xref: /freebsd/sys/contrib/openzfs/module/zstd/lib/compress/zstd_ldm.h (revision 61145dc2b94f12f6a47344fb9aac702321880e43)
1*61145dc2SMartin Matuska // SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-only
2c03c5b1cSMartin Matuska /*
3c03c5b1cSMartin Matuska  * Copyright (c) 2016-2020, Yann Collet, Facebook, Inc.
4c03c5b1cSMartin Matuska  * All rights reserved.
5c03c5b1cSMartin Matuska  *
6c03c5b1cSMartin Matuska  * This source code is licensed under both the BSD-style license (found in the
7c03c5b1cSMartin Matuska  * LICENSE file in the root directory of this source tree) and the GPLv2 (found
8c03c5b1cSMartin Matuska  * in the COPYING file in the root directory of this source tree).
9c03c5b1cSMartin Matuska  * You may select, at your option, one of the above-listed licenses.
10c03c5b1cSMartin Matuska  */
11c03c5b1cSMartin Matuska 
12c03c5b1cSMartin Matuska #ifndef ZSTD_LDM_H
13c03c5b1cSMartin Matuska #define ZSTD_LDM_H
14c03c5b1cSMartin Matuska 
15c03c5b1cSMartin Matuska #if defined (__cplusplus)
16c03c5b1cSMartin Matuska extern "C" {
17c03c5b1cSMartin Matuska #endif
18c03c5b1cSMartin Matuska 
19c03c5b1cSMartin Matuska #include "zstd_compress_internal.h"   /* ldmParams_t, U32 */
20c03c5b1cSMartin Matuska #include "../zstd.h"   /* ZSTD_CCtx, size_t */
21c03c5b1cSMartin Matuska 
22c03c5b1cSMartin Matuska /*-*************************************
23c03c5b1cSMartin Matuska *  Long distance matching
24c03c5b1cSMartin Matuska ***************************************/
25c03c5b1cSMartin Matuska 
26c03c5b1cSMartin Matuska #define ZSTD_LDM_DEFAULT_WINDOW_LOG ZSTD_WINDOWLOG_LIMIT_DEFAULT
27c03c5b1cSMartin Matuska 
28c03c5b1cSMartin Matuska void ZSTD_ldm_fillHashTable(
29c03c5b1cSMartin Matuska             ldmState_t* state, const BYTE* ip,
30c03c5b1cSMartin Matuska             const BYTE* iend, ldmParams_t const* params);
31c03c5b1cSMartin Matuska 
32c03c5b1cSMartin Matuska /**
33c03c5b1cSMartin Matuska  * ZSTD_ldm_generateSequences():
34c03c5b1cSMartin Matuska  *
35c03c5b1cSMartin Matuska  * Generates the sequences using the long distance match finder.
36c03c5b1cSMartin Matuska  * Generates long range matching sequences in `sequences`, which parse a prefix
37c03c5b1cSMartin Matuska  * of the source. `sequences` must be large enough to store every sequence,
38c03c5b1cSMartin Matuska  * which can be checked with `ZSTD_ldm_getMaxNbSeq()`.
39c03c5b1cSMartin Matuska  * @returns 0 or an error code.
40c03c5b1cSMartin Matuska  *
41c03c5b1cSMartin Matuska  * NOTE: The user must have called ZSTD_window_update() for all of the input
42c03c5b1cSMartin Matuska  * they have, even if they pass it to ZSTD_ldm_generateSequences() in chunks.
43c03c5b1cSMartin Matuska  * NOTE: This function returns an error if it runs out of space to store
44c03c5b1cSMartin Matuska  *       sequences.
45c03c5b1cSMartin Matuska  */
46c03c5b1cSMartin Matuska size_t ZSTD_ldm_generateSequences(
47c03c5b1cSMartin Matuska             ldmState_t* ldms, rawSeqStore_t* sequences,
48c03c5b1cSMartin Matuska             ldmParams_t const* params, void const* src, size_t srcSize);
49c03c5b1cSMartin Matuska 
50c03c5b1cSMartin Matuska /**
51c03c5b1cSMartin Matuska  * ZSTD_ldm_blockCompress():
52c03c5b1cSMartin Matuska  *
53c03c5b1cSMartin Matuska  * Compresses a block using the predefined sequences, along with a secondary
54c03c5b1cSMartin Matuska  * block compressor. The literals section of every sequence is passed to the
55c03c5b1cSMartin Matuska  * secondary block compressor, and those sequences are interspersed with the
56c03c5b1cSMartin Matuska  * predefined sequences. Returns the length of the last literals.
