VOL-2112 move to voltha-lib-go
Change-Id: Ic1af08003c1d2c698c0cce371e64f47b47b8d875
diff --git a/vendor/github.com/DataDog/zstd/.travis.yml b/vendor/github.com/DataDog/zstd/.travis.yml
index c5aa33d..629470c 100644
--- a/vendor/github.com/DataDog/zstd/.travis.yml
+++ b/vendor/github.com/DataDog/zstd/.travis.yml
@@ -1,9 +1,10 @@
+dist: xenial
language: go
go:
- - 1.9.x
- 1.10.x
- 1.11.x
+ - 1.12.x
os:
- linux
diff --git a/vendor/github.com/DataDog/zstd/README.md b/vendor/github.com/DataDog/zstd/README.md
index 6c02e16..b32c3e7 100644
--- a/vendor/github.com/DataDog/zstd/README.md
+++ b/vendor/github.com/DataDog/zstd/README.md
@@ -2,8 +2,8 @@
[C Zstd Homepage](https://github.com/Cyan4973/zstd)
-The current headers and C files are from *v1.3.4* (Commit
-[2555975](https://github.com/facebook/zstd/releases/tag/v1.3.4)).
+The current headers and C files are from *v1.4.1* (Commit
+[52181f8](https://github.com/facebook/zstd/releases/tag/v1.4.1)).
## Usage
diff --git a/vendor/github.com/DataDog/zstd/bitstream.h b/vendor/github.com/DataDog/zstd/bitstream.h
index f7f389f..d955bd6 100644
--- a/vendor/github.com/DataDog/zstd/bitstream.h
+++ b/vendor/github.com/DataDog/zstd/bitstream.h
@@ -1,8 +1,7 @@
/* ******************************************************************
bitstream
Part of FSE library
- header file (to include)
- Copyright (C) 2013-2017, Yann Collet.
+ Copyright (C) 2013-present, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
@@ -49,21 +48,10 @@
* Dependencies
******************************************/
#include "mem.h" /* unaligned access routines */
+#include "debug.h" /* assert(), DEBUGLOG(), RAWLOG() */
#include "error_private.h" /* error codes and messages */
-/*-*************************************
-* Debug
-***************************************/
-#if defined(BIT_DEBUG) && (BIT_DEBUG>=1)
-# include <assert.h>
-#else
-# ifndef assert
-# define assert(condition) ((void)0)
-# endif
-#endif
-
-
/*=========================================
* Target specific
=========================================*/
@@ -83,8 +71,7 @@
* A critical property of these streams is that they encode and decode in **reverse** direction.
* So the first bit sequence you add will be the last to be read, like a LIFO stack.
*/
-typedef struct
-{
+typedef struct {
size_t bitContainer;
unsigned bitPos;
char* startPtr;
@@ -118,8 +105,7 @@
/*-********************************************
* bitStream decoding API (read backward)
**********************************************/
-typedef struct
-{
+typedef struct {
size_t bitContainer;
unsigned bitsConsumed;
const char* ptr;
@@ -236,7 +222,8 @@
}
/*! BIT_addBitsFast() :
- * works only if `value` is _clean_, meaning all high bits above nbBits are 0 */
+ * works only if `value` is _clean_,
+ * meaning all high bits above nbBits are 0 */
MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC,
size_t value, unsigned nbBits)
{
@@ -352,17 +339,10 @@
MEM_STATIC size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits)
{
-#if defined(__BMI__) && defined(__GNUC__) && __GNUC__*1000+__GNUC_MINOR__ >= 4008 /* experimental */
-# if defined(__x86_64__)
- if (sizeof(bitContainer)==8)
- return _bextr_u64(bitContainer, start, nbBits);
- else
-# endif
- return _bextr_u32(bitContainer, start, nbBits);
-#else
+ U32 const regMask = sizeof(bitContainer)*8 - 1;
+ /* if start > regMask, bitstream is corrupted, and result is undefined */
assert(nbBits < BIT_MASK_SIZE);
- return (bitContainer >> start) & BIT_mask[nbBits];
-#endif
+ return (bitContainer >> (start & regMask)) & BIT_mask[nbBits];
}
MEM_STATIC size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
@@ -379,9 +359,13 @@
* @return : value extracted */
MEM_STATIC size_t BIT_lookBits(const BIT_DStream_t* bitD, U32 nbBits)
{
-#if defined(__BMI__) && defined(__GNUC__) /* experimental; fails if bitD->bitsConsumed + nbBits > sizeof(bitD->bitContainer)*8 */
+ /* arbitrate between double-shift and shift+mask */
+#if 1
+ /* if bitD->bitsConsumed + nbBits > sizeof(bitD->bitContainer)*8,
+ * bitstream is likely corrupted, and result is undefined */
return BIT_getMiddleBits(bitD->bitContainer, (sizeof(bitD->bitContainer)*8) - bitD->bitsConsumed - nbBits, nbBits);
#else
+ /* this code path is slower on my os-x laptop */
U32 const regMask = sizeof(bitD->bitContainer)*8 - 1;
return ((bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> 1) >> ((regMask-nbBits) & regMask);
#endif
@@ -405,7 +389,7 @@
* Read (consume) next n bits from local register and update.
* Pay attention to not read more than nbBits contained into local register.
* @return : extracted value. */
-MEM_STATIC size_t BIT_readBits(BIT_DStream_t* bitD, U32 nbBits)
+MEM_STATIC size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits)
{
size_t const value = BIT_lookBits(bitD, nbBits);
BIT_skipBits(bitD, nbBits);
@@ -414,7 +398,7 @@
/*! BIT_readBitsFast() :
* unsafe version; only works only if nbBits >= 1 */
-MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, U32 nbBits)
+MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits)
{
size_t const value = BIT_lookBitsFast(bitD, nbBits);
assert(nbBits >= 1);
diff --git a/vendor/github.com/DataDog/zstd/compiler.h b/vendor/github.com/DataDog/zstd/compiler.h
index e90a3bc..87bf51a 100644
--- a/vendor/github.com/DataDog/zstd/compiler.h
+++ b/vendor/github.com/DataDog/zstd/compiler.h
@@ -15,6 +15,8 @@
* Compiler specifics
*********************************************************/
/* force inlining */
+
+#if !defined(ZSTD_NO_INLINE)
#if defined (__GNUC__) || defined(__cplusplus) || defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L /* C99 */
# define INLINE_KEYWORD inline
#else
@@ -29,9 +31,16 @@
# define FORCE_INLINE_ATTR
#endif
+#else
+
+#define INLINE_KEYWORD
+#define FORCE_INLINE_ATTR
+
+#endif
+
/**
* FORCE_INLINE_TEMPLATE is used to define C "templates", which take constant
- * parameters. They must be inlined for the compiler to elimininate the constant
+ * parameters. They must be inlined for the compiler to eliminate the constant
* branches.
*/
#define FORCE_INLINE_TEMPLATE static INLINE_KEYWORD FORCE_INLINE_ATTR
@@ -77,9 +86,9 @@
* Enabled for clang & gcc >=4.8 on x86 when BMI2 isn't enabled by default.
*/
#ifndef DYNAMIC_BMI2
- #if (defined(__clang__) && __has_attribute(__target__)) \
+ #if ((defined(__clang__) && __has_attribute(__target__)) \
|| (defined(__GNUC__) \
- && (__GNUC__ >= 5 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 8))) \
+ && (__GNUC__ >= 5 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 8)))) \
&& (defined(__x86_64__) || defined(_M_X86)) \
&& !defined(__BMI2__)
# define DYNAMIC_BMI2 1
@@ -88,14 +97,41 @@
#endif
#endif
-/* prefetch */
-#if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86)) /* _mm_prefetch() is not defined outside of x86/x64 */
-# include <mmintrin.h> /* https://msdn.microsoft.com/fr-fr/library/84szxsww(v=vs.90).aspx */
-# define PREFETCH(ptr) _mm_prefetch((const char*)ptr, _MM_HINT_T0)
-#elif defined(__GNUC__)
-# define PREFETCH(ptr) __builtin_prefetch(ptr, 0, 0)
+/* prefetch
+ * can be disabled, by declaring NO_PREFETCH build macro */
+#if defined(NO_PREFETCH)
+# define PREFETCH_L1(ptr) (void)(ptr) /* disabled */
+# define PREFETCH_L2(ptr) (void)(ptr) /* disabled */
#else
-# define PREFETCH(ptr) /* disabled */
+# if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86)) /* _mm_prefetch() is not defined outside of x86/x64 */
+# include <mmintrin.h> /* https://msdn.microsoft.com/fr-fr/library/84szxsww(v=vs.90).aspx */
+# define PREFETCH_L1(ptr) _mm_prefetch((const char*)(ptr), _MM_HINT_T0)
+# define PREFETCH_L2(ptr) _mm_prefetch((const char*)(ptr), _MM_HINT_T1)
+# elif defined(__GNUC__) && ( (__GNUC__ >= 4) || ( (__GNUC__ == 3) && (__GNUC_MINOR__ >= 1) ) )
+# define PREFETCH_L1(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 3 /* locality */)
+# define PREFETCH_L2(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 2 /* locality */)
+# else
+# define PREFETCH_L1(ptr) (void)(ptr) /* disabled */
+# define PREFETCH_L2(ptr) (void)(ptr) /* disabled */
+# endif
+#endif /* NO_PREFETCH */
+
+#define CACHELINE_SIZE 64
+
+#define PREFETCH_AREA(p, s) { \
+ const char* const _ptr = (const char*)(p); \
+ size_t const _size = (size_t)(s); \
+ size_t _pos; \
+ for (_pos=0; _pos<_size; _pos+=CACHELINE_SIZE) { \
+ PREFETCH_L2(_ptr + _pos); \
+ } \
+}
+
+/* vectorization */
+#if !defined(__clang__) && defined(__GNUC__)
+# define DONT_VECTORIZE __attribute__((optimize("no-tree-vectorize")))
+#else
+# define DONT_VECTORIZE
#endif
/* disable warnings */
diff --git a/vendor/github.com/DataDog/zstd/cover.c b/vendor/github.com/DataDog/zstd/cover.c
index b5a3957..6219967 100644
--- a/vendor/github.com/DataDog/zstd/cover.c
+++ b/vendor/github.com/DataDog/zstd/cover.c
@@ -29,6 +29,7 @@
#include "mem.h" /* read */
#include "pool.h"
#include "threading.h"
+#include "cover.h"
#include "zstd_internal.h" /* includes zstd.h */
#ifndef ZDICT_STATIC_LINKING_ONLY
#define ZDICT_STATIC_LINKING_ONLY
@@ -38,7 +39,8 @@
/*-*************************************
* Constants
***************************************/
-#define COVER_MAX_SAMPLES_SIZE (sizeof(size_t) == 8 ? ((U32)-1) : ((U32)1 GB))
+#define COVER_MAX_SAMPLES_SIZE (sizeof(size_t) == 8 ? ((unsigned)-1) : ((unsigned)1 GB))
+#define DEFAULT_SPLITPOINT 1.0
/*-*************************************
* Console display
@@ -184,7 +186,7 @@
}
/**
- * Destroyes a map that is inited with COVER_map_init().
+ * Destroys a map that is inited with COVER_map_init().
*/
static void COVER_map_destroy(COVER_map_t *map) {
if (map->data) {
@@ -203,6 +205,8 @@
size_t *offsets;
const size_t *samplesSizes;
size_t nbSamples;
+ size_t nbTrainSamples;
+ size_t nbTestSamples;
U32 *suffix;
size_t suffixSize;
U32 *freqs;
@@ -220,9 +224,9 @@
/**
* Returns the sum of the sample sizes.
*/
-static size_t COVER_sum(const size_t *samplesSizes, unsigned nbSamples) {
+size_t COVER_sum(const size_t *samplesSizes, unsigned nbSamples) {
size_t sum = 0;
- size_t i;
+ unsigned i;
for (i = 0; i < nbSamples; ++i) {
sum += samplesSizes[i];
}
@@ -377,14 +381,6 @@
ctx->suffix[dmerId] = freq;
}
-/**
- * A segment is a range in the source as well as the score of the segment.
- */
-typedef struct {
- U32 begin;
- U32 end;
- U32 score;
-} COVER_segment_t;
/**
* Selects the best segment in an epoch.
@@ -395,7 +391,7 @@
*
* Score(S) = F(S_1) + F(S_2) + ... + F(S_{k-d+1})
*
- * Once the dmer d is in the dictionay we set F(d) = 0.
+ * Once the dmer d is in the dictionary we set F(d) = 0.
*/
static COVER_segment_t COVER_selectSegment(const COVER_ctx_t *ctx, U32 *freqs,
COVER_map_t *activeDmers, U32 begin,
@@ -439,7 +435,7 @@
U32 *delDmerOcc = COVER_map_at(activeDmers, delDmer);
activeSegment.begin += 1;
*delDmerOcc -= 1;
- /* If this is the last occurence of the dmer, subtract its score */
+ /* If this is the last occurrence of the dmer, subtract its score */
if (*delDmerOcc == 0) {
COVER_map_remove(activeDmers, delDmer);
activeSegment.score -= freqs[delDmer];
@@ -494,6 +490,10 @@
if (parameters.d > parameters.k) {
return 0;
}
+ /* 0 < splitPoint <= 1 */
+ if (parameters.splitPoint <= 0 || parameters.splitPoint > 1){
+ return 0;
+ }
return 1;
}
@@ -526,30 +526,49 @@
* Prepare a context for dictionary building.
* The context is only dependent on the parameter `d` and can used multiple
* times.
- * Returns 1 on success or zero on error.
+ * Returns 0 on success or error code on error.
* The context must be destroyed with `COVER_ctx_destroy()`.
*/
-static int COVER_ctx_init(COVER_ctx_t *ctx, const void *samplesBuffer,
+static size_t COVER_ctx_init(COVER_ctx_t *ctx, const void *samplesBuffer,
const size_t *samplesSizes, unsigned nbSamples,
- unsigned d) {
+ unsigned d, double splitPoint) {
const BYTE *const samples = (const BYTE *)samplesBuffer;
const size_t totalSamplesSize = COVER_sum(samplesSizes, nbSamples);
+ /* Split samples into testing and training sets */
+ const unsigned nbTrainSamples = splitPoint < 1.0 ? (unsigned)((double)nbSamples * splitPoint) : nbSamples;
+ const unsigned nbTestSamples = splitPoint < 1.0 ? nbSamples - nbTrainSamples : nbSamples;
+ const size_t trainingSamplesSize = splitPoint < 1.0 ? COVER_sum(samplesSizes, nbTrainSamples) : totalSamplesSize;
+ const size_t testSamplesSize = splitPoint < 1.0 ? COVER_sum(samplesSizes + nbTrainSamples, nbTestSamples) : totalSamplesSize;
/* Checks */
if (totalSamplesSize < MAX(d, sizeof(U64)) ||
totalSamplesSize >= (size_t)COVER_MAX_SAMPLES_SIZE) {
DISPLAYLEVEL(1, "Total samples size is too large (%u MB), maximum size is %u MB\n",
- (U32)(totalSamplesSize>>20), (COVER_MAX_SAMPLES_SIZE >> 20));
- return 0;
+ (unsigned)(totalSamplesSize>>20), (COVER_MAX_SAMPLES_SIZE >> 20));
+ return ERROR(srcSize_wrong);
+ }
+ /* Check if there are at least 5 training samples */
+ if (nbTrainSamples < 5) {
+ DISPLAYLEVEL(1, "Total number of training samples is %u and is invalid.", nbTrainSamples);
+ return ERROR(srcSize_wrong);
+ }
+ /* Check if there's testing sample */
+ if (nbTestSamples < 1) {
+ DISPLAYLEVEL(1, "Total number of testing samples is %u and is invalid.", nbTestSamples);
+ return ERROR(srcSize_wrong);
}
/* Zero the context */
memset(ctx, 0, sizeof(*ctx));
- DISPLAYLEVEL(2, "Training on %u samples of total size %u\n", nbSamples,
- (U32)totalSamplesSize);
+ DISPLAYLEVEL(2, "Training on %u samples of total size %u\n", nbTrainSamples,
+ (unsigned)trainingSamplesSize);
+ DISPLAYLEVEL(2, "Testing on %u samples of total size %u\n", nbTestSamples,
+ (unsigned)testSamplesSize);
ctx->samples = samples;
ctx->samplesSizes = samplesSizes;
ctx->nbSamples = nbSamples;
+ ctx->nbTrainSamples = nbTrainSamples;
+ ctx->nbTestSamples = nbTestSamples;
/* Partial suffix array */
- ctx->suffixSize = totalSamplesSize - MAX(d, sizeof(U64)) + 1;
+ ctx->suffixSize = trainingSamplesSize - MAX(d, sizeof(U64)) + 1;
ctx->suffix = (U32 *)malloc(ctx->suffixSize * sizeof(U32));
/* Maps index to the dmerID */
ctx->dmerAt = (U32 *)malloc(ctx->suffixSize * sizeof(U32));
@@ -558,12 +577,12 @@
if (!ctx->suffix || !ctx->dmerAt || !ctx->offsets) {
DISPLAYLEVEL(1, "Failed to allocate scratch buffers\n");
COVER_ctx_destroy(ctx);
- return 0;
+ return ERROR(memory_allocation);
}
ctx->freqs = NULL;
ctx->d = d;
- /* Fill offsets from the samlesSizes */
+ /* Fill offsets from the samplesSizes */
{
U32 i;
ctx->offsets[0] = 0;
@@ -581,10 +600,17 @@
for (i = 0; i < ctx->suffixSize; ++i) {
ctx->suffix[i] = i;
}
- /* qsort doesn't take an opaque pointer, so pass as a global */
+ /* qsort doesn't take an opaque pointer, so pass as a global.
+ * On OpenBSD qsort() is not guaranteed to be stable, their mergesort() is.
+ */
g_ctx = ctx;
+#if defined(__OpenBSD__)
+ mergesort(ctx->suffix, ctx->suffixSize, sizeof(U32),
+ (ctx->d <= 8 ? &COVER_strict_cmp8 : &COVER_strict_cmp));
+#else
qsort(ctx->suffix, ctx->suffixSize, sizeof(U32),
(ctx->d <= 8 ? &COVER_strict_cmp8 : &COVER_strict_cmp));
+#endif
}
DISPLAYLEVEL(2, "Computing frequencies\n");
/* For each dmer group (group of positions with the same first d bytes):
@@ -598,7 +624,40 @@
(ctx->d <= 8 ? &COVER_cmp8 : &COVER_cmp), &COVER_group);
ctx->freqs = ctx->suffix;
ctx->suffix = NULL;
- return 1;
+ return 0;
+}
+
+void COVER_warnOnSmallCorpus(size_t maxDictSize, size_t nbDmers, int displayLevel)
+{
+ const double ratio = (double)nbDmers / maxDictSize;
+ if (ratio >= 10) {
+ return;
+ }
+ LOCALDISPLAYLEVEL(displayLevel, 1,
+ "WARNING: The maximum dictionary size %u is too large "
+ "compared to the source size %u! "
+ "size(source)/size(dictionary) = %f, but it should be >= "
+ "10! This may lead to a subpar dictionary! We recommend "
+ "training on sources at least 10x, and up to 100x the "
+ "size of the dictionary!\n", (U32)maxDictSize,
+ (U32)nbDmers, ratio);
+}
+
+COVER_epoch_info_t COVER_computeEpochs(U32 maxDictSize,
+ U32 nbDmers, U32 k, U32 passes)
+{
+ const U32 minEpochSize = k * 10;
+ COVER_epoch_info_t epochs;
+ epochs.num = MAX(1, maxDictSize / k / passes);
+ epochs.size = nbDmers / epochs.num;
+ if (epochs.size >= minEpochSize) {
+ assert(epochs.size * epochs.num <= nbDmers);
+ return epochs;
+ }
+ epochs.size = MIN(minEpochSize, nbDmers);
+ epochs.num = nbDmers / epochs.size;
+ assert(epochs.size * epochs.num <= nbDmers);
+ return epochs;
}
/**
@@ -610,28 +669,34 @@
ZDICT_cover_params_t parameters) {
BYTE *const dict = (BYTE *)dictBuffer;
size_t tail = dictBufferCapacity;
- /* Divide the data up into epochs of equal size.
- * We will select at least one segment from each epoch.
- */
- const U32 epochs = (U32)(dictBufferCapacity / parameters.k);
- const U32 epochSize = (U32)(ctx->suffixSize / epochs);
+ /* Divide the data into epochs. We will select one segment from each epoch. */
+ const COVER_epoch_info_t epochs = COVER_computeEpochs(
+ (U32)dictBufferCapacity, (U32)ctx->suffixSize, parameters.k, 4);
+ const size_t maxZeroScoreRun = MAX(10, MIN(100, epochs.num >> 3));
+ size_t zeroScoreRun = 0;
size_t epoch;
- DISPLAYLEVEL(2, "Breaking content into %u epochs of size %u\n", epochs,
- epochSize);
+ DISPLAYLEVEL(2, "Breaking content into %u epochs of size %u\n",
+ (U32)epochs.num, (U32)epochs.size);
/* Loop through the epochs until there are no more segments or the dictionary
* is full.
*/
- for (epoch = 0; tail > 0; epoch = (epoch + 1) % epochs) {
- const U32 epochBegin = (U32)(epoch * epochSize);
- const U32 epochEnd = epochBegin + epochSize;
+ for (epoch = 0; tail > 0; epoch = (epoch + 1) % epochs.num) {
+ const U32 epochBegin = (U32)(epoch * epochs.size);
+ const U32 epochEnd = epochBegin + epochs.size;
size_t segmentSize;
/* Select a segment */
COVER_segment_t segment = COVER_selectSegment(
ctx, freqs, activeDmers, epochBegin, epochEnd, parameters);
- /* If the segment covers no dmers, then we are out of content */
+ /* If the segment covers no dmers, then we are out of content.
+ * There may be new content in other epochs, for continue for some time.
+ */
if (segment.score == 0) {
- break;
+ if (++zeroScoreRun >= maxZeroScoreRun) {
+ break;
+ }
+ continue;
}
+ zeroScoreRun = 0;
/* Trim the segment if necessary and if it is too small then we are done */
segmentSize = MIN(segment.end - segment.begin + parameters.d - 1, tail);
if (segmentSize < parameters.d) {
@@ -644,7 +709,7 @@
memcpy(dict + tail, ctx->samples + segment.begin, segmentSize);
DISPLAYUPDATE(
2, "\r%u%% ",
- (U32)(((dictBufferCapacity - tail) * 100) / dictBufferCapacity));
+ (unsigned)(((dictBufferCapacity - tail) * 100) / dictBufferCapacity));
}
DISPLAYLEVEL(2, "\r%79s\r", "");
return tail;
@@ -658,17 +723,17 @@
BYTE* const dict = (BYTE*)dictBuffer;
COVER_ctx_t ctx;
COVER_map_t activeDmers;
-
+ parameters.splitPoint = 1.0;
/* Initialize global data */
g_displayLevel = parameters.zParams.notificationLevel;
/* Checks */
if (!COVER_checkParameters(parameters, dictBufferCapacity)) {
DISPLAYLEVEL(1, "Cover parameters incorrect\n");
- return ERROR(GENERIC);
+ return ERROR(parameter_outOfBound);
}
if (nbSamples == 0) {
DISPLAYLEVEL(1, "Cover must have at least one input file\n");
- return ERROR(GENERIC);
+ return ERROR(srcSize_wrong);
}
if (dictBufferCapacity < ZDICT_DICTSIZE_MIN) {
DISPLAYLEVEL(1, "dictBufferCapacity must be at least %u\n",
@@ -676,14 +741,18 @@
return ERROR(dstSize_tooSmall);
}
/* Initialize context and activeDmers */
- if (!COVER_ctx_init(&ctx, samplesBuffer, samplesSizes, nbSamples,
- parameters.d)) {
- return ERROR(GENERIC);
+ {
+ size_t const initVal = COVER_ctx_init(&ctx, samplesBuffer, samplesSizes, nbSamples,
+ parameters.d, parameters.splitPoint);
+ if (ZSTD_isError(initVal)) {
+ return initVal;
+ }
}
+ COVER_warnOnSmallCorpus(dictBufferCapacity, ctx.suffixSize, g_displayLevel);
if (!COVER_map_init(&activeDmers, parameters.k - parameters.d + 1)) {
DISPLAYLEVEL(1, "Failed to allocate dmer map: out of memory\n");
COVER_ctx_destroy(&ctx);
- return ERROR(GENERIC);
+ return ERROR(memory_allocation);
}
DISPLAYLEVEL(2, "Building dictionary\n");
@@ -696,7 +765,7 @@
samplesBuffer, samplesSizes, nbSamples, parameters.zParams);
if (!ZSTD_isError(dictionarySize)) {
DISPLAYLEVEL(2, "Constructed dictionary of size %u\n",
- (U32)dictionarySize);
+ (unsigned)dictionarySize);
}
COVER_ctx_destroy(&ctx);
COVER_map_destroy(&activeDmers);
@@ -704,28 +773,65 @@
}
}
-/**
- * COVER_best_t is used for two purposes:
- * 1. Synchronizing threads.
- * 2. Saving the best parameters and dictionary.
- *
- * All of the methods except COVER_best_init() are thread safe if zstd is
- * compiled with multithreaded support.
- */
-typedef struct COVER_best_s {
- ZSTD_pthread_mutex_t mutex;
- ZSTD_pthread_cond_t cond;
- size_t liveJobs;
- void *dict;
- size_t dictSize;
- ZDICT_cover_params_t parameters;
- size_t compressedSize;
-} COVER_best_t;
+
+
+size_t COVER_checkTotalCompressedSize(const ZDICT_cover_params_t parameters,
+ const size_t *samplesSizes, const BYTE *samples,
+ size_t *offsets,
+ size_t nbTrainSamples, size_t nbSamples,
+ BYTE *const dict, size_t dictBufferCapacity) {
+ size_t totalCompressedSize = ERROR(GENERIC);
+ /* Pointers */
+ ZSTD_CCtx *cctx;
+ ZSTD_CDict *cdict;
+ void *dst;
+ /* Local variables */
+ size_t dstCapacity;
+ size_t i;
+ /* Allocate dst with enough space to compress the maximum sized sample */
+ {
+ size_t maxSampleSize = 0;
+ i = parameters.splitPoint < 1.0 ? nbTrainSamples : 0;
+ for (; i < nbSamples; ++i) {
+ maxSampleSize = MAX(samplesSizes[i], maxSampleSize);
+ }
+ dstCapacity = ZSTD_compressBound(maxSampleSize);
+ dst = malloc(dstCapacity);
+ }
+ /* Create the cctx and cdict */
+ cctx = ZSTD_createCCtx();
+ cdict = ZSTD_createCDict(dict, dictBufferCapacity,
+ parameters.zParams.compressionLevel);
+ if (!dst || !cctx || !cdict) {
+ goto _compressCleanup;
+ }
+ /* Compress each sample and sum their sizes (or error) */
+ totalCompressedSize = dictBufferCapacity;
+ i = parameters.splitPoint < 1.0 ? nbTrainSamples : 0;
+ for (; i < nbSamples; ++i) {
+ const size_t size = ZSTD_compress_usingCDict(
+ cctx, dst, dstCapacity, samples + offsets[i],
+ samplesSizes[i], cdict);
+ if (ZSTD_isError(size)) {
+ totalCompressedSize = size;
+ goto _compressCleanup;
+ }
+ totalCompressedSize += size;
+ }
+_compressCleanup:
+ ZSTD_freeCCtx(cctx);
+ ZSTD_freeCDict(cdict);
+ if (dst) {
+ free(dst);
+ }
+ return totalCompressedSize;
+}
+
/**
* Initialize the `COVER_best_t`.
*/
-static void COVER_best_init(COVER_best_t *best) {
+void COVER_best_init(COVER_best_t *best) {
if (best==NULL) return; /* compatible with init on NULL */
(void)ZSTD_pthread_mutex_init(&best->mutex, NULL);
(void)ZSTD_pthread_cond_init(&best->cond, NULL);
@@ -739,7 +845,7 @@
/**
* Wait until liveJobs == 0.
*/
-static void COVER_best_wait(COVER_best_t *best) {
+void COVER_best_wait(COVER_best_t *best) {
if (!best) {
return;
}
@@ -753,7 +859,7 @@
/**
* Call COVER_best_wait() and then destroy the COVER_best_t.
*/
-static void COVER_best_destroy(COVER_best_t *best) {
+void COVER_best_destroy(COVER_best_t *best) {
if (!best) {
return;
}
@@ -769,7 +875,7 @@
* Called when a thread is about to be launched.
* Increments liveJobs.
*/
-static void COVER_best_start(COVER_best_t *best) {
+void COVER_best_start(COVER_best_t *best) {
if (!best) {
return;
}
@@ -783,9 +889,11 @@
* Decrements liveJobs and signals any waiting threads if liveJobs == 0.
* If this dictionary is the best so far save it and its parameters.
*/
-static void COVER_best_finish(COVER_best_t *best, size_t compressedSize,
- ZDICT_cover_params_t parameters, void *dict,
- size_t dictSize) {
+void COVER_best_finish(COVER_best_t *best, ZDICT_cover_params_t parameters,
+ COVER_dictSelection_t selection) {
+ void* dict = selection.dictContent;
+ size_t compressedSize = selection.totalCompressedSize;
+ size_t dictSize = selection.dictSize;
if (!best) {
return;
}
@@ -805,19 +913,129 @@
if (!best->dict) {
best->compressedSize = ERROR(GENERIC);
best->dictSize = 0;
+ ZSTD_pthread_cond_signal(&best->cond);
+ ZSTD_pthread_mutex_unlock(&best->mutex);
return;
}
}
/* Save the dictionary, parameters, and size */
+ if (!dict) {
+ return;
+ }
memcpy(best->dict, dict, dictSize);
best->dictSize = dictSize;
best->parameters = parameters;
best->compressedSize = compressedSize;
}
- ZSTD_pthread_mutex_unlock(&best->mutex);
if (liveJobs == 0) {
ZSTD_pthread_cond_broadcast(&best->cond);
}
+ ZSTD_pthread_mutex_unlock(&best->mutex);
+ }
+}
+
+COVER_dictSelection_t COVER_dictSelectionError(size_t error) {
+ COVER_dictSelection_t selection = { NULL, 0, error };
+ return selection;
+}
+
+unsigned COVER_dictSelectionIsError(COVER_dictSelection_t selection) {
+ return (ZSTD_isError(selection.totalCompressedSize) || !selection.dictContent);
+}
+
+void COVER_dictSelectionFree(COVER_dictSelection_t selection){
+ free(selection.dictContent);
+}
+
+COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent,
+ size_t dictContentSize, const BYTE* samplesBuffer, const size_t* samplesSizes, unsigned nbFinalizeSamples,
+ size_t nbCheckSamples, size_t nbSamples, ZDICT_cover_params_t params, size_t* offsets, size_t totalCompressedSize) {
+
+ size_t largestDict = 0;
+ size_t largestCompressed = 0;
+ BYTE* customDictContentEnd = customDictContent + dictContentSize;
+
+ BYTE * largestDictbuffer = (BYTE *)malloc(dictContentSize);
+ BYTE * candidateDictBuffer = (BYTE *)malloc(dictContentSize);
+ double regressionTolerance = ((double)params.shrinkDictMaxRegression / 100.0) + 1.00;
+
+ if (!largestDictbuffer || !candidateDictBuffer) {
+ free(largestDictbuffer);
+ free(candidateDictBuffer);
+ return COVER_dictSelectionError(dictContentSize);
+ }
+
+ /* Initial dictionary size and compressed size */
+ memcpy(largestDictbuffer, customDictContent, dictContentSize);
+ dictContentSize = ZDICT_finalizeDictionary(
+ largestDictbuffer, dictContentSize, customDictContent, dictContentSize,
+ samplesBuffer, samplesSizes, nbFinalizeSamples, params.zParams);
+
+ if (ZDICT_isError(dictContentSize)) {
+ free(largestDictbuffer);
+ free(candidateDictBuffer);
+ return COVER_dictSelectionError(dictContentSize);
+ }
+
+ totalCompressedSize = COVER_checkTotalCompressedSize(params, samplesSizes,
+ samplesBuffer, offsets,
+ nbCheckSamples, nbSamples,
+ largestDictbuffer, dictContentSize);
+
+ if (ZSTD_isError(totalCompressedSize)) {
+ free(largestDictbuffer);
+ free(candidateDictBuffer);
+ return COVER_dictSelectionError(totalCompressedSize);
+ }
+
+ if (params.shrinkDict == 0) {
+ COVER_dictSelection_t selection = { largestDictbuffer, dictContentSize, totalCompressedSize };
+ free(candidateDictBuffer);
+ return selection;
+ }
+
+ largestDict = dictContentSize;
+ largestCompressed = totalCompressedSize;
+ dictContentSize = ZDICT_DICTSIZE_MIN;
+
+ /* Largest dict is initially at least ZDICT_DICTSIZE_MIN */
+ while (dictContentSize < largestDict) {
+ memcpy(candidateDictBuffer, largestDictbuffer, largestDict);
+ dictContentSize = ZDICT_finalizeDictionary(
+ candidateDictBuffer, dictContentSize, customDictContentEnd - dictContentSize, dictContentSize,
+ samplesBuffer, samplesSizes, nbFinalizeSamples, params.zParams);
+
+ if (ZDICT_isError(dictContentSize)) {
+ free(largestDictbuffer);
+ free(candidateDictBuffer);
+ return COVER_dictSelectionError(dictContentSize);
+
+ }
+
+ totalCompressedSize = COVER_checkTotalCompressedSize(params, samplesSizes,
+ samplesBuffer, offsets,
+ nbCheckSamples, nbSamples,
+ candidateDictBuffer, dictContentSize);
+
+ if (ZSTD_isError(totalCompressedSize)) {
+ free(largestDictbuffer);
+ free(candidateDictBuffer);
+ return COVER_dictSelectionError(totalCompressedSize);
+ }
+
+ if (totalCompressedSize <= largestCompressed * regressionTolerance) {
+ COVER_dictSelection_t selection = { candidateDictBuffer, dictContentSize, totalCompressedSize };
+ free(largestDictbuffer);
+ return selection;
+ }
+ dictContentSize *= 2;
+ }
+ dictContentSize = largestDict;
+ totalCompressedSize = largestCompressed;
+ {
+ COVER_dictSelection_t selection = { largestDictbuffer, dictContentSize, totalCompressedSize };
+ free(candidateDictBuffer);
+ return selection;
}
}
@@ -832,7 +1050,7 @@
} COVER_tryParameters_data_t;
/**
- * Tries a set of parameters and upates the COVER_best_t with the results.
+ * Tries a set of parameters and updates the COVER_best_t with the results.
* This function is thread safe if zstd is compiled with multithreaded support.
* It takes its parameters as an *OWNING* opaque pointer to support threading.
*/
@@ -846,6 +1064,7 @@
/* Allocate space for hash table, dict, and freqs */
COVER_map_t activeDmers;
BYTE *const dict = (BYTE * const)malloc(dictBufferCapacity);
+ COVER_dictSelection_t selection = COVER_dictSelectionError(ERROR(GENERIC));
U32 *freqs = (U32 *)malloc(ctx->suffixSize * sizeof(U32));
if (!COVER_map_init(&activeDmers, parameters.k - parameters.d + 1)) {
DISPLAYLEVEL(1, "Failed to allocate dmer map: out of memory\n");
@@ -861,68 +1080,21 @@
{
const size_t tail = COVER_buildDictionary(ctx, freqs, &activeDmers, dict,
dictBufferCapacity, parameters);
- dictBufferCapacity = ZDICT_finalizeDictionary(
- dict, dictBufferCapacity, dict + tail, dictBufferCapacity - tail,
- ctx->samples, ctx->samplesSizes, (unsigned)ctx->nbSamples,
- parameters.zParams);
- if (ZDICT_isError(dictBufferCapacity)) {
- DISPLAYLEVEL(1, "Failed to finalize dictionary\n");
+ selection = COVER_selectDict(dict + tail, dictBufferCapacity - tail,
+ ctx->samples, ctx->samplesSizes, (unsigned)ctx->nbTrainSamples, ctx->nbTrainSamples, ctx->nbSamples, parameters, ctx->offsets,
+ totalCompressedSize);
+
+ if (COVER_dictSelectionIsError(selection)) {
+ DISPLAYLEVEL(1, "Failed to select dictionary\n");
goto _cleanup;
}
}
- /* Check total compressed size */
- {
- /* Pointers */
- ZSTD_CCtx *cctx;
- ZSTD_CDict *cdict;
- void *dst;
- /* Local variables */
- size_t dstCapacity;
- size_t i;
- /* Allocate dst with enough space to compress the maximum sized sample */
- {
- size_t maxSampleSize = 0;
- for (i = 0; i < ctx->nbSamples; ++i) {
- maxSampleSize = MAX(ctx->samplesSizes[i], maxSampleSize);
- }
- dstCapacity = ZSTD_compressBound(maxSampleSize);
- dst = malloc(dstCapacity);
- }
- /* Create the cctx and cdict */
- cctx = ZSTD_createCCtx();
- cdict = ZSTD_createCDict(dict, dictBufferCapacity,
- parameters.zParams.compressionLevel);
- if (!dst || !cctx || !cdict) {
- goto _compressCleanup;
- }
- /* Compress each sample and sum their sizes (or error) */
- totalCompressedSize = dictBufferCapacity;
- for (i = 0; i < ctx->nbSamples; ++i) {
- const size_t size = ZSTD_compress_usingCDict(
- cctx, dst, dstCapacity, ctx->samples + ctx->offsets[i],
- ctx->samplesSizes[i], cdict);
- if (ZSTD_isError(size)) {
- totalCompressedSize = ERROR(GENERIC);
- goto _compressCleanup;
- }
- totalCompressedSize += size;
- }
- _compressCleanup:
- ZSTD_freeCCtx(cctx);
- ZSTD_freeCDict(cdict);
- if (dst) {
- free(dst);
- }
- }
-
_cleanup:
- COVER_best_finish(data->best, totalCompressedSize, parameters, dict,
- dictBufferCapacity);
+ free(dict);
+ COVER_best_finish(data->best, parameters, selection);
free(data);
COVER_map_destroy(&activeDmers);
- if (dict) {
- free(dict);
- }
+ COVER_dictSelectionFree(selection);
if (freqs) {
free(freqs);
}
@@ -934,6 +1106,8 @@
ZDICT_cover_params_t *parameters) {
/* constants */
const unsigned nbThreads = parameters->nbThreads;
+ const double splitPoint =
+ parameters->splitPoint <= 0.0 ? DEFAULT_SPLITPOINT : parameters->splitPoint;
const unsigned kMinD = parameters->d == 0 ? 6 : parameters->d;
const unsigned kMaxD = parameters->d == 0 ? 8 : parameters->d;
const unsigned kMinK = parameters->k == 0 ? 50 : parameters->k;
@@ -942,6 +1116,7 @@
const unsigned kStepSize = MAX((kMaxK - kMinK) / kSteps, 1);
const unsigned kIterations =
(1 + (kMaxD - kMinD) / 2) * (1 + (kMaxK - kMinK) / kStepSize);
+ const unsigned shrinkDict = 0;
/* Local variables */
const int displayLevel = parameters->zParams.notificationLevel;
unsigned iteration = 1;
@@ -949,15 +1124,20 @@
unsigned k;
COVER_best_t best;
POOL_ctx *pool = NULL;
+ int warned = 0;
/* Checks */
+ if (splitPoint <= 0 || splitPoint > 1) {
+ LOCALDISPLAYLEVEL(displayLevel, 1, "Incorrect parameters\n");
+ return ERROR(parameter_outOfBound);
+ }
if (kMinK < kMaxD || kMaxK < kMinK) {
LOCALDISPLAYLEVEL(displayLevel, 1, "Incorrect parameters\n");
- return ERROR(GENERIC);
+ return ERROR(parameter_outOfBound);
}
if (nbSamples == 0) {
DISPLAYLEVEL(1, "Cover must have at least one input file\n");
- return ERROR(GENERIC);
+ return ERROR(srcSize_wrong);
}
if (dictBufferCapacity < ZDICT_DICTSIZE_MIN) {
DISPLAYLEVEL(1, "dictBufferCapacity must be at least %u\n",
@@ -981,11 +1161,18 @@
/* Initialize the context for this value of d */
COVER_ctx_t ctx;
LOCALDISPLAYLEVEL(displayLevel, 3, "d=%u\n", d);
- if (!COVER_ctx_init(&ctx, samplesBuffer, samplesSizes, nbSamples, d)) {
- LOCALDISPLAYLEVEL(displayLevel, 1, "Failed to initialize context\n");
- COVER_best_destroy(&best);
- POOL_free(pool);
- return ERROR(GENERIC);
+ {
+ const size_t initVal = COVER_ctx_init(&ctx, samplesBuffer, samplesSizes, nbSamples, d, splitPoint);
+ if (ZSTD_isError(initVal)) {
+ LOCALDISPLAYLEVEL(displayLevel, 1, "Failed to initialize context\n");
+ COVER_best_destroy(&best);
+ POOL_free(pool);
+ return initVal;
+ }
+ }
+ if (!warned) {
+ COVER_warnOnSmallCorpus(dictBufferCapacity, ctx.suffixSize, displayLevel);
+ warned = 1;
}
/* Loop through k reusing the same context */
for (k = kMinK; k <= kMaxK; k += kStepSize) {
@@ -998,7 +1185,7 @@
COVER_best_destroy(&best);
COVER_ctx_destroy(&ctx);
POOL_free(pool);
- return ERROR(GENERIC);
+ return ERROR(memory_allocation);
}
data->ctx = &ctx;
data->best = &best;
@@ -1006,7 +1193,9 @@
data->parameters = *parameters;
data->parameters.k = k;
data->parameters.d = d;
+ data->parameters.splitPoint = splitPoint;
data->parameters.steps = kSteps;
+ data->parameters.shrinkDict = shrinkDict;
data->parameters.zParams.notificationLevel = g_displayLevel;
/* Check the parameters */
if (!COVER_checkParameters(data->parameters, dictBufferCapacity)) {
@@ -1023,7 +1212,7 @@
}
/* Print status */
LOCALDISPLAYUPDATE(displayLevel, 2, "\r%u%% ",
- (U32)((iteration * 100) / kIterations));
+ (unsigned)((iteration * 100) / kIterations));
++iteration;
}
COVER_best_wait(&best);
diff --git a/vendor/github.com/DataDog/zstd/cover.h b/vendor/github.com/DataDog/zstd/cover.h
new file mode 100644
index 0000000..d9e0636
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/cover.h
@@ -0,0 +1,147 @@
+#include <stdio.h> /* fprintf */
+#include <stdlib.h> /* malloc, free, qsort */
+#include <string.h> /* memset */
+#include <time.h> /* clock */
+#include "mem.h" /* read */
+#include "pool.h"
+#include "threading.h"
+#include "zstd_internal.h" /* includes zstd.h */
+#ifndef ZDICT_STATIC_LINKING_ONLY
+#define ZDICT_STATIC_LINKING_ONLY
+#endif
+#include "zdict.h"
+
+/**
+ * COVER_best_t is used for two purposes:
+ * 1. Synchronizing threads.
+ * 2. Saving the best parameters and dictionary.
+ *
+ * All of the methods except COVER_best_init() are thread safe if zstd is
+ * compiled with multithreaded support.
+ */
+typedef struct COVER_best_s {
+ ZSTD_pthread_mutex_t mutex;
+ ZSTD_pthread_cond_t cond;
+ size_t liveJobs;
+ void *dict;
+ size_t dictSize;
+ ZDICT_cover_params_t parameters;
+ size_t compressedSize;
+} COVER_best_t;
+
+/**
+ * A segment is a range in the source as well as the score of the segment.
+ */
+typedef struct {
+ U32 begin;
+ U32 end;
+ U32 score;
+} COVER_segment_t;
+
+/**
+ *Number of epochs and size of each epoch.
+ */
+typedef struct {
+ U32 num;
+ U32 size;
+} COVER_epoch_info_t;
+
+/**
+ * Struct used for the dictionary selection function.
+ */
+typedef struct COVER_dictSelection {
+ BYTE* dictContent;
+ size_t dictSize;
+ size_t totalCompressedSize;
+} COVER_dictSelection_t;
+
+/**
+ * Computes the number of epochs and the size of each epoch.
+ * We will make sure that each epoch gets at least 10 * k bytes.
+ *
+ * The COVER algorithms divide the data up into epochs of equal size and
+ * select one segment from each epoch.
+ *
+ * @param maxDictSize The maximum allowed dictionary size.
+ * @param nbDmers The number of dmers we are training on.
+ * @param k The parameter k (segment size).
+ * @param passes The target number of passes over the dmer corpus.
+ * More passes means a better dictionary.
+ */
+COVER_epoch_info_t COVER_computeEpochs(U32 maxDictSize, U32 nbDmers,
+ U32 k, U32 passes);
+
+/**
+ * Warns the user when their corpus is too small.
+ */
+void COVER_warnOnSmallCorpus(size_t maxDictSize, size_t nbDmers, int displayLevel);
+
+/**
+ * Checks total compressed size of a dictionary
+ */
+size_t COVER_checkTotalCompressedSize(const ZDICT_cover_params_t parameters,
+ const size_t *samplesSizes, const BYTE *samples,
+ size_t *offsets,
+ size_t nbTrainSamples, size_t nbSamples,
+ BYTE *const dict, size_t dictBufferCapacity);
+
+/**
+ * Returns the sum of the sample sizes.
+ */
+size_t COVER_sum(const size_t *samplesSizes, unsigned nbSamples) ;
+
+/**
+ * Initialize the `COVER_best_t`.
+ */
+void COVER_best_init(COVER_best_t *best);
+
+/**
+ * Wait until liveJobs == 0.
+ */
+void COVER_best_wait(COVER_best_t *best);
+
+/**
+ * Call COVER_best_wait() and then destroy the COVER_best_t.
+ */
+void COVER_best_destroy(COVER_best_t *best);
+
+/**
+ * Called when a thread is about to be launched.
+ * Increments liveJobs.
+ */
+void COVER_best_start(COVER_best_t *best);
+
+/**
+ * Called when a thread finishes executing, both on error or success.
+ * Decrements liveJobs and signals any waiting threads if liveJobs == 0.
+ * If this dictionary is the best so far save it and its parameters.
+ */
+void COVER_best_finish(COVER_best_t *best, ZDICT_cover_params_t parameters,
+ COVER_dictSelection_t selection);
+/**
+ * Error function for COVER_selectDict function. Checks if the return
+ * value is an error.
+ */
+unsigned COVER_dictSelectionIsError(COVER_dictSelection_t selection);
+
+ /**
+ * Error function for COVER_selectDict function. Returns a struct where
+ * return.totalCompressedSize is a ZSTD error.
+ */
+COVER_dictSelection_t COVER_dictSelectionError(size_t error);
+
+/**
+ * Always call after selectDict is called to free up used memory from
+ * newly created dictionary.
+ */
+void COVER_dictSelectionFree(COVER_dictSelection_t selection);
+
+/**
+ * Called to finalize the dictionary and select one based on whether or not
+ * the shrink-dict flag was enabled. If enabled the dictionary used is the
+ * smallest dictionary within a specified regression of the compressed size
+ * from the largest dictionary.
+ */
+ COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent,
+ size_t dictContentSize, const BYTE* samplesBuffer, const size_t* samplesSizes, unsigned nbFinalizeSamples,
+ size_t nbCheckSamples, size_t nbSamples, ZDICT_cover_params_t params, size_t* offsets, size_t totalCompressedSize);
diff --git a/vendor/github.com/DataDog/zstd/cpu.h b/vendor/github.com/DataDog/zstd/cpu.h
index 4eb48e3..5f0923f 100644
--- a/vendor/github.com/DataDog/zstd/cpu.h
+++ b/vendor/github.com/DataDog/zstd/cpu.h
@@ -36,7 +36,7 @@
U32 f1d = 0;
U32 f7b = 0;
U32 f7c = 0;
-#ifdef _MSC_VER
+#if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_IX86))
int reg[4];
__cpuid((int*)reg, 0);
{
@@ -72,14 +72,13 @@
"cpuid\n\t"
"popl %%ebx\n\t"
: "=a"(f1a), "=c"(f1c), "=d"(f1d)
- : "a"(1)
- :);
+ : "a"(1));
}
if (n >= 7) {
__asm__(
"pushl %%ebx\n\t"
"cpuid\n\t"
- "movl %%ebx, %%eax\n\r"
+ "movl %%ebx, %%eax\n\t"
"popl %%ebx"
: "=a"(f7b), "=c"(f7c)
: "a"(7), "c"(0)
diff --git a/vendor/github.com/DataDog/zstd/debug.c b/vendor/github.com/DataDog/zstd/debug.c
new file mode 100644
index 0000000..3ebdd1c
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/debug.c
@@ -0,0 +1,44 @@
+/* ******************************************************************
+ debug
+ Part of FSE library
+ Copyright (C) 2013-present, Yann Collet.
+
+ BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+ Redistribution and use in source and binary forms, with or without
+ modification, are permitted provided that the following conditions are
+ met:
+
+ * Redistributions of source code must retain the above copyright
+ notice, this list of conditions and the following disclaimer.
+ * Redistributions in binary form must reproduce the above
+ copyright notice, this list of conditions and the following disclaimer
+ in the documentation and/or other materials provided with the
+ distribution.
+
+ THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+ You can contact the author at :
+ - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
+****************************************************************** */
+
+
+/*
+ * This module only hosts one global variable
+ * which can be used to dynamically influence the verbosity of traces,
+ * such as DEBUGLOG and RAWLOG
+ */
+
+#include "debug.h"
+
+int g_debuglevel = DEBUGLEVEL;
diff --git a/vendor/github.com/DataDog/zstd/debug.h b/vendor/github.com/DataDog/zstd/debug.h
new file mode 100644
index 0000000..b4fc89d
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/debug.h
@@ -0,0 +1,134 @@
+/* ******************************************************************
+ debug
+ Part of FSE library
+ Copyright (C) 2013-present, Yann Collet.
+
+ BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+ Redistribution and use in source and binary forms, with or without
+ modification, are permitted provided that the following conditions are
+ met:
+
+ * Redistributions of source code must retain the above copyright
+ notice, this list of conditions and the following disclaimer.
+ * Redistributions in binary form must reproduce the above
+ copyright notice, this list of conditions and the following disclaimer
+ in the documentation and/or other materials provided with the
+ distribution.
+
+ THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+ You can contact the author at :
+ - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
+****************************************************************** */
+
+
+/*
+ * The purpose of this header is to enable debug functions.
+ * They regroup assert(), DEBUGLOG() and RAWLOG() for run-time,
+ * and DEBUG_STATIC_ASSERT() for compile-time.
+ *
+ * By default, DEBUGLEVEL==0, which means run-time debug is disabled.
+ *
+ * Level 1 enables assert() only.
+ * Starting level 2, traces can be generated and pushed to stderr.
+ * The higher the level, the more verbose the traces.
+ *
+ * It's possible to dynamically adjust level using variable g_debug_level,
+ * which is only declared if DEBUGLEVEL>=2,
+ * and is a global variable, not multi-thread protected (use with care)
+ */
+
+#ifndef DEBUG_H_12987983217
+#define DEBUG_H_12987983217
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+
+/* static assert is triggered at compile time, leaving no runtime artefact.
+ * static assert only works with compile-time constants.
+ * Also, this variant can only be used inside a function. */
+#define DEBUG_STATIC_ASSERT(c) (void)sizeof(char[(c) ? 1 : -1])
+
+
+/* DEBUGLEVEL is expected to be defined externally,
+ * typically through compiler command line.
+ * Value must be a number. */
+#ifndef DEBUGLEVEL
+# define DEBUGLEVEL 0
+#endif
+
+
+/* DEBUGFILE can be defined externally,
+ * typically through compiler command line.
+ * note : currently useless.
+ * Value must be stderr or stdout */
+#ifndef DEBUGFILE
+# define DEBUGFILE stderr
+#endif
+
+
+/* recommended values for DEBUGLEVEL :
+ * 0 : release mode, no debug, all run-time checks disabled
+ * 1 : enables assert() only, no display
+ * 2 : reserved, for currently active debug path
+ * 3 : events once per object lifetime (CCtx, CDict, etc.)
+ * 4 : events once per frame
+ * 5 : events once per block
+ * 6 : events once per sequence (verbose)
+ * 7+: events at every position (*very* verbose)
+ *
+ * It's generally inconvenient to output traces > 5.
+ * In which case, it's possible to selectively trigger high verbosity levels
+ * by modifying g_debug_level.
+ */
+
+#if (DEBUGLEVEL>=1)
+# include <assert.h>
+#else
+# ifndef assert /* assert may be already defined, due to prior #include <assert.h> */
+# define assert(condition) ((void)0) /* disable assert (default) */
+# endif
+#endif
+
+#if (DEBUGLEVEL>=2)
+# include <stdio.h>
+extern int g_debuglevel; /* the variable is only declared,
+ it actually lives in debug.c,
+ and is shared by the whole process.
+ It's not thread-safe.
+ It's useful when enabling very verbose levels
+ on selective conditions (such as position in src) */
+
+# define RAWLOG(l, ...) { \
+ if (l<=g_debuglevel) { \
+ fprintf(stderr, __VA_ARGS__); \
+ } }
+# define DEBUGLOG(l, ...) { \
+ if (l<=g_debuglevel) { \
+ fprintf(stderr, __FILE__ ": " __VA_ARGS__); \
+ fprintf(stderr, " \n"); \
+ } }
+#else
+# define RAWLOG(l, ...) {} /* disabled */
+# define DEBUGLOG(l, ...) {} /* disabled */
+#endif
+
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif /* DEBUG_H_12987983217 */
diff --git a/vendor/github.com/DataDog/zstd/divsufsort.c b/vendor/github.com/DataDog/zstd/divsufsort.c
index 60cceb0..ead9220 100644
--- a/vendor/github.com/DataDog/zstd/divsufsort.c
+++ b/vendor/github.com/DataDog/zstd/divsufsort.c
@@ -1637,7 +1637,7 @@
if(0 <= c2) { BUCKET_B(c2, c1) = k - SA; }
k = SA + BUCKET_B(c2 = c0, c1);
}
- assert(k < j);
+ assert(k < j); assert(k != NULL);
*k-- = s;
} else {
assert(((s == 0) && (T[s] == c1)) || (s < 0));
@@ -1701,7 +1701,7 @@
if(0 <= c2) { BUCKET_B(c2, c1) = k - SA; }
k = SA + BUCKET_B(c2 = c0, c1);
}
- assert(k < j);
+ assert(k < j); assert(k != NULL);
*k-- = s;
} else if(s != 0) {
*j = ~s;
@@ -1785,7 +1785,7 @@
if(0 <= c2) { BUCKET_B(c2, c1) = k - SA; }
k = SA + BUCKET_B(c2 = c0, c1);
}
- assert(k < j);
+ assert(k < j); assert(k != NULL);
*k-- = s;
} else if(s != 0) {
*j = ~s;
diff --git a/vendor/github.com/DataDog/zstd/entropy_common.c b/vendor/github.com/DataDog/zstd/entropy_common.c
index b37a082..b12944e 100644
--- a/vendor/github.com/DataDog/zstd/entropy_common.c
+++ b/vendor/github.com/DataDog/zstd/entropy_common.c
@@ -72,7 +72,21 @@
unsigned charnum = 0;
int previous0 = 0;
- if (hbSize < 4) return ERROR(srcSize_wrong);
+ if (hbSize < 4) {
+ /* This function only works when hbSize >= 4 */
+ char buffer[4];
+ memset(buffer, 0, sizeof(buffer));
+ memcpy(buffer, headerBuffer, hbSize);
+ { size_t const countSize = FSE_readNCount(normalizedCounter, maxSVPtr, tableLogPtr,
+ buffer, sizeof(buffer));
+ if (FSE_isError(countSize)) return countSize;
+ if (countSize > hbSize) return ERROR(corruption_detected);
+ return countSize;
+ } }
+ assert(hbSize >= 4);
+
+ /* init */
+ memset(normalizedCounter, 0, (*maxSVPtr+1) * sizeof(normalizedCounter[0])); /* all symbols not present in NCount have a frequency of 0 */
bitStream = MEM_readLE32(ip);
nbBits = (bitStream & 0xF) + FSE_MIN_TABLELOG; /* extract tableLog */
if (nbBits > FSE_TABLELOG_ABSOLUTE_MAX) return ERROR(tableLog_tooLarge);
@@ -105,6 +119,7 @@
if (n0 > *maxSVPtr) return ERROR(maxSymbolValue_tooSmall);
while (charnum < n0) normalizedCounter[charnum++] = 0;
if ((ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
+ assert((bitCount >> 3) <= 3); /* For first condition to work */
ip += bitCount>>3;
bitCount &= 7;
bitStream = MEM_readLE32(ip) >> bitCount;
diff --git a/vendor/github.com/DataDog/zstd/error_private.c b/vendor/github.com/DataDog/zstd/error_private.c
index d004ee6..7c1bb67 100644
--- a/vendor/github.com/DataDog/zstd/error_private.c
+++ b/vendor/github.com/DataDog/zstd/error_private.c
@@ -14,6 +14,10 @@
const char* ERR_getErrorString(ERR_enum code)
{
+#ifdef ZSTD_STRIP_ERROR_STRINGS
+ (void)code;
+ return "Error strings stripped";
+#else
static const char* const notErrorCode = "Unspecified error code";
switch( code )
{
@@ -39,10 +43,12 @@
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";
/* 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(maxCode):
default: return notErrorCode;
}
+#endif
}
diff --git a/vendor/github.com/DataDog/zstd/fastcover.c b/vendor/github.com/DataDog/zstd/fastcover.c
new file mode 100644
index 0000000..941bb5a
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/fastcover.c
@@ -0,0 +1,747 @@
+/*-*************************************
+* Dependencies
+***************************************/
+#include <stdio.h> /* fprintf */
+#include <stdlib.h> /* malloc, free, qsort */
+#include <string.h> /* memset */
+#include <time.h> /* clock */
+
+#include "mem.h" /* read */
+#include "pool.h"
+#include "threading.h"
+#include "cover.h"
+#include "zstd_internal.h" /* includes zstd.h */
+#ifndef ZDICT_STATIC_LINKING_ONLY
+#define ZDICT_STATIC_LINKING_ONLY
+#endif
+#include "zdict.h"
+
+
+/*-*************************************
+* Constants
+***************************************/
+#define FASTCOVER_MAX_SAMPLES_SIZE (sizeof(size_t) == 8 ? ((unsigned)-1) : ((unsigned)1 GB))
+#define FASTCOVER_MAX_F 31
+#define FASTCOVER_MAX_ACCEL 10
+#define DEFAULT_SPLITPOINT 0.75
+#define DEFAULT_F 20
+#define DEFAULT_ACCEL 1
+
+
+/*-*************************************
+* Console display
+***************************************/
+static int g_displayLevel = 2;
+#define DISPLAY(...) \
+ { \
+ fprintf(stderr, __VA_ARGS__); \
+ fflush(stderr); \
+ }
+#define LOCALDISPLAYLEVEL(displayLevel, l, ...) \
+ if (displayLevel >= l) { \
+ DISPLAY(__VA_ARGS__); \
+ } /* 0 : no display; 1: errors; 2: default; 3: details; 4: debug */
+#define DISPLAYLEVEL(l, ...) LOCALDISPLAYLEVEL(g_displayLevel, l, __VA_ARGS__)
+
+#define LOCALDISPLAYUPDATE(displayLevel, l, ...) \
+ if (displayLevel >= l) { \
+ if ((clock() - g_time > refreshRate) || (displayLevel >= 4)) { \
+ g_time = clock(); \
+ DISPLAY(__VA_ARGS__); \
+ } \
+ }
+#define DISPLAYUPDATE(l, ...) LOCALDISPLAYUPDATE(g_displayLevel, l, __VA_ARGS__)
+static const clock_t refreshRate = CLOCKS_PER_SEC * 15 / 100;
+static clock_t g_time = 0;
+
+
+/*-*************************************
+* Hash Functions
+***************************************/
+static const U64 prime6bytes = 227718039650203ULL;
+static size_t ZSTD_hash6(U64 u, U32 h) { return (size_t)(((u << (64-48)) * prime6bytes) >> (64-h)) ; }
+static size_t ZSTD_hash6Ptr(const void* p, U32 h) { return ZSTD_hash6(MEM_readLE64(p), h); }
+
+static const U64 prime8bytes = 0xCF1BBCDCB7A56463ULL;
+static size_t ZSTD_hash8(U64 u, U32 h) { return (size_t)(((u) * prime8bytes) >> (64-h)) ; }
+static size_t ZSTD_hash8Ptr(const void* p, U32 h) { return ZSTD_hash8(MEM_readLE64(p), h); }
+
+
+/**
+ * Hash the d-byte value pointed to by p and mod 2^f
+ */
+static size_t FASTCOVER_hashPtrToIndex(const void* p, U32 h, unsigned d) {
+ if (d == 6) {
+ return ZSTD_hash6Ptr(p, h) & ((1 << h) - 1);
+ }
+ return ZSTD_hash8Ptr(p, h) & ((1 << h) - 1);
+}
+
+
+/*-*************************************
+* Acceleration
+***************************************/
+typedef struct {
+ unsigned finalize; /* Percentage of training samples used for ZDICT_finalizeDictionary */
+ unsigned skip; /* Number of dmer skipped between each dmer counted in computeFrequency */
+} FASTCOVER_accel_t;
+
+
+static const FASTCOVER_accel_t FASTCOVER_defaultAccelParameters[FASTCOVER_MAX_ACCEL+1] = {
+ { 100, 0 }, /* accel = 0, should not happen because accel = 0 defaults to accel = 1 */
+ { 100, 0 }, /* accel = 1 */
+ { 50, 1 }, /* accel = 2 */
+ { 34, 2 }, /* accel = 3 */
+ { 25, 3 }, /* accel = 4 */
+ { 20, 4 }, /* accel = 5 */
+ { 17, 5 }, /* accel = 6 */
+ { 14, 6 }, /* accel = 7 */
+ { 13, 7 }, /* accel = 8 */
+ { 11, 8 }, /* accel = 9 */
+ { 10, 9 }, /* accel = 10 */
+};
+
+
+/*-*************************************
+* Context
+***************************************/
+typedef struct {
+ const BYTE *samples;
+ size_t *offsets;
+ const size_t *samplesSizes;
+ size_t nbSamples;
+ size_t nbTrainSamples;
+ size_t nbTestSamples;
+ size_t nbDmers;
+ U32 *freqs;
+ unsigned d;
+ unsigned f;
+ FASTCOVER_accel_t accelParams;
+} FASTCOVER_ctx_t;
+
+
+/*-*************************************
+* Helper functions
+***************************************/
+/**
+ * Selects the best segment in an epoch.
+ * Segments of are scored according to the function:
+ *
+ * Let F(d) be the frequency of all dmers with hash value d.
+ * Let S_i be hash value of the dmer at position i of segment S which has length k.
+ *
+ * Score(S) = F(S_1) + F(S_2) + ... + F(S_{k-d+1})
+ *
+ * Once the dmer with hash value d is in the dictionary we set F(d) = 0.
+ */
+static COVER_segment_t FASTCOVER_selectSegment(const FASTCOVER_ctx_t *ctx,
+ U32 *freqs, U32 begin, U32 end,
+ ZDICT_cover_params_t parameters,
+ U16* segmentFreqs) {
+ /* Constants */
+ const U32 k = parameters.k;
+ const U32 d = parameters.d;
+ const U32 f = ctx->f;
+ const U32 dmersInK = k - d + 1;
+
+ /* Try each segment (activeSegment) and save the best (bestSegment) */
+ COVER_segment_t bestSegment = {0, 0, 0};
+ COVER_segment_t activeSegment;
+
+ /* Reset the activeDmers in the segment */
+ /* The activeSegment starts at the beginning of the epoch. */
+ activeSegment.begin = begin;
+ activeSegment.end = begin;
+ activeSegment.score = 0;
+
+ /* Slide the activeSegment through the whole epoch.
+ * Save the best segment in bestSegment.
+ */
+ while (activeSegment.end < end) {
+ /* Get hash value of current dmer */
+ const size_t idx = FASTCOVER_hashPtrToIndex(ctx->samples + activeSegment.end, f, d);
+
+ /* Add frequency of this index to score if this is the first occurrence of index in active segment */
+ if (segmentFreqs[idx] == 0) {
+ activeSegment.score += freqs[idx];
+ }
+ /* Increment end of segment and segmentFreqs*/
+ activeSegment.end += 1;
+ segmentFreqs[idx] += 1;
+ /* If the window is now too large, drop the first position */
+ if (activeSegment.end - activeSegment.begin == dmersInK + 1) {
+ /* Get hash value of the dmer to be eliminated from active segment */
+ const size_t delIndex = FASTCOVER_hashPtrToIndex(ctx->samples + activeSegment.begin, f, d);
+ segmentFreqs[delIndex] -= 1;
+ /* Subtract frequency of this index from score if this is the last occurrence of this index in active segment */
+ if (segmentFreqs[delIndex] == 0) {
+ activeSegment.score -= freqs[delIndex];
+ }
+ /* Increment start of segment */
+ activeSegment.begin += 1;
+ }
+
+ /* If this segment is the best so far save it */
+ if (activeSegment.score > bestSegment.score) {
+ bestSegment = activeSegment;
+ }
+ }
+
+ /* Zero out rest of segmentFreqs array */
+ while (activeSegment.begin < end) {
+ const size_t delIndex = FASTCOVER_hashPtrToIndex(ctx->samples + activeSegment.begin, f, d);
+ segmentFreqs[delIndex] -= 1;
+ activeSegment.begin += 1;
+ }
+
+ {
+ /* Zero the frequency of hash value of each dmer covered by the chosen segment. */
+ U32 pos;
+ for (pos = bestSegment.begin; pos != bestSegment.end; ++pos) {
+ const size_t i = FASTCOVER_hashPtrToIndex(ctx->samples + pos, f, d);
+ freqs[i] = 0;
+ }
+ }
+
+ return bestSegment;
+}
+
+
+static int FASTCOVER_checkParameters(ZDICT_cover_params_t parameters,
+ size_t maxDictSize, unsigned f,
+ unsigned accel) {
+ /* k, d, and f are required parameters */
+ if (parameters.d == 0 || parameters.k == 0) {
+ return 0;
+ }
+ /* d has to be 6 or 8 */
+ if (parameters.d != 6 && parameters.d != 8) {
+ return 0;
+ }
+ /* k <= maxDictSize */
+ if (parameters.k > maxDictSize) {
+ return 0;
+ }
+ /* d <= k */
+ if (parameters.d > parameters.k) {
+ return 0;
+ }
+ /* 0 < f <= FASTCOVER_MAX_F*/
+ if (f > FASTCOVER_MAX_F || f == 0) {
+ return 0;
+ }
+ /* 0 < splitPoint <= 1 */
+ if (parameters.splitPoint <= 0 || parameters.splitPoint > 1) {
+ return 0;
+ }
+ /* 0 < accel <= 10 */
+ if (accel > 10 || accel == 0) {
+ return 0;
+ }
+ return 1;
+}
+
+
+/**
+ * Clean up a context initialized with `FASTCOVER_ctx_init()`.
+ */
+static void
+FASTCOVER_ctx_destroy(FASTCOVER_ctx_t* ctx)
+{
+ if (!ctx) return;
+
+ free(ctx->freqs);
+ ctx->freqs = NULL;
+
+ free(ctx->offsets);
+ ctx->offsets = NULL;
+}
+
+
+/**
+ * Calculate for frequency of hash value of each dmer in ctx->samples
+ */
+static void
+FASTCOVER_computeFrequency(U32* freqs, const FASTCOVER_ctx_t* ctx)
+{
+ const unsigned f = ctx->f;
+ const unsigned d = ctx->d;
+ const unsigned skip = ctx->accelParams.skip;
+ const unsigned readLength = MAX(d, 8);
+ size_t i;
+ assert(ctx->nbTrainSamples >= 5);
+ assert(ctx->nbTrainSamples <= ctx->nbSamples);
+ for (i = 0; i < ctx->nbTrainSamples; i++) {
+ size_t start = ctx->offsets[i]; /* start of current dmer */
+ size_t const currSampleEnd = ctx->offsets[i+1];
+ while (start + readLength <= currSampleEnd) {
+ const size_t dmerIndex = FASTCOVER_hashPtrToIndex(ctx->samples + start, f, d);
+ freqs[dmerIndex]++;
+ start = start + skip + 1;
+ }
+ }
+}
+
+
+/**
+ * Prepare a context for dictionary building.
+ * The context is only dependent on the parameter `d` and can used multiple
+ * times.
+ * Returns 0 on success or error code on error.
+ * The context must be destroyed with `FASTCOVER_ctx_destroy()`.
