xref: /dflybsd-src/contrib/zstd/lib/compress/zstd_ldm.h (revision a28cd43d19e8b720a6c852a4bbc5ae147a26165a)
1a28cd43dSSascha Wildner /*
2a28cd43dSSascha Wildner  * Copyright (c) 2016-2020, Yann Collet, Facebook, Inc.
3a28cd43dSSascha Wildner  * All rights reserved.
4a28cd43dSSascha Wildner  *
5a28cd43dSSascha Wildner  * This source code is licensed under both the BSD-style license (found in the
6a28cd43dSSascha Wildner  * LICENSE file in the root directory of this source tree) and the GPLv2 (found
7a28cd43dSSascha Wildner  * in the COPYING file in the root directory of this source tree).
8a28cd43dSSascha Wildner  * You may select, at your option, one of the above-listed licenses.
9a28cd43dSSascha Wildner  */
10a28cd43dSSascha Wildner 
11a28cd43dSSascha Wildner #ifndef ZSTD_LDM_H
12a28cd43dSSascha Wildner #define ZSTD_LDM_H
13a28cd43dSSascha Wildner 
14a28cd43dSSascha Wildner #if defined (__cplusplus)
15a28cd43dSSascha Wildner extern "C" {
16a28cd43dSSascha Wildner #endif
17a28cd43dSSascha Wildner 
18a28cd43dSSascha Wildner #include "zstd_compress_internal.h"   /* ldmParams_t, U32 */
19a28cd43dSSascha Wildner #include "../zstd.h"   /* ZSTD_CCtx, size_t */
20a28cd43dSSascha Wildner 
21a28cd43dSSascha Wildner /*-*************************************
22a28cd43dSSascha Wildner *  Long distance matching
23a28cd43dSSascha Wildner ***************************************/
24a28cd43dSSascha Wildner 
25a28cd43dSSascha Wildner #define ZSTD_LDM_DEFAULT_WINDOW_LOG ZSTD_WINDOWLOG_LIMIT_DEFAULT
26a28cd43dSSascha Wildner 
27a28cd43dSSascha Wildner void ZSTD_ldm_fillHashTable(
28a28cd43dSSascha Wildner             ldmState_t* state, const BYTE* ip,
29a28cd43dSSascha Wildner             const BYTE* iend, ldmParams_t const* params);
30a28cd43dSSascha Wildner 
31a28cd43dSSascha Wildner /**
32a28cd43dSSascha Wildner  * ZSTD_ldm_generateSequences():
33a28cd43dSSascha Wildner  *
34a28cd43dSSascha Wildner  * Generates the sequences using the long distance match finder.
35a28cd43dSSascha Wildner  * Generates long range matching sequences in `sequences`, which parse a prefix
36a28cd43dSSascha Wildner  * of the source. `sequences` must be large enough to store every sequence,
37a28cd43dSSascha Wildner  * which can be checked with `ZSTD_ldm_getMaxNbSeq()`.
38a28cd43dSSascha Wildner  * @returns 0 or an error code.
39a28cd43dSSascha Wildner  *
40a28cd43dSSascha Wildner  * NOTE: The user must have called ZSTD_window_update() for all of the input
41a28cd43dSSascha Wildner  * they have, even if they pass it to ZSTD_ldm_generateSequences() in chunks.
42a28cd43dSSascha Wildner  * NOTE: This function returns an error if it runs out of space to store
43a28cd43dSSascha Wildner  *       sequences.
44a28cd43dSSascha Wildner  */
45a28cd43dSSascha Wildner size_t ZSTD_ldm_generateSequences(
46a28cd43dSSascha Wildner             ldmState_t* ldms, rawSeqStore_t* sequences,
47a28cd43dSSascha Wildner             ldmParams_t const* params, void const* src, size_t srcSize);
48a28cd43dSSascha Wildner 
49a28cd43dSSascha Wildner /**
50a28cd43dSSascha Wildner  * ZSTD_ldm_blockCompress():
51a28cd43dSSascha Wildner  *
52a28cd43dSSascha Wildner  * Compresses a block using the predefined sequences, along with a secondary
53a28cd43dSSascha Wildner  * block compressor. The literals section of every sequence is passed to the
54a28cd43dSSascha Wildner  * secondary block compressor, and those sequences are interspersed with the
55a28cd43dSSascha Wildner  * predefined sequences. Returns the length of the last literals.
56a28cd43dSSascha Wildner  * Updates `rawSeqStore.pos` to indicate how many sequences have been consumed.
