refactor(compress): move LDM fast-table policy to Rust

Move the strategy classification that decides whether LDM preparation fills
fast or double-fast tables into Rust.  C retains the private match-state,
configuration-specific table-fill leaves, and the existing excluded-DFAST
assertion.  Add enum-value ABI checks plus focused tests for valid and invalid
strategy classifications.

Test Plan:
- `ulimit -v 41943040; cargo +nightly fmt --manifest-path rust/Cargo.toml --all -- --check`
- `ulimit -v 41943040; CARGO_BUILD_JOBS=1 cargo clippy --manifest-path rust/Cargo.toml --all-targets -- -D warnings`
- `ulimit -v 41943040; make -j1`
- `ulimit -v 41943040; make -j1 -C tests test`
- `ulimit -v 41943040; CARGO_BUILD_JOBS=1 cargo clippy --manifest-path rust/cli/Cargo.toml --all-targets -- -D warnings`
- `ulimit -v 41943040; CARGO_BUILD_JOBS=1 cargo test --manifest-path rust/cli/Cargo.toml --all-targets`

The integrated checks ran at the combined working-tree tip under a serial
40 GiB virtual-memory limit. Standalone root Rust unit linking remains
unavailable because the crate imports C-owned bridge symbols without a Cargo
build/link setup.
This commit is contained in:
2026-07-20 09:49:57 +02:00
parent 4409ef1e16
commit 7ea0cbd96e
2 changed files with 61 additions and 10 deletions
+14 -10
View File
@@ -248,6 +248,8 @@ ZSTD_RUST_LDM_OFFSET_CHECK(
#undef ZSTD_RUST_LDM_OFFSET_CHECK
typedef char ZSTD_rust_ldm_rep_count[(ZSTD_REP_NUM == 3) ? 1 : -1];
typedef char ZSTD_rust_ldm_strategy_values_check[
(ZSTD_fast == 1 && ZSTD_dfast == 2 && ZSTD_btultra2 == 9) ? 1 : -1];
void ZSTD_rust_ldm_adjustParameters(
void* params, U32 windowLog, int strategy,
@@ -270,6 +272,14 @@ size_t ZSTD_rust_ldm_blockCompress(
const void* src, size_t srcSize, U32 minMatch, int strategy);
U32 ZSTD_rust_ldm_limitTableUpdate(U32 curr, U32 nextToUpdate);
enum {
ZSTD_RUST_LDM_FAST_TABLES_NONE = 0,
ZSTD_RUST_LDM_FAST_TABLES_FAST = 1,
ZSTD_RUST_LDM_FAST_TABLES_DFAST = 2
};
int ZSTD_rust_ldm_fastTableKind(int strategy);
const U64* ZSTD_ldm_rust_gearTable(void);
void ZSTD_ldm_rust_prepareBlock(void* blockContext, const void* anchor);
size_t ZSTD_ldm_rust_compressLiterals(
@@ -289,24 +299,18 @@ typedef struct {
static void ZSTD_rust_ldm_fillFastTables(ZSTD_MatchState_t* ms, const BYTE* end)
{
switch (ms->cParams.strategy) {
case ZSTD_fast:
switch (ZSTD_rust_ldm_fastTableKind((int)ms->cParams.strategy)) {
case ZSTD_RUST_LDM_FAST_TABLES_FAST:
ZSTD_fillHashTable(ms, end, ZSTD_dtlm_fast, ZSTD_tfp_forCCtx);
break;
case ZSTD_dfast:
case ZSTD_RUST_LDM_FAST_TABLES_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:
case ZSTD_RUST_LDM_FAST_TABLES_NONE:
break;
default:
assert(0);
+47
View File
@@ -23,8 +23,16 @@ const LDM_MIN_MATCH_LENGTH: u32 = 64;
const HASH_READ_SIZE: usize = 8;
const ZSTD_REP_NUM: usize = 3;
const ZSTD_WINDOW_START_INDEX: u32 = 2;
const ZSTD_FAST_STRATEGY: c_int = 1;
const ZSTD_DFAST_STRATEGY: c_int = 2;
const ZSTD_STRATEGY_MAX: c_int = 9;
const ZSTD_BTOPT: c_int = 7;
const LDM_FAST_TABLES_NONE: c_int = 0;
const LDM_FAST_TABLES_FAST: c_int = 1;
const LDM_FAST_TABLES_DFAST: c_int = 2;
const LDM_FAST_TABLES_INVALID: c_int = -1;
#[repr(C)]
#[derive(Clone, Copy)]
struct LdmEntry {
@@ -196,6 +204,20 @@ pub extern "C" fn ZSTD_rust_ldm_limitTableUpdate(curr: u32, next_to_update: u32)
}
}
/// Classify the C-owned table update needed before an LDM literal segment.
///
/// The table fill itself stays in C because it owns the private match state;
/// Rust owns only this deterministic strategy policy.
#[no_mangle]
pub extern "C" fn ZSTD_rust_ldm_fastTableKind(strategy: c_int) -> c_int {
match strategy {
ZSTD_FAST_STRATEGY => LDM_FAST_TABLES_FAST,
ZSTD_DFAST_STRATEGY => LDM_FAST_TABLES_DFAST,
3..=ZSTD_STRATEGY_MAX => LDM_FAST_TABLES_NONE,
_ => LDM_FAST_TABLES_INVALID,
}
}
#[inline]
unsafe fn ldm_bucket(hash_table: *mut LdmEntry, hash: u32, bucket_size_log: u32) -> *mut LdmEntry {
unsafe { hash_table.add((hash as usize) << bucket_size_log) }
@@ -1075,6 +1097,31 @@ mod tests {
assert!(use_optimal_parser(9));
}
#[test]
fn fast_table_kind_preserves_c_strategy_dispatch() {
assert_eq!(
ZSTD_rust_ldm_fastTableKind(ZSTD_FAST_STRATEGY),
LDM_FAST_TABLES_FAST
);
assert_eq!(
ZSTD_rust_ldm_fastTableKind(ZSTD_DFAST_STRATEGY),
LDM_FAST_TABLES_DFAST
);
for strategy in 3..=ZSTD_STRATEGY_MAX {
assert_eq!(ZSTD_rust_ldm_fastTableKind(strategy), LDM_FAST_TABLES_NONE);
}
}
#[test]
fn fast_table_kind_rejects_invalid_strategy_values() {
assert_eq!(ZSTD_rust_ldm_fastTableKind(0), LDM_FAST_TABLES_INVALID);
assert_eq!(
ZSTD_rust_ldm_fastTableKind(ZSTD_STRATEGY_MAX + 1),
LDM_FAST_TABLES_INVALID
);
assert_eq!(ZSTD_rust_ldm_fastTableKind(-1), LDM_FAST_TABLES_INVALID);
}
#[test]
fn raw_sequence_skipping_merges_short_tail_matches() {
let mut sequences = [