zhangengru/tracy
1
0
Fork 0
mirror of https://github.com/wolfpld/tracy.git synced 2025-03-20 07:40:02 +08:00
Code Issues Projects Wiki Activity

Update zstd to 1.5.4.

Browse Source
This commit is contained in:
Bartosz Taudul
2023-02-10 20:25:40 +01:00
parent 69d0510255
commit 0d6fb703b4
66 changed files with 4541 additions and 3026 deletions
Download Patch File Download Diff File Expand all files Collapse all files
zstd/common/bits.h
+175
View File
@@ -0,0 +1,175 @@
/*
* 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.
*/
#ifndef ZSTD_BITS_H
#define ZSTD_BITS_H
#include "mem.h"
MEM_STATIC unsigned ZSTD_countTrailingZeros32_fallback(U32 val)
{
assert(val != 0);
{
static const int DeBruijnBytePos[32] = {0, 1, 28, 2, 29, 14, 24, 3,
30, 22, 20, 15, 25, 17, 4, 8,
31, 27, 13, 23, 21, 19, 16, 7,
26, 12, 18, 6, 11, 5, 10, 9};
return DeBruijnBytePos[((U32) ((val & -(S32) val) * 0x077CB531U)) >> 27];
}
}
MEM_STATIC unsigned ZSTD_countTrailingZeros32(U32 val)
{
assert(val != 0);
# if defined(_MSC_VER)
# if STATIC_BMI2 == 1
return _tzcnt_u32(val);
# else
if (val != 0) {
unsigned long r;
_BitScanForward(&r, val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4)
return (unsigned)__builtin_ctz(val);
# else
return ZSTD_countTrailingZeros32_fallback(val);
# endif
}
MEM_STATIC unsigned ZSTD_countLeadingZeros32_fallback(U32 val) {
assert(val != 0);
{
static const U32 DeBruijnClz[32] = {0, 9, 1, 10, 13, 21, 2, 29,
11, 14, 16, 18, 22, 25, 3, 30,
8, 12, 20, 28, 15, 17, 24, 7,
19, 27, 23, 6, 26, 5, 4, 31};
val |= val >> 1;
val |= val >> 2;
val |= val >> 4;
val |= val >> 8;
val |= val >> 16;
return 31 - DeBruijnClz[(val * 0x07C4ACDDU) >> 27];
}
}
MEM_STATIC unsigned ZSTD_countLeadingZeros32(U32 val)
{
assert(val != 0);
# if defined(_MSC_VER)
# if STATIC_BMI2 == 1
return _lzcnt_u32(val);
# else
if (val != 0) {
unsigned long r;
_BitScanReverse(&r, val);
return (unsigned)(31 - r);
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4)
return (unsigned)__builtin_clz(val);
# else
return ZSTD_countLeadingZeros32_fallback(val);
# endif
}
MEM_STATIC unsigned ZSTD_countTrailingZeros64(U64 val)
{
assert(val != 0);
# if defined(_MSC_VER) && defined(_WIN64)
# if STATIC_BMI2 == 1
return _tzcnt_u64(val);
# else
if (val != 0) {
unsigned long r;
_BitScanForward64(&r, val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4) && defined(__LP64__)
return (unsigned)__builtin_ctzll(val);
# else
{
U32 mostSignificantWord = (U32)(val >> 32);
U32 leastSignificantWord = (U32)val;
if (leastSignificantWord == 0) {
return 32 + ZSTD_countTrailingZeros32(mostSignificantWord);
} else {
return ZSTD_countTrailingZeros32(leastSignificantWord);
}
}
# endif
}
MEM_STATIC unsigned ZSTD_countLeadingZeros64(U64 val)
{
assert(val != 0);
# if defined(_MSC_VER) && defined(_WIN64)
# if STATIC_BMI2 == 1
return _lzcnt_u64(val);
# else
if (val != 0) {
unsigned long r;
_BitScanReverse64(&r, val);
return (unsigned)(63 - r);
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4)
return (unsigned)(__builtin_clzll(val));
# else
{
U32 mostSignificantWord = (U32)(val >> 32);
U32 leastSignificantWord = (U32)val;
if (mostSignificantWord == 0) {
return 32 + ZSTD_countLeadingZeros32(leastSignificantWord);
} else {
return ZSTD_countLeadingZeros32(mostSignificantWord);
}
}
# endif
}
MEM_STATIC unsigned ZSTD_NbCommonBytes(size_t val)
{
if (MEM_isLittleEndian()) {
if (MEM_64bits()) {
return ZSTD_countTrailingZeros64((U64)val) >> 3;
} else {
return ZSTD_countTrailingZeros32((U32)val) >> 3;
}
} else { /* Big Endian CPU */
if (MEM_64bits()) {
return ZSTD_countLeadingZeros64((U64)val) >> 3;
} else {
return ZSTD_countLeadingZeros32((U32)val) >> 3;
}
}
}
MEM_STATIC unsigned ZSTD_highbit32(U32 val) /* compress, dictBuilder, decodeCorpus */
{
assert(val != 0);
return 31 - ZSTD_countLeadingZeros32(val);
}
#endif /* ZSTD_BITS_H */
zstd/common/bitstream.h
+18 -59
View File
@@ -1,7 +1,7 @@
/* ******************************************************************
* bitstream
* Part of FSE library
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
@@ -30,14 +30,15 @@ extern "C" {
#include "compiler.h" /* UNLIKELY() */
#include "debug.h" /* assert(), DEBUGLOG(), RAWLOG() */
#include "error_private.h" /* error codes and messages */
#include "bits.h" /* ZSTD_highbit32 */
/*=========================================
* Target specific
=========================================*/
#ifndef ZSTD_NO_INTRINSICS
# if defined(__BMI__) && defined(__GNUC__)
# include <immintrin.h> /* support for bextr (experimental) */
# if (defined(__BMI__) || defined(__BMI2__)) && defined(__GNUC__)
# include <immintrin.h> /* support for bextr (experimental)/bzhi */
# elif defined(__ICCARM__)
# include <intrinsics.h>
# endif
@@ -132,48 +133,6 @@ MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC);
MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits);
/* faster, but works only if nbBits >= 1 */
/*-**************************************************************
* Internal functions
****************************************************************/
MEM_STATIC unsigned BIT_highbit32 (U32 val)
{
assert(val != 0);
{
# if defined(_MSC_VER) /* Visual */
# if STATIC_BMI2 == 1
return _lzcnt_u32(val) ^ 31;
# else
if (val != 0) {
unsigned long r;
_BitScanReverse(&r, val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 3) /* Use GCC Intrinsic */
return __builtin_clz (val) ^ 31;
# elif defined(__ICCARM__) /* IAR Intrinsic */
return 31 - __CLZ(val);
# else /* Software version */
static const unsigned DeBruijnClz[32] = { 0, 9, 1, 10, 13, 21, 2, 29,
11, 14, 16, 18, 22, 25, 3, 30,
8, 12, 20, 28, 15, 17, 24, 7,
19, 27, 23, 6, 26, 5, 4, 31 };
U32 v = val;
v |= v >> 1;
v |= v >> 2;
v |= v >> 4;
v |= v >> 8;
v |= v >> 16;
return DeBruijnClz[ (U32) (v * 0x07C4ACDDU) >> 27];
# endif
}
}
/*===== Local Constants =====*/
static const unsigned BIT_mask[] = {
0, 1, 3, 7, 0xF, 0x1F,
@@ -203,6 +162,16 @@ MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC,
return 0;
}
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
{
#if defined(STATIC_BMI2) && STATIC_BMI2 == 1 && !defined(ZSTD_NO_INTRINSICS)
return _bzhi_u64(bitContainer, nbBits);
#else
assert(nbBits < BIT_MASK_SIZE);
return bitContainer & BIT_mask[nbBits];
#endif
}
/*! BIT_addBits() :
* can add up to 31 bits into `bitC`.
* Note : does not check for register overflow ! */
@@ -212,7 +181,7 @@ MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC,
DEBUG_STATIC_ASSERT(BIT_MASK_SIZE == 32);
assert(nbBits < BIT_MASK_SIZE);
assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8);
bitC->bitContainer |= (value & BIT_mask[nbBits]) << bitC->bitPos;
bitC->bitContainer |= BIT_getLowerBits(value, nbBits) << bitC->bitPos;
bitC->bitPos += nbBits;
}
@@ -291,7 +260,7 @@ MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, si
bitD->ptr = (const char*)srcBuffer + srcSize - sizeof(bitD->bitContainer);
bitD->bitContainer = MEM_readLEST(bitD->ptr);
{ BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1];
bitD->bitsConsumed = lastByte ? 8 - BIT_highbit32(lastByte) : 0; /* ensures bitsConsumed is always set */
bitD->bitsConsumed = lastByte ? 8 - ZSTD_highbit32(lastByte) : 0; /* ensures bitsConsumed is always set */
if (lastByte == 0) return ERROR(GENERIC); /* endMark not present */ }
} else {
bitD->ptr = bitD->start;
@@ -319,7 +288,7 @@ MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, si
default: break;
}
{ BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1];
bitD->bitsConsumed = lastByte ? 8 - BIT_highbit32(lastByte) : 0;
bitD->bitsConsumed = lastByte ? 8 - ZSTD_highbit32(lastByte) : 0;
if (lastByte == 0) return ERROR(corruption_detected); /* endMark not present */
}
bitD->bitsConsumed += (U32)(sizeof(bitD->bitContainer) - srcSize)*8;
@@ -350,16 +319,6 @@ MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getMiddleBits(size_t bitContainer, U32 c
#endif
}
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
{
#if defined(STATIC_BMI2) && STATIC_BMI2 == 1
return _bzhi_u64(bitContainer, nbBits);
#else
assert(nbBits < BIT_MASK_SIZE);
return bitContainer & BIT_mask[nbBits];
#endif
}
/*! BIT_lookBits() :
* Provides next n bits from local register.