57a28cd43dSSascha Wildner  * `rawSeqStore.seq` may also be updated to split the last sequence between two
58a28cd43dSSascha Wildner  * blocks.
59a28cd43dSSascha Wildner  * @return The length of the last literals.
60a28cd43dSSascha Wildner  *
61a28cd43dSSascha Wildner  * NOTE: The source must be at most the maximum block size, but the predefined
62a28cd43dSSascha Wildner  * sequences can be any size, and may be longer than the block. In the case that
63a28cd43dSSascha Wildner  * they are longer than the block, the last sequences may need to be split into
64a28cd43dSSascha Wildner  * two. We handle that case correctly, and update `rawSeqStore` appropriately.
65a28cd43dSSascha Wildner  * NOTE: This function does not return any errors.
66a28cd43dSSascha Wildner  */
67a28cd43dSSascha Wildner size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
68a28cd43dSSascha Wildner             ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
69a28cd43dSSascha Wildner             void const* src, size_t srcSize);
70a28cd43dSSascha Wildner 
71a28cd43dSSascha Wildner /**
72a28cd43dSSascha Wildner  * ZSTD_ldm_skipSequences():
73a28cd43dSSascha Wildner  *
74a28cd43dSSascha Wildner  * Skip past `srcSize` bytes worth of sequences in `rawSeqStore`.
75a28cd43dSSascha Wildner  * Avoids emitting matches less than `minMatch` bytes.
76a28cd43dSSascha Wildner  * Must be called for data with is not passed to ZSTD_ldm_blockCompress().
77a28cd43dSSascha Wildner  */
78a28cd43dSSascha Wildner void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize,
79a28cd43dSSascha Wildner     U32 const minMatch);
80a28cd43dSSascha Wildner 
81a28cd43dSSascha Wildner /* ZSTD_ldm_skipRawSeqStoreBytes():
82a28cd43dSSascha Wildner  * Moves forward in rawSeqStore by nbBytes, updating fields 'pos' and 'posInSequence'.
83a28cd43dSSascha Wildner  * Not to be used in conjunction with ZSTD_ldm_skipSequences().
84a28cd43dSSascha Wildner  * Must be called for data with is not passed to ZSTD_ldm_blockCompress().
85a28cd43dSSascha Wildner  */
86a28cd43dSSascha Wildner void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes);
87a28cd43dSSascha Wildner 
88a28cd43dSSascha Wildner /** ZSTD_ldm_getTableSize() :
89a28cd43dSSascha Wildner  *  Estimate the space needed for long distance matching tables or 0 if LDM is
90a28cd43dSSascha Wildner  *  disabled.
91a28cd43dSSascha Wildner  */
92a28cd43dSSascha Wildner size_t ZSTD_ldm_getTableSize(ldmParams_t params);
93a28cd43dSSascha Wildner 
94a28cd43dSSascha Wildner /** ZSTD_ldm_getSeqSpace() :
95a28cd43dSSascha Wildner  *  Return an upper bound on the number of sequences that can be produced by
96a28cd43dSSascha Wildner  *  the long distance matcher, or 0 if LDM is disabled.
97a28cd43dSSascha Wildner  */
98a28cd43dSSascha Wildner size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize);
99a28cd43dSSascha Wildner 
100a28cd43dSSascha Wildner /** ZSTD_ldm_adjustParameters() :
101a28cd43dSSascha Wildner  *  If the params->hashRateLog is not set, set it to its default value based on
102a28cd43dSSascha Wildner  *  windowLog and params->hashLog.
103a28cd43dSSascha Wildner  *
104a28cd43dSSascha Wildner  *  Ensures that params->bucketSizeLog is <= params->hashLog (setting it to
105a28cd43dSSascha Wildner  *  params->hashLog if it is not).
106a28cd43dSSascha Wildner  *
107a28cd43dSSascha Wildner  *  Ensures that the minMatchLength >= targetLength during optimal parsing.
108a28cd43dSSascha Wildner  */
109a28cd43dSSascha Wildner void ZSTD_ldm_adjustParameters(ldmParams_t* params,
110a28cd43dSSascha Wildner                                ZSTD_compressionParameters const* cParams);
111a28cd43dSSascha Wildner 
112a28cd43dSSascha Wildner #if defined (__cplusplus)
113a28cd43dSSascha Wildner }
114a28cd43dSSascha Wildner #endif
115a28cd43dSSascha Wildner 
116a28cd43dSSascha Wildner #endif /* ZSTD_FAST_H */
117