+ */
+static size_t
+FASTCOVER_ctx_init(FASTCOVER_ctx_t* ctx,
+ const void* samplesBuffer,
+ const size_t* samplesSizes, unsigned nbSamples,
+ unsigned d, double splitPoint, unsigned f,
+ FASTCOVER_accel_t accelParams)
+{
+ const BYTE* const samples = (const BYTE*)samplesBuffer;
+ const size_t totalSamplesSize = COVER_sum(samplesSizes, nbSamples);
+ /* Split samples into testing and training sets */
+ const unsigned nbTrainSamples = splitPoint < 1.0 ? (unsigned)((double)nbSamples * splitPoint) : nbSamples;
+ const unsigned nbTestSamples = splitPoint < 1.0 ? nbSamples - nbTrainSamples : nbSamples;
+ const size_t trainingSamplesSize = splitPoint < 1.0 ? COVER_sum(samplesSizes, nbTrainSamples) : totalSamplesSize;
+ const size_t testSamplesSize = splitPoint < 1.0 ? COVER_sum(samplesSizes + nbTrainSamples, nbTestSamples) : totalSamplesSize;
+
+ /* Checks */
+ if (totalSamplesSize < MAX(d, sizeof(U64)) ||
+ totalSamplesSize >= (size_t)FASTCOVER_MAX_SAMPLES_SIZE) {
+ DISPLAYLEVEL(1, "Total samples size is too large (%u MB), maximum size is %u MB\n",
+ (unsigned)(totalSamplesSize >> 20), (FASTCOVER_MAX_SAMPLES_SIZE >> 20));
+ return ERROR(srcSize_wrong);
+ }
+
+ /* Check if there are at least 5 training samples */
+ if (nbTrainSamples < 5) {
+ DISPLAYLEVEL(1, "Total number of training samples is %u and is invalid\n", nbTrainSamples);
+ return ERROR(srcSize_wrong);
+ }
+
+ /* Check if there's testing sample */
+ if (nbTestSamples < 1) {
+ DISPLAYLEVEL(1, "Total number of testing samples is %u and is invalid.\n", nbTestSamples);
+ return ERROR(srcSize_wrong);
+ }
+
+ /* Zero the context */
+ memset(ctx, 0, sizeof(*ctx));
+ DISPLAYLEVEL(2, "Training on %u samples of total size %u\n", nbTrainSamples,
+ (unsigned)trainingSamplesSize);
+ DISPLAYLEVEL(2, "Testing on %u samples of total size %u\n", nbTestSamples,
+ (unsigned)testSamplesSize);
+
+ ctx->samples = samples;
+ ctx->samplesSizes = samplesSizes;
+ ctx->nbSamples = nbSamples;
+ ctx->nbTrainSamples = nbTrainSamples;
+ ctx->nbTestSamples = nbTestSamples;
+ ctx->nbDmers = trainingSamplesSize - MAX(d, sizeof(U64)) + 1;
+ ctx->d = d;
+ ctx->f = f;
+ ctx->accelParams = accelParams;
+
+ /* The offsets of each file */
+ ctx->offsets = (size_t*)calloc((nbSamples + 1), sizeof(size_t));
+ if (ctx->offsets == NULL) {
+ DISPLAYLEVEL(1, "Failed to allocate scratch buffers \n");
+ FASTCOVER_ctx_destroy(ctx);
+ return ERROR(memory_allocation);
+ }
+
+ /* Fill offsets from the samplesSizes */
+ { U32 i;
+ ctx->offsets[0] = 0;
+ assert(nbSamples >= 5);
+ for (i = 1; i <= nbSamples; ++i) {
+ ctx->offsets[i] = ctx->offsets[i - 1] + samplesSizes[i - 1];
+ }
+ }
+
+ /* Initialize frequency array of size 2^f */
+ ctx->freqs = (U32*)calloc(((U64)1 << f), sizeof(U32));
+ if (ctx->freqs == NULL) {
+ DISPLAYLEVEL(1, "Failed to allocate frequency table \n");
+ FASTCOVER_ctx_destroy(ctx);
+ return ERROR(memory_allocation);
+ }
+
+ DISPLAYLEVEL(2, "Computing frequencies\n");
+ FASTCOVER_computeFrequency(ctx->freqs, ctx);
+
+ return 0;
+}
+
+
+/**
+ * Given the prepared context build the dictionary.
+ */
+static size_t
+FASTCOVER_buildDictionary(const FASTCOVER_ctx_t* ctx,
+ U32* freqs,
+ void* dictBuffer, size_t dictBufferCapacity,
+ ZDICT_cover_params_t parameters,
+ U16* segmentFreqs)
+{
+ BYTE *const dict = (BYTE *)dictBuffer;
+ size_t tail = dictBufferCapacity;
+ /* Divide the data into epochs. We will select one segment from each epoch. */
+ const COVER_epoch_info_t epochs = COVER_computeEpochs(
+ (U32)dictBufferCapacity, (U32)ctx->nbDmers, parameters.k, 1);
+ const size_t maxZeroScoreRun = 10;
+ size_t zeroScoreRun = 0;
+ size_t epoch;
+ DISPLAYLEVEL(2, "Breaking content into %u epochs of size %u\n",
+ (U32)epochs.num, (U32)epochs.size);
+ /* Loop through the epochs until there are no more segments or the dictionary
+ * is full.
+ */
+ for (epoch = 0; tail > 0; epoch = (epoch + 1) % epochs.num) {
+ const U32 epochBegin = (U32)(epoch * epochs.size);
+ const U32 epochEnd = epochBegin + epochs.size;
+ size_t segmentSize;
+ /* Select a segment */
+ COVER_segment_t segment = FASTCOVER_selectSegment(
+ ctx, freqs, epochBegin, epochEnd, parameters, segmentFreqs);
+
+ /* If the segment covers no dmers, then we are out of content.
+ * There may be new content in other epochs, for continue for some time.
+ */
+ if (segment.score == 0) {
+ if (++zeroScoreRun >= maxZeroScoreRun) {
+ break;
+ }
+ continue;
+ }
+ zeroScoreRun = 0;
+
+ /* Trim the segment if necessary and if it is too small then we are done */
+ segmentSize = MIN(segment.end - segment.begin + parameters.d - 1, tail);
+ if (segmentSize < parameters.d) {
+ break;
+ }
+
+ /* We fill the dictionary from the back to allow the best segments to be
+ * referenced with the smallest offsets.
+ */
+ tail -= segmentSize;
+ memcpy(dict + tail, ctx->samples + segment.begin, segmentSize);
+ DISPLAYUPDATE(
+ 2, "\r%u%% ",
+ (unsigned)(((dictBufferCapacity - tail) * 100) / dictBufferCapacity));
+ }
+ DISPLAYLEVEL(2, "\r%79s\r", "");
+ return tail;
+}
+
+/**
+ * Parameters for FASTCOVER_tryParameters().
+ */
+typedef struct FASTCOVER_tryParameters_data_s {
+ const FASTCOVER_ctx_t* ctx;
+ COVER_best_t* best;
+ size_t dictBufferCapacity;
+ ZDICT_cover_params_t parameters;
+} FASTCOVER_tryParameters_data_t;
+
+
+/**
+ * Tries a set of parameters and updates the COVER_best_t with the results.
+ * This function is thread safe if zstd is compiled with multithreaded support.
+ * It takes its parameters as an *OWNING* opaque pointer to support threading.
+ */
+static void FASTCOVER_tryParameters(void *opaque)
+{
+ /* Save parameters as local variables */
+ FASTCOVER_tryParameters_data_t *const data = (FASTCOVER_tryParameters_data_t *)opaque;
+ const FASTCOVER_ctx_t *const ctx = data->ctx;
+ const ZDICT_cover_params_t parameters = data->parameters;
+ size_t dictBufferCapacity = data->dictBufferCapacity;
+ size_t totalCompressedSize = ERROR(GENERIC);
+ /* Initialize array to keep track of frequency of dmer within activeSegment */
+ U16* segmentFreqs = (U16 *)calloc(((U64)1 << ctx->f), sizeof(U16));
+ /* Allocate space for hash table, dict, and freqs */
+ BYTE *const dict = (BYTE * const)malloc(dictBufferCapacity);
+ COVER_dictSelection_t selection = COVER_dictSelectionError(ERROR(GENERIC));
+ U32 *freqs = (U32*) malloc(((U64)1 << ctx->f) * sizeof(U32));
+ if (!segmentFreqs || !dict || !freqs) {
+ DISPLAYLEVEL(1, "Failed to allocate buffers: out of memory\n");
+ goto _cleanup;
+ }
+ /* Copy the frequencies because we need to modify them */
+ memcpy(freqs, ctx->freqs, ((U64)1 << ctx->f) * sizeof(U32));
+ /* Build the dictionary */
+ { const size_t tail = FASTCOVER_buildDictionary(ctx, freqs, dict, dictBufferCapacity,
+ parameters, segmentFreqs);
+
+ const unsigned nbFinalizeSamples = (unsigned)(ctx->nbTrainSamples * ctx->accelParams.finalize / 100);
+ selection = COVER_selectDict(dict + tail, dictBufferCapacity - tail,
+ ctx->samples, ctx->samplesSizes, nbFinalizeSamples, ctx->nbTrainSamples, ctx->nbSamples, parameters, ctx->offsets,
+ totalCompressedSize);
+
+ if (COVER_dictSelectionIsError(selection)) {
+ DISPLAYLEVEL(1, "Failed to select dictionary\n");
+ goto _cleanup;
+ }
+ }
+_cleanup:
+ free(dict);
+ COVER_best_finish(data->best, parameters, selection);
+ free(data);
+ free(segmentFreqs);
+ COVER_dictSelectionFree(selection);
+ free(freqs);
+}
+
+
+static void
+FASTCOVER_convertToCoverParams(ZDICT_fastCover_params_t fastCoverParams,
+ ZDICT_cover_params_t* coverParams)
+{
+ coverParams->k = fastCoverParams.k;
+ coverParams->d = fastCoverParams.d;
+ coverParams->steps = fastCoverParams.steps;
+ coverParams->nbThreads = fastCoverParams.nbThreads;
+ coverParams->splitPoint = fastCoverParams.splitPoint;
+ coverParams->zParams = fastCoverParams.zParams;
+ coverParams->shrinkDict = fastCoverParams.shrinkDict;
+}
+
+
+static void
+FASTCOVER_convertToFastCoverParams(ZDICT_cover_params_t coverParams,
+ ZDICT_fastCover_params_t* fastCoverParams,
+ unsigned f, unsigned accel)
+{
+ fastCoverParams->k = coverParams.k;
+ fastCoverParams->d = coverParams.d;
+ fastCoverParams->steps = coverParams.steps;
+ fastCoverParams->nbThreads = coverParams.nbThreads;
+ fastCoverParams->splitPoint = coverParams.splitPoint;
+ fastCoverParams->f = f;
+ fastCoverParams->accel = accel;
+ fastCoverParams->zParams = coverParams.zParams;
+ fastCoverParams->shrinkDict = coverParams.shrinkDict;
+}
+
+
+ZDICTLIB_API size_t
+ZDICT_trainFromBuffer_fastCover(void* dictBuffer, size_t dictBufferCapacity,
+ const void* samplesBuffer,
+ const size_t* samplesSizes, unsigned nbSamples,
+ ZDICT_fastCover_params_t parameters)
+{
+ BYTE* const dict = (BYTE*)dictBuffer;
+ FASTCOVER_ctx_t ctx;
+ ZDICT_cover_params_t coverParams;
+ FASTCOVER_accel_t accelParams;
+ /* Initialize global data */
+ g_displayLevel = parameters.zParams.notificationLevel;
+ /* Assign splitPoint and f if not provided */
+ parameters.splitPoint = 1.0;
+ parameters.f = parameters.f == 0 ? DEFAULT_F : parameters.f;
+ parameters.accel = parameters.accel == 0 ? DEFAULT_ACCEL : parameters.accel;
+ /* Convert to cover parameter */
+ memset(&coverParams, 0 , sizeof(coverParams));
+ FASTCOVER_convertToCoverParams(parameters, &coverParams);
+ /* Checks */
+ if (!FASTCOVER_checkParameters(coverParams, dictBufferCapacity, parameters.f,
+ parameters.accel)) {
+ DISPLAYLEVEL(1, "FASTCOVER parameters incorrect\n");
+ return ERROR(parameter_outOfBound);
+ }
+ if (nbSamples == 0) {
+ DISPLAYLEVEL(1, "FASTCOVER must have at least one input file\n");
+ return ERROR(srcSize_wrong);
+ }
+ if (dictBufferCapacity < ZDICT_DICTSIZE_MIN) {
+ DISPLAYLEVEL(1, "dictBufferCapacity must be at least %u\n",
+ ZDICT_DICTSIZE_MIN);
+ return ERROR(dstSize_tooSmall);
+ }
+ /* Assign corresponding FASTCOVER_accel_t to accelParams*/
+ accelParams = FASTCOVER_defaultAccelParameters[parameters.accel];
+ /* Initialize context */
+ {
+ size_t const initVal = FASTCOVER_ctx_init(&ctx, samplesBuffer, samplesSizes, nbSamples,
+ coverParams.d, parameters.splitPoint, parameters.f,
+ accelParams);
+ if (ZSTD_isError(initVal)) {
+ DISPLAYLEVEL(1, "Failed to initialize context\n");
+ return initVal;
+ }
+ }
+ COVER_warnOnSmallCorpus(dictBufferCapacity, ctx.nbDmers, g_displayLevel);
+ /* Build the dictionary */
+ DISPLAYLEVEL(2, "Building dictionary\n");
+ {
+ /* Initialize array to keep track of frequency of dmer within activeSegment */
+ U16* segmentFreqs = (U16 *)calloc(((U64)1 << parameters.f), sizeof(U16));
+ const size_t tail = FASTCOVER_buildDictionary(&ctx, ctx.freqs, dictBuffer,
+ dictBufferCapacity, coverParams, segmentFreqs);
+ const unsigned nbFinalizeSamples = (unsigned)(ctx.nbTrainSamples * ctx.accelParams.finalize / 100);
+ const size_t dictionarySize = ZDICT_finalizeDictionary(
+ dict, dictBufferCapacity, dict + tail, dictBufferCapacity - tail,
+ samplesBuffer, samplesSizes, nbFinalizeSamples, coverParams.zParams);
+ if (!ZSTD_isError(dictionarySize)) {
+ DISPLAYLEVEL(2, "Constructed dictionary of size %u\n",
+ (unsigned)dictionarySize);
+ }
+ FASTCOVER_ctx_destroy(&ctx);
+ free(segmentFreqs);
+ return dictionarySize;
+ }
+}
+
+
+ZDICTLIB_API size_t
+ZDICT_optimizeTrainFromBuffer_fastCover(
+ void* dictBuffer, size_t dictBufferCapacity,
+ const void* samplesBuffer,
+ const size_t* samplesSizes, unsigned nbSamples,
+ ZDICT_fastCover_params_t* parameters)
+{
+ ZDICT_cover_params_t coverParams;
+ FASTCOVER_accel_t accelParams;
+ /* constants */
+ const unsigned nbThreads = parameters->nbThreads;
+ const double splitPoint =
+ parameters->splitPoint <= 0.0 ? DEFAULT_SPLITPOINT : parameters->splitPoint;
+ const unsigned kMinD = parameters->d == 0 ? 6 : parameters->d;
+ const unsigned kMaxD = parameters->d == 0 ? 8 : parameters->d;
+ const unsigned kMinK = parameters->k == 0 ? 50 : parameters->k;
+ const unsigned kMaxK = parameters->k == 0 ? 2000 : parameters->k;
+ const unsigned kSteps = parameters->steps == 0 ? 40 : parameters->steps;
+ const unsigned kStepSize = MAX((kMaxK - kMinK) / kSteps, 1);
+ const unsigned kIterations =
+ (1 + (kMaxD - kMinD) / 2) * (1 + (kMaxK - kMinK) / kStepSize);
+ const unsigned f = parameters->f == 0 ? DEFAULT_F : parameters->f;
+ const unsigned accel = parameters->accel == 0 ? DEFAULT_ACCEL : parameters->accel;
+ const unsigned shrinkDict = 0;
+ /* Local variables */
+ const int displayLevel = parameters->zParams.notificationLevel;
+ unsigned iteration = 1;
+ unsigned d;
+ unsigned k;
+ COVER_best_t best;
+ POOL_ctx *pool = NULL;
+ int warned = 0;
+ /* Checks */
+ if (splitPoint <= 0 || splitPoint > 1) {
+ LOCALDISPLAYLEVEL(displayLevel, 1, "Incorrect splitPoint\n");
+ return ERROR(parameter_outOfBound);
+ }
+ if (accel == 0 || accel > FASTCOVER_MAX_ACCEL) {
+ LOCALDISPLAYLEVEL(displayLevel, 1, "Incorrect accel\n");
+ return ERROR(parameter_outOfBound);
+ }
+ if (kMinK < kMaxD || kMaxK < kMinK) {
+ LOCALDISPLAYLEVEL(displayLevel, 1, "Incorrect k\n");
+ return ERROR(parameter_outOfBound);
+ }
+ if (nbSamples == 0) {
+ LOCALDISPLAYLEVEL(displayLevel, 1, "FASTCOVER must have at least one input file\n");
+ return ERROR(srcSize_wrong);
+ }
+ if (dictBufferCapacity < ZDICT_DICTSIZE_MIN) {
+ LOCALDISPLAYLEVEL(displayLevel, 1, "dictBufferCapacity must be at least %u\n",
+ ZDICT_DICTSIZE_MIN);
+ return ERROR(dstSize_tooSmall);
+ }
+ if (nbThreads > 1) {
+ pool = POOL_create(nbThreads, 1);
+ if (!pool) {
+ return ERROR(memory_allocation);
+ }
+ }
+ /* Initialization */
+ COVER_best_init(&best);
+ memset(&coverParams, 0 , sizeof(coverParams));
+ FASTCOVER_convertToCoverParams(*parameters, &coverParams);
+ accelParams = FASTCOVER_defaultAccelParameters[accel];
+ /* Turn down global display level to clean up display at level 2 and below */
+ g_displayLevel = displayLevel == 0 ? 0 : displayLevel - 1;
+ /* Loop through d first because each new value needs a new context */
+ LOCALDISPLAYLEVEL(displayLevel, 2, "Trying %u different sets of parameters\n",
+ kIterations);
+ for (d = kMinD; d <= kMaxD; d += 2) {
+ /* Initialize the context for this value of d */
+ FASTCOVER_ctx_t ctx;
+ LOCALDISPLAYLEVEL(displayLevel, 3, "d=%u\n", d);
+ {
+ size_t const initVal = FASTCOVER_ctx_init(&ctx, samplesBuffer, samplesSizes, nbSamples, d, splitPoint, f, accelParams);
+ if (ZSTD_isError(initVal)) {
+ LOCALDISPLAYLEVEL(displayLevel, 1, "Failed to initialize context\n");
+ COVER_best_destroy(&best);
+ POOL_free(pool);
+ return initVal;
+ }
+ }
+ if (!warned) {
+ COVER_warnOnSmallCorpus(dictBufferCapacity, ctx.nbDmers, displayLevel);
+ warned = 1;
+ }
+ /* Loop through k reusing the same context */
+ for (k = kMinK; k <= kMaxK; k += kStepSize) {
+ /* Prepare the arguments */
+ FASTCOVER_tryParameters_data_t *data = (FASTCOVER_tryParameters_data_t *)malloc(
+ sizeof(FASTCOVER_tryParameters_data_t));
+ LOCALDISPLAYLEVEL(displayLevel, 3, "k=%u\n", k);
+ if (!data) {
+ LOCALDISPLAYLEVEL(displayLevel, 1, "Failed to allocate parameters\n");
+ COVER_best_destroy(&best);
+ FASTCOVER_ctx_destroy(&ctx);
+ POOL_free(pool);
+ return ERROR(memory_allocation);
+ }
+ data->ctx = &ctx;
+ data->best = &best;
+ data->dictBufferCapacity = dictBufferCapacity;
+ data->parameters = coverParams;
+ data->parameters.k = k;
+ data->parameters.d = d;
+ data->parameters.splitPoint = splitPoint;
+ data->parameters.steps = kSteps;
+ data->parameters.shrinkDict = shrinkDict;
+ data->parameters.zParams.notificationLevel = g_displayLevel;
+ /* Check the parameters */
+ if (!FASTCOVER_checkParameters(data->parameters, dictBufferCapacity,
+ data->ctx->f, accel)) {
+ DISPLAYLEVEL(1, "FASTCOVER parameters incorrect\n");
+ free(data);
+ continue;
+ }
+ /* Call the function and pass ownership of data to it */
+ COVER_best_start(&best);
+ if (pool) {
+ POOL_add(pool, &FASTCOVER_tryParameters, data);
+ } else {
+ FASTCOVER_tryParameters(data);
+ }
+ /* Print status */
+ LOCALDISPLAYUPDATE(displayLevel, 2, "\r%u%% ",
+ (unsigned)((iteration * 100) / kIterations));
+ ++iteration;
+ }
+ COVER_best_wait(&best);
+ FASTCOVER_ctx_destroy(&ctx);
+ }
+ LOCALDISPLAYLEVEL(displayLevel, 2, "\r%79s\r", "");
+ /* Fill the output buffer and parameters with output of the best parameters */
+ {
+ const size_t dictSize = best.dictSize;
+ if (ZSTD_isError(best.compressedSize)) {
+ const size_t compressedSize = best.compressedSize;
+ COVER_best_destroy(&best);
+ POOL_free(pool);
+ return compressedSize;
+ }
+ FASTCOVER_convertToFastCoverParams(best.parameters, parameters, f, accel);
+ memcpy(dictBuffer, best.dict, dictSize);
+ COVER_best_destroy(&best);
+ POOL_free(pool);
+ return dictSize;
+ }
+
+}
diff --git a/vendor/github.com/DataDog/zstd/fse.h b/vendor/github.com/DataDog/zstd/fse.h
index 6a1d272..811c670 100644
--- a/vendor/github.com/DataDog/zstd/fse.h
+++ b/vendor/github.com/DataDog/zstd/fse.h
@@ -72,6 +72,7 @@
#define FSE_VERSION_NUMBER (FSE_VERSION_MAJOR *100*100 + FSE_VERSION_MINOR *100 + FSE_VERSION_RELEASE)
FSE_PUBLIC_API unsigned FSE_versionNumber(void); /**< library version number; to be used when checking dll version */
+
/*-****************************************
* FSE simple functions
******************************************/
@@ -129,7 +130,7 @@
******************************************/
/*!
FSE_compress() does the following:
-1. count symbol occurrence from source[] into table count[]
+1. count symbol occurrence from source[] into table count[] (see hist.h)
2. normalize counters so that sum(count[]) == Power_of_2 (2^tableLog)
3. save normalized counters to memory buffer using writeNCount()
4. build encoding table 'CTable' from normalized counters
@@ -147,15 +148,6 @@
/* *** COMPRESSION *** */
-/*! FSE_count():
- Provides the precise count of each byte within a table 'count'.
- 'count' is a table of unsigned int, of minimum size (*maxSymbolValuePtr+1).
- *maxSymbolValuePtr will be updated if detected smaller than initial value.
- @return : the count of the most frequent symbol (which is not identified).
- if return == srcSize, there is only one symbol.
- Can also return an error code, which can be tested with FSE_isError(). */
-FSE_PUBLIC_API size_t FSE_count(unsigned* count, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize);
-
/*! FSE_optimalTableLog():
dynamically downsize 'tableLog' when conditions are met.
It saves CPU time, by using smaller tables, while preserving or even improving compression ratio.
@@ -167,7 +159,8 @@
'normalizedCounter' is a table of short, of minimum size (maxSymbolValue+1).
@return : tableLog,
or an errorCode, which can be tested using FSE_isError() */
-FSE_PUBLIC_API size_t FSE_normalizeCount(short* normalizedCounter, unsigned tableLog, const unsigned* count, size_t srcSize, unsigned maxSymbolValue);
+FSE_PUBLIC_API size_t FSE_normalizeCount(short* normalizedCounter, unsigned tableLog,
+ const unsigned* count, size_t srcSize, unsigned maxSymbolValue);
/*! FSE_NCountWriteBound():
Provides the maximum possible size of an FSE normalized table, given 'maxSymbolValue' and 'tableLog'.
@@ -178,8 +171,9 @@
Compactly save 'normalizedCounter' into 'buffer'.
@return : size of the compressed table,
or an errorCode, which can be tested using FSE_isError(). */
-FSE_PUBLIC_API size_t FSE_writeNCount (void* buffer, size_t bufferSize, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog);
-
+FSE_PUBLIC_API size_t FSE_writeNCount (void* buffer, size_t bufferSize,
+ const short* normalizedCounter,
+ unsigned maxSymbolValue, unsigned tableLog);
/*! Constructor and Destructor of FSE_CTable.
Note that FSE_CTable size depends on 'tableLog' and 'maxSymbolValue' */
@@ -250,7 +244,9 @@
@return : size read from 'rBuffer',
or an errorCode, which can be tested using FSE_isError().
maxSymbolValuePtr[0] and tableLogPtr[0] will also be updated with their respective values */
-FSE_PUBLIC_API size_t FSE_readNCount (short* normalizedCounter, unsigned* maxSymbolValuePtr, unsigned* tableLogPtr, const void* rBuffer, size_t rBuffSize);
+FSE_PUBLIC_API size_t FSE_readNCount (short* normalizedCounter,
+ unsigned* maxSymbolValuePtr, unsigned* tableLogPtr,
+ const void* rBuffer, size_t rBuffSize);
/*! Constructor and Destructor of FSE_DTable.
Note that its size depends on 'tableLog' */
@@ -325,33 +321,8 @@
/* *****************************************
-* FSE advanced API
-*******************************************/
-/* FSE_count_wksp() :
- * Same as FSE_count(), but using an externally provided scratch buffer.
- * `workSpace` size must be table of >= `1024` unsigned
- */
-size_t FSE_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
- const void* source, size_t sourceSize, unsigned* workSpace);
-
-/** FSE_countFast() :
- * same as FSE_count(), but blindly trusts that all byte values within src are <= *maxSymbolValuePtr
- */
-size_t FSE_countFast(unsigned* count, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize);
-
-/* FSE_countFast_wksp() :
- * Same as FSE_countFast(), but using an externally provided scratch buffer.
- * `workSpace` must be a table of minimum `1024` unsigned
- */
-size_t FSE_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned* workSpace);
-
-/*! FSE_count_simple() :
- * Same as FSE_countFast(), but does not use any additional memory (not even on stack).
- * This function is unsafe, and will segfault if any value within `src` is `> *maxSymbolValuePtr` (presuming it's also the size of `count`).
-*/
-size_t FSE_count_simple(unsigned* count, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize);
-
-
+ * FSE advanced API
+ ***************************************** */
unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus);
/**< same as FSE_optimalTableLog(), which used `minus==2` */
@@ -387,7 +358,7 @@
typedef enum {
FSE_repeat_none, /**< Cannot use the previous table */
FSE_repeat_check, /**< Can use the previous table but it must be checked */
- FSE_repeat_valid /**< Can use the previous table and it is asumed to be valid */
+ FSE_repeat_valid /**< Can use the previous table and it is assumed to be valid */
} FSE_repeat;
/* *****************************************
@@ -541,7 +512,7 @@
const U32 tableLog = MEM_read16(ptr);
statePtr->value = (ptrdiff_t)1<<tableLog;
statePtr->stateTable = u16ptr+2;
- statePtr->symbolTT = ((const U32*)ct + 1 + (tableLog ? (1<<(tableLog-1)) : 1));
+ statePtr->symbolTT = ct + 1 + (tableLog ? (1<<(tableLog-1)) : 1);
statePtr->stateLog = tableLog;
}
@@ -560,7 +531,7 @@
}
}
-MEM_STATIC void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* statePtr, U32 symbol)
+MEM_STATIC void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* statePtr, unsigned symbol)
{
FSE_symbolCompressionTransform const symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol];
const U16* const stateTable = (const U16*)(statePtr->stateTable);
@@ -576,6 +547,39 @@
}
+/* 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)
+ * 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)
+{
+ const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr;
+ return (symbolTT[symbolValue].deltaNbBits + ((1<<16)-1)) >> 16;
+}
+
+/* FSE_bitCost() :
+ * Approximate symbol cost, as fractional value, using fixed-point format (accuracyLog fractional bits)
+ * 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_bitCost(const void* symbolTTPtr, U32 tableLog, U32 symbolValue, U32 accuracyLog)
+{
+ const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr;
+ U32 const minNbBits = symbolTT[symbolValue].deltaNbBits >> 16;
+ U32 const threshold = (minNbBits+1) << 16;
+ assert(tableLog < 16);
+ assert(accuracyLog < 31-tableLog); /* ensure enough room for renormalization double shift */
+ { U32 const tableSize = 1 << tableLog;
+ U32 const deltaFromThreshold = threshold - (symbolTT[symbolValue].deltaNbBits + tableSize);
+ U32 const normalizedDeltaFromThreshold = (deltaFromThreshold << accuracyLog) >> tableLog; /* linear interpolation (very approximate) */
+ U32 const bitMultiplier = 1 << accuracyLog;
+ assert(symbolTT[symbolValue].deltaNbBits + tableSize <= threshold);
+ assert(normalizedDeltaFromThreshold <= bitMultiplier);
+ return (minNbBits+1)*bitMultiplier - normalizedDeltaFromThreshold;
+ }
+}
+
+
/* ====== Decompression ====== */
typedef struct {
diff --git a/vendor/github.com/DataDog/zstd/fse_compress.c b/vendor/github.com/DataDog/zstd/fse_compress.c
index cb8f1fa..68b47e1 100644
--- a/vendor/github.com/DataDog/zstd/fse_compress.c
+++ b/vendor/github.com/DataDog/zstd/fse_compress.c
@@ -1,6 +1,6 @@
/* ******************************************************************
FSE : Finite State Entropy encoder
- Copyright (C) 2013-2015, Yann Collet.
+ Copyright (C) 2013-present, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
@@ -37,9 +37,11 @@
****************************************************************/
#include <stdlib.h> /* malloc, free, qsort */
#include <string.h> /* memcpy, memset */
-#include <stdio.h> /* printf (debug) */
-#include "bitstream.h"
#include "compiler.h"
+#include "mem.h" /* U32, U16, etc. */
+#include "debug.h" /* assert, DEBUGLOG */
+#include "hist.h" /* HIST_count_wksp */
+#include "bitstream.h"
#define FSE_STATIC_LINKING_ONLY
#include "fse.h"
#include "error_private.h"
@@ -49,7 +51,6 @@
* Error Management
****************************************************************/
#define FSE_isError ERR_isError
-#define FSE_STATIC_ASSERT(c) { enum { FSE_static_assert = 1/(int)(!!(c)) }; } /* use only *after* variable declarations */
/* **************************************************************
@@ -82,7 +83,9 @@
* wkspSize should be sized to handle worst case situation, which is `1<<max_tableLog * sizeof(FSE_FUNCTION_TYPE)`
* workSpace must also be properly aligned with FSE_FUNCTION_TYPE requirements
*/
-size_t FSE_buildCTable_wksp(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
+size_t FSE_buildCTable_wksp(FSE_CTable* ct,
+ const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog,
+ void* workSpace, size_t wkspSize)
{
U32 const tableSize = 1 << tableLog;
U32 const tableMask = tableSize - 1;
@@ -100,14 +103,19 @@
if (((size_t)1 << tableLog) * sizeof(FSE_FUNCTION_TYPE) > wkspSize) return ERROR(tableLog_tooLarge);
tableU16[-2] = (U16) tableLog;
tableU16[-1] = (U16) maxSymbolValue;
+ assert(tableLog < 16); /* required for threshold strategy to work */
/* For explanations on how to distribute symbol values over the table :
- * http://fastcompression.blogspot.fr/2014/02/fse-distributing-symbol-values.html */
+ * http://fastcompression.blogspot.fr/2014/02/fse-distributing-symbol-values.html */
+
+ #ifdef __clang_analyzer__
+ memset(tableSymbol, 0, sizeof(*tableSymbol) * tableSize); /* useless initialization, just to keep scan-build happy */
+ #endif
/* symbol start positions */
{ U32 u;
cumul[0] = 0;
- for (u=1; u<=maxSymbolValue+1; u++) {
+ for (u=1; u <= maxSymbolValue+1; u++) {
if (normalizedCounter[u-1]==-1) { /* Low proba symbol */
cumul[u] = cumul[u-1] + 1;
tableSymbol[highThreshold--] = (FSE_FUNCTION_TYPE)(u-1);
@@ -121,14 +129,16 @@
{ U32 position = 0;
U32 symbol;
for (symbol=0; symbol<=maxSymbolValue; symbol++) {
- int nbOccurences;
- for (nbOccurences=0; nbOccurences<normalizedCounter[symbol]; nbOccurences++) {
+ int nbOccurrences;
+ int const freq = normalizedCounter[symbol];
+ for (nbOccurrences=0; nbOccurrences<freq; nbOccurrences++) {
tableSymbol[position] = (FSE_FUNCTION_TYPE)symbol;
position = (position + step) & tableMask;
- while (position > highThreshold) position = (position + step) & tableMask; /* Low proba area */
+ while (position > highThreshold)
+ position = (position + step) & tableMask; /* Low proba area */
} }
- if (position!=0) return ERROR(GENERIC); /* Must have gone through all positions */
+ assert(position==0); /* Must have initialized all positions */
}
/* Build table */
@@ -143,7 +153,10 @@
for (s=0; s<=maxSymbolValue; s++) {
switch (normalizedCounter[s])
{
- case 0: break;
+ case 0:
+ /* filling nonetheless, for compatibility with FSE_getMaxNbBits() */
+ symbolTT[s].deltaNbBits = ((tableLog+1) << 16) - (1<<tableLog);
+ break;
case -1:
case 1:
@@ -160,6 +173,18 @@
total += normalizedCounter[s];
} } } }
+#if 0 /* debug : symbol costs */
+ DEBUGLOG(5, "\n --- table statistics : ");
+ { U32 symbol;
+ for (symbol=0; symbol<=maxSymbolValue; symbol++) {
+ DEBUGLOG(5, "%3u: w=%3i, maxBits=%u, fracBits=%.2f",
+ symbol, normalizedCounter[symbol],
+ FSE_getMaxNbBits(symbolTT, symbol),
+ (double)FSE_bitCost(symbolTT, tableLog, symbol, 8) / 256);
+ }
+ }
+#endif
+
return 0;
}
@@ -174,8 +199,9 @@
#ifndef FSE_COMMONDEFS_ONLY
+
/*-**************************************************************
-* FSE NCount encoding-decoding
+* FSE NCount encoding
****************************************************************/
size_t FSE_NCountWriteBound(unsigned maxSymbolValue, unsigned tableLog)
{
@@ -183,9 +209,10 @@
return maxSymbolValue ? maxHeaderSize : FSE_NCOUNTBOUND; /* maxSymbolValue==0 ? use default */
}
-static size_t FSE_writeNCount_generic (void* header, size_t headerBufferSize,
- const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog,
- unsigned writeIsSafe)
+static size_t
+FSE_writeNCount_generic (void* header, size_t headerBufferSize,
+ const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog,
+ unsigned writeIsSafe)
{
BYTE* const ostart = (BYTE*) header;
BYTE* out = ostart;
@@ -194,13 +221,12 @@
const int tableSize = 1 << tableLog;
int remaining;
int threshold;
- U32 bitStream;
- int bitCount;
- unsigned charnum = 0;
- int previous0 = 0;
+ U32 bitStream = 0;
+ int bitCount = 0;
+ unsigned symbol = 0;
+ unsigned const alphabetSize = maxSymbolValue + 1;
+ int previousIs0 = 0;
- bitStream = 0;
- bitCount = 0;
/* Table Size */
bitStream += (tableLog-FSE_MIN_TABLELOG) << bitCount;
bitCount += 4;
@@ -210,48 +236,53 @@
threshold = tableSize;
nbBits = tableLog+1;
- while (remaining>1) { /* stops at 1 */
- if (previous0) {
- unsigned start = charnum;
- while (!normalizedCounter[charnum]) charnum++;
- while (charnum >= start+24) {
+ while ((symbol < alphabetSize) && (remaining>1)) { /* stops at 1 */
+ if (previousIs0) {
+ unsigned start = symbol;
+ while ((symbol < alphabetSize) && !normalizedCounter[symbol]) symbol++;
+ if (symbol == alphabetSize) break; /* incorrect distribution */
+ while (symbol >= start+24) {
start+=24;
bitStream += 0xFFFFU << bitCount;
- if ((!writeIsSafe) && (out > oend-2)) return ERROR(dstSize_tooSmall); /* Buffer overflow */
+ if ((!writeIsSafe) && (out > oend-2))
+ return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE) bitStream;
out[1] = (BYTE)(bitStream>>8);
out+=2;
bitStream>>=16;
}
- while (charnum >= start+3) {
+ while (symbol >= start+3) {
start+=3;
bitStream += 3 << bitCount;
bitCount += 2;
}
- bitStream += (charnum-start) << bitCount;
+ bitStream += (symbol-start) << bitCount;
bitCount += 2;
if (bitCount>16) {
- if ((!writeIsSafe) && (out > oend - 2)) return ERROR(dstSize_tooSmall); /* Buffer overflow */
+ if ((!writeIsSafe) && (out > oend - 2))
+ return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE)bitStream;
out[1] = (BYTE)(bitStream>>8);
out += 2;
bitStream >>= 16;
bitCount -= 16;
} }
- { int count = normalizedCounter[charnum++];
- int const max = (2*threshold-1)-remaining;
+ { int count = normalizedCounter[symbol++];
+ int const max = (2*threshold-1) - remaining;
remaining -= count < 0 ? -count : count;
count++; /* +1 for extra accuracy */
- if (count>=threshold) count += max; /* [0..max[ [max..threshold[ (...) [threshold+max 2*threshold[ */
+ if (count>=threshold)
+ count += max; /* [0..max[ [max..threshold[ (...) [threshold+max 2*threshold[ */
bitStream += count << bitCount;
bitCount += nbBits;
bitCount -= (count<max);
- previous0 = (count==1);
+ previousIs0 = (count==1);
if (remaining<1) return ERROR(GENERIC);
while (remaining<threshold) { nbBits--; threshold>>=1; }
}
if (bitCount>16) {
- if ((!writeIsSafe) && (out > oend - 2)) return ERROR(dstSize_tooSmall); /* Buffer overflow */
+ if ((!writeIsSafe) && (out > oend - 2))
+ return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE)bitStream;
out[1] = (BYTE)(bitStream>>8);
out += 2;
@@ -259,19 +290,23 @@
bitCount -= 16;
} }
+ if (remaining != 1)
+ return ERROR(GENERIC); /* incorrect normalized distribution */
+ assert(symbol <= alphabetSize);
+
/* flush remaining bitStream */
- if ((!writeIsSafe) && (out > oend - 2)) return ERROR(dstSize_tooSmall); /* Buffer overflow */
+ if ((!writeIsSafe) && (out > oend - 2))
+ return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE)bitStream;
out[1] = (BYTE)(bitStream>>8);
out+= (bitCount+7) /8;
- if (charnum > maxSymbolValue + 1) return ERROR(GENERIC);
-
return (out-ostart);
}
-size_t FSE_writeNCount (void* buffer, size_t bufferSize, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog)
+size_t FSE_writeNCount (void* buffer, size_t bufferSize,
+ const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog)
{
if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); /* Unsupported */
if (tableLog < FSE_MIN_TABLELOG) return ERROR(GENERIC); /* Unsupported */
@@ -279,179 +314,13 @@
if (bufferSize < FSE_NCountWriteBound(maxSymbolValue, tableLog))
return FSE_writeNCount_generic(buffer, bufferSize, normalizedCounter, maxSymbolValue, tableLog, 0);
- return FSE_writeNCount_generic(buffer, bufferSize, normalizedCounter, maxSymbolValue, tableLog, 1);
+ return FSE_writeNCount_generic(buffer, bufferSize, normalizedCounter, maxSymbolValue, tableLog, 1 /* write in buffer is safe */);
}
-
-/*-**************************************************************
-* Counting histogram
-****************************************************************/
-/*! FSE_count_simple
- This function counts byte values within `src`, and store the histogram into table `count`.
- It doesn't use any additional memory.
- But this function is unsafe : it doesn't check that all values within `src` can fit into `count`.
- For this reason, prefer using a table `count` with 256 elements.
- @return : count of most numerous element.
-*/
-size_t FSE_count_simple(unsigned* count, unsigned* maxSymbolValuePtr,
- const void* src, size_t srcSize)
-{
- const BYTE* ip = (const BYTE*)src;
- const BYTE* const end = ip + srcSize;
- unsigned maxSymbolValue = *maxSymbolValuePtr;
- unsigned max=0;
-
- memset(count, 0, (maxSymbolValue+1)*sizeof(*count));
- if (srcSize==0) { *maxSymbolValuePtr = 0; return 0; }
-
- while (ip<end) {
- assert(*ip <= maxSymbolValue);
- count[*ip++]++;
- }
-
- while (!count[maxSymbolValue]) maxSymbolValue--;
- *maxSymbolValuePtr = maxSymbolValue;
-
- { U32 s; for (s=0; s<=maxSymbolValue; s++) if (count[s] > max) max = count[s]; }
-
- return (size_t)max;
-}
-
-
-/* FSE_count_parallel_wksp() :
- * Same as FSE_count_parallel(), but using an externally provided scratch buffer.
- * `workSpace` size must be a minimum of `1024 * sizeof(unsigned)`.
- * @return : largest histogram frequency, or an error code (notably when histogram would be larger than *maxSymbolValuePtr). */
-static size_t FSE_count_parallel_wksp(
- unsigned* count, unsigned* maxSymbolValuePtr,
- const void* source, size_t sourceSize,
- unsigned checkMax, unsigned* const workSpace)
-{
- const BYTE* ip = (const BYTE*)source;
- const BYTE* const iend = ip+sourceSize;
- unsigned maxSymbolValue = *maxSymbolValuePtr;
- unsigned max=0;
- U32* const Counting1 = workSpace;
- U32* const Counting2 = Counting1 + 256;
- U32* const Counting3 = Counting2 + 256;
- U32* const Counting4 = Counting3 + 256;
-
- memset(workSpace, 0, 4*256*sizeof(unsigned));
-
- /* safety checks */
- if (!sourceSize) {
- memset(count, 0, maxSymbolValue + 1);
- *maxSymbolValuePtr = 0;
- return 0;
- }
- if (!maxSymbolValue) maxSymbolValue = 255; /* 0 == default */
-
- /* by stripes of 16 bytes */
- { U32 cached = MEM_read32(ip); ip += 4;
- while (ip < iend-15) {
- U32 c = cached; cached = MEM_read32(ip); ip += 4;
- Counting1[(BYTE) c ]++;
- Counting2[(BYTE)(c>>8) ]++;
- Counting3[(BYTE)(c>>16)]++;
- Counting4[ c>>24 ]++;
- c = cached; cached = MEM_read32(ip); ip += 4;
- Counting1[(BYTE) c ]++;
- Counting2[(BYTE)(c>>8) ]++;
- Counting3[(BYTE)(c>>16)]++;
- Counting4[ c>>24 ]++;
- c = cached; cached = MEM_read32(ip); ip += 4;
- Counting1[(BYTE) c ]++;
- Counting2[(BYTE)(c>>8) ]++;
- Counting3[(BYTE)(c>>16)]++;
- Counting4[ c>>24 ]++;
- c = cached; cached = MEM_read32(ip); ip += 4;
- Counting1[(BYTE) c ]++;
- Counting2[(BYTE)(c>>8) ]++;
- Counting3[(BYTE)(c>>16)]++;
- Counting4[ c>>24 ]++;
- }
- ip-=4;
- }
-
- /* finish last symbols */
- while (ip<iend) Counting1[*ip++]++;
-
- if (checkMax) { /* verify stats will fit into destination table */
- U32 s; for (s=255; s>maxSymbolValue; s--) {
- Counting1[s] += Counting2[s] + Counting3[s] + Counting4[s];
- if (Counting1[s]) return ERROR(maxSymbolValue_tooSmall);
- } }
-
- { U32 s;
- if (maxSymbolValue > 255) maxSymbolValue = 255;
- for (s=0; s<=maxSymbolValue; s++) {
- count[s] = Counting1[s] + Counting2[s] + Counting3[s] + Counting4[s];
- if (count[s] > max) max = count[s];
- } }
-
- while (!count[maxSymbolValue]) maxSymbolValue--;
- *maxSymbolValuePtr = maxSymbolValue;
- return (size_t)max;
-}
-
-/* FSE_countFast_wksp() :
- * Same as FSE_countFast(), but using an externally provided scratch buffer.
- * `workSpace` size must be table of >= `1024` unsigned */
-size_t FSE_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
- const void* source, size_t sourceSize,
- unsigned* workSpace)
-{
- if (sourceSize < 1500) /* heuristic threshold */
- return FSE_count_simple(count, maxSymbolValuePtr, source, sourceSize);
- return FSE_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, 0, workSpace);
-}
-
-/* fast variant (unsafe : won't check if src contains values beyond count[] limit) */
-size_t FSE_countFast(unsigned* count, unsigned* maxSymbolValuePtr,
- const void* source, size_t sourceSize)
-{
- unsigned tmpCounters[1024];
- return FSE_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, tmpCounters);
-}
-
-/* FSE_count_wksp() :
- * Same as FSE_count(), but using an externally provided scratch buffer.
- * `workSpace` size must be table of >= `1024` unsigned */
-size_t FSE_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
- const void* source, size_t sourceSize, unsigned* workSpace)
-{
- if (*maxSymbolValuePtr < 255)
- return FSE_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, 1, workSpace);
- *maxSymbolValuePtr = 255;
- return FSE_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, workSpace);
-}
-
-size_t FSE_count(unsigned* count, unsigned* maxSymbolValuePtr,
- const void* src, size_t srcSize)
-{
- unsigned tmpCounters[1024];
- return FSE_count_wksp(count, maxSymbolValuePtr, src, srcSize, tmpCounters);
-}
-
-
-
/*-**************************************************************
* FSE Compression Code
****************************************************************/
-/*! FSE_sizeof_CTable() :
- FSE_CTable is a variable size structure which contains :
- `U16 tableLog;`
- `U16 maxSymbolValue;`
- `U16 nextStateNumber[1 << tableLog];` // This size is variable
- `FSE_symbolCompressionTransform symbolTT[maxSymbolValue+1];` // This size is variable
-Allocation is manual (C standard does not support variable-size structures).
-*/
-size_t FSE_sizeof_CTable (unsigned maxSymbolValue, unsigned tableLog)
-{
- if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge);
- return FSE_CTABLE_SIZE_U32 (tableLog, maxSymbolValue) * sizeof(U32);
-}
FSE_CTable* FSE_createCTable (unsigned maxSymbolValue, unsigned tableLog)
{
@@ -466,7 +335,7 @@
/* provides the minimum logSize to safely represent a distribution */
static unsigned FSE_minTableLog(size_t srcSize, unsigned maxSymbolValue)
{
- U32 minBitsSrc = BIT_highbit32((U32)(srcSize - 1)) + 1;
+ U32 minBitsSrc = BIT_highbit32((U32)(srcSize)) + 1;
U32 minBitsSymbols = BIT_highbit32(maxSymbolValue) + 2;
U32 minBits = minBitsSrc < minBitsSymbols ? minBitsSrc : minBitsSymbols;
assert(srcSize > 1); /* Not supported, RLE should be used instead */
@@ -529,6 +398,9 @@
}
ToDistribute = (1 << tableLog) - distributed;
+ if (ToDistribute == 0)
+ return 0;
+
if ((total / ToDistribute) > lowOne) {
/* risk of rounding to zero */
lowOne = (U32)((total * 3) / (ToDistribute * 2));
@@ -629,11 +501,11 @@
U32 s;
U32 nTotal = 0;
for (s=0; s<=maxSymbolValue; s++)
- printf("%3i: %4i \n", s, normalizedCounter[s]);
+ RAWLOG(2, "%3i: %4i \n", s, normalizedCounter[s]);
for (s=0; s<=maxSymbolValue; s++)
nTotal += abs(normalizedCounter[s]);
if (nTotal != (1U<<tableLog))
- printf("Warning !!! Total == %u != %u !!!", nTotal, 1U<<tableLog);
+ RAWLOG(2, "Warning !!! Total == %u != %u !!!", nTotal, 1U<<tableLog);
getchar();
}
#endif
@@ -786,7 +658,7 @@
BYTE* op = ostart;
BYTE* const oend = ostart + dstSize;
- U32 count[FSE_MAX_SYMBOL_VALUE+1];
+ unsigned count[FSE_MAX_SYMBOL_VALUE+1];
S16 norm[FSE_MAX_SYMBOL_VALUE+1];
FSE_CTable* CTable = (FSE_CTable*)workSpace;
size_t const CTableSize = FSE_CTABLE_SIZE_U32(tableLog, maxSymbolValue);
@@ -800,7 +672,7 @@
if (!tableLog) tableLog = FSE_DEFAULT_TABLELOG;
/* Scan input and build symbol stats */
- { CHECK_V_F(maxCount, FSE_count_wksp(count, &maxSymbolValue, src, srcSize, (unsigned*)scratchBuffer) );
+ { CHECK_V_F(maxCount, HIST_count_wksp(count, &maxSymbolValue, src, srcSize, scratchBuffer, scratchBufferSize) );
if (maxCount == srcSize) return 1; /* only a single symbol in src : rle */
if (maxCount == 1) return 0; /* each symbol present maximum once => not compressible */
if (maxCount < (srcSize >> 7)) return 0; /* Heuristic : not compressible enough */
@@ -835,7 +707,7 @@
size_t FSE_compress2 (void* dst, size_t dstCapacity, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog)
{
fseWkspMax_t scratchBuffer;
- FSE_STATIC_ASSERT(sizeof(scratchBuffer) >= FSE_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)); /* compilation failures here means scratchBuffer is not large enough */
+ DEBUG_STATIC_ASSERT(sizeof(scratchBuffer) >= FSE_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)); /* compilation failures here means scratchBuffer is not large enough */
if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge);
return FSE_compress_wksp(dst, dstCapacity, src, srcSize, maxSymbolValue, tableLog, &scratchBuffer, sizeof(scratchBuffer));
}
diff --git a/vendor/github.com/DataDog/zstd/fse_decompress.c b/vendor/github.com/DataDog/zstd/fse_decompress.c
index 4c66c3b..72bbead 100644
--- a/vendor/github.com/DataDog/zstd/fse_decompress.c
+++ b/vendor/github.com/DataDog/zstd/fse_decompress.c
@@ -49,7 +49,7 @@
* Error Management
****************************************************************/
#define FSE_isError ERR_isError
-#define FSE_STATIC_ASSERT(c) { enum { FSE_static_assert = 1/(int)(!!(c)) }; } /* use only *after* variable declarations */
+#define FSE_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c) /* use only *after* variable declarations */
/* check and forward error code */
#define CHECK_F(f) { size_t const e = f; if (FSE_isError(e)) return e; }
diff --git a/vendor/github.com/DataDog/zstd/hist.c b/vendor/github.com/DataDog/zstd/hist.c
new file mode 100644
index 0000000..45b7bab
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/hist.c
@@ -0,0 +1,203 @@
+/* ******************************************************************
+ hist : Histogram functions
+ part of Finite State Entropy project
+ Copyright (C) 2013-present, Yann Collet.
+
+ BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+ Redistribution and use in source and binary forms, with or without
+ modification, are permitted provided that the following conditions are
+ met:
+
+ * Redistributions of source code must retain the above copyright
+ notice, this list of conditions and the following disclaimer.
+ * Redistributions in binary form must reproduce the above
+ copyright notice, this list of conditions and the following disclaimer
+ in the documentation and/or other materials provided with the
+ distribution.
+
+ THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+ You can contact the author at :
+ - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ - Public forum : https://groups.google.com/forum/#!forum/lz4c
+****************************************************************** */
+
+/* --- dependencies --- */
+#include "mem.h" /* U32, BYTE, etc. */
+#include "debug.h" /* assert, DEBUGLOG */
+#include "error_private.h" /* ERROR */
+#include "hist.h"
+
+
+/* --- Error management --- */
+unsigned HIST_isError(size_t code) { return ERR_isError(code); }
+
+/*-**************************************************************
+ * Histogram functions
+ ****************************************************************/
+unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* src, size_t srcSize)
+{
+ const BYTE* ip = (const BYTE*)src;
+ const BYTE* const end = ip + srcSize;
+ unsigned maxSymbolValue = *maxSymbolValuePtr;
+ unsigned largestCount=0;
+
+ memset(count, 0, (maxSymbolValue+1) * sizeof(*count));
+ if (srcSize==0) { *maxSymbolValuePtr = 0; return 0; }
+
+ while (ip<end) {
+ assert(*ip <= maxSymbolValue);
+ count[*ip++]++;
+ }
+
+ while (!count[maxSymbolValue]) maxSymbolValue--;
+ *maxSymbolValuePtr = maxSymbolValue;
+
+ { U32 s;
+ for (s=0; s<=maxSymbolValue; s++)
+ if (count[s] > largestCount) largestCount = count[s];
+ }
+
+ return largestCount;
+}
+
+typedef enum { trustInput, checkMaxSymbolValue } HIST_checkInput_e;
+
+/* HIST_count_parallel_wksp() :
+ * store histogram into 4 intermediate tables, recombined at the end.
+ * this design makes better use of OoO cpus,
+ * and is noticeably faster when some values are heavily repeated.
+ * But it needs some additional workspace for intermediate tables.
+ * `workSpace` size must be a table of size >= HIST_WKSP_SIZE_U32.
+ * @return : largest histogram frequency,
+ * or an error code (notably when histogram would be larger than *maxSymbolValuePtr). */
+static size_t HIST_count_parallel_wksp(
+ unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* source, size_t sourceSize,
+ HIST_checkInput_e check,
+ U32* const workSpace)
+{
+ const BYTE* ip = (const BYTE*)source;
+ const BYTE* const iend = ip+sourceSize;
+ unsigned maxSymbolValue = *maxSymbolValuePtr;
+ unsigned max=0;
+ U32* const Counting1 = workSpace;
+ U32* const Counting2 = Counting1 + 256;
+ U32* const Counting3 = Counting2 + 256;
+ U32* const Counting4 = Counting3 + 256;
+
+ memset(workSpace, 0, 4*256*sizeof(unsigned));
+
+ /* safety checks */
+ if (!sourceSize) {
+ memset(count, 0, maxSymbolValue + 1);
+ *maxSymbolValuePtr = 0;
+ return 0;
+ }
+ if (!maxSymbolValue) maxSymbolValue = 255; /* 0 == default */
+
+ /* by stripes of 16 bytes */
+ { U32 cached = MEM_read32(ip); ip += 4;
+ while (ip < iend-15) {
+ U32 c = cached; cached = MEM_read32(ip); ip += 4;
+ Counting1[(BYTE) c ]++;
+ Counting2[(BYTE)(c>>8) ]++;
+ Counting3[(BYTE)(c>>16)]++;
+ Counting4[ c>>24 ]++;
+ c = cached; cached = MEM_read32(ip); ip += 4;
+ Counting1[(BYTE) c ]++;
+ Counting2[(BYTE)(c>>8) ]++;
+ Counting3[(BYTE)(c>>16)]++;
+ Counting4[ c>>24 ]++;
+ c = cached; cached = MEM_read32(ip); ip += 4;
+ Counting1[(BYTE) c ]++;
+ Counting2[(BYTE)(c>>8) ]++;
+ Counting3[(BYTE)(c>>16)]++;
+ Counting4[ c>>24 ]++;
+ c = cached; cached = MEM_read32(ip); ip += 4;
+ Counting1[(BYTE) c ]++;
+ Counting2[(BYTE)(c>>8) ]++;
+ Counting3[(BYTE)(c>>16)]++;
+ Counting4[ c>>24 ]++;
+ }
+ ip-=4;
+ }
+
+ /* finish last symbols */
+ while (ip<iend) Counting1[*ip++]++;
+
+ if (check) { /* verify stats will fit into destination table */
+ U32 s; for (s=255; s>maxSymbolValue; s--) {
+ Counting1[s] += Counting2[s] + Counting3[s] + Counting4[s];
+ if (Counting1[s]) return ERROR(maxSymbolValue_tooSmall);
+ } }
+
+ { U32 s;
+ if (maxSymbolValue > 255) maxSymbolValue = 255;
+ for (s=0; s<=maxSymbolValue; s++) {
+ count[s] = Counting1[s] + Counting2[s] + Counting3[s] + Counting4[s];
+ if (count[s] > max) max = count[s];
+ } }
+
+ while (!count[maxSymbolValue]) maxSymbolValue--;
+ *maxSymbolValuePtr = maxSymbolValue;
+ return (size_t)max;
+}
+
+/* HIST_countFast_wksp() :
+ * Same as HIST_countFast(), but using an externally provided scratch buffer.
+ * `workSpace` is a writable buffer which must be 4-bytes aligned,
+ * `workSpaceSize` must be >= HIST_WKSP_SIZE
+ */
+size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* source, size_t sourceSize,
+ void* workSpace, size_t workSpaceSize)
+{
+ if (sourceSize < 1500) /* heuristic threshold */
+ return HIST_count_simple(count, maxSymbolValuePtr, source, sourceSize);
+ if ((size_t)workSpace & 3) return ERROR(GENERIC); /* must be aligned on 4-bytes boundaries */
+ if (workSpaceSize < HIST_WKSP_SIZE) return ERROR(workSpace_tooSmall);
+ return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, trustInput, (U32*)workSpace);
+}
+
+/* fast variant (unsafe : won't check if src contains values beyond count[] limit) */
+size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* source, size_t sourceSize)
+{
+ unsigned tmpCounters[HIST_WKSP_SIZE_U32];
+ return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, tmpCounters, sizeof(tmpCounters));
+}
+
+/* HIST_count_wksp() :
+ * Same as HIST_count(), but using an externally provided scratch buffer.
+ * `workSpace` size must be table of >= HIST_WKSP_SIZE_U32 unsigned */
+size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* source, size_t sourceSize,
+ void* workSpace, size_t workSpaceSize)
+{
+ if ((size_t)workSpace & 3) return ERROR(GENERIC); /* must be aligned on 4-bytes boundaries */
+ if (workSpaceSize < HIST_WKSP_SIZE) return ERROR(workSpace_tooSmall);
+ if (*maxSymbolValuePtr < 255)
+ return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, checkMaxSymbolValue, (U32*)workSpace);
+ *maxSymbolValuePtr = 255;
+ return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, workSpace, workSpaceSize);
+}
+
+size_t HIST_count(unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* src, size_t srcSize)
+{
+ unsigned tmpCounters[HIST_WKSP_SIZE_U32];
+ return HIST_count_wksp(count, maxSymbolValuePtr, src, srcSize, tmpCounters, sizeof(tmpCounters));
+}
diff --git a/vendor/github.com/DataDog/zstd/hist.h b/vendor/github.com/DataDog/zstd/hist.h
new file mode 100644
index 0000000..8b38935
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/hist.h
@@ -0,0 +1,95 @@
+/* ******************************************************************
+ hist : Histogram functions
+ part of Finite State Entropy project
+ Copyright (C) 2013-present, Yann Collet.
+
+ BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+ Redistribution and use in source and binary forms, with or without
+ modification, are permitted provided that the following conditions are
+ met:
+
+ * Redistributions of source code must retain the above copyright
+ notice, this list of conditions and the following disclaimer.
+ * Redistributions in binary form must reproduce the above
+ copyright notice, this list of conditions and the following disclaimer
+ in the documentation and/or other materials provided with the
+ distribution.
+
+ THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+ You can contact the author at :
+ - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ - Public forum : https://groups.google.com/forum/#!forum/lz4c
+****************************************************************** */
+
+/* --- dependencies --- */
+#include <stddef.h> /* size_t */
+
+
+/* --- simple histogram functions --- */
+
+/*! HIST_count():
+ * Provides the precise count of each byte within a table 'count'.
+ * 'count' is a table of unsigned int, of minimum size (*maxSymbolValuePtr+1).
+ * Updates *maxSymbolValuePtr with actual largest symbol value detected.
+ * @return : count of the most frequent symbol (which isn't identified).
+ * or an error code, which can be tested using HIST_isError().
+ * note : if return == srcSize, there is only one symbol.
+ */
+size_t HIST_count(unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* src, size_t srcSize);
+
+unsigned HIST_isError(size_t code); /**< tells if a return value is an error code */
+
+
+/* --- advanced histogram functions --- */
+
+#define HIST_WKSP_SIZE_U32 1024
+#define HIST_WKSP_SIZE (HIST_WKSP_SIZE_U32 * sizeof(unsigned))
+/** HIST_count_wksp() :
+ * Same as HIST_count(), but using an externally provided scratch buffer.
+ * Benefit is this function will use very little stack space.
+ * `workSpace` is a writable buffer which must be 4-bytes aligned,
+ * `workSpaceSize` must be >= HIST_WKSP_SIZE
+ */
+size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* src, size_t srcSize,
+ void* workSpace, size_t workSpaceSize);
+
+/** HIST_countFast() :
+ * same as HIST_count(), but blindly trusts that all byte values within src are <= *maxSymbolValuePtr.
+ * This function is unsafe, and will segfault if any value within `src` is `> *maxSymbolValuePtr`
+ */
+size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* src, size_t srcSize);
+
+/** HIST_countFast_wksp() :
+ * Same as HIST_countFast(), but using an externally provided scratch buffer.
+ * `workSpace` is a writable buffer which must be 4-bytes aligned,
+ * `workSpaceSize` must be >= HIST_WKSP_SIZE
+ */
+size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* src, size_t srcSize,
+ void* workSpace, size_t workSpaceSize);
+
+/*! HIST_count_simple() :
+ * Same as HIST_countFast(), this function is unsafe,
+ * and will segfault if any value within `src` is `> *maxSymbolValuePtr`.
+ * It is also a bit slower for large inputs.
+ * However, it does not need any additional memory (not even on stack).
+ * @return : count of the most frequent symbol.
+ * Note this function doesn't produce any error (i.e. it must succeed).
+ */
+unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr,
+ const void* src, size_t srcSize);
diff --git a/vendor/github.com/DataDog/zstd/huf.h b/vendor/github.com/DataDog/zstd/huf.h
index b4645b4..6b572c4 100644
--- a/vendor/github.com/DataDog/zstd/huf.h
+++ b/vendor/github.com/DataDog/zstd/huf.h
@@ -1,7 +1,7 @@
/* ******************************************************************
- Huffman coder, part of New Generation Entropy library
- header file
- Copyright (C) 2013-2016, Yann Collet.
+ huff0 huffman codec,
+ part of Finite State Entropy library
+ Copyright (C) 2013-present, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
@@ -163,25 +163,29 @@
/* static allocation of HUF's DTable */
typedef U32 HUF_DTable;
#define HUF_DTABLE_SIZE(maxTableLog) (1 + (1<<(maxTableLog)))
-#define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \
+#define HUF_CREATE_STATIC_DTABLEX1(DTable, maxTableLog) \
HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = { ((U32)((maxTableLog)-1) * 0x01000001) }
-#define HUF_CREATE_STATIC_DTABLEX4(DTable, maxTableLog) \
+#define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \
HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = { ((U32)(maxTableLog) * 0x01000001) }
/* ****************************************
* Advanced decompression functions
******************************************/
-size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol decoder */
-size_t HUF_decompress4X4 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */
+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_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol 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); /**< single-symbol decoder */
-size_t HUF_decompress4X4_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */
-size_t HUF_decompress4X4_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< double-symbols decoder */
+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
/* ****************************************
@@ -208,7 +212,7 @@
typedef enum {
HUF_repeat_none, /**< Cannot use the previous table */
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 asumed to be valid */
+ 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.
@@ -227,7 +231,9 @@
*/
#define HUF_CTABLE_WORKSPACE_SIZE_U32 (2*HUF_SYMBOLVALUE_MAX +1 +1)
#define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))
-size_t HUF_buildCTable_wksp (HUF_CElt* tree, const U32* count, U32 maxSymbolValue, U32 maxNbBits, void* workSpace, size_t wkspSize);
+size_t HUF_buildCTable_wksp (HUF_CElt* tree,
+ const unsigned* count, U32 maxSymbolValue, U32 maxNbBits,
+ void* workSpace, size_t wkspSize);
/*! HUF_readStats() :
* Read compact Huffman tree, saved by HUF_writeCTable().
@@ -242,10 +248,15 @@
* Loading a CTable saved with HUF_writeCTable() */
size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize);
+/** HUF_getNbBits() :
+ * Read nbBits from CTable symbolTable, for symbol `symbolValue` presumed <= HUF_SYMBOLVALUE_MAX
+ * Note 1 : is not inlined, as HUF_CElt definition is private
+ * Note 2 : const void* used, so that it can provide a statically allocated table as argument (which uses type U32) */
+U32 HUF_getNbBits(const void* symbolTable, U32 symbolValue);
/*
* HUF_decompress() does the following:
- * 1. select the decompression algorithm (X2, X4) based on pre-computed heuristics
+ * 1. select the decompression algorithm (X1, X2) based on pre-computed heuristics
* 2. build Huffman table from save, using HUF_readDTableX?()
* 3. decode 1 or 4 segments in parallel using HUF_decompress?X?_usingDTable()
*/
@@ -253,13 +264,13 @@
/** HUF_selectDecoder() :
* Tells which decoder is likely to decode faster,
* based on a set of pre-computed metrics.
- * @return : 0==HUF_decompress4X2, 1==HUF_decompress4X4 .
+ * @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 .
* Assumption : 0 < dstSize <= 128 KB */
U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize);
/**
* The minimum workspace size for the `workSpace` used in
- * HUF_readDTableX2_wksp() and HUF_readDTableX4_wksp().
+ * HUF_readDTableX1_wksp() and HUF_readDTableX2_wksp().
*
* The space used depends on HUF_TABLELOG_MAX, ranging from ~1500 bytes when
* HUF_TABLE_LOG_MAX=12 to ~1850 bytes when HUF_TABLE_LOG_MAX=15.
@@ -270,14 +281,22 @@
#define HUF_DECOMPRESS_WORKSPACE_SIZE (2 << 10)
#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);
-size_t HUF_readDTableX4 (HUF_DTable* DTable, const void* src, size_t srcSize);
-size_t HUF_readDTableX4_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);
-size_t HUF_decompress4X4_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#endif
/* ====================== */
@@ -298,25 +317,37 @@
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);
-size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* single-symbol decoder */
-size_t HUF_decompress1X4 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* double-symbol decoder */
+size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* single-symbol decoder */
+#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);
-size_t HUF_decompress1X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol 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); /**< single-symbol decoder */
-size_t HUF_decompress1X4_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */
-size_t HUF_decompress1X4_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< double-symbols decoder */
+#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_decompress1X4_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#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_decompress1X2_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
+#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);
+#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);
diff --git a/vendor/github.com/DataDog/zstd/huf_compress.c b/vendor/github.com/DataDog/zstd/huf_compress.c
index 83230b4..f074f1e 100644
--- a/vendor/github.com/DataDog/zstd/huf_compress.c
+++ b/vendor/github.com/DataDog/zstd/huf_compress.c
@@ -45,8 +45,9 @@
****************************************************************/
#include <string.h> /* memcpy, memset */
#include <stdio.h> /* printf (debug) */
-#include "bitstream.h"
#include "compiler.h"
+#include "bitstream.h"
+#include "hist.h"
#define FSE_STATIC_LINKING_ONLY /* FSE_optimalTableLog_internal */
#include "fse.h" /* header compression */
#define HUF_STATIC_LINKING_ONLY
@@ -58,7 +59,7 @@
* Error Management
****************************************************************/
#define HUF_isError ERR_isError
-#define HUF_STATIC_ASSERT(c) { enum { HUF_static_assert = 1/(int)(!!(c)) }; } /* use only *after* variable declarations */
+#define HUF_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c) /* use only *after* variable declarations */
#define CHECK_V_F(e, f) size_t const e = f; if (ERR_isError(e)) return e
#define CHECK_F(f) { CHECK_V_F(_var_err__, f); }
@@ -81,28 +82,28 @@
* Note : all elements within weightTable are supposed to be <= HUF_TABLELOG_MAX.
*/
#define MAX_FSE_TABLELOG_FOR_HUFF_HEADER 6
-size_t HUF_compressWeights (void* dst, size_t dstSize, const void* weightTable, size_t wtSize)
+static size_t HUF_compressWeights (void* dst, size_t dstSize, const void* weightTable, size_t wtSize)
{
BYTE* const ostart = (BYTE*) dst;
BYTE* op = ostart;
BYTE* const oend = ostart + dstSize;
- U32 maxSymbolValue = HUF_TABLELOG_MAX;
+ unsigned maxSymbolValue = HUF_TABLELOG_MAX;
U32 tableLog = MAX_FSE_TABLELOG_FOR_HUFF_HEADER;
FSE_CTable CTable[FSE_CTABLE_SIZE_U32(MAX_FSE_TABLELOG_FOR_HUFF_HEADER, HUF_TABLELOG_MAX)];
BYTE scratchBuffer[1<<MAX_FSE_TABLELOG_FOR_HUFF_HEADER];
- U32 count[HUF_TABLELOG_MAX+1];
+ unsigned count[HUF_TABLELOG_MAX+1];
S16 norm[HUF_TABLELOG_MAX+1];
/* init conditions */
if (wtSize <= 1) return 0; /* Not compressible */
/* Scan input and build symbol stats */
- { CHECK_V_F(maxCount, FSE_count_simple(count, &maxSymbolValue, weightTable, wtSize) );
+ { unsigned const maxCount = HIST_count_simple(count, &maxSymbolValue, weightTable, wtSize); /* never fails */
if (maxCount == wtSize) return 1; /* only a single symbol in src : rle */
- if (maxCount == 1) return 0; /* each symbol present maximum once => not compressible */
+ if (maxCount == 1) return 0; /* each symbol present maximum once => not compressible */
}
tableLog = FSE_optimalTableLog(tableLog, wtSize, maxSymbolValue);
@@ -133,7 +134,7 @@
`CTable` : Huffman tree to save, using huf representation.
@return : size of saved CTable */
size_t HUF_writeCTable (void* dst, size_t maxDstSize,
- const HUF_CElt* CTable, U32 maxSymbolValue, U32 huffLog)
+ const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog)
{
BYTE bitsToWeight[HUF_TABLELOG_MAX + 1]; /* precomputed conversion table */
BYTE huffWeight[HUF_SYMBOLVALUE_MAX];
@@ -168,7 +169,7 @@
}
-size_t HUF_readCTable (HUF_CElt* CTable, U32* maxSymbolValuePtr, const void* src, size_t srcSize)
+size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize)
{
BYTE huffWeight[HUF_SYMBOLVALUE_MAX + 1]; /* init not required, even though some static analyzer may complain */
U32 rankVal[HUF_TABLELOG_ABSOLUTEMAX + 1]; /* large enough for values from 0 to 16 */
@@ -216,6 +217,13 @@
return readSize;
}
+U32 HUF_getNbBits(const void* symbolTable, U32 symbolValue)
+{
+ const HUF_CElt* table = (const HUF_CElt*)symbolTable;
+ assert(symbolValue <= HUF_SYMBOLVALUE_MAX);
+ return table[symbolValue].nbBits;
+}
+
typedef struct nodeElt_s {
U32 count;
@@ -307,7 +315,7 @@
U32 current;
} rankPos;
-static void HUF_sort(nodeElt* huffNode, const U32* count, U32 maxSymbolValue)
+static void HUF_sort(nodeElt* huffNode, const unsigned* count, U32 maxSymbolValue)
{
rankPos rank[32];
U32 n;
@@ -339,7 +347,7 @@
*/
#define STARTNODE (HUF_SYMBOLVALUE_MAX+1)
typedef nodeElt huffNodeTable[HUF_CTABLE_WORKSPACE_SIZE_U32];
-size_t HUF_buildCTable_wksp (HUF_CElt* tree, const U32* count, U32 maxSymbolValue, U32 maxNbBits, void* workSpace, size_t wkspSize)
+size_t HUF_buildCTable_wksp (HUF_CElt* tree, const unsigned* count, U32 maxSymbolValue, U32 maxNbBits, void* workSpace, size_t wkspSize)
{
nodeElt* const huffNode0 = (nodeElt*)workSpace;
nodeElt* const huffNode = huffNode0+1;
@@ -413,7 +421,7 @@
* @return : maxNbBits
* Note : count is used before tree is written, so they can safely overlap
*/
-size_t HUF_buildCTable (HUF_CElt* tree, const U32* count, U32 maxSymbolValue, U32 maxNbBits)
+size_t HUF_buildCTable (HUF_CElt* tree, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits)
{
huffNodeTable nodeTable;
return HUF_buildCTable_wksp(tree, count, maxSymbolValue, maxNbBits, nodeTable, sizeof(nodeTable));
@@ -602,13 +610,14 @@
return HUF_compress4X_usingCTable_internal(dst, dstSize, src, srcSize, CTable, /* bmi2 */ 0);
}
+typedef enum { HUF_singleStream, HUF_fourStreams } HUF_nbStreams_e;
static size_t HUF_compressCTable_internal(
BYTE* const ostart, BYTE* op, BYTE* const oend,
const void* src, size_t srcSize,
- unsigned singleStream, const HUF_CElt* CTable, const int bmi2)
+ HUF_nbStreams_e nbStreams, const HUF_CElt* CTable, const int bmi2)
{
- size_t const cSize = singleStream ?
