropriately.
65*c03c5b1cSMartin Matuska  * NOTE: This function does not return any errors.
66*c03c5b1cSMartin Matuska  */
67*c03c5b1cSMartin Matuska size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
68*c03c5b1cSMartin Matuska             ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
69*c03c5b1cSMartin Matuska             void const* src, size_t srcSize);
70*c03c5b1cSMartin Matuska 
71*c03c5b1cSMartin Matuska /**
72*c03c5b1cSMartin Matuska  * ZSTD_ldm_skipSequences():
73*c03c5b1cSMartin Matuska  *
74*c03c5b1cSMartin Matuska  * Skip past `srcSize` bytes worth of sequences in `rawSeqStore`.
75*c03c5b1cSMartin Matuska  * Avoids emitting matches less than `minMatch` bytes.
76*c03c5b1cSMartin Matuska  * Must be called for data with is not passed to ZSTD_ldm_blockCompress().
77*c03c5b1cSMartin Matuska  */
78*c03c5b1cSMartin Matuska void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize,
79*c03c5b1cSMartin Matuska     U32 const minMatch);
80*c03c5b1cSMartin Matuska 
81*c03c5b1cSMartin Matuska 
82*c03c5b1cSMartin Matuska /** ZSTD_ldm_getTableSize() :
83*c03c5b1cSMartin Matuska  *  Estimate the space needed for long distance matching tables or 0 if LDM is
84*c03c5b1cSMartin Matuska  *  disabled.
85*c03c5b1cSMartin Matuska  */
86*c03c5b1cSMartin Matuska size_t ZSTD_ldm_getTableSize(ldmParams_t params);
87*c03c5b1cSMartin Matuska 
88*c03c5b1cSMartin Matuska /** ZSTD_ldm_getSeqSpace() :
89*c03c5b1cSMartin Matuska  *  Return an upper bound on the number of sequences that can be produced by
90*c03c5b1cSMartin Matuska  *  the long distance matcher, or 0 if LDM is disabled.
91*c03c5b1cSMartin Matuska  */
92*c03c5b1cSMartin Matuska size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize);
93*c03c5b1cSMartin Matuska 
94*c03c5b1cSMartin Matuska /** ZSTD_ldm_adjustParameters() :
95*c03c5b1cSMartin Matuska  *  If the params->hashRateLog is not set, set it to its default value based on
96*c03c5b1cSMartin Matuska  *  windowLog and params->hashLog.
97*c03c5b1cSMartin Matuska  *
98*c03c5b1cSMartin Matuska  *  Ensures that params->bucketSizeLog is <= params->hashLog (setting it to
99*c03c5b1cSMartin Matuska  *  params->hashLog if it is not).
100*c03c5b1cSMartin Matuska  *
101*c03c5b1cSMartin Matuska  *  Ensures that the minMatchLength >= targetLength during optimal parsing.
102*c03c5b1cSMartin Matuska  */
103*c03c5b1cSMartin Matuska void ZSTD_ldm_adjustParameters(ldmParams_t* params,
104*c03c5b1cSMartin Matuska                                ZSTD_compressionParameters const* cParams);
105*c03c5b1cSMartin Matuska 
106*c03c5b1cSMartin Matuska #if defined (__cplusplus)
107*c03c5b1cSMartin Matuska }
108*c03c5b1cSMartin Matuska #endif
109*c03c5b1cSMartin Matuska 
110*c03c5b1cSMartin Matuska #endif /* ZSTD_FAST_H */
111