* local register is not modified.
@@ -406,7 +365,7 @@ MEM_STATIC FORCE_INLINE_ATTR size_t BIT_readBits(BIT_DStream_t* bitD, unsigned n
}
/*! BIT_readBitsFast() :
* unsafe version; only works only if nbBits >= 1 */
* unsafe version; only works if nbBits >= 1 */
MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits)
{
size_t const value = BIT_lookBitsFast(bitD, nbBits);
zstd/common/compiler.h
+22 -3
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
@@ -165,6 +165,12 @@
#define UNLIKELY(x) (x)
#endif
#if __has_builtin(__builtin_unreachable) || (defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 5)))
# define ZSTD_UNREACHABLE { assert(0), __builtin_unreachable(); }
#else
# define ZSTD_UNREACHABLE { assert(0); }
#endif
/* disable warnings */
#ifdef _MSC_VER /* Visual Studio */
# include <intrin.h> /* For Visual 2005 */
@@ -181,6 +187,8 @@
# ifdef __AVX2__ //MSVC does not have a BMI2 specific flag, but every CPU that supports AVX2 also supports BMI2
# define STATIC_BMI2 1
# endif
# elif defined(__BMI2__) && defined(__x86_64__) && defined(__GNUC__)
# define STATIC_BMI2 1
# endif
#endif
@@ -273,7 +281,18 @@
* Sanitizer
*****************************************************************/
#if ZSTD_MEMORY_SANITIZER
/* Issue #3240 reports an ASAN failure on an llvm-mingw build. Out of an
* abundance of caution, disable our custom poisoning on mingw. */
#ifdef __MINGW32__
#ifndef ZSTD_ASAN_DONT_POISON_WORKSPACE
#define ZSTD_ASAN_DONT_POISON_WORKSPACE 1
#endif
#ifndef ZSTD_MSAN_DONT_POISON_WORKSPACE
#define ZSTD_MSAN_DONT_POISON_WORKSPACE 1
#endif
#endif
#if ZSTD_MEMORY_SANITIZER && !defined(ZSTD_MSAN_DONT_POISON_WORKSPACE)
/* Not all platforms that support msan provide sanitizers/msan_interface.h.
* We therefore declare the functions we need ourselves, rather than trying to
* include the header file... */
@@ -294,7 +313,7 @@ void __msan_poison(const volatile void *a, size_t size);
intptr_t __msan_test_shadow(const volatile void *x, size_t size);
#endif
#if ZSTD_ADDRESS_SANITIZER
#if ZSTD_ADDRESS_SANITIZER && !defined(ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* Not all platforms that support asan provide sanitizers/asan_interface.h.
* We therefore declare the functions we need ourselves, rather than trying to
* include the header file... */
zstd/common/cpu.h
+1 -1
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
zstd/common/debug.c
+1 -1
View File
@@ -1,7 +1,7 @@
/* ******************************************************************
* debug
* Part of FSE library
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
zstd/common/debug.h
+1 -1
View File
@@ -1,7 +1,7 @@
/* ******************************************************************
* debug
* Part of FSE library
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
zstd/common/entropy_common.c
+12 -40
View File
@@ -1,6 +1,6 @@
/* ******************************************************************
* Common functions of New Generation Entropy library
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy
@@ -19,8 +19,8 @@
#include "error_private.h" /* ERR_*, ERROR */
#define FSE_STATIC_LINKING_ONLY /* FSE_MIN_TABLELOG */
#include "fse.h"
#define HUF_STATIC_LINKING_ONLY /* HUF_TABLELOG_ABSOLUTEMAX */
#include "huf.h"
#include "bits.h" /* ZSDT_highbit32, ZSTD_countTrailingZeros32 */
/*=== Version ===*/
@@ -38,34 +38,6 @@ const char* HUF_getErrorName(size_t code) { return ERR_getErrorName(code); }
/*-**************************************************************
* FSE NCount encoding-decoding
****************************************************************/
static U32 FSE_ctz(U32 val)
{
assert(val != 0);
{
# if defined(_MSC_VER) /* Visual */
if (val != 0) {
unsigned long r;
_BitScanForward(&r, val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# elif defined(__GNUC__) && (__GNUC__ >= 3) /* GCC Intrinsic */
return __builtin_ctz(val);
# elif defined(__ICCARM__) /* IAR Intrinsic */
return __CTZ(val);
# else /* Software version */
U32 count = 0;
while ((val & 1) == 0) {
val >>= 1;
++count;
}
return count;
# endif
}
}
FORCE_INLINE_TEMPLATE
size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize)
@@ -113,7 +85,7 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne
* repeat.
* Avoid UB by setting the high bit to 1.
*/
int repeats = FSE_ctz(~bitStream | 0x80000000) >> 1;
int repeats = ZSTD_countTrailingZeros32(~bitStream | 0x80000000) >> 1;
while (repeats >= 12) {
charnum += 3 * 12;
if (LIKELY(ip <= iend-7)) {
@@ -124,7 +96,7 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne
ip = iend - 4;
}
bitStream = MEM_readLE32(ip) >> bitCount;
repeats = FSE_ctz(~bitStream | 0x80000000) >> 1;
repeats = ZSTD_countTrailingZeros32(~bitStream | 0x80000000) >> 1;
}
charnum += 3 * repeats;
bitStream >>= 2 * repeats;
@@ -189,7 +161,7 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne
* know that threshold > 1.
*/
if (remaining <= 1) break;
nbBits = BIT_highbit32(remaining) + 1;
nbBits = ZSTD_highbit32(remaining) + 1;
threshold = 1 << (nbBits - 1);
}
if (charnum >= maxSV1) break;
@@ -264,7 +236,7 @@ size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, U32* rankStats,
const void* src, size_t srcSize)
{
U32 wksp[HUF_READ_STATS_WORKSPACE_SIZE_U32];
return HUF_readStats_wksp(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, wksp, sizeof(wksp), /* bmi2 */ 0);
return HUF_readStats_wksp(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, wksp, sizeof(wksp), /* flags */ 0);
}
FORCE_INLINE_TEMPLATE size_t
@@ -312,14 +284,14 @@ HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats,
if (weightTotal == 0) return ERROR(corruption_detected);
/* get last non-null symbol weight (implied, total must be 2^n) */
{ U32 const tableLog = BIT_highbit32(weightTotal) + 1;
{ U32 const tableLog = ZSTD_highbit32(weightTotal) + 1;
if (tableLog > HUF_TABLELOG_MAX) return ERROR(corruption_detected);
*tableLogPtr = tableLog;
/* determine last weight */
{ U32 const total = 1 << tableLog;
U32 const rest = total - weightTotal;
U32 const verif = 1 << BIT_highbit32(rest);
U32 const lastWeight = BIT_highbit32(rest) + 1;
U32 const verif = 1 << ZSTD_highbit32(rest);
U32 const lastWeight = ZSTD_highbit32(rest) + 1;
if (verif != rest) return ERROR(corruption_detected); /* last value must be a clean power of 2 */
huffWeight[oSize] = (BYTE)lastWeight;
rankStats[lastWeight]++;
@@ -356,13 +328,13 @@ size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workSpace, size_t wkspSize,
int bmi2)
int flags)
{
#if DYNAMIC_BMI2
if (bmi2) {
if (flags & HUF_flags_bmi2) {
return HUF_readStats_body_bmi2(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize);
}
#endif
(void)bmi2;
(void)flags;
return HUF_readStats_body_default(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize);
}
zstd/common/error_private.c
+9 -2
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
@@ -27,9 +27,11 @@ const char* ERR_getErrorString(ERR_enum code)
case PREFIX(version_unsupported): return "Version not supported";
case PREFIX(frameParameter_unsupported): return "Unsupported frame parameter";
case PREFIX(frameParameter_windowTooLarge): return "Frame requires too much memory for decoding";
case PREFIX(corruption_detected): return "Corrupted block detected";
case PREFIX(corruption_detected): return "Data corruption detected";
case PREFIX(checksum_wrong): return "Restored data doesn't match checksum";
case PREFIX(literals_headerWrong): return "Header of Literals' block doesn't respect format specification";
case PREFIX(parameter_unsupported): return "Unsupported parameter";
case PREFIX(parameter_combination_unsupported): return "Unsupported combination of parameters";
case PREFIX(parameter_outOfBound): return "Parameter is out of bound";
case PREFIX(init_missing): return "Context should be init first";
case PREFIX(memory_allocation): return "Allocation error : not enough memory";
@@ -38,17 +40,22 @@ const char* ERR_getErrorString(ERR_enum code)
case PREFIX(tableLog_tooLarge): return "tableLog requires too much memory : unsupported";
case PREFIX(maxSymbolValue_tooLarge): return "Unsupported max Symbol Value : too large";
case PREFIX(maxSymbolValue_tooSmall): return "Specified maxSymbolValue is too small";
case PREFIX(stabilityCondition_notRespected): return "pledged buffer stability condition is not respected";
case PREFIX(dictionary_corrupted): return "Dictionary is corrupted";
case PREFIX(dictionary_wrong): return "Dictionary mismatch";
case PREFIX(dictionaryCreation_failed): return "Cannot create Dictionary from provided samples";
case PREFIX(dstSize_tooSmall): return "Destination buffer is too small";
case PREFIX(srcSize_wrong): return "Src size is incorrect";
case PREFIX(dstBuffer_null): return "Operation on NULL destination buffer";
case PREFIX(noForwardProgress_destFull): return "Operation made no progress over multiple calls, due to output buffer being full";
case PREFIX(noForwardProgress_inputEmpty): return "Operation made no progress over multiple calls, due to input being empty";
/* following error codes are not stable and may be removed or changed in a future version */
case PREFIX(frameIndex_tooLarge): return "Frame index is too large";
case PREFIX(seekableIO): return "An I/O error occurred when reading/seeking";
case PREFIX(dstBuffer_wrong): return "Destination buffer is wrong";
case PREFIX(srcBuffer_wrong): return "Source buffer is wrong";
case PREFIX(sequenceProducer_failed): return "Block-level external sequence producer returned an error code";
case PREFIX(externalSequences_invalid): return "External sequences are not valid";
case PREFIX(maxCode):
default: return notErrorCode;
}
zstd/common/error_private.h
+1 -1
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
zstd/common/fse.h
+5 -83
View File
@@ -1,7 +1,7 @@
/* ******************************************************************
* FSE : Finite State Entropy codec
* Public Prototypes declaration
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
@@ -53,34 +53,6 @@ extern "C" {
FSE_PUBLIC_API unsigned FSE_versionNumber(void); /**< library version number; to be used when checking dll version */
/*-****************************************
* FSE simple functions
******************************************/
/*! FSE_compress() :
Compress content of buffer 'src', of size 'srcSize', into destination buffer 'dst'.