+ size_t const cSize = (nbStreams==HUF_singleStream) ?
HUF_compress1X_usingCTable_internal(op, oend - op, src, srcSize, CTable, bmi2) :
HUF_compress4X_usingCTable_internal(op, oend - op, src, srcSize, CTable, bmi2);
if (HUF_isError(cSize)) { return cSize; }
@@ -620,21 +629,21 @@
}
typedef struct {
- U32 count[HUF_SYMBOLVALUE_MAX + 1];
+ unsigned count[HUF_SYMBOLVALUE_MAX + 1];
HUF_CElt CTable[HUF_SYMBOLVALUE_MAX + 1];
huffNodeTable nodeTable;
} HUF_compress_tables_t;
/* HUF_compress_internal() :
* `workSpace` must a table of at least HUF_WORKSPACE_SIZE_U32 unsigned */
-static size_t HUF_compress_internal (
- void* dst, size_t dstSize,
- const void* src, size_t srcSize,
- unsigned maxSymbolValue, unsigned huffLog,
- unsigned singleStream,
- void* workSpace, size_t wkspSize,
- HUF_CElt* oldHufTable, HUF_repeat* repeat, int preferRepeat,
- const int bmi2)
+static size_t
+HUF_compress_internal (void* dst, size_t dstSize,
+ const void* src, size_t srcSize,
+ unsigned maxSymbolValue, unsigned huffLog,
+ HUF_nbStreams_e nbStreams,
+ void* workSpace, size_t wkspSize,
+ HUF_CElt* oldHufTable, HUF_repeat* repeat, int preferRepeat,
+ const int bmi2)
{
HUF_compress_tables_t* const table = (HUF_compress_tables_t*)workSpace;
BYTE* const ostart = (BYTE*)dst;
@@ -643,7 +652,7 @@
/* checks & inits */
if (((size_t)workSpace & 3) != 0) return ERROR(GENERIC); /* must be aligned on 4-bytes boundaries */
- if (wkspSize < sizeof(*table)) return ERROR(workSpace_tooSmall);
+ if (wkspSize < HUF_WORKSPACE_SIZE) return ERROR(workSpace_tooSmall);
if (!srcSize) return 0; /* Uncompressed */
if (!dstSize) return 0; /* cannot fit anything within dst budget */
if (srcSize > HUF_BLOCKSIZE_MAX) return ERROR(srcSize_wrong); /* current block size limit */
@@ -656,13 +665,13 @@
if (preferRepeat && repeat && *repeat == HUF_repeat_valid) {
return HUF_compressCTable_internal(ostart, op, oend,
src, srcSize,
- singleStream, oldHufTable, bmi2);
+ nbStreams, oldHufTable, bmi2);
}
/* Scan input and build symbol stats */
- { CHECK_V_F(largest, FSE_count_wksp (table->count, &maxSymbolValue, (const BYTE*)src, srcSize, table->count) );
+ { CHECK_V_F(largest, HIST_count_wksp (table->count, &maxSymbolValue, (const BYTE*)src, srcSize, workSpace, wkspSize) );
if (largest == srcSize) { *ostart = ((const BYTE*)src)[0]; return 1; } /* single symbol, rle */
- if (largest <= (srcSize >> 7)+1) return 0; /* heuristic : probably not compressible enough */
+ if (largest <= (srcSize >> 7)+4) return 0; /* heuristic : probably not compressible enough */
}
/* Check validity of previous table */
@@ -675,14 +684,15 @@
if (preferRepeat && repeat && *repeat != HUF_repeat_none) {
return HUF_compressCTable_internal(ostart, op, oend,
src, srcSize,
- singleStream, oldHufTable, bmi2);
+ nbStreams, oldHufTable, bmi2);
}
/* Build Huffman Tree */
huffLog = HUF_optimalTableLog(huffLog, srcSize, maxSymbolValue);
- { CHECK_V_F(maxBits, HUF_buildCTable_wksp(table->CTable, table->count,
- maxSymbolValue, huffLog,
- table->nodeTable, sizeof(table->nodeTable)) );
+ { size_t const maxBits = HUF_buildCTable_wksp(table->CTable, table->count,
+ maxSymbolValue, huffLog,
+ table->nodeTable, sizeof(table->nodeTable));
+ CHECK_F(maxBits);
huffLog = (U32)maxBits;
/* Zero unused symbols in CTable, so we can check it for validity */
memset(table->CTable + (maxSymbolValue + 1), 0,
@@ -698,7 +708,7 @@
if (oldSize <= hSize + newSize || hSize + 12 >= srcSize) {
return HUF_compressCTable_internal(ostart, op, oend,
src, srcSize,
- singleStream, oldHufTable, bmi2);
+ nbStreams, oldHufTable, bmi2);
} }
/* Use the new huffman table */
@@ -710,7 +720,7 @@
}
return HUF_compressCTable_internal(ostart, op, oend,
src, srcSize,
- singleStream, table->CTable, bmi2);
+ nbStreams, table->CTable, bmi2);
}
@@ -720,7 +730,7 @@
void* workSpace, size_t wkspSize)
{
return HUF_compress_internal(dst, dstSize, src, srcSize,
- maxSymbolValue, huffLog, 1 /*single stream*/,
+ maxSymbolValue, huffLog, HUF_singleStream,
workSpace, wkspSize,
NULL, NULL, 0, 0 /*bmi2*/);
}
@@ -732,7 +742,7 @@
HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2)
{
return HUF_compress_internal(dst, dstSize, src, srcSize,
- maxSymbolValue, huffLog, 1 /*single stream*/,
+ maxSymbolValue, huffLog, HUF_singleStream,
workSpace, wkspSize, hufTable,
repeat, preferRepeat, bmi2);
}
@@ -754,7 +764,7 @@
void* workSpace, size_t wkspSize)
{
return HUF_compress_internal(dst, dstSize, src, srcSize,
- maxSymbolValue, huffLog, 0 /*4 streams*/,
+ maxSymbolValue, huffLog, HUF_fourStreams,
workSpace, wkspSize,
NULL, NULL, 0, 0 /*bmi2*/);
}
@@ -769,7 +779,7 @@
HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2)
{
return HUF_compress_internal(dst, dstSize, src, srcSize,
- maxSymbolValue, huffLog, 0 /* 4 streams */,
+ maxSymbolValue, huffLog, HUF_fourStreams,
workSpace, wkspSize,
hufTable, repeat, preferRepeat, bmi2);
}
diff --git a/vendor/github.com/DataDog/zstd/huf_decompress.c b/vendor/github.com/DataDog/zstd/huf_decompress.c
index 73f5c46..3f8bd29 100644
--- a/vendor/github.com/DataDog/zstd/huf_decompress.c
+++ b/vendor/github.com/DataDog/zstd/huf_decompress.c
@@ -1,6 +1,7 @@
/* ******************************************************************
- Huffman decoder, part of New Generation Entropy library
- Copyright (C) 2013-2016, Yann Collet.
+ huff0 huffman decoder,
+ part of Finite State Entropy library
+ Copyright (C) 2013-present, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
@@ -29,26 +30,37 @@
You can contact the author at :
- FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy
- - Public forum : https://groups.google.com/forum/#!forum/lz4c
****************************************************************** */
/* **************************************************************
* Dependencies
****************************************************************/
#include <string.h> /* memcpy, memset */
-#include "bitstream.h" /* BIT_* */
#include "compiler.h"
-#include "fse.h" /* header compression */
+#include "bitstream.h" /* BIT_* */
+#include "fse.h" /* to compress headers */
#define HUF_STATIC_LINKING_ONLY
#include "huf.h"
#include "error_private.h"
+/* **************************************************************
+* Macros
+****************************************************************/
+
+/* These two optional macros force the use one way or another of the two
+ * Huffman decompression implementations. You can't force in both directions
+ * at the same time.
+ */
+#if defined(HUF_FORCE_DECOMPRESS_X1) && \
+ defined(HUF_FORCE_DECOMPRESS_X2)
+#error "Cannot force the use of the X1 and X2 decoders at the same time!"
+#endif
+
/* **************************************************************
* Error Management
****************************************************************/
#define HUF_isError ERR_isError
-#define HUF_STATIC_ASSERT(c) { enum { HUF_static_assert = 1/(int)(!!(c)) }; } /* use only *after* variable declarations */
#define CHECK_F(f) { size_t const err_ = (f); if (HUF_isError(err_)) return err_; }
@@ -59,6 +71,51 @@
#define HUF_ALIGN_MASK(x, mask) (((x) + (mask)) & ~(mask))
+/* **************************************************************
+* BMI2 Variant Wrappers
+****************************************************************/
+#if DYNAMIC_BMI2
+
+#define HUF_DGEN(fn) \
+ \
+ static size_t fn##_default( \
+ void* dst, size_t dstSize, \
+ const void* cSrc, size_t cSrcSize, \
+ const HUF_DTable* DTable) \
+ { \
+ return fn##_body(dst, dstSize, cSrc, cSrcSize, DTable); \
+ } \
+ \
+ static TARGET_ATTRIBUTE("bmi2") size_t fn##_bmi2( \
+ void* dst, size_t dstSize, \
+ const void* cSrc, size_t cSrcSize, \
+ const HUF_DTable* DTable) \
+ { \
+ return fn##_body(dst, dstSize, cSrc, cSrcSize, DTable); \
+ } \
+ \
+ static size_t fn(void* dst, size_t dstSize, void const* cSrc, \
+ size_t cSrcSize, HUF_DTable const* DTable, int bmi2) \
+ { \
+ if (bmi2) { \
+ return fn##_bmi2(dst, dstSize, cSrc, cSrcSize, DTable); \
+ } \
+ return fn##_default(dst, dstSize, cSrc, cSrcSize, DTable); \
+ }
+
+#else
+
+#define HUF_DGEN(fn) \
+ static size_t fn(void* dst, size_t dstSize, void const* cSrc, \
+ size_t cSrcSize, HUF_DTable const* DTable, int bmi2) \
+ { \
+ (void)bmi2; \
+ return fn##_body(dst, dstSize, cSrc, cSrcSize, DTable); \
+ }
+
+#endif
+
+
/*-***************************/
/* generic DTableDesc */
/*-***************************/
@@ -72,18 +129,20 @@
}
+#ifndef HUF_FORCE_DECOMPRESS_X2
+
/*-***************************/
/* single-symbol decoding */
/*-***************************/
-typedef struct { BYTE byte; BYTE nbBits; } HUF_DEltX2; /* single-symbol decoding */
+typedef struct { BYTE byte; BYTE nbBits; } HUF_DEltX1; /* single-symbol decoding */
-size_t HUF_readDTableX2_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize)
+size_t HUF_readDTableX1_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize)
{
U32 tableLog = 0;
U32 nbSymbols = 0;
size_t iSize;
void* const dtPtr = DTable + 1;
- HUF_DEltX2* const dt = (HUF_DEltX2*)dtPtr;
+ HUF_DEltX1* const dt = (HUF_DEltX1*)dtPtr;
U32* rankVal;
BYTE* huffWeight;
@@ -96,7 +155,7 @@
if ((spaceUsed32 << 2) > wkspSize) return ERROR(tableLog_tooLarge);
- HUF_STATIC_ASSERT(sizeof(DTableDesc) == sizeof(HUF_DTable));
+ DEBUG_STATIC_ASSERT(sizeof(DTableDesc) == sizeof(HUF_DTable));
/* memset(huffWeight, 0, sizeof(huffWeight)); */ /* is not necessary, even though some analyzer complain ... */
iSize = HUF_readStats(huffWeight, HUF_SYMBOLVALUE_MAX + 1, rankVal, &nbSymbols, &tableLog, src, srcSize);
@@ -124,7 +183,7 @@
U32 const w = huffWeight[n];
U32 const length = (1 << w) >> 1;
U32 u;
- HUF_DEltX2 D;
+ HUF_DEltX1 D;
D.byte = (BYTE)n; D.nbBits = (BYTE)(tableLog + 1 - w);
for (u = rankVal[w]; u < rankVal[w] + length; u++)
dt[u] = D;
@@ -134,17 +193,15 @@
return iSize;
}
-size_t HUF_readDTableX2(HUF_DTable* DTable, const void* src, size_t srcSize)
+size_t HUF_readDTableX1(HUF_DTable* DTable, const void* src, size_t srcSize)
{
U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32];
- return HUF_readDTableX2_wksp(DTable, src, srcSize,
+ return HUF_readDTableX1_wksp(DTable, src, srcSize,
workSpace, sizeof(workSpace));
}
-typedef struct { U16 sequence; BYTE nbBits; BYTE length; } HUF_DEltX4; /* double-symbols decoding */
-
FORCE_INLINE_TEMPLATE BYTE
-HUF_decodeSymbolX2(BIT_DStream_t* Dstream, const HUF_DEltX2* dt, const U32 dtLog)
+HUF_decodeSymbolX1(BIT_DStream_t* Dstream, const HUF_DEltX1* dt, const U32 dtLog)
{
size_t const val = BIT_lookBitsFast(Dstream, dtLog); /* note : dtLog >= 1 */
BYTE const c = dt[val].byte;
@@ -152,44 +209,44 @@
return c;
}
-#define HUF_DECODE_SYMBOLX2_0(ptr, DStreamPtr) \
- *ptr++ = HUF_decodeSymbolX2(DStreamPtr, dt, dtLog)
+#define HUF_DECODE_SYMBOLX1_0(ptr, DStreamPtr) \
+ *ptr++ = HUF_decodeSymbolX1(DStreamPtr, dt, dtLog)
-#define HUF_DECODE_SYMBOLX2_1(ptr, DStreamPtr) \
+#define HUF_DECODE_SYMBOLX1_1(ptr, DStreamPtr) \
if (MEM_64bits() || (HUF_TABLELOG_MAX<=12)) \
- HUF_DECODE_SYMBOLX2_0(ptr, DStreamPtr)
+ HUF_DECODE_SYMBOLX1_0(ptr, DStreamPtr)
-#define HUF_DECODE_SYMBOLX2_2(ptr, DStreamPtr) \
+#define HUF_DECODE_SYMBOLX1_2(ptr, DStreamPtr) \
if (MEM_64bits()) \
- HUF_DECODE_SYMBOLX2_0(ptr, DStreamPtr)
+ HUF_DECODE_SYMBOLX1_0(ptr, DStreamPtr)
HINT_INLINE size_t
-HUF_decodeStreamX2(BYTE* p, BIT_DStream_t* const bitDPtr, BYTE* const pEnd, const HUF_DEltX2* const dt, const U32 dtLog)
+HUF_decodeStreamX1(BYTE* p, BIT_DStream_t* const bitDPtr, BYTE* const pEnd, const HUF_DEltX1* const dt, const U32 dtLog)
{
BYTE* const pStart = p;
/* up to 4 symbols at a time */
while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p < pEnd-3)) {
- HUF_DECODE_SYMBOLX2_2(p, bitDPtr);
- HUF_DECODE_SYMBOLX2_1(p, bitDPtr);
- HUF_DECODE_SYMBOLX2_2(p, bitDPtr);
- HUF_DECODE_SYMBOLX2_0(p, bitDPtr);
+ HUF_DECODE_SYMBOLX1_2(p, bitDPtr);
+ HUF_DECODE_SYMBOLX1_1(p, bitDPtr);
+ HUF_DECODE_SYMBOLX1_2(p, bitDPtr);
+ HUF_DECODE_SYMBOLX1_0(p, bitDPtr);
}
/* [0-3] symbols remaining */
if (MEM_32bits())
while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p < pEnd))
- HUF_DECODE_SYMBOLX2_0(p, bitDPtr);
+ HUF_DECODE_SYMBOLX1_0(p, bitDPtr);
/* no more data to retrieve from bitstream, no need to reload */
while (p < pEnd)
- HUF_DECODE_SYMBOLX2_0(p, bitDPtr);
+ HUF_DECODE_SYMBOLX1_0(p, bitDPtr);
return pEnd-pStart;
}
FORCE_INLINE_TEMPLATE size_t
-HUF_decompress1X2_usingDTable_internal_body(
+HUF_decompress1X1_usingDTable_internal_body(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
const HUF_DTable* DTable)
@@ -197,14 +254,14 @@
BYTE* op = (BYTE*)dst;
BYTE* const oend = op + dstSize;
const void* dtPtr = DTable + 1;
- const HUF_DEltX2* const dt = (const HUF_DEltX2*)dtPtr;
+ const HUF_DEltX1* const dt = (const HUF_DEltX1*)dtPtr;
BIT_DStream_t bitD;
DTableDesc const dtd = HUF_getDTableDesc(DTable);
U32 const dtLog = dtd.tableLog;
CHECK_F( BIT_initDStream(&bitD, cSrc, cSrcSize) );
- HUF_decodeStreamX2(op, &bitD, oend, dt, dtLog);
+ HUF_decodeStreamX1(op, &bitD, oend, dt, dtLog);
if (!BIT_endOfDStream(&bitD)) return ERROR(corruption_detected);
@@ -212,7 +269,7 @@
}
FORCE_INLINE_TEMPLATE size_t
-HUF_decompress4X2_usingDTable_internal_body(
+HUF_decompress4X1_usingDTable_internal_body(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
const HUF_DTable* DTable)
@@ -224,7 +281,7 @@
BYTE* const ostart = (BYTE*) dst;
BYTE* const oend = ostart + dstSize;
const void* const dtPtr = DTable + 1;
- const HUF_DEltX2* const dt = (const HUF_DEltX2*)dtPtr;
+ const HUF_DEltX1* const dt = (const HUF_DEltX1*)dtPtr;
/* Init */
BIT_DStream_t bitD1;
@@ -260,22 +317,22 @@
/* up to 16 symbols per loop (4 symbols per stream) in 64-bit mode */
endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
while ( (endSignal==BIT_DStream_unfinished) && (op4<(oend-3)) ) {
- HUF_DECODE_SYMBOLX2_2(op1, &bitD1);
- HUF_DECODE_SYMBOLX2_2(op2, &bitD2);
- HUF_DECODE_SYMBOLX2_2(op3, &bitD3);
- HUF_DECODE_SYMBOLX2_2(op4, &bitD4);
- HUF_DECODE_SYMBOLX2_1(op1, &bitD1);
- HUF_DECODE_SYMBOLX2_1(op2, &bitD2);
- HUF_DECODE_SYMBOLX2_1(op3, &bitD3);
- HUF_DECODE_SYMBOLX2_1(op4, &bitD4);
- HUF_DECODE_SYMBOLX2_2(op1, &bitD1);
- HUF_DECODE_SYMBOLX2_2(op2, &bitD2);
- HUF_DECODE_SYMBOLX2_2(op3, &bitD3);
- HUF_DECODE_SYMBOLX2_2(op4, &bitD4);
- HUF_DECODE_SYMBOLX2_0(op1, &bitD1);
- HUF_DECODE_SYMBOLX2_0(op2, &bitD2);
- HUF_DECODE_SYMBOLX2_0(op3, &bitD3);
- HUF_DECODE_SYMBOLX2_0(op4, &bitD4);
+ HUF_DECODE_SYMBOLX1_2(op1, &bitD1);
+ HUF_DECODE_SYMBOLX1_2(op2, &bitD2);
+ HUF_DECODE_SYMBOLX1_2(op3, &bitD3);
+ HUF_DECODE_SYMBOLX1_2(op4, &bitD4);
+ HUF_DECODE_SYMBOLX1_1(op1, &bitD1);
+ HUF_DECODE_SYMBOLX1_1(op2, &bitD2);
+ HUF_DECODE_SYMBOLX1_1(op3, &bitD3);
+ HUF_DECODE_SYMBOLX1_1(op4, &bitD4);
+ HUF_DECODE_SYMBOLX1_2(op1, &bitD1);
+ HUF_DECODE_SYMBOLX1_2(op2, &bitD2);
+ HUF_DECODE_SYMBOLX1_2(op3, &bitD3);
+ HUF_DECODE_SYMBOLX1_2(op4, &bitD4);
+ HUF_DECODE_SYMBOLX1_0(op1, &bitD1);
+ HUF_DECODE_SYMBOLX1_0(op2, &bitD2);
+ HUF_DECODE_SYMBOLX1_0(op3, &bitD3);
+ HUF_DECODE_SYMBOLX1_0(op4, &bitD4);
BIT_reloadDStream(&bitD1);
BIT_reloadDStream(&bitD2);
BIT_reloadDStream(&bitD3);
@@ -291,191 +348,10 @@
/* note : op4 supposed already verified within main loop */
/* finish bitStreams one by one */
- HUF_decodeStreamX2(op1, &bitD1, opStart2, dt, dtLog);
- HUF_decodeStreamX2(op2, &bitD2, opStart3, dt, dtLog);
- HUF_decodeStreamX2(op3, &bitD3, opStart4, dt, dtLog);
- HUF_decodeStreamX2(op4, &bitD4, oend, dt, dtLog);
-
- /* check */
- { U32 const endCheck = BIT_endOfDStream(&bitD1) & BIT_endOfDStream(&bitD2) & BIT_endOfDStream(&bitD3) & BIT_endOfDStream(&bitD4);
- if (!endCheck) return ERROR(corruption_detected); }
-
- /* decoded size */
- return dstSize;
- }
-}
-
-
-FORCE_INLINE_TEMPLATE U32
-HUF_decodeSymbolX4(void* op, BIT_DStream_t* DStream, const HUF_DEltX4* dt, const U32 dtLog)
-{
- size_t const val = BIT_lookBitsFast(DStream, dtLog); /* note : dtLog >= 1 */
- memcpy(op, dt+val, 2);
- BIT_skipBits(DStream, dt[val].nbBits);
- return dt[val].length;
-}
-
-FORCE_INLINE_TEMPLATE U32
-HUF_decodeLastSymbolX4(void* op, BIT_DStream_t* DStream, const HUF_DEltX4* dt, const U32 dtLog)
-{
- size_t const val = BIT_lookBitsFast(DStream, dtLog); /* note : dtLog >= 1 */
- memcpy(op, dt+val, 1);
- if (dt[val].length==1) BIT_skipBits(DStream, dt[val].nbBits);
- else {
- if (DStream->bitsConsumed < (sizeof(DStream->bitContainer)*8)) {
- BIT_skipBits(DStream, dt[val].nbBits);
- if (DStream->bitsConsumed > (sizeof(DStream->bitContainer)*8))
- /* ugly hack; works only because it's the last symbol. Note : can't easily extract nbBits from just this symbol */
- DStream->bitsConsumed = (sizeof(DStream->bitContainer)*8);
- } }
- return 1;
-}
-
-#define HUF_DECODE_SYMBOLX4_0(ptr, DStreamPtr) \
- ptr += HUF_decodeSymbolX4(ptr, DStreamPtr, dt, dtLog)
-
-#define HUF_DECODE_SYMBOLX4_1(ptr, DStreamPtr) \
- if (MEM_64bits() || (HUF_TABLELOG_MAX<=12)) \
- ptr += HUF_decodeSymbolX4(ptr, DStreamPtr, dt, dtLog)
-
-#define HUF_DECODE_SYMBOLX4_2(ptr, DStreamPtr) \
- if (MEM_64bits()) \
- ptr += HUF_decodeSymbolX4(ptr, DStreamPtr, dt, dtLog)
-
-HINT_INLINE size_t
-HUF_decodeStreamX4(BYTE* p, BIT_DStream_t* bitDPtr, BYTE* const pEnd,
- const HUF_DEltX4* const dt, const U32 dtLog)
-{
- BYTE* const pStart = p;
-
- /* up to 8 symbols at a time */
- while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p < pEnd-(sizeof(bitDPtr->bitContainer)-1))) {
- HUF_DECODE_SYMBOLX4_2(p, bitDPtr);
- HUF_DECODE_SYMBOLX4_1(p, bitDPtr);
- HUF_DECODE_SYMBOLX4_2(p, bitDPtr);
- HUF_DECODE_SYMBOLX4_0(p, bitDPtr);
- }
-
- /* closer to end : up to 2 symbols at a time */
- while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p <= pEnd-2))
- HUF_DECODE_SYMBOLX4_0(p, bitDPtr);
-
- while (p <= pEnd-2)
- HUF_DECODE_SYMBOLX4_0(p, bitDPtr); /* no need to reload : reached the end of DStream */
-
- if (p < pEnd)
- p += HUF_decodeLastSymbolX4(p, bitDPtr, dt, dtLog);
-
- return p-pStart;
-}
-
-FORCE_INLINE_TEMPLATE size_t
-HUF_decompress1X4_usingDTable_internal_body(
- void* dst, size_t dstSize,
- const void* cSrc, size_t cSrcSize,
- const HUF_DTable* DTable)
-{
- BIT_DStream_t bitD;
-
- /* Init */
- CHECK_F( BIT_initDStream(&bitD, cSrc, cSrcSize) );
-
- /* decode */
- { BYTE* const ostart = (BYTE*) dst;
- BYTE* const oend = ostart + dstSize;
- const void* const dtPtr = DTable+1; /* force compiler to not use strict-aliasing */
- const HUF_DEltX4* const dt = (const HUF_DEltX4*)dtPtr;
- DTableDesc const dtd = HUF_getDTableDesc(DTable);
- HUF_decodeStreamX4(ostart, &bitD, oend, dt, dtd.tableLog);
- }
-
- /* check */
- if (!BIT_endOfDStream(&bitD)) return ERROR(corruption_detected);
-
- /* decoded size */
- return dstSize;
-}
-
-
-FORCE_INLINE_TEMPLATE size_t
-HUF_decompress4X4_usingDTable_internal_body(
- void* dst, size_t dstSize,
- const void* cSrc, size_t cSrcSize,
- const HUF_DTable* DTable)
-{
- if (cSrcSize < 10) return ERROR(corruption_detected); /* strict minimum : jump table + 1 byte per stream */
-
- { const BYTE* const istart = (const BYTE*) cSrc;
- BYTE* const ostart = (BYTE*) dst;
- BYTE* const oend = ostart + dstSize;
- const void* const dtPtr = DTable+1;
- const HUF_DEltX4* const dt = (const HUF_DEltX4*)dtPtr;
-
- /* Init */
- BIT_DStream_t bitD1;
- BIT_DStream_t bitD2;
- BIT_DStream_t bitD3;
- BIT_DStream_t bitD4;
- size_t const length1 = MEM_readLE16(istart);
- size_t const length2 = MEM_readLE16(istart+2);
- size_t const length3 = MEM_readLE16(istart+4);
- size_t const length4 = cSrcSize - (length1 + length2 + length3 + 6);
- const BYTE* const istart1 = istart + 6; /* jumpTable */
- const BYTE* const istart2 = istart1 + length1;
- const BYTE* const istart3 = istart2 + length2;
- const BYTE* const istart4 = istart3 + length3;
- size_t const segmentSize = (dstSize+3) / 4;
- BYTE* const opStart2 = ostart + segmentSize;
- BYTE* const opStart3 = opStart2 + segmentSize;
- BYTE* const opStart4 = opStart3 + segmentSize;
- BYTE* op1 = ostart;
- BYTE* op2 = opStart2;
- BYTE* op3 = opStart3;
- BYTE* op4 = opStart4;
- U32 endSignal;
- DTableDesc const dtd = HUF_getDTableDesc(DTable);
- U32 const dtLog = dtd.tableLog;
-
- if (length4 > cSrcSize) return ERROR(corruption_detected); /* overflow */
- CHECK_F( BIT_initDStream(&bitD1, istart1, length1) );
- CHECK_F( BIT_initDStream(&bitD2, istart2, length2) );
- CHECK_F( BIT_initDStream(&bitD3, istart3, length3) );
- CHECK_F( BIT_initDStream(&bitD4, istart4, length4) );
-
- /* 16-32 symbols per loop (4-8 symbols per stream) */
- endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
- for ( ; (endSignal==BIT_DStream_unfinished) & (op4<(oend-(sizeof(bitD4.bitContainer)-1))) ; ) {
- HUF_DECODE_SYMBOLX4_2(op1, &bitD1);
- HUF_DECODE_SYMBOLX4_2(op2, &bitD2);
- HUF_DECODE_SYMBOLX4_2(op3, &bitD3);
- HUF_DECODE_SYMBOLX4_2(op4, &bitD4);
- HUF_DECODE_SYMBOLX4_1(op1, &bitD1);
- HUF_DECODE_SYMBOLX4_1(op2, &bitD2);
- HUF_DECODE_SYMBOLX4_1(op3, &bitD3);
- HUF_DECODE_SYMBOLX4_1(op4, &bitD4);
- HUF_DECODE_SYMBOLX4_2(op1, &bitD1);
- HUF_DECODE_SYMBOLX4_2(op2, &bitD2);
- HUF_DECODE_SYMBOLX4_2(op3, &bitD3);
- HUF_DECODE_SYMBOLX4_2(op4, &bitD4);
- HUF_DECODE_SYMBOLX4_0(op1, &bitD1);
- HUF_DECODE_SYMBOLX4_0(op2, &bitD2);
- HUF_DECODE_SYMBOLX4_0(op3, &bitD3);
- HUF_DECODE_SYMBOLX4_0(op4, &bitD4);
-
- endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
- }
-
- /* check corruption */
- if (op1 > opStart2) return ERROR(corruption_detected);
- if (op2 > opStart3) return ERROR(corruption_detected);
- if (op3 > opStart4) return ERROR(corruption_detected);
- /* note : op4 already verified within main loop */
-
- /* finish bitStreams one by one */
- HUF_decodeStreamX4(op1, &bitD1, opStart2, dt, dtLog);
- HUF_decodeStreamX4(op2, &bitD2, opStart3, dt, dtLog);
- HUF_decodeStreamX4(op3, &bitD3, opStart4, dt, dtLog);
- HUF_decodeStreamX4(op4, &bitD4, oend, dt, dtLog);
+ HUF_decodeStreamX1(op1, &bitD1, opStart2, dt, dtLog);
+ HUF_decodeStreamX1(op2, &bitD2, opStart3, dt, dtLog);
+ HUF_decodeStreamX1(op3, &bitD3, opStart4, dt, dtLog);
+ HUF_decodeStreamX1(op4, &bitD4, oend, dt, dtLog);
/* check */
{ U32 const endCheck = BIT_endOfDStream(&bitD1) & BIT_endOfDStream(&bitD2) & BIT_endOfDStream(&bitD3) & BIT_endOfDStream(&bitD4);
@@ -491,153 +367,119 @@
const void *cSrc,
size_t cSrcSize,
const HUF_DTable *DTable);
-#if DYNAMIC_BMI2
-#define X(fn) \
- \
- static size_t fn##_default( \
- void* dst, size_t dstSize, \
- const void* cSrc, size_t cSrcSize, \
- const HUF_DTable* DTable) \
- { \
- return fn##_body(dst, dstSize, cSrc, cSrcSize, DTable); \
- } \
- \
- static TARGET_ATTRIBUTE("bmi2") size_t fn##_bmi2( \
- void* dst, size_t dstSize, \
- const void* cSrc, size_t cSrcSize, \
- const HUF_DTable* DTable) \
- { \
- return fn##_body(dst, dstSize, cSrc, cSrcSize, DTable); \
- } \
- \
- static size_t fn(void* dst, size_t dstSize, void const* cSrc, \
- size_t cSrcSize, HUF_DTable const* DTable, int bmi2) \
- { \
- if (bmi2) { \
- return fn##_bmi2(dst, dstSize, cSrc, cSrcSize, DTable); \
- } \
- return fn##_default(dst, dstSize, cSrc, cSrcSize, DTable); \
- }
-
-#else
-
-#define X(fn) \
- static size_t fn(void* dst, size_t dstSize, void const* cSrc, \
- size_t cSrcSize, HUF_DTable const* DTable, int bmi2) \
- { \
- (void)bmi2; \
- return fn##_body(dst, dstSize, cSrc, cSrcSize, DTable); \
- }
-
-#endif
-
-X(HUF_decompress1X2_usingDTable_internal)
-X(HUF_decompress4X2_usingDTable_internal)
-X(HUF_decompress1X4_usingDTable_internal)
-X(HUF_decompress4X4_usingDTable_internal)
-
-#undef X
+HUF_DGEN(HUF_decompress1X1_usingDTable_internal)
+HUF_DGEN(HUF_decompress4X1_usingDTable_internal)
-size_t HUF_decompress1X2_usingDTable(
+
+size_t HUF_decompress1X1_usingDTable(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
const HUF_DTable* DTable)
{
DTableDesc dtd = HUF_getDTableDesc(DTable);
if (dtd.tableType != 0) return ERROR(GENERIC);
- return HUF_decompress1X2_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+ return HUF_decompress1X1_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
}
-size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* DCtx, void* dst, size_t dstSize,
+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)
{
const BYTE* ip = (const BYTE*) cSrc;
- size_t const hSize = HUF_readDTableX2_wksp(DCtx, cSrc, cSrcSize, workSpace, wkspSize);
+ size_t const hSize = HUF_readDTableX1_wksp(DCtx, cSrc, cSrcSize, workSpace, wkspSize);
if (HUF_isError(hSize)) return hSize;
if (hSize >= cSrcSize) return ERROR(srcSize_wrong);
ip += hSize; cSrcSize -= hSize;
- return HUF_decompress1X2_usingDTable_internal(dst, dstSize, ip, cSrcSize, DCtx, /* bmi2 */ 0);
+ return HUF_decompress1X1_usingDTable_internal(dst, dstSize, ip, cSrcSize, DCtx, /* bmi2 */ 0);
}
-size_t HUF_decompress1X2_DCtx(HUF_DTable* DCtx, void* dst, size_t dstSize,
+size_t HUF_decompress1X1_DCtx(HUF_DTable* DCtx, void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize)
{
U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32];
- return HUF_decompress1X2_DCtx_wksp(DCtx, dst, dstSize, cSrc, cSrcSize,
+ return HUF_decompress1X1_DCtx_wksp(DCtx, dst, dstSize, cSrc, cSrcSize,
workSpace, sizeof(workSpace));
}
-size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
+size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
{
- HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_TABLELOG_MAX);
- return HUF_decompress1X2_DCtx (DTable, dst, dstSize, cSrc, cSrcSize);
+ HUF_CREATE_STATIC_DTABLEX1(DTable, HUF_TABLELOG_MAX);
+ return HUF_decompress1X1_DCtx (DTable, dst, dstSize, cSrc, cSrcSize);
}
-size_t HUF_decompress4X2_usingDTable(
+size_t HUF_decompress4X1_usingDTable(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
const HUF_DTable* DTable)
{
DTableDesc dtd = HUF_getDTableDesc(DTable);
if (dtd.tableType != 0) return ERROR(GENERIC);
- return HUF_decompress4X2_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+ return HUF_decompress4X1_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
}
-static size_t HUF_decompress4X2_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize,
+static size_t HUF_decompress4X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
void* workSpace, size_t wkspSize, int bmi2)
{
const BYTE* ip = (const BYTE*) cSrc;
- size_t const hSize = HUF_readDTableX2_wksp (dctx, cSrc, cSrcSize,
+ size_t const hSize = HUF_readDTableX1_wksp (dctx, cSrc, cSrcSize,
workSpace, wkspSize);
if (HUF_isError(hSize)) return hSize;
if (hSize >= cSrcSize) return ERROR(srcSize_wrong);
ip += hSize; cSrcSize -= hSize;
- return HUF_decompress4X2_usingDTable_internal(dst, dstSize, ip, cSrcSize, dctx, bmi2);
+ return HUF_decompress4X1_usingDTable_internal(dst, dstSize, ip, cSrcSize, dctx, bmi2);
}
-size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize,
+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)
{
- return HUF_decompress4X2_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, 0);
+ return HUF_decompress4X1_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, 0);
}
-size_t HUF_decompress4X2_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
+size_t HUF_decompress4X1_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
{
U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32];
- return HUF_decompress4X2_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize,
+ return HUF_decompress4X1_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize,
workSpace, sizeof(workSpace));
}
-size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
+size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
{
- HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_TABLELOG_MAX);
- return HUF_decompress4X2_DCtx(DTable, dst, dstSize, cSrc, cSrcSize);
+ HUF_CREATE_STATIC_DTABLEX1(DTable, HUF_TABLELOG_MAX);
+ return HUF_decompress4X1_DCtx(DTable, dst, dstSize, cSrc, cSrcSize);
}
+#endif /* HUF_FORCE_DECOMPRESS_X2 */
+
+
+#ifndef HUF_FORCE_DECOMPRESS_X1
/* *************************/
/* double-symbols decoding */
/* *************************/
-typedef struct { BYTE symbol; BYTE weight; } sortedSymbol_t;
-/* HUF_fillDTableX4Level2() :
+typedef struct { U16 sequence; BYTE nbBits; BYTE length; } HUF_DEltX2; /* double-symbols decoding */
+typedef struct { BYTE symbol; BYTE weight; } sortedSymbol_t;
+typedef U32 rankValCol_t[HUF_TABLELOG_MAX + 1];
+typedef rankValCol_t rankVal_t[HUF_TABLELOG_MAX];
+
+
+/* HUF_fillDTableX2Level2() :
* `rankValOrigin` must be a table of at least (HUF_TABLELOG_MAX + 1) U32 */
-static void HUF_fillDTableX4Level2(HUF_DEltX4* DTable, U32 sizeLog, const U32 consumed,
+static void HUF_fillDTableX2Level2(HUF_DEltX2* DTable, U32 sizeLog, const U32 consumed,
const U32* rankValOrigin, const int minWeight,
const sortedSymbol_t* sortedSymbols, const U32 sortedListSize,
U32 nbBitsBaseline, U16 baseSeq)
{
- HUF_DEltX4 DElt;
+ HUF_DEltX2 DElt;
U32 rankVal[HUF_TABLELOG_MAX + 1];
/* get pre-calculated rankVal */
@@ -672,10 +514,8 @@
} }
}
-typedef U32 rankValCol_t[HUF_TABLELOG_MAX + 1];
-typedef rankValCol_t rankVal_t[HUF_TABLELOG_MAX];
-static void HUF_fillDTableX4(HUF_DEltX4* DTable, const U32 targetLog,
+static void HUF_fillDTableX2(HUF_DEltX2* DTable, const U32 targetLog,
const sortedSymbol_t* sortedList, const U32 sortedListSize,
const U32* rankStart, rankVal_t rankValOrigin, const U32 maxWeight,
const U32 nbBitsBaseline)
@@ -700,12 +540,12 @@
int minWeight = nbBits + scaleLog;
if (minWeight < 1) minWeight = 1;
sortedRank = rankStart[minWeight];
- HUF_fillDTableX4Level2(DTable+start, targetLog-nbBits, nbBits,
+ HUF_fillDTableX2Level2(DTable+start, targetLog-nbBits, nbBits,
rankValOrigin[nbBits], minWeight,
sortedList+sortedRank, sortedListSize-sortedRank,
nbBitsBaseline, symbol);
} else {
- HUF_DEltX4 DElt;
+ HUF_DEltX2 DElt;
MEM_writeLE16(&(DElt.sequence), symbol);
DElt.nbBits = (BYTE)(nbBits);
DElt.length = 1;
@@ -717,16 +557,16 @@
}
}
-size_t HUF_readDTableX4_wksp(HUF_DTable* DTable, const void* src,
- size_t srcSize, void* workSpace,
- size_t wkspSize)
+size_t HUF_readDTableX2_wksp(HUF_DTable* DTable,
+ const void* src, size_t srcSize,
+ void* workSpace, size_t wkspSize)
{
U32 tableLog, maxW, sizeOfSort, nbSymbols;
DTableDesc dtd = HUF_getDTableDesc(DTable);
U32 const maxTableLog = dtd.maxTableLog;
size_t iSize;
void* dtPtr = DTable+1; /* force compiler to avoid strict-aliasing */
- HUF_DEltX4* const dt = (HUF_DEltX4*)dtPtr;
+ HUF_DEltX2* const dt = (HUF_DEltX2*)dtPtr;
U32 *rankStart;
rankValCol_t* rankVal;
@@ -752,7 +592,7 @@
rankStart = rankStart0 + 1;
memset(rankStats, 0, sizeof(U32) * (2 * HUF_TABLELOG_MAX + 2 + 1));
- HUF_STATIC_ASSERT(sizeof(HUF_DEltX4) == sizeof(HUF_DTable)); /* if compiler fails here, assertion is wrong */
+ DEBUG_STATIC_ASSERT(sizeof(HUF_DEltX2) == sizeof(HUF_DTable)); /* if compiler fails here, assertion is wrong */
if (maxTableLog > HUF_TABLELOG_MAX) return ERROR(tableLog_tooLarge);
/* memset(weightList, 0, sizeof(weightList)); */ /* is not necessary, even though some analyzer complain ... */
@@ -806,7 +646,7 @@
rankValPtr[w] = rankVal0[w] >> consumed;
} } } }
- HUF_fillDTableX4(dt, maxTableLog,
+ HUF_fillDTableX2(dt, maxTableLog,
sortedSymbol, sizeOfSort,
rankStart0, rankVal, maxW,
tableLog+1);
@@ -817,112 +657,308 @@
return iSize;
}
-size_t HUF_readDTableX4(HUF_DTable* DTable, const void* src, size_t srcSize)
+size_t HUF_readDTableX2(HUF_DTable* DTable, const void* src, size_t srcSize)
{
U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32];
- return HUF_readDTableX4_wksp(DTable, src, srcSize,
+ return HUF_readDTableX2_wksp(DTable, src, srcSize,
workSpace, sizeof(workSpace));
}
-size_t HUF_decompress1X4_usingDTable(
+
+FORCE_INLINE_TEMPLATE U32
+HUF_decodeSymbolX2(void* op, BIT_DStream_t* DStream, const HUF_DEltX2* dt, const U32 dtLog)
+{
+ size_t const val = BIT_lookBitsFast(DStream, dtLog); /* note : dtLog >= 1 */
+ memcpy(op, dt+val, 2);
+ BIT_skipBits(DStream, dt[val].nbBits);
+ return dt[val].length;
+}
+
+FORCE_INLINE_TEMPLATE U32
+HUF_decodeLastSymbolX2(void* op, BIT_DStream_t* DStream, const HUF_DEltX2* dt, const U32 dtLog)
+{
+ size_t const val = BIT_lookBitsFast(DStream, dtLog); /* note : dtLog >= 1 */
+ memcpy(op, dt+val, 1);
+ if (dt[val].length==1) BIT_skipBits(DStream, dt[val].nbBits);
+ else {
+ if (DStream->bitsConsumed < (sizeof(DStream->bitContainer)*8)) {
+ BIT_skipBits(DStream, dt[val].nbBits);
+ if (DStream->bitsConsumed > (sizeof(DStream->bitContainer)*8))
+ /* ugly hack; works only because it's the last symbol. Note : can't easily extract nbBits from just this symbol */
+ DStream->bitsConsumed = (sizeof(DStream->bitContainer)*8);
+ } }
+ return 1;
+}
+
+#define HUF_DECODE_SYMBOLX2_0(ptr, DStreamPtr) \
+ ptr += HUF_decodeSymbolX2(ptr, DStreamPtr, dt, dtLog)
+
+#define HUF_DECODE_SYMBOLX2_1(ptr, DStreamPtr) \
+ if (MEM_64bits() || (HUF_TABLELOG_MAX<=12)) \
+ ptr += HUF_decodeSymbolX2(ptr, DStreamPtr, dt, dtLog)
+
+#define HUF_DECODE_SYMBOLX2_2(ptr, DStreamPtr) \
+ if (MEM_64bits()) \
+ ptr += HUF_decodeSymbolX2(ptr, DStreamPtr, dt, dtLog)
+
+HINT_INLINE size_t
+HUF_decodeStreamX2(BYTE* p, BIT_DStream_t* bitDPtr, BYTE* const pEnd,
+ const HUF_DEltX2* const dt, const U32 dtLog)
+{
+ BYTE* const pStart = p;
+
+ /* up to 8 symbols at a time */
+ while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p < pEnd-(sizeof(bitDPtr->bitContainer)-1))) {
+ HUF_DECODE_SYMBOLX2_2(p, bitDPtr);
+ HUF_DECODE_SYMBOLX2_1(p, bitDPtr);
+ HUF_DECODE_SYMBOLX2_2(p, bitDPtr);
+ HUF_DECODE_SYMBOLX2_0(p, bitDPtr);
+ }
+
+ /* closer to end : up to 2 symbols at a time */
+ while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p <= pEnd-2))
+ HUF_DECODE_SYMBOLX2_0(p, bitDPtr);
+
+ while (p <= pEnd-2)
+ HUF_DECODE_SYMBOLX2_0(p, bitDPtr); /* no need to reload : reached the end of DStream */
+
+ if (p < pEnd)
+ p += HUF_decodeLastSymbolX2(p, bitDPtr, dt, dtLog);
+
+ return p-pStart;
+}
+
+FORCE_INLINE_TEMPLATE size_t
+HUF_decompress1X2_usingDTable_internal_body(
+ void* dst, size_t dstSize,
+ const void* cSrc, size_t cSrcSize,
+ const HUF_DTable* DTable)
+{
+ BIT_DStream_t bitD;
+
+ /* Init */
+ CHECK_F( BIT_initDStream(&bitD, cSrc, cSrcSize) );
+
+ /* decode */
+ { BYTE* const ostart = (BYTE*) dst;
+ BYTE* const oend = ostart + dstSize;
+ const void* const dtPtr = DTable+1; /* force compiler to not use strict-aliasing */
+ const HUF_DEltX2* const dt = (const HUF_DEltX2*)dtPtr;
+ DTableDesc const dtd = HUF_getDTableDesc(DTable);
+ HUF_decodeStreamX2(ostart, &bitD, oend, dt, dtd.tableLog);
+ }
+
+ /* check */
+ if (!BIT_endOfDStream(&bitD)) return ERROR(corruption_detected);
+
+ /* decoded size */
+ return dstSize;
+}
+
+
+FORCE_INLINE_TEMPLATE size_t
+HUF_decompress4X2_usingDTable_internal_body(
+ void* dst, size_t dstSize,
+ const void* cSrc, size_t cSrcSize,
+ const HUF_DTable* DTable)
+{
+ if (cSrcSize < 10) return ERROR(corruption_detected); /* strict minimum : jump table + 1 byte per stream */
+
+ { const BYTE* const istart = (const BYTE*) cSrc;
+ BYTE* const ostart = (BYTE*) dst;
+ BYTE* const oend = ostart + dstSize;
+ const void* const dtPtr = DTable+1;
+ const HUF_DEltX2* const dt = (const HUF_DEltX2*)dtPtr;
+
+ /* Init */
+ BIT_DStream_t bitD1;
+ BIT_DStream_t bitD2;
+ BIT_DStream_t bitD3;
+ BIT_DStream_t bitD4;
+ size_t const length1 = MEM_readLE16(istart);
+ size_t const length2 = MEM_readLE16(istart+2);
+ size_t const length3 = MEM_readLE16(istart+4);
+ size_t const length4 = cSrcSize - (length1 + length2 + length3 + 6);
+ const BYTE* const istart1 = istart + 6; /* jumpTable */
+ const BYTE* const istart2 = istart1 + length1;
+ const BYTE* const istart3 = istart2 + length2;
+ const BYTE* const istart4 = istart3 + length3;
+ size_t const segmentSize = (dstSize+3) / 4;
+ BYTE* const opStart2 = ostart + segmentSize;
+ BYTE* const opStart3 = opStart2 + segmentSize;
+ BYTE* const opStart4 = opStart3 + segmentSize;
+ BYTE* op1 = ostart;
+ BYTE* op2 = opStart2;
+ BYTE* op3 = opStart3;
+ BYTE* op4 = opStart4;
+ U32 endSignal;
+ DTableDesc const dtd = HUF_getDTableDesc(DTable);
+ U32 const dtLog = dtd.tableLog;
+
+ if (length4 > cSrcSize) return ERROR(corruption_detected); /* overflow */
+ CHECK_F( BIT_initDStream(&bitD1, istart1, length1) );
+ CHECK_F( BIT_initDStream(&bitD2, istart2, length2) );
+ CHECK_F( BIT_initDStream(&bitD3, istart3, length3) );
+ CHECK_F( BIT_initDStream(&bitD4, istart4, length4) );
+
+ /* 16-32 symbols per loop (4-8 symbols per stream) */
+ endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
+ for ( ; (endSignal==BIT_DStream_unfinished) & (op4<(oend-(sizeof(bitD4.bitContainer)-1))) ; ) {
+ HUF_DECODE_SYMBOLX2_2(op1, &bitD1);
+ HUF_DECODE_SYMBOLX2_2(op2, &bitD2);
+ HUF_DECODE_SYMBOLX2_2(op3, &bitD3);
+ HUF_DECODE_SYMBOLX2_2(op4, &bitD4);
+ HUF_DECODE_SYMBOLX2_1(op1, &bitD1);
+ HUF_DECODE_SYMBOLX2_1(op2, &bitD2);
+ HUF_DECODE_SYMBOLX2_1(op3, &bitD3);
+ HUF_DECODE_SYMBOLX2_1(op4, &bitD4);
+ HUF_DECODE_SYMBOLX2_2(op1, &bitD1);
+ HUF_DECODE_SYMBOLX2_2(op2, &bitD2);
+ HUF_DECODE_SYMBOLX2_2(op3, &bitD3);
+ HUF_DECODE_SYMBOLX2_2(op4, &bitD4);
+ HUF_DECODE_SYMBOLX2_0(op1, &bitD1);
+ HUF_DECODE_SYMBOLX2_0(op2, &bitD2);
+ HUF_DECODE_SYMBOLX2_0(op3, &bitD3);
+ HUF_DECODE_SYMBOLX2_0(op4, &bitD4);
+
+ endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
+ }
+
+ /* check corruption */
+ if (op1 > opStart2) return ERROR(corruption_detected);
+ if (op2 > opStart3) return ERROR(corruption_detected);
+ if (op3 > opStart4) return ERROR(corruption_detected);
+ /* note : op4 already verified within main loop */
+
+ /* finish bitStreams one by one */
+ HUF_decodeStreamX2(op1, &bitD1, opStart2, dt, dtLog);
+ HUF_decodeStreamX2(op2, &bitD2, opStart3, dt, dtLog);
+ HUF_decodeStreamX2(op3, &bitD3, opStart4, dt, dtLog);
+ HUF_decodeStreamX2(op4, &bitD4, oend, dt, dtLog);
+
+ /* check */
+ { U32 const endCheck = BIT_endOfDStream(&bitD1) & BIT_endOfDStream(&bitD2) & BIT_endOfDStream(&bitD3) & BIT_endOfDStream(&bitD4);
+ if (!endCheck) return ERROR(corruption_detected); }
+
+ /* decoded size */
+ return dstSize;
+ }
+}
+
+HUF_DGEN(HUF_decompress1X2_usingDTable_internal)
+HUF_DGEN(HUF_decompress4X2_usingDTable_internal)
+
+size_t HUF_decompress1X2_usingDTable(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
const HUF_DTable* DTable)
{
DTableDesc dtd = HUF_getDTableDesc(DTable);
if (dtd.tableType != 1) return ERROR(GENERIC);
- return HUF_decompress1X4_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+ return HUF_decompress1X2_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
}
-size_t HUF_decompress1X4_DCtx_wksp(HUF_DTable* DCtx, void* dst, size_t dstSize,
+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)
{
const BYTE* ip = (const BYTE*) cSrc;
- size_t const hSize = HUF_readDTableX4_wksp(DCtx, cSrc, cSrcSize,
+ size_t const hSize = HUF_readDTableX2_wksp(DCtx, cSrc, cSrcSize,
workSpace, wkspSize);
if (HUF_isError(hSize)) return hSize;
if (hSize >= cSrcSize) return ERROR(srcSize_wrong);
ip += hSize; cSrcSize -= hSize;
- return HUF_decompress1X4_usingDTable_internal(dst, dstSize, ip, cSrcSize, DCtx, /* bmi2 */ 0);
+ return HUF_decompress1X2_usingDTable_internal(dst, dstSize, ip, cSrcSize, DCtx, /* bmi2 */ 0);
}
-size_t HUF_decompress1X4_DCtx(HUF_DTable* DCtx, void* dst, size_t dstSize,
+size_t HUF_decompress1X2_DCtx(HUF_DTable* DCtx, void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize)
{
U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32];
- return HUF_decompress1X4_DCtx_wksp(DCtx, dst, dstSize, cSrc, cSrcSize,
+ return HUF_decompress1X2_DCtx_wksp(DCtx, dst, dstSize, cSrc, cSrcSize,
workSpace, sizeof(workSpace));
}
-size_t HUF_decompress1X4 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
+size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
{
- HUF_CREATE_STATIC_DTABLEX4(DTable, HUF_TABLELOG_MAX);
- return HUF_decompress1X4_DCtx(DTable, dst, dstSize, cSrc, cSrcSize);
+ HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_TABLELOG_MAX);
+ return HUF_decompress1X2_DCtx(DTable, dst, dstSize, cSrc, cSrcSize);
}
-size_t HUF_decompress4X4_usingDTable(
+size_t HUF_decompress4X2_usingDTable(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
const HUF_DTable* DTable)
{
DTableDesc dtd = HUF_getDTableDesc(DTable);
if (dtd.tableType != 1) return ERROR(GENERIC);
- return HUF_decompress4X4_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+ return HUF_decompress4X2_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
}
-static size_t HUF_decompress4X4_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize,
+static size_t HUF_decompress4X2_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
void* workSpace, size_t wkspSize, int bmi2)
{
const BYTE* ip = (const BYTE*) cSrc;
- size_t hSize = HUF_readDTableX4_wksp(dctx, cSrc, cSrcSize,
+ size_t hSize = HUF_readDTableX2_wksp(dctx, cSrc, cSrcSize,
workSpace, wkspSize);
if (HUF_isError(hSize)) return hSize;
if (hSize >= cSrcSize) return ERROR(srcSize_wrong);
ip += hSize; cSrcSize -= hSize;
- return HUF_decompress4X4_usingDTable_internal(dst, dstSize, ip, cSrcSize, dctx, bmi2);
+ return HUF_decompress4X2_usingDTable_internal(dst, dstSize, ip, cSrcSize, dctx, bmi2);
}
-size_t HUF_decompress4X4_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize,
+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)
{
- return HUF_decompress4X4_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, /* bmi2 */ 0);
+ return HUF_decompress4X2_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, /* bmi2 */ 0);
}
-size_t HUF_decompress4X4_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize,
+size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize)
{
U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32];
- return HUF_decompress4X4_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize,
+ return HUF_decompress4X2_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize,
workSpace, sizeof(workSpace));
}
-size_t HUF_decompress4X4 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
+size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
{
- HUF_CREATE_STATIC_DTABLEX4(DTable, HUF_TABLELOG_MAX);
- return HUF_decompress4X4_DCtx(DTable, dst, dstSize, cSrc, cSrcSize);
+ HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_TABLELOG_MAX);
+ return HUF_decompress4X2_DCtx(DTable, dst, dstSize, cSrc, cSrcSize);
}
+#endif /* HUF_FORCE_DECOMPRESS_X1 */
-/* ********************************/
-/* Generic decompression selector */
-/* ********************************/
+
+/* ***********************************/
+/* Universal decompression selectors */
+/* ***********************************/
size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize,
const void* cSrc, size_t cSrcSize,
const HUF_DTable* DTable)
{
DTableDesc const dtd = HUF_getDTableDesc(DTable);
- return dtd.tableType ? HUF_decompress1X4_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0) :
- HUF_decompress1X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+#if defined(HUF_FORCE_DECOMPRESS_X1)
+ (void)dtd;
+ assert(dtd.tableType == 0);
+ return HUF_decompress1X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+#elif defined(HUF_FORCE_DECOMPRESS_X2)
+ (void)dtd;
+ assert(dtd.tableType == 1);
+ return HUF_decompress1X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+#else
+ return dtd.tableType ? HUF_decompress1X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0) :
+ HUF_decompress1X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+#endif
}
size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize,
@@ -930,11 +966,22 @@
const HUF_DTable* DTable)
{
DTableDesc const dtd = HUF_getDTableDesc(DTable);
- return dtd.tableType ? HUF_decompress4X4_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0) :
- HUF_decompress4X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+#if defined(HUF_FORCE_DECOMPRESS_X1)
+ (void)dtd;
+ assert(dtd.tableType == 0);
+ return HUF_decompress4X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+#elif defined(HUF_FORCE_DECOMPRESS_X2)
+ (void)dtd;
+ assert(dtd.tableType == 1);
+ return HUF_decompress4X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+#else
+ return dtd.tableType ? HUF_decompress4X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0) :
+ HUF_decompress4X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0);
+#endif
}
+#if !defined(HUF_FORCE_DECOMPRESS_X1) && !defined(HUF_FORCE_DECOMPRESS_X2)
typedef struct { U32 tableTime; U32 decode256Time; } algo_time_t;
static const algo_time_t algoTime[16 /* Quantization */][3 /* single, double, quad */] =
{
@@ -956,16 +1003,26 @@
{{1455,128}, {2422,124}, {4174,124}}, /* Q ==14 : 87-93% */
{{ 722,128}, {1891,145}, {1936,146}}, /* Q ==15 : 93-99% */
};
+#endif
/** HUF_selectDecoder() :
* Tells which decoder is likely to decode faster,
* based on a set of pre-computed metrics.
- * @return : 0==HUF_decompress4X2, 1==HUF_decompress4X4 .
+ * @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 .
* Assumption : 0 < dstSize <= 128 KB */
U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize)
{
assert(dstSize > 0);
- assert(dstSize <= 128 KB);
+ assert(dstSize <= 128*1024);
+#if defined(HUF_FORCE_DECOMPRESS_X1)
+ (void)dstSize;
+ (void)cSrcSize;
+ return 0;
+#elif defined(HUF_FORCE_DECOMPRESS_X2)
+ (void)dstSize;
+ (void)cSrcSize;
+ return 1;
+#else
/* decoder timing evaluation */
{ U32 const Q = (cSrcSize >= dstSize) ? 15 : (U32)(cSrcSize * 16 / dstSize); /* Q < 16 */
U32 const D256 = (U32)(dstSize >> 8);
@@ -973,14 +1030,18 @@
U32 DTime1 = algoTime[Q][1].tableTime + (algoTime[Q][1].decode256Time * D256);
DTime1 += DTime1 >> 3; /* advantage to algorithm using less memory, to reduce cache eviction */
return DTime1 < DTime0;
-} }
+ }
+#endif
+}
typedef size_t (*decompressionAlgo)(void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);
size_t HUF_decompress (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
{
- static const decompressionAlgo decompress[2] = { HUF_decompress4X2, HUF_decompress4X4 };
+#if !defined(HUF_FORCE_DECOMPRESS_X1) && !defined(HUF_FORCE_DECOMPRESS_X2)
+ static const decompressionAlgo decompress[2] = { HUF_decompress4X1, HUF_decompress4X2 };
+#endif
/* validation checks */
if (dstSize == 0) return ERROR(dstSize_tooSmall);
@@ -989,7 +1050,17 @@
if (cSrcSize == 1) { memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */
{ U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize);
+#if defined(HUF_FORCE_DECOMPRESS_X1)
+ (void)algoNb;
+ assert(algoNb == 0);
+ return HUF_decompress4X1(dst, dstSize, cSrc, cSrcSize);
+#elif defined(HUF_FORCE_DECOMPRESS_X2)
+ (void)algoNb;
+ assert(algoNb == 1);
+ return HUF_decompress4X2(dst, dstSize, cSrc, cSrcSize);
+#else
return decompress[algoNb](dst, dstSize, cSrc, cSrcSize);
+#endif
}
}
@@ -1002,8 +1073,18 @@
if (cSrcSize == 1) { memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */
{ U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize);
- return algoNb ? HUF_decompress4X4_DCtx(dctx, dst, dstSize, cSrc, cSrcSize) :
- HUF_decompress4X2_DCtx(dctx, dst, dstSize, cSrc, cSrcSize) ;
+#if defined(HUF_FORCE_DECOMPRESS_X1)
+ (void)algoNb;
+ assert(algoNb == 0);
+ return HUF_decompress4X1_DCtx(dctx, dst, dstSize, cSrc, cSrcSize);
+#elif defined(HUF_FORCE_DECOMPRESS_X2)
+ (void)algoNb;
+ assert(algoNb == 1);
+ return HUF_decompress4X2_DCtx(dctx, dst, dstSize, cSrc, cSrcSize);
+#else
+ return algoNb ? HUF_decompress4X2_DCtx(dctx, dst, dstSize, cSrc, cSrcSize) :
+ HUF_decompress4X1_DCtx(dctx, dst, dstSize, cSrc, cSrcSize) ;
+#endif
}
}
@@ -1025,8 +1106,19 @@
if (cSrcSize == 0) return ERROR(corruption_detected);
{ U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize);
- return algoNb ? HUF_decompress4X4_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize):
- HUF_decompress4X2_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize);
+#if defined(HUF_FORCE_DECOMPRESS_X1)
+ (void)algoNb;
+ assert(algoNb == 0);
+ return HUF_decompress4X1_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize);
+#elif defined(HUF_FORCE_DECOMPRESS_X2)
+ (void)algoNb;
+ assert(algoNb == 1);
+ return HUF_decompress4X2_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize);
+#else
+ return algoNb ? HUF_decompress4X2_DCtx_wksp(dctx, dst, dstSize, cSrc,
+ cSrcSize, workSpace, wkspSize):
+ HUF_decompress4X1_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize);
+#endif
}
}
@@ -1041,10 +1133,22 @@
if (cSrcSize == 1) { memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */
{ U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize);
- return algoNb ? HUF_decompress1X4_DCtx_wksp(dctx, dst, dstSize, cSrc,
- cSrcSize, workSpace, wkspSize):
- HUF_decompress1X2_DCtx_wksp(dctx, dst, dstSize, cSrc,
+#if defined(HUF_FORCE_DECOMPRESS_X1)
+ (void)algoNb;
+ assert(algoNb == 0);
+ return HUF_decompress1X1_DCtx_wksp(dctx, dst, dstSize, cSrc,
cSrcSize, workSpace, wkspSize);
+#elif defined(HUF_FORCE_DECOMPRESS_X2)
+ (void)algoNb;
+ assert(algoNb == 1);
+ return HUF_decompress1X2_DCtx_wksp(dctx, dst, dstSize, cSrc,
+ cSrcSize, workSpace, wkspSize);
+#else
+ return algoNb ? HUF_decompress1X2_DCtx_wksp(dctx, dst, dstSize, cSrc,
+ cSrcSize, workSpace, wkspSize):
+ HUF_decompress1X1_DCtx_wksp(dctx, dst, dstSize, cSrc,
+ cSrcSize, workSpace, wkspSize);
+#endif
}
}
@@ -1060,27 +1164,49 @@
size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2)
{
DTableDesc const dtd = HUF_getDTableDesc(DTable);
- return dtd.tableType ? HUF_decompress1X4_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2) :
- HUF_decompress1X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2);
+#if defined(HUF_FORCE_DECOMPRESS_X1)
+ (void)dtd;
+ assert(dtd.tableType == 0);
+ return HUF_decompress1X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2);
+#elif defined(HUF_FORCE_DECOMPRESS_X2)
+ (void)dtd;
+ assert(dtd.tableType == 1);
+ return HUF_decompress1X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2);
+#else
+ return dtd.tableType ? HUF_decompress1X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2) :
+ HUF_decompress1X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2);
+#endif
}
-size_t HUF_decompress1X2_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2)
+#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)
{
const BYTE* ip = (const BYTE*) cSrc;
- size_t const hSize = HUF_readDTableX2_wksp(dctx, cSrc, cSrcSize, workSpace, wkspSize);
+ size_t const hSize = HUF_readDTableX1_wksp(dctx, cSrc, cSrcSize, workSpace, wkspSize);
if (HUF_isError(hSize)) return hSize;
if (hSize >= cSrcSize) return ERROR(srcSize_wrong);
ip += hSize; cSrcSize -= hSize;
- return HUF_decompress1X2_usingDTable_internal(dst, dstSize, ip, cSrcSize, dctx, bmi2);
+ return HUF_decompress1X1_usingDTable_internal(dst, dstSize, ip, cSrcSize, dctx, bmi2);
}
+#endif
size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2)
{
DTableDesc const dtd = HUF_getDTableDesc(DTable);
- return dtd.tableType ? HUF_decompress4X4_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2) :
- HUF_decompress4X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2);
+#if defined(HUF_FORCE_DECOMPRESS_X1)
+ (void)dtd;
+ assert(dtd.tableType == 0);
+ return HUF_decompress4X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2);
+#elif defined(HUF_FORCE_DECOMPRESS_X2)
+ (void)dtd;
+ assert(dtd.tableType == 1);
+ return HUF_decompress4X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2);
+#else
+ return dtd.tableType ? HUF_decompress4X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2) :
+ HUF_decompress4X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2);
+#endif
}
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)
@@ -1090,7 +1216,17 @@
if (cSrcSize == 0) return ERROR(corruption_detected);
{ U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize);
- return algoNb ? HUF_decompress4X4_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, bmi2) :
- HUF_decompress4X2_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, bmi2);
+#if defined(HUF_FORCE_DECOMPRESS_X1)
+ (void)algoNb;
+ assert(algoNb == 0);
+ return HUF_decompress4X1_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, bmi2);
+#elif defined(HUF_FORCE_DECOMPRESS_X2)
+ (void)algoNb;
+ assert(algoNb == 1);
+ return HUF_decompress4X2_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, bmi2);
+#else
+ return algoNb ? HUF_decompress4X2_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, bmi2) :
+ HUF_decompress4X1_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, bmi2);
+#endif
}
}
diff --git a/vendor/github.com/DataDog/zstd/mem.h b/vendor/github.com/DataDog/zstd/mem.h
index 47d2300..5da2487 100644
--- a/vendor/github.com/DataDog/zstd/mem.h
+++ b/vendor/github.com/DataDog/zstd/mem.h
@@ -39,6 +39,10 @@
# define MEM_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */
#endif
+#ifndef __has_builtin
+# define __has_builtin(x) 0 /* compat. with non-clang compilers */
+#endif
+
/* code only tested on 32 and 64 bits systems */
#define MEM_STATIC_ASSERT(c) { enum { MEM_static_assert = 1/(int)(!!(c)) }; }
MEM_STATIC void MEM_check(void) { MEM_STATIC_ASSERT((sizeof(size_t)==4) || (sizeof(size_t)==8)); }
@@ -57,11 +61,23 @@
typedef uint64_t U64;
typedef int64_t S64;
#else
+# include <limits.h>
+#if CHAR_BIT != 8
+# error "this implementation requires char to be exactly 8-bit type"
+#endif
typedef unsigned char BYTE;
+#if USHRT_MAX != 65535
+# error "this implementation requires short to be exactly 16-bit type"
+#endif
typedef unsigned short U16;
typedef signed short S16;
+#if UINT_MAX != 4294967295
+# error "this implementation requires int to be exactly 32-bit type"
+#endif
typedef unsigned int U32;
typedef signed int S32;
+/* note : there are no limits defined for long long type in C90.
+ * limits exist in C99, however, in such case, <stdint.h> is preferred */
typedef unsigned long long U64;
typedef signed long long S64;
#endif
@@ -186,7 +202,8 @@
{
#if defined(_MSC_VER) /* Visual Studio */
return _byteswap_ulong(in);
-#elif defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)
+#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \
+ || (defined(__clang__) && __has_builtin(__builtin_bswap32))
return __builtin_bswap32(in);
#else
return ((in << 24) & 0xff000000 ) |
@@ -200,7 +217,8 @@
{
#if defined(_MSC_VER) /* Visual Studio */
return _byteswap_uint64(in);
-#elif defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)
+#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \
+ || (defined(__clang__) && __has_builtin(__builtin_bswap64))
return __builtin_bswap64(in);
#else
return ((in << 56) & 0xff00000000000000ULL) |
diff --git a/vendor/github.com/DataDog/zstd/pool.c b/vendor/github.com/DataDog/zstd/pool.c
index 773488b..7a82945 100644
--- a/vendor/github.com/DataDog/zstd/pool.c
+++ b/vendor/github.com/DataDog/zstd/pool.c
@@ -10,9 +10,10 @@
/* ====== Dependencies ======= */
-#include <stddef.h> /* size_t */
-#include "pool.h"
+#include <stddef.h> /* size_t */
+#include "debug.h" /* assert */
#include "zstd_internal.h" /* ZSTD_malloc, ZSTD_free */
+#include "pool.h"
/* ====== Compiler specifics ====== */
#if defined(_MSC_VER)
@@ -33,8 +34,9 @@
struct POOL_ctx_s {
ZSTD_customMem customMem;
/* Keep track of the threads */
- ZSTD_pthread_t *threads;
- size_t numThreads;
+ ZSTD_pthread_t* threads;
+ size_t threadCapacity;
+ size_t threadLimit;
/* The queue is a circular buffer */
POOL_job *queue;
@@ -58,10 +60,10 @@
};
/* POOL_thread() :
- Work thread for the thread pool.
- Waits for jobs and executes them.
- @returns : NULL on failure else non-null.
-*/
+ * Work thread for the thread pool.
+ * Waits for jobs and executes them.
+ * @returns : NULL on failure else non-null.
