Files
zstd-rs/lib/compress/zstd_ldm.c
T
ddidderr fe7e24c770 feat(rust): migrate high-level runtime paths
Move long-distance matching and high-level decompression from C shims into
Rust. The decoder now owns context, dictionary, parameter, one-shot, and
buffered streaming state while C retains allocation/configuration, legacy,
and trace leaves.

Move CLI parsing, safety policy, and dispatch into a separate Rust static
archive. Keeping it separate prevents library builds from retaining FIO
symbols, while C continues to own file opening, replacement, and I/O.
Program targets now select matching compression/decompression archives.

The remaining C boundary is intentional: high-level compression, optimal
parsing, dictionary building, legacy callbacks, and CLI file I/O still need
migration.

Test Plan:
- cargo test --all-targets (native and i686)
- cargo test --all-targets in rust/cli (native and i686)
- CLI crate compression-only and decompression-only feature tests
- native and i686 fuzzer/zstreamtest runs, plus legacy and dictionary tests
- ZSTD_C_PREDICT and ZSTD_HEAPMODE=0 fuzzer coverage
- library, dynamic-link, and program-target build/round-trip matrix

Refs: rust/README.md
2026-07-11 09:03:41 +02:00

344 lines
12 KiB
C

/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* The LDM algorithms live in rust/src/zstd_ldm.rs. This file retains private
* match-state extraction and block-compressor dispatch only. */
#include "zstd_ldm.h"
#include "../common/debug.h"
#include "zstd_fast.h"
#include "zstd_double_fast.h"
#include "zstd_ldm_geartab.h"
typedef struct {
U32 offset;
U32 checksum;
} ZSTD_rust_ldm_entry_layout;
typedef struct {
U32 offset;
U32 litLength;
U32 matchLength;
} ZSTD_rust_raw_seq_layout;
typedef struct {
rawSeq* seq;
size_t pos;
size_t posInSequence;
size_t size;
size_t capacity;
} ZSTD_rust_raw_seq_store_layout;
typedef struct {
ZSTD_ParamSwitch_e enableLdm;
U32 hashLog;
U32 bucketSizeLog;
U32 minMatchLength;
U32 hashRateLog;
U32 windowLog;
} ZSTD_rust_ldm_params_layout;
typedef struct {
BYTE const* nextSrc;
BYTE const* base;
BYTE const* dictBase;
U32 dictLimit;
U32 lowLimit;
U32 nbOverflowCorrections;
} ZSTD_rust_ldm_window_layout;
typedef char ZSTD_rust_ldm_entry_layout_check[
sizeof(ldmEntry_t) == sizeof(ZSTD_rust_ldm_entry_layout) ? 1 : -1];
typedef char ZSTD_rust_raw_seq_layout_check[
sizeof(rawSeq) == sizeof(ZSTD_rust_raw_seq_layout) ? 1 : -1];
typedef char ZSTD_rust_raw_seq_store_layout_check[
sizeof(RawSeqStore_t) == sizeof(ZSTD_rust_raw_seq_store_layout) ? 1 : -1];
typedef char ZSTD_rust_ldm_params_layout_check[
sizeof(ldmParams_t) == sizeof(ZSTD_rust_ldm_params_layout) ? 1 : -1];
typedef char ZSTD_rust_ldm_window_layout_check[
sizeof(ZSTD_window_t) == sizeof(ZSTD_rust_ldm_window_layout) ? 1 : -1];
typedef char ZSTD_rust_param_switch_layout_check[
sizeof(ZSTD_ParamSwitch_e) == sizeof(int) ? 1 : -1];
#define ZSTD_RUST_LDM_OFFSET_CHECK(name, c_type, c_field, rust_type, rust_field) \
typedef char name[offsetof(c_type, c_field) == offsetof(rust_type, rust_field) ? 1 : -1]
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_entry_offset_check,
ldmEntry_t, offset,
ZSTD_rust_ldm_entry_layout, offset);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_entry_checksum_check,
ldmEntry_t, checksum,
ZSTD_rust_ldm_entry_layout, checksum);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_raw_seq_offset_check,
rawSeq, offset,
ZSTD_rust_raw_seq_layout, offset);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_raw_seq_lit_length_check,
rawSeq, litLength,
ZSTD_rust_raw_seq_layout, litLength);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_raw_seq_match_length_check,
rawSeq, matchLength,
ZSTD_rust_raw_seq_layout, matchLength);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_raw_seq_store_seq_check,
RawSeqStore_t, seq,
ZSTD_rust_raw_seq_store_layout, seq);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_raw_seq_store_pos_check,
RawSeqStore_t, pos,
ZSTD_rust_raw_seq_store_layout, pos);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_raw_seq_store_pos_in_sequence_check,