'dst' buffer must be already allocated. Compression runs faster is dstCapacity >= FSE_compressBound(srcSize).
@return : size of compressed data (<= dstCapacity).
Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!!
if return == 1, srcData is a single byte symbol * srcSize times. Use RLE compression instead.
if FSE_isError(return), compression failed (more details using FSE_getErrorName())
*/
FSE_PUBLIC_API size_t FSE_compress(void* dst, size_t dstCapacity,
const void* src, size_t srcSize);
/*! FSE_decompress():
Decompress FSE data from buffer 'cSrc', of size 'cSrcSize',
into already allocated destination buffer 'dst', of size 'dstCapacity'.
@return : size of regenerated data (<= maxDstSize),
or an error code, which can be tested using FSE_isError() .
** Important ** : FSE_decompress() does not decompress non-compressible nor RLE data !!!
Why ? : making this distinction requires a header.
Header management is intentionally delegated to the user layer, which can better manage special cases.
*/
FSE_PUBLIC_API size_t FSE_decompress(void* dst, size_t dstCapacity,
const void* cSrc, size_t cSrcSize);
/*-*****************************************
* Tool functions
******************************************/
@@ -91,20 +63,6 @@ FSE_PUBLIC_API unsigned FSE_isError(size_t code); /* tells if a return
FSE_PUBLIC_API const char* FSE_getErrorName(size_t code); /* provides error code string (useful for debugging) */
/*-*****************************************
* FSE advanced functions
******************************************/
/*! FSE_compress2() :
Same as FSE_compress(), but allows the selection of 'maxSymbolValue' and 'tableLog'
Both parameters can be defined as '0' to mean : use default value
@return : size of compressed data
Special values : if return == 0, srcData is not compressible => Nothing is stored within cSrc !!!
if return == 1, srcData is a single byte symbol * srcSize times. Use RLE compression.
if FSE_isError(return), it's an error code.
*/
FSE_PUBLIC_API size_t FSE_compress2 (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog);
/*-*****************************************
* FSE detailed API
******************************************/
@@ -164,8 +122,6 @@ FSE_PUBLIC_API size_t FSE_writeNCount (void* buffer, size_t bufferSize,
/*! Constructor and Destructor of FSE_CTable.
Note that FSE_CTable size depends on 'tableLog' and 'maxSymbolValue' */
typedef unsigned FSE_CTable; /* don't allocate that. It's only meant to be more restrictive than void* */
FSE_PUBLIC_API FSE_CTable* FSE_createCTable (unsigned maxSymbolValue, unsigned tableLog);
FSE_PUBLIC_API void FSE_freeCTable (FSE_CTable* ct);
/*! FSE_buildCTable():
Builds `ct`, which must be already allocated, using FSE_createCTable().
@@ -241,23 +197,7 @@ FSE_PUBLIC_API size_t FSE_readNCount_bmi2(short* normalizedCounter,
unsigned* maxSymbolValuePtr, unsigned* tableLogPtr,
const void* rBuffer, size_t rBuffSize, int bmi2);
/*! Constructor and Destructor of FSE_DTable.
Note that its size depends on 'tableLog' */
typedef unsigned FSE_DTable; /* don't allocate that. It's just a way to be more restrictive than void* */
FSE_PUBLIC_API FSE_DTable* FSE_createDTable(unsigned tableLog);
FSE_PUBLIC_API void FSE_freeDTable(FSE_DTable* dt);
/*! FSE_buildDTable():
Builds 'dt', which must be already allocated, using FSE_createDTable().
return : 0, or an errorCode, which can be tested using FSE_isError() */
FSE_PUBLIC_API size_t FSE_buildDTable (FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog);
/*! FSE_decompress_usingDTable():
Decompress compressed source `cSrc` of size `cSrcSize` using `dt`
into `dst` which must be already allocated.
@return : size of regenerated data (necessarily <= `dstCapacity`),
or an errorCode, which can be tested using FSE_isError() */
FSE_PUBLIC_API size_t FSE_decompress_usingDTable(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, const FSE_DTable* dt);
/*!
Tutorial :
@@ -320,16 +260,6 @@ If there is an error, the function will return an error code, which can be teste
unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus);
/**< same as FSE_optimalTableLog(), which used `minus==2` */
/* FSE_compress_wksp() :
* Same as FSE_compress2(), but using an externally allocated scratch buffer (`workSpace`).
* FSE_COMPRESS_WKSP_SIZE_U32() provides the minimum size required for `workSpace` as a table of FSE_CTable.
*/
#define FSE_COMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ( FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) + ((maxTableLog > 12) ? (1 << (maxTableLog - 2)) : 1024) )
size_t FSE_compress_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
size_t FSE_buildCTable_raw (FSE_CTable* ct, unsigned nbBits);
/**< build a fake FSE_CTable, designed for a flat distribution, where each symbol uses nbBits */
size_t FSE_buildCTable_rle (FSE_CTable* ct, unsigned char symbolValue);
/**< build a fake FSE_CTable, designed to compress always the same symbolValue */
@@ -347,19 +277,11 @@ size_t FSE_buildCTable_wksp(FSE_CTable* ct, const short* normalizedCounter, unsi
FSE_PUBLIC_API size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
/**< Same as FSE_buildDTable(), using an externally allocated `workspace` produced with `FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxSymbolValue)` */
size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits);
/**< build a fake FSE_DTable, designed to read a flat distribution where each symbol uses nbBits */
size_t FSE_buildDTable_rle (FSE_DTable* dt, unsigned char symbolValue);
/**< build a fake FSE_DTable, designed to always generate the same symbolValue */
#define FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) (FSE_DTABLE_SIZE_U32(maxTableLog) + FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) + (FSE_MAX_SYMBOL_VALUE + 1) / 2 + 1)
#define FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) (FSE_DTABLE_SIZE_U32(maxTableLog) + 1 + FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) + (FSE_MAX_SYMBOL_VALUE + 1) / 2 + 1)
#define FSE_DECOMPRESS_WKSP_SIZE(maxTableLog, maxSymbolValue) (FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(unsigned))
size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize);
/**< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DECOMPRESS_WKSP_SIZE_U32(maxLog, maxSymbolValue)` */
size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2);
/**< Same as FSE_decompress_wksp() but with dynamic BMI2 support. Pass 1 if your CPU supports BMI2 or 0 if it doesn't. */
/**< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DECOMPRESS_WKSP_SIZE_U32(maxLog, maxSymbolValue)`.
* Set bmi2 to 1 if your CPU supports BMI2 or 0 if it doesn't */
typedef enum {
FSE_repeat_none, /**< Cannot use the previous table */
@@ -555,7 +477,7 @@ MEM_STATIC void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* statePt
/* FSE_getMaxNbBits() :
* Approximate maximum cost of a symbol, in bits.
* Fractional get rounded up (i.e : a symbol with a normalized frequency of 3 gives the same result as a frequency of 2)
* Fractional get rounded up (i.e. a symbol with a normalized frequency of 3 gives the same result as a frequency of 2)
* note 1 : assume symbolValue is valid (<= maxSymbolValue)
* note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */
MEM_STATIC U32 FSE_getMaxNbBits(const void* symbolTTPtr, U32 symbolValue)
zstd/common/fse_decompress.c
+7 -99
View File
@@ -1,6 +1,6 @@
/* ******************************************************************
* FSE : Finite State Entropy decoder
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
@@ -24,6 +24,7 @@
#include "error_private.h"
#define ZSTD_DEPS_NEED_MALLOC
#include "zstd_deps.h"
#include "bits.h" /* ZSTD_highbit32 */
/* **************************************************************
@@ -55,19 +56,6 @@
#define FSE_FUNCTION_NAME(X,Y) FSE_CAT(X,Y)
#define FSE_TYPE_NAME(X,Y) FSE_CAT(X,Y)
/* Function templates */
FSE_DTable* FSE_createDTable (unsigned tableLog)
{
if (tableLog > FSE_TABLELOG_ABSOLUTE_MAX) tableLog = FSE_TABLELOG_ABSOLUTE_MAX;
return (FSE_DTable*)ZSTD_malloc( FSE_DTABLE_SIZE_U32(tableLog) * sizeof (U32) );
}
void FSE_freeDTable (FSE_DTable* dt)
{
ZSTD_free(dt);
}
static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
{
void* const tdPtr = dt+1; /* because *dt is unsigned, 32-bits aligned on 32-bits */
@@ -127,10 +115,10 @@ static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCo
}
}
/* Now we spread those positions across the table.