+ */
static void* POOL_thread(void* opaque) {
POOL_ctx* const ctx = (POOL_ctx*)opaque;
if (!ctx) { return NULL; }
@@ -69,50 +71,55 @@
/* Lock the mutex and wait for a non-empty queue or until shutdown */
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
- while (ctx->queueEmpty && !ctx->shutdown) {
+ while ( ctx->queueEmpty
+ || (ctx->numThreadsBusy >= ctx->threadLimit) ) {
+ if (ctx->shutdown) {
+ /* even if !queueEmpty, (possible if numThreadsBusy >= threadLimit),
+ * a few threads will be shutdown while !queueEmpty,
+ * but enough threads will remain active to finish the queue */
+ ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+ return opaque;
+ }
ZSTD_pthread_cond_wait(&ctx->queuePopCond, &ctx->queueMutex);
}
- /* empty => shutting down: so stop */
- if (ctx->queueEmpty) {
- ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
- return opaque;
- }
/* Pop a job off the queue */
{ POOL_job const job = ctx->queue[ctx->queueHead];
ctx->queueHead = (ctx->queueHead + 1) % ctx->queueSize;
ctx->numThreadsBusy++;
ctx->queueEmpty = ctx->queueHead == ctx->queueTail;
/* Unlock the mutex, signal a pusher, and run the job */
- ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
ZSTD_pthread_cond_signal(&ctx->queuePushCond);
+ ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
job.function(job.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_mutex_lock(&ctx->queueMutex);
- ctx->numThreadsBusy--;
- ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
ZSTD_pthread_cond_signal(&ctx->queuePushCond);
- } }
+ }
+ ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+ }
} /* for (;;) */
- /* Unreachable */
+ assert(0); /* Unreachable */
}
POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem);
}
-POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customMem) {
+POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
+ ZSTD_customMem customMem) {
POOL_ctx* ctx;
- /* Check the parameters */
+ /* Check parameters */
if (!numThreads) { return NULL; }
/* Allocate the context and zero initialize */
ctx = (POOL_ctx*)ZSTD_calloc(sizeof(POOL_ctx), customMem);
if (!ctx) { return NULL; }
/* Initialize the job queue.
- * It needs one extra space since one space is wasted to differentiate empty
- * and full queues.
+ * It needs one extra space since one space is wasted to differentiate
+ * empty and full queues.
*/
ctx->queueSize = queueSize + 1;
ctx->queue = (POOL_job*)ZSTD_malloc(ctx->queueSize * sizeof(POOL_job), customMem);
@@ -126,7 +133,7 @@
ctx->shutdown = 0;
/* Allocate space for the thread handles */
ctx->threads = (ZSTD_pthread_t*)ZSTD_malloc(numThreads * sizeof(ZSTD_pthread_t), customMem);
- ctx->numThreads = 0;
+ ctx->threadCapacity = 0;
ctx->customMem = customMem;
/* Check for errors */
if (!ctx->threads || !ctx->queue) { POOL_free(ctx); return NULL; }
@@ -134,11 +141,12 @@
{ size_t i;
for (i = 0; i < numThreads; ++i) {
if (ZSTD_pthread_create(&ctx->threads[i], NULL, &POOL_thread, ctx)) {
- ctx->numThreads = i;
+ ctx->threadCapacity = i;
POOL_free(ctx);
return NULL;
} }
- ctx->numThreads = numThreads;
+ ctx->threadCapacity = numThreads;
+ ctx->threadLimit = numThreads;
}
return ctx;
}
@@ -156,8 +164,8 @@
ZSTD_pthread_cond_broadcast(&ctx->queuePopCond);
/* Join all of the threads */
{ size_t i;
- for (i = 0; i < ctx->numThreads; ++i) {
- ZSTD_pthread_join(ctx->threads[i], NULL);
+ for (i = 0; i < ctx->threadCapacity; ++i) {
+ ZSTD_pthread_join(ctx->threads[i], NULL); /* note : could fail */
} }
}
@@ -172,24 +180,68 @@
ZSTD_free(ctx, ctx->customMem);
}
+
+
size_t POOL_sizeof(POOL_ctx *ctx) {
if (ctx==NULL) return 0; /* supports sizeof NULL */
return sizeof(*ctx)
+ ctx->queueSize * sizeof(POOL_job)
- + ctx->numThreads * sizeof(ZSTD_pthread_t);
+ + ctx->threadCapacity * sizeof(ZSTD_pthread_t);
+}
+
+
+/* @return : 0 on success, 1 on error */
+static int POOL_resize_internal(POOL_ctx* ctx, size_t numThreads)
+{
+ if (numThreads <= ctx->threadCapacity) {
+ if (!numThreads) return 1;
+ ctx->threadLimit = numThreads;
+ return 0;
+ }
+ /* numThreads > threadCapacity */
+ { ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_malloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem);
+ if (!threadPool) return 1;
+ /* replace existing thread pool */
+ memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(*threadPool));
+ ZSTD_free(ctx->threads, ctx->customMem);
+ ctx->threads = threadPool;
+ /* Initialize additional threads */
+ { size_t threadId;
+ for (threadId = ctx->threadCapacity; threadId < numThreads; ++threadId) {
+ if (ZSTD_pthread_create(&threadPool[threadId], NULL, &POOL_thread, ctx)) {
+ ctx->threadCapacity = threadId;
+ return 1;
+ } }
+ } }
+ /* successfully expanded */
+ ctx->threadCapacity = numThreads;
+ ctx->threadLimit = numThreads;
+ return 0;
+}
+
+/* @return : 0 on success, 1 on error */
+int POOL_resize(POOL_ctx* ctx, size_t numThreads)
+{
+ int result;
+ if (ctx==NULL) return 1;
+ ZSTD_pthread_mutex_lock(&ctx->queueMutex);
+ result = POOL_resize_internal(ctx, numThreads);
+ ZSTD_pthread_cond_broadcast(&ctx->queuePopCond);
+ ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+ return result;
}
/**
* Returns 1 if the queue is full and 0 otherwise.
*
- * If the queueSize is 1 (the pool was created with an intended queueSize of 0),
- * then a queue is empty if there is a thread free and no job is waiting.
+ * When queueSize is 1 (pool was created with an intended queueSize of 0),
+ * then a queue is empty if there is a thread free _and_ no job is waiting.
*/
static int isQueueFull(POOL_ctx const* ctx) {
if (ctx->queueSize > 1) {
return ctx->queueHead == ((ctx->queueTail + 1) % ctx->queueSize);
} else {
- return ctx->numThreadsBusy == ctx->numThreads ||
+ return (ctx->numThreadsBusy == ctx->threadLimit) ||
!ctx->queueEmpty;
}
}
@@ -263,6 +315,11 @@
(void)ctx;
}
+int POOL_resize(POOL_ctx* ctx, size_t numThreads) {
+ (void)ctx; (void)numThreads;
+ return 0;
+}
+
void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque) {
(void)ctx;
function(opaque);
diff --git a/vendor/github.com/DataDog/zstd/pool.h b/vendor/github.com/DataDog/zstd/pool.h
index a57e9b4..458d37f 100644
--- a/vendor/github.com/DataDog/zstd/pool.h
+++ b/vendor/github.com/DataDog/zstd/pool.h
@@ -30,40 +30,50 @@
*/
POOL_ctx* POOL_create(size_t numThreads, size_t queueSize);
-POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customMem);
+POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
+ ZSTD_customMem customMem);
/*! POOL_free() :
- Free a thread pool returned by POOL_create().
-*/
+ * Free a thread pool returned by POOL_create().
+ */
void POOL_free(POOL_ctx* ctx);
+/*! POOL_resize() :
+ * Expands or shrinks pool's number of threads.
+ * This is more efficient than releasing + creating a new context,
+ * since it tries to preserve and re-use existing threads.
+ * `numThreads` must be at least 1.
+ * @return : 0 when resize was successful,
+ * !0 (typically 1) if there is an error.
+ * note : only numThreads can be resized, queueSize remains unchanged.
+ */
+int POOL_resize(POOL_ctx* ctx, size_t numThreads);
+
/*! POOL_sizeof() :
- return memory usage of pool returned by POOL_create().
-*/
+ * @return threadpool memory usage
+ * note : compatible with NULL (returns 0 in this case)
+ */
size_t POOL_sizeof(POOL_ctx* ctx);
/*! POOL_function :
- The function type that can be added to a thread pool.
-*/
+ * The function type that can be added to a thread pool.
+ */
typedef void (*POOL_function)(void*);
-/*! POOL_add_function :
- The function type for a generic thread pool add function.
-*/
-typedef void (*POOL_add_function)(void*, POOL_function, void*);
/*! POOL_add() :
- Add the job `function(opaque)` to the thread pool. `ctx` must be valid.
- Possibly blocks until there is room in the queue.
- Note : The function may be executed asynchronously, so `opaque` must live until the function has been completed.
-*/
+ * Add the job `function(opaque)` to the thread pool. `ctx` must be valid.
+ * Possibly blocks until there is room in the queue.
+ * Note : The function may be executed asynchronously,
+ * therefore, `opaque` must live until function has been completed.
+ */
void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque);
/*! POOL_tryAdd() :
- Add the job `function(opaque)` to the thread pool if a worker is available.
- return immediately otherwise.
- @return : 1 if successful, 0 if not.
-*/
+ * Add the job `function(opaque)` to thread pool _if_ a worker is available.
+ * Returns immediately even if not (does not block).
+ * @return : 1 if successful, 0 if not.
+ */
int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque);
diff --git a/vendor/github.com/DataDog/zstd/threading.c b/vendor/github.com/DataDog/zstd/threading.c
index 8be8c8d..f3d4fa8 100644
--- a/vendor/github.com/DataDog/zstd/threading.c
+++ b/vendor/github.com/DataDog/zstd/threading.c
@@ -14,8 +14,8 @@
* This file will hold wrapper for systems, which do not support pthreads
*/
-/* create fake symbol to avoid empty trnaslation unit warning */
-int g_ZSTD_threading_useles_symbol;
+/* create fake symbol to avoid empty translation unit warning */
+int g_ZSTD_threading_useless_symbol;
#if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
diff --git a/vendor/github.com/DataDog/zstd/xxhash.c b/vendor/github.com/DataDog/zstd/xxhash.c
index 9d9c0e9..30599aa 100644
--- a/vendor/github.com/DataDog/zstd/xxhash.c
+++ b/vendor/github.com/DataDog/zstd/xxhash.c
@@ -66,10 +66,10 @@
/* #define XXH_ACCEPT_NULL_INPUT_POINTER 1 */
/*!XXH_FORCE_NATIVE_FORMAT :
- * By default, xxHash library provides endian-independant Hash values, based on little-endian convention.
+ * By default, xxHash library provides endian-independent Hash values, based on little-endian convention.
* Results are therefore identical for little-endian and big-endian CPU.
* This comes at a performance cost for big-endian CPU, since some swapping is required to emulate little-endian format.
- * Should endian-independance be of no importance for your application, you may set the #define below to 1,
+ * Should endian-independence be of no importance for your application, you may set the #define below to 1,
* to improve speed for Big-endian CPU.
* This option has no impact on Little_Endian CPU.
*/
@@ -98,6 +98,7 @@
/* Modify the local functions below should you wish to use some other memory routines */
/* for malloc(), free() */
#include <stdlib.h>
+#include <stddef.h> /* size_t */
static void* XXH_malloc(size_t s) { return malloc(s); }
static void XXH_free (void* p) { free(p); }
/* for memcpy() */
diff --git a/vendor/github.com/DataDog/zstd/zdict.c b/vendor/github.com/DataDog/zstd/zdict.c
index 7d24e49..ee21ee1 100644
--- a/vendor/github.com/DataDog/zstd/zdict.c
+++ b/vendor/github.com/DataDog/zstd/zdict.c
@@ -255,15 +255,15 @@
}
{ int i;
- U32 searchLength;
+ U32 mml;
U32 refinedStart = start;
U32 refinedEnd = end;
DISPLAYLEVEL(4, "\n");
- DISPLAYLEVEL(4, "found %3u matches of length >= %i at pos %7u ", (U32)(end-start), MINMATCHLENGTH, (U32)pos);
+ DISPLAYLEVEL(4, "found %3u matches of length >= %i at pos %7u ", (unsigned)(end-start), MINMATCHLENGTH, (unsigned)pos);
DISPLAYLEVEL(4, "\n");
- for (searchLength = MINMATCHLENGTH ; ; searchLength++) {
+ for (mml = MINMATCHLENGTH ; ; mml++) {
BYTE currentChar = 0;
U32 currentCount = 0;
U32 currentID = refinedStart;
@@ -271,13 +271,13 @@
U32 selectedCount = 0;
U32 selectedID = currentID;
for (id =refinedStart; id < refinedEnd; id++) {
- if (b[suffix[id] + searchLength] != currentChar) {
+ if (b[suffix[id] + mml] != currentChar) {
if (currentCount > selectedCount) {
selectedCount = currentCount;
selectedID = currentID;
}
currentID = id;
- currentChar = b[ suffix[id] + searchLength];
+ currentChar = b[ suffix[id] + mml];
currentCount = 0;
}
currentCount ++;
@@ -293,7 +293,7 @@
refinedEnd = refinedStart + selectedCount;
}
- /* evaluate gain based on new ref */
+ /* evaluate gain based on new dict */
start = refinedStart;
pos = suffix[refinedStart];
end = start;
@@ -341,8 +341,8 @@
for (i=MINMATCHLENGTH; i<=(int)maxLength; i++)
savings[i] = savings[i-1] + (lengthList[i] * (i-3));
- DISPLAYLEVEL(4, "Selected ref at position %u, of length %u : saves %u (ratio: %.2f) \n",
- (U32)pos, (U32)maxLength, savings[maxLength], (double)savings[maxLength] / maxLength);
+ DISPLAYLEVEL(4, "Selected dict at position %u, of length %u : saves %u (ratio: %.2f) \n",
+ (unsigned)pos, (unsigned)maxLength, (unsigned)savings[maxLength], (double)savings[maxLength] / maxLength);
solution.pos = (U32)pos;
solution.length = (U32)maxLength;
@@ -497,7 +497,7 @@
static size_t ZDICT_trainBuffer_legacy(dictItem* dictList, U32 dictListSize,
const void* const buffer, size_t bufferSize, /* buffer must end with noisy guard band */
const size_t* fileSizes, unsigned nbFiles,
- U32 minRatio, U32 notificationLevel)
+ unsigned minRatio, U32 notificationLevel)
{
int* const suffix0 = (int*)malloc((bufferSize+2)*sizeof(*suffix0));
int* const suffix = suffix0+1;
@@ -523,11 +523,11 @@
memset(doneMarks, 0, bufferSize+16);
/* limit sample set size (divsufsort limitation)*/
- if (bufferSize > ZDICT_MAX_SAMPLES_SIZE) DISPLAYLEVEL(3, "sample set too large : reduced to %u MB ...\n", (U32)(ZDICT_MAX_SAMPLES_SIZE>>20));
+ if (bufferSize > ZDICT_MAX_SAMPLES_SIZE) DISPLAYLEVEL(3, "sample set too large : reduced to %u MB ...\n", (unsigned)(ZDICT_MAX_SAMPLES_SIZE>>20));
while (bufferSize > ZDICT_MAX_SAMPLES_SIZE) bufferSize -= fileSizes[--nbFiles];
/* sort */
- DISPLAYLEVEL(2, "sorting %u files of total size %u MB ...\n", nbFiles, (U32)(bufferSize>>20));
+ DISPLAYLEVEL(2, "sorting %u files of total size %u MB ...\n", nbFiles, (unsigned)(bufferSize>>20));
{ int const divSuftSortResult = divsufsort((const unsigned char*)buffer, suffix, (int)bufferSize, 0);
if (divSuftSortResult != 0) { result = ERROR(GENERIC); goto _cleanup; }
}
@@ -581,7 +581,7 @@
typedef struct
{
- ZSTD_CCtx* ref; /* contains reference to dictionary */
+ ZSTD_CDict* dict; /* dictionary */
ZSTD_CCtx* zc; /* working context */
void* workPlace; /* must be ZSTD_BLOCKSIZE_MAX allocated */
} EStats_ress_t;
@@ -589,7 +589,7 @@
#define MAXREPOFFSET 1024
static void ZDICT_countEStats(EStats_ress_t esr, ZSTD_parameters params,
- U32* countLit, U32* offsetcodeCount, U32* matchlengthCount, U32* litlengthCount, U32* repOffsets,
+ unsigned* countLit, unsigned* offsetcodeCount, unsigned* matchlengthCount, unsigned* litlengthCount, U32* repOffsets,
const void* src, size_t srcSize,
U32 notificationLevel)
{
@@ -597,11 +597,12 @@
size_t cSize;
if (srcSize > blockSizeMax) srcSize = blockSizeMax; /* protection vs large samples */
- { size_t const errorCode = ZSTD_copyCCtx(esr.zc, esr.ref, 0);
- if (ZSTD_isError(errorCode)) { DISPLAYLEVEL(1, "warning : ZSTD_copyCCtx failed \n"); return; }
+ { size_t const errorCode = ZSTD_compressBegin_usingCDict(esr.zc, esr.dict);
+ if (ZSTD_isError(errorCode)) { DISPLAYLEVEL(1, "warning : ZSTD_compressBegin_usingCDict failed \n"); return; }
+
}
cSize = ZSTD_compressBlock(esr.zc, esr.workPlace, ZSTD_BLOCKSIZE_MAX, src, srcSize);
- if (ZSTD_isError(cSize)) { DISPLAYLEVEL(3, "warning : could not compress sample size %u \n", (U32)srcSize); return; }
+ if (ZSTD_isError(cSize)) { DISPLAYLEVEL(3, "warning : could not compress sample size %u \n", (unsigned)srcSize); return; }
if (cSize) { /* if == 0; block is not compressible */
const seqStore_t* const seqStorePtr = ZSTD_getSeqStore(esr.zc);
@@ -670,7 +671,7 @@
* rewrite `countLit` to contain a mostly flat but still compressible distribution of literals.
* necessary to avoid generating a non-compressible distribution that HUF_writeCTable() cannot encode.
*/
-static void ZDICT_flatLit(U32* countLit)
+static void ZDICT_flatLit(unsigned* countLit)
{
int u;
for (u=1; u<256; u++) countLit[u] = 2;
@@ -686,18 +687,18 @@
const void* dictBuffer, size_t dictBufferSize,
unsigned notificationLevel)
{
- U32 countLit[256];
+ unsigned countLit[256];
HUF_CREATE_STATIC_CTABLE(hufTable, 255);
- U32 offcodeCount[OFFCODE_MAX+1];
+ unsigned offcodeCount[OFFCODE_MAX+1];
short offcodeNCount[OFFCODE_MAX+1];
U32 offcodeMax = ZSTD_highbit32((U32)(dictBufferSize + 128 KB));
- U32 matchLengthCount[MaxML+1];
+ unsigned matchLengthCount[MaxML+1];
short matchLengthNCount[MaxML+1];
- U32 litLengthCount[MaxLL+1];
+ unsigned litLengthCount[MaxLL+1];
short litLengthNCount[MaxLL+1];
U32 repOffset[MAXREPOFFSET];
offsetCount_t bestRepOffset[ZSTD_REP_NUM+1];
- EStats_ress_t esr;
+ EStats_ress_t esr = { NULL, NULL, NULL };
ZSTD_parameters params;
U32 u, huffLog = 11, Offlog = OffFSELog, mlLog = MLFSELog, llLog = LLFSELog, total;
size_t pos = 0, errorCode;
@@ -708,14 +709,6 @@
/* init */
DEBUGLOG(4, "ZDICT_analyzeEntropy");
- esr.ref = ZSTD_createCCtx();
- esr.zc = ZSTD_createCCtx();
- esr.workPlace = malloc(ZSTD_BLOCKSIZE_MAX);
- if (!esr.ref || !esr.zc || !esr.workPlace) {
- eSize = ERROR(memory_allocation);
- DISPLAYLEVEL(1, "Not enough memory \n");
- goto _cleanup;
- }
if (offcodeMax>OFFCODE_MAX) { eSize = ERROR(dictionaryCreation_failed); goto _cleanup; } /* too large dictionary */
for (u=0; u<256; u++) countLit[u] = 1; /* any character must be described */
for (u=0; u<=offcodeMax; u++) offcodeCount[u] = 1;
@@ -724,14 +717,17 @@
memset(repOffset, 0, sizeof(repOffset));
repOffset[1] = repOffset[4] = repOffset[8] = 1;
memset(bestRepOffset, 0, sizeof(bestRepOffset));
- if (compressionLevel<=0) compressionLevel = g_compressionLevel_default;
+ if (compressionLevel==0) compressionLevel = g_compressionLevel_default;
params = ZSTD_getParams(compressionLevel, averageSampleSize, dictBufferSize);
- { size_t const beginResult = ZSTD_compressBegin_advanced(esr.ref, dictBuffer, dictBufferSize, params, 0);
- if (ZSTD_isError(beginResult)) {
- DISPLAYLEVEL(1, "error : ZSTD_compressBegin_advanced() failed : %s \n", ZSTD_getErrorName(beginResult));
- eSize = ERROR(GENERIC);
- goto _cleanup;
- } }
+
+ esr.dict = ZSTD_createCDict_advanced(dictBuffer, dictBufferSize, ZSTD_dlm_byRef, ZSTD_dct_rawContent, params.cParams, ZSTD_defaultCMem);
+ esr.zc = ZSTD_createCCtx();
+ esr.workPlace = malloc(ZSTD_BLOCKSIZE_MAX);
+ if (!esr.dict || !esr.zc || !esr.workPlace) {
+ eSize = ERROR(memory_allocation);
+ DISPLAYLEVEL(1, "Not enough memory \n");
+ goto _cleanup;
+ }
/* collect stats on all samples */
for (u=0; u<nbFiles; u++) {
@@ -745,7 +741,7 @@
/* analyze, build stats, starting with literals */
{ size_t maxNbBits = HUF_buildCTable (hufTable, countLit, 255, huffLog);
if (HUF_isError(maxNbBits)) {
- eSize = ERROR(GENERIC);
+ eSize = maxNbBits;
DISPLAYLEVEL(1, " HUF_buildCTable error \n");
goto _cleanup;
}
@@ -768,7 +764,7 @@
total=0; for (u=0; u<=offcodeMax; u++) total+=offcodeCount[u];
errorCode = FSE_normalizeCount(offcodeNCount, Offlog, offcodeCount, total, offcodeMax);
if (FSE_isError(errorCode)) {
- eSize = ERROR(GENERIC);
+ eSize = errorCode;
DISPLAYLEVEL(1, "FSE_normalizeCount error with offcodeCount \n");
goto _cleanup;
}
@@ -777,7 +773,7 @@
total=0; for (u=0; u<=MaxML; u++) total+=matchLengthCount[u];
errorCode = FSE_normalizeCount(matchLengthNCount, mlLog, matchLengthCount, total, MaxML);
if (FSE_isError(errorCode)) {
- eSize = ERROR(GENERIC);
+ eSize = errorCode;
DISPLAYLEVEL(1, "FSE_normalizeCount error with matchLengthCount \n");
goto _cleanup;
}
@@ -786,7 +782,7 @@
total=0; for (u=0; u<=MaxLL; u++) total+=litLengthCount[u];
errorCode = FSE_normalizeCount(litLengthNCount, llLog, litLengthCount, total, MaxLL);
if (FSE_isError(errorCode)) {
- eSize = ERROR(GENERIC);
+ eSize = errorCode;
DISPLAYLEVEL(1, "FSE_normalizeCount error with litLengthCount \n");
goto _cleanup;
}
@@ -795,7 +791,7 @@
/* write result to buffer */
{ size_t const hhSize = HUF_writeCTable(dstPtr, maxDstSize, hufTable, 255, huffLog);
if (HUF_isError(hhSize)) {
- eSize = ERROR(GENERIC);
+ eSize = hhSize;
DISPLAYLEVEL(1, "HUF_writeCTable error \n");
goto _cleanup;
}
@@ -806,7 +802,7 @@
{ size_t const ohSize = FSE_writeNCount(dstPtr, maxDstSize, offcodeNCount, OFFCODE_MAX, Offlog);
if (FSE_isError(ohSize)) {
- eSize = ERROR(GENERIC);
+ eSize = ohSize;
DISPLAYLEVEL(1, "FSE_writeNCount error with offcodeNCount \n");
goto _cleanup;
}
@@ -817,7 +813,7 @@
{ size_t const mhSize = FSE_writeNCount(dstPtr, maxDstSize, matchLengthNCount, MaxML, mlLog);
if (FSE_isError(mhSize)) {
- eSize = ERROR(GENERIC);
+ eSize = mhSize;
DISPLAYLEVEL(1, "FSE_writeNCount error with matchLengthNCount \n");
goto _cleanup;
}
@@ -828,7 +824,7 @@
{ size_t const lhSize = FSE_writeNCount(dstPtr, maxDstSize, litLengthNCount, MaxLL, llLog);
if (FSE_isError(lhSize)) {
- eSize = ERROR(GENERIC);
+ eSize = lhSize;
DISPLAYLEVEL(1, "FSE_writeNCount error with litlengthNCount \n");
goto _cleanup;
}
@@ -838,7 +834,7 @@
}
if (maxDstSize<12) {
- eSize = ERROR(GENERIC);
+ eSize = ERROR(dstSize_tooSmall);
DISPLAYLEVEL(1, "not enough space to write RepOffsets \n");
goto _cleanup;
}
@@ -856,7 +852,7 @@
eSize += 12;
_cleanup:
- ZSTD_freeCCtx(esr.ref);
+ ZSTD_freeCDict(esr.dict);
ZSTD_freeCCtx(esr.zc);
free(esr.workPlace);
@@ -867,13 +863,13 @@
size_t ZDICT_finalizeDictionary(void* dictBuffer, size_t dictBufferCapacity,
const void* customDictContent, size_t dictContentSize,
- const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
- ZDICT_params_t params)
+ const void* samplesBuffer, const size_t* samplesSizes,
+ unsigned nbSamples, ZDICT_params_t params)
{
size_t hSize;
#define HBUFFSIZE 256 /* should prove large enough for all entropy headers */
BYTE header[HBUFFSIZE];
- int const compressionLevel = (params.compressionLevel <= 0) ? g_compressionLevel_default : params.compressionLevel;
+ int const compressionLevel = (params.compressionLevel == 0) ? g_compressionLevel_default : params.compressionLevel;
U32 const notificationLevel = params.notificationLevel;
/* check conditions */
@@ -914,11 +910,12 @@
}
-size_t ZDICT_addEntropyTablesFromBuffer_advanced(void* dictBuffer, size_t dictContentSize, size_t dictBufferCapacity,
- const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
- ZDICT_params_t params)
+static size_t ZDICT_addEntropyTablesFromBuffer_advanced(
+ void* dictBuffer, size_t dictContentSize, size_t dictBufferCapacity,
+ const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
+ ZDICT_params_t params)
{
- int const compressionLevel = (params.compressionLevel <= 0) ? g_compressionLevel_default : params.compressionLevel;
+ int const compressionLevel = (params.compressionLevel == 0) ? g_compressionLevel_default : params.compressionLevel;
U32 const notificationLevel = params.notificationLevel;
size_t hSize = 8;
@@ -947,7 +944,11 @@
return MIN(dictBufferCapacity, hSize+dictContentSize);
}
-
+/* Hidden declaration for dbio.c */
+size_t ZDICT_trainFromBuffer_unsafe_legacy(
+ void* dictBuffer, size_t maxDictSize,
+ const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
+ ZDICT_legacy_params_t params);
/*! ZDICT_trainFromBuffer_unsafe_legacy() :
* Warning : `samplesBuffer` must be followed by noisy guard band.
* @return : size of dictionary, or an error code which can be tested with ZDICT_isError()
@@ -982,31 +983,33 @@
/* display best matches */
if (params.zParams.notificationLevel>= 3) {
- U32 const nb = MIN(25, dictList[0].pos);
- U32 const dictContentSize = ZDICT_dictSize(dictList);
- U32 u;
- DISPLAYLEVEL(3, "\n %u segments found, of total size %u \n", dictList[0].pos-1, dictContentSize);
+ unsigned const nb = MIN(25, dictList[0].pos);
+ unsigned const dictContentSize = ZDICT_dictSize(dictList);
+ unsigned u;
+ DISPLAYLEVEL(3, "\n %u segments found, of total size %u \n", (unsigned)dictList[0].pos-1, dictContentSize);
DISPLAYLEVEL(3, "list %u best segments \n", nb-1);
for (u=1; u<nb; u++) {
- U32 const pos = dictList[u].pos;
- U32 const length = dictList[u].length;
+ unsigned const pos = dictList[u].pos;
+ unsigned const length = dictList[u].length;
U32 const printedLength = MIN(40, length);
- if ((pos > samplesBuffSize) || ((pos + length) > samplesBuffSize))
+ if ((pos > samplesBuffSize) || ((pos + length) > samplesBuffSize)) {
+ free(dictList);
return ERROR(GENERIC); /* should never happen */
+ }
DISPLAYLEVEL(3, "%3u:%3u bytes at pos %8u, savings %7u bytes |",
- u, length, pos, dictList[u].savings);
+ u, length, pos, (unsigned)dictList[u].savings);
ZDICT_printHex((const char*)samplesBuffer+pos, printedLength);
DISPLAYLEVEL(3, "| \n");
} }
/* create dictionary */
- { U32 dictContentSize = ZDICT_dictSize(dictList);
+ { unsigned dictContentSize = ZDICT_dictSize(dictList);
if (dictContentSize < ZDICT_CONTENTSIZE_MIN) { free(dictList); return ERROR(dictionaryCreation_failed); } /* dictionary content too small */
if (dictContentSize < targetDictSize/4) {
- DISPLAYLEVEL(2, "! warning : selected content significantly smaller than requested (%u < %u) \n", dictContentSize, (U32)maxDictSize);
+ DISPLAYLEVEL(2, "! warning : selected content significantly smaller than requested (%u < %u) \n", dictContentSize, (unsigned)maxDictSize);
if (samplesBuffSize < 10 * targetDictSize)
- DISPLAYLEVEL(2, "! consider increasing the number of samples (total size : %u MB)\n", (U32)(samplesBuffSize>>20));
+ DISPLAYLEVEL(2, "! consider increasing the number of samples (total size : %u MB)\n", (unsigned)(samplesBuffSize>>20));
if (minRep > MINRATIO) {
DISPLAYLEVEL(2, "! consider increasing selectivity to produce larger dictionary (-s%u) \n", selectivity+1);
DISPLAYLEVEL(2, "! note : larger dictionaries are not necessarily better, test its efficiency on samples \n");
@@ -1014,9 +1017,9 @@
}
if ((dictContentSize > targetDictSize*3) && (nbSamples > 2*MINRATIO) && (selectivity>1)) {
- U32 proposedSelectivity = selectivity-1;
+ unsigned proposedSelectivity = selectivity-1;
while ((nbSamples >> proposedSelectivity) <= MINRATIO) { proposedSelectivity--; }
- DISPLAYLEVEL(2, "! note : calculated dictionary significantly larger than requested (%u > %u) \n", dictContentSize, (U32)maxDictSize);
+ DISPLAYLEVEL(2, "! note : calculated dictionary significantly larger than requested (%u > %u) \n", dictContentSize, (unsigned)maxDictSize);
DISPLAYLEVEL(2, "! consider increasing dictionary size, or produce denser dictionary (-s%u) \n", proposedSelectivity);
DISPLAYLEVEL(2, "! always test dictionary efficiency on real samples \n");
}
@@ -1082,17 +1085,17 @@
size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCapacity,
const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples)
{
- ZDICT_cover_params_t params;
+ ZDICT_fastCover_params_t params;
DEBUGLOG(3, "ZDICT_trainFromBuffer");
memset(¶ms, 0, sizeof(params));
params.d = 8;
params.steps = 4;
/* Default to level 6 since no compression level information is available */
- params.zParams.compressionLevel = 6;
-#if defined(ZSTD_DEBUG) && (ZSTD_DEBUG>=1)
- params.zParams.notificationLevel = ZSTD_DEBUG;
+ params.zParams.compressionLevel = 3;
+#if defined(DEBUGLEVEL) && (DEBUGLEVEL>=1)
+ params.zParams.notificationLevel = DEBUGLEVEL;
#endif
- return ZDICT_optimizeTrainFromBuffer_cover(dictBuffer, dictBufferCapacity,
+ return ZDICT_optimizeTrainFromBuffer_fastCover(dictBuffer, dictBufferCapacity,
samplesBuffer, samplesSizes, nbSamples,
¶ms);
}
diff --git a/vendor/github.com/DataDog/zstd/zdict.h b/vendor/github.com/DataDog/zstd/zdict.h
index ad459c2..37978ec 100644
--- a/vendor/github.com/DataDog/zstd/zdict.h
+++ b/vendor/github.com/DataDog/zstd/zdict.h
@@ -39,20 +39,27 @@
/*! ZDICT_trainFromBuffer():
* Train a dictionary from an array of samples.
- * Redirect towards ZDICT_optimizeTrainFromBuffer_cover() single-threaded, with d=8 and steps=4.
+ * Redirect towards ZDICT_optimizeTrainFromBuffer_fastCover() single-threaded, with d=8, steps=4,
+ * f=20, and accel=1.
* Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
* supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
* The resulting dictionary will be saved into `dictBuffer`.
* @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
* or an error code, which can be tested with ZDICT_isError().
- * Note: ZDICT_trainFromBuffer() requires about 9 bytes of memory for each input byte.
+ * Note: Dictionary training will fail if there are not enough samples to construct a
+ * dictionary, or if most of the samples are too small (< 8 bytes being the lower limit).
+ * If dictionary training fails, you should use zstd without a dictionary, as the dictionary
+ * would've been ineffective anyways. If you believe your samples would benefit from a dictionary
+ * please open an issue with details, and we can look into it.
+ * Note: ZDICT_trainFromBuffer()'s memory usage is about 6 MB.
* Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
* It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
* In general, it's recommended to provide a few thousands samples, though this can vary a lot.
* It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
*/
ZDICTLIB_API size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCapacity,
- const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples);
+ const void* samplesBuffer,
+ const size_t* samplesSizes, unsigned nbSamples);
/*====== Helper functions ======*/
@@ -84,11 +91,27 @@
typedef struct {
unsigned k; /* Segment size : constraint: 0 < k : Reasonable range [16, 2048+] */
unsigned d; /* dmer size : constraint: 0 < d <= k : Reasonable range [6, 16] */
- unsigned steps; /* Number of steps : Only used for optimization : 0 means default (32) : Higher means more parameters checked */
+ unsigned steps; /* Number of steps : Only used for optimization : 0 means default (40) : Higher means more parameters checked */
unsigned nbThreads; /* Number of threads : constraint: 0 < nbThreads : 1 means single-threaded : Only used for optimization : Ignored if ZSTD_MULTITHREAD is not defined */
+ double splitPoint; /* Percentage of samples used for training: Only used for optimization : the first nbSamples * splitPoint samples will be used to training, the last nbSamples * (1 - splitPoint) samples will be used for testing, 0 means default (1.0), 1.0 when all samples are used for both training and testing */
+ unsigned shrinkDict; /* Train dictionaries to shrink in size starting from the minimum size and selects the smallest dictionary that is shrinkDictMaxRegression% worse than the largest dictionary. 0 means no shrinking and 1 means shrinking */
+ unsigned shrinkDictMaxRegression; /* Sets shrinkDictMaxRegression so that a smaller dictionary can be at worse shrinkDictMaxRegression% worse than the max dict size dictionary. */
ZDICT_params_t zParams;
} ZDICT_cover_params_t;
+typedef struct {
+ unsigned k; /* Segment size : constraint: 0 < k : Reasonable range [16, 2048+] */
+ unsigned d; /* dmer size : constraint: 0 < d <= k : Reasonable range [6, 16] */
+ unsigned f; /* log of size of frequency array : constraint: 0 < f <= 31 : 1 means default(20)*/
+ unsigned steps; /* Number of steps : Only used for optimization : 0 means default (40) : Higher means more parameters checked */
+ unsigned nbThreads; /* Number of threads : constraint: 0 < nbThreads : 1 means single-threaded : Only used for optimization : Ignored if ZSTD_MULTITHREAD is not defined */
+ double splitPoint; /* Percentage of samples used for training: Only used for optimization : the first nbSamples * splitPoint samples will be used to training, the last nbSamples * (1 - splitPoint) samples will be used for testing, 0 means default (0.75), 1.0 when all samples are used for both training and testing */
+ unsigned accel; /* Acceleration level: constraint: 0 < accel <= 10, higher means faster and less accurate, 0 means default(1) */
+ unsigned shrinkDict; /* Train dictionaries to shrink in size starting from the minimum size and selects the smallest dictionary that is shrinkDictMaxRegression% worse than the largest dictionary. 0 means no shrinking and 1 means shrinking */
+ unsigned shrinkDictMaxRegression; /* Sets shrinkDictMaxRegression so that a smaller dictionary can be at worse shrinkDictMaxRegression% worse than the max dict size dictionary. */
+
+ ZDICT_params_t zParams;
+} ZDICT_fastCover_params_t;
/*! ZDICT_trainFromBuffer_cover():
* Train a dictionary from an array of samples using the COVER algorithm.
@@ -97,6 +120,7 @@
* The resulting dictionary will be saved into `dictBuffer`.
* @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
* or an error code, which can be tested with ZDICT_isError().
+ * See ZDICT_trainFromBuffer() for details on failure modes.
* Note: ZDICT_trainFromBuffer_cover() requires about 9 bytes of memory for each input byte.
* Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
* It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
@@ -115,13 +139,14 @@
* dictionary constructed with those parameters is stored in `dictBuffer`.
*
* All of the parameters d, k, steps are optional.
- * If d is non-zero then we don't check multiple values of d, otherwise we check d = {6, 8, 10, 12, 14, 16}.
+ * If d is non-zero then we don't check multiple values of d, otherwise we check d = {6, 8}.
* if steps is zero it defaults to its default value.
- * If k is non-zero then we don't check multiple values of k, otherwise we check steps values in [16, 2048].
+ * If k is non-zero then we don't check multiple values of k, otherwise we check steps values in [50, 2000].
*
* @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
- * or an error code, which can be tested with ZDICT_isError().
- * On success `*parameters` contains the parameters selected.
+ * or an error code, which can be tested with ZDICT_isError().
+ * On success `*parameters` contains the parameters selected.
+ * See ZDICT_trainFromBuffer() for details on failure modes.
* Note: ZDICT_optimizeTrainFromBuffer_cover() requires about 8 bytes of memory for each input byte and additionally another 5 bytes of memory for each byte of memory for each thread.
*/
ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_cover(
@@ -129,6 +154,50 @@
const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
ZDICT_cover_params_t* parameters);
+/*! ZDICT_trainFromBuffer_fastCover():
+ * Train a dictionary from an array of samples using a modified version of COVER algorithm.
+ * Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
+ * supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
+ * d and k are required.
+ * All other parameters are optional, will use default values if not provided
+ * The resulting dictionary will be saved into `dictBuffer`.
+ * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
+ * or an error code, which can be tested with ZDICT_isError().
+ * See ZDICT_trainFromBuffer() for details on failure modes.
+ * Note: ZDICT_trainFromBuffer_fastCover() requires 6 * 2^f bytes of memory.
+ * Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
+ * It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
+ * In general, it's recommended to provide a few thousands samples, though this can vary a lot.
+ * It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
+ */
+ZDICTLIB_API size_t ZDICT_trainFromBuffer_fastCover(void *dictBuffer,
+ size_t dictBufferCapacity, const void *samplesBuffer,
+ const size_t *samplesSizes, unsigned nbSamples,
+ ZDICT_fastCover_params_t parameters);
+
+/*! ZDICT_optimizeTrainFromBuffer_fastCover():
+ * The same requirements as above hold for all the parameters except `parameters`.
+ * This function tries many parameter combinations (specifically, k and d combinations)
+ * and picks the best parameters. `*parameters` is filled with the best parameters found,
+ * dictionary constructed with those parameters is stored in `dictBuffer`.
+ * All of the parameters d, k, steps, f, and accel are optional.
+ * If d is non-zero then we don't check multiple values of d, otherwise we check d = {6, 8}.
+ * if steps is zero it defaults to its default value.
+ * If k is non-zero then we don't check multiple values of k, otherwise we check steps values in [50, 2000].
+ * If f is zero, default value of 20 is used.
+ * If accel is zero, default value of 1 is used.
+ *
+ * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
+ * or an error code, which can be tested with ZDICT_isError().
+ * On success `*parameters` contains the parameters selected.
+ * See ZDICT_trainFromBuffer() for details on failure modes.
+ * Note: ZDICT_optimizeTrainFromBuffer_fastCover() requires about 6 * 2^f bytes of memory for each thread.
+ */
+ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_fastCover(void* dictBuffer,
+ size_t dictBufferCapacity, const void* samplesBuffer,
+ const size_t* samplesSizes, unsigned nbSamples,
+ ZDICT_fastCover_params_t* parameters);
+
/*! ZDICT_finalizeDictionary():
* Given a custom content as a basis for dictionary, and a set of samples,
* finalize dictionary by adding headers and statistics.
@@ -140,7 +209,7 @@
* maxDictSize must be >= dictContentSize, and must be >= ZDICT_DICTSIZE_MIN bytes.
*
* @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`),
- * or an error code, which can be tested by ZDICT_isError().
+ * or an error code, which can be tested by ZDICT_isError().
* Note: ZDICT_finalizeDictionary() will push notifications into stderr if instructed to, using notificationLevel>0.
* Note 2: dictBuffer and dictContent can overlap
*/
@@ -164,6 +233,7 @@
* `parameters` is optional and can be provided with values set to 0 to mean "default".
* @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
* or an error code, which can be tested with ZDICT_isError().
+ * See ZDICT_trainFromBuffer() for details on failure modes.
* Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
* It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
* In general, it's recommended to provide a few thousands samples, though this can vary a lot.
diff --git a/vendor/github.com/DataDog/zstd/zstd.go b/vendor/github.com/DataDog/zstd/zstd.go
index e9953d6..b6af4eb 100644
--- a/vendor/github.com/DataDog/zstd/zstd.go
+++ b/vendor/github.com/DataDog/zstd/zstd.go
@@ -23,6 +23,7 @@
"bytes"
"errors"
"io/ioutil"
+ "runtime"
"unsafe"
)
@@ -84,6 +85,7 @@
C.size_t(len(src)),
C.int(level))
+ runtime.KeepAlive(src)
written := int(cWritten)
// Check if the return is an Error code
if err := getError(written); err != nil {
@@ -107,6 +109,7 @@
C.uintptr_t(uintptr(unsafe.Pointer(&src[0]))),
C.size_t(len(src)))
+ runtime.KeepAlive(src)
written := int(cWritten)
// Check error
if err := getError(written); err != nil {
@@ -115,7 +118,7 @@
return dst[:written], nil
}
- if dst == nil {
+ if len(dst) == 0 {
// Attempt to use zStd to determine decompressed size (may result in error or 0)
size := int(C.size_t(C.ZSTD_getDecompressedSize(unsafe.Pointer(&src[0]), C.size_t(len(src)))))
diff --git a/vendor/github.com/DataDog/zstd/zstd.h b/vendor/github.com/DataDog/zstd/zstd.h
index 6405da6..a1910ee 100644
--- a/vendor/github.com/DataDog/zstd/zstd.h
+++ b/vendor/github.com/DataDog/zstd/zstd.h
@@ -35,40 +35,73 @@
#endif
-/*******************************************************************************************************
+/*******************************************************************************
Introduction
- zstd, short for Zstandard, is a fast lossless compression algorithm,
- targeting real-time compression scenarios at zlib-level and better compression ratios.
- The zstd compression library provides in-memory compression and decompression functions.
- The library supports compression levels from 1 up to ZSTD_maxCLevel() which is currently 22.
- Levels >= 20, labeled `--ultra`, should be used with caution, as they require more memory.
+ zstd, short for Zstandard, is a fast lossless compression algorithm, targeting
+ real-time compression scenarios at zlib-level and better compression ratios.
+ The zstd compression library provides in-memory compression and decompression
+ functions.
+
+ The library supports regular compression levels from 1 up to ZSTD_maxCLevel(),
+ which is currently 22. Levels >= 20, labeled `--ultra`, should be used with
+ caution, as they require more memory. The library also offers negative
+ compression levels, which extend the range of speed vs. ratio preferences.
+ The lower the level, the faster the speed (at the cost of compression).
+
Compression can be done in:
- a single step (described as Simple API)
- a single step, reusing a context (described as Explicit context)
- unbounded multiple steps (described as Streaming compression)
- The compression ratio achievable on small data can be highly improved using a dictionary in:
- - a single step (described as Simple dictionary API)
- - a single step, reusing a dictionary (described as Bulk-processing dictionary API)
- Advanced experimental functions can be accessed using #define ZSTD_STATIC_LINKING_ONLY before including zstd.h.
- Advanced experimental APIs shall never be used with a dynamic library.
- They are not "stable", their definition may change in the future. Only static linking is allowed.
-*********************************************************************************************************/
+ The compression ratio achievable on small data can be highly improved using
+ a dictionary. Dictionary compression can be performed in:
+ - a single step (described as Simple dictionary API)
+ - a single step, reusing a dictionary (described as Bulk-processing
+ dictionary API)
+
+ Advanced experimental functions can be accessed using
+ `#define ZSTD_STATIC_LINKING_ONLY` before including zstd.h.
+
+ Advanced experimental APIs should never be used with a dynamically-linked
+ library. They are not "stable"; their definitions or signatures may change in
+ the future. Only static linking is allowed.
+*******************************************************************************/
/*------ Version ------*/
#define ZSTD_VERSION_MAJOR 1
-#define ZSTD_VERSION_MINOR 3
-#define ZSTD_VERSION_RELEASE 4
+#define ZSTD_VERSION_MINOR 4
+#define ZSTD_VERSION_RELEASE 1
#define ZSTD_VERSION_NUMBER (ZSTD_VERSION_MAJOR *100*100 + ZSTD_VERSION_MINOR *100 + ZSTD_VERSION_RELEASE)
-ZSTDLIB_API unsigned ZSTD_versionNumber(void); /**< useful to check dll version */
+ZSTDLIB_API unsigned ZSTD_versionNumber(void); /**< to check runtime library version */
#define ZSTD_LIB_VERSION ZSTD_VERSION_MAJOR.ZSTD_VERSION_MINOR.ZSTD_VERSION_RELEASE
#define ZSTD_QUOTE(str) #str
#define ZSTD_EXPAND_AND_QUOTE(str) ZSTD_QUOTE(str)
#define ZSTD_VERSION_STRING ZSTD_EXPAND_AND_QUOTE(ZSTD_LIB_VERSION)
-ZSTDLIB_API const char* ZSTD_versionString(void); /* added in v1.3.0 */
+ZSTDLIB_API const char* ZSTD_versionString(void); /* requires v1.3.0+ */
+
+/* *************************************
+ * Default constant
+ ***************************************/
+#ifndef ZSTD_CLEVEL_DEFAULT
+# define ZSTD_CLEVEL_DEFAULT 3
+#endif
+
+/* *************************************
+ * Constants
+ ***************************************/
+
+/* All magic numbers are supposed read/written to/from files/memory using little-endian convention */
+#define ZSTD_MAGICNUMBER 0xFD2FB528 /* valid since v0.8.0 */
+#define ZSTD_MAGIC_DICTIONARY 0xEC30A437 /* valid since v0.7.0 */
+#define ZSTD_MAGIC_SKIPPABLE_START 0x184D2A50 /* all 16 values, from 0x184D2A50 to 0x184D2A5F, signal the beginning of a skippable frame */
+#define ZSTD_MAGIC_SKIPPABLE_MASK 0xFFFFFFF0
+
+#define ZSTD_BLOCKSIZELOG_MAX 17
+#define ZSTD_BLOCKSIZE_MAX (1<<ZSTD_BLOCKSIZELOG_MAX)
+
/***************************************
@@ -92,11 +125,11 @@
ZSTDLIB_API size_t ZSTD_decompress( void* dst, size_t dstCapacity,
const void* src, size_t compressedSize);
-/*! ZSTD_getFrameContentSize() : added in v1.3.0
+/*! ZSTD_getFrameContentSize() : requires v1.3.0+
* `src` should point to the start of a ZSTD encoded frame.
* `srcSize` must be at least as large as the frame header.
* hint : any size >= `ZSTD_frameHeaderSize_max` is large enough.
- * @return : - decompressed size of the frame in `src`, if known
+ * @return : - decompressed size of `src` frame content, if known
* - ZSTD_CONTENTSIZE_UNKNOWN if the size cannot be determined
* - ZSTD_CONTENTSIZE_ERROR if an error occurred (e.g. invalid magic number, srcSize too small)
* note 1 : a 0 return value means the frame is valid but "empty".
@@ -106,7 +139,8 @@
* Optionally, application can rely on some implicit limit,
* as ZSTD_decompress() only needs an upper bound of decompressed size.
* (For example, data could be necessarily cut into blocks <= 16 KB).
- * note 3 : decompressed size is always present when compression is done with ZSTD_compress()
+ * note 3 : decompressed size is always present when compression is completed using single-pass functions,
+ * such as ZSTD_compress(), ZSTD_compressCCtx() ZSTD_compress_usingDict() or ZSTD_compress_usingCDict().
* note 4 : decompressed size can be very large (64-bits value),
* potentially larger than what local system can handle as a single memory segment.
* In which case, it's necessary to use streaming mode to decompress data.
@@ -123,16 +157,24 @@
* Both functions work the same way, but ZSTD_getDecompressedSize() blends
* "empty", "unknown" and "error" results to the same return value (0),
* while ZSTD_getFrameContentSize() gives them separate return values.
- * `src` is the start of a zstd compressed frame.
- * @return : content size to be decompressed, as a 64-bits value _if known and not empty_, 0 otherwise. */
+ * @return : decompressed size of `src` frame content _if known and not empty_, 0 otherwise. */
ZSTDLIB_API unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize);
+/*! ZSTD_findFrameCompressedSize() :
+ * `src` should point to the start of a ZSTD frame or skippable frame.
+ * `srcSize` must be >= first frame size
+ * @return : the compressed size of the first frame starting at `src`,
+ * suitable to pass as `srcSize` to `ZSTD_decompress` or similar,
+ * or an error code if input is invalid */
+ZSTDLIB_API size_t ZSTD_findFrameCompressedSize(const void* src, size_t srcSize);
+
/*====== Helper functions ======*/
#define ZSTD_COMPRESSBOUND(srcSize) ((srcSize) + ((srcSize)>>8) + (((srcSize) < (128<<10)) ? (((128<<10) - (srcSize)) >> 11) /* margin, from 64 to 0 */ : 0)) /* this formula ensures that bound(A) + bound(B) <= bound(A+B) as long as A and B >= 128 KB */
ZSTDLIB_API size_t ZSTD_compressBound(size_t srcSize); /*!< maximum compressed size in worst case single-pass scenario */
ZSTDLIB_API unsigned ZSTD_isError(size_t code); /*!< tells if a `size_t` function result is an error code */
ZSTDLIB_API const char* ZSTD_getErrorName(size_t code); /*!< provides readable string from an error code */
+ZSTDLIB_API int ZSTD_minCLevel(void); /*!< minimum negative compression level allowed */
ZSTDLIB_API int ZSTD_maxCLevel(void); /*!< maximum compression level available */
@@ -141,16 +183,23 @@
***************************************/
/*= Compression context
* When compressing many times,
- * it is recommended to allocate a context just once, and re-use it for each successive compression operation.
+ * it is recommended to allocate a context just once,
+ * and re-use it for each successive compression operation.
* This will make workload friendlier for system's memory.
- * Use one context per thread for parallel execution in multi-threaded environments. */
+ * Note : re-using context is just a speed / resource optimization.
+ * It doesn't change the compression ratio, which remains identical.
+ * Note 2 : In multi-threaded environments,
+ * use one different context per thread for parallel execution.
+ */
typedef struct ZSTD_CCtx_s ZSTD_CCtx;
ZSTDLIB_API ZSTD_CCtx* ZSTD_createCCtx(void);
ZSTDLIB_API size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx);
/*! ZSTD_compressCCtx() :
- * Same as ZSTD_compress(), requires an allocated ZSTD_CCtx (see ZSTD_createCCtx()). */
-ZSTDLIB_API size_t ZSTD_compressCCtx(ZSTD_CCtx* ctx,
+ * Same as ZSTD_compress(), using an explicit ZSTD_CCtx
+ * The function will compress at requested compression level,
+ * ignoring any other parameter */
+ZSTDLIB_API size_t ZSTD_compressCCtx(ZSTD_CCtx* cctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
int compressionLevel);
@@ -166,82 +215,328 @@
ZSTDLIB_API size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx);
/*! ZSTD_decompressDCtx() :
- * Same as ZSTD_decompress(), requires an allocated ZSTD_DCtx (see ZSTD_createDCtx()) */
-ZSTDLIB_API size_t ZSTD_decompressDCtx(ZSTD_DCtx* ctx,
+ * Same as ZSTD_decompress(),
+ * requires an allocated ZSTD_DCtx.
+ * Compatible with sticky parameters.
+ */
+ZSTDLIB_API size_t ZSTD_decompressDCtx(ZSTD_DCtx* dctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize);
-/**************************
-* Simple dictionary API
-***************************/
-/*! ZSTD_compress_usingDict() :
- * Compression using a predefined Dictionary (see dictBuilder/zdict.h).
- * Note : This function loads the dictionary, resulting in significant startup delay.
- * Note : When `dict == NULL || dictSize < 8` no dictionary is used. */
-ZSTDLIB_API size_t ZSTD_compress_usingDict(ZSTD_CCtx* ctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- const void* dict,size_t dictSize,
- int compressionLevel);
+/***************************************
+* Advanced compression API
+***************************************/
-/*! ZSTD_decompress_usingDict() :
- * Decompression using a predefined Dictionary (see dictBuilder/zdict.h).
- * Dictionary must be identical to the one used during compression.
- * Note : This function loads the dictionary, resulting in significant startup delay.
- * Note : When `dict == NULL || dictSize < 8` no dictionary is used. */
-ZSTDLIB_API size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- const void* dict,size_t dictSize);
+/* API design :
+ * Parameters are pushed one by one into an existing context,
+ * using ZSTD_CCtx_set*() functions.
+ * Pushed parameters are sticky : they are valid for next compressed frame, and any subsequent frame.
+ * "sticky" parameters are applicable to `ZSTD_compress2()` and `ZSTD_compressStream*()` !
+ * They do not apply to "simple" one-shot variants such as ZSTD_compressCCtx()
+ *
+ * It's possible to reset all parameters to "default" using ZSTD_CCtx_reset().
+ *
+ * This API supercedes all other "advanced" API entry points in the experimental section.
+ * In the future, we expect to remove from experimental API entry points which are redundant with this API.
+ */
-/**********************************
- * Bulk processing dictionary API
- *********************************/
-typedef struct ZSTD_CDict_s ZSTD_CDict;
-
-/*! ZSTD_createCDict() :
- * When compressing multiple messages / blocks with the same dictionary, it's recommended to load it just once.
- * ZSTD_createCDict() will create a digested dictionary, ready to start future compression operations without startup delay.
- * ZSTD_CDict can be created once and shared by multiple threads concurrently, since its usage is read-only.
- * `dictBuffer` can be released after ZSTD_CDict creation, since its content is copied within CDict */
-ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict(const void* dictBuffer, size_t dictSize,
- int compressionLevel);
-
-/*! ZSTD_freeCDict() :
- * Function frees memory allocated by ZSTD_createCDict(). */
-ZSTDLIB_API size_t ZSTD_freeCDict(ZSTD_CDict* CDict);
-
-/*! ZSTD_compress_usingCDict() :
- * Compression using a digested Dictionary.
- * Faster startup than ZSTD_compress_usingDict(), recommended when same dictionary is used multiple times.
- * Note that compression level is decided during dictionary creation.
- * Frame parameters are hardcoded (dictID=yes, contentSize=yes, checksum=no) */
-ZSTDLIB_API size_t ZSTD_compress_usingCDict(ZSTD_CCtx* cctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- const ZSTD_CDict* cdict);
+/* Compression strategies, listed from fastest to strongest */
+typedef enum { ZSTD_fast=1,
+ ZSTD_dfast=2,
+ ZSTD_greedy=3,
+ ZSTD_lazy=4,
+ ZSTD_lazy2=5,
+ ZSTD_btlazy2=6,
+ ZSTD_btopt=7,
+ ZSTD_btultra=8,
+ ZSTD_btultra2=9
+ /* note : new strategies _might_ be added in the future.
+ Only the order (from fast to strong) is guaranteed */
+} ZSTD_strategy;
-typedef struct ZSTD_DDict_s ZSTD_DDict;
+typedef enum {
-/*! ZSTD_createDDict() :
- * Create a digested dictionary, ready to start decompression operation without startup delay.
- * dictBuffer can be released after DDict creation, as its content is copied inside DDict */
-ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict(const void* dictBuffer, size_t dictSize);
+ /* compression parameters
+ * Note: When compressing with a ZSTD_CDict these parameters are superseded
+ * by the parameters used to construct the ZSTD_CDict. See ZSTD_CCtx_refCDict()
+ * for more info (superseded-by-cdict). */
+ ZSTD_c_compressionLevel=100, /* Update all compression parameters according to pre-defined cLevel table
+ * Default level is ZSTD_CLEVEL_DEFAULT==3.
+ * Special: value 0 means default, which is controlled by ZSTD_CLEVEL_DEFAULT.
+ * Note 1 : it's possible to pass a negative compression level.
+ * Note 2 : setting a level sets all default values of other compression parameters */
+ ZSTD_c_windowLog=101, /* Maximum allowed back-reference distance, expressed as power of 2.
+ * Must be clamped between ZSTD_WINDOWLOG_MIN and ZSTD_WINDOWLOG_MAX.
+ * Special: value 0 means "use default windowLog".
+ * Note: Using a windowLog greater than ZSTD_WINDOWLOG_LIMIT_DEFAULT
+ * requires explicitly allowing such window size at decompression stage if using streaming. */
+ ZSTD_c_hashLog=102, /* Size of the initial probe table, as a power of 2.
+ * Resulting memory usage is (1 << (hashLog+2)).
+ * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX.
+ * Larger tables improve compression ratio of strategies <= dFast,
+ * and improve speed of strategies > dFast.
+ * Special: value 0 means "use default hashLog". */
+ ZSTD_c_chainLog=103, /* Size of the multi-probe search table, as a power of 2.
+ * Resulting memory usage is (1 << (chainLog+2)).
+ * Must be clamped between ZSTD_CHAINLOG_MIN and ZSTD_CHAINLOG_MAX.
+ * Larger tables result in better and slower compression.
+ * This parameter is useless when using "fast" strategy.
+ * It's still useful when using "dfast" strategy,
+ * in which case it defines a secondary probe table.
+ * Special: value 0 means "use default chainLog". */
+ ZSTD_c_searchLog=104, /* Number of search attempts, as a power of 2.
+ * More attempts result in better and slower compression.
+ * This parameter is useless when using "fast" and "dFast" strategies.
+ * Special: value 0 means "use default searchLog". */
+ ZSTD_c_minMatch=105, /* Minimum size of searched matches.
+ * Note that Zstandard can still find matches of smaller size,
+ * it just tweaks its search algorithm to look for this size and larger.
+ * Larger values increase compression and decompression speed, but decrease ratio.
+ * Must be clamped between ZSTD_MINMATCH_MIN and ZSTD_MINMATCH_MAX.
+ * Note that currently, for all strategies < btopt, effective minimum is 4.
+ * , for all strategies > fast, effective maximum is 6.
+ * Special: value 0 means "use default minMatchLength". */
+ ZSTD_c_targetLength=106, /* Impact of this field depends on strategy.
+ * For strategies btopt, btultra & btultra2:
+ * Length of Match considered "good enough" to stop search.
+ * Larger values make compression stronger, and slower.
+ * For strategy fast:
+ * Distance between match sampling.
+ * Larger values make compression faster, and weaker.
+ * Special: value 0 means "use default targetLength". */
+ ZSTD_c_strategy=107, /* See ZSTD_strategy enum definition.
+ * The higher the value of selected strategy, the more complex it is,
+ * resulting in stronger and slower compression.
+ * Special: value 0 means "use default strategy". */
-/*! ZSTD_freeDDict() :
- * Function frees memory allocated with ZSTD_createDDict() */
-ZSTDLIB_API size_t ZSTD_freeDDict(ZSTD_DDict* ddict);
+ /* LDM mode parameters */
+ ZSTD_c_enableLongDistanceMatching=160, /* Enable long distance matching.
+ * This parameter is designed to improve compression ratio
+ * for large inputs, by finding large matches at long distance.
+ * It increases memory usage and window size.
+ * Note: enabling this parameter increases default ZSTD_c_windowLog to 128 MB
+ * except when expressly set to a different value. */
+ ZSTD_c_ldmHashLog=161, /* Size of the table for long distance matching, as a power of 2.
+ * Larger values increase memory usage and compression ratio,
+ * but decrease compression speed.
+ * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX
+ * default: windowlog - 7.
+ * Special: value 0 means "automatically determine hashlog". */
+ ZSTD_c_ldmMinMatch=162, /* Minimum match size for long distance matcher.
+ * Larger/too small values usually decrease compression ratio.
+ * Must be clamped between ZSTD_LDM_MINMATCH_MIN and ZSTD_LDM_MINMATCH_MAX.
+ * Special: value 0 means "use default value" (default: 64). */
+ ZSTD_c_ldmBucketSizeLog=163, /* Log size of each bucket in the LDM hash table for collision resolution.
+ * Larger values improve collision resolution but decrease compression speed.
+ * The maximum value is ZSTD_LDM_BUCKETSIZELOG_MAX.
+ * Special: value 0 means "use default value" (default: 3). */
+ ZSTD_c_ldmHashRateLog=164, /* Frequency of inserting/looking up entries into the LDM hash table.
+ * Must be clamped between 0 and (ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN).
+ * Default is MAX(0, (windowLog - ldmHashLog)), optimizing hash table usage.
+ * Larger values improve compression speed.
+ * Deviating far from default value will likely result in a compression ratio decrease.
+ * Special: value 0 means "automatically determine hashRateLog". */
-/*! ZSTD_decompress_usingDDict() :
- * Decompression using a digested Dictionary.
- * Faster startup than ZSTD_decompress_usingDict(), recommended when same dictionary is used multiple times. */
-ZSTDLIB_API size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- const ZSTD_DDict* ddict);
+ /* frame parameters */
+ ZSTD_c_contentSizeFlag=200, /* Content size will be written into frame header _whenever known_ (default:1)
+ * Content size must be known at the beginning of compression.
+ * This is automatically the case when using ZSTD_compress2(),
+ * For streaming variants, content size must be provided with ZSTD_CCtx_setPledgedSrcSize() */
+ ZSTD_c_checksumFlag=201, /* A 32-bits checksum of content is written at end of frame (default:0) */
+ ZSTD_c_dictIDFlag=202, /* When applicable, dictionary's ID is written into frame header (default:1) */
+
+ /* multi-threading parameters */
+ /* These parameters are only useful if multi-threading is enabled (compiled with build macro ZSTD_MULTITHREAD).
+ * They return an error otherwise. */
+ ZSTD_c_nbWorkers=400, /* Select how many threads will be spawned to compress in parallel.
+ * When nbWorkers >= 1, triggers asynchronous mode when used with ZSTD_compressStream*() :
+ * ZSTD_compressStream*() consumes input and flush output if possible, but immediately gives back control to caller,
+ * while compression work is performed in parallel, within worker threads.
+ * (note : a strong exception to this rule is when first invocation of ZSTD_compressStream2() sets ZSTD_e_end :
+ * in which case, ZSTD_compressStream2() delegates to ZSTD_compress2(), which is always a blocking call).
+ * More workers improve speed, but also increase memory usage.
+ * Default value is `0`, aka "single-threaded mode" : no worker is spawned, compression is performed inside Caller's thread, all invocations are blocking */
+ ZSTD_c_jobSize=401, /* Size of a compression job. This value is enforced only when nbWorkers >= 1.
+ * Each compression job is completed in parallel, so this value can indirectly impact the nb of active threads.
+ * 0 means default, which is dynamically determined based on compression parameters.
+ * Job size must be a minimum of overlap size, or 1 MB, whichever is largest.
+ * The minimum size is automatically and transparently enforced */
+ ZSTD_c_overlapLog=402, /* Control the overlap size, as a fraction of window size.
+ * The overlap size is an amount of data reloaded from previous job at the beginning of a new job.
+ * It helps preserve compression ratio, while each job is compressed in parallel.
+ * This value is enforced only when nbWorkers >= 1.
+ * Larger values increase compression ratio, but decrease speed.
+ * Possible values range from 0 to 9 :
+ * - 0 means "default" : value will be determined by the library, depending on strategy
+ * - 1 means "no overlap"
+ * - 9 means "full overlap", using a full window size.
+ * Each intermediate rank increases/decreases load size by a factor 2 :
+ * 9: full window; 8: w/2; 7: w/4; 6: w/8; 5:w/16; 4: w/32; 3:w/64; 2:w/128; 1:no overlap; 0:default
+ * default value varies between 6 and 9, depending on strategy */
+
+ /* note : additional experimental parameters are also available
+ * within the experimental section of the API.
+ * At the time of this writing, they include :
+ * ZSTD_c_rsyncable
+ * ZSTD_c_format
+ * ZSTD_c_forceMaxWindow
+ * ZSTD_c_forceAttachDict
+ * ZSTD_c_literalCompressionMode
+ * ZSTD_c_targetCBlockSize
+ * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them.
+ * note : never ever use experimentalParam? names directly;
+ * also, the enums values themselves are unstable and can still change.
+ */
+ ZSTD_c_experimentalParam1=500,
+ ZSTD_c_experimentalParam2=10,
+ ZSTD_c_experimentalParam3=1000,
+ ZSTD_c_experimentalParam4=1001,
+ ZSTD_c_experimentalParam5=1002,
+ ZSTD_c_experimentalParam6=1003,
+} ZSTD_cParameter;
+
+typedef struct {
+ size_t error;
+ int lowerBound;
+ int upperBound;
+} ZSTD_bounds;
+
+/*! ZSTD_cParam_getBounds() :
+ * All parameters must belong to an interval with lower and upper bounds,
+ * otherwise they will either trigger an error or be automatically clamped.
+ * @return : a structure, ZSTD_bounds, which contains
+ * - an error status field, which must be tested using ZSTD_isError()
+ * - lower and upper bounds, both inclusive
+ */
+ZSTDLIB_API ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter cParam);
+
+/*! ZSTD_CCtx_setParameter() :
+ * Set one compression parameter, selected by enum ZSTD_cParameter.
+ * All parameters have valid bounds. Bounds can be queried using ZSTD_cParam_getBounds().
+ * Providing a value beyond bound will either clamp it, or trigger an error (depending on parameter).
+ * Setting a parameter is generally only possible during frame initialization (before starting compression).
+ * Exception : when using multi-threading mode (nbWorkers >= 1),
+ * the following parameters can be updated _during_ compression (within same frame):
+ * => compressionLevel, hashLog, chainLog, searchLog, minMatch, targetLength and strategy.
+ * new parameters will be active for next job only (after a flush()).
+ * @return : an error code (which can be tested using ZSTD_isError()).
+ */
+ZSTDLIB_API size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int value);
+
+/*! ZSTD_CCtx_setPledgedSrcSize() :
+ * Total input data size to be compressed as a single frame.
+ * Value will be written in frame header, unless if explicitly forbidden using ZSTD_c_contentSizeFlag.
+ * This value will also be controlled at end of frame, and trigger an error if not respected.
+ * @result : 0, or an error code (which can be tested with ZSTD_isError()).
+ * Note 1 : pledgedSrcSize==0 actually means zero, aka an empty frame.
+ * In order to mean "unknown content size", pass constant ZSTD_CONTENTSIZE_UNKNOWN.
+ * ZSTD_CONTENTSIZE_UNKNOWN is default value for any new frame.
+ * Note 2 : pledgedSrcSize is only valid once, for the next frame.
+ * It's discarded at the end of the frame, and replaced by ZSTD_CONTENTSIZE_UNKNOWN.
+ * Note 3 : Whenever all input data is provided and consumed in a single round,
+ * for example with ZSTD_compress2(),
+ * or invoking immediately ZSTD_compressStream2(,,,ZSTD_e_end),
+ * this value is automatically overridden by srcSize instead.
+ */
+ZSTDLIB_API size_t ZSTD_CCtx_setPledgedSrcSize(ZSTD_CCtx* cctx, unsigned long long pledgedSrcSize);
+
+typedef enum {
+ ZSTD_reset_session_only = 1,
+ ZSTD_reset_parameters = 2,
+ ZSTD_reset_session_and_parameters = 3
+} ZSTD_ResetDirective;
+
+/*! ZSTD_CCtx_reset() :
+ * There are 2 different things that can be reset, independently or jointly :
+ * - The session : will stop compressing current frame, and make CCtx ready to start a new one.
+ * Useful after an error, or to interrupt any ongoing compression.
+ * Any internal data not yet flushed is cancelled.
+ * Compression parameters and dictionary remain unchanged.
+ * They will be used to compress next frame.
+ * Resetting session never fails.
+ * - The parameters : changes all parameters back to "default".
+ * This removes any reference to any dictionary too.
+ * Parameters can only be changed between 2 sessions (i.e. no compression is currently ongoing)
+ * otherwise the reset fails, and function returns an error value (which can be tested using ZSTD_isError())
+ * - Both : similar to resetting the session, followed by resetting parameters.
+ */
+ZSTDLIB_API size_t ZSTD_CCtx_reset(ZSTD_CCtx* cctx, ZSTD_ResetDirective reset);
+
+/*! ZSTD_compress2() :
+ * Behave the same as ZSTD_compressCCtx(), but compression parameters are set using the advanced API.
+ * ZSTD_compress2() always starts a new frame.
+ * Should cctx hold data from a previously unfinished frame, everything about it is forgotten.
+ * - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*()
+ * - The function is always blocking, returns when compression is completed.
+ * Hint : compression runs faster if `dstCapacity` >= `ZSTD_compressBound(srcSize)`.
+ * @return : compressed size written into `dst` (<= `dstCapacity),
+ * or an error code if it fails (which can be tested using ZSTD_isError()).
+ */
+ZSTDLIB_API size_t ZSTD_compress2( ZSTD_CCtx* cctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize);
+
+
+/***************************************
+* Advanced decompression API
+***************************************/
+
+/* The advanced API pushes parameters one by one into an existing DCtx context.
+ * Parameters are sticky, and remain valid for all following frames
+ * using the same DCtx context.
+ * It's possible to reset parameters to default values using ZSTD_DCtx_reset().
+ * Note : This API is compatible with existing ZSTD_decompressDCtx() and ZSTD_decompressStream().
+ * Therefore, no new decompression function is necessary.
+ */
+
+typedef enum {
+
+ ZSTD_d_windowLogMax=100, /* Select a size limit (in power of 2) beyond which
+ * the streaming API will refuse to allocate memory buffer
+ * in order to protect the host from unreasonable memory requirements.
+ * This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode.
+ * By default, a decompression context accepts window sizes <= (1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT).
+ * Special: value 0 means "use default maximum windowLog". */
+
+ /* note : additional experimental parameters are also available
+ * within the experimental section of the API.
+ * At the time of this writing, they include :
+ * ZSTD_c_format
+ * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them.