RawSeqStore_t, posInSequence,
ZSTD_rust_raw_seq_store_layout, posInSequence);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_raw_seq_store_size_check,
RawSeqStore_t, size,
ZSTD_rust_raw_seq_store_layout, size);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_raw_seq_store_capacity_check,
RawSeqStore_t, capacity,
ZSTD_rust_raw_seq_store_layout, capacity);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_params_enable_check,
ldmParams_t, enableLdm,
ZSTD_rust_ldm_params_layout, enableLdm);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_params_hash_log_check,
ldmParams_t, hashLog,
ZSTD_rust_ldm_params_layout, hashLog);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_params_bucket_size_log_check,
ldmParams_t, bucketSizeLog,
ZSTD_rust_ldm_params_layout, bucketSizeLog);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_params_min_match_length_check,
ldmParams_t, minMatchLength,
ZSTD_rust_ldm_params_layout, minMatchLength);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_params_hash_rate_log_check,
ldmParams_t, hashRateLog,
ZSTD_rust_ldm_params_layout, hashRateLog);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_params_window_log_check,
ldmParams_t, windowLog,
ZSTD_rust_ldm_params_layout, windowLog);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_window_next_src_check,
ZSTD_window_t, nextSrc,
ZSTD_rust_ldm_window_layout, nextSrc);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_window_base_check,
ZSTD_window_t, base,
ZSTD_rust_ldm_window_layout, base);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_window_dict_base_check,
ZSTD_window_t, dictBase,
ZSTD_rust_ldm_window_layout, dictBase);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_window_dict_limit_check,
ZSTD_window_t, dictLimit,
ZSTD_rust_ldm_window_layout, dictLimit);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_window_low_limit_check,
ZSTD_window_t, lowLimit,
ZSTD_rust_ldm_window_layout, lowLimit);
ZSTD_RUST_LDM_OFFSET_CHECK(
ZSTD_rust_ldm_window_overflow_corrections_check,
ZSTD_window_t, nbOverflowCorrections,
ZSTD_rust_ldm_window_layout, nbOverflowCorrections);
#undef ZSTD_RUST_LDM_OFFSET_CHECK
typedef char ZSTD_rust_ldm_rep_count[(ZSTD_REP_NUM == 3) ? 1 : -1];
void ZSTD_rust_ldm_adjustParameters(
void* params, U32 windowLog, int strategy,
U32 hashLogMax, U32 bucketSizeLogMax, int btultra);
size_t ZSTD_rust_ldm_getTableSize(
const void* params, int enableLdm, size_t redzoneSize);
size_t ZSTD_rust_ldm_getMaxNbSeq(
const void* params, int enableLdm, size_t maxChunkSize);
void ZSTD_rust_ldm_fillHashTable(
void* hashTable, BYTE* bucketOffsets, const BYTE* base,
const BYTE* ip, const BYTE* iend, const void* params);
size_t ZSTD_rust_ldm_generateSequences(
void* hashTable, BYTE* bucketOffsets, void* window, U32* loadedDictEnd,
void* rawSeqStore, const void* params,
const void* src, size_t srcSize, int overflowCorrectFrequently);
void ZSTD_rust_ldm_skipSequences(void* rawSeqStore, size_t srcSize, U32 minMatch);
void ZSTD_rust_ldm_skipRawSeqStoreBytes(void* rawSeqStore, size_t nbBytes);
size_t ZSTD_rust_ldm_blockCompress(
void* rawSeqStore, void* blockContext, void* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize, U32 minMatch, int useOptimalParser);
const U64* ZSTD_ldm_rust_gearTable(void);
void ZSTD_ldm_rust_prepareBlock(void* blockContext, const void* anchor);
size_t ZSTD_ldm_rust_compressLiterals(
void* blockContext, void* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize);
void ZSTD_ldm_rust_storeSeq(
void* seqStore, size_t litLength, const void* literals, const void* litLimit,
U32 offBase, size_t matchLength);
void ZSTD_ldm_rust_setLdmSeqStore(void* blockContext, const void* rawSeqStore);
const U64* ZSTD_ldm_rust_gearTable(void)
{
return ZSTD_ldm_gearTab;
}
typedef struct {
ZSTD_MatchState_t* ms;
ZSTD_BlockCompressor_f blockCompressor;
} ZSTD_rust_ldm_block_context;
static void ZSTD_rust_ldm_limitTableUpdate(ZSTD_MatchState_t* ms, const BYTE* anchor)
{
U32 const curr = (U32)(anchor - ms->window.base);