* The benefit of doing it in two stages is that we avoid the the
* The benefit of doing it in two stages is that we avoid the
* variable size inner loop, which caused lots of branch misses.
* Now we can run through all the positions without any branch misses.
* We unroll the loop twice, since that is what emperically worked best.
* We unroll the loop twice, since that is what empirically worked best.
*/
{
size_t position = 0;
@@ -166,7 +154,7 @@ static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCo
for (u=0; u<tableSize; u++) {
FSE_FUNCTION_TYPE const symbol = (FSE_FUNCTION_TYPE)(tableDecode[u].symbol);
U32 const nextState = symbolNext[symbol]++;
tableDecode[u].nbBits = (BYTE) (tableLog - BIT_highbit32(nextState) );
tableDecode[u].nbBits = (BYTE) (tableLog - ZSTD_highbit32(nextState) );
tableDecode[u].newState = (U16) ( (nextState << tableDecode[u].nbBits) - tableSize);
} }
@@ -184,49 +172,6 @@ size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsi
/*-*******************************************************
* Decompression (Byte symbols)
*********************************************************/
size_t FSE_buildDTable_rle (FSE_DTable* dt, BYTE symbolValue)
{
void* ptr = dt;
FSE_DTableHeader* const DTableH = (FSE_DTableHeader*)ptr;
void* dPtr = dt + 1;
FSE_decode_t* const cell = (FSE_decode_t*)dPtr;
DTableH->tableLog = 0;
DTableH->fastMode = 0;
cell->newState = 0;
cell->symbol = symbolValue;
cell->nbBits = 0;
return 0;
}
size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits)
{
void* ptr = dt;
FSE_DTableHeader* const DTableH = (FSE_DTableHeader*)ptr;
void* dPtr = dt + 1;
FSE_decode_t* const dinfo = (FSE_decode_t*)dPtr;
const unsigned tableSize = 1 << nbBits;
const unsigned tableMask = tableSize - 1;
const unsigned maxSV1 = tableMask+1;
unsigned s;
/* Sanity checks */
if (nbBits < 1) return ERROR(GENERIC); /* min size */
/* Build Decoding Table */
DTableH->tableLog = (U16)nbBits;
DTableH->fastMode = 1;
for (s=0; s<maxSV1; s++) {
dinfo[s].newState = 0;
dinfo[s].symbol = (BYTE)s;
dinfo[s].nbBits = (BYTE)nbBits;
}
return 0;
}
FORCE_INLINE_TEMPLATE size_t FSE_decompress_usingDTable_generic(
void* dst, size_t maxDstSize,
@@ -290,26 +235,6 @@ FORCE_INLINE_TEMPLATE size_t FSE_decompress_usingDTable_generic(
return op-ostart;
}
size_t FSE_decompress_usingDTable(void* dst, size_t originalSize,
const void* cSrc, size_t cSrcSize,
const FSE_DTable* dt)
{
const void* ptr = dt;
const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr;
const U32 fastMode = DTableH->fastMode;
/* select fast mode (static) */
if (fastMode) return FSE_decompress_usingDTable_generic(dst, originalSize, cSrc, cSrcSize, dt, 1);
return FSE_decompress_usingDTable_generic(dst, originalSize, cSrc, cSrcSize, dt, 0);
}
size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize)
{
return FSE_decompress_wksp_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, /* bmi2 */ 0);
}
typedef struct {
short ncount[FSE_MAX_SYMBOL_VALUE + 1];
FSE_DTable dtable[1]; /* Dynamically sized */
@@ -342,7 +267,8 @@ FORCE_INLINE_TEMPLATE size_t FSE_decompress_wksp_body(
}
if (FSE_DECOMPRESS_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(tableLog_tooLarge);
workSpace = wksp->dtable + FSE_DTABLE_SIZE_U32(tableLog);
assert(sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog) <= wkspSize);
workSpace = (BYTE*)workSpace + sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog);
wkspSize -= sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog);
CHECK_F( FSE_buildDTable_internal(wksp->dtable, wksp->ncount, maxSymbolValue, tableLog, workSpace, wkspSize) );
@@ -382,22 +308,4 @@ size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc,
return FSE_decompress_wksp_body_default(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize);
}
typedef FSE_DTable DTable_max_t[FSE_DTABLE_SIZE_U32(FSE_MAX_TABLELOG)];
#ifndef ZSTD_NO_UNUSED_FUNCTIONS
size_t FSE_buildDTable(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog) {
U32 wksp[FSE_BUILD_DTABLE_WKSP_SIZE_U32(FSE_TABLELOG_ABSOLUTE_MAX, FSE_MAX_SYMBOL_VALUE)];
return FSE_buildDTable_wksp(dt, normalizedCounter, maxSymbolValue, tableLog, wksp, sizeof(wksp));
}
size_t FSE_decompress(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize)
{
/* Static analyzer seems unable to understand this table will be properly initialized later */
U32 wksp[FSE_DECOMPRESS_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)];
return FSE_decompress_wksp(dst, dstCapacity, cSrc, cSrcSize, FSE_MAX_TABLELOG, wksp, sizeof(wksp));
}
#endif
#endif /* FSE_COMMONDEFS_ONLY */
zstd/common/huf.h
+71 -162
View File
@@ -1,7 +1,7 @@
/* ******************************************************************
* huff0 huffman codec,
* part of Finite State Entropy library
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
@@ -21,100 +21,23 @@ extern "C" {
/* *** Dependencies *** */
#include "zstd_deps.h" /* size_t */
/* *** library symbols visibility *** */
/* Note : when linking with -fvisibility=hidden on gcc, or by default on Visual,
* HUF symbols remain "private" (internal symbols for library only).
* Set macro FSE_DLL_EXPORT to 1 if you want HUF symbols visible on DLL interface */
#if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4)
# define HUF_PUBLIC_API __attribute__ ((visibility ("default")))
#elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) /* Visual expected */
# define HUF_PUBLIC_API __declspec(dllexport)
#elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1)
# define HUF_PUBLIC_API __declspec(dllimport) /* not required, just to generate faster code (saves a function pointer load from IAT and an indirect jump) */
#else
# define HUF_PUBLIC_API
#endif
/* ========================== */
/* *** simple functions *** */
/* ========================== */
/** HUF_compress() :
* Compress content from buffer 'src', of size 'srcSize', into buffer 'dst'.
* 'dst' buffer must be already allocated.
* Compression runs faster if `dstCapacity` >= HUF_compressBound(srcSize).
* `srcSize` must be <= `HUF_BLOCKSIZE_MAX` == 128 KB.
* @return : size of compressed data (<= `dstCapacity`).
* Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!!
* if HUF_isError(return), compression failed (more details using HUF_getErrorName())
*/
HUF_PUBLIC_API size_t HUF_compress(void* dst, size_t dstCapacity,
const void* src, size_t srcSize);
/** HUF_decompress() :
* Decompress HUF data from buffer 'cSrc', of size 'cSrcSize',
* into already allocated buffer 'dst', of minimum size 'dstSize'.
* `originalSize` : **must** be the ***exact*** size of original (uncompressed) data.
* Note : in contrast with FSE, HUF_decompress can regenerate
* RLE (cSrcSize==1) and uncompressed (cSrcSize==dstSize) data,
* because it knows size to regenerate (originalSize).
* @return : size of regenerated data (== originalSize),
* or an error code, which can be tested using HUF_isError()
*/
HUF_PUBLIC_API size_t HUF_decompress(void* dst, size_t originalSize,
const void* cSrc, size_t cSrcSize);
/* *** Tool functions *** */
#define HUF_BLOCKSIZE_MAX (128 * 1024) /**< maximum input size for a single block compressed with HUF_compress */
HUF_PUBLIC_API size_t HUF_compressBound(size_t size); /**< maximum compressed size (worst case) */
/* Error Management */
HUF_PUBLIC_API unsigned HUF_isError(size_t code); /**< tells if a return value is an error code */
HUF_PUBLIC_API const char* HUF_getErrorName(size_t code); /**< provides error code string (useful for debugging) */
/* *** Advanced function *** */
/** HUF_compress2() :
* Same as HUF_compress(), but offers control over `maxSymbolValue` and `tableLog`.
* `maxSymbolValue` must be <= HUF_SYMBOLVALUE_MAX .
* `tableLog` must be `<= HUF_TABLELOG_MAX` . */
HUF_PUBLIC_API size_t HUF_compress2 (void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
unsigned maxSymbolValue, unsigned tableLog);
/** HUF_compress4X_wksp() :
* Same as HUF_compress2(), but uses externally allocated `workSpace`.
* `workspace` must be at least as large as HUF_WORKSPACE_SIZE */
#define HUF_WORKSPACE_SIZE ((8 << 10) + 512 /* sorting scratch space */)
#define HUF_WORKSPACE_SIZE_U64 (HUF_WORKSPACE_SIZE / sizeof(U64))
HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
unsigned maxSymbolValue, unsigned tableLog,
void* workSpace, size_t wkspSize);
#endif /* HUF_H_298734234 */
/* ******************************************************************
* WARNING !!