57c03c5b1cSMartin Matuska  * Updates `rawSeqStore.pos` to indicate how many sequences have been consumed.
58c03c5b1cSMartin Matuska  * `rawSeqStore.seq` may also be updated to split the last sequence between two
59c03c5b1cSMartin Matuska  * blocks.
60c03c5b1cSMartin Matuska  * @return The length of the last literals.
61c03c5b1cSMartin Matuska  *
62c03c5b1cSMartin Matuska  * NOTE: The source must be at most the maximum block size, but the predefined
63c03c5b1cSMartin Matuska  * sequences can be any size, and may be longer than the block. In the case that
64c03c5b1cSMartin Matuska  * they are longer than the block, the last sequences may need to be split into
65c03c5b1cSMartin Matuska  * two. We handle that case correctly, and update `rawSeqStore` appropriately.
66c03c5b1cSMartin Matuska  * NOTE: This function does not return any errors.
67c03c5b1cSMartin Matuska  */
68c03c5b1cSMartin Matuska size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
69c03c5b1cSMartin Matuska             ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
70c03c5b1cSMartin Matuska             void const* src, size_t srcSize);
71c03c5b1cSMartin Matuska 
72c03c5b1cSMartin Matuska /**
73c03c5b1cSMartin Matuska  * ZSTD_ldm_skipSequences():
74c03c5b1cSMartin Matuska  *
75c03c5b1cSMartin Matuska  * Skip past `srcSize` bytes worth of sequences in `rawSeqStore`.
76c03c5b1cSMartin Matuska  * Avoids emitting matches less than `minMatch` bytes.
77c03c5b1cSMartin Matuska  * Must be called for data with is not passed to ZSTD_ldm_blockCompress().
78c03c5b1cSMartin Matuska  */
79c03c5b1cSMartin Matuska void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize,
80c03c5b1cSMartin Matuska     U32 const minMatch);
81c03c5b1cSMartin Matuska 
82c03c5b1cSMartin Matuska 
83c03c5b1cSMartin Matuska /** ZSTD_ldm_getTableSize() :
84c03c5b1cSMartin Matuska  *  Estimate the space needed for long distance matching tables or 0 if LDM is
85c03c5b1cSMartin Matuska  *  disabled.
86c03c5b1cSMartin Matuska  */
87c03c5b1cSMartin Matuska size_t ZSTD_ldm_getTableSize(ldmParams_t params);
88c03c5b1cSMartin Matuska 
89c03c5b1cSMartin Matuska /** ZSTD_ldm_getSeqSpace() :
90c03c5b1cSMartin Matuska  *  Return an upper bound on the number of sequences that can be produced by
91c03c5b1cSMartin Matuska  *  the long distance matcher, or 0 if LDM is disabled.
92c03c5b1cSMartin Matuska  */
93c03c5b1cSMartin Matuska size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize);
94c03c5b1cSMartin Matuska 
95c03c5b1cSMartin Matuska /** ZSTD_ldm_adjustParameters() :
96c03c5b1cSMartin Matuska  *  If the params->hashRateLog is not set, set it to its default value based on
97c03c5b1cSMartin Matuska  *  windowLog and params->hashLog.
98c03c5b1cSMartin Matuska  *
99c03c5b1cSMartin Matuska  *  Ensures that params->bucketSizeLog is <= params->hashLog (setting it to
100c03c5b1cSMartin Matuska  *  params->hashLog if it is not).
101c03c5b1cSMartin Matuska  *
102c03c5b1cSMartin Matuska  *  Ensures that the minMatchLength >= targetLength during optimal parsing.
103c03c5b1cSMartin Matuska  */
104c03c5b1cSMartin Matuska void ZSTD_ldm_adjustParameters(ldmParams_t* params,
105c03c5b1cSMartin Matuska                                ZSTD_compressionParameters const* cParams);
106c03c5b1cSMartin Matuska 
107c03c5b1cSMartin Matuska #if defined (__cplusplus)
108c03c5b1cSMartin Matuska }
109c03c5b1cSMartin Matuska #endif
110c03c5b1cSMartin Matuska 
111c03c5b1cSMartin Matuska #endif /* ZSTD_FAST_H */
112