if (curr > ms->nextToUpdate + 1024) {
ms->nextToUpdate = curr - MIN(512, curr - ms->nextToUpdate - 1024);
}
}
static void ZSTD_rust_ldm_fillFastTables(ZSTD_MatchState_t* ms, const BYTE* end)
{
switch (ms->cParams.strategy) {
case ZSTD_fast:
ZSTD_fillHashTable(ms, end, ZSTD_dtlm_fast, ZSTD_tfp_forCCtx);
break;
case ZSTD_dfast:
#ifndef ZSTD_EXCLUDE_DFAST_BLOCK_COMPRESSOR
ZSTD_fillDoubleHashTable(ms, end, ZSTD_dtlm_fast, ZSTD_tfp_forCCtx);
#else
assert(0);
#endif
break;
case ZSTD_greedy:
case ZSTD_lazy:
case ZSTD_lazy2:
case ZSTD_btlazy2:
case ZSTD_btopt:
case ZSTD_btultra:
case ZSTD_btultra2:
break;
default:
assert(0);
}
}
void ZSTD_ldm_rust_prepareBlock(void* blockContext, const void* anchor)
{
ZSTD_rust_ldm_block_context* const context = (ZSTD_rust_ldm_block_context*)blockContext;
ZSTD_rust_ldm_limitTableUpdate(context->ms, (const BYTE*)anchor);
ZSTD_rust_ldm_fillFastTables(context->ms, (const BYTE*)anchor);
}
size_t ZSTD_ldm_rust_compressLiterals(
void* blockContext, void* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize)
{
ZSTD_rust_ldm_block_context* const context = (ZSTD_rust_ldm_block_context*)blockContext;
return context->blockCompressor(context->ms, (SeqStore_t*)seqStore, rep, src, srcSize);
}
void ZSTD_ldm_rust_storeSeq(
void* seqStore, size_t litLength, const void* literals, const void* litLimit,
U32 offBase, size_t matchLength)
{
ZSTD_storeSeq((SeqStore_t*)seqStore, litLength, (const BYTE*)literals,
(const BYTE*)litLimit, offBase, matchLength);
}
void ZSTD_ldm_rust_setLdmSeqStore(void* blockContext, const void* rawSeqStore)
{
ZSTD_rust_ldm_block_context* const context = (ZSTD_rust_ldm_block_context*)blockContext;
context->ms->ldmSeqStore = (const RawSeqStore_t*)rawSeqStore;
}
void ZSTD_ldm_adjustParameters(ldmParams_t* params,
const ZSTD_compressionParameters* cParams)
{
ZSTD_rust_ldm_adjustParameters(
params, cParams->windowLog, (int)cParams->strategy,
ZSTD_HASHLOG_MAX, ZSTD_LDM_BUCKETSIZELOG_MAX, (int)ZSTD_btultra);
}
size_t ZSTD_ldm_getTableSize(ldmParams_t params)
{
#if ZSTD_ADDRESS_SANITIZER && !defined(ZSTD_ASAN_DONT_POISON_WORKSPACE)
size_t const redzoneSize = ZSTD_CWKSP_ASAN_REDZONE_SIZE;
#else
size_t const redzoneSize = 0;
#endif
return ZSTD_rust_ldm_getTableSize(
&params, params.enableLdm == ZSTD_ps_enable, redzoneSize);
}
size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize)
{
return ZSTD_rust_ldm_getMaxNbSeq(
&params, params.enableLdm == ZSTD_ps_enable, maxChunkSize);
}
void ZSTD_ldm_fillHashTable(
ldmState_t* ldmState, const BYTE* ip,
const BYTE* iend, ldmParams_t const* params)
{
ZSTD_rust_ldm_fillHashTable(
ldmState->hashTable, ldmState->bucketOffsets, ldmState->window.base,
ip, iend, params);
}
size_t ZSTD_ldm_generateSequences(
ldmState_t* ldmState, RawSeqStore_t* sequences,
ldmParams_t const* params, void const* src, size_t srcSize)
{
assert(ldmState->window.nextSrc >= (const BYTE*)src + srcSize);
assert(sequences->pos <= sequences->size);
assert(sequences->size <= sequences->capacity);
return ZSTD_rust_ldm_generateSequences(
ldmState->hashTable, ldmState->bucketOffsets, &ldmState->window,
&ldmState->loadedDictEnd, sequences, params, src, srcSize,
ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY);
}
void ZSTD_ldm_skipSequences(RawSeqStore_t* rawSeqStore, size_t srcSize,
U32 const minMatch)
{
ZSTD_rust_ldm_skipSequences(rawSeqStore, srcSize, minMatch);
}
void ZSTD_ldm_skipRawSeqStoreBytes(RawSeqStore_t* rawSeqStore, size_t nbBytes)
{
ZSTD_rust_ldm_skipRawSeqStoreBytes(rawSeqStore, nbBytes);
}
size_t ZSTD_ldm_blockCompress(RawSeqStore_t* rawSeqStore,
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_ParamSwitch_e useRowMatchFinder,
void const* src, size_t srcSize)
{
ZSTD_rust_ldm_block_context context;
context.ms = ms;
context.blockCompressor = ZSTD_selectBlockCompressor(
ms->cParams.strategy, useRowMatchFinder, ZSTD_matchState_dictMode(ms));
assert(context.blockCompressor != NULL);
return ZSTD_rust_ldm_blockCompress(
rawSeqStore, &context, seqStore, rep, src, srcSize,
ms->cParams.minMatch, ms->cParams.strategy >= ZSTD_btopt);
}