* The following section contains advanced and experimental definitions
* which shall never be used in the context of a dynamic library,
* because they are not guaranteed to remain stable in the future.
* Only consider them in association with static linking.
* *****************************************************************/
#if defined(HUF_STATIC_LINKING_ONLY) && !defined(HUF_H_HUF_STATIC_LINKING_ONLY)
#define HUF_H_HUF_STATIC_LINKING_ONLY
/* *** Dependencies *** */
#include "mem.h" /* U32 */
#include "mem.h" /* U32 */
#define FSE_STATIC_LINKING_ONLY
#include "fse.h"
/* *** Tool functions *** */
#define HUF_BLOCKSIZE_MAX (128 * 1024) /**< maximum input size for a single block compressed with HUF_compress */
size_t HUF_compressBound(size_t size); /**< maximum compressed size (worst case) */
/* Error Management */
unsigned HUF_isError(size_t code); /**< tells if a return value is an error code */
const char* HUF_getErrorName(size_t code); /**< provides error code string (useful for debugging) */
#define HUF_WORKSPACE_SIZE ((8 << 10) + 512 /* sorting scratch space */)
#define HUF_WORKSPACE_SIZE_U64 (HUF_WORKSPACE_SIZE / sizeof(U64))
/* *** Constants *** */
#define HUF_TABLELOG_MAX 12 /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */
#define HUF_TABLELOG_DEFAULT 11 /* default tableLog value when none specified */
@@ -154,25 +77,49 @@ typedef U32 HUF_DTable;
/* ****************************************
* Advanced decompression functions
******************************************/
size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol decoder */
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */
#endif
size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< decodes RLE and uncompressed */
size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< considers RLE and uncompressed as errors */
size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< considers RLE and uncompressed as errors */
size_t HUF_decompress4X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol decoder */
size_t HUF_decompress4X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< single-symbol decoder */
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */
size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< double-symbols decoder */
#endif
/**
* Huffman flags bitset.
* For all flags, 0 is the default value.
*/
typedef enum {
/**
* If compiled with DYNAMIC_BMI2: Set flag only if the CPU supports BMI2 at runtime.
* Otherwise: Ignored.
*/
HUF_flags_bmi2 = (1 << 0),
/**
* If set: Test possible table depths to find the one that produces the smallest header + encoded size.
* If unset: Use heuristic to find the table depth.
*/
HUF_flags_optimalDepth = (1 << 1),
/**
* If set: If the previous table can encode the input, always reuse the previous table.
* If unset: If the previous table can encode the input, reuse the previous table if it results in a smaller output.
*/
HUF_flags_preferRepeat = (1 << 2),
/**
* If set: Sample the input and check if the sample is uncompressible, if it is then don't attempt to compress.
* If unset: Always histogram the entire input.
*/
HUF_flags_suspectUncompressible = (1 << 3),
/**
* If set: Don't use assembly implementations
* If unset: Allow using assembly implementations
*/
HUF_flags_disableAsm = (1 << 4),
/**
* If set: Don't use the fast decoding loop, always use the fallback decoding loop.
* If unset: Use the fast decoding loop when possible.
*/
HUF_flags_disableFast = (1 << 5)
} HUF_flags_e;
/* ****************************************
* HUF detailed API
* ****************************************/
#define HUF_OPTIMAL_DEPTH_THRESHOLD ZSTD_btultra
/*! HUF_compress() does the following:
* 1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h")
@@ -185,12 +132,12 @@ size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize,
* For example, it's possible to compress several blocks using the same 'CTable',
* or to save and regenerate 'CTable' using external methods.
*/
unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue);
size_t HUF_buildCTable (HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits); /* @return : maxNbBits; CTable and count can overlap. In which case, CTable will overwrite count content */
size_t HUF_writeCTable (void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog);
unsigned HUF_minTableLog(unsigned symbolCardinality);
unsigned HUF_cardinality(const unsigned* count, unsigned maxSymbolValue);
unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, void* workSpace,
size_t wkspSize, HUF_CElt* table, const unsigned* count, int flags); /* table is used as scratch space for building and testing tables, not a return value */
size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize);
size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
size_t HUF_compress4X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2);
size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int flags);
size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
@@ -199,6 +146,7 @@ typedef enum {
HUF_repeat_check, /**< Can use the previous table but it must be checked. Note : The previous table must have been constructed by HUF_compress{1, 4}X_repeat */
HUF_repeat_valid /**< Can use the previous table and it is assumed to be valid */
} HUF_repeat;
/** HUF_compress4X_repeat() :
* Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
* If it uses hufTable it does not modify hufTable or repeat.
@@ -209,13 +157,13 @@ size_t HUF_compress4X_repeat(void* dst, size_t dstSize,
const void* src, size_t srcSize,
unsigned maxSymbolValue, unsigned tableLog,
void* workSpace, size_t wkspSize, /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible);
HUF_CElt* hufTable, HUF_repeat* repeat, int flags);
/** HUF_buildCTable_wksp() :
* Same as HUF_buildCTable(), but using externally allocated scratch buffer.
* `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE.
*/
#define HUF_CTABLE_WORKSPACE_SIZE_U32 (2*HUF_SYMBOLVALUE_MAX +1 +1)
#define HUF_CTABLE_WORKSPACE_SIZE_U32 ((4 * (HUF_SYMBOLVALUE_MAX + 1)) + 192)
#define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))
size_t HUF_buildCTable_wksp (HUF_CElt* tree,
const unsigned* count, U32 maxSymbolValue, U32 maxNbBits,
@@ -241,7 +189,7 @@ size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize,
U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workspace, size_t wkspSize,
int bmi2);
int flags);
/** HUF_readCTable() :
* Loading a CTable saved with HUF_writeCTable() */
@@ -279,32 +227,12 @@ U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize);
#define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9))
#define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))
#ifndef HUF_FORCE_DECOMPRESS_X2
size_t HUF_readDTableX1 (HUF_DTable* DTable, const void* src, size_t srcSize);
size_t HUF_readDTableX1_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
#endif
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_readDTableX2 (HUF_DTable* DTable, const void* src, size_t srcSize);
size_t HUF_readDTableX2_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
#endif
size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
#ifndef HUF_FORCE_DECOMPRESS_X2
size_t HUF_decompress4X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
#endif
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_decompress4X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
#endif
/* ====================== */
/* single stream variants */
/* ====================== */
size_t HUF_compress1X (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog);
size_t HUF_compress1X_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); /**< `workSpace` must be a table of at least HUF_WORKSPACE_SIZE_U64 U64 */
size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
size_t HUF_compress1X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2);
size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int flags);
/** HUF_compress1X_repeat() :
* Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
* If it uses hufTable it does not modify hufTable or repeat.
@@ -315,49 +243,30 @@ size_t HUF_compress1X_repeat(void* dst, size_t dstSize,
const void* src, size_t srcSize,
unsigned maxSymbolValue, unsigned tableLog,
void* workSpace, size_t wkspSize, /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible);
HUF_CElt* hufTable, HUF_repeat* repeat, int flags);
size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* single-symbol decoder */
size_t HUF_decompress1X_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* double-symbol decoder */
#endif
size_t HUF_decompress1X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);
size_t HUF_decompress1X_DCtx_wksp (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);
#ifndef HUF_FORCE_DECOMPRESS_X2
size_t HUF_decompress1X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol decoder */
size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< single-symbol decoder */
#endif
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_decompress1X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */
size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< double-symbols decoder */
#endif
size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); /**< automatic selection of sing or double symbol decoder, based on DTable */
#ifndef HUF_FORCE_DECOMPRESS_X2
size_t HUF_decompress1X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
#endif
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_decompress1X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags); /**< double-symbols decoder */
#endif
/* BMI2 variants.
* If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
*/
size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int flags);
#ifndef HUF_FORCE_DECOMPRESS_X2
size_t HUF_decompress1X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
#endif
size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
size_t HUF_decompress4X_hufOnly_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int flags);
size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
#ifndef HUF_FORCE_DECOMPRESS_X2
size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2);
size_t HUF_readDTableX1_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int flags);
#endif
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2);
size_t HUF_readDTableX2_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int flags);
#endif
#endif /* HUF_STATIC_LINKING_ONLY */
#endif /* HUF_H_298734234 */
#if defined (__cplusplus)
}
zstd/common/mem.h
+39 -46
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
@@ -133,21 +133,15 @@ MEM_STATIC size_t MEM_swapST(size_t in);
/*-**************************************************************
* Memory I/O Implementation
*****************************************************************/
/* MEM_FORCE_MEMORY_ACCESS :
* By default, access to unaligned memory is controlled by `memcpy()`, which is safe and portable.
* Unfortunately, on some target/compiler combinations, the generated assembly is sub-optimal.
* The below switch allow to select different access method for improved performance.
* Method 0 (default) : use `memcpy()`. Safe and portable.
* Method 1 : `__packed` statement. It depends on compiler extension (i.e., not portable).
* This method is safe if your compiler supports it, and *generally* as fast or faster than `memcpy`.
/* MEM_FORCE_MEMORY_ACCESS : For accessing unaligned memory:
* Method 0 : always use `memcpy()`. Safe and portable.
* Method 1 : Use compiler extension to set unaligned access.
* Method 2 : direct access. This method is portable but violate C standard.
* It can generate buggy code on targets depending on alignment.
* In some circumstances, it's the only known way to get the most performance (i.e. GCC + ARMv6)
* See http://fastcompression.blogspot.fr/2015/08/accessing-unaligned-memory.html for details.
