ddidderr/zstd-rs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

feat(compress): move match table reduction to Rust

Browse Source 
Port the 16-cell match-table reduction leaf while preserving the window-index threshold, wrapping subtraction, and btlazy2 unsorted marker. Keep ZSTD_reduceIndex and its strategy-specific table selection in C behind the narrow Rust ABI.

Test Plan: cargo test --manifest-path rust/Cargo.toml --no-default-features --features compression (180 passed); root clippy, bench clippy, and test clippy; make -B -C lib -j2 lib; make -B -C tests -j2 test-zstream (84 named tests plus 6,845 and 9,628 fuzz cases passed).
This commit is contained in:
ddidderr
2026-07-18 04:43:26 +02:00
parent 5362112c9c
commit 2cd0b29790
2 changed files with 96 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/compress/zstd_compress.c
+5 -46
View File
@@ -52,6 +52,8 @@ size_t ZSTD_compressStream2_c(ZSTD_CCtx* cctx,
ZSTD_EndDirective endOp);
int ZSTD_rust_simpleCompress2Level(const void* cctx);
int ZSTD_rust_simpleCompressStream2Level(const void* cctx);
void ZSTD_rust_reduceTable(U32* table, U32 size, U32 reducerValue,
int preserveMark);
/* Context-free compression-parameter selection and sizing leaves live in
* Rust (rust/src/zstd_compress_params.rs). This file retains
@@ -2076,63 +2078,20 @@ size_t ZSTD_copyCCtx(ZSTD_CCtx* dstCCtx, const ZSTD_CCtx* srcCCtx, unsigned long
}
#define ZSTD_ROWSIZE 16
/*! ZSTD_reduceTable() :
* reduce table indexes by `reducerValue`, or squash to zero.
* PreserveMark preserves "unsorted mark" for btlazy2 strategy.
* It must be set to a clear 0/1 value, to remove branch during inlining.
* Presume table size is a multiple of ZSTD_ROWSIZE
* Presume table size is a multiple of 16 cells
* to help auto-vectorization */
FORCE_INLINE_TEMPLATE void
ZSTD_reduceTable_internal (U32* const table, U32 const size, U32 const reducerValue, int const preserveMark)
{
int const nbRows = (int)size / ZSTD_ROWSIZE;
int cellNb = 0;
int rowNb;
/* Protect special index values < ZSTD_WINDOW_START_INDEX. */
U32 const reducerThreshold = reducerValue + ZSTD_WINDOW_START_INDEX;
assert((size & (ZSTD_ROWSIZE-1)) == 0); /* multiple of ZSTD_ROWSIZE */
assert(size < (1U<<31)); /* can be cast to int */
#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE)
/* To validate that the table reuse logic is sound, and that we don't
* access table space that we haven't cleaned, we re-"poison" the table
* space every time we mark it dirty.
*
* This function however is intended to operate on those dirty tables and
* re-clean them. So when this function is used correctly, we can unpoison
* the memory it operated on. This introduces a blind spot though, since
* if we now try to operate on __actually__ poisoned memory, we will not
* detect that. */
__msan_unpoison(table, size * sizeof(U32));
#endif
for (rowNb=0 ; rowNb < nbRows ; rowNb++) {
int column;
for (column=0; column<ZSTD_ROWSIZE; column++) {
U32 newVal;
if (preserveMark && table[cellNb] == ZSTD_DUBT_UNSORTED_MARK) {
/* This write is pointless, but is required(?) for the compiler
* to auto-vectorize the loop. */
newVal = ZSTD_DUBT_UNSORTED_MARK;
} else if (table[cellNb] < reducerThreshold) {
newVal = 0;
} else {
newVal = table[cellNb] - reducerValue;
}
table[cellNb] = newVal;
cellNb++;
} }
}
static void ZSTD_reduceTable(U32* const table, U32 const size, U32 const reducerValue)
{
ZSTD_reduceTable_internal(table, size, reducerValue, 0);
ZSTD_rust_reduceTable(table, size, reducerValue, 0);
}
static void ZSTD_reduceTable_btlazy2(U32* const table, U32 const size, U32 const reducerValue)
{
ZSTD_reduceTable_internal(table, size, reducerValue, 1);
ZSTD_rust_reduceTable(table, size, reducerValue, 1);
}
/*! ZSTD_reduceIndex() :