+ * note : never ever use experimentalParam? names directly
+ */
+ ZSTD_d_experimentalParam1=1000
+
+} ZSTD_dParameter;
+
+/*! ZSTD_dParam_getBounds() :
+ * All parameters must belong to an interval with lower and upper bounds,
+ * otherwise they will either trigger an error or be automatically clamped.
+ * @return : a structure, ZSTD_bounds, which contains
+ * - an error status field, which must be tested using ZSTD_isError()
+ * - both lower and upper bounds, inclusive
+ */
+ZSTDLIB_API ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam);
+
+/*! ZSTD_DCtx_setParameter() :
+ * Set one compression parameter, selected by enum ZSTD_dParameter.
+ * All parameters have valid bounds. Bounds can be queried using ZSTD_dParam_getBounds().
+ * Providing a value beyond bound will either clamp it, or trigger an error (depending on parameter).
+ * Setting a parameter is only possible during frame initialization (before starting decompression).
+ * @return : 0, or an error code (which can be tested using ZSTD_isError()).
+ */
+ZSTDLIB_API size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int value);
+
+/*! ZSTD_DCtx_reset() :
+ * Return a DCtx to clean state.
+ * Session and parameters can be reset jointly or separately.
+ * Parameters can only be reset when no active frame is being decompressed.
+ * @return : 0, or an error code, which can be tested with ZSTD_isError()
+ */
+ZSTDLIB_API size_t ZSTD_DCtx_reset(ZSTD_DCtx* dctx, ZSTD_ResetDirective reset);
/****************************
@@ -268,57 +563,153 @@
* A ZSTD_CStream object is required to track streaming operation.
* Use ZSTD_createCStream() and ZSTD_freeCStream() to create/release resources.
* ZSTD_CStream objects can be reused multiple times on consecutive compression operations.
-* It is recommended to re-use ZSTD_CStream in situations where many streaming operations will be achieved consecutively,
-* since it will play nicer with system's memory, by re-using already allocated memory.
-* Use one separate ZSTD_CStream per thread for parallel execution.
+* It is recommended to re-use ZSTD_CStream since it will play nicer with system's memory, by re-using already allocated memory.
*
-* Start a new compression by initializing ZSTD_CStream.
-* Use ZSTD_initCStream() to start a new compression operation.
-* Use ZSTD_initCStream_usingDict() or ZSTD_initCStream_usingCDict() for a compression which requires a dictionary (experimental section)
+* For parallel execution, use one separate ZSTD_CStream per thread.
*
-* Use ZSTD_compressStream() repetitively to consume input stream.
-* The function will automatically update both `pos` fields.
-* Note that it may not consume the entire input, in which case `pos < size`,
-* and it's up to the caller to present again remaining data.
-* @return : a size hint, preferred nb of bytes to use as input for next function call
-* or an error code, which can be tested using ZSTD_isError().
-* Note 1 : it's just a hint, to help latency a little, any other value will work fine.
-* Note 2 : size hint is guaranteed to be <= ZSTD_CStreamInSize()
+* note : since v1.3.0, ZSTD_CStream and ZSTD_CCtx are the same thing.
*
-* At any moment, it's possible to flush whatever data remains within internal buffer, using ZSTD_flushStream().
-* `output->pos` will be updated.
-* Note that some content might still be left within internal buffer if `output->size` is too small.
-* @return : nb of bytes still present within internal buffer (0 if it's empty)
+* Parameters are sticky : when starting a new compression on the same context,
+* it will re-use the same sticky parameters as previous compression session.
+* When in doubt, it's recommended to fully initialize the context before usage.
+* Use ZSTD_CCtx_reset() to reset the context and ZSTD_CCtx_setParameter(),
+* ZSTD_CCtx_setPledgedSrcSize(), or ZSTD_CCtx_loadDictionary() and friends to
+* set more specific parameters, the pledged source size, or load a dictionary.
+*
+* Use ZSTD_compressStream2() with ZSTD_e_continue as many times as necessary to
+* consume input stream. The function will automatically update both `pos`
+* fields within `input` and `output`.
+* Note that the function may not consume the entire input, for example, because
+* the output buffer is already full, in which case `input.pos < input.size`.
+* The caller must check if input has been entirely consumed.
+* If not, the caller must make some room to receive more compressed data,
+* and then present again remaining input data.
+* note: ZSTD_e_continue is guaranteed to make some forward progress when called,
+* but doesn't guarantee maximal forward progress. This is especially relevant
+* when compressing with multiple threads. The call won't block if it can
+* consume some input, but if it can't it will wait for some, but not all,
+* output to be flushed.
+* @return : provides a minimum amount of data remaining to be flushed from internal buffers
+* or an error code, which can be tested using ZSTD_isError().
+*
+* At any moment, it's possible to flush whatever data might remain stuck within internal buffer,
+* using ZSTD_compressStream2() with ZSTD_e_flush. `output->pos` will be updated.
+* Note that, if `output->size` is too small, a single invocation with ZSTD_e_flush might not be enough (return code > 0).
+* In which case, make some room to receive more compressed data, and call again ZSTD_compressStream2() with ZSTD_e_flush.
+* You must continue calling ZSTD_compressStream2() with ZSTD_e_flush until it returns 0, at which point you can change the
+* operation.
+* note: ZSTD_e_flush will flush as much output as possible, meaning when compressing with multiple threads, it will
+* block until the flush is complete or the output buffer is full.
+* @return : 0 if internal buffers are entirely flushed,
+* >0 if some data still present within internal buffer (the value is minimal estimation of remaining size),
* or an error code, which can be tested using ZSTD_isError().
*
-* ZSTD_endStream() instructs to finish a frame.
+* Calling ZSTD_compressStream2() with ZSTD_e_end instructs to finish a frame.
* It will perform a flush and write frame epilogue.
* The epilogue is required for decoders to consider a frame completed.
-* ZSTD_endStream() may not be able to flush full data if `output->size` is too small.
-* In which case, call again ZSTD_endStream() to complete the flush.
+* flush operation is the same, and follows same rules as calling ZSTD_compressStream2() with ZSTD_e_flush.
+* You must continue calling ZSTD_compressStream2() with ZSTD_e_end until it returns 0, at which point you are free to
+* start a new frame.
+* note: ZSTD_e_end will flush as much output as possible, meaning when compressing with multiple threads, it will
+* block until the flush is complete or the output buffer is full.
* @return : 0 if frame fully completed and fully flushed,
- or >0 if some data is still present within internal buffer
- (value is minimum size estimation for remaining data to flush, but it could be more)
+* >0 if some data still present within internal buffer (the value is minimal estimation of remaining size),
* or an error code, which can be tested using ZSTD_isError().
*
* *******************************************************************/
typedef ZSTD_CCtx ZSTD_CStream; /**< CCtx and CStream are now effectively same object (>= v1.3.0) */
- /* Continue to distinguish them for compatibility with versions <= v1.2.0 */
+ /* Continue to distinguish them for compatibility with older versions <= v1.2.0 */
/*===== ZSTD_CStream management functions =====*/
ZSTDLIB_API ZSTD_CStream* ZSTD_createCStream(void);
ZSTDLIB_API size_t ZSTD_freeCStream(ZSTD_CStream* zcs);
/*===== Streaming compression functions =====*/
-ZSTDLIB_API size_t ZSTD_initCStream(ZSTD_CStream* zcs, int compressionLevel);
-ZSTDLIB_API size_t ZSTD_compressStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output, ZSTD_inBuffer* input);
-ZSTDLIB_API size_t ZSTD_flushStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output);
-ZSTDLIB_API size_t ZSTD_endStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output);
+typedef enum {
+ ZSTD_e_continue=0, /* collect more data, encoder decides when to output compressed result, for optimal compression ratio */
+ ZSTD_e_flush=1, /* flush any data provided so far,
+ * it creates (at least) one new block, that can be decoded immediately on reception;
+ * frame will continue: any future data can still reference previously compressed data, improving compression.
+ * note : multithreaded compression will block to flush as much output as possible. */
+ ZSTD_e_end=2 /* flush any remaining data _and_ close current frame.
+ * note that frame is only closed after compressed data is fully flushed (return value == 0).
+ * After that point, any additional data starts a new frame.
+ * note : each frame is independent (does not reference any content from previous frame).
+ : note : multithreaded compression will block to flush as much output as possible. */
+} ZSTD_EndDirective;
+/*! ZSTD_compressStream2() :
+ * Behaves about the same as ZSTD_compressStream, with additional control on end directive.
+ * - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*()
+ * - Compression parameters cannot be changed once compression is started (save a list of exceptions in multi-threading mode)
+ * - output->pos must be <= dstCapacity, input->pos must be <= srcSize
+ * - output->pos and input->pos will be updated. They are guaranteed to remain below their respective limit.
+ * - When nbWorkers==0 (default), function is blocking : it completes its job before returning to caller.
+ * - When nbWorkers>=1, function is non-blocking : it just acquires a copy of input, and distributes jobs to internal worker threads, flush whatever is available,
+ * and then immediately returns, just indicating that there is some data remaining to be flushed.
+ * The function nonetheless guarantees forward progress : it will return only after it reads or write at least 1+ byte.
+ * - Exception : if the first call requests a ZSTD_e_end directive and provides enough dstCapacity, the function delegates to ZSTD_compress2() which is always blocking.
+ * - @return provides a minimum amount of data remaining to be flushed from internal buffers
+ * or an error code, which can be tested using ZSTD_isError().
+ * if @return != 0, flush is not fully completed, there is still some data left within internal buffers.
+ * This is useful for ZSTD_e_flush, since in this case more flushes are necessary to empty all buffers.
+ * For ZSTD_e_end, @return == 0 when internal buffers are fully flushed and frame is completed.
+ * - after a ZSTD_e_end directive, if internal buffer is not fully flushed (@return != 0),
+ * only ZSTD_e_end or ZSTD_e_flush operations are allowed.
+ * Before starting a new compression job, or changing compression parameters,
+ * it is required to fully flush internal buffers.
+ */
+ZSTDLIB_API size_t ZSTD_compressStream2( ZSTD_CCtx* cctx,
+ ZSTD_outBuffer* output,
+ ZSTD_inBuffer* input,
+ ZSTD_EndDirective endOp);
+
+
+/* These buffer sizes are softly recommended.
+ * They are not required : ZSTD_compressStream*() happily accepts any buffer size, for both input and output.
+ * Respecting the recommended size just makes it a bit easier for ZSTD_compressStream*(),
+ * reducing the amount of memory shuffling and buffering, resulting in minor performance savings.
+ *
+ * However, note that these recommendations are from the perspective of a C caller program.
+ * If the streaming interface is invoked from some other language,
+ * especially managed ones such as Java or Go, through a foreign function interface such as jni or cgo,
+ * a major performance rule is to reduce crossing such interface to an absolute minimum.
+ * It's not rare that performance ends being spent more into the interface, rather than compression itself.
+ * In which cases, prefer using large buffers, as large as practical,
+ * for both input and output, to reduce the nb of roundtrips.
+ */
ZSTDLIB_API size_t ZSTD_CStreamInSize(void); /**< recommended size for input buffer */
-ZSTDLIB_API size_t ZSTD_CStreamOutSize(void); /**< recommended size for output buffer. Guarantee to successfully flush at least one complete compressed block in all circumstances. */
+ZSTDLIB_API size_t ZSTD_CStreamOutSize(void); /**< recommended size for output buffer. Guarantee to successfully flush at least one complete compressed block. */
+/* *****************************************************************************
+ * This following is a legacy streaming API.
+ * It can be replaced by ZSTD_CCtx_reset() and ZSTD_compressStream2().
+ * It is redundant, but remains fully supported.
+ * Advanced parameters and dictionary compression can only be used through the
+ * new API.
+ ******************************************************************************/
+
+/*!
+ * Equivalent to:
+ *
+ * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
+ * ZSTD_CCtx_refCDict(zcs, NULL); // clear the dictionary (if any)
+ * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel);
+ */
+ZSTDLIB_API size_t ZSTD_initCStream(ZSTD_CStream* zcs, int compressionLevel);
+/*!
+ * Alternative for ZSTD_compressStream2(zcs, output, input, ZSTD_e_continue).
+ * NOTE: The return value is different. ZSTD_compressStream() returns a hint for
+ * the next read size (if non-zero and not an error). ZSTD_compressStream2()
+ * returns the minimum nb of bytes left to flush (if non-zero and not an error).
+ */
+ZSTDLIB_API size_t ZSTD_compressStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output, ZSTD_inBuffer* input);
+/*! Equivalent to ZSTD_compressStream2(zcs, output, &emptyInput, ZSTD_e_flush). */
+ZSTDLIB_API size_t ZSTD_flushStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output);
+/*! Equivalent to ZSTD_compressStream2(zcs, output, &emptyInput, ZSTD_e_end). */
+ZSTDLIB_API size_t ZSTD_endStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output);
+
/*-***************************************************************************
* Streaming decompression - HowTo
@@ -327,101 +718,374 @@
* Use ZSTD_createDStream() and ZSTD_freeDStream() to create/release resources.
* ZSTD_DStream objects can be re-used multiple times.
*
-* Use ZSTD_initDStream() to start a new decompression operation,
-* or ZSTD_initDStream_usingDict() if decompression requires a dictionary.
-* @return : recommended first input size
+* Use ZSTD_initDStream() to start a new decompression operation.
+* @return : recommended first input size
+* Alternatively, use advanced API to set specific properties.
*
* Use ZSTD_decompressStream() repetitively to consume your input.
* The function will update both `pos` fields.
* If `input.pos < input.size`, some input has not been consumed.
* It's up to the caller to present again remaining data.
+* The function tries to flush all data decoded immediately, respecting output buffer size.
* If `output.pos < output.size`, decoder has flushed everything it could.
-* @return : 0 when a frame is completely decoded and fully flushed,
-* an error code, which can be tested using ZSTD_isError(),
-* any other value > 0, which means there is still some decoding to do to complete current frame.
-* The return value is a suggested next input size (a hint to improve latency) that will never load more than the current frame.
+* But if `output.pos == output.size`, there might be some data left within internal buffers.,
+* In which case, call ZSTD_decompressStream() again to flush whatever remains in the buffer.
+* Note : with no additional input provided, amount of data flushed is necessarily <= ZSTD_BLOCKSIZE_MAX.
+* @return : 0 when a frame is completely decoded and fully flushed,
+* or an error code, which can be tested using ZSTD_isError(),
+* or any other value > 0, which means there is still some decoding or flushing to do to complete current frame :
+* the return value is a suggested next input size (just a hint for better latency)
+* that will never request more than the remaining frame size.
* *******************************************************************************/
typedef ZSTD_DCtx ZSTD_DStream; /**< DCtx and DStream are now effectively same object (>= v1.3.0) */
- /* For compatibility with versions <= v1.2.0, continue to consider them separated. */
+ /* For compatibility with versions <= v1.2.0, prefer differentiating them. */
/*===== ZSTD_DStream management functions =====*/
ZSTDLIB_API ZSTD_DStream* ZSTD_createDStream(void);
ZSTDLIB_API size_t ZSTD_freeDStream(ZSTD_DStream* zds);
/*===== Streaming decompression functions =====*/
+
+/* This function is redundant with the advanced API and equivalent to:
+ *
+ * ZSTD_DCtx_reset(zds);
+ * ZSTD_DCtx_refDDict(zds, NULL);
+ */
ZSTDLIB_API size_t ZSTD_initDStream(ZSTD_DStream* zds);
+
ZSTDLIB_API size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inBuffer* input);
ZSTDLIB_API size_t ZSTD_DStreamInSize(void); /*!< recommended size for input buffer */
ZSTDLIB_API size_t ZSTD_DStreamOutSize(void); /*!< recommended size for output buffer. Guarantee to successfully flush at least one complete block in all circumstances. */
+
+/**************************
+* Simple dictionary API
+***************************/
+/*! ZSTD_compress_usingDict() :
+ * Compression at an explicit compression level using a Dictionary.
+ * A dictionary can be any arbitrary data segment (also called a prefix),
+ * or a buffer with specified information (see dictBuilder/zdict.h).
+ * Note : This function loads the dictionary, resulting in significant startup delay.
+ * It's intended for a dictionary used only once.
+ * Note 2 : When `dict == NULL || dictSize < 8` no dictionary is used. */
+ZSTDLIB_API size_t ZSTD_compress_usingDict(ZSTD_CCtx* ctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ const void* dict,size_t dictSize,
+ int compressionLevel);
+
+/*! ZSTD_decompress_usingDict() :
+ * Decompression using a known Dictionary.
+ * Dictionary must be identical to the one used during compression.
+ * Note : This function loads the dictionary, resulting in significant startup delay.
+ * It's intended for a dictionary used only once.
+ * Note : When `dict == NULL || dictSize < 8` no dictionary is used. */
+ZSTDLIB_API size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ const void* dict,size_t dictSize);
+
+
+/***********************************
+ * Bulk processing dictionary API
+ **********************************/
+typedef struct ZSTD_CDict_s ZSTD_CDict;
+
+/*! ZSTD_createCDict() :
+ * When compressing multiple messages / blocks using the same dictionary, it's recommended to load it only once.
+ * ZSTD_createCDict() will create a digested dictionary, ready to start future compression operations without startup cost.
+ * ZSTD_CDict can be created once and shared by multiple threads concurrently, since its usage is read-only.
+ * `dictBuffer` can be released after ZSTD_CDict creation, because its content is copied within CDict.
+ * Consider experimental function `ZSTD_createCDict_byReference()` if you prefer to not duplicate `dictBuffer` content.
+ * Note : A ZSTD_CDict can be created from an empty dictBuffer, but it is inefficient when used to compress small data. */
+ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict(const void* dictBuffer, size_t dictSize,
+ int compressionLevel);
+
+/*! ZSTD_freeCDict() :
+ * Function frees memory allocated by ZSTD_createCDict(). */
+ZSTDLIB_API size_t ZSTD_freeCDict(ZSTD_CDict* CDict);
+
+/*! ZSTD_compress_usingCDict() :
+ * Compression using a digested Dictionary.
+ * Recommended when same dictionary is used multiple times.
+ * Note : compression level is _decided at dictionary creation time_,
+ * and frame parameters are hardcoded (dictID=yes, contentSize=yes, checksum=no) */
+ZSTDLIB_API size_t ZSTD_compress_usingCDict(ZSTD_CCtx* cctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ const ZSTD_CDict* cdict);
+
+
+typedef struct ZSTD_DDict_s ZSTD_DDict;
+
+/*! ZSTD_createDDict() :
+ * Create a digested dictionary, ready to start decompression operation without startup delay.
+ * dictBuffer can be released after DDict creation, as its content is copied inside DDict. */
+ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict(const void* dictBuffer, size_t dictSize);
+
+/*! ZSTD_freeDDict() :
+ * Function frees memory allocated with ZSTD_createDDict() */
+ZSTDLIB_API size_t ZSTD_freeDDict(ZSTD_DDict* ddict);
+
+/*! ZSTD_decompress_usingDDict() :
+ * Decompression using a digested Dictionary.
+ * Recommended when same dictionary is used multiple times. */
+ZSTDLIB_API size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ const ZSTD_DDict* ddict);
+
+
+/********************************
+ * Dictionary helper functions
+ *******************************/
+
+/*! ZSTD_getDictID_fromDict() :
+ * Provides the dictID stored within dictionary.
+ * if @return == 0, the dictionary is not conformant with Zstandard specification.
+ * It can still be loaded, but as a content-only dictionary. */
+ZSTDLIB_API unsigned ZSTD_getDictID_fromDict(const void* dict, size_t dictSize);
+
+/*! ZSTD_getDictID_fromDDict() :
+ * Provides the dictID of the dictionary loaded into `ddict`.
+ * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty.
+ * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */
+ZSTDLIB_API unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict);
+
+/*! ZSTD_getDictID_fromFrame() :
+ * Provides the dictID required to decompressed the frame stored within `src`.
+ * If @return == 0, the dictID could not be decoded.
+ * This could for one of the following reasons :
+ * - The frame does not require a dictionary to be decoded (most common case).
+ * - The frame was built with dictID intentionally removed. Whatever dictionary is necessary is a hidden information.
+ * Note : this use case also happens when using a non-conformant dictionary.
+ * - `srcSize` is too small, and as a result, the frame header could not be decoded (only possible if `srcSize < ZSTD_FRAMEHEADERSIZE_MAX`).
+ * - This is not a Zstandard frame.
+ * When identifying the exact failure cause, it's possible to use ZSTD_getFrameHeader(), which will provide a more precise error code. */
+ZSTDLIB_API unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize);
+
+
+/*******************************************************************************
+ * Advanced dictionary and prefix API
+ *
+ * This API allows dictionaries to be used with ZSTD_compress2(),
+ * ZSTD_compressStream2(), and ZSTD_decompress(). Dictionaries are sticky, and
+ * only reset with the context is reset with ZSTD_reset_parameters or
+ * ZSTD_reset_session_and_parameters. Prefixes are single-use.
+ ******************************************************************************/
+
+
+/*! ZSTD_CCtx_loadDictionary() :
+ * Create an internal CDict from `dict` buffer.
+ * Decompression will have to use same dictionary.
+ * @result : 0, or an error code (which can be tested with ZSTD_isError()).
+ * Special: Loading a NULL (or 0-size) dictionary invalidates previous dictionary,
+ * meaning "return to no-dictionary mode".
+ * Note 1 : Dictionary is sticky, it will be used for all future compressed frames.
+ * To return to "no-dictionary" situation, load a NULL dictionary (or reset parameters).
+ * Note 2 : Loading a dictionary involves building tables.
+ * It's also a CPU consuming operation, with non-negligible impact on latency.
+ * Tables are dependent on compression parameters, and for this reason,
+ * compression parameters can no longer be changed after loading a dictionary.
+ * Note 3 :`dict` content will be copied internally.
+ * Use experimental ZSTD_CCtx_loadDictionary_byReference() to reference content instead.
+ * In such a case, dictionary buffer must outlive its users.
+ * Note 4 : Use ZSTD_CCtx_loadDictionary_advanced()
+ * to precisely select how dictionary content must be interpreted. */
+ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary(ZSTD_CCtx* cctx, const void* dict, size_t dictSize);
+
+/*! ZSTD_CCtx_refCDict() :
+ * Reference a prepared dictionary, to be used for all next compressed frames.
+ * Note that compression parameters are enforced from within CDict,
+ * and supersede any compression parameter previously set within CCtx.
+ * The parameters ignored are labled as "superseded-by-cdict" in the ZSTD_cParameter enum docs.
+ * The ignored parameters will be used again if the CCtx is returned to no-dictionary mode.
+ * The dictionary will remain valid for future compressed frames using same CCtx.
+ * @result : 0, or an error code (which can be tested with ZSTD_isError()).
+ * Special : Referencing a NULL CDict means "return to no-dictionary mode".
+ * Note 1 : Currently, only one dictionary can be managed.
+ * Referencing a new dictionary effectively "discards" any previous one.
+ * Note 2 : CDict is just referenced, its lifetime must outlive its usage within CCtx. */
+ZSTDLIB_API size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict);
+
+/*! ZSTD_CCtx_refPrefix() :
+ * Reference a prefix (single-usage dictionary) for next compressed frame.
+ * A prefix is **only used once**. Tables are discarded at end of frame (ZSTD_e_end).
+ * Decompression will need same prefix to properly regenerate data.
+ * Compressing with a prefix is similar in outcome as performing a diff and compressing it,
+ * but performs much faster, especially during decompression (compression speed is tunable with compression level).
+ * @result : 0, or an error code (which can be tested with ZSTD_isError()).
+ * Special: Adding any prefix (including NULL) invalidates any previous prefix or dictionary
+ * Note 1 : Prefix buffer is referenced. It **must** outlive compression.
+ * Its content must remain unmodified during compression.
+ * Note 2 : If the intention is to diff some large src data blob with some prior version of itself,
+ * ensure that the window size is large enough to contain the entire source.
+ * See ZSTD_c_windowLog.
+ * Note 3 : Referencing a prefix involves building tables, which are dependent on compression parameters.
+ * It's a CPU consuming operation, with non-negligible impact on latency.
+ * If there is a need to use the same prefix multiple times, consider loadDictionary instead.
+ * Note 4 : By default, the prefix is interpreted as raw content (ZSTD_dm_rawContent).
+ * Use experimental ZSTD_CCtx_refPrefix_advanced() to alter dictionary interpretation. */
+ZSTDLIB_API size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx,
+ const void* prefix, size_t prefixSize);
+
+/*! ZSTD_DCtx_loadDictionary() :
+ * Create an internal DDict from dict buffer,
+ * to be used to decompress next frames.
+ * The dictionary remains valid for all future frames, until explicitly invalidated.
+ * @result : 0, or an error code (which can be tested with ZSTD_isError()).
+ * Special : Adding a NULL (or 0-size) dictionary invalidates any previous dictionary,
+ * meaning "return to no-dictionary mode".
+ * Note 1 : Loading a dictionary involves building tables,
+ * which has a non-negligible impact on CPU usage and latency.
+ * It's recommended to "load once, use many times", to amortize the cost
+ * Note 2 :`dict` content will be copied internally, so `dict` can be released after loading.
+ * Use ZSTD_DCtx_loadDictionary_byReference() to reference dictionary content instead.
+ * Note 3 : Use ZSTD_DCtx_loadDictionary_advanced() to take control of
+ * how dictionary content is loaded and interpreted.
+ */
+ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize);
+
+/*! ZSTD_DCtx_refDDict() :
+ * Reference a prepared dictionary, to be used to decompress next frames.
+ * The dictionary remains active for decompression of future frames using same DCtx.
+ * @result : 0, or an error code (which can be tested with ZSTD_isError()).
+ * Note 1 : Currently, only one dictionary can be managed.
+ * Referencing a new dictionary effectively "discards" any previous one.
+ * Special: referencing a NULL DDict means "return to no-dictionary mode".
+ * Note 2 : DDict is just referenced, its lifetime must outlive its usage from DCtx.
+ */
+ZSTDLIB_API size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict);
+
+/*! ZSTD_DCtx_refPrefix() :
+ * Reference a prefix (single-usage dictionary) to decompress next frame.
+ * This is the reverse operation of ZSTD_CCtx_refPrefix(),
+ * and must use the same prefix as the one used during compression.
+ * Prefix is **only used once**. Reference is discarded at end of frame.
+ * End of frame is reached when ZSTD_decompressStream() returns 0.
+ * @result : 0, or an error code (which can be tested with ZSTD_isError()).
+ * Note 1 : Adding any prefix (including NULL) invalidates any previously set prefix or dictionary
+ * Note 2 : Prefix buffer is referenced. It **must** outlive decompression.
+ * Prefix buffer must remain unmodified up to the end of frame,
+ * reached when ZSTD_decompressStream() returns 0.
+ * Note 3 : By default, the prefix is treated as raw content (ZSTD_dm_rawContent).
+ * Use ZSTD_CCtx_refPrefix_advanced() to alter dictMode (Experimental section)
+ * Note 4 : Referencing a raw content prefix has almost no cpu nor memory cost.
+ * A full dictionary is more costly, as it requires building tables.
+ */
+ZSTDLIB_API size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx,
+ const void* prefix, size_t prefixSize);
+
+/* === Memory management === */
+
+/*! ZSTD_sizeof_*() :
+ * These functions give the _current_ memory usage of selected object.
+ * Note that object memory usage can evolve (increase or decrease) over time. */
+ZSTDLIB_API size_t ZSTD_sizeof_CCtx(const ZSTD_CCtx* cctx);
+ZSTDLIB_API size_t ZSTD_sizeof_DCtx(const ZSTD_DCtx* dctx);
+ZSTDLIB_API size_t ZSTD_sizeof_CStream(const ZSTD_CStream* zcs);
+ZSTDLIB_API size_t ZSTD_sizeof_DStream(const ZSTD_DStream* zds);
+ZSTDLIB_API size_t ZSTD_sizeof_CDict(const ZSTD_CDict* cdict);
+ZSTDLIB_API size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict);
+
#endif /* ZSTD_H_235446 */
-
-/****************************************************************************************
- * START OF ADVANCED AND EXPERIMENTAL FUNCTIONS
- * The definitions in this section are considered experimental.
- * They should never be used with a dynamic library, as prototypes may change in the future.
+/* **************************************************************************************
+ * ADVANCED AND EXPERIMENTAL FUNCTIONS
+ ****************************************************************************************
+ * The definitions in the following section are considered experimental.
* They are provided for advanced scenarios.
+ * They should never be used with a dynamic library, as prototypes may change in the future.
* Use them only in association with static linking.
* ***************************************************************************************/
#if defined(ZSTD_STATIC_LINKING_ONLY) && !defined(ZSTD_H_ZSTD_STATIC_LINKING_ONLY)
#define ZSTD_H_ZSTD_STATIC_LINKING_ONLY
-/* --- Constants ---*/
-#define ZSTD_MAGICNUMBER 0xFD2FB528 /* >= v0.8.0 */
-#define ZSTD_MAGIC_SKIPPABLE_START 0x184D2A50U
-#define ZSTD_MAGIC_DICTIONARY 0xEC30A437 /* >= v0.7.0 */
+/****************************************************************************************
+ * experimental API (static linking only)
+ ****************************************************************************************
+ * The following symbols and constants
+ * are not planned to join "stable API" status in the near future.
+ * They can still change in future versions.
+ * Some of them are planned to remain in the static_only section indefinitely.
+ * Some of them might be removed in the future (especially when redundant with existing stable functions)
+ * ***************************************************************************************/
-#define ZSTD_WINDOWLOG_MAX_32 30
-#define ZSTD_WINDOWLOG_MAX_64 31
-#define ZSTD_WINDOWLOG_MAX ((unsigned)(sizeof(size_t) == 4 ? ZSTD_WINDOWLOG_MAX_32 : ZSTD_WINDOWLOG_MAX_64))
-#define ZSTD_WINDOWLOG_MIN 10
-#define ZSTD_HASHLOG_MAX ((ZSTD_WINDOWLOG_MAX < 30) ? ZSTD_WINDOWLOG_MAX : 30)
-#define ZSTD_HASHLOG_MIN 6
-#define ZSTD_CHAINLOG_MAX_32 29
-#define ZSTD_CHAINLOG_MAX_64 30
-#define ZSTD_CHAINLOG_MAX ((unsigned)(sizeof(size_t) == 4 ? ZSTD_CHAINLOG_MAX_32 : ZSTD_CHAINLOG_MAX_64))
-#define ZSTD_CHAINLOG_MIN ZSTD_HASHLOG_MIN
-#define ZSTD_HASHLOG3_MAX 17
-#define ZSTD_SEARCHLOG_MAX (ZSTD_WINDOWLOG_MAX-1)
-#define ZSTD_SEARCHLOG_MIN 1
-#define ZSTD_SEARCHLENGTH_MAX 7 /* only for ZSTD_fast, other strategies are limited to 6 */
-#define ZSTD_SEARCHLENGTH_MIN 3 /* only for ZSTD_btopt, other strategies are limited to 4 */
-#define ZSTD_TARGETLENGTH_MIN 1 /* only used by btopt, btultra and btfast */
-#define ZSTD_LDM_MINMATCH_MIN 4
-#define ZSTD_LDM_MINMATCH_MAX 4096
-#define ZSTD_LDM_BUCKETSIZELOG_MAX 8
-
-#define ZSTD_FRAMEHEADERSIZE_PREFIX 5 /* minimum input size to know frame header size */
+#define ZSTD_FRAMEHEADERSIZE_PREFIX 5 /* minimum input size required to query frame header size */
#define ZSTD_FRAMEHEADERSIZE_MIN 6
-#define ZSTD_FRAMEHEADERSIZE_MAX 18 /* for static allocation */
-static const size_t ZSTD_frameHeaderSize_prefix = ZSTD_FRAMEHEADERSIZE_PREFIX;
-static const size_t ZSTD_frameHeaderSize_min = ZSTD_FRAMEHEADERSIZE_MIN;
-static const size_t ZSTD_frameHeaderSize_max = ZSTD_FRAMEHEADERSIZE_MAX;
-static const size_t ZSTD_skippableHeaderSize = 8; /* magic number + skippable frame length */
+#define ZSTD_FRAMEHEADERSIZE_MAX 18 /* can be useful for static allocation */
+#define ZSTD_SKIPPABLEHEADERSIZE 8
+
+/* compression parameter bounds */
+#define ZSTD_WINDOWLOG_MAX_32 30
+#define ZSTD_WINDOWLOG_MAX_64 31
+#define ZSTD_WINDOWLOG_MAX ((int)(sizeof(size_t) == 4 ? ZSTD_WINDOWLOG_MAX_32 : ZSTD_WINDOWLOG_MAX_64))
+#define ZSTD_WINDOWLOG_MIN 10
+#define ZSTD_HASHLOG_MAX ((ZSTD_WINDOWLOG_MAX < 30) ? ZSTD_WINDOWLOG_MAX : 30)
+#define ZSTD_HASHLOG_MIN 6
+#define ZSTD_CHAINLOG_MAX_32 29
+#define ZSTD_CHAINLOG_MAX_64 30
+#define ZSTD_CHAINLOG_MAX ((int)(sizeof(size_t) == 4 ? ZSTD_CHAINLOG_MAX_32 : ZSTD_CHAINLOG_MAX_64))
+#define ZSTD_CHAINLOG_MIN ZSTD_HASHLOG_MIN
+#define ZSTD_SEARCHLOG_MAX (ZSTD_WINDOWLOG_MAX-1)
+#define ZSTD_SEARCHLOG_MIN 1
+#define ZSTD_MINMATCH_MAX 7 /* only for ZSTD_fast, other strategies are limited to 6 */
+#define ZSTD_MINMATCH_MIN 3 /* only for ZSTD_btopt+, faster strategies are limited to 4 */
+#define ZSTD_TARGETLENGTH_MAX ZSTD_BLOCKSIZE_MAX
+#define ZSTD_TARGETLENGTH_MIN 0 /* note : comparing this constant to an unsigned results in a tautological test */
+#define ZSTD_STRATEGY_MIN ZSTD_fast
+#define ZSTD_STRATEGY_MAX ZSTD_btultra2
-/*--- Advanced types ---*/
-typedef enum { ZSTD_fast=1, ZSTD_dfast, ZSTD_greedy, ZSTD_lazy, ZSTD_lazy2,
- ZSTD_btlazy2, ZSTD_btopt, ZSTD_btultra } ZSTD_strategy; /* from faster to stronger */
+#define ZSTD_OVERLAPLOG_MIN 0
+#define ZSTD_OVERLAPLOG_MAX 9
+
+#define ZSTD_WINDOWLOG_LIMIT_DEFAULT 27 /* by default, the streaming decoder will refuse any frame
+ * requiring larger than (1<<ZSTD_WINDOWLOG_LIMIT_DEFAULT) window size,
+ * to preserve host's memory from unreasonable requirements.
+ * This limit can be overridden using ZSTD_DCtx_setParameter(,ZSTD_d_windowLogMax,).
+ * The limit does not apply for one-pass decoders (such as ZSTD_decompress()), since no additional memory is allocated */
+
+
+/* LDM parameter bounds */
+#define ZSTD_LDM_HASHLOG_MIN ZSTD_HASHLOG_MIN
+#define ZSTD_LDM_HASHLOG_MAX ZSTD_HASHLOG_MAX
+#define ZSTD_LDM_MINMATCH_MIN 4
+#define ZSTD_LDM_MINMATCH_MAX 4096
+#define ZSTD_LDM_BUCKETSIZELOG_MIN 1
+#define ZSTD_LDM_BUCKETSIZELOG_MAX 8
+#define ZSTD_LDM_HASHRATELOG_MIN 0
+#define ZSTD_LDM_HASHRATELOG_MAX (ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN)
+
+/* Advanced parameter bounds */
+#define ZSTD_TARGETCBLOCKSIZE_MIN 64
+#define ZSTD_TARGETCBLOCKSIZE_MAX ZSTD_BLOCKSIZE_MAX
+
+/* internal */
+#define ZSTD_HASHLOG3_MAX 17
+
+
+/* --- Advanced types --- */
+
+typedef struct ZSTD_CCtx_params_s ZSTD_CCtx_params;
typedef struct {
- unsigned windowLog; /**< largest match distance : larger == more compression, more memory needed during decompression */
- unsigned chainLog; /**< fully searched segment : larger == more compression, slower, more memory (useless for fast) */
- unsigned hashLog; /**< dispatch table : larger == faster, more memory */
- unsigned searchLog; /**< nb of searches : larger == more compression, slower */
- unsigned searchLength; /**< match length searched : larger == faster decompression, sometimes less compression */
- unsigned targetLength; /**< acceptable match size for optimal parser (only) : larger == more compression, slower */
- ZSTD_strategy strategy;
+ unsigned windowLog; /**< largest match distance : larger == more compression, more memory needed during decompression */
+ unsigned chainLog; /**< fully searched segment : larger == more compression, slower, more memory (useless for fast) */
+ unsigned hashLog; /**< dispatch table : larger == faster, more memory */
+ unsigned searchLog; /**< nb of searches : larger == more compression, slower */
+ unsigned minMatch; /**< match length searched : larger == faster decompression, sometimes less compression */
+ unsigned targetLength; /**< acceptable match size for optimal parser (only) : larger == more compression, slower */
+ ZSTD_strategy strategy; /**< see ZSTD_strategy definition above */
} ZSTD_compressionParameters;
typedef struct {
- unsigned contentSizeFlag; /**< 1: content size will be in frame header (when known) */
- unsigned checksumFlag; /**< 1: generate a 32-bits checksum at end of frame, for error detection */
- unsigned noDictIDFlag; /**< 1: no dictID will be saved into frame header (if dictionary compression) */
+ int contentSizeFlag; /**< 1: content size will be in frame header (when known) */
+ int checksumFlag; /**< 1: generate a 32-bits checksum using XXH64 algorithm at end of frame, for error detection */
+ int noDictIDFlag; /**< 1: no dictID will be saved into frame header (dictID is only useful for dictionary compression) */
} ZSTD_frameParameters;
typedef struct {
@@ -429,37 +1093,83 @@
ZSTD_frameParameters fParams;
} ZSTD_parameters;
-typedef struct ZSTD_CCtx_params_s ZSTD_CCtx_params;
-
typedef enum {
- ZSTD_dct_auto=0, /* dictionary is "full" when starting with ZSTD_MAGIC_DICTIONARY, otherwise it is "rawContent" */
- ZSTD_dct_rawContent, /* ensures dictionary is always loaded as rawContent, even if it starts with ZSTD_MAGIC_DICTIONARY */
- ZSTD_dct_fullDict /* refuses to load a dictionary if it does not respect Zstandard's specification */
+ ZSTD_dct_auto = 0, /* dictionary is "full" when starting with ZSTD_MAGIC_DICTIONARY, otherwise it is "rawContent" */
+ ZSTD_dct_rawContent = 1, /* ensures dictionary is always loaded as rawContent, even if it starts with ZSTD_MAGIC_DICTIONARY */
+ ZSTD_dct_fullDict = 2 /* refuses to load a dictionary if it does not respect Zstandard's specification, starting with ZSTD_MAGIC_DICTIONARY */
} ZSTD_dictContentType_e;
typedef enum {
- ZSTD_dlm_byCopy = 0, /**< Copy dictionary content internally */
- ZSTD_dlm_byRef, /**< Reference dictionary content -- the dictionary buffer must outlive its users. */
+ ZSTD_dlm_byCopy = 0, /**< Copy dictionary content internally */
+ ZSTD_dlm_byRef = 1, /**< Reference dictionary content -- the dictionary buffer must outlive its users. */
} ZSTD_dictLoadMethod_e;
+typedef enum {
+ /* Opened question : should we have a format ZSTD_f_auto ?
+ * Today, it would mean exactly the same as ZSTD_f_zstd1.
+ * But, in the future, should several formats become supported,
+ * on the compression side, it would mean "default format".
+ * On the decompression side, it would mean "automatic format detection",
+ * so that ZSTD_f_zstd1 would mean "accept *only* zstd frames".
+ * Since meaning is a little different, another option could be to define different enums for compression and decompression.
+ * This question could be kept for later, when there are actually multiple formats to support,
+ * but there is also the question of pinning enum values, and pinning value `0` is especially important */
+ ZSTD_f_zstd1 = 0, /* zstd frame format, specified in zstd_compression_format.md (default) */
+ ZSTD_f_zstd1_magicless = 1, /* Variant of zstd frame format, without initial 4-bytes magic number.
+ * Useful to save 4 bytes per generated frame.
+ * Decoder cannot recognise automatically this format, requiring this instruction. */
+} ZSTD_format_e;
+
+typedef enum {
+ /* Note: this enum and the behavior it controls are effectively internal
+ * implementation details of the compressor. They are expected to continue
+ * to evolve and should be considered only in the context of extremely
+ * advanced performance tuning.
+ *
+ * Zstd currently supports the use of a CDict in two ways:
+ *
+ * - The contents of the CDict can be copied into the working context. This
+ * means that the compression can search both the dictionary and input
+ * while operating on a single set of internal tables. This makes
+ * the compression faster per-byte of input. However, the initial copy of
+ * the CDict's tables incurs a fixed cost at the beginning of the
+ * compression. For small compressions (< 8 KB), that copy can dominate
+ * the cost of the compression.
+ *
+ * - The CDict's tables can be used in-place. In this model, compression is
+ * slower per input byte, because the compressor has to search two sets of
+ * tables. However, this model incurs no start-up cost (as long as the
+ * working context's tables can be reused). For small inputs, this can be
+ * faster than copying the CDict's tables.
+ *
+ * Zstd has a simple internal heuristic that selects which strategy to use
+ * at the beginning of a compression. However, if experimentation shows that
+ * Zstd is making poor choices, it is possible to override that choice with
+ * this enum.
+ */
+ ZSTD_dictDefaultAttach = 0, /* Use the default heuristic. */
+ ZSTD_dictForceAttach = 1, /* Never copy the dictionary. */
+ ZSTD_dictForceCopy = 2, /* Always copy the dictionary. */
+} ZSTD_dictAttachPref_e;
+
+typedef enum {
+ ZSTD_lcm_auto = 0, /**< Automatically determine the compression mode based on the compression level.
+ * Negative compression levels will be uncompressed, and positive compression
+ * levels will be compressed. */
+ ZSTD_lcm_huffman = 1, /**< Always attempt Huffman compression. Uncompressed literals will still be
+ * emitted if Huffman compression is not profitable. */
+ ZSTD_lcm_uncompressed = 2, /**< Always emit uncompressed literals. */
+} ZSTD_literalCompressionMode_e;
/***************************************
* Frame size functions
***************************************/
-/*! ZSTD_findFrameCompressedSize() :
- * `src` should point to the start of a ZSTD encoded frame or skippable frame
- * `srcSize` must be >= first frame size
- * @return : the compressed size of the first frame starting at `src`,
- * suitable to pass to `ZSTD_decompress` or similar,
- * or an error code if input is invalid */
-ZSTDLIB_API size_t ZSTD_findFrameCompressedSize(const void* src, size_t srcSize);
-
/*! ZSTD_findDecompressedSize() :
- * `src` should point the start of a series of ZSTD encoded and/or skippable frames
+ * `src` should point to the start of a series of ZSTD encoded and/or skippable frames
* `srcSize` must be the _exact_ size of this series
- * (i.e. there should be a frame boundary exactly at `srcSize` bytes after `src`)
+ * (i.e. there should be a frame boundary at `src + srcSize`)
* @return : - decompressed size of all data in all successive frames
* - if the decompressed size cannot be determined: ZSTD_CONTENTSIZE_UNKNOWN
* - if an error occurred: ZSTD_CONTENTSIZE_ERROR
@@ -479,10 +1189,25 @@
* however it does mean that all frame data must be present and valid. */
ZSTDLIB_API unsigned long long ZSTD_findDecompressedSize(const void* src, size_t srcSize);
+/*! ZSTD_decompressBound() :
+ * `src` should point to the start of a series of ZSTD encoded and/or skippable frames
+ * `srcSize` must be the _exact_ size of this series
+ * (i.e. there should be a frame boundary at `src + srcSize`)
+ * @return : - upper-bound for the decompressed size of all data in all successive frames
+ * - if an error occured: ZSTD_CONTENTSIZE_ERROR
+ *
+ * note 1 : an error can occur if `src` contains an invalid or incorrectly formatted frame.
+ * note 2 : the upper-bound is exact when the decompressed size field is available in every ZSTD encoded frame of `src`.
+ * in this case, `ZSTD_findDecompressedSize` and `ZSTD_decompressBound` return the same value.
+ * note 3 : when the decompressed size field isn't available, the upper-bound for that frame is calculated by:
+ * upper-bound = # blocks * min(128 KB, Window_Size)
+ */
+ZSTDLIB_API unsigned long long ZSTD_decompressBound(const void* src, size_t srcSize);
+
/*! ZSTD_frameHeaderSize() :
-* `src` should point to the start of a ZSTD frame
-* `srcSize` must be >= ZSTD_frameHeaderSize_prefix.
-* @return : size of the Frame Header */
+ * srcSize must be >= ZSTD_FRAMEHEADERSIZE_PREFIX.
+ * @return : size of the Frame Header,
+ * or an error code (if srcSize is too small) */
ZSTDLIB_API size_t ZSTD_frameHeaderSize(const void* src, size_t srcSize);
@@ -490,16 +1215,6 @@
* Memory management
***************************************/
-/*! ZSTD_sizeof_*() :
- * These functions give the current memory usage of selected object.
- * Object memory usage can evolve when re-used. */
-ZSTDLIB_API size_t ZSTD_sizeof_CCtx(const ZSTD_CCtx* cctx);
-ZSTDLIB_API size_t ZSTD_sizeof_DCtx(const ZSTD_DCtx* dctx);
-ZSTDLIB_API size_t ZSTD_sizeof_CStream(const ZSTD_CStream* zcs);
-ZSTDLIB_API size_t ZSTD_sizeof_DStream(const ZSTD_DStream* zds);
-ZSTDLIB_API size_t ZSTD_sizeof_CDict(const ZSTD_CDict* cdict);
-ZSTDLIB_API size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict);
-
/*! ZSTD_estimate*() :
* These functions make it possible to estimate memory usage
* of a future {D,C}Ctx, before its creation.
@@ -507,7 +1222,7 @@
* It will also consider src size to be arbitrarily "large", which is worst case.
* If srcSize is known to always be small, ZSTD_estimateCCtxSize_usingCParams() can provide a tighter estimation.
* ZSTD_estimateCCtxSize_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel.
- * ZSTD_estimateCCtxSize_usingCCtxParams() can be used in tandem with ZSTD_CCtxParam_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_p_nbWorkers is >= 1.
+ * ZSTD_estimateCCtxSize_usingCCtxParams() can be used in tandem with ZSTD_CCtxParams_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_c_nbWorkers is >= 1.
* Note : CCtx size estimation is only correct for single-threaded compression. */
ZSTDLIB_API size_t ZSTD_estimateCCtxSize(int compressionLevel);
ZSTDLIB_API size_t ZSTD_estimateCCtxSize_usingCParams(ZSTD_compressionParameters cParams);
@@ -519,7 +1234,7 @@
* It will also consider src size to be arbitrarily "large", which is worst case.
* If srcSize is known to always be small, ZSTD_estimateCStreamSize_usingCParams() can provide a tighter estimation.
* ZSTD_estimateCStreamSize_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel.
- * ZSTD_estimateCStreamSize_usingCCtxParams() can be used in tandem with ZSTD_CCtxParam_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_p_nbWorkers is >= 1.
+ * ZSTD_estimateCStreamSize_usingCCtxParams() can be used in tandem with ZSTD_CCtxParams_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_c_nbWorkers is >= 1.
* Note : CStream size estimation is only correct for single-threaded compression.
* ZSTD_DStream memory budget depends on window Size.
* This information can be passed manually, using ZSTD_estimateDStreamSize,
@@ -582,6 +1297,7 @@
ZSTD_dictLoadMethod_e dictLoadMethod,
ZSTD_dictContentType_e dictContentType);
+
/*! Custom memory allocation :
* These prototypes make it possible to pass your own allocation/free functions.
* ZSTD_customMem is provided at creation time, using ZSTD_create*_advanced() variants listed below.
@@ -616,46 +1332,201 @@
/*! ZSTD_createCDict_byReference() :
* Create a digested dictionary for compression
- * Dictionary content is simply referenced, and therefore stays in dictBuffer.
- * It is important that dictBuffer outlives CDict, it must remain read accessible throughout the lifetime of CDict */
+ * Dictionary content is just referenced, not duplicated.
+ * As a consequence, `dictBuffer` **must** outlive CDict,
+ * and its content must remain unmodified throughout the lifetime of CDict. */
ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict_byReference(const void* dictBuffer, size_t dictSize, int compressionLevel);
/*! ZSTD_getCParams() :
-* @return ZSTD_compressionParameters structure for a selected compression level and estimated srcSize.
-* `estimatedSrcSize` value is optional, select 0 if not known */
+ * @return ZSTD_compressionParameters structure for a selected compression level and estimated srcSize.
+ * `estimatedSrcSize` value is optional, select 0 if not known */
ZSTDLIB_API ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize);
/*! ZSTD_getParams() :
-* same as ZSTD_getCParams(), but @return a full `ZSTD_parameters` object instead of sub-component `ZSTD_compressionParameters`.
-* All fields of `ZSTD_frameParameters` are set to default : contentSize=1, checksum=0, noDictID=0 */
+ * same as ZSTD_getCParams(), but @return a full `ZSTD_parameters` object instead of sub-component `ZSTD_compressionParameters`.
+ * All fields of `ZSTD_frameParameters` are set to default : contentSize=1, checksum=0, noDictID=0 */
ZSTDLIB_API ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize);
/*! ZSTD_checkCParams() :
-* Ensure param values remain within authorized range */
+ * Ensure param values remain within authorized range.
+ * @return 0 on success, or an error code (can be checked with ZSTD_isError()) */
ZSTDLIB_API size_t ZSTD_checkCParams(ZSTD_compressionParameters params);
/*! ZSTD_adjustCParams() :
* optimize params for a given `srcSize` and `dictSize`.
- * both values are optional, select `0` if unknown. */
+ * `srcSize` can be unknown, in which case use ZSTD_CONTENTSIZE_UNKNOWN.
+ * `dictSize` must be `0` when there is no dictionary.
+ * cPar can be invalid : all parameters will be clamped within valid range in the @return struct.
+ * This function never fails (wide contract) */
ZSTDLIB_API ZSTD_compressionParameters ZSTD_adjustCParams(ZSTD_compressionParameters cPar, unsigned long long srcSize, size_t dictSize);
/*! ZSTD_compress_advanced() :
-* Same as ZSTD_compress_usingDict(), with fine-tune control over each compression parameter */
-ZSTDLIB_API size_t ZSTD_compress_advanced (ZSTD_CCtx* cctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- const void* dict,size_t dictSize,
- ZSTD_parameters params);
+ * Same as ZSTD_compress_usingDict(), with fine-tune control over compression parameters (by structure) */
+ZSTDLIB_API size_t ZSTD_compress_advanced(ZSTD_CCtx* cctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ const void* dict,size_t dictSize,
+ ZSTD_parameters params);
/*! ZSTD_compress_usingCDict_advanced() :
-* Same as ZSTD_compress_usingCDict(), with fine-tune control over frame parameters */
+ * Same as ZSTD_compress_usingCDict(), with fine-tune control over frame parameters */
ZSTDLIB_API size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- const ZSTD_CDict* cdict, ZSTD_frameParameters fParams);
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ const ZSTD_CDict* cdict,
+ ZSTD_frameParameters fParams);
-/*--- Advanced decompression functions ---*/
+/*! ZSTD_CCtx_loadDictionary_byReference() :
+ * Same as ZSTD_CCtx_loadDictionary(), but dictionary content is referenced, instead of being copied into CCtx.
+ * It saves some memory, but also requires that `dict` outlives its usage within `cctx` */
+ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary_byReference(ZSTD_CCtx* cctx, const void* dict, size_t dictSize);
+
+/*! ZSTD_CCtx_loadDictionary_advanced() :
+ * Same as ZSTD_CCtx_loadDictionary(), but gives finer control over
+ * how to load the dictionary (by copy ? by reference ?)
+ * and how to interpret it (automatic ? force raw mode ? full mode only ?) */
+ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType);
+
+/*! ZSTD_CCtx_refPrefix_advanced() :
+ * Same as ZSTD_CCtx_refPrefix(), but gives finer control over
+ * how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?) */
+ZSTDLIB_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType);
+
+/* === experimental parameters === */
+/* these parameters can be used with ZSTD_setParameter()
+ * they are not guaranteed to remain supported in the future */
+
+ /* Enables rsyncable mode,
+ * which makes compressed files more rsync friendly
+ * by adding periodic synchronization points to the compressed data.
+ * The target average block size is ZSTD_c_jobSize / 2.
+ * It's possible to modify the job size to increase or decrease
+ * the granularity of the synchronization point.
+ * Once the jobSize is smaller than the window size,
+ * it will result in compression ratio degradation.
+ * NOTE 1: rsyncable mode only works when multithreading is enabled.
+ * NOTE 2: rsyncable performs poorly in combination with long range mode,
+ * since it will decrease the effectiveness of synchronization points,
+ * though mileage may vary.
+ * NOTE 3: Rsyncable mode limits maximum compression speed to ~400 MB/s.
+ * If the selected compression level is already running significantly slower,
+ * the overall speed won't be significantly impacted.
+ */
+ #define ZSTD_c_rsyncable ZSTD_c_experimentalParam1
+
+/* Select a compression format.
+ * The value must be of type ZSTD_format_e.
+ * See ZSTD_format_e enum definition for details */
+#define ZSTD_c_format ZSTD_c_experimentalParam2
+
+/* Force back-reference distances to remain < windowSize,
+ * even when referencing into Dictionary content (default:0) */
+#define ZSTD_c_forceMaxWindow ZSTD_c_experimentalParam3
+
+/* Controls whether the contents of a CDict
+ * are used in place, or copied into the working context.
+ * Accepts values from the ZSTD_dictAttachPref_e enum.
+ * See the comments on that enum for an explanation of the feature. */
+#define ZSTD_c_forceAttachDict ZSTD_c_experimentalParam4
+
+/* Controls how the literals are compressed (default is auto).
+ * The value must be of type ZSTD_literalCompressionMode_e.
+ * See ZSTD_literalCompressionMode_t enum definition for details.
+ */
+#define ZSTD_c_literalCompressionMode ZSTD_c_experimentalParam5
+
+/* Tries to fit compressed block size to be around targetCBlockSize.
+ * No target when targetCBlockSize == 0.
+ * There is no guarantee on compressed block size (default:0) */
+#define ZSTD_c_targetCBlockSize ZSTD_c_experimentalParam6
+
+/*! ZSTD_CCtx_getParameter() :
+ * Get the requested compression parameter value, selected by enum ZSTD_cParameter,
+ * and store it into int* value.
+ * @return : 0, or an error code (which can be tested with ZSTD_isError()).
+ */
+ZSTDLIB_API size_t ZSTD_CCtx_getParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int* value);
+
+
+/*! ZSTD_CCtx_params :
+ * Quick howto :
+ * - ZSTD_createCCtxParams() : Create a ZSTD_CCtx_params structure
+ * - ZSTD_CCtxParams_setParameter() : Push parameters one by one into
+ * an existing ZSTD_CCtx_params structure.
+ * This is similar to
+ * ZSTD_CCtx_setParameter().
+ * - ZSTD_CCtx_setParametersUsingCCtxParams() : Apply parameters to
+ * an existing CCtx.
+ * These parameters will be applied to
+ * all subsequent frames.
+ * - ZSTD_compressStream2() : Do compression using the CCtx.
+ * - ZSTD_freeCCtxParams() : Free the memory.
+ *
+ * This can be used with ZSTD_estimateCCtxSize_advanced_usingCCtxParams()
+ * for static allocation of CCtx for single-threaded compression.
+ */
+ZSTDLIB_API ZSTD_CCtx_params* ZSTD_createCCtxParams(void);
+ZSTDLIB_API size_t ZSTD_freeCCtxParams(ZSTD_CCtx_params* params);
+
+/*! ZSTD_CCtxParams_reset() :
+ * Reset params to default values.
+ */
+ZSTDLIB_API size_t ZSTD_CCtxParams_reset(ZSTD_CCtx_params* params);
+
+/*! ZSTD_CCtxParams_init() :
+ * Initializes the compression parameters of cctxParams according to
+ * compression level. All other parameters are reset to their default values.
+ */
+ZSTDLIB_API size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int compressionLevel);
+
+/*! ZSTD_CCtxParams_init_advanced() :
+ * Initializes the compression and frame parameters of cctxParams according to
+ * params. All other parameters are reset to their default values.
+ */
+ZSTDLIB_API size_t ZSTD_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params);
+
+/*! ZSTD_CCtxParams_setParameter() :
+ * Similar to ZSTD_CCtx_setParameter.
+ * Set one compression parameter, selected by enum ZSTD_cParameter.
+ * Parameters must be applied to a ZSTD_CCtx using ZSTD_CCtx_setParametersUsingCCtxParams().
+ * @result : 0, or an error code (which can be tested with ZSTD_isError()).
+ */
+ZSTDLIB_API size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int value);
+
+/*! ZSTD_CCtxParams_getParameter() :
+ * Similar to ZSTD_CCtx_getParameter.
+ * Get the requested value of one compression parameter, selected by enum ZSTD_cParameter.
+ * @result : 0, or an error code (which can be tested with ZSTD_isError()).
+ */
+ZSTDLIB_API size_t ZSTD_CCtxParams_getParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int* value);
+
+/*! ZSTD_CCtx_setParametersUsingCCtxParams() :
+ * Apply a set of ZSTD_CCtx_params to the compression context.
+ * This can be done even after compression is started,
+ * if nbWorkers==0, this will have no impact until a new compression is started.
+ * if nbWorkers>=1, new parameters will be picked up at next job,
+ * with a few restrictions (windowLog, pledgedSrcSize, nbWorkers, jobSize, and overlapLog are not updated).
+ */
+ZSTDLIB_API size_t ZSTD_CCtx_setParametersUsingCCtxParams(
+ ZSTD_CCtx* cctx, const ZSTD_CCtx_params* params);
+
+/*! ZSTD_compressStream2_simpleArgs() :
+ * Same as ZSTD_compressStream2(),
+ * but using only integral types as arguments.
+ * This variant might be helpful for binders from dynamic languages
+ * which have troubles handling structures containing memory pointers.
+ */
+ZSTDLIB_API size_t ZSTD_compressStream2_simpleArgs (
+ ZSTD_CCtx* cctx,
+ void* dst, size_t dstCapacity, size_t* dstPos,
+ const void* src, size_t srcSize, size_t* srcPos,
+ ZSTD_EndDirective endOp);
+
+
+/***************************************
+* Advanced decompression functions
+***************************************/
/*! ZSTD_isFrame() :
* Tells if the content of `buffer` starts with a valid Frame Identifier.
@@ -671,78 +1542,201 @@
* it must remain read accessible throughout the lifetime of DDict */
ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize);
+/*! ZSTD_DCtx_loadDictionary_byReference() :
+ * Same as ZSTD_DCtx_loadDictionary(),
+ * but references `dict` content instead of copying it into `dctx`.
+ * This saves memory if `dict` remains around.,
+ * However, it's imperative that `dict` remains accessible (and unmodified) while being used, so it must outlive decompression. */
+ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary_byReference(ZSTD_DCtx* dctx, const void* dict, size_t dictSize);
-/*! ZSTD_getDictID_fromDict() :
- * Provides the dictID stored within dictionary.
- * if @return == 0, the dictionary is not conformant with Zstandard specification.
- * It can still be loaded, but as a content-only dictionary. */
-ZSTDLIB_API unsigned ZSTD_getDictID_fromDict(const void* dict, size_t dictSize);
+/*! ZSTD_DCtx_loadDictionary_advanced() :
+ * Same as ZSTD_DCtx_loadDictionary(),
+ * but gives direct control over
+ * how to load the dictionary (by copy ? by reference ?)
+ * and how to interpret it (automatic ? force raw mode ? full mode only ?). */
+ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType);
-/*! ZSTD_getDictID_fromDDict() :
- * Provides the dictID of the dictionary loaded into `ddict`.
- * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty.
- * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */
-ZSTDLIB_API unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict);
+/*! ZSTD_DCtx_refPrefix_advanced() :
+ * Same as ZSTD_DCtx_refPrefix(), but gives finer control over
+ * how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?) */
+ZSTDLIB_API size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType);
-/*! ZSTD_getDictID_fromFrame() :
- * Provides the dictID required to decompressed the frame stored within `src`.
- * If @return == 0, the dictID could not be decoded.
- * This could for one of the following reasons :
- * - The frame does not require a dictionary to be decoded (most common case).
- * - The frame was built with dictID intentionally removed. Whatever dictionary is necessary is a hidden information.
- * Note : this use case also happens when using a non-conformant dictionary.
- * - `srcSize` is too small, and as a result, the frame header could not be decoded (only possible if `srcSize < ZSTD_FRAMEHEADERSIZE_MAX`).
- * - This is not a Zstandard frame.
- * When identifying the exact failure cause, it's possible to use ZSTD_getFrameHeader(), which will provide a more precise error code. */
-ZSTDLIB_API unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize);
+/*! ZSTD_DCtx_setMaxWindowSize() :
+ * Refuses allocating internal buffers for frames requiring a window size larger than provided limit.
+ * This protects a decoder context from reserving too much memory for itself (potential attack scenario).
+ * This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode.
+ * By default, a decompression context accepts all window sizes <= (1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT)
+ * @return : 0, or an error code (which can be tested using ZSTD_isError()).
+ */
+ZSTDLIB_API size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize);
+
+/* ZSTD_d_format
+ * experimental parameter,
+ * allowing selection between ZSTD_format_e input compression formats
+ */
+#define ZSTD_d_format ZSTD_d_experimentalParam1
+
+/*! ZSTD_DCtx_setFormat() :
+ * Instruct the decoder context about what kind of data to decode next.
+ * This instruction is mandatory to decode data without a fully-formed header,
+ * such ZSTD_f_zstd1_magicless for example.
+ * @return : 0, or an error code (which can be tested using ZSTD_isError()). */
+ZSTDLIB_API size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format);
+
+/*! ZSTD_decompressStream_simpleArgs() :
+ * Same as ZSTD_decompressStream(),
+ * but using only integral types as arguments.
+ * This can be helpful for binders from dynamic languages
+ * which have troubles handling structures containing memory pointers.
+ */
+ZSTDLIB_API size_t ZSTD_decompressStream_simpleArgs (
+ ZSTD_DCtx* dctx,
+ void* dst, size_t dstCapacity, size_t* dstPos,
+ const void* src, size_t srcSize, size_t* srcPos);
/********************************************************************
* Advanced streaming functions
+* Warning : most of these functions are now redundant with the Advanced API.
+* Once Advanced API reaches "stable" status,
+* redundant functions will be deprecated, and then at some point removed.
********************************************************************/
/*===== Advanced Streaming compression functions =====*/
-ZSTDLIB_API size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, int compressionLevel, unsigned long long pledgedSrcSize); /**< pledgedSrcSize must be correct. If it is not known at init time, use ZSTD_CONTENTSIZE_UNKNOWN. Note that, for compatibility with older programs, "0" also disables frame content size field. It may be enabled in the future. */
-ZSTDLIB_API size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel); /**< creates of an internal CDict (incompatible with static CCtx), except if dict == NULL or dictSize < 8, in which case no dict is used. Note: dict is loaded with ZSTD_dm_auto (treated as a full zstd dictionary if it begins with ZSTD_MAGIC_DICTIONARY, else as raw content) and ZSTD_dlm_byCopy.*/
+/**! ZSTD_initCStream_srcSize() :
+ * This function is deprecated, and equivalent to:
+ * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
+ * ZSTD_CCtx_refCDict(zcs, NULL); // clear the dictionary (if any)
+ * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel);
+ * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize);
+ *
+ * pledgedSrcSize must be correct. If it is not known at init time, use
+ * ZSTD_CONTENTSIZE_UNKNOWN. Note that, for compatibility with older programs,
+ * "0" also disables frame content size field. It may be enabled in the future.
+ */
+ZSTDLIB_API size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, int compressionLevel, unsigned long long pledgedSrcSize);
+/**! ZSTD_initCStream_usingDict() :
+ * This function is deprecated, and is equivalent to:
+ * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
+ * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel);
+ * ZSTD_CCtx_loadDictionary(zcs, dict, dictSize);
+ *
+ * Creates of an internal CDict (incompatible with static CCtx), except if
+ * dict == NULL or dictSize < 8, in which case no dict is used.
+ * Note: dict is loaded with ZSTD_dm_auto (treated as a full zstd dictionary if
+ * it begins with ZSTD_MAGIC_DICTIONARY, else as raw content) and ZSTD_dlm_byCopy.
+ */
+ZSTDLIB_API size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel);
+/**! ZSTD_initCStream_advanced() :
+ * This function is deprecated, and is approximately equivalent to:
+ * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
+ * ZSTD_CCtx_setZstdParams(zcs, params); // Set the zstd params and leave the rest as-is
+ * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize);
+ * ZSTD_CCtx_loadDictionary(zcs, dict, dictSize);
+ *
+ * pledgedSrcSize must be correct. If srcSize is not known at init time, use
+ * value ZSTD_CONTENTSIZE_UNKNOWN. dict is loaded with ZSTD_dm_auto and ZSTD_dlm_byCopy.
+ */
ZSTDLIB_API size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, const void* dict, size_t dictSize,
- ZSTD_parameters params, unsigned long long pledgedSrcSize); /**< pledgedSrcSize must be correct. If srcSize is not known at init time, use value ZSTD_CONTENTSIZE_UNKNOWN. dict is loaded with ZSTD_dm_auto and ZSTD_dlm_byCopy. */
-ZSTDLIB_API size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict); /**< note : cdict will just be referenced, and must outlive compression session */
-ZSTDLIB_API size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, const ZSTD_CDict* cdict, ZSTD_frameParameters fParams, unsigned long long pledgedSrcSize); /**< same as ZSTD_initCStream_usingCDict(), with control over frame parameters. pledgedSrcSize must be correct. If srcSize is not known at init time, use value ZSTD_CONTENTSIZE_UNKNOWN. */
+ ZSTD_parameters params, unsigned long long pledgedSrcSize);
+/**! ZSTD_initCStream_usingCDict() :
+ * This function is deprecated, and equivalent to:
+ * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
+ * ZSTD_CCtx_refCDict(zcs, cdict);
+ *
+ * note : cdict will just be referenced, and must outlive compression session
+ */
+ZSTDLIB_API size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict);
+/**! ZSTD_initCStream_usingCDict_advanced() :
+ * This function is deprecated, and is approximately equivalent to:
+ * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
+ * ZSTD_CCtx_setZstdFrameParams(zcs, fParams); // Set the zstd frame params and leave the rest as-is
+ * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize);
+ * ZSTD_CCtx_refCDict(zcs, cdict);
+ *
+ * same as ZSTD_initCStream_usingCDict(), with control over frame parameters.
+ * pledgedSrcSize must be correct. If srcSize is not known at init time, use
+ * value ZSTD_CONTENTSIZE_UNKNOWN.
+ */
+ZSTDLIB_API size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, const ZSTD_CDict* cdict, ZSTD_frameParameters fParams, unsigned long long pledgedSrcSize);
/*! ZSTD_resetCStream() :
- * start a new compression job, using same parameters from previous job.
- * This is typically useful to skip dictionary loading stage, since it will re-use it in-place..
+ * This function is deprecated, and is equivalent to:
+ * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
+ * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize);
+ *
+ * start a new frame, using same parameters from previous frame.
+ * This is typically useful to skip dictionary loading stage, since it will re-use it in-place.
* Note that zcs must be init at least once before using ZSTD_resetCStream().
* If pledgedSrcSize is not known at reset time, use macro ZSTD_CONTENTSIZE_UNKNOWN.
* If pledgedSrcSize > 0, its value must be correct, as it will be written in header, and controlled at the end.
* For the time being, pledgedSrcSize==0 is interpreted as "srcSize unknown" for compatibility with older programs,
* but it will change to mean "empty" in future version, so use macro ZSTD_CONTENTSIZE_UNKNOWN instead.
- * @return : 0, or an error code (which can be tested using ZSTD_isError()) */
+ * @return : 0, or an error code (which can be tested using ZSTD_isError())
+ */
ZSTDLIB_API size_t ZSTD_resetCStream(ZSTD_CStream* zcs, unsigned long long pledgedSrcSize);
typedef struct {
- unsigned long long ingested;
- unsigned long long consumed;
- unsigned long long produced;
+ unsigned long long ingested; /* nb input bytes read and buffered */
+ unsigned long long consumed; /* nb input bytes actually compressed */
+ unsigned long long produced; /* nb of compressed bytes generated and buffered */
+ unsigned long long flushed; /* nb of compressed bytes flushed : not provided; can be tracked from caller side */
+ unsigned currentJobID; /* MT only : latest started job nb */
+ unsigned nbActiveWorkers; /* MT only : nb of workers actively compressing at probe time */
} ZSTD_frameProgression;
-/* ZSTD_getFrameProgression():
+/* ZSTD_getFrameProgression() :
* tells how much data has been ingested (read from input)
* consumed (input actually compressed) and produced (output) for current frame.
- * Therefore, (ingested - consumed) is amount of input data buffered internally, not yet compressed.
- * Can report progression inside worker threads (multi-threading and non-blocking mode).
+ * Note : (ingested - consumed) is amount of input data buffered internally, not yet compressed.
+ * Aggregates progression inside active worker threads.
*/
-ZSTD_frameProgression ZSTD_getFrameProgression(const ZSTD_CCtx* cctx);
+ZSTDLIB_API ZSTD_frameProgression ZSTD_getFrameProgression(const ZSTD_CCtx* cctx);
+/*! ZSTD_toFlushNow() :
+ * Tell how many bytes are ready to be flushed immediately.
+ * Useful for multithreading scenarios (nbWorkers >= 1).
+ * Probe the oldest active job, defined as oldest job not yet entirely flushed,
+ * and check its output buffer.
+ * @return : amount of data stored in oldest job and ready to be flushed immediately.
+ * if @return == 0, it means either :
+ * + there is no active job (could be checked with ZSTD_frameProgression()), or
+ * + oldest job is still actively compressing data,
+ * but everything it has produced has also been flushed so far,
+ * therefore flush speed is limited by production speed of oldest job
+ * irrespective of the speed of concurrent (and newer) jobs.