* Prefer these methods in priority order (0 > 1 > 2)
* Default : method 1 if supported, else method 0
*/
#ifndef MEM_FORCE_MEMORY_ACCESS /* can be defined externally, on command line for example */
# if defined(__INTEL_COMPILER) || defined(__GNUC__) || defined(__ICCARM__)
# ifdef __GNUC__
# define MEM_FORCE_MEMORY_ACCESS 1
# endif
#endif
@@ -190,30 +184,19 @@ MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(U64*)memPtr = value; }
#elif defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==1)
/* __pack instructions are safer, but compiler specific, hence potentially problematic for some compilers */
/* currently only defined for gcc and icc */
#if defined(_MSC_VER) || (defined(__INTEL_COMPILER) && defined(WIN32))
__pragma( pack(push, 1) )
typedef struct { U16 v; } unalign16;
typedef struct { U32 v; } unalign32;
typedef struct { U64 v; } unalign64;
typedef struct { size_t v; } unalignArch;
__pragma( pack(pop) )
#else
typedef struct { U16 v; } __attribute__((packed)) unalign16;
typedef struct { U32 v; } __attribute__((packed)) unalign32;
typedef struct { U64 v; } __attribute__((packed)) unalign64;
typedef struct { size_t v; } __attribute__((packed)) unalignArch;
#endif
typedef __attribute__((aligned(1))) U16 unalign16;
typedef __attribute__((aligned(1))) U32 unalign32;
typedef __attribute__((aligned(1))) U64 unalign64;
typedef __attribute__((aligned(1))) size_t unalignArch;
MEM_STATIC U16 MEM_read16(const void* ptr) { return ((const unalign16*)ptr)->v; }
MEM_STATIC U32 MEM_read32(const void* ptr) { return ((const unalign32*)ptr)->v; }
MEM_STATIC U64 MEM_read64(const void* ptr) { return ((const unalign64*)ptr)->v; }
MEM_STATIC size_t MEM_readST(const void* ptr) { return ((const unalignArch*)ptr)->v; }
MEM_STATIC U16 MEM_read16(const void* ptr) { return *(const unalign16*)ptr; }
MEM_STATIC U32 MEM_read32(const void* ptr) { return *(const unalign32*)ptr; }
MEM_STATIC U64 MEM_read64(const void* ptr) { return *(const unalign64*)ptr; }
MEM_STATIC size_t MEM_readST(const void* ptr) { return *(const unalignArch*)ptr; }
MEM_STATIC void MEM_write16(void* memPtr, U16 value) { ((unalign16*)memPtr)->v = value; }
MEM_STATIC void MEM_write32(void* memPtr, U32 value) { ((unalign32*)memPtr)->v = value; }
MEM_STATIC void MEM_write64(void* memPtr, U64 value) { ((unalign64*)memPtr)->v = value; }
MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(unalign16*)memPtr = value; }
MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(unalign32*)memPtr = value; }
MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(unalign64*)memPtr = value; }
#else
@@ -257,6 +240,14 @@ MEM_STATIC void MEM_write64(void* memPtr, U64 value)
#endif /* MEM_FORCE_MEMORY_ACCESS */
MEM_STATIC U32 MEM_swap32_fallback(U32 in)
{
return ((in << 24) & 0xff000000 ) |
((in << 8) & 0x00ff0000 ) |
((in >> 8) & 0x0000ff00 ) |
((in >> 24) & 0x000000ff );
}
MEM_STATIC U32 MEM_swap32(U32 in)
{
#if defined(_MSC_VER) /* Visual Studio */
@@ -265,13 +256,22 @@ MEM_STATIC U32 MEM_swap32(U32 in)
|| (defined(__clang__) && __has_builtin(__builtin_bswap32))
return __builtin_bswap32(in);
#else
return ((in << 24) & 0xff000000 ) |
((in << 8) & 0x00ff0000 ) |
((in >> 8) & 0x0000ff00 ) |
((in >> 24) & 0x000000ff );
return MEM_swap32_fallback(in);
#endif
}
MEM_STATIC U64 MEM_swap64_fallback(U64 in)
{
return ((in << 56) & 0xff00000000000000ULL) |
((in << 40) & 0x00ff000000000000ULL) |
((in << 24) & 0x0000ff0000000000ULL) |
((in << 8) & 0x000000ff00000000ULL) |
((in >> 8) & 0x00000000ff000000ULL) |
((in >> 24) & 0x0000000000ff0000ULL) |
((in >> 40) & 0x000000000000ff00ULL) |
((in >> 56) & 0x00000000000000ffULL);
}
MEM_STATIC U64 MEM_swap64(U64 in)
{
#if defined(_MSC_VER) /* Visual Studio */
@@ -280,14 +280,7 @@ MEM_STATIC U64 MEM_swap64(U64 in)
|| (defined(__clang__) && __has_builtin(__builtin_bswap64))
return __builtin_bswap64(in);
#else
return ((in << 56) & 0xff00000000000000ULL) |
((in << 40) & 0x00ff000000000000ULL) |
((in << 24) & 0x0000ff0000000000ULL) |
((in << 8) & 0x000000ff00000000ULL) |
((in >> 8) & 0x00000000ff000000ULL) |
((in >> 24) & 0x0000000000ff0000ULL) |
((in >> 40) & 0x000000000000ff00ULL) |
((in >> 56) & 0x00000000000000ffULL);
return MEM_swap64_fallback(in);
#endif
}
zstd/common/pool.c
+26 -10
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
@@ -12,7 +12,7 @@
/* ====== Dependencies ======= */
#include "zstd_deps.h" /* size_t */
#include "debug.h" /* assert */
#include "zstd_internal.h" /* ZSTD_customMalloc, ZSTD_customFree */
#include "zstd_internal.h" /* ZSTD_customCalloc, ZSTD_customFree */
#include "pool.h"
/* ====== Compiler specifics ====== */
@@ -96,9 +96,7 @@ static void* POOL_thread(void* opaque) {
/* If the intended queue size was 0, signal after finishing job */
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
ctx->numThreadsBusy--;
if (ctx->queueSize == 1) {
ZSTD_pthread_cond_signal(&ctx->queuePushCond);
}
ZSTD_pthread_cond_signal(&ctx->queuePushCond);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
}
} /* for (;;) */
@@ -128,7 +126,7 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
* empty and full queues.
*/
ctx->queueSize = queueSize + 1;
ctx->queue = (POOL_job*)ZSTD_customMalloc(ctx->queueSize * sizeof(POOL_job), customMem);
ctx->queue = (POOL_job*)ZSTD_customCalloc(ctx->queueSize * sizeof(POOL_job), customMem);
ctx->queueHead = 0;
ctx->queueTail = 0;
ctx->numThreadsBusy = 0;
@@ -142,7 +140,7 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
}
ctx->shutdown = 0;
/* Allocate space for the thread handles */
ctx->threads = (ZSTD_pthread_t*)ZSTD_customMalloc(numThreads * sizeof(ZSTD_pthread_t), customMem);
ctx->threads = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), customMem);
ctx->threadCapacity = 0;
ctx->customMem = customMem;
/* Check for errors */
@@ -175,7 +173,7 @@ static void POOL_join(POOL_ctx* ctx) {
/* Join all of the threads */
{ size_t i;
for (i = 0; i < ctx->threadCapacity; ++i) {
ZSTD_pthread_join(ctx->threads[i], NULL); /* note : could fail */
ZSTD_pthread_join(ctx->threads[i]); /* note : could fail */
} }
}
@@ -190,6 +188,17 @@ void POOL_free(POOL_ctx *ctx) {
ZSTD_customFree(ctx, ctx->customMem);
}
/*! POOL_joinJobs() :
* Waits for all queued jobs to finish executing.
*/
void POOL_joinJobs(POOL_ctx* ctx) {
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
while(!ctx->queueEmpty || ctx->numThreadsBusy > 0) {
ZSTD_pthread_cond_wait(&ctx->queuePushCond, &ctx->queueMutex);
}
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
}
void ZSTD_freeThreadPool (ZSTD_threadPool* pool) {
POOL_free (pool);
}
@@ -211,7 +220,7 @@ static int POOL_resize_internal(POOL_ctx* ctx, size_t numThreads)
return 0;
}
/* numThreads > threadCapacity */
{ ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customMalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem);
{ ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem);
if (!threadPool) return 1;
/* replace existing thread pool */
ZSTD_memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(*threadPool));
@@ -262,7 +271,9 @@ static int isQueueFull(POOL_ctx const* ctx) {
static void
POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque)
{
POOL_job const job = {function, opaque};
POOL_job job;
job.function = function;
job.opaque = opaque;
assert(ctx != NULL);
if (ctx->shutdown) return;
@@ -330,6 +341,11 @@ void POOL_free(POOL_ctx* ctx) {
(void)ctx;
}
void POOL_joinJobs(POOL_ctx* ctx){
assert(!ctx || ctx == &g_poolCtx);
(void)ctx;
}
int POOL_resize(POOL_ctx* ctx, size_t numThreads) {
(void)ctx; (void)numThreads;
return 0;
zstd/common/pool.h
+7 -1
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
@@ -38,6 +38,12 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
*/
void POOL_free(POOL_ctx* ctx);
/*! POOL_joinJobs() :
* Waits for all queued jobs to finish executing.
*/
void POOL_joinJobs(POOL_ctx* ctx);
/*! POOL_resize() :
* Expands or shrinks pool's number of threads.