+ */
+ZSTDLIB_API size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx);
/*===== Advanced Streaming decompression functions =====*/
-typedef enum { DStream_p_maxWindowSize } ZSTD_DStreamParameter_e;
-ZSTDLIB_API size_t ZSTD_setDStreamParameter(ZSTD_DStream* zds, ZSTD_DStreamParameter_e paramType, unsigned paramValue); /* obsolete : this API will be removed in a future version */
-ZSTDLIB_API size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize); /**< note: no dictionary will be used if dict == NULL or dictSize < 8 */
-ZSTDLIB_API size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* zds, const ZSTD_DDict* ddict); /**< note : ddict is referenced, it must outlive decompression session */
-ZSTDLIB_API size_t ZSTD_resetDStream(ZSTD_DStream* zds); /**< re-use decompression parameters from previous init; saves dictionary loading */
+/**
+ * This function is deprecated, and is equivalent to:
+ *
+ * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only);
+ * ZSTD_DCtx_loadDictionary(zds, dict, dictSize);
+ *
+ * note: no dictionary will be used if dict == NULL or dictSize < 8
+ */
+ZSTDLIB_API size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize);
+/**
+ * This function is deprecated, and is equivalent to:
+ *
+ * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only);
+ * ZSTD_DCtx_refDDict(zds, ddict);
+ *
+ * note : ddict is referenced, it must outlive decompression session
+ */
+ZSTDLIB_API size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* zds, const ZSTD_DDict* ddict);
+/**
+ * This function is deprecated, and is equivalent to:
+ *
+ * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only);
+ *
+ * re-use decompression parameters from previous init; saves dictionary loading
+ */
+ZSTDLIB_API size_t ZSTD_resetDStream(ZSTD_DStream* zds);
/*********************************************************************
@@ -880,7 +1874,17 @@
unsigned dictID;
unsigned checksumFlag;
} ZSTD_frameHeader;
+
+/*! ZSTD_getFrameHeader() :
+ * decode Frame Header, or requires larger `srcSize`.
+ * @return : 0, `zfhPtr` is correctly filled,
+ * >0, `srcSize` is too small, value is wanted `srcSize` amount,
+ * or an error code, which can be tested using ZSTD_isError() */
ZSTDLIB_API size_t ZSTD_getFrameHeader(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize); /**< doesn't consume input */
+/*! ZSTD_getFrameHeader_advanced() :
+ * same as ZSTD_getFrameHeader(),
+ * with added capability to select a format (like ZSTD_f_zstd1_magicless) */
+ZSTDLIB_API size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format);
ZSTDLIB_API size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize); /**< when frame content size is not known, pass in frameContentSize == ZSTD_CONTENTSIZE_UNKNOWN */
ZSTDLIB_API size_t ZSTD_decompressBegin(ZSTD_DCtx* dctx);
@@ -897,462 +1901,6 @@
-/* ============================================ */
-/** New advanced API (experimental) */
-/* ============================================ */
-
-/* notes on API design :
- * In this proposal, parameters are pushed one by one into an existing context,
- * and then applied on all subsequent compression jobs.
- * When no parameter is ever provided, CCtx is created with compression level ZSTD_CLEVEL_DEFAULT.
- *
- * This API is intended to replace all others advanced / experimental API entry points.
- * But it stands a reasonable chance to become "stable", after a reasonable testing period.
- */
-
-/* note on naming convention :
- * Initially, the API favored names like ZSTD_setCCtxParameter() .
- * In this proposal, convention is changed towards ZSTD_CCtx_setParameter() .
- * The main driver is that it identifies more clearly the target object type.
- * It feels clearer when considering multiple targets :
- * ZSTD_CDict_setParameter() (rather than ZSTD_setCDictParameter())
- * ZSTD_CCtxParams_setParameter() (rather than ZSTD_setCCtxParamsParameter() )
- * etc...
- */
-
-/* note on enum design :
- * All enum will be pinned to explicit values before reaching "stable API" status */
-
-typedef enum {
- /* Opened question : should we have a format ZSTD_f_auto ?
- * Today, it would mean exactly the same as ZSTD_f_zstd1.
- * But, in the future, should several formats become supported,
- * on the compression side, it would mean "default format".
- * On the decompression side, it would mean "automatic format detection",
- * so that ZSTD_f_zstd1 would mean "accept *only* zstd frames".
- * Since meaning is a little different, another option could be to define different enums for compression and decompression.
- * This question could be kept for later, when there are actually multiple formats to support,
- * but there is also the question of pinning enum values, and pinning value `0` is especially important */
- ZSTD_f_zstd1 = 0, /* zstd frame format, specified in zstd_compression_format.md (default) */
- ZSTD_f_zstd1_magicless, /* Variant of zstd frame format, without initial 4-bytes magic number.
- * Useful to save 4 bytes per generated frame.
- * Decoder cannot recognise automatically this format, requiring instructions. */
-} ZSTD_format_e;
-
-typedef enum {
- /* compression format */
- ZSTD_p_format = 10, /* See ZSTD_format_e enum definition.
- * Cast selected format as unsigned for ZSTD_CCtx_setParameter() compatibility. */
-
- /* compression parameters */
- ZSTD_p_compressionLevel=100, /* Update all compression parameters according to pre-defined cLevel table
- * Default level is ZSTD_CLEVEL_DEFAULT==3.
- * Special: value 0 means "do not change cLevel".
- * Note 1 : it's possible to pass a negative compression level by casting it to unsigned type.
- * Note 2 : setting a level sets all default values of other compression parameters.
- * Note 3 : setting compressionLevel automatically updates ZSTD_p_compressLiterals. */
- ZSTD_p_windowLog, /* Maximum allowed back-reference distance, expressed as power of 2.
- * Must be clamped between ZSTD_WINDOWLOG_MIN and ZSTD_WINDOWLOG_MAX.
- * Special: value 0 means "use default windowLog".
- * Note: Using a window size greater than ZSTD_MAXWINDOWSIZE_DEFAULT (default: 2^27)
- * requires explicitly allowing such window size during decompression stage. */
- ZSTD_p_hashLog, /* Size of the probe table, as a power of 2.
- * Resulting table size is (1 << (hashLog+2)).
- * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX.
- * Larger tables improve compression ratio of strategies <= dFast,
- * and improve speed of strategies > dFast.
- * Special: value 0 means "use default hashLog". */
- ZSTD_p_chainLog, /* Size of the full-search table, as a power of 2.
- * Resulting table size is (1 << (chainLog+2)).
- * Larger tables result in better and slower compression.
- * This parameter is useless when using "fast" strategy.
- * Special: value 0 means "use default chainLog". */
- ZSTD_p_searchLog, /* Number of search attempts, as a power of 2.
- * More attempts result in better and slower compression.
- * This parameter is useless when using "fast" and "dFast" strategies.
- * Special: value 0 means "use default searchLog". */
- ZSTD_p_minMatch, /* Minimum size of searched matches (note : repCode matches can be smaller).
- * Larger values make faster compression and decompression, but decrease ratio.
- * Must be clamped between ZSTD_SEARCHLENGTH_MIN and ZSTD_SEARCHLENGTH_MAX.
- * Note that currently, for all strategies < btopt, effective minimum is 4.
- * , for all strategies > fast, effective maximum is 6.
- * Special: value 0 means "use default minMatchLength". */
- ZSTD_p_targetLength, /* Impact of this field depends on strategy.
- * For strategies btopt & btultra:
- * Length of Match considered "good enough" to stop search.
- * Larger values make compression stronger, and slower.
- * For strategy fast:
- * Distance between match sampling.
- * Larger values make compression faster, and weaker.
- * Special: value 0 means "use default targetLength". */
- ZSTD_p_compressionStrategy, /* See ZSTD_strategy enum definition.
- * Cast selected strategy as unsigned for ZSTD_CCtx_setParameter() compatibility.
- * The higher the value of selected strategy, the more complex it is,
- * resulting in stronger and slower compression.
- * Special: value 0 means "use default strategy". */
-
- ZSTD_p_enableLongDistanceMatching=160, /* Enable long distance matching.
- * This parameter is designed to improve compression ratio
- * for large inputs, by finding large matches at long distance.
- * It increases memory usage and window size.
- * Note: enabling this parameter increases ZSTD_p_windowLog to 128 MB
- * except when expressly set to a different value. */
- ZSTD_p_ldmHashLog, /* Size of the table for long distance matching, as a power of 2.
- * Larger values increase memory usage and compression ratio,
- * but decrease compression speed.
- * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX
- * default: windowlog - 7.
- * Special: value 0 means "automatically determine hashlog". */
- ZSTD_p_ldmMinMatch, /* Minimum match size for long distance matcher.
- * Larger/too small values usually decrease compression ratio.
- * Must be clamped between ZSTD_LDM_MINMATCH_MIN and ZSTD_LDM_MINMATCH_MAX.
- * Special: value 0 means "use default value" (default: 64). */
- ZSTD_p_ldmBucketSizeLog, /* Log size of each bucket in the LDM hash table for collision resolution.
- * Larger values improve collision resolution but decrease compression speed.
- * The maximum value is ZSTD_LDM_BUCKETSIZELOG_MAX .
- * Special: value 0 means "use default value" (default: 3). */
- ZSTD_p_ldmHashEveryLog, /* Frequency of inserting/looking up entries in the LDM hash table.
- * Must be clamped between 0 and (ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN).
- * Default is MAX(0, (windowLog - ldmHashLog)), optimizing hash table usage.
- * Larger values improve compression speed.
- * Deviating far from default value will likely result in a compression ratio decrease.
- * Special: value 0 means "automatically determine hashEveryLog". */
-
- /* frame parameters */
- ZSTD_p_contentSizeFlag=200, /* Content size will be written into frame header _whenever known_ (default:1)
- * Content size must be known at the beginning of compression,
- * it is provided using ZSTD_CCtx_setPledgedSrcSize() */
- ZSTD_p_checksumFlag, /* A 32-bits checksum of content is written at end of frame (default:0) */
- ZSTD_p_dictIDFlag, /* When applicable, dictionary's ID is written into frame header (default:1) */
-
- /* multi-threading parameters */
- /* These parameters are only useful if multi-threading is enabled (ZSTD_MULTITHREAD).
- * They return an error otherwise. */
- ZSTD_p_nbWorkers=400, /* Select how many threads will be spawned to compress in parallel.
- * When nbWorkers >= 1, triggers asynchronous mode :
- * ZSTD_compress_generic() consumes some input, flush some output if possible, and immediately gives back control to caller,
- * while compression work is performed in parallel, within worker threads.
- * (note : a strong exception to this rule is when first invocation sets ZSTD_e_end : it becomes a blocking call).
- * More workers improve speed, but also increase memory usage.
- * Default value is `0`, aka "single-threaded mode" : no worker is spawned, compression is performed inside Caller's thread, all invocations are blocking */
- ZSTD_p_jobSize, /* Size of a compression job. This value is enforced only in non-blocking mode.
- * Each compression job is completed in parallel, so this value indirectly controls the nb of active threads.
- * 0 means default, which is dynamically determined based on compression parameters.
- * Job size must be a minimum of overlapSize, or 1 MB, whichever is largest.
- * The minimum size is automatically and transparently enforced */
- ZSTD_p_overlapSizeLog, /* Size of previous input reloaded at the beginning of each job.
- * 0 => no overlap, 6(default) => use 1/8th of windowSize, >=9 => use full windowSize */
-
- /* =================================================================== */
- /* experimental parameters - no stability guaranteed */
- /* =================================================================== */
-
- ZSTD_p_compressLiterals=1000, /* control huffman compression of literals (enabled) by default.
- * disabling it improves speed and decreases compression ratio by a large amount.
- * note : this setting is automatically updated when changing compression level.
- * positive compression levels set ZSTD_p_compressLiterals to 1.
- * negative compression levels set ZSTD_p_compressLiterals to 0. */
-
- ZSTD_p_forceMaxWindow=1100, /* Force back-reference distances to remain < windowSize,
- * even when referencing into Dictionary content (default:0) */
-
-} ZSTD_cParameter;
-
-
-/*! ZSTD_CCtx_setParameter() :
- * Set one compression parameter, selected by enum ZSTD_cParameter.
- * Setting a parameter is generally only possible during frame initialization (before starting compression),
- * except for a few exceptions which can be updated during compression: compressionLevel, hashLog, chainLog, searchLog, minMatch, targetLength and strategy.
- * Note : when `value` is an enum, cast it to unsigned for proper type checking.
- * @result : informational value (typically, value being set clamped correctly),
- * or an error code (which can be tested with ZSTD_isError()). */
-ZSTDLIB_API size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, unsigned value);
-
-/*! ZSTD_CCtx_setPledgedSrcSize() :
- * Total input data size to be compressed as a single frame.
- * This value will be controlled at the end, and result in error if not respected.
- * @result : 0, or an error code (which can be tested with ZSTD_isError()).
- * Note 1 : 0 means zero, empty.
- * In order to mean "unknown content size", pass constant ZSTD_CONTENTSIZE_UNKNOWN.
- * ZSTD_CONTENTSIZE_UNKNOWN is default value for any new compression job.
- * Note 2 : If all data is provided and consumed in a single round,
- * this value is overriden by srcSize instead. */
-ZSTDLIB_API size_t ZSTD_CCtx_setPledgedSrcSize(ZSTD_CCtx* cctx, unsigned long long pledgedSrcSize);
-
-/*! ZSTD_CCtx_loadDictionary() :
- * Create an internal CDict from `dict` buffer.
- * Decompression will have to use same dictionary.
- * @result : 0, or an error code (which can be tested with ZSTD_isError()).
- * Special: Adding a NULL (or 0-size) dictionary invalidates previous dictionary,
- * meaning "return to no-dictionary mode".
- * Note 1 : Dictionary will be used for all future compression jobs.
- * To return to "no-dictionary" situation, load a NULL dictionary
- * Note 2 : Loading a dictionary involves building tables, which are dependent on compression parameters.
- * For this reason, compression parameters cannot be changed anymore after loading a dictionary.
- * It's also a CPU consuming operation, with non-negligible impact on latency.
- * Note 3 :`dict` content will be copied internally.
- * Use ZSTD_CCtx_loadDictionary_byReference() to reference dictionary content instead.
- * In such a case, dictionary buffer must outlive its users.
- * Note 4 : Use ZSTD_CCtx_loadDictionary_advanced()
- * to precisely select how dictionary content must be interpreted. */
-ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary(ZSTD_CCtx* cctx, const void* dict, size_t dictSize);
-ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary_byReference(ZSTD_CCtx* cctx, const void* dict, size_t dictSize);
-ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType);
-
-
-/*! ZSTD_CCtx_refCDict() :
- * Reference a prepared dictionary, to be used for all next compression jobs.
- * Note that compression parameters are enforced from within CDict,
- * and supercede any compression parameter previously set within CCtx.
- * The dictionary will remain valid for future compression jobs using same CCtx.
- * @result : 0, or an error code (which can be tested with ZSTD_isError()).
- * Special : adding a NULL CDict means "return to no-dictionary mode".
- * Note 1 : Currently, only one dictionary can be managed.
- * Adding a new dictionary effectively "discards" any previous one.
- * Note 2 : CDict is just referenced, its lifetime must outlive CCtx. */
-ZSTDLIB_API size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict);
-
-/*! ZSTD_CCtx_refPrefix() :
- * Reference a prefix (single-usage dictionary) for next compression job.
- * Decompression need same prefix to properly regenerate data.
- * Prefix is **only used once**. Tables are discarded at end of compression job.
- * Subsequent compression jobs will be done without prefix (if none is explicitly referenced).
- * If there is a need to use same prefix multiple times, consider embedding it into a ZSTD_CDict instead.
- * @result : 0, or an error code (which can be tested with ZSTD_isError()).
- * Special: Adding any prefix (including NULL) invalidates any previous prefix or dictionary
- * Note 1 : Prefix buffer is referenced. It must outlive compression job.
- * Note 2 : Referencing a prefix involves building tables, which are dependent on compression parameters.
- * It's a CPU consuming operation, with non-negligible impact on latency.
- * Note 3 : By default, the prefix is treated as raw content (ZSTD_dm_rawContent).
- * Use ZSTD_CCtx_refPrefix_advanced() to alter dictMode. */
-ZSTDLIB_API size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize);
-ZSTDLIB_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType);
-
-/*! ZSTD_CCtx_reset() :
- * Return a CCtx to clean state.
- * Useful after an error, or to interrupt an ongoing compression job and start a new one.
- * Any internal data not yet flushed is cancelled.
- * Dictionary (if any) is dropped.
- * All parameters are back to default values.
- * It's possible to modify compression parameters after a reset.
- */
-ZSTDLIB_API void ZSTD_CCtx_reset(ZSTD_CCtx* cctx);
-
-
-
-typedef enum {
- ZSTD_e_continue=0, /* collect more data, encoder decides when to output compressed result, for optimal conditions */
- ZSTD_e_flush, /* flush any data provided so far - frame will continue, future data can still reference previous data for better compression */
- ZSTD_e_end /* flush any remaining data and close current frame. Any additional data starts a new frame. */
-} ZSTD_EndDirective;
-
-/*! ZSTD_compress_generic() :
- * Behave about the same as ZSTD_compressStream. To note :
- * - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_setParameter()
- * - Compression parameters cannot be changed once compression is started.
- * - outpot->pos must be <= dstCapacity, input->pos must be <= srcSize
- * - outpot->pos and input->pos will be updated. They are guaranteed to remain below their respective limit.
- * - In single-thread mode (default), function is blocking : it completed its job before returning to caller.
- * - In multi-thread mode, function is non-blocking : it just acquires a copy of input, and distribute job to internal worker threads,
- * and then immediately returns, just indicating that there is some data remaining to be flushed.
- * The function nonetheless guarantees forward progress : it will return only after it reads or write at least 1+ byte.
- * - Exception : in multi-threading mode, if the first call requests a ZSTD_e_end directive, it is blocking : it will complete compression before giving back control to caller.
- * - @return provides a minimum amount of data remaining to be flushed from internal buffers
- * or an error code, which can be tested using ZSTD_isError().
- * if @return != 0, flush is not fully completed, there is still some data left within internal buffers.
- * This is useful for ZSTD_e_flush, since in this case more flushes are necessary to empty all buffers.
- * For ZSTD_e_end, @return == 0 when internal buffers are fully flushed and frame is completed.
- * - after a ZSTD_e_end directive, if internal buffer is not fully flushed (@return != 0),
- * only ZSTD_e_end or ZSTD_e_flush operations are allowed.
- * Before starting a new compression job, or changing compression parameters,
- * it is required to fully flush internal buffers.
- */
-ZSTDLIB_API size_t ZSTD_compress_generic (ZSTD_CCtx* cctx,
- ZSTD_outBuffer* output,
- ZSTD_inBuffer* input,
- ZSTD_EndDirective endOp);
-
-
-/*! ZSTD_compress_generic_simpleArgs() :
- * Same as ZSTD_compress_generic(),
- * but using only integral types as arguments.
- * Argument list is larger than ZSTD_{in,out}Buffer,
- * but can be helpful for binders from dynamic languages
- * which have troubles handling structures containing memory pointers.
- */
-ZSTDLIB_API size_t ZSTD_compress_generic_simpleArgs (
- ZSTD_CCtx* cctx,
- void* dst, size_t dstCapacity, size_t* dstPos,
- const void* src, size_t srcSize, size_t* srcPos,
- ZSTD_EndDirective endOp);
-
-
-/*! ZSTD_CCtx_params :
- * Quick howto :
- * - ZSTD_createCCtxParams() : Create a ZSTD_CCtx_params structure
- * - ZSTD_CCtxParam_setParameter() : Push parameters one by one into
- * an existing ZSTD_CCtx_params structure.
- * This is similar to
- * ZSTD_CCtx_setParameter().
- * - ZSTD_CCtx_setParametersUsingCCtxParams() : Apply parameters to
- * an existing CCtx.
- * These parameters will be applied to
- * all subsequent compression jobs.
- * - ZSTD_compress_generic() : Do compression using the CCtx.
- * - ZSTD_freeCCtxParams() : Free the memory.
- *
- * This can be used with ZSTD_estimateCCtxSize_advanced_usingCCtxParams()
- * for static allocation for single-threaded compression.
- */
-ZSTDLIB_API ZSTD_CCtx_params* ZSTD_createCCtxParams(void);
-ZSTDLIB_API size_t ZSTD_freeCCtxParams(ZSTD_CCtx_params* params);
-
-
-/*! ZSTD_CCtxParams_reset() :
- * Reset params to default values.
- */
-ZSTDLIB_API size_t ZSTD_CCtxParams_reset(ZSTD_CCtx_params* params);
-
-/*! ZSTD_CCtxParams_init() :
- * Initializes the compression parameters of cctxParams according to
- * compression level. All other parameters are reset to their default values.
- */
-ZSTDLIB_API size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int compressionLevel);
-
-/*! ZSTD_CCtxParams_init_advanced() :
- * Initializes the compression and frame parameters of cctxParams according to
- * params. All other parameters are reset to their default values.
- */
-ZSTDLIB_API size_t ZSTD_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params);
-
-
-/*! ZSTD_CCtxParam_setParameter() :
- * Similar to ZSTD_CCtx_setParameter.
- * Set one compression parameter, selected by enum ZSTD_cParameter.
- * Parameters must be applied to a ZSTD_CCtx using ZSTD_CCtx_setParametersUsingCCtxParams().
- * Note : when `value` is an enum, cast it to unsigned for proper type checking.
- * @result : 0, or an error code (which can be tested with ZSTD_isError()).
- */
-ZSTDLIB_API size_t ZSTD_CCtxParam_setParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, unsigned value);
-
-/*! ZSTD_CCtx_setParametersUsingCCtxParams() :
- * Apply a set of ZSTD_CCtx_params to the compression context.
- * This can be done even after compression is started,
- * if nbWorkers==0, this will have no impact until a new compression is started.
- * if nbWorkers>=1, new parameters will be picked up at next job,
- * with a few restrictions (windowLog, pledgedSrcSize, nbWorkers, jobSize, and overlapLog are not updated).
- */
-ZSTDLIB_API size_t ZSTD_CCtx_setParametersUsingCCtxParams(
- ZSTD_CCtx* cctx, const ZSTD_CCtx_params* params);
-
-
-/*=== Advanced parameters for decompression API ===*/
-
-/* The following parameters must be set after creating a ZSTD_DCtx* (or ZSTD_DStream*) object,
- * but before starting decompression of a frame.
- */
-
-/*! ZSTD_DCtx_loadDictionary() :
- * Create an internal DDict from dict buffer,
- * to be used to decompress next frames.
- * @result : 0, or an error code (which can be tested with ZSTD_isError()).
- * Special : Adding a NULL (or 0-size) dictionary invalidates any previous dictionary,
- * meaning "return to no-dictionary mode".
- * Note 1 : `dict` content will be copied internally.
- * Use ZSTD_DCtx_loadDictionary_byReference()
- * to reference dictionary content instead.
- * In which case, the dictionary buffer must outlive its users.
- * Note 2 : Loading a dictionary involves building tables,
- * which has a non-negligible impact on CPU usage and latency.
- * Note 3 : Use ZSTD_DCtx_loadDictionary_advanced() to select
- * how dictionary content will be interpreted and loaded.
- */
-ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize);
-ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary_byReference(ZSTD_DCtx* dctx, const void* dict, size_t dictSize);
-ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType);
-
-
-/*! ZSTD_DCtx_refDDict() :
- * Reference a prepared dictionary, to be used to decompress next frames.
- * The dictionary remains active for decompression of future frames using same DCtx.
- * @result : 0, or an error code (which can be tested with ZSTD_isError()).
- * Note 1 : Currently, only one dictionary can be managed.
- * Referencing a new dictionary effectively "discards" any previous one.
- * Special : adding a NULL DDict means "return to no-dictionary mode".
- * Note 2 : DDict is just referenced, its lifetime must outlive its usage from DCtx.
- */
-ZSTDLIB_API size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict);
-
-
-/*! ZSTD_DCtx_refPrefix() :
- * Reference a prefix (single-usage dictionary) for next compression job.
- * Prefix is **only used once**. It must be explicitly referenced before each frame.
- * If there is a need to use same prefix multiple times, consider embedding it into a ZSTD_DDict instead.
- * @result : 0, or an error code (which can be tested with ZSTD_isError()).
- * Note 1 : Adding any prefix (including NULL) invalidates any previously set prefix or dictionary
- * Note 2 : Prefix buffer is referenced. It must outlive compression job.
- * Note 3 : By default, the prefix is treated as raw content (ZSTD_dm_rawContent).
- * Use ZSTD_CCtx_refPrefix_advanced() to alter dictMode.
- * Note 4 : Referencing a raw content prefix has almost no cpu nor memory cost.
- */
-ZSTDLIB_API size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize);
-ZSTDLIB_API size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType);
-
-
-/*! ZSTD_DCtx_setMaxWindowSize() :
- * Refuses allocating internal buffers for frames requiring a window size larger than provided limit.
- * This is useful to prevent a decoder context from reserving too much memory for itself (potential attack scenario).
- * This parameter is only useful in streaming mode, since no internal buffer is allocated in direct mode.
- * By default, a decompression context accepts all window sizes <= (1 << ZSTD_WINDOWLOG_MAX)
- * @return : 0, or an error code (which can be tested using ZSTD_isError()).
- */
-ZSTDLIB_API size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize);
-
-
-/*! ZSTD_DCtx_setFormat() :
- * Instruct the decoder context about what kind of data to decode next.
- * This instruction is mandatory to decode data without a fully-formed header,
- * such ZSTD_f_zstd1_magicless for example.
- * @return : 0, or an error code (which can be tested using ZSTD_isError()).
- */
-ZSTDLIB_API size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format);
-
-
-/*! ZSTD_decompress_generic() :
- * Behave the same as ZSTD_decompressStream.
- * Decompression parameters cannot be changed once decompression is started.
- * @return : an error code, which can be tested using ZSTD_isError()
- * if >0, a hint, nb of expected input bytes for next invocation.
- * `0` means : a frame has just been fully decoded and flushed.
- */
-ZSTDLIB_API size_t ZSTD_decompress_generic(ZSTD_DCtx* dctx,
- ZSTD_outBuffer* output,
- ZSTD_inBuffer* input);
-
-
-/*! ZSTD_decompress_generic_simpleArgs() :
- * Same as ZSTD_decompress_generic(),
- * but using only integral types as arguments.
- * Argument list is larger than ZSTD_{in,out}Buffer,
- * but can be helpful for binders from dynamic languages
- * which have troubles handling structures containing memory pointers.
- */
-ZSTDLIB_API size_t ZSTD_decompress_generic_simpleArgs (
- ZSTD_DCtx* dctx,
- void* dst, size_t dstCapacity, size_t* dstPos,
- const void* src, size_t srcSize, size_t* srcPos);
-
-
-/*! ZSTD_DCtx_reset() :
- * Return a DCtx to clean state.
- * If a decompression was ongoing, any internal data not yet flushed is cancelled.
- * All parameters are back to default values, including sticky ones.
- * Dictionary (if any) is dropped.
- * Parameters can be modified again after a reset.
- */
-ZSTDLIB_API void ZSTD_DCtx_reset(ZSTD_DCtx* dctx);
-
-
/* ============================ */
/** Block level API */
@@ -1372,10 +1920,10 @@
+ copyCCtx() and copyDCtx() can be used too
- Block size is limited, it must be <= ZSTD_getBlockSize() <= ZSTD_BLOCKSIZE_MAX == 128 KB
+ If input is larger than a block size, it's necessary to split input data into multiple blocks
- + For inputs larger than a single block size, consider using the regular ZSTD_compress() instead.
+ + For inputs larger than a single block, really consider using regular ZSTD_compress() instead.
Frame metadata is not that costly, and quickly becomes negligible as source size grows larger.
- When a block is considered not compressible enough, ZSTD_compressBlock() result will be zero.
- In which case, nothing is produced into `dst`.
+ In which case, nothing is produced into `dst` !
+ User must test for such outcome and deal directly with uncompressed data
+ ZSTD_decompressBlock() doesn't accept uncompressed data as input !!!
+ In case of multiple successive blocks, should some of them be uncompressed,
@@ -1383,8 +1931,6 @@
Use ZSTD_insertBlock() for such a case.
*/
-#define ZSTD_BLOCKSIZELOG_MAX 17
-#define ZSTD_BLOCKSIZE_MAX (1<<ZSTD_BLOCKSIZELOG_MAX) /* define, for static allocation */
/*===== Raw zstd block functions =====*/
ZSTDLIB_API size_t ZSTD_getBlockSize (const ZSTD_CCtx* cctx);
ZSTDLIB_API size_t ZSTD_compressBlock (ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
diff --git a/vendor/github.com/DataDog/zstd/zstd_common.c b/vendor/github.com/DataDog/zstd/zstd_common.c
index bccc948..667f4a2 100644
--- a/vendor/github.com/DataDog/zstd/zstd_common.c
+++ b/vendor/github.com/DataDog/zstd/zstd_common.c
@@ -30,8 +30,10 @@
/*-****************************************
* ZSTD Error Management
******************************************/
+#undef ZSTD_isError /* defined within zstd_internal.h */
/*! ZSTD_isError() :
- * tells if a return value is an error code */
+ * tells if a return value is an error code
+ * symbol is required for external callers */
unsigned ZSTD_isError(size_t code) { return ERR_isError(code); }
/*! ZSTD_getErrorName() :
@@ -46,11 +48,6 @@
* provides error code string from enum */
const char* ZSTD_getErrorString(ZSTD_ErrorCode code) { return ERR_getErrorString(code); }
-/*! g_debuglog_enable :
- * turn on/off debug traces (global switch) */
-#if defined(ZSTD_DEBUG) && (ZSTD_DEBUG >= 2)
-int g_debuglog_enable = 1;
-#endif
/*=**************************************************************
diff --git a/vendor/github.com/DataDog/zstd/zstd_compress.c b/vendor/github.com/DataDog/zstd/zstd_compress.c
index 2aa26da..1476512 100644
--- a/vendor/github.com/DataDog/zstd/zstd_compress.c
+++ b/vendor/github.com/DataDog/zstd/zstd_compress.c
@@ -8,21 +8,14 @@
* You may select, at your option, one of the above-listed licenses.
*/
-
-/*-*************************************
-* Tuning parameters
-***************************************/
-#ifndef ZSTD_CLEVEL_DEFAULT
-# define ZSTD_CLEVEL_DEFAULT 3
-#endif
-
-
/*-*************************************
* Dependencies
***************************************/
+#include <limits.h> /* INT_MAX */
#include <string.h> /* memset */
#include "cpu.h"
#include "mem.h"
+#include "hist.h" /* HIST_countFast_wksp */
#define FSE_STATIC_LINKING_ONLY /* FSE_encodeSymbol */
#include "fse.h"
#define HUF_STATIC_LINKING_ONLY
@@ -54,7 +47,6 @@
size_t workspaceSize;
ZSTD_matchState_t matchState;
ZSTD_compressedBlockState_t cBlockState;
- ZSTD_compressionParameters cParams;
ZSTD_customMem customMem;
U32 dictID;
}; /* typedef'd to ZSTD_CDict within "zstd.h" */
@@ -64,17 +56,26 @@
return ZSTD_createCCtx_advanced(ZSTD_defaultCMem);
}
+static void ZSTD_initCCtx(ZSTD_CCtx* cctx, ZSTD_customMem memManager)
+{
+ assert(cctx != NULL);
+ memset(cctx, 0, sizeof(*cctx));
+ cctx->customMem = memManager;
+ cctx->bmi2 = ZSTD_cpuid_bmi2(ZSTD_cpuid());
+ { size_t const err = ZSTD_CCtx_reset(cctx, ZSTD_reset_parameters);
+ assert(!ZSTD_isError(err));
+ (void)err;
+ }
+}
+
ZSTD_CCtx* ZSTD_createCCtx_advanced(ZSTD_customMem customMem)
{
ZSTD_STATIC_ASSERT(zcss_init==0);
ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_UNKNOWN==(0ULL - 1));
if (!customMem.customAlloc ^ !customMem.customFree) return NULL;
- { ZSTD_CCtx* const cctx = (ZSTD_CCtx*)ZSTD_calloc(sizeof(ZSTD_CCtx), customMem);
+ { ZSTD_CCtx* const cctx = (ZSTD_CCtx*)ZSTD_malloc(sizeof(ZSTD_CCtx), customMem);
if (!cctx) return NULL;
- cctx->customMem = customMem;
- cctx->requestedParams.compressionLevel = ZSTD_CLEVEL_DEFAULT;
- cctx->requestedParams.fParams.contentSizeFlag = 1;
- cctx->bmi2 = ZSTD_cpuid_bmi2(ZSTD_cpuid());
+ ZSTD_initCCtx(cctx, customMem);
return cctx;
}
}
@@ -102,17 +103,44 @@
return cctx;
}
-size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx)
+/**
+ * Clears and frees all of the dictionaries in the CCtx.
+ */
+static void ZSTD_clearAllDicts(ZSTD_CCtx* cctx)
{
- if (cctx==NULL) return 0; /* support free on NULL */
- if (cctx->staticSize) return ERROR(memory_allocation); /* not compatible with static CCtx */
+ ZSTD_free(cctx->localDict.dictBuffer, cctx->customMem);
+ ZSTD_freeCDict(cctx->localDict.cdict);
+ memset(&cctx->localDict, 0, sizeof(cctx->localDict));
+ memset(&cctx->prefixDict, 0, sizeof(cctx->prefixDict));
+ cctx->cdict = NULL;
+}
+
+static size_t ZSTD_sizeof_localDict(ZSTD_localDict dict)
+{
+ size_t const bufferSize = dict.dictBuffer != NULL ? dict.dictSize : 0;
+ size_t const cdictSize = ZSTD_sizeof_CDict(dict.cdict);
+ return bufferSize + cdictSize;
+}
+
+static void ZSTD_freeCCtxContent(ZSTD_CCtx* cctx)
+{
+ assert(cctx != NULL);
+ assert(cctx->staticSize == 0);
ZSTD_free(cctx->workSpace, cctx->customMem); cctx->workSpace = NULL;
- ZSTD_freeCDict(cctx->cdictLocal); cctx->cdictLocal = NULL;
+ ZSTD_clearAllDicts(cctx);
#ifdef ZSTD_MULTITHREAD
ZSTDMT_freeCCtx(cctx->mtctx); cctx->mtctx = NULL;
#endif
+}
+
+size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx)
+{
+ if (cctx==NULL) return 0; /* support free on NULL */
+ RETURN_ERROR_IF(cctx->staticSize, memory_allocation,
+ "not compatible with static CCtx");
+ ZSTD_freeCCtxContent(cctx);
ZSTD_free(cctx, cctx->customMem);
- return 0; /* reserved as a potential error code in the future */
+ return 0;
}
@@ -121,7 +149,7 @@
#ifdef ZSTD_MULTITHREAD
return ZSTDMT_sizeof_CCtx(cctx->mtctx);
#else
- (void) cctx;
+ (void)cctx;
return 0;
#endif
}
@@ -131,7 +159,7 @@
{
if (cctx==NULL) return 0; /* support sizeof on NULL */
return sizeof(*cctx) + cctx->workSpaceSize
- + ZSTD_sizeof_CDict(cctx->cdictLocal)
+ + ZSTD_sizeof_localDict(cctx->localDict)
+ ZSTD_sizeof_mtctx(cctx);
}
@@ -143,21 +171,6 @@
/* private API call, for dictBuilder only */
const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx) { return &(ctx->seqStore); }
-ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams(
- const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize)
-{
- ZSTD_compressionParameters cParams = ZSTD_getCParams(CCtxParams->compressionLevel, srcSizeHint, dictSize);
- if (CCtxParams->ldmParams.enableLdm) cParams.windowLog = ZSTD_LDM_DEFAULT_WINDOW_LOG;
- if (CCtxParams->cParams.windowLog) cParams.windowLog = CCtxParams->cParams.windowLog;
- if (CCtxParams->cParams.hashLog) cParams.hashLog = CCtxParams->cParams.hashLog;
- if (CCtxParams->cParams.chainLog) cParams.chainLog = CCtxParams->cParams.chainLog;
- if (CCtxParams->cParams.searchLog) cParams.searchLog = CCtxParams->cParams.searchLog;
- if (CCtxParams->cParams.searchLength) cParams.searchLength = CCtxParams->cParams.searchLength;
- if (CCtxParams->cParams.targetLength) cParams.targetLength = CCtxParams->cParams.targetLength;
- if (CCtxParams->cParams.strategy) cParams.strategy = CCtxParams->cParams.strategy;
- return cParams;
-}
-
static ZSTD_CCtx_params ZSTD_makeCCtxParamsFromCParams(
ZSTD_compressionParameters cParams)
{
@@ -202,7 +215,7 @@
}
size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int compressionLevel) {
- if (!cctxParams) { return ERROR(GENERIC); }
+ RETURN_ERROR_IF(!cctxParams, GENERIC);
memset(cctxParams, 0, sizeof(*cctxParams));
cctxParams->compressionLevel = compressionLevel;
cctxParams->fParams.contentSizeFlag = 1;
@@ -211,8 +224,8 @@
size_t ZSTD_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params)
{
- if (!cctxParams) { return ERROR(GENERIC); }
- CHECK_F( ZSTD_checkCParams(params.cParams) );
+ RETURN_ERROR_IF(!cctxParams, GENERIC);
+ FORWARD_IF_ERROR( ZSTD_checkCParams(params.cParams) );
memset(cctxParams, 0, sizeof(*cctxParams));
cctxParams->cParams = params.cParams;
cctxParams->fParams = params.fParams;
@@ -234,249 +247,553 @@
return ret;
}
-#define CLAMPCHECK(val,min,max) { \
- if (((val)<(min)) | ((val)>(max))) { \
- return ERROR(parameter_outOfBound); \
-} }
+ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter param)
+{
+ ZSTD_bounds bounds = { 0, 0, 0 };
+
+ switch(param)
+ {
+ case ZSTD_c_compressionLevel:
+ bounds.lowerBound = ZSTD_minCLevel();
+ bounds.upperBound = ZSTD_maxCLevel();
+ return bounds;
+
+ case ZSTD_c_windowLog:
+ bounds.lowerBound = ZSTD_WINDOWLOG_MIN;
+ bounds.upperBound = ZSTD_WINDOWLOG_MAX;
+ return bounds;
+
+ case ZSTD_c_hashLog:
+ bounds.lowerBound = ZSTD_HASHLOG_MIN;
+ bounds.upperBound = ZSTD_HASHLOG_MAX;
+ return bounds;
+
+ case ZSTD_c_chainLog:
+ bounds.lowerBound = ZSTD_CHAINLOG_MIN;
+ bounds.upperBound = ZSTD_CHAINLOG_MAX;
+ return bounds;
+
+ case ZSTD_c_searchLog:
+ bounds.lowerBound = ZSTD_SEARCHLOG_MIN;
+ bounds.upperBound = ZSTD_SEARCHLOG_MAX;
+ return bounds;
+
+ case ZSTD_c_minMatch:
+ bounds.lowerBound = ZSTD_MINMATCH_MIN;
+ bounds.upperBound = ZSTD_MINMATCH_MAX;
+ return bounds;
+
+ case ZSTD_c_targetLength:
+ bounds.lowerBound = ZSTD_TARGETLENGTH_MIN;
+ bounds.upperBound = ZSTD_TARGETLENGTH_MAX;
+ return bounds;
+
+ case ZSTD_c_strategy:
+ bounds.lowerBound = ZSTD_STRATEGY_MIN;
+ bounds.upperBound = ZSTD_STRATEGY_MAX;
+ return bounds;
+
+ case ZSTD_c_contentSizeFlag:
+ bounds.lowerBound = 0;
+ bounds.upperBound = 1;
+ return bounds;
+
+ case ZSTD_c_checksumFlag:
+ bounds.lowerBound = 0;
+ bounds.upperBound = 1;
+ return bounds;
+
+ case ZSTD_c_dictIDFlag:
+ bounds.lowerBound = 0;
+ bounds.upperBound = 1;
+ return bounds;
+
+ case ZSTD_c_nbWorkers:
+ bounds.lowerBound = 0;
+#ifdef ZSTD_MULTITHREAD
+ bounds.upperBound = ZSTDMT_NBWORKERS_MAX;
+#else
+ bounds.upperBound = 0;
+#endif
+ return bounds;
+
+ case ZSTD_c_jobSize:
+ bounds.lowerBound = 0;
+#ifdef ZSTD_MULTITHREAD
+ bounds.upperBound = ZSTDMT_JOBSIZE_MAX;
+#else
+ bounds.upperBound = 0;
+#endif
+ return bounds;
+
+ case ZSTD_c_overlapLog:
+ bounds.lowerBound = ZSTD_OVERLAPLOG_MIN;
+ bounds.upperBound = ZSTD_OVERLAPLOG_MAX;
+ return bounds;
+
+ case ZSTD_c_enableLongDistanceMatching:
+ bounds.lowerBound = 0;
+ bounds.upperBound = 1;
+ return bounds;
+
+ case ZSTD_c_ldmHashLog:
+ bounds.lowerBound = ZSTD_LDM_HASHLOG_MIN;
+ bounds.upperBound = ZSTD_LDM_HASHLOG_MAX;
+ return bounds;
+
+ case ZSTD_c_ldmMinMatch:
+ bounds.lowerBound = ZSTD_LDM_MINMATCH_MIN;
+ bounds.upperBound = ZSTD_LDM_MINMATCH_MAX;
+ return bounds;
+
+ case ZSTD_c_ldmBucketSizeLog:
+ bounds.lowerBound = ZSTD_LDM_BUCKETSIZELOG_MIN;
+ bounds.upperBound = ZSTD_LDM_BUCKETSIZELOG_MAX;
+ return bounds;
+
+ case ZSTD_c_ldmHashRateLog:
+ bounds.lowerBound = ZSTD_LDM_HASHRATELOG_MIN;
+ bounds.upperBound = ZSTD_LDM_HASHRATELOG_MAX;
+ return bounds;
+
+ /* experimental parameters */
+ case ZSTD_c_rsyncable:
+ bounds.lowerBound = 0;
+ bounds.upperBound = 1;
+ return bounds;
+
+ case ZSTD_c_forceMaxWindow :
+ bounds.lowerBound = 0;
+ bounds.upperBound = 1;
+ return bounds;
+
+ case ZSTD_c_format:
+ ZSTD_STATIC_ASSERT(ZSTD_f_zstd1 < ZSTD_f_zstd1_magicless);
+ bounds.lowerBound = ZSTD_f_zstd1;
+ bounds.upperBound = ZSTD_f_zstd1_magicless; /* note : how to ensure at compile time that this is the highest value enum ? */
+ return bounds;
+
+ case ZSTD_c_forceAttachDict:
+ ZSTD_STATIC_ASSERT(ZSTD_dictDefaultAttach < ZSTD_dictForceCopy);
+ bounds.lowerBound = ZSTD_dictDefaultAttach;
+ bounds.upperBound = ZSTD_dictForceCopy; /* note : how to ensure at compile time that this is the highest value enum ? */
+ return bounds;
+
+ case ZSTD_c_literalCompressionMode:
+ ZSTD_STATIC_ASSERT(ZSTD_lcm_auto < ZSTD_lcm_huffman && ZSTD_lcm_huffman < ZSTD_lcm_uncompressed);
+ bounds.lowerBound = ZSTD_lcm_auto;
+ bounds.upperBound = ZSTD_lcm_uncompressed;
+ return bounds;
+
+ case ZSTD_c_targetCBlockSize:
+ bounds.lowerBound = ZSTD_TARGETCBLOCKSIZE_MIN;
+ bounds.upperBound = ZSTD_TARGETCBLOCKSIZE_MAX;
+ return bounds;
+
+ default:
+ { ZSTD_bounds const boundError = { ERROR(parameter_unsupported), 0, 0 };
+ return boundError;
+ }
+ }
+}
+
+/* ZSTD_cParam_withinBounds:
+ * @return 1 if value is within cParam bounds,
+ * 0 otherwise */
+static int ZSTD_cParam_withinBounds(ZSTD_cParameter cParam, int value)
+{
+ ZSTD_bounds const bounds = ZSTD_cParam_getBounds(cParam);
+ if (ZSTD_isError(bounds.error)) return 0;
+ if (value < bounds.lowerBound) return 0;
+ if (value > bounds.upperBound) return 0;
+ return 1;
+}
+
+/* ZSTD_cParam_clampBounds:
+ * Clamps the value into the bounded range.
+ */
+static size_t ZSTD_cParam_clampBounds(ZSTD_cParameter cParam, int* value)
+{
+ ZSTD_bounds const bounds = ZSTD_cParam_getBounds(cParam);
+ if (ZSTD_isError(bounds.error)) return bounds.error;
+ if (*value < bounds.lowerBound) *value = bounds.lowerBound;
+ if (*value > bounds.upperBound) *value = bounds.upperBound;
+ return 0;
+}
+
+#define BOUNDCHECK(cParam, val) { \
+ RETURN_ERROR_IF(!ZSTD_cParam_withinBounds(cParam,val), \
+ parameter_outOfBound); \
+}
static int ZSTD_isUpdateAuthorized(ZSTD_cParameter param)
{
switch(param)
{
- case ZSTD_p_compressionLevel:
- case ZSTD_p_hashLog:
- case ZSTD_p_chainLog:
- case ZSTD_p_searchLog:
- case ZSTD_p_minMatch:
- case ZSTD_p_targetLength:
- case ZSTD_p_compressionStrategy:
- case ZSTD_p_compressLiterals:
+ case ZSTD_c_compressionLevel:
+ case ZSTD_c_hashLog:
+ case ZSTD_c_chainLog:
+ case ZSTD_c_searchLog:
+ case ZSTD_c_minMatch:
+ case ZSTD_c_targetLength:
+ case ZSTD_c_strategy:
return 1;
- case ZSTD_p_format:
- case ZSTD_p_windowLog:
- case ZSTD_p_contentSizeFlag:
- case ZSTD_p_checksumFlag:
- case ZSTD_p_dictIDFlag:
- case ZSTD_p_forceMaxWindow :
- case ZSTD_p_nbWorkers:
- case ZSTD_p_jobSize:
- case ZSTD_p_overlapSizeLog:
- case ZSTD_p_enableLongDistanceMatching:
- case ZSTD_p_ldmHashLog:
- case ZSTD_p_ldmMinMatch:
- case ZSTD_p_ldmBucketSizeLog:
- case ZSTD_p_ldmHashEveryLog:
+ case ZSTD_c_format:
+ case ZSTD_c_windowLog:
+ case ZSTD_c_contentSizeFlag:
+ case ZSTD_c_checksumFlag:
+ case ZSTD_c_dictIDFlag:
+ case ZSTD_c_forceMaxWindow :
+ case ZSTD_c_nbWorkers:
+ case ZSTD_c_jobSize:
+ case ZSTD_c_overlapLog:
+ case ZSTD_c_rsyncable:
+ case ZSTD_c_enableLongDistanceMatching:
+ case ZSTD_c_ldmHashLog:
+ case ZSTD_c_ldmMinMatch:
+ case ZSTD_c_ldmBucketSizeLog:
+ case ZSTD_c_ldmHashRateLog:
+ case ZSTD_c_forceAttachDict:
+ case ZSTD_c_literalCompressionMode:
+ case ZSTD_c_targetCBlockSize:
default:
return 0;
}
}
-size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, unsigned value)
+size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int value)
{
- DEBUGLOG(4, "ZSTD_CCtx_setParameter (%u, %u)", (U32)param, value);
+ DEBUGLOG(4, "ZSTD_CCtx_setParameter (%i, %i)", (int)param, value);
if (cctx->streamStage != zcss_init) {
if (ZSTD_isUpdateAuthorized(param)) {
cctx->cParamsChanged = 1;
} else {
- return ERROR(stage_wrong);
+ RETURN_ERROR(stage_wrong);
} }
switch(param)
{
- case ZSTD_p_format :
- return ZSTD_CCtxParam_setParameter(&cctx->requestedParams, param, value);
+ case ZSTD_c_nbWorkers:
+ RETURN_ERROR_IF((value!=0) && cctx->staticSize, parameter_unsupported,
+ "MT not compatible with static alloc");
+ break;
- case ZSTD_p_compressionLevel:
- if (cctx->cdict) return ERROR(stage_wrong);
- return ZSTD_CCtxParam_setParameter(&cctx->requestedParams, param, value);
+ case ZSTD_c_compressionLevel:
+ case ZSTD_c_windowLog:
+ case ZSTD_c_hashLog:
+ case ZSTD_c_chainLog:
+ case ZSTD_c_searchLog:
+ case ZSTD_c_minMatch:
+ case ZSTD_c_targetLength:
+ case ZSTD_c_strategy:
+ case ZSTD_c_ldmHashRateLog:
+ case ZSTD_c_format:
+ case ZSTD_c_contentSizeFlag:
+ case ZSTD_c_checksumFlag:
+ case ZSTD_c_dictIDFlag:
+ case ZSTD_c_forceMaxWindow:
+ case ZSTD_c_forceAttachDict:
+ case ZSTD_c_literalCompressionMode:
+ case ZSTD_c_jobSize:
+ case ZSTD_c_overlapLog:
+ case ZSTD_c_rsyncable:
+ case ZSTD_c_enableLongDistanceMatching:
+ case ZSTD_c_ldmHashLog:
+ case ZSTD_c_ldmMinMatch:
+ case ZSTD_c_ldmBucketSizeLog:
+ case ZSTD_c_targetCBlockSize:
+ break;
- case ZSTD_p_windowLog:
- case ZSTD_p_hashLog:
- case ZSTD_p_chainLog:
- case ZSTD_p_searchLog:
- case ZSTD_p_minMatch:
- case ZSTD_p_targetLength:
- case ZSTD_p_compressionStrategy:
- if (cctx->cdict) return ERROR(stage_wrong);
- return ZSTD_CCtxParam_setParameter(&cctx->requestedParams, param, value);
-
- case ZSTD_p_compressLiterals:
- case ZSTD_p_contentSizeFlag:
- case ZSTD_p_checksumFlag:
- case ZSTD_p_dictIDFlag:
- return ZSTD_CCtxParam_setParameter(&cctx->requestedParams, param, value);
-
- case ZSTD_p_forceMaxWindow : /* Force back-references to remain < windowSize,
- * even when referencing into Dictionary content.
- * default : 0 when using a CDict, 1 when using a Prefix */
- return ZSTD_CCtxParam_setParameter(&cctx->requestedParams, param, value);
-
- case ZSTD_p_nbWorkers:
- if ((value>0) && cctx->staticSize) {
- return ERROR(parameter_unsupported); /* MT not compatible with static alloc */
- }
- return ZSTD_CCtxParam_setParameter(&cctx->requestedParams, param, value);
-
- case ZSTD_p_jobSize:
- case ZSTD_p_overlapSizeLog:
- return ZSTD_CCtxParam_setParameter(&cctx->requestedParams, param, value);
-
- case ZSTD_p_enableLongDistanceMatching:
- case ZSTD_p_ldmHashLog:
- case ZSTD_p_ldmMinMatch:
- case ZSTD_p_ldmBucketSizeLog:
- case ZSTD_p_ldmHashEveryLog:
- if (cctx->cdict) return ERROR(stage_wrong);
- return ZSTD_CCtxParam_setParameter(&cctx->requestedParams, param, value);
-
- default: return ERROR(parameter_unsupported);
+ default: RETURN_ERROR(parameter_unsupported);
}
+ return ZSTD_CCtxParams_setParameter(&cctx->requestedParams, param, value);
}
-size_t ZSTD_CCtxParam_setParameter(
- ZSTD_CCtx_params* CCtxParams, ZSTD_cParameter param, unsigned value)
+size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* CCtxParams,
+ ZSTD_cParameter param, int value)
{
- DEBUGLOG(4, "ZSTD_CCtxParam_setParameter (%u, %u)", (U32)param, value);
+ DEBUGLOG(4, "ZSTD_CCtxParams_setParameter (%i, %i)", (int)param, value);
switch(param)
{
- case ZSTD_p_format :
- if (value > (unsigned)ZSTD_f_zstd1_magicless)
- return ERROR(parameter_unsupported);
+ case ZSTD_c_format :
+ BOUNDCHECK(ZSTD_c_format, value);
CCtxParams->format = (ZSTD_format_e)value;
return (size_t)CCtxParams->format;
- case ZSTD_p_compressionLevel : {
- int cLevel = (int)value; /* cast expected to restore negative sign */
- if (cLevel > ZSTD_maxCLevel()) cLevel = ZSTD_maxCLevel();
- if (cLevel) { /* 0 : does not change current level */
- CCtxParams->disableLiteralCompression = (cLevel<0); /* negative levels disable huffman */
- CCtxParams->compressionLevel = cLevel;
+ case ZSTD_c_compressionLevel : {
+ FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(param, &value));
+ if (value) { /* 0 : does not change current level */
+ CCtxParams->compressionLevel = value;
}
if (CCtxParams->compressionLevel >= 0) return CCtxParams->compressionLevel;
return 0; /* return type (size_t) cannot represent negative values */
}
- case ZSTD_p_windowLog :
- if (value>0) /* 0 => use default */
- CLAMPCHECK(value, ZSTD_WINDOWLOG_MIN, ZSTD_WINDOWLOG_MAX);
+ case ZSTD_c_windowLog :
+ if (value!=0) /* 0 => use default */
+ BOUNDCHECK(ZSTD_c_windowLog, value);
CCtxParams->cParams.windowLog = value;
return CCtxParams->cParams.windowLog;
- case ZSTD_p_hashLog :
- if (value>0) /* 0 => use default */
- CLAMPCHECK(value, ZSTD_HASHLOG_MIN, ZSTD_HASHLOG_MAX);
+ case ZSTD_c_hashLog :
+ if (value!=0) /* 0 => use default */
+ BOUNDCHECK(ZSTD_c_hashLog, value);
CCtxParams->cParams.hashLog = value;
return CCtxParams->cParams.hashLog;
- case ZSTD_p_chainLog :
- if (value>0) /* 0 => use default */
- CLAMPCHECK(value, ZSTD_CHAINLOG_MIN, ZSTD_CHAINLOG_MAX);
+ case ZSTD_c_chainLog :
+ if (value!=0) /* 0 => use default */
+ BOUNDCHECK(ZSTD_c_chainLog, value);
CCtxParams->cParams.chainLog = value;
return CCtxParams->cParams.chainLog;
- case ZSTD_p_searchLog :
- if (value>0) /* 0 => use default */
- CLAMPCHECK(value, ZSTD_SEARCHLOG_MIN, ZSTD_SEARCHLOG_MAX);
+ case ZSTD_c_searchLog :
+ if (value!=0) /* 0 => use default */
+ BOUNDCHECK(ZSTD_c_searchLog, value);
CCtxParams->cParams.searchLog = value;
return value;
- case ZSTD_p_minMatch :
- if (value>0) /* 0 => use default */
- CLAMPCHECK(value, ZSTD_SEARCHLENGTH_MIN, ZSTD_SEARCHLENGTH_MAX);
- CCtxParams->cParams.searchLength = value;
- return CCtxParams->cParams.searchLength;
+ case ZSTD_c_minMatch :
+ if (value!=0) /* 0 => use default */
+ BOUNDCHECK(ZSTD_c_minMatch, value);
+ CCtxParams->cParams.minMatch = value;
+ return CCtxParams->cParams.minMatch;
- case ZSTD_p_targetLength :
- /* all values are valid. 0 => use default */
+ case ZSTD_c_targetLength :
+ BOUNDCHECK(ZSTD_c_targetLength, value);
CCtxParams->cParams.targetLength = value;
return CCtxParams->cParams.targetLength;
- case ZSTD_p_compressionStrategy :
- if (value>0) /* 0 => use default */
- CLAMPCHECK(value, (unsigned)ZSTD_fast, (unsigned)ZSTD_btultra);
+ case ZSTD_c_strategy :
+ if (value!=0) /* 0 => use default */
+ BOUNDCHECK(ZSTD_c_strategy, value);
CCtxParams->cParams.strategy = (ZSTD_strategy)value;
return (size_t)CCtxParams->cParams.strategy;
- case ZSTD_p_compressLiterals:
- CCtxParams->disableLiteralCompression = !value;
- return !CCtxParams->disableLiteralCompression;
-
- case ZSTD_p_contentSizeFlag :
+ case ZSTD_c_contentSizeFlag :
/* Content size written in frame header _when known_ (default:1) */
- DEBUGLOG(4, "set content size flag = %u", (value>0));
- CCtxParams->fParams.contentSizeFlag = value > 0;
+ DEBUGLOG(4, "set content size flag = %u", (value!=0));
+ CCtxParams->fParams.contentSizeFlag = value != 0;
return CCtxParams->fParams.contentSizeFlag;
- case ZSTD_p_checksumFlag :
+ case ZSTD_c_checksumFlag :
/* A 32-bits content checksum will be calculated and written at end of frame (default:0) */
- CCtxParams->fParams.checksumFlag = value > 0;
+ CCtxParams->fParams.checksumFlag = value != 0;
return CCtxParams->fParams.checksumFlag;
- case ZSTD_p_dictIDFlag : /* When applicable, dictionary's dictID is provided in frame header (default:1) */
- DEBUGLOG(4, "set dictIDFlag = %u", (value>0));
+ case ZSTD_c_dictIDFlag : /* When applicable, dictionary's dictID is provided in frame header (default:1) */
+ DEBUGLOG(4, "set dictIDFlag = %u", (value!=0));
CCtxParams->fParams.noDictIDFlag = !value;
return !CCtxParams->fParams.noDictIDFlag;
- case ZSTD_p_forceMaxWindow :
- CCtxParams->forceWindow = (value > 0);
+ case ZSTD_c_forceMaxWindow :
+ CCtxParams->forceWindow = (value != 0);
return CCtxParams->forceWindow;
- case ZSTD_p_nbWorkers :
+ case ZSTD_c_forceAttachDict : {
+ const ZSTD_dictAttachPref_e pref = (ZSTD_dictAttachPref_e)value;
+ BOUNDCHECK(ZSTD_c_forceAttachDict, pref);
+ CCtxParams->attachDictPref = pref;
+ return CCtxParams->attachDictPref;
+ }
+
+ case ZSTD_c_literalCompressionMode : {
+ const ZSTD_literalCompressionMode_e lcm = (ZSTD_literalCompressionMode_e)value;
+ BOUNDCHECK(ZSTD_c_literalCompressionMode, lcm);
+ CCtxParams->literalCompressionMode = lcm;
+ return CCtxParams->literalCompressionMode;
+ }
+
+ case ZSTD_c_nbWorkers :
#ifndef ZSTD_MULTITHREAD
- if (value>0) return ERROR(parameter_unsupported);
+ RETURN_ERROR_IF(value!=0, parameter_unsupported, "not compiled with multithreading");
return 0;
#else
- return ZSTDMT_CCtxParam_setNbWorkers(CCtxParams, value);
+ FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(param, &value));
+ CCtxParams->nbWorkers = value;
+ return CCtxParams->nbWorkers;
#endif
- case ZSTD_p_jobSize :
+ case ZSTD_c_jobSize :
#ifndef ZSTD_MULTITHREAD
- return ERROR(parameter_unsupported);
+ RETURN_ERROR_IF(value!=0, parameter_unsupported, "not compiled with multithreading");
+ return 0;
#else
- return ZSTDMT_CCtxParam_setMTCtxParameter(CCtxParams, ZSTDMT_p_jobSize, value);
+ /* Adjust to the minimum non-default value. */
+ if (value != 0 && value < ZSTDMT_JOBSIZE_MIN)
+ value = ZSTDMT_JOBSIZE_MIN;
+ FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(param, &value));
+ assert(value >= 0);
+ CCtxParams->jobSize = value;
+ return CCtxParams->jobSize;
#endif
- case ZSTD_p_overlapSizeLog :
+ case ZSTD_c_overlapLog :
#ifndef ZSTD_MULTITHREAD
- return ERROR(parameter_unsupported);
+ RETURN_ERROR_IF(value!=0, parameter_unsupported, "not compiled with multithreading");
+ return 0;
#else
- return ZSTDMT_CCtxParam_setMTCtxParameter(CCtxParams, ZSTDMT_p_overlapSectionLog, value);
+ FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(ZSTD_c_overlapLog, &value));
+ CCtxParams->overlapLog = value;
+ return CCtxParams->overlapLog;
#endif
- case ZSTD_p_enableLongDistanceMatching :
- CCtxParams->ldmParams.enableLdm = (value>0);
+ case ZSTD_c_rsyncable :
+#ifndef ZSTD_MULTITHREAD
+ RETURN_ERROR_IF(value!=0, parameter_unsupported, "not compiled with multithreading");
+ return 0;
+#else
+ FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(ZSTD_c_overlapLog, &value));
+ CCtxParams->rsyncable = value;
+ return CCtxParams->rsyncable;
+#endif
+
+ case ZSTD_c_enableLongDistanceMatching :
+ CCtxParams->ldmParams.enableLdm = (value!=0);
return CCtxParams->ldmParams.enableLdm;
- case ZSTD_p_ldmHashLog :
- if (value>0) /* 0 ==> auto */
- CLAMPCHECK(value, ZSTD_HASHLOG_MIN, ZSTD_HASHLOG_MAX);
+ case ZSTD_c_ldmHashLog :
+ if (value!=0) /* 0 ==> auto */
+ BOUNDCHECK(ZSTD_c_ldmHashLog, value);
CCtxParams->ldmParams.hashLog = value;
return CCtxParams->ldmParams.hashLog;
- case ZSTD_p_ldmMinMatch :
- if (value>0) /* 0 ==> default */
- CLAMPCHECK(value, ZSTD_LDM_MINMATCH_MIN, ZSTD_LDM_MINMATCH_MAX);
+ case ZSTD_c_ldmMinMatch :
+ if (value!=0) /* 0 ==> default */
+ BOUNDCHECK(ZSTD_c_ldmMinMatch, value);
CCtxParams->ldmParams.minMatchLength = value;
return CCtxParams->ldmParams.minMatchLength;
- case ZSTD_p_ldmBucketSizeLog :
- if (value > ZSTD_LDM_BUCKETSIZELOG_MAX)
- return ERROR(parameter_outOfBound);
+ case ZSTD_c_ldmBucketSizeLog :
+ if (value!=0) /* 0 ==> default */
+ BOUNDCHECK(ZSTD_c_ldmBucketSizeLog, value);
CCtxParams->ldmParams.bucketSizeLog = value;
return CCtxParams->ldmParams.bucketSizeLog;
- case ZSTD_p_ldmHashEveryLog :
- if (value > ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN)
- return ERROR(parameter_outOfBound);
- CCtxParams->ldmParams.hashEveryLog = value;
- return CCtxParams->ldmParams.hashEveryLog;
+ case ZSTD_c_ldmHashRateLog :
+ RETURN_ERROR_IF(value > ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN,
+ parameter_outOfBound);
+ CCtxParams->ldmParams.hashRateLog = value;
+ return CCtxParams->ldmParams.hashRateLog;
- default: return ERROR(parameter_unsupported);
+ case ZSTD_c_targetCBlockSize :
+ if (value!=0) /* 0 ==> default */
+ BOUNDCHECK(ZSTD_c_targetCBlockSize, value);
+ CCtxParams->targetCBlockSize = value;
+ return CCtxParams->targetCBlockSize;
+
+ default: RETURN_ERROR(parameter_unsupported, "unknown parameter");
}
}
+size_t ZSTD_CCtx_getParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int* value)
+{
+ return ZSTD_CCtxParams_getParameter(&cctx->requestedParams, param, value);
+}
+
+size_t ZSTD_CCtxParams_getParameter(
+ ZSTD_CCtx_params* CCtxParams, ZSTD_cParameter param, int* value)
+{
+ switch(param)
+ {
+ case ZSTD_c_format :
+ *value = CCtxParams->format;
+ break;
+ case ZSTD_c_compressionLevel :
+ *value = CCtxParams->compressionLevel;
+ break;
+ case ZSTD_c_windowLog :
+ *value = (int)CCtxParams->cParams.windowLog;
+ break;
+ case ZSTD_c_hashLog :
+ *value = (int)CCtxParams->cParams.hashLog;
+ break;
+ case ZSTD_c_chainLog :
+ *value = (int)CCtxParams->cParams.chainLog;
+ break;
+ case ZSTD_c_searchLog :
+ *value = CCtxParams->cParams.searchLog;
+ break;
+ case ZSTD_c_minMatch :
+ *value = CCtxParams->cParams.minMatch;
+ break;
+ case ZSTD_c_targetLength :
+ *value = CCtxParams->cParams.targetLength;
+ break;
+ case ZSTD_c_strategy :
+ *value = (unsigned)CCtxParams->cParams.strategy;
+ break;
+ case ZSTD_c_contentSizeFlag :
+ *value = CCtxParams->fParams.contentSizeFlag;
+ break;
+ case ZSTD_c_checksumFlag :
+ *value = CCtxParams->fParams.checksumFlag;
+ break;
+ case ZSTD_c_dictIDFlag :
+ *value = !CCtxParams->fParams.noDictIDFlag;
+ break;
+ case ZSTD_c_forceMaxWindow :
+ *value = CCtxParams->forceWindow;
+ break;
+ case ZSTD_c_forceAttachDict :
+ *value = CCtxParams->attachDictPref;
+ break;
+ case ZSTD_c_literalCompressionMode :
+ *value = CCtxParams->literalCompressionMode;
+ break;
+ case ZSTD_c_nbWorkers :
+#ifndef ZSTD_MULTITHREAD
+ assert(CCtxParams->nbWorkers == 0);
+#endif
+ *value = CCtxParams->nbWorkers;
+ break;
+ case ZSTD_c_jobSize :
+#ifndef ZSTD_MULTITHREAD
+ RETURN_ERROR(parameter_unsupported, "not compiled with multithreading");
+#else
+ assert(CCtxParams->jobSize <= INT_MAX);
+ *value = (int)CCtxParams->jobSize;
+ break;
+#endif
+ case ZSTD_c_overlapLog :
+#ifndef ZSTD_MULTITHREAD
+ RETURN_ERROR(parameter_unsupported, "not compiled with multithreading");
+#else
+ *value = CCtxParams->overlapLog;
+ break;
+#endif
+ case ZSTD_c_rsyncable :
+#ifndef ZSTD_MULTITHREAD
+ RETURN_ERROR(parameter_unsupported, "not compiled with multithreading");
+#else
+ *value = CCtxParams->rsyncable;
+ break;
+#endif
+ case ZSTD_c_enableLongDistanceMatching :
+ *value = CCtxParams->ldmParams.enableLdm;
+ break;
+ case ZSTD_c_ldmHashLog :
+ *value = CCtxParams->ldmParams.hashLog;
+ break;
+ case ZSTD_c_ldmMinMatch :
+ *value = CCtxParams->ldmParams.minMatchLength;
+ break;
+ case ZSTD_c_ldmBucketSizeLog :
+ *value = CCtxParams->ldmParams.bucketSizeLog;
+ break;
+ case ZSTD_c_ldmHashRateLog :
+ *value = CCtxParams->ldmParams.hashRateLog;
+ break;
+ case ZSTD_c_targetCBlockSize :
+ *value = (int)CCtxParams->targetCBlockSize;
+ break;
+ default: RETURN_ERROR(parameter_unsupported, "unknown parameter");
+ }
+ return 0;
+}
+
/** ZSTD_CCtx_setParametersUsingCCtxParams() :
* just applies `params` into `cctx`
* no action is performed, parameters are merely stored.
@@ -487,8 +804,9 @@
size_t ZSTD_CCtx_setParametersUsingCCtxParams(
ZSTD_CCtx* cctx, const ZSTD_CCtx_params* params)
{
- if (cctx->streamStage != zcss_init) return ERROR(stage_wrong);
- if (cctx->cdict) return ERROR(stage_wrong);
+ DEBUGLOG(4, "ZSTD_CCtx_setParametersUsingCCtxParams");
+ RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong);
+ RETURN_ERROR_IF(cctx->cdict, stage_wrong);
cctx->requestedParams = *params;
return 0;
@@ -497,33 +815,71 @@
ZSTDLIB_API size_t ZSTD_CCtx_setPledgedSrcSize(ZSTD_CCtx* cctx, unsigned long long pledgedSrcSize)
{
DEBUGLOG(4, "ZSTD_CCtx_setPledgedSrcSize to %u bytes", (U32)pledgedSrcSize);
- if (cctx->streamStage != zcss_init) return ERROR(stage_wrong);
+ RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong);
cctx->pledgedSrcSizePlusOne = pledgedSrcSize+1;
return 0;
}
+/**
+ * Initializes the local dict using the requested parameters.
+ * NOTE: This does not use the pledged src size, because it may be used for more
+ * than one compression.
+ */
+static size_t ZSTD_initLocalDict(ZSTD_CCtx* cctx)
+{
+ ZSTD_localDict* const dl = &cctx->localDict;
+ ZSTD_compressionParameters const cParams = ZSTD_getCParamsFromCCtxParams(
+ &cctx->requestedParams, 0, dl->dictSize);
+ if (dl->dict == NULL) {
+ /* No local dictionary. */
+ assert(dl->dictBuffer == NULL);
+ assert(dl->cdict == NULL);
+ assert(dl->dictSize == 0);
+ return 0;
+ }
+ if (dl->cdict != NULL) {
+ assert(cctx->cdict == dl->cdict);
+ /* Local dictionary already initialized. */
+ return 0;
+ }
+ assert(dl->dictSize > 0);
+ assert(cctx->cdict == NULL);
+ assert(cctx->prefixDict.dict == NULL);
+
+ dl->cdict = ZSTD_createCDict_advanced(
+ dl->dict,
+ dl->dictSize,
+ ZSTD_dlm_byRef,
+ dl->dictContentType,
+ cParams,
+ cctx->customMem);
+ RETURN_ERROR_IF(!dl->cdict, memory_allocation);
+ cctx->cdict = dl->cdict;
+ return 0;
+}
+
size_t ZSTD_CCtx_loadDictionary_advanced(
ZSTD_CCtx* cctx, const void* dict, size_t dictSize,
ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType)
{
- if (cctx->streamStage != zcss_init) return ERROR(stage_wrong);
- if (cctx->staticSize) return ERROR(memory_allocation); /* no malloc for static CCtx */
+ RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong);
+ RETURN_ERROR_IF(cctx->staticSize, memory_allocation,
+ "no malloc for static CCtx");
DEBUGLOG(4, "ZSTD_CCtx_loadDictionary_advanced (size: %u)", (U32)dictSize);
- ZSTD_freeCDict(cctx->cdictLocal); /* in case one already exists */
- if (dict==NULL || dictSize==0) { /* no dictionary mode */
- cctx->cdictLocal = NULL;
- cctx->cdict = NULL;
+ ZSTD_clearAllDicts(cctx); /* in case one already exists */
+ if (dict == NULL || dictSize == 0) /* no dictionary mode */
+ return 0;
+ if (dictLoadMethod == ZSTD_dlm_byRef) {
+ cctx->localDict.dict = dict;
} else {
- ZSTD_compressionParameters const cParams =
- ZSTD_getCParamsFromCCtxParams(&cctx->requestedParams, cctx->pledgedSrcSizePlusOne-1, dictSize);
- cctx->cdictLocal = ZSTD_createCDict_advanced(
- dict, dictSize,
- dictLoadMethod, dictContentType,
- cParams, cctx->customMem);
- cctx->cdict = cctx->cdictLocal;
- if (cctx->cdictLocal == NULL)
- return ERROR(memory_allocation);
+ void* dictBuffer = ZSTD_malloc(dictSize, cctx->customMem);
+ RETURN_ERROR_IF(!dictBuffer, memory_allocation);
+ memcpy(dictBuffer, dict, dictSize);
+ cctx->localDict.dictBuffer = dictBuffer;
+ cctx->localDict.dict = dictBuffer;
}
+ cctx->localDict.dictSize = dictSize;
+ cctx->localDict.dictContentType = dictContentType;
return 0;
}
@@ -543,9 +899,10 @@
size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict)
{
- if (cctx->streamStage != zcss_init) return ERROR(stage_wrong);
+ RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong);
+ /* Free the existing local cdict (if any) to save memory. */
+ ZSTD_clearAllDicts(cctx);
cctx->cdict = cdict;
- memset(&cctx->prefixDict, 0, sizeof(cctx->prefixDict)); /* exclusive */
return 0;
}
@@ -557,61 +914,67 @@
size_t ZSTD_CCtx_refPrefix_advanced(
ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType)
{
- if (cctx->streamStage != zcss_init) return ERROR(stage_wrong);
- cctx->cdict = NULL; /* prefix discards any prior cdict */
+ RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong);
+ ZSTD_clearAllDicts(cctx);
cctx->prefixDict.dict = prefix;
cctx->prefixDict.dictSize = prefixSize;
cctx->prefixDict.dictContentType = dictContentType;
return 0;
}
-static void ZSTD_startNewCompression(ZSTD_CCtx* cctx)
-{
- cctx->streamStage = zcss_init;
- cctx->pledgedSrcSizePlusOne = 0;
-}
-
/*! ZSTD_CCtx_reset() :
* Also dumps dictionary */
-void ZSTD_CCtx_reset(ZSTD_CCtx* cctx)
+size_t ZSTD_CCtx_reset(ZSTD_CCtx* cctx, ZSTD_ResetDirective reset)
{
- ZSTD_startNewCompression(cctx);
- cctx->cdict = NULL;
+ if ( (reset == ZSTD_reset_session_only)
+ || (reset == ZSTD_reset_session_and_parameters) ) {
+ cctx->streamStage = zcss_init;
+ cctx->pledgedSrcSizePlusOne = 0;
+ }
+ if ( (reset == ZSTD_reset_parameters)
+ || (reset == ZSTD_reset_session_and_parameters) ) {
+ RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong);
+ ZSTD_clearAllDicts(cctx);
+ return ZSTD_CCtxParams_reset(&cctx->requestedParams);
+ }
+ return 0;
}
+
/** ZSTD_checkCParams() :
control CParam values remain within authorized range.
@return : 0, or an error code if one value is beyond authorized range */
size_t ZSTD_checkCParams(ZSTD_compressionParameters cParams)
{
- CLAMPCHECK(cParams.windowLog, ZSTD_WINDOWLOG_MIN, ZSTD_WINDOWLOG_MAX);
- CLAMPCHECK(cParams.chainLog, ZSTD_CHAINLOG_MIN, ZSTD_CHAINLOG_MAX);
- CLAMPCHECK(cParams.hashLog, ZSTD_HASHLOG_MIN, ZSTD_HASHLOG_MAX);
- CLAMPCHECK(cParams.searchLog, ZSTD_SEARCHLOG_MIN, ZSTD_SEARCHLOG_MAX);
- CLAMPCHECK(cParams.searchLength, ZSTD_SEARCHLENGTH_MIN, ZSTD_SEARCHLENGTH_MAX);
- if ((U32)(cParams.targetLength) < ZSTD_TARGETLENGTH_MIN)
- return ERROR(parameter_unsupported);
- if ((U32)(cParams.strategy) > (U32)ZSTD_btultra)
- return ERROR(parameter_unsupported);
+ BOUNDCHECK(ZSTD_c_windowLog, (int)cParams.windowLog);
+ BOUNDCHECK(ZSTD_c_chainLog, (int)cParams.chainLog);
+ BOUNDCHECK(ZSTD_c_hashLog, (int)cParams.hashLog);
+ BOUNDCHECK(ZSTD_c_searchLog, (int)cParams.searchLog);
+ BOUNDCHECK(ZSTD_c_minMatch, (int)cParams.minMatch);
+ BOUNDCHECK(ZSTD_c_targetLength,(int)cParams.targetLength);
+ BOUNDCHECK(ZSTD_c_strategy, cParams.strategy);
return 0;
}
/** ZSTD_clampCParams() :
* make CParam values within valid range.
* @return : valid CParams */
-static ZSTD_compressionParameters ZSTD_clampCParams(ZSTD_compressionParameters cParams)
+static ZSTD_compressionParameters
+ZSTD_clampCParams(ZSTD_compressionParameters cParams)
{
-# define CLAMP(val,min,max) { \
- if (val<min) val=min; \
- else if (val>max) val=max; \
+# define CLAMP_TYPE(cParam, val, type) { \
+ ZSTD_bounds const bounds = ZSTD_cParam_getBounds(cParam); \
+ if ((int)val<bounds.lowerBound) val=(type)bounds.lowerBound; \
+ else if ((int)val>bounds.upperBound) val=(type)bounds.upperBound; \
}
- CLAMP(cParams.windowLog, ZSTD_WINDOWLOG_MIN, ZSTD_WINDOWLOG_MAX);
- CLAMP(cParams.chainLog, ZSTD_CHAINLOG_MIN, ZSTD_CHAINLOG_MAX);
- CLAMP(cParams.hashLog, ZSTD_HASHLOG_MIN, ZSTD_HASHLOG_MAX);
- CLAMP(cParams.searchLog, ZSTD_SEARCHLOG_MIN, ZSTD_SEARCHLOG_MAX);
- CLAMP(cParams.searchLength, ZSTD_SEARCHLENGTH_MIN, ZSTD_SEARCHLENGTH_MAX);
- if ((U32)(cParams.targetLength) < ZSTD_TARGETLENGTH_MIN) cParams.targetLength = ZSTD_TARGETLENGTH_MIN;
- if ((U32)(cParams.strategy) > (U32)ZSTD_btultra) cParams.strategy = ZSTD_btultra;
+# define CLAMP(cParam, val) CLAMP_TYPE(cParam, val, unsigned)
+ CLAMP(ZSTD_c_windowLog, cParams.windowLog);
+ CLAMP(ZSTD_c_chainLog, cParams.chainLog);
+ CLAMP(ZSTD_c_hashLog, cParams.hashLog);
+ CLAMP(ZSTD_c_searchLog, cParams.searchLog);
+ CLAMP(ZSTD_c_minMatch, cParams.minMatch);
+ CLAMP(ZSTD_c_targetLength,cParams.targetLength);
+ CLAMP_TYPE(ZSTD_c_strategy,cParams.strategy, ZSTD_strategy);
return cParams;
}
@@ -624,17 +987,21 @@
}
/** ZSTD_adjustCParams_internal() :
- optimize `cPar` for a given input (`srcSize` and `dictSize`).
- mostly downsizing to reduce memory consumption and initialization latency.
- Both `srcSize` and `dictSize` are optional (use 0 if unknown).
- Note : cPar is considered validated at this stage. Use ZSTD_checkCParams() to ensure that condition. */
-ZSTD_compressionParameters ZSTD_adjustCParams_internal(ZSTD_compressionParameters cPar, unsigned long long srcSize, size_t dictSize)
+ * optimize `cPar` for a specified input (`srcSize` and `dictSize`).
+ * mostly downsize to reduce memory consumption and initialization latency.
+ * `srcSize` can be ZSTD_CONTENTSIZE_UNKNOWN when not known.
+ * note : for the time being, `srcSize==0` means "unknown" too, for compatibility with older convention.
+ * condition : cPar is presumed validated (can be checked using ZSTD_checkCParams()). */
+static ZSTD_compressionParameters
+ZSTD_adjustCParams_internal(ZSTD_compressionParameters cPar,
+ unsigned long long srcSize,
+ size_t dictSize)
{
static const U64 minSrcSize = 513; /* (1<<9) + 1 */
static const U64 maxWindowResize = 1ULL << (ZSTD_WINDOWLOG_MAX-1);
assert(ZSTD_checkCParams(cPar)==0);
- if (dictSize && (srcSize+1<2) /* srcSize unknown */ )
+ if (dictSize && (srcSize+1<2) /* ZSTD_CONTENTSIZE_UNKNOWN and 0 mean "unknown" */ )
srcSize = minSrcSize; /* presumed small when there is a dictionary */
else if (srcSize == 0)
srcSize = ZSTD_CONTENTSIZE_UNKNOWN; /* 0 == unknown : presumed large */
@@ -648,35 +1015,55 @@
ZSTD_highbit32(tSize-1) + 1;
if (cPar.windowLog > srcLog) cPar.windowLog = srcLog;
}
- if (cPar.hashLog > cPar.windowLog) cPar.hashLog = cPar.windowLog;
+ if (cPar.hashLog > cPar.windowLog+1) cPar.hashLog = cPar.windowLog+1;
{ U32 const cycleLog = ZSTD_cycleLog(cPar.chainLog, cPar.strategy);
if (cycleLog > cPar.windowLog)
cPar.chainLog -= (cycleLog - cPar.windowLog);
}
if (cPar.windowLog < ZSTD_WINDOWLOG_ABSOLUTEMIN)
- cPar.windowLog = ZSTD_WINDOWLOG_ABSOLUTEMIN; /* required for frame header */
+ cPar.windowLog = ZSTD_WINDOWLOG_ABSOLUTEMIN; /* minimum wlog required for valid frame header */
return cPar;
}
-ZSTD_compressionParameters ZSTD_adjustCParams(ZSTD_compressionParameters cPar, unsigned long long srcSize, size_t dictSize)
+ZSTD_compressionParameters
+ZSTD_adjustCParams(ZSTD_compressionParameters cPar,
+ unsigned long long srcSize,
+ size_t dictSize)
{
- cPar = ZSTD_clampCParams(cPar);
+ cPar = ZSTD_clampCParams(cPar); /* resulting cPar is necessarily valid (all parameters within range) */
return ZSTD_adjustCParams_internal(cPar, srcSize, dictSize);
}
-static size_t ZSTD_sizeof_matchState(ZSTD_compressionParameters const* cParams, const U32 forCCtx)
+ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams(
+ const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize)
+{
+ ZSTD_compressionParameters cParams = ZSTD_getCParams(CCtxParams->compressionLevel, srcSizeHint, dictSize);
+ if (CCtxParams->ldmParams.enableLdm) cParams.windowLog = ZSTD_LDM_DEFAULT_WINDOW_LOG;
+ if (CCtxParams->cParams.windowLog) cParams.windowLog = CCtxParams->cParams.windowLog;
+ if (CCtxParams->cParams.hashLog) cParams.hashLog = CCtxParams->cParams.hashLog;
+ if (CCtxParams->cParams.chainLog) cParams.chainLog = CCtxParams->cParams.chainLog;
+ if (CCtxParams->cParams.searchLog) cParams.searchLog = CCtxParams->cParams.searchLog;
+ if (CCtxParams->cParams.minMatch) cParams.minMatch = CCtxParams->cParams.minMatch;
+ if (CCtxParams->cParams.targetLength) cParams.targetLength = CCtxParams->cParams.targetLength;
+ if (CCtxParams->cParams.strategy) cParams.strategy = CCtxParams->cParams.strategy;
+ assert(!ZSTD_checkCParams(cParams));
+ return ZSTD_adjustCParams_internal(cParams, srcSizeHint, dictSize);
+}
+
+static size_t
+ZSTD_sizeof_matchState(const ZSTD_compressionParameters* const cParams,
+ const U32 forCCtx)
{
size_t const chainSize = (cParams->strategy == ZSTD_fast) ? 0 : ((size_t)1 << cParams->chainLog);
size_t const hSize = ((size_t)1) << cParams->hashLog;
- U32 const hashLog3 = (forCCtx && cParams->searchLength==3) ? MIN(ZSTD_HASHLOG3_MAX, cParams->windowLog) : 0;
+ U32 const hashLog3 = (forCCtx && cParams->minMatch==3) ? MIN(ZSTD_HASHLOG3_MAX, cParams->windowLog) : 0;
size_t const h3Size = ((size_t)1) << hashLog3;
size_t const tableSpace = (chainSize + hSize + h3Size) * sizeof(U32);
size_t const optPotentialSpace = ((MaxML+1) + (MaxLL+1) + (MaxOff+1) + (1<<Litbits)) * sizeof(U32)
+ (ZSTD_OPT_NUM+1) * (sizeof(ZSTD_match_t)+sizeof(ZSTD_optimal_t));
- size_t const optSpace = (forCCtx && ((cParams->strategy == ZSTD_btopt) ||
- (cParams->strategy == ZSTD_btultra)))
+ size_t const optSpace = (forCCtx && (cParams->strategy >= ZSTD_btopt))
? optPotentialSpace
: 0;
DEBUGLOG(4, "chainSize: %u - hSize: %u - h3Size: %u",
@@ -686,14 +1073,13 @@
size_t ZSTD_estimateCCtxSize_usingCCtxParams(const ZSTD_CCtx_params* params)
{
- /* Estimate CCtx size is supported for single-threaded compression only. */
- if (params->nbWorkers > 0) { return ERROR(GENERIC); }
+ RETURN_ERROR_IF(params->nbWorkers > 0, GENERIC, "Estimate CCtx size is supported for single-threaded compression only.");
{ ZSTD_compressionParameters const cParams =
ZSTD_getCParamsFromCCtxParams(params, 0, 0);
size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, (size_t)1 << cParams.windowLog);
- U32 const divider = (cParams.searchLength==3) ? 3 : 4;
+ U32 const divider = (cParams.minMatch==3) ? 3 : 4;
size_t const maxNbSeq = blockSize / divider;
- size_t const tokenSpace = blockSize + 11*maxNbSeq;
+ size_t const tokenSpace = WILDCOPY_OVERLENGTH + blockSize + 11*maxNbSeq;
size_t const entropySpace = HUF_WORKSPACE_SIZE;
size_t const blockStateSpace = 2 * sizeof(ZSTD_compressedBlockState_t);
size_t const matchStateSize = ZSTD_sizeof_matchState(&cParams, /* forCCtx */ 1);
@@ -726,7 +1112,7 @@
{
int level;
size_t memBudget = 0;
- for (level=1; level<=compressionLevel; level++) {
+ for (level=MIN(compressionLevel, 1); level<=compressionLevel; level++) {
size_t const newMB = ZSTD_estimateCCtxSize_internal(level);
if (newMB > memBudget) memBudget = newMB;
}
@@ -735,10 +1121,12 @@
size_t ZSTD_estimateCStreamSize_usingCCtxParams(const ZSTD_CCtx_params* params)
{
- if (params->nbWorkers > 0) { return ERROR(GENERIC); }
- { size_t const CCtxSize = ZSTD_estimateCCtxSize_usingCCtxParams(params);
- size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, (size_t)1 << params->cParams.windowLog);
- size_t const inBuffSize = ((size_t)1 << params->cParams.windowLog) + blockSize;
+ RETURN_ERROR_IF(params->nbWorkers > 0, GENERIC, "Estimate CCtx size is supported for single-threaded compression only.");
+ { ZSTD_compressionParameters const cParams =
+ ZSTD_getCParamsFromCCtxParams(params, 0, 0);
+ size_t const CCtxSize = ZSTD_estimateCCtxSize_usingCCtxParams(params);
+ size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, (size_t)1 << cParams.windowLog);
+ size_t const inBuffSize = ((size_t)1 << cParams.windowLog) + blockSize;
size_t const outBuffSize = ZSTD_compressBound(blockSize) + 1;
size_t const streamingSize = inBuffSize + outBuffSize;
@@ -752,15 +1140,17 @@
return ZSTD_estimateCStreamSize_usingCCtxParams(¶ms);
}
-static size_t ZSTD_estimateCStreamSize_internal(int compressionLevel) {
+static size_t ZSTD_estimateCStreamSize_internal(int compressionLevel)
+{
ZSTD_compressionParameters const cParams = ZSTD_getCParams(compressionLevel, 0, 0);
return ZSTD_estimateCStreamSize_usingCParams(cParams);
}
-size_t ZSTD_estimateCStreamSize(int compressionLevel) {
+size_t ZSTD_estimateCStreamSize(int compressionLevel)
+{
int level;
size_t memBudget = 0;
- for (level=1; level<=compressionLevel; level++) {
+ for (level=MIN(compressionLevel, 1); level<=compressionLevel; level++) {
size_t const newMB = ZSTD_estimateCStreamSize_internal(level);
if (newMB > memBudget) memBudget = newMB;
}
@@ -786,9 +1176,27 @@
fp.ingested = cctx->consumedSrcSize + buffered;
fp.consumed = cctx->consumedSrcSize;
fp.produced = cctx->producedCSize;
+ fp.flushed = cctx->producedCSize; /* simplified; some data might still be left within streaming output buffer */
+ fp.currentJobID = 0;
+ fp.nbActiveWorkers = 0;
return fp;
} }
+/*! ZSTD_toFlushNow()
+ * Only useful for multithreading scenarios currently (nbWorkers >= 1).
+ */
+size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx)
+{
+#ifdef ZSTD_MULTITHREAD
+ if (cctx->appliedParams.nbWorkers > 0) {
+ return ZSTDMT_toFlushNow(cctx->mtctx);
+ }
+#endif
+ (void)cctx;
+ return 0; /* over-simplification; could also check if context is currently running in streaming mode, and in which case, report how many bytes are left to be flushed within output buffer */
+}
+
+
static U32 ZSTD_equivalentCParams(ZSTD_compressionParameters cParams1,
ZSTD_compressionParameters cParams2)
@@ -796,7 +1204,21 @@
return (cParams1.hashLog == cParams2.hashLog)
& (cParams1.chainLog == cParams2.chainLog)
& (cParams1.strategy == cParams2.strategy) /* opt parser space */
- & ((cParams1.searchLength==3) == (cParams2.searchLength==3)); /* hashlog3 space */
+ & ((cParams1.minMatch==3) == (cParams2.minMatch==3)); /* hashlog3 space */
+}
+
+static void ZSTD_assertEqualCParams(ZSTD_compressionParameters cParams1,
+ ZSTD_compressionParameters cParams2)
+{
+ (void)cParams1;
+ (void)cParams2;
+ assert(cParams1.windowLog == cParams2.windowLog);
+ assert(cParams1.chainLog == cParams2.chainLog);
+ assert(cParams1.hashLog == cParams2.hashLog);
+ assert(cParams1.searchLog == cParams2.searchLog);
+ assert(cParams1.minMatch == cParams2.minMatch);
+ assert(cParams1.targetLength == cParams2.targetLength);
+ assert(cParams1.strategy == cParams2.strategy);
}
/** The parameters are equivalent if ldm is not enabled in both sets or
@@ -809,7 +1231,7 @@
ldmParams1.hashLog == ldmParams2.hashLog &&
ldmParams1.bucketSizeLog == ldmParams2.bucketSizeLog &&
ldmParams1.minMatchLength == ldmParams2.minMatchLength &&
- ldmParams1.hashEveryLog == ldmParams2.hashEveryLog);
+ ldmParams1.hashRateLog == ldmParams2.hashRateLog);
}
typedef enum { ZSTDb_not_buffered, ZSTDb_buffered } ZSTD_buffered_policy_e;
@@ -817,33 +1239,51 @@
/* ZSTD_sufficientBuff() :
* check internal buffers exist for streaming if buffPol == ZSTDb_buffered .
* Note : they are assumed to be correctly sized if ZSTD_equivalentCParams()==1 */
-static U32 ZSTD_sufficientBuff(size_t bufferSize1, size_t blockSize1,
+static U32 ZSTD_sufficientBuff(size_t bufferSize1, size_t maxNbSeq1,
+ size_t maxNbLit1,
ZSTD_buffered_policy_e buffPol2,
ZSTD_compressionParameters cParams2,
U64 pledgedSrcSize)
{
size_t const windowSize2 = MAX(1, (size_t)MIN(((U64)1 << cParams2.windowLog), pledgedSrcSize));
size_t const blockSize2 = MIN(ZSTD_BLOCKSIZE_MAX, windowSize2);
+ size_t const maxNbSeq2 = blockSize2 / ((cParams2.minMatch == 3) ? 3 : 4);
+ size_t const maxNbLit2 = blockSize2;
size_t const neededBufferSize2 = (buffPol2==ZSTDb_buffered) ? windowSize2 + blockSize2 : 0;
- DEBUGLOG(4, "ZSTD_sufficientBuff: is windowSize2=%u <= wlog1=%u",
- (U32)windowSize2, cParams2.windowLog);
- DEBUGLOG(4, "ZSTD_sufficientBuff: is blockSize2=%u <= blockSize1=%u",
- (U32)blockSize2, (U32)blockSize1);
- return (blockSize2 <= blockSize1) /* seqStore space depends on blockSize */
+ DEBUGLOG(4, "ZSTD_sufficientBuff: is neededBufferSize2=%u <= bufferSize1=%u",
+ (U32)neededBufferSize2, (U32)bufferSize1);
+ DEBUGLOG(4, "ZSTD_sufficientBuff: is maxNbSeq2=%u <= maxNbSeq1=%u",
+ (U32)maxNbSeq2, (U32)maxNbSeq1);
+ DEBUGLOG(4, "ZSTD_sufficientBuff: is maxNbLit2=%u <= maxNbLit1=%u",
+ (U32)maxNbLit2, (U32)maxNbLit1);
+ return (maxNbLit2 <= maxNbLit1)
+ & (maxNbSeq2 <= maxNbSeq1)
& (neededBufferSize2 <= bufferSize1);
}
/** Equivalence for resetCCtx purposes */
static U32 ZSTD_equivalentParams(ZSTD_CCtx_params params1,
ZSTD_CCtx_params params2,
- size_t buffSize1, size_t blockSize1,
+ size_t buffSize1,
+ size_t maxNbSeq1, size_t maxNbLit1,
ZSTD_buffered_policy_e buffPol2,
U64 pledgedSrcSize)
{
DEBUGLOG(4, "ZSTD_equivalentParams: pledgedSrcSize=%u", (U32)pledgedSrcSize);
- return ZSTD_equivalentCParams(params1.cParams, params2.cParams) &&
- ZSTD_equivalentLdmParams(params1.ldmParams, params2.ldmParams) &&
- ZSTD_sufficientBuff(buffSize1, blockSize1, buffPol2, params2.cParams, pledgedSrcSize);
+ if (!ZSTD_equivalentCParams(params1.cParams, params2.cParams)) {
+ DEBUGLOG(4, "ZSTD_equivalentCParams() == 0");
+ return 0;
+ }
+ if (!ZSTD_equivalentLdmParams(params1.ldmParams, params2.ldmParams)) {
+ DEBUGLOG(4, "ZSTD_equivalentLdmParams() == 0");
+ return 0;
+ }
+ if (!ZSTD_sufficientBuff(buffSize1, maxNbSeq1, maxNbLit1, buffPol2,
+ params2.cParams, pledgedSrcSize)) {
+ DEBUGLOG(4, "ZSTD_sufficientBuff() == 0");
+ return 0;
+ }
+ return 1;
}
static void ZSTD_reset_compressedBlockState(ZSTD_compressedBlockState_t* bs)
@@ -851,23 +1291,24 @@
int i;
for (i = 0; i < ZSTD_REP_NUM; ++i)
bs->rep[i] = repStartValue[i];
- bs->entropy.hufCTable_repeatMode = HUF_repeat_none;
- bs->entropy.offcode_repeatMode = FSE_repeat_none;
- bs->entropy.matchlength_repeatMode = FSE_repeat_none;
- bs->entropy.litlength_repeatMode = FSE_repeat_none;
+ bs->entropy.huf.repeatMode = HUF_repeat_none;
+ bs->entropy.fse.offcode_repeatMode = FSE_repeat_none;
+ bs->entropy.fse.matchlength_repeatMode = FSE_repeat_none;
+ bs->entropy.fse.litlength_repeatMode = FSE_repeat_none;
}
/*! ZSTD_invalidateMatchState()
- * Invalidate all the matches in the match finder tables.
- * Requires nextSrc and base to be set (can be NULL).
+ * Invalidate all the matches in the match finder tables.
+ * Requires nextSrc and base to be set (can be NULL).
*/
static void ZSTD_invalidateMatchState(ZSTD_matchState_t* ms)
{
ZSTD_window_clear(&ms->window);
- ms->nextToUpdate = ms->window.dictLimit + 1;
+ ms->nextToUpdate = ms->window.dictLimit;
ms->loadedDictEnd = 0;
ms->opt.litLengthSum = 0; /* force reset of btopt stats */
+ ms->dictMatchState = NULL;
}
/*! ZSTD_continueCCtx() :
@@ -880,6 +1321,7 @@
cctx->blockSize = blockSize; /* previous block size could be different even for same windowLog, due to pledgedSrcSize */
cctx->appliedParams = params;
+ cctx->blockState.matchState.cParams = params.cParams;
cctx->pledgedSrcSizePlusOne = pledgedSrcSize+1;
cctx->consumedSrcSize = 0;
cctx->producedCSize = 0;
@@ -900,11 +1342,17 @@
typedef enum { ZSTDcrp_continue, ZSTDcrp_noMemset } ZSTD_compResetPolicy_e;
-static void* ZSTD_reset_matchState(ZSTD_matchState_t* ms, void* ptr, ZSTD_compressionParameters const* cParams, ZSTD_compResetPolicy_e const crp, U32 const forCCtx)
+typedef enum { ZSTD_resetTarget_CDict, ZSTD_resetTarget_CCtx } ZSTD_resetTarget_e;
+
+static void*
+ZSTD_reset_matchState(ZSTD_matchState_t* ms,
+ void* ptr,
+ const ZSTD_compressionParameters* cParams,
+ ZSTD_compResetPolicy_e const crp, ZSTD_resetTarget_e const forWho)
{
size_t const chainSize = (cParams->strategy == ZSTD_fast) ? 0 : ((size_t)1 << cParams->chainLog);
size_t const hSize = ((size_t)1) << cParams->hashLog;
- U32 const hashLog3 = (forCCtx && cParams->searchLength==3) ? MIN(ZSTD_HASHLOG3_MAX, cParams->windowLog) : 0;
+ U32 const hashLog3 = ((forWho == ZSTD_resetTarget_CCtx) && cParams->minMatch==3) ? MIN(ZSTD_HASHLOG3_MAX, cParams->windowLog) : 0;
size_t const h3Size = ((size_t)1) << hashLog3;
size_t const tableSpace = (chainSize + hSize + h3Size) * sizeof(U32);
@@ -912,12 +1360,15 @@
ms->hashLog3 = hashLog3;
memset(&ms->window, 0, sizeof(ms->window));
+ ms->window.dictLimit = 1; /* start from 1, so that 1st position is valid */
+ ms->window.lowLimit = 1; /* it ensures first and later CCtx usages compress the same */
+ ms->window.nextSrc = ms->window.base + 1; /* see issue #1241 */
ZSTD_invalidateMatchState(ms);
/* opt parser space */
- if (forCCtx && ((cParams->strategy == ZSTD_btopt) | (cParams->strategy == ZSTD_btultra))) {
+ if ((forWho == ZSTD_resetTarget_CCtx) && (cParams->strategy >= ZSTD_btopt)) {
DEBUGLOG(4, "reserving optimal parser space");
- ms->opt.litFreq = (U32*)ptr;
+ ms->opt.litFreq = (unsigned*)ptr;
ms->opt.litLengthFreq = ms->opt.litFreq + (1<<Litbits);
ms->opt.matchLengthFreq = ms->opt.litLengthFreq + (MaxLL+1);
ms->opt.offCodeFreq = ms->opt.matchLengthFreq + (MaxML+1);
@@ -937,14 +1388,37 @@
ms->hashTable3 = ms->chainTable + chainSize;
ptr = ms->hashTable3 + h3Size;
+ ms->cParams = *cParams;
+
assert(((size_t)ptr & 3) == 0);
return ptr;
}
+/* ZSTD_indexTooCloseToMax() :
+ * minor optimization : prefer memset() rather than reduceIndex()
+ * which is measurably slow in some circumstances (reported for Visual Studio).
+ * Works when re-using a context for a lot of smallish inputs :
+ * if all inputs are smaller than ZSTD_INDEXOVERFLOW_MARGIN,
+ * memset() will be triggered before reduceIndex().
+ */
+#define ZSTD_INDEXOVERFLOW_MARGIN (16 MB)
+static int ZSTD_indexTooCloseToMax(ZSTD_window_t w)
+{
+ return (size_t)(w.nextSrc - w.base) > (ZSTD_CURRENT_MAX - ZSTD_INDEXOVERFLOW_MARGIN);
+}
+
+#define ZSTD_WORKSPACETOOLARGE_FACTOR 3 /* define "workspace is too large" as this number of times larger than needed */
+#define ZSTD_WORKSPACETOOLARGE_MAXDURATION 128 /* when workspace is continuously too large
+ * during at least this number of times,
+ * context's memory usage is considered wasteful,
+ * because it's sized to handle a worst case scenario which rarely happens.
+ * In which case, resize it down to free some memory */
+
/*! ZSTD_resetCCtx_internal() :
note : `params` are assumed fully validated at this stage */
static size_t ZSTD_resetCCtx_internal(ZSTD_CCtx* zc,
- ZSTD_CCtx_params params, U64 pledgedSrcSize,
+ ZSTD_CCtx_params params,
+ U64 const pledgedSrcSize,
ZSTD_compResetPolicy_e const crp,
ZSTD_buffered_policy_e const zbuff)
{
@@ -954,34 +1428,43 @@
if (crp == ZSTDcrp_continue) {
if (ZSTD_equivalentParams(zc->appliedParams, params,
- zc->inBuffSize, zc->blockSize,
- zbuff, pledgedSrcSize)) {
- DEBUGLOG(4, "ZSTD_equivalentParams()==1 -> continue mode (wLog1=%u, blockSize1=%u)",
- zc->appliedParams.cParams.windowLog, (U32)zc->blockSize);
- return ZSTD_continueCCtx(zc, params, pledgedSrcSize);
- } }
+ zc->inBuffSize,
+ zc->seqStore.maxNbSeq, zc->seqStore.maxNbLit,
+ zbuff, pledgedSrcSize) ) {
+ DEBUGLOG(4, "ZSTD_equivalentParams()==1 -> consider continue mode");
+ zc->workSpaceOversizedDuration += (zc->workSpaceOversizedDuration > 0); /* if it was too large, it still is */
+ if (zc->workSpaceOversizedDuration <= ZSTD_WORKSPACETOOLARGE_MAXDURATION) {
+ DEBUGLOG(4, "continue mode confirmed (wLog1=%u, blockSize1=%zu)",
+ zc->appliedParams.cParams.windowLog, zc->blockSize);
+ if (ZSTD_indexTooCloseToMax(zc->blockState.matchState.window)) {
+ /* prefer a reset, faster than a rescale */
+ ZSTD_reset_matchState(&zc->blockState.matchState,
+ zc->entropyWorkspace + HUF_WORKSPACE_SIZE_U32,
+ ¶ms.cParams,
+ crp, ZSTD_resetTarget_CCtx);
+ }
+ return ZSTD_continueCCtx(zc, params, pledgedSrcSize);
+ } } }
DEBUGLOG(4, "ZSTD_equivalentParams()==0 -> reset CCtx");
if (params.ldmParams.enableLdm) {
/* Adjust long distance matching parameters */
- params.ldmParams.windowLog = params.cParams.windowLog;
ZSTD_ldm_adjustParameters(¶ms.ldmParams, ¶ms.cParams);
assert(params.ldmParams.hashLog >= params.ldmParams.bucketSizeLog);
- assert(params.ldmParams.hashEveryLog < 32);
- zc->ldmState.hashPower =
- ZSTD_ldm_getHashPower(params.ldmParams.minMatchLength);
+ assert(params.ldmParams.hashRateLog < 32);
+ zc->ldmState.hashPower = ZSTD_rollingHash_primePower(params.ldmParams.minMatchLength);
}
{ size_t const windowSize = MAX(1, (size_t)MIN(((U64)1 << params.cParams.windowLog), pledgedSrcSize));
size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, windowSize);
- U32 const divider = (params.cParams.searchLength==3) ? 3 : 4;
+ U32 const divider = (params.cParams.minMatch==3) ? 3 : 4;
size_t const maxNbSeq = blockSize / divider;
- size_t const tokenSpace = blockSize + 11*maxNbSeq;
+ size_t const tokenSpace = WILDCOPY_OVERLENGTH + blockSize + 11*maxNbSeq;
size_t const buffOutSize = (zbuff==ZSTDb_buffered) ? ZSTD_compressBound(blockSize)+1 : 0;
size_t const buffInSize = (zbuff==ZSTDb_buffered) ? windowSize + blockSize : 0;
size_t const matchStateSize = ZSTD_sizeof_matchState(¶ms.cParams, /* forCCtx */ 1);
size_t const maxNbLdmSeq = ZSTD_ldm_getMaxNbSeq(params.ldmParams, blockSize);
- void* ptr;
+ void* ptr; /* used to partition workSpace */
/* Check if workSpace is large enough, alloc a new one if needed */
{ size_t const entropySpace = HUF_WORKSPACE_SIZE;
@@ -993,25 +1476,33 @@
size_t const neededSpace = entropySpace + blockStateSpace + ldmSpace +
ldmSeqSpace + matchStateSize + tokenSpace +
bufferSpace;
- DEBUGLOG(4, "Need %uKB workspace, including %uKB for match state, and %uKB for buffers",
- (U32)(neededSpace>>10), (U32)(matchStateSize>>10), (U32)(bufferSpace>>10));
- DEBUGLOG(4, "windowSize: %u - blockSize: %u", (U32)windowSize, (U32)blockSize);
- if (zc->workSpaceSize < neededSpace) { /* too small : resize */
- DEBUGLOG(4, "Need to update workSpaceSize from %uK to %uK",
- (unsigned)(zc->workSpaceSize>>10),
- (unsigned)(neededSpace>>10));
- /* static cctx : no resize, error out */
- if (zc->staticSize) return ERROR(memory_allocation);
+ int const workSpaceTooSmall = zc->workSpaceSize < neededSpace;
+ int const workSpaceTooLarge = zc->workSpaceSize > ZSTD_WORKSPACETOOLARGE_FACTOR * neededSpace;
+ int const workSpaceWasteful = workSpaceTooLarge && (zc->workSpaceOversizedDuration > ZSTD_WORKSPACETOOLARGE_MAXDURATION);
+ zc->workSpaceOversizedDuration = workSpaceTooLarge ? zc->workSpaceOversizedDuration+1 : 0;
+
+ DEBUGLOG(4, "Need %zuKB workspace, including %zuKB for match state, and %zuKB for buffers",
+ neededSpace>>10, matchStateSize>>10, bufferSpace>>10);
+ DEBUGLOG(4, "windowSize: %zu - blockSize: %zu", windowSize, blockSize);
+
+ if (workSpaceTooSmall || workSpaceWasteful) {
+ DEBUGLOG(4, "Resize workSpaceSize from %zuKB to %zuKB",
+ zc->workSpaceSize >> 10,
+ neededSpace >> 10);
+
+ RETURN_ERROR_IF(zc->staticSize, memory_allocation, "static cctx : no resize");
zc->workSpaceSize = 0;
ZSTD_free(zc->workSpace, zc->customMem);
zc->workSpace = ZSTD_malloc(neededSpace, zc->customMem);
- if (zc->workSpace == NULL) return ERROR(memory_allocation);
+ RETURN_ERROR_IF(zc->workSpace == NULL, memory_allocation);
zc->workSpaceSize = neededSpace;
- ptr = zc->workSpace;
+ zc->workSpaceOversizedDuration = 0;
- /* Statically sized space. entropyWorkspace never moves (but prev/next block swap places) */
+ /* Statically sized space.
+ * entropyWorkspace never moves,
+ * though prev/next block swap places */
assert(((size_t)zc->workSpace & 3) == 0); /* ensure correct alignment */
assert(zc->workSpaceSize >= 2 * sizeof(ZSTD_compressedBlockState_t));
zc->blockState.prevCBlock = (ZSTD_compressedBlockState_t*)zc->workSpace;
@@ -1022,13 +1513,14 @@
/* init params */
zc->appliedParams = params;
+ zc->blockState.matchState.cParams = params.cParams;
zc->pledgedSrcSizePlusOne = pledgedSrcSize+1;
zc->consumedSrcSize = 0;
zc->producedCSize = 0;
if (pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN)
zc->appliedParams.fParams.contentSizeFlag = 0;
DEBUGLOG(4, "pledged content size : %u ; flag : %u",
- (U32)pledgedSrcSize, zc->appliedParams.fParams.contentSizeFlag);
+ (unsigned)pledgedSrcSize, zc->appliedParams.fParams.contentSizeFlag);
zc->blockSize = blockSize;
XXH64_reset(&zc->xxhState, 0);
@@ -1037,7 +1529,10 @@
ZSTD_reset_compressedBlockState(zc->blockState.prevCBlock);
- ptr = zc->entropyWorkspace + HUF_WORKSPACE_SIZE_U32;
+ ptr = ZSTD_reset_matchState(&zc->blockState.matchState,
+ zc->entropyWorkspace + HUF_WORKSPACE_SIZE_U32,
+ ¶ms.cParams,
+ crp, ZSTD_resetTarget_CCtx);
/* ldm hash table */
/* initialize bucketOffsets table later for pointer alignment */
@@ -1055,16 +1550,19 @@
}
assert(((size_t)ptr & 3) == 0); /* ensure ptr is properly aligned */
- ptr = ZSTD_reset_matchState(&zc->blockState.matchState, ptr, ¶ms.cParams, crp, /* forCCtx */ 1);
-
/* sequences storage */
+ zc->seqStore.maxNbSeq = maxNbSeq;
zc->seqStore.sequencesStart = (seqDef*)ptr;
ptr = zc->seqStore.sequencesStart + maxNbSeq;
zc->seqStore.llCode = (BYTE*) ptr;
zc->seqStore.mlCode = zc->seqStore.llCode + maxNbSeq;
zc->seqStore.ofCode = zc->seqStore.mlCode + maxNbSeq;
zc->seqStore.litStart = zc->seqStore.ofCode + maxNbSeq;
- ptr = zc->seqStore.litStart + blockSize;
+ /* ZSTD_wildcopy() is used to copy into the literals buffer,
+ * so we have to oversize the buffer by WILDCOPY_OVERLENGTH bytes.
+ */
+ zc->seqStore.maxNbLit = blockSize;
+ ptr = zc->seqStore.litStart + blockSize + WILDCOPY_OVERLENGTH;
/* ldm bucketOffsets table */
if (params.ldmParams.enableLdm) {
@@ -1098,28 +1596,109 @@
assert(!ZSTD_window_hasExtDict(cctx->blockState.matchState.window));
}
-static size_t ZSTD_resetCCtx_usingCDict(ZSTD_CCtx* cctx,
+/* These are the approximate sizes for each strategy past which copying the
+ * dictionary tables into the working context is faster than using them
+ * in-place.
+ */
+static const size_t attachDictSizeCutoffs[ZSTD_STRATEGY_MAX+1] = {
+ 8 KB, /* unused */
+ 8 KB, /* ZSTD_fast */
+ 16 KB, /* ZSTD_dfast */
+ 32 KB, /* ZSTD_greedy */
+ 32 KB, /* ZSTD_lazy */
+ 32 KB, /* ZSTD_lazy2 */
+ 32 KB, /* ZSTD_btlazy2 */
+ 32 KB, /* ZSTD_btopt */
+ 8 KB, /* ZSTD_btultra */
+ 8 KB /* ZSTD_btultra2 */
+};
+
+static int ZSTD_shouldAttachDict(const ZSTD_CDict* cdict,
+ ZSTD_CCtx_params params,
+ U64 pledgedSrcSize)
+{
+ size_t cutoff = attachDictSizeCutoffs[cdict->matchState.cParams.strategy];
+ return ( pledgedSrcSize <= cutoff
+ || pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN
+ || params.attachDictPref == ZSTD_dictForceAttach )
+ && params.attachDictPref != ZSTD_dictForceCopy
+ && !params.forceWindow; /* dictMatchState isn't correctly
+ * handled in _enforceMaxDist */
+}
+
+static size_t
+ZSTD_resetCCtx_byAttachingCDict(ZSTD_CCtx* cctx,
+ const ZSTD_CDict* cdict,
+ ZSTD_CCtx_params params,
+ U64 pledgedSrcSize,
+ ZSTD_buffered_policy_e zbuff)
+{
+ { const ZSTD_compressionParameters* const cdict_cParams = &cdict->matchState.cParams;
+ unsigned const windowLog = params.cParams.windowLog;
+ assert(windowLog != 0);
+ /* Resize working context table params for input only, since the dict
+ * has its own tables. */
+ params.cParams = ZSTD_adjustCParams_internal(*cdict_cParams, pledgedSrcSize, 0);
+ params.cParams.windowLog = windowLog;
+ ZSTD_resetCCtx_internal(cctx, params, pledgedSrcSize,
+ ZSTDcrp_continue, zbuff);
+ assert(cctx->appliedParams.cParams.strategy == cdict_cParams->strategy);
+ }
+
+ { const U32 cdictEnd = (U32)( cdict->matchState.window.nextSrc
+ - cdict->matchState.window.base);
+ const U32 cdictLen = cdictEnd - cdict->matchState.window.dictLimit;
+ if (cdictLen == 0) {
+ /* don't even attach dictionaries with no contents */
+ DEBUGLOG(4, "skipping attaching empty dictionary");
+ } else {
+ DEBUGLOG(4, "attaching dictionary into context");
+ cctx->blockState.matchState.dictMatchState = &cdict->matchState;
+
+ /* prep working match state so dict matches never have negative indices
+ * when they are translated to the working context's index space. */
+ if (cctx->blockState.matchState.window.dictLimit < cdictEnd) {
+ cctx->blockState.matchState.window.nextSrc =
+ cctx->blockState.matchState.window.base + cdictEnd;
+ ZSTD_window_clear(&cctx->blockState.matchState.window);
+ }
+ /* loadedDictEnd is expressed within the referential of the active context */
+ cctx->blockState.matchState.loadedDictEnd = cctx->blockState.matchState.window.dictLimit;
+ } }
+
+ cctx->dictID = cdict->dictID;
+
+ /* copy block state */
+ memcpy(cctx->blockState.prevCBlock, &cdict->cBlockState, sizeof(cdict->cBlockState));
+
+ return 0;
+}
+
+static size_t ZSTD_resetCCtx_byCopyingCDict(ZSTD_CCtx* cctx,
const ZSTD_CDict* cdict,
- unsigned windowLog,
- ZSTD_frameParameters fParams,
+ ZSTD_CCtx_params params,
U64 pledgedSrcSize,
ZSTD_buffered_policy_e zbuff)
{
- { ZSTD_CCtx_params params = cctx->requestedParams;
+ const ZSTD_compressionParameters *cdict_cParams = &cdict->matchState.cParams;
+
+ DEBUGLOG(4, "copying dictionary into context");
+
+ { unsigned const windowLog = params.cParams.windowLog;
+ assert(windowLog != 0);
/* Copy only compression parameters related to tables. */
- params.cParams = cdict->cParams;
- if (windowLog) params.cParams.windowLog = windowLog;
- params.fParams = fParams;
+ params.cParams = *cdict_cParams;
+ params.cParams.windowLog = windowLog;
ZSTD_resetCCtx_internal(cctx, params, pledgedSrcSize,
ZSTDcrp_noMemset, zbuff);
- assert(cctx->appliedParams.cParams.strategy == cdict->cParams.strategy);
- assert(cctx->appliedParams.cParams.hashLog == cdict->cParams.hashLog);
- assert(cctx->appliedParams.cParams.chainLog == cdict->cParams.chainLog);
+ assert(cctx->appliedParams.cParams.strategy == cdict_cParams->strategy);
+ assert(cctx->appliedParams.cParams.hashLog == cdict_cParams->hashLog);
+ assert(cctx->appliedParams.cParams.chainLog == cdict_cParams->chainLog);
}
/* copy tables */
- { size_t const chainSize = (cdict->cParams.strategy == ZSTD_fast) ? 0 : ((size_t)1 << cdict->cParams.chainLog);
- size_t const hSize = (size_t)1 << cdict->cParams.hashLog;
+ { size_t const chainSize = (cdict_cParams->strategy == ZSTD_fast) ? 0 : ((size_t)1 << cdict_cParams->chainLog);
+ size_t const hSize = (size_t)1 << cdict_cParams->hashLog;
size_t const tableSpace = (chainSize + hSize) * sizeof(U32);
assert((U32*)cctx->blockState.matchState.chainTable == (U32*)cctx->blockState.matchState.hashTable + hSize); /* chainTable must follow hashTable */
assert((U32*)cctx->blockState.matchState.hashTable3 == (U32*)cctx->blockState.matchState.chainTable + chainSize);
@@ -1127,6 +1706,7 @@
assert((U32*)cdict->matchState.hashTable3 == (U32*)cdict->matchState.chainTable + chainSize);
memcpy(cctx->blockState.matchState.hashTable, cdict->matchState.hashTable, tableSpace); /* presumes all tables follow each other */
}
+
/* Zero the hashTable3, since the cdict never fills it */
{ size_t const h3Size = (size_t)1 << cctx->blockState.matchState.hashLog3;
assert(cdict->matchState.hashLog3 == 0);
@@ -1134,14 +1714,13 @@
}
/* copy dictionary offsets */
- {
- ZSTD_matchState_t const* srcMatchState = &cdict->matchState;
+ { ZSTD_matchState_t const* srcMatchState = &cdict->matchState;
ZSTD_matchState_t* dstMatchState = &cctx->blockState.matchState;
dstMatchState->window = srcMatchState->window;
dstMatchState->nextToUpdate = srcMatchState->nextToUpdate;
- dstMatchState->nextToUpdate3= srcMatchState->nextToUpdate3;
dstMatchState->loadedDictEnd= srcMatchState->loadedDictEnd;
}
+
cctx->dictID = cdict->dictID;
/* copy block state */
@@ -1150,6 +1729,28 @@
return 0;
}
+/* We have a choice between copying the dictionary context into the working
+ * context, or referencing the dictionary context from the working context
+ * in-place. We decide here which strategy to use. */
+static size_t ZSTD_resetCCtx_usingCDict(ZSTD_CCtx* cctx,
+ const ZSTD_CDict* cdict,
+ ZSTD_CCtx_params params,
+ U64 pledgedSrcSize,
+ ZSTD_buffered_policy_e zbuff)
+{
+
+ DEBUGLOG(4, "ZSTD_resetCCtx_usingCDict (pledgedSrcSize=%u)",
+ (unsigned)pledgedSrcSize);
+
+ if (ZSTD_shouldAttachDict(cdict, params, pledgedSrcSize)) {
+ return ZSTD_resetCCtx_byAttachingCDict(
+ cctx, cdict, params, pledgedSrcSize, zbuff);
+ } else {
+ return ZSTD_resetCCtx_byCopyingCDict(
+ cctx, cdict, params, pledgedSrcSize, zbuff);
+ }
+}
+
/*! ZSTD_copyCCtx_internal() :
* Duplicate an existing context `srcCCtx` into another one `dstCCtx`.
* Only works during stage ZSTDcs_init (i.e. after creation, but before first call to ZSTD_compressContinue()).
@@ -1164,7 +1765,7 @@
ZSTD_buffered_policy_e zbuff)
{
DEBUGLOG(5, "ZSTD_copyCCtx_internal");
- if (srcCCtx->stage!=ZSTDcs_init) return ERROR(stage_wrong);
+ RETURN_ERROR_IF(srcCCtx->stage!=ZSTDcs_init, stage_wrong);
memcpy(&dstCCtx->customMem, &srcCCtx->customMem, sizeof(ZSTD_customMem));
{ ZSTD_CCtx_params params = dstCCtx->requestedParams;
@@ -1192,11 +1793,10 @@
/* copy dictionary offsets */
{
- ZSTD_matchState_t const* srcMatchState = &srcCCtx->blockState.matchState;
+ const ZSTD_matchState_t* srcMatchState = &srcCCtx->blockState.matchState;
ZSTD_matchState_t* dstMatchState = &dstCCtx->blockState.matchState;
dstMatchState->window = srcMatchState->window;
dstMatchState->nextToUpdate = srcMatchState->nextToUpdate;
- dstMatchState->nextToUpdate3= srcMatchState->nextToUpdate3;
dstMatchState->loadedDictEnd= srcMatchState->loadedDictEnd;
}
dstCCtx->dictID = srcCCtx->dictID;
@@ -1266,16 +1866,15 @@
/*! ZSTD_reduceIndex() :
* rescale all indexes to avoid future overflow (indexes are U32) */
-static void ZSTD_reduceIndex (ZSTD_CCtx* zc, const U32 reducerValue)
+static void ZSTD_reduceIndex (ZSTD_matchState_t* ms, ZSTD_CCtx_params const* params, const U32 reducerValue)
{
- ZSTD_matchState_t* const ms = &zc->blockState.matchState;
- { U32 const hSize = (U32)1 << zc->appliedParams.cParams.hashLog;
+ { U32 const hSize = (U32)1 << params->cParams.hashLog;
ZSTD_reduceTable(ms->hashTable, hSize, reducerValue);
}
- if (zc->appliedParams.cParams.strategy != ZSTD_fast) {
- U32 const chainSize = (U32)1 << zc->appliedParams.cParams.chainLog;
- if (zc->appliedParams.cParams.strategy == ZSTD_btlazy2)
+ if (params->cParams.strategy != ZSTD_fast) {
+ U32 const chainSize = (U32)1 << params->cParams.chainLog;
+ if (params->cParams.strategy == ZSTD_btlazy2)
ZSTD_reduceTable_btlazy2(ms->chainTable, chainSize, reducerValue);
else
ZSTD_reduceTable(ms->chainTable, chainSize, reducerValue);
@@ -1294,21 +1893,22 @@
/* See doc/zstd_compression_format.md for detailed format description */
-size_t ZSTD_noCompressBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize)
+static size_t ZSTD_noCompressBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize, U32 lastBlock)
{
- if (srcSize + ZSTD_blockHeaderSize > dstCapacity) return ERROR(dstSize_tooSmall);
+ U32 const cBlockHeader24 = lastBlock + (((U32)bt_raw)<<1) + (U32)(srcSize << 3);
+ RETURN_ERROR_IF(srcSize + ZSTD_blockHeaderSize > dstCapacity,
+ dstSize_tooSmall);
+ MEM_writeLE24(dst, cBlockHeader24);
memcpy((BYTE*)dst + ZSTD_blockHeaderSize, src, srcSize);
- MEM_writeLE24(dst, (U32)(srcSize << 2) + (U32)bt_raw);
- return ZSTD_blockHeaderSize+srcSize;
+ return ZSTD_blockHeaderSize + srcSize;
}
-
static size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
BYTE* const ostart = (BYTE* const)dst;
U32 const flSize = 1 + (srcSize>31) + (srcSize>4095);
- if (srcSize + flSize > dstCapacity) return ERROR(dstSize_tooSmall);
+ RETURN_ERROR_IF(srcSize + flSize > dstCapacity, dstSize_tooSmall);
switch(flSize)
{
@@ -1356,16 +1956,27 @@
}
-static size_t ZSTD_minGain(size_t srcSize) { return (srcSize >> 6) + 2; }
+/* ZSTD_minGain() :
+ * minimum compression required
+ * to generate a compress block or a compressed literals section.
+ * note : use same formula for both situations */
+static size_t ZSTD_minGain(size_t srcSize, ZSTD_strategy strat)
+{
+ U32 const minlog = (strat>=ZSTD_btultra) ? (U32)(strat) - 1 : 6;
+ ZSTD_STATIC_ASSERT(ZSTD_btultra == 8);
+ assert(ZSTD_cParam_withinBounds(ZSTD_c_strategy, strat));
+ return (srcSize >> minlog) + 2;
+}
-static size_t ZSTD_compressLiterals (ZSTD_entropyCTables_t const* prevEntropy,
- ZSTD_entropyCTables_t* nextEntropy,
+static size_t ZSTD_compressLiterals (ZSTD_hufCTables_t const* prevHuf,
+ ZSTD_hufCTables_t* nextHuf,
ZSTD_strategy strategy, int disableLiteralCompression,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
- U32* workspace, const int bmi2)
+ void* workspace, size_t wkspSize,
+ const int bmi2)
{
- size_t const minGain = ZSTD_minGain(srcSize);
+ size_t const minGain = ZSTD_minGain(srcSize, strategy);
size_t const lhSize = 3 + (srcSize >= 1 KB) + (srcSize >= 16 KB);
BYTE* const ostart = (BYTE*)dst;
U32 singleStream = srcSize < 256;
@@ -1376,27 +1987,25 @@
disableLiteralCompression);
/* Prepare nextEntropy assuming reusing the existing table */
- nextEntropy->hufCTable_repeatMode = prevEntropy->hufCTable_repeatMode;
- memcpy(nextEntropy->hufCTable, prevEntropy->hufCTable,
- sizeof(prevEntropy->hufCTable));
+ memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
if (disableLiteralCompression)
return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
/* small ? don't even attempt compression (speed opt) */
# define COMPRESS_LITERALS_SIZE_MIN 63
- { size_t const minLitSize = (prevEntropy->hufCTable_repeatMode == HUF_repeat_valid) ? 6 : COMPRESS_LITERALS_SIZE_MIN;
+ { size_t const minLitSize = (prevHuf->repeatMode == HUF_repeat_valid) ? 6 : COMPRESS_LITERALS_SIZE_MIN;
if (srcSize <= minLitSize) return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
}
- if (dstCapacity < lhSize+1) return ERROR(dstSize_tooSmall); /* not enough space for compression */
- { HUF_repeat repeat = prevEntropy->hufCTable_repeatMode;
+ RETURN_ERROR_IF(dstCapacity < lhSize+1, dstSize_tooSmall, "not enough space for compression");
+ { HUF_repeat repeat = prevHuf->repeatMode;
int const preferRepeat = strategy < ZSTD_lazy ? srcSize <= 1024 : 0;
if (repeat == HUF_repeat_valid && lhSize == 3) singleStream = 1;
cLitSize = singleStream ? HUF_compress1X_repeat(ostart+lhSize, dstCapacity-lhSize, src, srcSize, 255, 11,
- workspace, HUF_WORKSPACE_SIZE, (HUF_CElt*)nextEntropy->hufCTable, &repeat, preferRepeat, bmi2)
+ workspace, wkspSize, (HUF_CElt*)nextHuf->CTable, &repeat, preferRepeat, bmi2)
: HUF_compress4X_repeat(ostart+lhSize, dstCapacity-lhSize, src, srcSize, 255, 11,
- workspace, HUF_WORKSPACE_SIZE, (HUF_CElt*)nextEntropy->hufCTable, &repeat, preferRepeat, bmi2);
+ workspace, wkspSize, (HUF_CElt*)nextHuf->CTable, &repeat, preferRepeat, bmi2);
if (repeat != HUF_repeat_none) {
/* reused the existing table */
hType = set_repeat;
@@ -1404,17 +2013,17 @@
}
if ((cLitSize==0) | (cLitSize >= srcSize - minGain) | ERR_isError(cLitSize)) {
- memcpy(nextEntropy->hufCTable, prevEntropy->hufCTable, sizeof(prevEntropy->hufCTable));
+ memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
}
if (cLitSize==1) {
- memcpy(nextEntropy->hufCTable, prevEntropy->hufCTable, sizeof(prevEntropy->hufCTable));
+ memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
return ZSTD_compressRleLiteralsBlock(dst, dstCapacity, src, srcSize);
}
if (hType == set_compressed) {
/* using a newly constructed table */
- nextEntropy->hufCTable_repeatMode = HUF_repeat_check;
+ nextHuf->repeatMode = HUF_repeat_check;
}
/* Build header */
@@ -1451,6 +2060,7 @@
BYTE* const mlCodeTable = seqStorePtr->mlCode;
U32 const nbSeq = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart);
U32 u;
+ assert(nbSeq <= seqStorePtr->maxNbSeq);
for (u=0; u<nbSeq; u++) {
U32 const llv = sequences[u].litLength;
U32 const mlv = sequences[u].matchLength;
@@ -1464,72 +2074,243 @@
mlCodeTable[seqStorePtr->longLengthPos] = MaxML;
}
+
+/**
+ * -log2(x / 256) lookup table for x in [0, 256).
+ * If x == 0: Return 0
+ * Else: Return floor(-log2(x / 256) * 256)
+ */
+static unsigned const kInverseProbabilityLog256[256] = {
+ 0, 2048, 1792, 1642, 1536, 1453, 1386, 1329, 1280, 1236, 1197, 1162,
+ 1130, 1100, 1073, 1047, 1024, 1001, 980, 960, 941, 923, 906, 889,
+ 874, 859, 844, 830, 817, 804, 791, 779, 768, 756, 745, 734,
+ 724, 714, 704, 694, 685, 676, 667, 658, 650, 642, 633, 626,
+ 618, 610, 603, 595, 588, 581, 574, 567, 561, 554, 548, 542,
+ 535, 529, 523, 517, 512, 506, 500, 495, 489, 484, 478, 473,
+ 468, 463, 458, 453, 448, 443, 438, 434, 429, 424, 420, 415,
+ 411, 407, 402, 398, 394, 390, 386, 382, 377, 373, 370, 366,
+ 362, 358, 354, 350, 347, 343, 339, 336, 332, 329, 325, 322,
+ 318, 315, 311, 308, 305, 302, 298, 295, 292, 289, 286, 282,
+ 279, 276, 273, 270, 267, 264, 261, 258, 256, 253, 250, 247,
+ 244, 241, 239, 236, 233, 230, 228, 225, 222, 220, 217, 215,
+ 212, 209, 207, 204, 202, 199, 197, 194, 192, 190, 187, 185,
+ 182, 180, 178, 175, 173, 171, 168, 166, 164, 162, 159, 157,
+ 155, 153, 151, 149, 146, 144, 142, 140, 138, 136, 134, 132,
+ 130, 128, 126, 123, 121, 119, 117, 115, 114, 112, 110, 108,
+ 106, 104, 102, 100, 98, 96, 94, 93, 91, 89, 87, 85,
+ 83, 82, 80, 78, 76, 74, 73, 71, 69, 67, 66, 64,
+ 62, 61, 59, 57, 55, 54, 52, 50, 49, 47, 46, 44,
+ 42, 41, 39, 37, 36, 34, 33, 31, 30, 28, 26, 25,
+ 23, 22, 20, 19, 17, 16, 14, 13, 11, 10, 8, 7,
+ 5, 4, 2, 1,
+};
+
+
+/**
+ * Returns the cost in bits of encoding the distribution described by count
+ * using the entropy bound.
+ */
+static size_t ZSTD_entropyCost(unsigned const* count, unsigned const max, size_t const total)
+{
+ unsigned cost = 0;
+ unsigned s;
+ for (s = 0; s <= max; ++s) {
+ unsigned norm = (unsigned)((256 * count[s]) / total);
+ if (count[s] != 0 && norm == 0)
+ norm = 1;
+ assert(count[s] < total);
+ cost += count[s] * kInverseProbabilityLog256[norm];
+ }
+ return cost >> 8;
+}
+
+
+/**
+ * Returns the cost in bits of encoding the distribution in count using the
+ * table described by norm. The max symbol support by norm is assumed >= max.
+ * norm must be valid for every symbol with non-zero probability in count.
+ */
+static size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog,
+ unsigned const* count, unsigned const max)
+{
+ unsigned const shift = 8 - accuracyLog;
+ size_t cost = 0;
+ unsigned s;
+ assert(accuracyLog <= 8);
+ for (s = 0; s <= max; ++s) {
+ unsigned const normAcc = norm[s] != -1 ? norm[s] : 1;
+ unsigned const norm256 = normAcc << shift;
+ assert(norm256 > 0);
+ assert(norm256 < 256);
+ cost += count[s] * kInverseProbabilityLog256[norm256];
+ }
+ return cost >> 8;
+}
+
+
+static unsigned ZSTD_getFSEMaxSymbolValue(FSE_CTable const* ctable) {
+ void const* ptr = ctable;
+ U16 const* u16ptr = (U16 const*)ptr;
+ U32 const maxSymbolValue = MEM_read16(u16ptr + 1);
+ return maxSymbolValue;
+}
+
+
+/**
+ * Returns the cost in bits of encoding the distribution in count using ctable.
+ * Returns an error if ctable cannot represent all the symbols in count.
+ */
+static size_t ZSTD_fseBitCost(
+ FSE_CTable const* ctable,
+ unsigned const* count,
+ unsigned const max)
+{
+ unsigned const kAccuracyLog = 8;
+ size_t cost = 0;
+ unsigned s;
+ FSE_CState_t cstate;
+ FSE_initCState(&cstate, ctable);
+ RETURN_ERROR_IF(ZSTD_getFSEMaxSymbolValue(ctable) < max, GENERIC,
+ "Repeat FSE_CTable has maxSymbolValue %u < %u",
+ ZSTD_getFSEMaxSymbolValue(ctable), max);
+ for (s = 0; s <= max; ++s) {
+ unsigned const tableLog = cstate.stateLog;
+ unsigned const badCost = (tableLog + 1) << kAccuracyLog;
+ unsigned const bitCost = FSE_bitCost(cstate.symbolTT, tableLog, s, kAccuracyLog);
+ if (count[s] == 0)
+ continue;
+ RETURN_ERROR_IF(bitCost >= badCost, GENERIC,
+ "Repeat FSE_CTable has Prob[%u] == 0", s);
+ cost += count[s] * bitCost;
+ }
+ return cost >> kAccuracyLog;
+}
+
+/**
+ * Returns the cost in bytes of encoding the normalized count header.
+ * Returns an error if any of the helper functions return an error.
+ */
+static size_t ZSTD_NCountCost(unsigned const* count, unsigned const max,
+ size_t const nbSeq, unsigned const FSELog)
+{
+ BYTE wksp[FSE_NCOUNTBOUND];
+ S16 norm[MaxSeq + 1];
+ const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max);
+ FORWARD_IF_ERROR(FSE_normalizeCount(norm, tableLog, count, nbSeq, max));
+ return FSE_writeNCount(wksp, sizeof(wksp), norm, max, tableLog);
+}
+
+
typedef enum {
ZSTD_defaultDisallowed = 0,
ZSTD_defaultAllowed = 1
} ZSTD_defaultPolicy_e;
-MEM_STATIC
-symbolEncodingType_e ZSTD_selectEncodingType(
- FSE_repeat* repeatMode, size_t const mostFrequent, size_t nbSeq,
- U32 defaultNormLog, ZSTD_defaultPolicy_e const isDefaultAllowed)
+MEM_STATIC symbolEncodingType_e
+ZSTD_selectEncodingType(
+ FSE_repeat* repeatMode, unsigned const* count, unsigned const max,
+ size_t const mostFrequent, size_t nbSeq, unsigned const FSELog,
+ FSE_CTable const* prevCTable,
+ short const* defaultNorm, U32 defaultNormLog,
+ ZSTD_defaultPolicy_e const isDefaultAllowed,
+ ZSTD_strategy const strategy)
{
-#define MIN_SEQ_FOR_DYNAMIC_FSE 64
-#define MAX_SEQ_FOR_STATIC_FSE 1000
ZSTD_STATIC_ASSERT(ZSTD_defaultDisallowed == 0 && ZSTD_defaultAllowed != 0);
- if ((mostFrequent == nbSeq) && (!isDefaultAllowed || nbSeq > 2)) {
+ if (mostFrequent == nbSeq) {
+ *repeatMode = FSE_repeat_none;
+ if (isDefaultAllowed && nbSeq <= 2) {
+ /* Prefer set_basic over set_rle when there are 2 or less symbols,
+ * since RLE uses 1 byte, but set_basic uses 5-6 bits per symbol.
+ * If basic encoding isn't possible, always choose RLE.
+ */
+ DEBUGLOG(5, "Selected set_basic");
+ return set_basic;
+ }
DEBUGLOG(5, "Selected set_rle");
- /* Prefer set_basic over set_rle when there are 2 or less symbols,
- * since RLE uses 1 byte, but set_basic uses 5-6 bits per symbol.
- * If basic encoding isn't possible, always choose RLE.
- */
- *repeatMode = FSE_repeat_check;
return set_rle;
}
- if ( isDefaultAllowed
- && (*repeatMode == FSE_repeat_valid) && (nbSeq < MAX_SEQ_FOR_STATIC_FSE)) {
- DEBUGLOG(5, "Selected set_repeat");
- return set_repeat;
- }
- if ( isDefaultAllowed
- && ((nbSeq < MIN_SEQ_FOR_DYNAMIC_FSE) || (mostFrequent < (nbSeq >> (defaultNormLog-1)))) ) {
- DEBUGLOG(5, "Selected set_basic");
- /* The format allows default tables to be repeated, but it isn't useful.
- * When using simple heuristics to select encoding type, we don't want
- * to confuse these tables with dictionaries. When running more careful
- * analysis, we don't need to waste time checking both repeating tables
- * and default tables.
- */
- *repeatMode = FSE_repeat_none;
- return set_basic;
+ if (strategy < ZSTD_lazy) {
+ if (isDefaultAllowed) {
+ size_t const staticFse_nbSeq_max = 1000;
+ size_t const mult = 10 - strategy;
+ size_t const baseLog = 3;
+ size_t const dynamicFse_nbSeq_min = (((size_t)1 << defaultNormLog) * mult) >> baseLog; /* 28-36 for offset, 56-72 for lengths */
+ assert(defaultNormLog >= 5 && defaultNormLog <= 6); /* xx_DEFAULTNORMLOG */
+ assert(mult <= 9 && mult >= 7);
+ if ( (*repeatMode == FSE_repeat_valid)
+ && (nbSeq < staticFse_nbSeq_max) ) {
+ DEBUGLOG(5, "Selected set_repeat");
+ return set_repeat;
+ }
+ if ( (nbSeq < dynamicFse_nbSeq_min)
+ || (mostFrequent < (nbSeq >> (defaultNormLog-1))) ) {
+ DEBUGLOG(5, "Selected set_basic");
+ /* The format allows default tables to be repeated, but it isn't useful.
+ * When using simple heuristics to select encoding type, we don't want
+ * to confuse these tables with dictionaries. When running more careful
+ * analysis, we don't need to waste time checking both repeating tables
+ * and default tables.