* This is more efficient than releasing + creating a new context,
zstd/common/portability_macros.h
+22 -3
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
@@ -12,7 +12,7 @@
#define ZSTD_PORTABILITY_MACROS_H
/**
* This header file contains macro defintions to support portability.
* This header file contains macro definitions to support portability.
* This header is shared between C and ASM code, so it MUST only
* contain macro definitions. It MUST not contain any C code.
*
@@ -88,7 +88,7 @@
#endif
/**
* Only enable assembly for GNUC comptabile compilers,
* Only enable assembly for GNUC compatible compilers,
* because other platforms may not support GAS assembly syntax.
*
* Only enable assembly for Linux / MacOS, other platforms may
@@ -134,4 +134,23 @@
# define ZSTD_ENABLE_ASM_X86_64_BMI2 0
#endif
/*
* For x86 ELF targets, add .note.gnu.property section for Intel CET in
* assembly sources when CET is enabled.
*
* Additionally, any function that may be called indirectly must begin
* with ZSTD_CET_ENDBRANCH.
*/
#if defined(__ELF__) && (defined(__x86_64__) || defined(__i386__)) \
&& defined(__has_include)
# if __has_include(<cet.h>)
# include <cet.h>
# define ZSTD_CET_ENDBRANCH _CET_ENDBR
# endif
#endif
#ifndef ZSTD_CET_ENDBRANCH
# define ZSTD_CET_ENDBRANCH
#endif
#endif /* ZSTD_PORTABILITY_MACROS_H */
zstd/common/threading.c
+68 -14
View File
@@ -23,8 +23,7 @@ int g_ZSTD_threading_useless_symbol;
#if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
/**
* Windows minimalist Pthread Wrapper, based on :
* http://www.cse.wustl.edu/~schmidt/win32-cv-1.html
* Windows minimalist Pthread Wrapper
*/
@@ -35,37 +34,92 @@ int g_ZSTD_threading_useless_symbol;
/* === Implementation === */
typedef struct {
void* (*start_routine)(void*);
void* arg;
int initialized;
ZSTD_pthread_cond_t initialized_cond;
ZSTD_pthread_mutex_t initialized_mutex;
} ZSTD_thread_params_t;
static unsigned __stdcall worker(void *arg)
{
ZSTD_pthread_t* const thread = (ZSTD_pthread_t*) arg;
thread->arg = thread->start_routine(thread->arg);
void* (*start_routine)(void*);
void* thread_arg;
/* Inialized thread_arg and start_routine and signal main thread that we don't need it
* to wait any longer.
*/
{
ZSTD_thread_params_t* thread_param = (ZSTD_thread_params_t*)arg;
thread_arg = thread_param->arg;
start_routine = thread_param->start_routine;
/* Signal main thread that we are running and do not depend on its memory anymore */
ZSTD_pthread_mutex_lock(&thread_param->initialized_mutex);
thread_param->initialized = 1;
ZSTD_pthread_cond_signal(&thread_param->initialized_cond);
ZSTD_pthread_mutex_unlock(&thread_param->initialized_mutex);
}
start_routine(thread_arg);
return 0;
}
int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused,
void* (*start_routine) (void*), void* arg)
{
ZSTD_thread_params_t thread_param;
(void)unused;
thread->arg = arg;
thread->start_routine = start_routine;
thread->handle = (HANDLE) _beginthreadex(NULL, 0, worker, thread, 0, NULL);
if (!thread->handle)
thread_param.start_routine = start_routine;
thread_param.arg = arg;
thread_param.initialized = 0;
*thread = NULL;
/* Setup thread initialization synchronization */
if(ZSTD_pthread_cond_init(&thread_param.initialized_cond, NULL)) {
/* Should never happen on Windows */
return -1;
}
if(ZSTD_pthread_mutex_init(&thread_param.initialized_mutex, NULL)) {
/* Should never happen on Windows */
ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
return -1;
}
/* Spawn thread */
*thread = (HANDLE)_beginthreadex(NULL, 0, worker, &thread_param, 0, NULL);
if (!thread) {
ZSTD_pthread_mutex_destroy(&thread_param.initialized_mutex);
ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
return errno;
else
return 0;
}
/* Wait for thread to be initialized */
ZSTD_pthread_mutex_lock(&thread_param.initialized_mutex);
while(!thread_param.initialized) {
ZSTD_pthread_cond_wait(&thread_param.initialized_cond, &thread_param.initialized_mutex);
}
ZSTD_pthread_mutex_unlock(&thread_param.initialized_mutex);
ZSTD_pthread_mutex_destroy(&thread_param.initialized_mutex);
ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
return 0;
}
int ZSTD_pthread_join(ZSTD_pthread_t thread, void **value_ptr)
int ZSTD_pthread_join(ZSTD_pthread_t thread)
{
DWORD result;
if (!thread.handle) return 0;
if (!thread) return 0;
result = WaitForSingleObject(thread, INFINITE);
CloseHandle(thread);
result = WaitForSingleObject(thread.handle, INFINITE);
switch (result) {
case WAIT_OBJECT_0:
if (value_ptr) *value_ptr = thread.arg;
return 0;
case WAIT_ABANDONED:
return EINVAL;
zstd/common/threading.h
+5 -10
View File
@@ -23,8 +23,7 @@ extern "C" {
#if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
/**
* Windows minimalist Pthread Wrapper, based on :
* http://www.cse.wustl.edu/~schmidt/win32-cv-1.html
* Windows minimalist Pthread Wrapper
*/
#ifdef WINVER
# undef WINVER
@@ -62,16 +61,12 @@ extern "C" {
#define ZSTD_pthread_cond_broadcast(a) WakeAllConditionVariable((a))
/* ZSTD_pthread_create() and ZSTD_pthread_join() */
typedef struct {
HANDLE handle;
void* (*start_routine)(void*);
void* arg;
} ZSTD_pthread_t;
typedef HANDLE ZSTD_pthread_t;
int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused,
void* (*start_routine) (void*), void* arg);
int ZSTD_pthread_join(ZSTD_pthread_t thread, void** value_ptr);
int ZSTD_pthread_join(ZSTD_pthread_t thread);
/**
* add here more wrappers as required
@@ -99,7 +94,7 @@ int ZSTD_pthread_join(ZSTD_pthread_t thread, void** value_ptr);
#define ZSTD_pthread_t pthread_t
#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
#define ZSTD_pthread_join(a, b) pthread_join((a),(b))
#define ZSTD_pthread_join(a) pthread_join((a),NULL)
#else /* DEBUGLEVEL >= 1 */
@@ -124,7 +119,7 @@ int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond);
#define ZSTD_pthread_t pthread_t
#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
#define ZSTD_pthread_join(a, b) pthread_join((a),(b))
#define ZSTD_pthread_join(a) pthread_join((a),NULL)
#endif
zstd/common/xxhash.c
+2 -2
View File
@@ -1,9 +1,9 @@
/*
* xxHash - Fast Hash algorithm
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - xxHash homepage: http://www.xxhash.com
* - xxHash homepage: https://cyan4973.github.io/xxHash/
* - xxHash source repository : https://github.com/Cyan4973/xxHash
*
* This source code is licensed under both the BSD-style license (found in the
zstd/common/xxhash.h
+8 -8
View File
@@ -1,9 +1,9 @@
/*
* xxHash - Fast Hash algorithm
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - xxHash homepage: http://www.xxhash.com
* - xxHash homepage: https://cyan4973.github.io/xxHash/
* - xxHash source repository : https://github.com/Cyan4973/xxHash
*
* This source code is licensed under both the BSD-style license (found in the
@@ -1314,7 +1314,7 @@ XXH3_128bits_reset_withSecretandSeed(XXH3_state_t* statePtr,
* care, as what works on one compiler/platform/optimization level may cause
* another to read garbage data or even crash.
*
* See http://fastcompression.blogspot.com/2015/08/accessing-unaligned-memory.html for details.
* See https://fastcompression.blogspot.com/2015/08/accessing-unaligned-memory.html for details.
*
* Prefer these methods in priority order (0 > 3 > 1 > 2)
*/
@@ -1534,7 +1534,7 @@ static void* XXH_memcpy(void* dest, const void* src, size_t size) { return ZSTD_
* @brief Used to prevent unwanted optimizations for @p var.
*
* It uses an empty GCC inline assembly statement with a register constraint
* which forces @p var into a general purpose register (eg eax, ebx, ecx
* which forces @p var into a general purpose register (e.g. eax, ebx, ecx
* on x86) and marks it as modified.
*
* This is used in a few places to avoid unwanted autovectorization (e.g.
@@ -1655,7 +1655,7 @@ static xxh_u32 XXH_read32(const void* ptr)
/*
* Portable and safe solution. Generally efficient.
* see: http://fastcompression.blogspot.com/2015/08/accessing-unaligned-memory.html
* see: https://fastcompression.blogspot.com/2015/08/accessing-unaligned-memory.html
*/
static xxh_u32 XXH_read32(const void* memPtr)
{
@@ -2296,7 +2296,7 @@ static xxh_u64 XXH_read64(const void* ptr)
/*
* Portable and safe solution. Generally efficient.
* see: http://fastcompression.blogspot.com/2015/08/accessing-unaligned-memory.html
* see: https://fastcompression.blogspot.com/2015/08/accessing-unaligned-memory.html
*/
static xxh_u64 XXH_read64(const void* memPtr)
{
@@ -2809,7 +2809,7 @@ enum XXH_VECTOR_TYPE /* fake enum */ {
* @ingroup tuning
* @brief Selects the minimum alignment for XXH3's accumulators.
*
* When using SIMD, this should match the alignment reqired for said vector
* When using SIMD, this should match the alignment required for said vector
* type, so, for example, 32 for AVX2.
*
* Default: Auto detected.
@@ -3026,7 +3026,7 @@ enum XXH_VECTOR_TYPE /* fake enum */ {
* have more than 2 NEON (F0/F1) micro-ops. If you are only using NEON instructions,
* you are only using 2/3 of the CPU bandwidth.
*
* This is even more noticable on the more advanced cores like the A76 which
* This is even more noticeable on the more advanced cores like the A76 which
* can dispatch 8 micro-ops per cycle, but still only 2 NEON micro-ops at once.
*
* Therefore, @ref XXH3_NEON_LANES lanes will be processed using NEON, and the
zstd/common/zstd_common.c
+1 -1
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
zstd/common/zstd_deps.h
+1 -1
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
zstd/common/zstd_internal.h
+17 -113
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Yann Collet, Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
@@ -28,7 +28,6 @@
#include "../zstd.h"
#define FSE_STATIC_LINKING_ONLY
#include "fse.h"
#define HUF_STATIC_LINKING_ONLY
#include "huf.h"
#ifndef XXH_STATIC_LINKING_ONLY
# define XXH_STATIC_LINKING_ONLY /* XXH64_state_t */
@@ -93,9 +92,9 @@ typedef enum { bt_raw, bt_rle, bt_compressed, bt_reserved } blockType_e;
#define ZSTD_FRAMECHECKSUMSIZE 4
#define MIN_SEQUENCES_SIZE 1 /* nbSeq==0 */
#define MIN_CBLOCK_SIZE (1 /*litCSize*/ + 1 /* RLE or RAW */ + MIN_SEQUENCES_SIZE /* nbSeq==0 */) /* for a non-null block */
#define MIN_CBLOCK_SIZE (1 /*litCSize*/ + 1 /* RLE or RAW */) /* for a non-null block */
#define MIN_LITERALS_FOR_4_STREAMS 6
#define HufLog 12
typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingType_e;
#define LONGNBSEQ 0x7F00
@@ -103,6 +102,7 @@ typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingTy
#define MINMATCH 3
#define Litbits 8
#define LitHufLog 11
#define MaxLit ((1<<Litbits) - 1)
#define MaxML 52
#define MaxLL 35
@@ -113,6 +113,8 @@ typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingTy
#define LLFSELog 9
#define OffFSELog 8
#define MaxFSELog MAX(MAX(MLFSELog, LLFSELog), OffFSELog)
#define MaxMLBits 16
#define MaxLLBits 16
#define ZSTD_MAX_HUF_HEADER_SIZE 128 /* header + <= 127 byte tree description */
/* Each table cannot take more than #symbols * FSELog bits */
@@ -235,12 +237,6 @@ void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e
* one COPY16() in the first call. Then, do two calls per loop since
* at that point it is more likely to have a high trip count.
*/
#ifdef __aarch64__
do {
COPY16(op, ip);
}
while (op < oend);
#else
ZSTD_copy16(op, ip);
if (16 >= length) return;
op += 16;
@@ -250,7 +246,6 @@ void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e
COPY16(op, ip);
}
while (op < oend);
#endif
}
}
@@ -299,11 +294,11 @@ typedef enum {
typedef struct {
seqDef* sequencesStart;
seqDef* sequences; /* ptr to end of sequences */
BYTE* litStart;
BYTE* lit; /* ptr to end of literals */
BYTE* llCode;
BYTE* mlCode;
BYTE* ofCode;
BYTE* litStart;
BYTE* lit; /* ptr to end of literals */
BYTE* llCode;
BYTE* mlCode;
BYTE* ofCode;
size_t maxNbSeq;
size_t maxNbLit;
@@ -311,8 +306,8 @@ typedef struct {
* in the seqStore that has a value larger than U16 (if it exists). To do so, we increment
* the existing value of the litLength or matchLength by 0x10000.
*/
ZSTD_longLengthType_e longLengthType;
U32 longLengthPos; /* Index of the sequence to apply long length modification to */
ZSTD_longLengthType_e longLengthType;
U32 longLengthPos; /* Index of the sequence to apply long length modification to */
} seqStore_t;
typedef struct {
@@ -331,10 +326,10 @@ MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore
seqLen.matchLength = seq->mlBase + MINMATCH;
if (seqStore->longLengthPos == (U32)(seq - seqStore->sequencesStart)) {
if (seqStore->longLengthType == ZSTD_llt_literalLength) {
seqLen.litLength += 0xFFFF;
seqLen.litLength += 0x10000;
}
if (seqStore->longLengthType == ZSTD_llt_matchLength) {
seqLen.matchLength += 0xFFFF;
seqLen.matchLength += 0x10000;
}
}
return seqLen;
@@ -347,12 +342,13 @@ MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore
* `decompressedBound != ZSTD_CONTENTSIZE_ERROR`
*/
typedef struct {
size_t nbBlocks;
size_t compressedSize;
unsigned long long decompressedBound;
} ZSTD_frameSizeInfo; /* decompress & legacy */
const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx); /* compress & dictBuilder */
void ZSTD_seqToCodes(const seqStore_t* seqStorePtr); /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */
int ZSTD_seqToCodes(const seqStore_t* seqStorePtr); /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */
/* custom memory allocation functions */
void* ZSTD_customMalloc(size_t size, ZSTD_customMem customMem);
@@ -360,98 +356,6 @@ void* ZSTD_customCalloc(size_t size, ZSTD_customMem customMem);
void ZSTD_customFree(void* ptr, ZSTD_customMem customMem);
MEM_STATIC U32 ZSTD_highbit32(U32 val) /* compress, dictBuilder, decodeCorpus */
{
assert(val != 0);
{
# if defined(_MSC_VER) /* Visual */
# if STATIC_BMI2 == 1
return _lzcnt_u32(val)^31;
# else
if (val != 0) {
unsigned long r;
_BitScanReverse(&r, val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 3) /* GCC Intrinsic */
return __builtin_clz (val) ^ 31;
# elif defined(__ICCARM__) /* IAR Intrinsic */
return 31 - __CLZ(val);
# else /* Software version */
static const U32 DeBruijnClz[32] = { 0, 9, 1, 10, 13, 21, 2, 29, 11, 14, 16, 18, 22, 25, 3, 30, 8, 12, 20, 28, 15, 17, 24, 7, 19, 27, 23, 6, 26, 5, 4, 31 };
U32 v = val;
v |= v >> 1;
v |= v >> 2;
v |= v >> 4;
v |= v >> 8;
v |= v >> 16;
return DeBruijnClz[(v * 0x07C4ACDDU) >> 27];
# endif
}
}
/**
* Counts the number of trailing zeros of a `size_t`.
* Most compilers should support CTZ as a builtin. A backup
* implementation is provided if the builtin isn't supported, but
* it may not be terribly efficient.
*/
MEM_STATIC unsigned ZSTD_countTrailingZeros(size_t val)
{
if (MEM_64bits()) {
# if defined(_MSC_VER) && defined(_WIN64)
# if STATIC_BMI2
return _tzcnt_u64(val);
# else
if (val != 0) {
unsigned long r;
_BitScanForward64(&r, (U64)val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4)
return __builtin_ctzll((U64)val);
# else
static const int DeBruijnBytePos[64] = { 0, 1, 2, 7, 3, 13, 8, 19,
4, 25, 14, 28, 9, 34, 20, 56,
5, 17, 26, 54, 15, 41, 29, 43,
10, 31, 38, 35, 21, 45, 49, 57,
63, 6, 12, 18, 24, 27, 33, 55,
16, 53, 40, 42, 30, 37, 44, 48,
62, 11, 23, 32, 52, 39, 36, 47,
61, 22, 51, 46, 60, 50, 59, 58 };
return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58];
# endif
} else { /* 32 bits */
# if defined(_MSC_VER)
if (val != 0) {
unsigned long r;
_BitScanForward(&r, (U32)val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# elif defined(__GNUC__) && (__GNUC__ >= 3)
return __builtin_ctz((U32)val);
# else
static const int DeBruijnBytePos[32] = { 0, 1, 28, 2, 29, 14, 24, 3,
30, 22, 20, 15, 25, 17, 4, 8,
31, 27, 13, 23, 21, 19, 16, 7,
26, 12, 18, 6, 11, 5, 10, 9 };
return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
# endif
}
}
/* ZSTD_invalidateRepCodes() :
* ensures next compression will not use repcodes from previous block.
* Note : only works with regular variant;
zstd/common/zstd_trace.h
+3 -3
View File
@@ -1,5 +1,5 @@
/*
* Copyright (c) Facebook, Inc.
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
@@ -21,13 +21,13 @@ extern "C" {
* For now, enable conservatively:
* - Only GNUC
* - Only ELF
* - Only x86-64 and i386
* - Only x86-64, i386 and aarch64
* Also, explicitly disable on platforms known not to work so they aren't
* forgotten in the future.
*/
#if !defined(ZSTD_HAVE_WEAK_SYMBOLS) && \
defined(__GNUC__) && defined(__ELF__) && \
(defined(__x86_64__) || defined(_M_X64) || defined(__i386__) || defined(_M_IX86)) && \
(defined(__x86_64__) || defined(_M_X64) || defined(__i386__) || defined(_M_IX86) || defined(__aarch64__)) && \
!defined(__APPLE__) && !defined(_WIN32) && !defined(__MINGW32__) && \
!defined(__CYGWIN__) && !defined(_AIX)
# define ZSTD_HAVE_WEAK_SYMBOLS 1