+ */
+ *repeatMode = FSE_repeat_none;
+ return set_basic;
+ }
+ }
+ } else {
+ size_t const basicCost = isDefaultAllowed ? ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, count, max) : ERROR(GENERIC);
+ size_t const repeatCost = *repeatMode != FSE_repeat_none ? ZSTD_fseBitCost(prevCTable, count, max) : ERROR(GENERIC);
+ size_t const NCountCost = ZSTD_NCountCost(count, max, nbSeq, FSELog);
+ size_t const compressedCost = (NCountCost << 3) + ZSTD_entropyCost(count, max, nbSeq);
+
+ if (isDefaultAllowed) {
+ assert(!ZSTD_isError(basicCost));
+ assert(!(*repeatMode == FSE_repeat_valid && ZSTD_isError(repeatCost)));
+ }
+ assert(!ZSTD_isError(NCountCost));
+ assert(compressedCost < ERROR(maxCode));
+ DEBUGLOG(5, "Estimated bit costs: basic=%u\trepeat=%u\tcompressed=%u",
+ (unsigned)basicCost, (unsigned)repeatCost, (unsigned)compressedCost);
+ if (basicCost <= repeatCost && basicCost <= compressedCost) {
+ DEBUGLOG(5, "Selected set_basic");
+ assert(isDefaultAllowed);
+ *repeatMode = FSE_repeat_none;
+ return set_basic;
+ }
+ if (repeatCost <= compressedCost) {
+ DEBUGLOG(5, "Selected set_repeat");
+ assert(!ZSTD_isError(repeatCost));
+ return set_repeat;
+ }
+ assert(compressedCost < basicCost && compressedCost < repeatCost);
}
DEBUGLOG(5, "Selected set_compressed");
*repeatMode = FSE_repeat_check;
return set_compressed;
}
-MEM_STATIC
-size_t ZSTD_buildCTable(void* dst, size_t dstCapacity,
- FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type,
- U32* count, U32 max,
- BYTE const* codeTable, size_t nbSeq,
- S16 const* defaultNorm, U32 defaultNormLog, U32 defaultMax,
- FSE_CTable const* prevCTable, size_t prevCTableSize,
- void* workspace, size_t workspaceSize)
+MEM_STATIC size_t
+ZSTD_buildCTable(void* dst, size_t dstCapacity,
+ FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type,
+ unsigned* count, U32 max,
+ const BYTE* codeTable, size_t nbSeq,
+ const S16* defaultNorm, U32 defaultNormLog, U32 defaultMax,
+ const FSE_CTable* prevCTable, size_t prevCTableSize,
+ void* workspace, size_t workspaceSize)
{
BYTE* op = (BYTE*)dst;
- BYTE const* const oend = op + dstCapacity;
+ const BYTE* const oend = op + dstCapacity;
+ DEBUGLOG(6, "ZSTD_buildCTable (dstCapacity=%u)", (unsigned)dstCapacity);
switch (type) {
case set_rle:
+ FORWARD_IF_ERROR(FSE_buildCTable_rle(nextCTable, (BYTE)max));
+ RETURN_ERROR_IF(dstCapacity==0, dstSize_tooSmall);
*op = codeTable[0];
- CHECK_F(FSE_buildCTable_rle(nextCTable, (BYTE)max));
return 1;
case set_repeat:
memcpy(nextCTable, prevCTable, prevCTableSize);
return 0;
case set_basic:
- CHECK_F(FSE_buildCTable_wksp(nextCTable, defaultNorm, defaultMax, defaultNormLog, workspace, workspaceSize)); /* note : could be pre-calculated */
+ FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, defaultNorm, defaultMax, defaultNormLog, workspace, workspaceSize)); /* note : could be pre-calculated */
return 0;
case set_compressed: {
S16 norm[MaxSeq + 1];
@@ -1540,14 +2321,14 @@
nbSeq_1--;
}
assert(nbSeq_1 > 1);
- CHECK_F(FSE_normalizeCount(norm, tableLog, count, nbSeq_1, max));
+ FORWARD_IF_ERROR(FSE_normalizeCount(norm, tableLog, count, nbSeq_1, max));
{ size_t const NCountSize = FSE_writeNCount(op, oend - op, norm, max, tableLog); /* overflow protected */
- if (FSE_isError(NCountSize)) return NCountSize;
- CHECK_F(FSE_buildCTable_wksp(nextCTable, norm, max, tableLog, workspace, workspaceSize));
+ FORWARD_IF_ERROR(NCountSize);
+ FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, norm, max, tableLog, workspace, workspaceSize));
return NCountSize;
}
}
- default: return assert(0), ERROR(GENERIC);
+ default: assert(0); RETURN_ERROR(GENERIC);
}
}
@@ -1564,7 +2345,12 @@
FSE_CState_t stateOffsetBits;
FSE_CState_t stateLitLength;
- CHECK_E(BIT_initCStream(&blockStream, dst, dstCapacity), dstSize_tooSmall); /* not enough space remaining */
+ RETURN_ERROR_IF(
+ ERR_isError(BIT_initCStream(&blockStream, dst, dstCapacity)),
+ dstSize_tooSmall, "not enough space remaining");
+ DEBUGLOG(6, "available space for bitstream : %i (dstCapacity=%u)",
+ (int)(blockStream.endPtr - blockStream.startPtr),
+ (unsigned)dstCapacity);
/* first symbols */
FSE_initCState2(&stateMatchLength, CTable_MatchLength, mlCodeTable[nbSeq-1]);
@@ -1597,9 +2383,9 @@
U32 const ofBits = ofCode;
U32 const mlBits = ML_bits[mlCode];
DEBUGLOG(6, "encoding: litlen:%2u - matchlen:%2u - offCode:%7u",
- sequences[n].litLength,
- sequences[n].matchLength + MINMATCH,
- sequences[n].offset);
+ (unsigned)sequences[n].litLength,
+ (unsigned)sequences[n].matchLength + MINMATCH,
+ (unsigned)sequences[n].offset);
/* 32b*/ /* 64b*/
/* (7)*/ /* (7)*/
FSE_encodeSymbol(&blockStream, &stateOffsetBits, ofCode); /* 15 */ /* 15 */
@@ -1624,6 +2410,7 @@
BIT_addBits(&blockStream, sequences[n].offset, ofBits); /* 31 */
}
BIT_flushBits(&blockStream); /* (7)*/
+ DEBUGLOG(7, "remaining space : %i", (int)(blockStream.endPtr - blockStream.ptr));
} }
DEBUGLOG(6, "ZSTD_encodeSequences: flushing ML state with %u bits", stateMatchLength.stateLog);
@@ -1634,7 +2421,7 @@
FSE_flushCState(&blockStream, &stateLitLength);
{ size_t const streamSize = BIT_closeCStream(&blockStream);
- if (streamSize==0) return ERROR(dstSize_tooSmall); /* not enough space */
+ RETURN_ERROR_IF(streamSize==0, dstSize_tooSmall, "not enough space");
return streamSize;
}
}
@@ -1674,13 +2461,14 @@
#endif
-size_t ZSTD_encodeSequences(
+static size_t ZSTD_encodeSequences(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2)
{
+ DEBUGLOG(5, "ZSTD_encodeSequences: dstCapacity = %u", (unsigned)dstCapacity);
#if DYNAMIC_BMI2
if (bmi2) {
return ZSTD_encodeSequences_bmi2(dst, dstCapacity,
@@ -1698,18 +2486,38 @@
sequences, nbSeq, longOffsets);
}
-MEM_STATIC size_t ZSTD_compressSequences_internal(seqStore_t* seqStorePtr,
- ZSTD_entropyCTables_t const* prevEntropy,
- ZSTD_entropyCTables_t* nextEntropy,
- ZSTD_CCtx_params const* cctxParams,
- void* dst, size_t dstCapacity, U32* workspace,
- const int bmi2)
+static int ZSTD_disableLiteralsCompression(const ZSTD_CCtx_params* cctxParams)
+{
+ switch (cctxParams->literalCompressionMode) {
+ case ZSTD_lcm_huffman:
+ return 0;
+ case ZSTD_lcm_uncompressed:
+ return 1;
+ default:
+ assert(0 /* impossible: pre-validated */);
+ /* fall-through */
+ case ZSTD_lcm_auto:
+ return (cctxParams->cParams.strategy == ZSTD_fast) && (cctxParams->cParams.targetLength > 0);
+ }
+}
+
+/* ZSTD_compressSequences_internal():
+ * actually compresses both literals and sequences */
+MEM_STATIC size_t
+ZSTD_compressSequences_internal(seqStore_t* seqStorePtr,
+ const ZSTD_entropyCTables_t* prevEntropy,
+ ZSTD_entropyCTables_t* nextEntropy,
+ const ZSTD_CCtx_params* cctxParams,
+ void* dst, size_t dstCapacity,
+ void* workspace, size_t wkspSize,
+ const int bmi2)
{
const int longOffsets = cctxParams->cParams.windowLog > STREAM_ACCUMULATOR_MIN;
- U32 count[MaxSeq+1];
- FSE_CTable* CTable_LitLength = nextEntropy->litlengthCTable;
- FSE_CTable* CTable_OffsetBits = nextEntropy->offcodeCTable;
- FSE_CTable* CTable_MatchLength = nextEntropy->matchlengthCTable;
+ ZSTD_strategy const strategy = cctxParams->cParams.strategy;
+ unsigned count[MaxSeq+1];
+ FSE_CTable* CTable_LitLength = nextEntropy->fse.litlengthCTable;
+ FSE_CTable* CTable_OffsetBits = nextEntropy->fse.offcodeCTable;
+ FSE_CTable* CTable_MatchLength = nextEntropy->fse.matchlengthCTable;
U32 LLtype, Offtype, MLtype; /* compressed, raw or rle */
const seqDef* const sequences = seqStorePtr->sequencesStart;
const BYTE* const ofCodeTable = seqStorePtr->ofCode;
@@ -1720,87 +2528,114 @@
BYTE* op = ostart;
size_t const nbSeq = seqStorePtr->sequences - seqStorePtr->sequencesStart;
BYTE* seqHead;
+ BYTE* lastNCount = NULL;
ZSTD_STATIC_ASSERT(HUF_WORKSPACE_SIZE >= (1<<MAX(MLFSELog,LLFSELog)));
+ DEBUGLOG(5, "ZSTD_compressSequences_internal");
/* Compress literals */
{ const BYTE* const literals = seqStorePtr->litStart;
size_t const litSize = seqStorePtr->lit - literals;
size_t const cSize = ZSTD_compressLiterals(
- prevEntropy, nextEntropy,
- cctxParams->cParams.strategy, cctxParams->disableLiteralCompression,
+ &prevEntropy->huf, &nextEntropy->huf,
+ cctxParams->cParams.strategy,
+ ZSTD_disableLiteralsCompression(cctxParams),
op, dstCapacity,
literals, litSize,
- workspace, bmi2);
- if (ZSTD_isError(cSize))
- return cSize;
+ workspace, wkspSize,
+ bmi2);
+ FORWARD_IF_ERROR(cSize);
assert(cSize <= dstCapacity);
op += cSize;
}
/* Sequences Header */
- if ((oend-op) < 3 /*max nbSeq Size*/ + 1 /*seqHead*/) return ERROR(dstSize_tooSmall);
+ RETURN_ERROR_IF((oend-op) < 3 /*max nbSeq Size*/ + 1 /*seqHead*/,
+ dstSize_tooSmall);
if (nbSeq < 0x7F)
*op++ = (BYTE)nbSeq;
else if (nbSeq < LONGNBSEQ)
op[0] = (BYTE)((nbSeq>>8) + 0x80), op[1] = (BYTE)nbSeq, op+=2;
else
op[0]=0xFF, MEM_writeLE16(op+1, (U16)(nbSeq - LONGNBSEQ)), op+=3;
+ assert(op <= oend);
if (nbSeq==0) {
- memcpy(nextEntropy->litlengthCTable, prevEntropy->litlengthCTable, sizeof(prevEntropy->litlengthCTable));
- nextEntropy->litlength_repeatMode = prevEntropy->litlength_repeatMode;
- memcpy(nextEntropy->offcodeCTable, prevEntropy->offcodeCTable, sizeof(prevEntropy->offcodeCTable));
- nextEntropy->offcode_repeatMode = prevEntropy->offcode_repeatMode;
- memcpy(nextEntropy->matchlengthCTable, prevEntropy->matchlengthCTable, sizeof(prevEntropy->matchlengthCTable));
- nextEntropy->matchlength_repeatMode = prevEntropy->matchlength_repeatMode;
- return op - ostart;
+ /* Copy the old tables over as if we repeated them */
+ memcpy(&nextEntropy->fse, &prevEntropy->fse, sizeof(prevEntropy->fse));
+ return op - ostart;
}
/* seqHead : flags for FSE encoding type */
seqHead = op++;
+ assert(op <= oend);
/* convert length/distances into codes */
ZSTD_seqToCodes(seqStorePtr);
/* build CTable for Literal Lengths */
- { U32 max = MaxLL;
- size_t const mostFrequent = FSE_countFast_wksp(count, &max, llCodeTable, nbSeq, workspace);
+ { unsigned max = MaxLL;
+ size_t const mostFrequent = HIST_countFast_wksp(count, &max, llCodeTable, nbSeq, workspace, wkspSize); /* can't fail */
DEBUGLOG(5, "Building LL table");
- nextEntropy->litlength_repeatMode = prevEntropy->litlength_repeatMode;
- LLtype = ZSTD_selectEncodingType(&nextEntropy->litlength_repeatMode, mostFrequent, nbSeq, LL_defaultNormLog, ZSTD_defaultAllowed);
+ nextEntropy->fse.litlength_repeatMode = prevEntropy->fse.litlength_repeatMode;
+ LLtype = ZSTD_selectEncodingType(&nextEntropy->fse.litlength_repeatMode,
+ count, max, mostFrequent, nbSeq,
+ LLFSELog, prevEntropy->fse.litlengthCTable,
+ LL_defaultNorm, LL_defaultNormLog,
+ ZSTD_defaultAllowed, strategy);
+ assert(set_basic < set_compressed && set_rle < set_compressed);
+ assert(!(LLtype < set_compressed && nextEntropy->fse.litlength_repeatMode != FSE_repeat_none)); /* We don't copy tables */
{ size_t const countSize = ZSTD_buildCTable(op, oend - op, CTable_LitLength, LLFSELog, (symbolEncodingType_e)LLtype,
- count, max, llCodeTable, nbSeq, LL_defaultNorm, LL_defaultNormLog, MaxLL,
- prevEntropy->litlengthCTable, sizeof(prevEntropy->litlengthCTable),
- workspace, HUF_WORKSPACE_SIZE);
- if (ZSTD_isError(countSize)) return countSize;
+ count, max, llCodeTable, nbSeq, LL_defaultNorm, LL_defaultNormLog, MaxLL,
+ prevEntropy->fse.litlengthCTable, sizeof(prevEntropy->fse.litlengthCTable),
+ workspace, wkspSize);
+ FORWARD_IF_ERROR(countSize);
+ if (LLtype == set_compressed)
+ lastNCount = op;
op += countSize;
+ assert(op <= oend);
} }
/* build CTable for Offsets */
- { U32 max = MaxOff;
- size_t const mostFrequent = FSE_countFast_wksp(count, &max, ofCodeTable, nbSeq, workspace);
+ { unsigned max = MaxOff;
+ size_t const mostFrequent = HIST_countFast_wksp(count, &max, ofCodeTable, nbSeq, workspace, wkspSize); /* can't fail */
/* We can only use the basic table if max <= DefaultMaxOff, otherwise the offsets are too large */
ZSTD_defaultPolicy_e const defaultPolicy = (max <= DefaultMaxOff) ? ZSTD_defaultAllowed : ZSTD_defaultDisallowed;
DEBUGLOG(5, "Building OF table");
- nextEntropy->offcode_repeatMode = prevEntropy->offcode_repeatMode;
- Offtype = ZSTD_selectEncodingType(&nextEntropy->offcode_repeatMode, mostFrequent, nbSeq, OF_defaultNormLog, defaultPolicy);
+ nextEntropy->fse.offcode_repeatMode = prevEntropy->fse.offcode_repeatMode;
+ Offtype = ZSTD_selectEncodingType(&nextEntropy->fse.offcode_repeatMode,
+ count, max, mostFrequent, nbSeq,
+ OffFSELog, prevEntropy->fse.offcodeCTable,
+ OF_defaultNorm, OF_defaultNormLog,
+ defaultPolicy, strategy);
+ assert(!(Offtype < set_compressed && nextEntropy->fse.offcode_repeatMode != FSE_repeat_none)); /* We don't copy tables */
{ size_t const countSize = ZSTD_buildCTable(op, oend - op, CTable_OffsetBits, OffFSELog, (symbolEncodingType_e)Offtype,
- count, max, ofCodeTable, nbSeq, OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff,
- prevEntropy->offcodeCTable, sizeof(prevEntropy->offcodeCTable),
- workspace, HUF_WORKSPACE_SIZE);
- if (ZSTD_isError(countSize)) return countSize;
+ count, max, ofCodeTable, nbSeq, OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff,
+ prevEntropy->fse.offcodeCTable, sizeof(prevEntropy->fse.offcodeCTable),
+ workspace, wkspSize);
+ FORWARD_IF_ERROR(countSize);
+ if (Offtype == set_compressed)
+ lastNCount = op;
op += countSize;
+ assert(op <= oend);
} }
/* build CTable for MatchLengths */
- { U32 max = MaxML;
- size_t const mostFrequent = FSE_countFast_wksp(count, &max, mlCodeTable, nbSeq, workspace);
- DEBUGLOG(5, "Building ML table");
- nextEntropy->matchlength_repeatMode = prevEntropy->matchlength_repeatMode;
- MLtype = ZSTD_selectEncodingType(&nextEntropy->matchlength_repeatMode, mostFrequent, nbSeq, ML_defaultNormLog, ZSTD_defaultAllowed);
+ { unsigned max = MaxML;
+ size_t const mostFrequent = HIST_countFast_wksp(count, &max, mlCodeTable, nbSeq, workspace, wkspSize); /* can't fail */
+ DEBUGLOG(5, "Building ML table (remaining space : %i)", (int)(oend-op));
+ nextEntropy->fse.matchlength_repeatMode = prevEntropy->fse.matchlength_repeatMode;
+ MLtype = ZSTD_selectEncodingType(&nextEntropy->fse.matchlength_repeatMode,
+ count, max, mostFrequent, nbSeq,
+ MLFSELog, prevEntropy->fse.matchlengthCTable,
+ ML_defaultNorm, ML_defaultNormLog,
+ ZSTD_defaultAllowed, strategy);
+ assert(!(MLtype < set_compressed && nextEntropy->fse.matchlength_repeatMode != FSE_repeat_none)); /* We don't copy tables */
{ size_t const countSize = ZSTD_buildCTable(op, oend - op, CTable_MatchLength, MLFSELog, (symbolEncodingType_e)MLtype,
- count, max, mlCodeTable, nbSeq, ML_defaultNorm, ML_defaultNormLog, MaxML,
- prevEntropy->matchlengthCTable, sizeof(prevEntropy->matchlengthCTable),
- workspace, HUF_WORKSPACE_SIZE);
- if (ZSTD_isError(countSize)) return countSize;
+ count, max, mlCodeTable, nbSeq, ML_defaultNorm, ML_defaultNormLog, MaxML,
+ prevEntropy->fse.matchlengthCTable, sizeof(prevEntropy->fse.matchlengthCTable),
+ workspace, wkspSize);
+ FORWARD_IF_ERROR(countSize);
+ if (MLtype == set_compressed)
+ lastNCount = op;
op += countSize;
+ assert(op <= oend);
} }
*seqHead = (BYTE)((LLtype<<6) + (Offtype<<4) + (MLtype<<2));
@@ -1812,65 +2647,104 @@
CTable_LitLength, llCodeTable,
sequences, nbSeq,
longOffsets, bmi2);
- if (ZSTD_isError(bitstreamSize)) return bitstreamSize;
+ FORWARD_IF_ERROR(bitstreamSize);
op += bitstreamSize;
+ assert(op <= oend);
+ /* zstd versions <= 1.3.4 mistakenly report corruption when
+ * FSE_readNCount() receives a buffer < 4 bytes.
+ * Fixed by https://github.com/facebook/zstd/pull/1146.
+ * This can happen when the last set_compressed table present is 2
+ * bytes and the bitstream is only one byte.
+ * In this exceedingly rare case, we will simply emit an uncompressed
+ * block, since it isn't worth optimizing.
+ */
+ if (lastNCount && (op - lastNCount) < 4) {
+ /* NCountSize >= 2 && bitstreamSize > 0 ==> lastCountSize == 3 */
+ assert(op - lastNCount == 3);
+ DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.3.4 by "
+ "emitting an uncompressed block.");
+ return 0;
+ }
}
+ DEBUGLOG(5, "compressed block size : %u", (unsigned)(op - ostart));
return op - ostart;
}
-MEM_STATIC size_t ZSTD_compressSequences(seqStore_t* seqStorePtr,
- ZSTD_entropyCTables_t const* prevEntropy,
- ZSTD_entropyCTables_t* nextEntropy,
- ZSTD_CCtx_params const* cctxParams,
- void* dst, size_t dstCapacity,
- size_t srcSize, U32* workspace, int bmi2)
+MEM_STATIC size_t
+ZSTD_compressSequences(seqStore_t* seqStorePtr,
+ const ZSTD_entropyCTables_t* prevEntropy,
+ ZSTD_entropyCTables_t* nextEntropy,
+ const ZSTD_CCtx_params* cctxParams,
+ void* dst, size_t dstCapacity,
+ size_t srcSize,
+ void* workspace, size_t wkspSize,
+ int bmi2)
{
size_t const cSize = ZSTD_compressSequences_internal(
- seqStorePtr, prevEntropy, nextEntropy, cctxParams, dst, dstCapacity,
- workspace, bmi2);
+ seqStorePtr, prevEntropy, nextEntropy, cctxParams,
+ dst, dstCapacity,
+ workspace, wkspSize, bmi2);
+ if (cSize == 0) return 0;
/* When srcSize <= dstCapacity, there is enough space to write a raw uncompressed block.
* Since we ran out of space, block must be not compressible, so fall back to raw uncompressed block.
*/
if ((cSize == ERROR(dstSize_tooSmall)) & (srcSize <= dstCapacity))
return 0; /* block not compressed */
- if (ZSTD_isError(cSize)) return cSize;
+ FORWARD_IF_ERROR(cSize);
/* Check compressibility */
- { size_t const maxCSize = srcSize - ZSTD_minGain(srcSize); /* note : fixed formula, maybe should depend on compression level, or strategy */
+ { size_t const maxCSize = srcSize - ZSTD_minGain(srcSize, cctxParams->cParams.strategy);
if (cSize >= maxCSize) return 0; /* block not compressed */
}
- /* We check that dictionaries have offset codes available for the first
- * block. After the first block, the offcode table might not have large
- * enough codes to represent the offsets in the data.
- */
- if (nextEntropy->offcode_repeatMode == FSE_repeat_valid)
- nextEntropy->offcode_repeatMode = FSE_repeat_check;
-
return cSize;
}
/* ZSTD_selectBlockCompressor() :
* Not static, but internal use only (used by long distance matcher)
* assumption : strat is a valid strategy */
-ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, int extDict)
+ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, ZSTD_dictMode_e dictMode)
{
- static const ZSTD_blockCompressor blockCompressor[2][(unsigned)ZSTD_btultra+1] = {
+ static const ZSTD_blockCompressor blockCompressor[3][ZSTD_STRATEGY_MAX+1] = {
{ ZSTD_compressBlock_fast /* default for 0 */,
- ZSTD_compressBlock_fast, ZSTD_compressBlock_doubleFast, ZSTD_compressBlock_greedy,
- ZSTD_compressBlock_lazy, ZSTD_compressBlock_lazy2, ZSTD_compressBlock_btlazy2,
- ZSTD_compressBlock_btopt, ZSTD_compressBlock_btultra },
+ ZSTD_compressBlock_fast,
+ ZSTD_compressBlock_doubleFast,
+ ZSTD_compressBlock_greedy,
+ ZSTD_compressBlock_lazy,
+ ZSTD_compressBlock_lazy2,
+ ZSTD_compressBlock_btlazy2,
+ ZSTD_compressBlock_btopt,
+ ZSTD_compressBlock_btultra,
+ ZSTD_compressBlock_btultra2 },
{ ZSTD_compressBlock_fast_extDict /* default for 0 */,
- ZSTD_compressBlock_fast_extDict, ZSTD_compressBlock_doubleFast_extDict, ZSTD_compressBlock_greedy_extDict,
- ZSTD_compressBlock_lazy_extDict,ZSTD_compressBlock_lazy2_extDict, ZSTD_compressBlock_btlazy2_extDict,
- ZSTD_compressBlock_btopt_extDict, ZSTD_compressBlock_btultra_extDict }
+ ZSTD_compressBlock_fast_extDict,
+ ZSTD_compressBlock_doubleFast_extDict,
+ ZSTD_compressBlock_greedy_extDict,
+ ZSTD_compressBlock_lazy_extDict,
+ ZSTD_compressBlock_lazy2_extDict,
+ ZSTD_compressBlock_btlazy2_extDict,
+ ZSTD_compressBlock_btopt_extDict,
+ ZSTD_compressBlock_btultra_extDict,
+ ZSTD_compressBlock_btultra_extDict },
+ { ZSTD_compressBlock_fast_dictMatchState /* default for 0 */,
+ ZSTD_compressBlock_fast_dictMatchState,
+ ZSTD_compressBlock_doubleFast_dictMatchState,
+ ZSTD_compressBlock_greedy_dictMatchState,
+ ZSTD_compressBlock_lazy_dictMatchState,
+ ZSTD_compressBlock_lazy2_dictMatchState,
+ ZSTD_compressBlock_btlazy2_dictMatchState,
+ ZSTD_compressBlock_btopt_dictMatchState,
+ ZSTD_compressBlock_btultra_dictMatchState,
+ ZSTD_compressBlock_btultra_dictMatchState }
};
+ ZSTD_blockCompressor selectedCompressor;
ZSTD_STATIC_ASSERT((unsigned)ZSTD_fast == 1);
- assert((U32)strat >= (U32)ZSTD_fast);
- assert((U32)strat <= (U32)ZSTD_btultra);
- return blockCompressor[extDict!=0][(U32)strat];
+ assert(ZSTD_cParam_withinBounds(ZSTD_c_strategy, strat));
+ selectedCompressor = blockCompressor[(int)dictMode][(int)strat];
+ assert(selectedCompressor != NULL);
+ return selectedCompressor;
}
static void ZSTD_storeLastLiterals(seqStore_t* seqStorePtr,
@@ -1880,36 +2754,47 @@
seqStorePtr->lit += lastLLSize;
}
-static void ZSTD_resetSeqStore(seqStore_t* ssPtr)
+void ZSTD_resetSeqStore(seqStore_t* ssPtr)
{
ssPtr->lit = ssPtr->litStart;
ssPtr->sequences = ssPtr->sequencesStart;
ssPtr->longLengthID = 0;
}
-static size_t ZSTD_compressBlock_internal(ZSTD_CCtx* zc,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize)
+typedef enum { ZSTDbss_compress, ZSTDbss_noCompress } ZSTD_buildSeqStore_e;
+
+static size_t ZSTD_buildSeqStore(ZSTD_CCtx* zc, const void* src, size_t srcSize)
{
ZSTD_matchState_t* const ms = &zc->blockState.matchState;
- DEBUGLOG(5, "ZSTD_compressBlock_internal (dstCapacity=%u, dictLimit=%u, nextToUpdate=%u)",
- (U32)dstCapacity, ms->window.dictLimit, ms->nextToUpdate);
+ DEBUGLOG(5, "ZSTD_buildSeqStore (srcSize=%zu)", srcSize);
+ assert(srcSize <= ZSTD_BLOCKSIZE_MAX);
+ /* Assert that we have correctly flushed the ctx params into the ms's copy */
+ ZSTD_assertEqualCParams(zc->appliedParams.cParams, ms->cParams);
if (srcSize < MIN_CBLOCK_SIZE+ZSTD_blockHeaderSize+1) {
- ZSTD_ldm_skipSequences(&zc->externSeqStore, srcSize, zc->appliedParams.cParams.searchLength);
- return 0; /* don't even attempt compression below a certain srcSize */
+ ZSTD_ldm_skipSequences(&zc->externSeqStore, srcSize, zc->appliedParams.cParams.minMatch);
+ return ZSTDbss_noCompress; /* don't even attempt compression below a certain srcSize */
}
ZSTD_resetSeqStore(&(zc->seqStore));
+ /* required for optimal parser to read stats from dictionary */
+ ms->opt.symbolCosts = &zc->blockState.prevCBlock->entropy;
+ /* tell the optimal parser how we expect to compress literals */
+ ms->opt.literalCompressionMode = zc->appliedParams.literalCompressionMode;
+ /* a gap between an attached dict and the current window is not safe,
+ * they must remain adjacent,
+ * and when that stops being the case, the dict must be unset */
+ assert(ms->dictMatchState == NULL || ms->loadedDictEnd == ms->window.dictLimit);
/* limited update after a very long match */
{ const BYTE* const base = ms->window.base;
const BYTE* const istart = (const BYTE*)src;
const U32 current = (U32)(istart-base);
+ if (sizeof(ptrdiff_t)==8) assert(istart - base < (ptrdiff_t)(U32)(-1)); /* ensure no overflow */
if (current > ms->nextToUpdate + 384)
ms->nextToUpdate = current - MIN(192, (U32)(current - ms->nextToUpdate - 384));
}
/* select and store sequences */
- { U32 const extDict = ZSTD_window_hasExtDict(ms->window);
+ { ZSTD_dictMode_e const dictMode = ZSTD_matchState_dictMode(ms);
size_t lastLLSize;
{ int i;
for (i = 0; i < ZSTD_REP_NUM; ++i)
@@ -1922,8 +2807,7 @@
ZSTD_ldm_blockCompress(&zc->externSeqStore,
ms, &zc->seqStore,
zc->blockState.nextCBlock->rep,
- &zc->appliedParams.cParams,
- src, srcSize, extDict);
+ src, srcSize);
assert(zc->externSeqStore.pos <= zc->externSeqStore.size);
} else if (zc->appliedParams.ldmParams.enableLdm) {
rawSeqStore_t ldmSeqStore = {NULL, 0, 0, 0};
@@ -1931,7 +2815,7 @@
ldmSeqStore.seq = zc->ldmSequences;
ldmSeqStore.capacity = zc->maxNbLdmSequences;
/* Updates ldmSeqStore.size */
- CHECK_F(ZSTD_ldm_generateSequences(&zc->ldmState, &ldmSeqStore,
+ FORWARD_IF_ERROR(ZSTD_ldm_generateSequences(&zc->ldmState, &ldmSeqStore,
&zc->appliedParams.ldmParams,
src, srcSize));
/* Updates ldmSeqStore.pos */
@@ -1939,30 +2823,73 @@
ZSTD_ldm_blockCompress(&ldmSeqStore,
ms, &zc->seqStore,
zc->blockState.nextCBlock->rep,
- &zc->appliedParams.cParams,
- src, srcSize, extDict);
+ src, srcSize);
assert(ldmSeqStore.pos == ldmSeqStore.size);
} else { /* not long range mode */
- ZSTD_blockCompressor const blockCompressor = ZSTD_selectBlockCompressor(zc->appliedParams.cParams.strategy, extDict);
- lastLLSize = blockCompressor(ms, &zc->seqStore, zc->blockState.nextCBlock->rep, &zc->appliedParams.cParams, src, srcSize);
+ ZSTD_blockCompressor const blockCompressor = ZSTD_selectBlockCompressor(zc->appliedParams.cParams.strategy, dictMode);
+ lastLLSize = blockCompressor(ms, &zc->seqStore, zc->blockState.nextCBlock->rep, src, srcSize);
}
{ const BYTE* const lastLiterals = (const BYTE*)src + srcSize - lastLLSize;
ZSTD_storeLastLiterals(&zc->seqStore, lastLiterals, lastLLSize);
} }
+ return ZSTDbss_compress;
+}
+
+static size_t ZSTD_compressBlock_internal(ZSTD_CCtx* zc,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize)
+{
+ size_t cSize;
+ DEBUGLOG(5, "ZSTD_compressBlock_internal (dstCapacity=%u, dictLimit=%u, nextToUpdate=%u)",
+ (unsigned)dstCapacity, (unsigned)zc->blockState.matchState.window.dictLimit, (unsigned)zc->blockState.matchState.nextToUpdate);
+
+ { const size_t bss = ZSTD_buildSeqStore(zc, src, srcSize);
+ FORWARD_IF_ERROR(bss);
+ if (bss == ZSTDbss_noCompress) { cSize = 0; goto out; }
+ }
/* encode sequences and literals */
- { size_t const cSize = ZSTD_compressSequences(&zc->seqStore,
- &zc->blockState.prevCBlock->entropy, &zc->blockState.nextCBlock->entropy,
- &zc->appliedParams,
- dst, dstCapacity,
- srcSize, zc->entropyWorkspace, zc->bmi2);
- if (ZSTD_isError(cSize) || cSize == 0) return cSize;
- /* confirm repcodes and entropy tables */
- { ZSTD_compressedBlockState_t* const tmp = zc->blockState.prevCBlock;
- zc->blockState.prevCBlock = zc->blockState.nextCBlock;
- zc->blockState.nextCBlock = tmp;
- }
- return cSize;
+ cSize = ZSTD_compressSequences(&zc->seqStore,
+ &zc->blockState.prevCBlock->entropy, &zc->blockState.nextCBlock->entropy,
+ &zc->appliedParams,
+ dst, dstCapacity,
+ srcSize,
+ zc->entropyWorkspace, HUF_WORKSPACE_SIZE /* statically allocated in resetCCtx */,
+ zc->bmi2);
+
+out:
+ if (!ZSTD_isError(cSize) && cSize != 0) {
+ /* confirm repcodes and entropy tables when emitting a compressed block */
+ ZSTD_compressedBlockState_t* const tmp = zc->blockState.prevCBlock;
+ zc->blockState.prevCBlock = zc->blockState.nextCBlock;
+ zc->blockState.nextCBlock = tmp;
+ }
+ /* We check that dictionaries have offset codes available for the first
+ * block. After the first block, the offcode table might not have large
+ * enough codes to represent the offsets in the data.
+ */
+ if (zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid)
+ zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check;
+
+ return cSize;
+}
+
+
+static void ZSTD_overflowCorrectIfNeeded(ZSTD_matchState_t* ms, ZSTD_CCtx_params const* params, void const* ip, void const* iend)
+{
+ if (ZSTD_window_needOverflowCorrection(ms->window, iend)) {
+ U32 const maxDist = (U32)1 << params->cParams.windowLog;
+ U32 const cycleLog = ZSTD_cycleLog(params->cParams.chainLog, params->cParams.strategy);
+ U32 const correction = ZSTD_window_correctOverflow(&ms->window, cycleLog, maxDist, ip);
+ ZSTD_STATIC_ASSERT(ZSTD_CHAINLOG_MAX <= 30);
+ ZSTD_STATIC_ASSERT(ZSTD_WINDOWLOG_MAX_32 <= 30);
+ ZSTD_STATIC_ASSERT(ZSTD_WINDOWLOG_MAX <= 31);
+ ZSTD_reduceIndex(ms, params, correction);
+ if (ms->nextToUpdate < correction) ms->nextToUpdate = 0;
+ else ms->nextToUpdate -= correction;
+ /* invalidate dictionaries on overflow correction */
+ ms->loadedDictEnd = 0;
+ ms->dictMatchState = NULL;
}
}
@@ -1985,9 +2912,9 @@
BYTE* const ostart = (BYTE*)dst;
BYTE* op = ostart;
U32 const maxDist = (U32)1 << cctx->appliedParams.cParams.windowLog;
- assert(cctx->appliedParams.cParams.windowLog <= 31);
+ assert(cctx->appliedParams.cParams.windowLog <= ZSTD_WINDOWLOG_MAX);
- DEBUGLOG(5, "ZSTD_compress_frameChunk (blockSize=%u)", (U32)blockSize);
+ DEBUGLOG(5, "ZSTD_compress_frameChunk (blockSize=%u)", (unsigned)blockSize);
if (cctx->appliedParams.fParams.checksumFlag && srcSize)
XXH64_update(&cctx->xxhState, src, srcSize);
@@ -1995,36 +2922,25 @@
ZSTD_matchState_t* const ms = &cctx->blockState.matchState;
U32 const lastBlock = lastFrameChunk & (blockSize >= remaining);
- if (dstCapacity < ZSTD_blockHeaderSize + MIN_CBLOCK_SIZE)
- return ERROR(dstSize_tooSmall); /* not enough space to store compressed block */
+ RETURN_ERROR_IF(dstCapacity < ZSTD_blockHeaderSize + MIN_CBLOCK_SIZE,
+ dstSize_tooSmall,
+ "not enough space to store compressed block");
if (remaining < blockSize) blockSize = remaining;
- if (ZSTD_window_needOverflowCorrection(ms->window, ip + blockSize)) {
- U32 const cycleLog = ZSTD_cycleLog(cctx->appliedParams.cParams.chainLog, cctx->appliedParams.cParams.strategy);
- U32 const correction = ZSTD_window_correctOverflow(&ms->window, cycleLog, maxDist, ip);
- ZSTD_STATIC_ASSERT(ZSTD_CHAINLOG_MAX <= 30);
- ZSTD_STATIC_ASSERT(ZSTD_WINDOWLOG_MAX_32 <= 30);
- ZSTD_STATIC_ASSERT(ZSTD_WINDOWLOG_MAX <= 31);
+ ZSTD_overflowCorrectIfNeeded(ms, &cctx->appliedParams, ip, ip + blockSize);
+ ZSTD_checkDictValidity(&ms->window, ip + blockSize, maxDist, &ms->loadedDictEnd, &ms->dictMatchState);
- ZSTD_reduceIndex(cctx, correction);
- if (ms->nextToUpdate < correction) ms->nextToUpdate = 0;
- else ms->nextToUpdate -= correction;
- ms->loadedDictEnd = 0;
- }
- ZSTD_window_enforceMaxDist(&ms->window, ip + blockSize, maxDist, &ms->loadedDictEnd);
+ /* Ensure hash/chain table insertion resumes no sooner than lowlimit */
if (ms->nextToUpdate < ms->window.lowLimit) ms->nextToUpdate = ms->window.lowLimit;
{ size_t cSize = ZSTD_compressBlock_internal(cctx,
op+ZSTD_blockHeaderSize, dstCapacity-ZSTD_blockHeaderSize,
ip, blockSize);
- if (ZSTD_isError(cSize)) return cSize;
+ FORWARD_IF_ERROR(cSize);
if (cSize == 0) { /* block is not compressible */
- U32 const cBlockHeader24 = lastBlock + (((U32)bt_raw)<<1) + (U32)(blockSize << 3);
- if (blockSize + ZSTD_blockHeaderSize > dstCapacity) return ERROR(dstSize_tooSmall);
- MEM_writeLE32(op, cBlockHeader24); /* 4th byte will be overwritten */
- memcpy(op + ZSTD_blockHeaderSize, ip, blockSize);
- cSize = ZSTD_blockHeaderSize + blockSize;
+ cSize = ZSTD_noCompressBlock(op, dstCapacity, ip, blockSize, lastBlock);
+ FORWARD_IF_ERROR(cSize);
} else {
U32 const cBlockHeader24 = lastBlock + (((U32)bt_compressed)<<1) + (U32)(cSize << 3);
MEM_writeLE24(op, cBlockHeader24);
@@ -2038,11 +2954,11 @@
assert(dstCapacity >= cSize);
dstCapacity -= cSize;
DEBUGLOG(5, "ZSTD_compress_frameChunk: adding a block of size %u",
- (U32)cSize);
+ (unsigned)cSize);
} }
if (lastFrameChunk && (op>ostart)) cctx->stage = ZSTDcs_ending;
- return op-ostart;
+ return (size_t)(op-ostart);
}
@@ -2057,18 +2973,19 @@
BYTE const windowLogByte = (BYTE)((params.cParams.windowLog - ZSTD_WINDOWLOG_ABSOLUTEMIN) << 3);
U32 const fcsCode = params.fParams.contentSizeFlag ?
(pledgedSrcSize>=256) + (pledgedSrcSize>=65536+256) + (pledgedSrcSize>=0xFFFFFFFFU) : 0; /* 0-3 */
- BYTE const frameHeaderDecriptionByte = (BYTE)(dictIDSizeCode + (checksumFlag<<2) + (singleSegment<<5) + (fcsCode<<6) );
+ BYTE const frameHeaderDescriptionByte = (BYTE)(dictIDSizeCode + (checksumFlag<<2) + (singleSegment<<5) + (fcsCode<<6) );
size_t pos=0;
- if (dstCapacity < ZSTD_frameHeaderSize_max) return ERROR(dstSize_tooSmall);
+ assert(!(params.fParams.contentSizeFlag && pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN));
+ RETURN_ERROR_IF(dstCapacity < ZSTD_FRAMEHEADERSIZE_MAX, dstSize_tooSmall);
DEBUGLOG(4, "ZSTD_writeFrameHeader : dictIDFlag : %u ; dictID : %u ; dictIDSizeCode : %u",
- !params.fParams.noDictIDFlag, dictID, dictIDSizeCode);
+ !params.fParams.noDictIDFlag, (unsigned)dictID, (unsigned)dictIDSizeCode);
if (params.format == ZSTD_f_zstd1) {
MEM_writeLE32(dst, ZSTD_MAGICNUMBER);
pos = 4;
}
- op[pos++] = frameHeaderDecriptionByte;
+ op[pos++] = frameHeaderDescriptionByte;
if (!singleSegment) op[pos++] = windowLogByte;
switch(dictIDSizeCode)
{
@@ -2092,11 +3009,11 @@
/* ZSTD_writeLastEmptyBlock() :
* output an empty Block with end-of-frame mark to complete a frame
* @return : size of data written into `dst` (== ZSTD_blockHeaderSize (defined in zstd_internal.h))
- * or an error code if `dstCapcity` is too small (<ZSTD_blockHeaderSize)
+ * or an error code if `dstCapacity` is too small (<ZSTD_blockHeaderSize)
*/
size_t ZSTD_writeLastEmptyBlock(void* dst, size_t dstCapacity)
{
- if (dstCapacity < ZSTD_blockHeaderSize) return ERROR(dstSize_tooSmall);
+ RETURN_ERROR_IF(dstCapacity < ZSTD_blockHeaderSize, dstSize_tooSmall);
{ U32 const cBlockHeader24 = 1 /*lastBlock*/ + (((U32)bt_raw)<<1); /* 0 size */
MEM_writeLE24(dst, cBlockHeader24);
return ZSTD_blockHeaderSize;
@@ -2105,10 +3022,9 @@
size_t ZSTD_referenceExternalSequences(ZSTD_CCtx* cctx, rawSeq* seq, size_t nbSeq)
{
- if (cctx->stage != ZSTDcs_init)
- return ERROR(stage_wrong);
- if (cctx->appliedParams.ldmParams.enableLdm)
- return ERROR(parameter_unsupported);
+ RETURN_ERROR_IF(cctx->stage != ZSTDcs_init, stage_wrong);
+ RETURN_ERROR_IF(cctx->appliedParams.ldmParams.enableLdm,
+ parameter_unsupported);
cctx->externSeqStore.seq = seq;
cctx->externSeqStore.size = nbSeq;
cctx->externSeqStore.capacity = nbSeq;
@@ -2122,17 +3038,19 @@
const void* src, size_t srcSize,
U32 frame, U32 lastFrameChunk)
{
- ZSTD_matchState_t* ms = &cctx->blockState.matchState;
+ ZSTD_matchState_t* const ms = &cctx->blockState.matchState;
size_t fhSize = 0;
DEBUGLOG(5, "ZSTD_compressContinue_internal, stage: %u, srcSize: %u",
- cctx->stage, (U32)srcSize);
- if (cctx->stage==ZSTDcs_created) return ERROR(stage_wrong); /* missing init (ZSTD_compressBegin) */
+ cctx->stage, (unsigned)srcSize);
+ RETURN_ERROR_IF(cctx->stage==ZSTDcs_created, stage_wrong,
+ "missing init (ZSTD_compressBegin)");
if (frame && (cctx->stage==ZSTDcs_init)) {
fhSize = ZSTD_writeFrameHeader(dst, dstCapacity, cctx->appliedParams,
cctx->pledgedSrcSizePlusOne-1, cctx->dictID);
- if (ZSTD_isError(fhSize)) return fhSize;
+ FORWARD_IF_ERROR(fhSize);
+ assert(fhSize <= dstCapacity);
dstCapacity -= fhSize;
dst = (char*)dst + fhSize;
cctx->stage = ZSTDcs_ongoing;
@@ -2143,22 +3061,31 @@
if (!ZSTD_window_update(&ms->window, src, srcSize)) {
ms->nextToUpdate = ms->window.dictLimit;
}
- if (cctx->appliedParams.ldmParams.enableLdm)
+ if (cctx->appliedParams.ldmParams.enableLdm) {
ZSTD_window_update(&cctx->ldmState.window, src, srcSize);
+ }
- DEBUGLOG(5, "ZSTD_compressContinue_internal (blockSize=%u)", (U32)cctx->blockSize);
+ if (!frame) {
+ /* overflow check and correction for block mode */
+ ZSTD_overflowCorrectIfNeeded(ms, &cctx->appliedParams, src, (BYTE const*)src + srcSize);
+ }
+
+ DEBUGLOG(5, "ZSTD_compressContinue_internal (blockSize=%u)", (unsigned)cctx->blockSize);
{ size_t const cSize = frame ?
ZSTD_compress_frameChunk (cctx, dst, dstCapacity, src, srcSize, lastFrameChunk) :
ZSTD_compressBlock_internal (cctx, dst, dstCapacity, src, srcSize);
- if (ZSTD_isError(cSize)) return cSize;
+ FORWARD_IF_ERROR(cSize);
cctx->consumedSrcSize += srcSize;
cctx->producedCSize += (cSize + fhSize);
- if (cctx->appliedParams.fParams.contentSizeFlag) { /* control src size */
- if (cctx->consumedSrcSize+1 > cctx->pledgedSrcSizePlusOne) {
- DEBUGLOG(4, "error : pledgedSrcSize = %u, while realSrcSize >= %u",
- (U32)cctx->pledgedSrcSizePlusOne-1, (U32)cctx->consumedSrcSize);
- return ERROR(srcSize_wrong);
- }
+ assert(!(cctx->appliedParams.fParams.contentSizeFlag && cctx->pledgedSrcSizePlusOne == 0));
+ if (cctx->pledgedSrcSizePlusOne != 0) { /* control src size */
+ ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_UNKNOWN == (unsigned long long)-1);
+ RETURN_ERROR_IF(
+ cctx->consumedSrcSize+1 > cctx->pledgedSrcSizePlusOne,
+ srcSize_wrong,
+ "error : pledgedSrcSize = %u, while realSrcSize >= %u",
+ (unsigned)cctx->pledgedSrcSizePlusOne-1,
+ (unsigned)cctx->consumedSrcSize);
}
return cSize + fhSize;
}
@@ -2168,7 +3095,7 @@
void* dst, size_t dstCapacity,
const void* src, size_t srcSize)
{
- DEBUGLOG(5, "ZSTD_compressContinue (srcSize=%u)", (U32)srcSize);
+ DEBUGLOG(5, "ZSTD_compressContinue (srcSize=%u)", (unsigned)srcSize);
return ZSTD_compressContinue_internal(cctx, dst, dstCapacity, src, srcSize, 1 /* frame mode */, 0 /* last chunk */);
}
@@ -2183,49 +3110,66 @@
size_t ZSTD_compressBlock(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
size_t const blockSizeMax = ZSTD_getBlockSize(cctx);
- if (srcSize > blockSizeMax) return ERROR(srcSize_wrong);
+ RETURN_ERROR_IF(srcSize > blockSizeMax, srcSize_wrong);
+
return ZSTD_compressContinue_internal(cctx, dst, dstCapacity, src, srcSize, 0 /* frame mode */, 0 /* last chunk */);
}
/*! ZSTD_loadDictionaryContent() :
* @return : 0, or an error code
*/
-static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms, ZSTD_CCtx_params const* params, const void* src, size_t srcSize)
+static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms,
+ ZSTD_CCtx_params const* params,
+ const void* src, size_t srcSize,
+ ZSTD_dictTableLoadMethod_e dtlm)
{
- const BYTE* const ip = (const BYTE*) src;
+ const BYTE* ip = (const BYTE*) src;
const BYTE* const iend = ip + srcSize;
- ZSTD_compressionParameters const* cParams = ¶ms->cParams;
ZSTD_window_update(&ms->window, src, srcSize);
ms->loadedDictEnd = params->forceWindow ? 0 : (U32)(iend - ms->window.base);
+ /* Assert that we the ms params match the params we're being given */
+ ZSTD_assertEqualCParams(params->cParams, ms->cParams);
+
if (srcSize <= HASH_READ_SIZE) return 0;
- switch(params->cParams.strategy)
- {
- case ZSTD_fast:
- ZSTD_fillHashTable(ms, cParams, iend);
- break;
- case ZSTD_dfast:
- ZSTD_fillDoubleHashTable(ms, cParams, iend);
- break;
+ while (iend - ip > HASH_READ_SIZE) {
+ size_t const remaining = iend - ip;
+ size_t const chunk = MIN(remaining, ZSTD_CHUNKSIZE_MAX);
+ const BYTE* const ichunk = ip + chunk;
- case ZSTD_greedy:
- case ZSTD_lazy:
- case ZSTD_lazy2:
- if (srcSize >= HASH_READ_SIZE)
- ZSTD_insertAndFindFirstIndex(ms, cParams, iend-HASH_READ_SIZE);
- break;
+ ZSTD_overflowCorrectIfNeeded(ms, params, ip, ichunk);
- case ZSTD_btlazy2: /* we want the dictionary table fully sorted */
- case ZSTD_btopt:
- case ZSTD_btultra:
- if (srcSize >= HASH_READ_SIZE)
- ZSTD_updateTree(ms, cParams, iend-HASH_READ_SIZE, iend);
- break;
+ switch(params->cParams.strategy)
+ {
+ case ZSTD_fast:
+ ZSTD_fillHashTable(ms, ichunk, dtlm);
+ break;
+ case ZSTD_dfast:
+ ZSTD_fillDoubleHashTable(ms, ichunk, dtlm);
+ break;
- default:
- assert(0); /* not possible : not a valid strategy id */
+ case ZSTD_greedy:
+ case ZSTD_lazy:
+ case ZSTD_lazy2:
+ if (chunk >= HASH_READ_SIZE)
+ ZSTD_insertAndFindFirstIndex(ms, ichunk-HASH_READ_SIZE);
+ break;
+
+ case ZSTD_btlazy2: /* we want the dictionary table fully sorted */
+ case ZSTD_btopt:
+ case ZSTD_btultra:
+ case ZSTD_btultra2:
+ if (chunk >= HASH_READ_SIZE)
+ ZSTD_updateTree(ms, ichunk-HASH_READ_SIZE, ichunk);
+ break;
+
+ default:
+ assert(0); /* not possible : not a valid strategy id */
+ }
+
+ ip = ichunk;
}
ms->nextToUpdate = (U32)(iend - ms->window.base);
@@ -2239,9 +3183,9 @@
NOTE: This behavior is not standard and could be improved in the future. */
static size_t ZSTD_checkDictNCount(short* normalizedCounter, unsigned dictMaxSymbolValue, unsigned maxSymbolValue) {
U32 s;
- if (dictMaxSymbolValue < maxSymbolValue) return ERROR(dictionary_corrupted);
+ RETURN_ERROR_IF(dictMaxSymbolValue < maxSymbolValue, dictionary_corrupted);
for (s = 0; s <= maxSymbolValue; ++s) {
- if (normalizedCounter[s] == 0) return ERROR(dictionary_corrupted);
+ RETURN_ERROR_IF(normalizedCounter[s] == 0, dictionary_corrupted);
}
return 0;
}
@@ -2256,7 +3200,12 @@
* assumptions : magic number supposed already checked
* dictSize supposed > 8
*/
-static size_t ZSTD_loadZstdDictionary(ZSTD_compressedBlockState_t* bs, ZSTD_matchState_t* ms, ZSTD_CCtx_params const* params, const void* dict, size_t dictSize, void* workspace)
+static size_t ZSTD_loadZstdDictionary(ZSTD_compressedBlockState_t* bs,
+ ZSTD_matchState_t* ms,
+ ZSTD_CCtx_params const* params,
+ const void* dict, size_t dictSize,
+ ZSTD_dictTableLoadMethod_e dtlm,
+ void* workspace)
{
const BYTE* dictPtr = (const BYTE*)dict;
const BYTE* const dictEnd = dictPtr + dictSize;
@@ -2265,53 +3214,65 @@
size_t dictID;
ZSTD_STATIC_ASSERT(HUF_WORKSPACE_SIZE >= (1<<MAX(MLFSELog,LLFSELog)));
+ assert(dictSize > 8);
+ assert(MEM_readLE32(dictPtr) == ZSTD_MAGIC_DICTIONARY);
dictPtr += 4; /* skip magic number */
dictID = params->fParams.noDictIDFlag ? 0 : MEM_readLE32(dictPtr);
dictPtr += 4;
{ unsigned maxSymbolValue = 255;
- size_t const hufHeaderSize = HUF_readCTable((HUF_CElt*)bs->entropy.hufCTable, &maxSymbolValue, dictPtr, dictEnd-dictPtr);
- if (HUF_isError(hufHeaderSize)) return ERROR(dictionary_corrupted);
- if (maxSymbolValue < 255) return ERROR(dictionary_corrupted);
+ size_t const hufHeaderSize = HUF_readCTable((HUF_CElt*)bs->entropy.huf.CTable, &maxSymbolValue, dictPtr, dictEnd-dictPtr);
+ RETURN_ERROR_IF(HUF_isError(hufHeaderSize), dictionary_corrupted);
+ RETURN_ERROR_IF(maxSymbolValue < 255, dictionary_corrupted);
dictPtr += hufHeaderSize;
}
{ unsigned offcodeLog;
size_t const offcodeHeaderSize = FSE_readNCount(offcodeNCount, &offcodeMaxValue, &offcodeLog, dictPtr, dictEnd-dictPtr);
- if (FSE_isError(offcodeHeaderSize)) return ERROR(dictionary_corrupted);
- if (offcodeLog > OffFSELog) return ERROR(dictionary_corrupted);
+ RETURN_ERROR_IF(FSE_isError(offcodeHeaderSize), dictionary_corrupted);
+ RETURN_ERROR_IF(offcodeLog > OffFSELog, dictionary_corrupted);
/* Defer checking offcodeMaxValue because we need to know the size of the dictionary content */
- CHECK_E( FSE_buildCTable_wksp(bs->entropy.offcodeCTable, offcodeNCount, offcodeMaxValue, offcodeLog, workspace, HUF_WORKSPACE_SIZE),
- dictionary_corrupted);
+ /* fill all offset symbols to avoid garbage at end of table */
+ RETURN_ERROR_IF(FSE_isError(FSE_buildCTable_wksp(
+ bs->entropy.fse.offcodeCTable,
+ offcodeNCount, MaxOff, offcodeLog,
+ workspace, HUF_WORKSPACE_SIZE)),
+ dictionary_corrupted);
dictPtr += offcodeHeaderSize;
}
{ short matchlengthNCount[MaxML+1];
unsigned matchlengthMaxValue = MaxML, matchlengthLog;
size_t const matchlengthHeaderSize = FSE_readNCount(matchlengthNCount, &matchlengthMaxValue, &matchlengthLog, dictPtr, dictEnd-dictPtr);
- if (FSE_isError(matchlengthHeaderSize)) return ERROR(dictionary_corrupted);
- if (matchlengthLog > MLFSELog) return ERROR(dictionary_corrupted);
+ RETURN_ERROR_IF(FSE_isError(matchlengthHeaderSize), dictionary_corrupted);
+ RETURN_ERROR_IF(matchlengthLog > MLFSELog, dictionary_corrupted);
/* Every match length code must have non-zero probability */
- CHECK_F( ZSTD_checkDictNCount(matchlengthNCount, matchlengthMaxValue, MaxML));
- CHECK_E( FSE_buildCTable_wksp(bs->entropy.matchlengthCTable, matchlengthNCount, matchlengthMaxValue, matchlengthLog, workspace, HUF_WORKSPACE_SIZE),
- dictionary_corrupted);
+ FORWARD_IF_ERROR( ZSTD_checkDictNCount(matchlengthNCount, matchlengthMaxValue, MaxML));
+ RETURN_ERROR_IF(FSE_isError(FSE_buildCTable_wksp(
+ bs->entropy.fse.matchlengthCTable,
+ matchlengthNCount, matchlengthMaxValue, matchlengthLog,
+ workspace, HUF_WORKSPACE_SIZE)),
+ dictionary_corrupted);
dictPtr += matchlengthHeaderSize;
}
{ short litlengthNCount[MaxLL+1];
unsigned litlengthMaxValue = MaxLL, litlengthLog;
size_t const litlengthHeaderSize = FSE_readNCount(litlengthNCount, &litlengthMaxValue, &litlengthLog, dictPtr, dictEnd-dictPtr);
- if (FSE_isError(litlengthHeaderSize)) return ERROR(dictionary_corrupted);
- if (litlengthLog > LLFSELog) return ERROR(dictionary_corrupted);
+ RETURN_ERROR_IF(FSE_isError(litlengthHeaderSize), dictionary_corrupted);
+ RETURN_ERROR_IF(litlengthLog > LLFSELog, dictionary_corrupted);
/* Every literal length code must have non-zero probability */
- CHECK_F( ZSTD_checkDictNCount(litlengthNCount, litlengthMaxValue, MaxLL));
- CHECK_E( FSE_buildCTable_wksp(bs->entropy.litlengthCTable, litlengthNCount, litlengthMaxValue, litlengthLog, workspace, HUF_WORKSPACE_SIZE),
- dictionary_corrupted);
+ FORWARD_IF_ERROR( ZSTD_checkDictNCount(litlengthNCount, litlengthMaxValue, MaxLL));
+ RETURN_ERROR_IF(FSE_isError(FSE_buildCTable_wksp(
+ bs->entropy.fse.litlengthCTable,
+ litlengthNCount, litlengthMaxValue, litlengthLog,
+ workspace, HUF_WORKSPACE_SIZE)),
+ dictionary_corrupted);
dictPtr += litlengthHeaderSize;
}
- if (dictPtr+12 > dictEnd) return ERROR(dictionary_corrupted);
+ RETURN_ERROR_IF(dictPtr+12 > dictEnd, dictionary_corrupted);
bs->rep[0] = MEM_readLE32(dictPtr+0);
bs->rep[1] = MEM_readLE32(dictPtr+4);
bs->rep[2] = MEM_readLE32(dictPtr+8);
@@ -2324,30 +3285,33 @@
offcodeMax = ZSTD_highbit32(maxOffset); /* Calculate minimum offset code required to represent maxOffset */
}
/* All offset values <= dictContentSize + 128 KB must be representable */
- CHECK_F (ZSTD_checkDictNCount(offcodeNCount, offcodeMaxValue, MIN(offcodeMax, MaxOff)));
+ FORWARD_IF_ERROR(ZSTD_checkDictNCount(offcodeNCount, offcodeMaxValue, MIN(offcodeMax, MaxOff)));
/* All repCodes must be <= dictContentSize and != 0*/
{ U32 u;
for (u=0; u<3; u++) {
- if (bs->rep[u] == 0) return ERROR(dictionary_corrupted);
- if (bs->rep[u] > dictContentSize) return ERROR(dictionary_corrupted);
+ RETURN_ERROR_IF(bs->rep[u] == 0, dictionary_corrupted);
+ RETURN_ERROR_IF(bs->rep[u] > dictContentSize, dictionary_corrupted);
} }
- bs->entropy.hufCTable_repeatMode = HUF_repeat_valid;
- bs->entropy.offcode_repeatMode = FSE_repeat_valid;
- bs->entropy.matchlength_repeatMode = FSE_repeat_valid;
- bs->entropy.litlength_repeatMode = FSE_repeat_valid;
- CHECK_F(ZSTD_loadDictionaryContent(ms, params, dictPtr, dictContentSize));
+ bs->entropy.huf.repeatMode = HUF_repeat_valid;
+ bs->entropy.fse.offcode_repeatMode = FSE_repeat_valid;
+ bs->entropy.fse.matchlength_repeatMode = FSE_repeat_valid;
+ bs->entropy.fse.litlength_repeatMode = FSE_repeat_valid;
+ FORWARD_IF_ERROR(ZSTD_loadDictionaryContent(ms, params, dictPtr, dictContentSize, dtlm));
return dictID;
}
}
/** ZSTD_compress_insertDictionary() :
* @return : dictID, or an error code */
-static size_t ZSTD_compress_insertDictionary(ZSTD_compressedBlockState_t* bs, ZSTD_matchState_t* ms,
- ZSTD_CCtx_params const* params,
- const void* dict, size_t dictSize,
- ZSTD_dictContentType_e dictContentType,
- void* workspace)
+static size_t
+ZSTD_compress_insertDictionary(ZSTD_compressedBlockState_t* bs,
+ ZSTD_matchState_t* ms,
+ const ZSTD_CCtx_params* params,
+ const void* dict, size_t dictSize,
+ ZSTD_dictContentType_e dictContentType,
+ ZSTD_dictTableLoadMethod_e dtlm,
+ void* workspace)
{
DEBUGLOG(4, "ZSTD_compress_insertDictionary (dictSize=%u)", (U32)dictSize);
if ((dict==NULL) || (dictSize<=8)) return 0;
@@ -2356,30 +3320,30 @@
/* dict restricted modes */
if (dictContentType == ZSTD_dct_rawContent)
- return ZSTD_loadDictionaryContent(ms, params, dict, dictSize);
+ return ZSTD_loadDictionaryContent(ms, params, dict, dictSize, dtlm);
if (MEM_readLE32(dict) != ZSTD_MAGIC_DICTIONARY) {
if (dictContentType == ZSTD_dct_auto) {
DEBUGLOG(4, "raw content dictionary detected");
- return ZSTD_loadDictionaryContent(ms, params, dict, dictSize);
+ return ZSTD_loadDictionaryContent(ms, params, dict, dictSize, dtlm);
}
- if (dictContentType == ZSTD_dct_fullDict)
- return ERROR(dictionary_wrong);
+ RETURN_ERROR_IF(dictContentType == ZSTD_dct_fullDict, dictionary_wrong);
assert(0); /* impossible */
}
/* dict as full zstd dictionary */
- return ZSTD_loadZstdDictionary(bs, ms, params, dict, dictSize, workspace);
+ return ZSTD_loadZstdDictionary(bs, ms, params, dict, dictSize, dtlm, workspace);
}
/*! ZSTD_compressBegin_internal() :
* @return : 0, or an error code */
-size_t ZSTD_compressBegin_internal(ZSTD_CCtx* cctx,
- const void* dict, size_t dictSize,
- ZSTD_dictContentType_e dictContentType,
- const ZSTD_CDict* cdict,
- ZSTD_CCtx_params params, U64 pledgedSrcSize,
- ZSTD_buffered_policy_e zbuff)
+static size_t ZSTD_compressBegin_internal(ZSTD_CCtx* cctx,
+ const void* dict, size_t dictSize,
+ ZSTD_dictContentType_e dictContentType,
+ ZSTD_dictTableLoadMethod_e dtlm,
+ const ZSTD_CDict* cdict,
+ ZSTD_CCtx_params params, U64 pledgedSrcSize,
+ ZSTD_buffered_policy_e zbuff)
{
DEBUGLOG(4, "ZSTD_compressBegin_internal: wlog=%u", params.cParams.windowLog);
/* params are supposed to be fully validated at this point */
@@ -2387,19 +3351,16 @@
assert(!((dict) && (cdict))); /* either dict or cdict, not both */
if (cdict && cdict->dictContentSize>0) {
- cctx->requestedParams = params;
- return ZSTD_resetCCtx_usingCDict(cctx, cdict, params.cParams.windowLog,
- params.fParams, pledgedSrcSize, zbuff);
+ return ZSTD_resetCCtx_usingCDict(cctx, cdict, params, pledgedSrcSize, zbuff);
}
- CHECK_F( ZSTD_resetCCtx_internal(cctx, params, pledgedSrcSize,
+ FORWARD_IF_ERROR( ZSTD_resetCCtx_internal(cctx, params, pledgedSrcSize,
ZSTDcrp_continue, zbuff) );
- {
- size_t const dictID = ZSTD_compress_insertDictionary(
+ { size_t const dictID = ZSTD_compress_insertDictionary(
cctx->blockState.prevCBlock, &cctx->blockState.matchState,
- ¶ms, dict, dictSize, dictContentType, cctx->entropyWorkspace);
- if (ZSTD_isError(dictID)) return dictID;
- assert(dictID <= (size_t)(U32)-1);
+ ¶ms, dict, dictSize, dictContentType, dtlm, cctx->entropyWorkspace);
+ FORWARD_IF_ERROR(dictID);
+ assert(dictID <= UINT_MAX);
cctx->dictID = (U32)dictID;
}
return 0;
@@ -2408,15 +3369,16 @@
size_t ZSTD_compressBegin_advanced_internal(ZSTD_CCtx* cctx,
const void* dict, size_t dictSize,
ZSTD_dictContentType_e dictContentType,
+ ZSTD_dictTableLoadMethod_e dtlm,
const ZSTD_CDict* cdict,
ZSTD_CCtx_params params,
unsigned long long pledgedSrcSize)
{
DEBUGLOG(4, "ZSTD_compressBegin_advanced_internal: wlog=%u", params.cParams.windowLog);
/* compression parameters verification and optimization */
- CHECK_F( ZSTD_checkCParams(params.cParams) );
+ FORWARD_IF_ERROR( ZSTD_checkCParams(params.cParams) );
return ZSTD_compressBegin_internal(cctx,
- dict, dictSize, dictContentType,
+ dict, dictSize, dictContentType, dtlm,
cdict,
params, pledgedSrcSize,
ZSTDb_not_buffered);
@@ -2431,7 +3393,7 @@
ZSTD_CCtx_params const cctxParams =
ZSTD_assignParamsToCCtxParams(cctx->requestedParams, params);
return ZSTD_compressBegin_advanced_internal(cctx,
- dict, dictSize, ZSTD_dct_auto,
+ dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast,
NULL /*cdict*/,
cctxParams, pledgedSrcSize);
}
@@ -2441,8 +3403,8 @@
ZSTD_parameters const params = ZSTD_getParams(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize);
ZSTD_CCtx_params const cctxParams =
ZSTD_assignParamsToCCtxParams(cctx->requestedParams, params);
- DEBUGLOG(4, "ZSTD_compressBegin_usingDict (dictSize=%u)", (U32)dictSize);
- return ZSTD_compressBegin_internal(cctx, dict, dictSize, ZSTD_dct_auto, NULL,
+ DEBUGLOG(4, "ZSTD_compressBegin_usingDict (dictSize=%u)", (unsigned)dictSize);
+ return ZSTD_compressBegin_internal(cctx, dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast, NULL,
cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, ZSTDb_not_buffered);
}
@@ -2462,12 +3424,12 @@
size_t fhSize = 0;
DEBUGLOG(4, "ZSTD_writeEpilogue");
- if (cctx->stage == ZSTDcs_created) return ERROR(stage_wrong); /* init missing */
+ RETURN_ERROR_IF(cctx->stage == ZSTDcs_created, stage_wrong, "init missing");
/* special case : empty frame */
if (cctx->stage == ZSTDcs_init) {
fhSize = ZSTD_writeFrameHeader(dst, dstCapacity, cctx->appliedParams, 0, 0);
- if (ZSTD_isError(fhSize)) return fhSize;
+ FORWARD_IF_ERROR(fhSize);
dstCapacity -= fhSize;
op += fhSize;
cctx->stage = ZSTDcs_ongoing;
@@ -2476,7 +3438,7 @@
if (cctx->stage != ZSTDcs_ending) {
/* write one last empty block, make it the "last" block */
U32 const cBlockHeader24 = 1 /* last block */ + (((U32)bt_raw)<<1) + 0;
- if (dstCapacity<4) return ERROR(dstSize_tooSmall);
+ RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall);
MEM_writeLE32(op, cBlockHeader24);
op += ZSTD_blockHeaderSize;
dstCapacity -= ZSTD_blockHeaderSize;
@@ -2484,8 +3446,8 @@
if (cctx->appliedParams.fParams.checksumFlag) {
U32 const checksum = (U32) XXH64_digest(&cctx->xxhState);
- if (dstCapacity<4) return ERROR(dstSize_tooSmall);
- DEBUGLOG(4, "ZSTD_writeEpilogue: write checksum : %08X", checksum);
+ RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall);
+ DEBUGLOG(4, "ZSTD_writeEpilogue: write checksum : %08X", (unsigned)checksum);
MEM_writeLE32(op, checksum);
op += 4;
}
@@ -2502,45 +3464,53 @@
size_t const cSize = ZSTD_compressContinue_internal(cctx,
dst, dstCapacity, src, srcSize,
1 /* frame mode */, 1 /* last chunk */);
- if (ZSTD_isError(cSize)) return cSize;
+ FORWARD_IF_ERROR(cSize);
endResult = ZSTD_writeEpilogue(cctx, (char*)dst + cSize, dstCapacity-cSize);
- if (ZSTD_isError(endResult)) return endResult;
- if (cctx->appliedParams.fParams.contentSizeFlag) { /* control src size */
+ FORWARD_IF_ERROR(endResult);
+ assert(!(cctx->appliedParams.fParams.contentSizeFlag && cctx->pledgedSrcSizePlusOne == 0));
+ if (cctx->pledgedSrcSizePlusOne != 0) { /* control src size */
+ ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_UNKNOWN == (unsigned long long)-1);
DEBUGLOG(4, "end of frame : controlling src size");
- if (cctx->pledgedSrcSizePlusOne != cctx->consumedSrcSize+1) {
- DEBUGLOG(4, "error : pledgedSrcSize = %u, while realSrcSize = %u",
- (U32)cctx->pledgedSrcSizePlusOne-1, (U32)cctx->consumedSrcSize);
- return ERROR(srcSize_wrong);
- } }
+ RETURN_ERROR_IF(
+ cctx->pledgedSrcSizePlusOne != cctx->consumedSrcSize+1,
+ srcSize_wrong,
+ "error : pledgedSrcSize = %u, while realSrcSize = %u",
+ (unsigned)cctx->pledgedSrcSizePlusOne-1,
+ (unsigned)cctx->consumedSrcSize);
+ }
return cSize + endResult;
}
static size_t ZSTD_compress_internal (ZSTD_CCtx* cctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- const void* dict,size_t dictSize,
- ZSTD_parameters params)
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ const void* dict,size_t dictSize,
+ ZSTD_parameters params)
{
ZSTD_CCtx_params const cctxParams =
ZSTD_assignParamsToCCtxParams(cctx->requestedParams, params);
DEBUGLOG(4, "ZSTD_compress_internal");
return ZSTD_compress_advanced_internal(cctx,
- dst, dstCapacity,
- src, srcSize,
- dict, dictSize,
- cctxParams);
+ dst, dstCapacity,
+ src, srcSize,
+ dict, dictSize,
+ cctxParams);
}
-size_t ZSTD_compress_advanced (ZSTD_CCtx* ctx,
+size_t ZSTD_compress_advanced (ZSTD_CCtx* cctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const void* dict,size_t dictSize,
ZSTD_parameters params)
{
DEBUGLOG(4, "ZSTD_compress_advanced");
- CHECK_F(ZSTD_checkCParams(params.cParams));
- return ZSTD_compress_internal(ctx, dst, dstCapacity, src, srcSize, dict, dictSize, params);
+ FORWARD_IF_ERROR(ZSTD_checkCParams(params.cParams));
+ return ZSTD_compress_internal(cctx,
+ dst, dstCapacity,
+ src, srcSize,
+ dict, dictSize,
+ params);
}
/* Internal */
@@ -2551,37 +3521,44 @@
const void* dict,size_t dictSize,
ZSTD_CCtx_params params)
{
- DEBUGLOG(4, "ZSTD_compress_advanced_internal (srcSize:%u)",
- (U32)srcSize);
- CHECK_F( ZSTD_compressBegin_internal(cctx, dict, dictSize, ZSTD_dct_auto, NULL,
- params, srcSize, ZSTDb_not_buffered) );
+ DEBUGLOG(4, "ZSTD_compress_advanced_internal (srcSize:%u)", (unsigned)srcSize);
+ FORWARD_IF_ERROR( ZSTD_compressBegin_internal(cctx,
+ dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast, NULL,
+ params, srcSize, ZSTDb_not_buffered) );
return ZSTD_compressEnd(cctx, dst, dstCapacity, src, srcSize);
}
-size_t ZSTD_compress_usingDict(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize,
- const void* dict, size_t dictSize, int compressionLevel)
+size_t ZSTD_compress_usingDict(ZSTD_CCtx* cctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ const void* dict, size_t dictSize,
+ int compressionLevel)
{
- ZSTD_parameters const params = ZSTD_getParams(compressionLevel, srcSize ? srcSize : 1, dict ? dictSize : 0);
+ ZSTD_parameters const params = ZSTD_getParams(compressionLevel, srcSize + (!srcSize), dict ? dictSize : 0);
ZSTD_CCtx_params cctxParams = ZSTD_assignParamsToCCtxParams(cctx->requestedParams, params);
assert(params.fParams.contentSizeFlag == 1);
- ZSTD_CCtxParam_setParameter(&cctxParams, ZSTD_p_compressLiterals, compressionLevel>=0);
return ZSTD_compress_advanced_internal(cctx, dst, dstCapacity, src, srcSize, dict, dictSize, cctxParams);
}
-size_t ZSTD_compressCCtx (ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize, int compressionLevel)
+size_t ZSTD_compressCCtx(ZSTD_CCtx* cctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ int compressionLevel)
{
- DEBUGLOG(4, "ZSTD_compressCCtx (srcSize=%u)", (U32)srcSize);
+ DEBUGLOG(4, "ZSTD_compressCCtx (srcSize=%u)", (unsigned)srcSize);
+ assert(cctx != NULL);
return ZSTD_compress_usingDict(cctx, dst, dstCapacity, src, srcSize, NULL, 0, compressionLevel);
}
-size_t ZSTD_compress(void* dst, size_t dstCapacity, const void* src, size_t srcSize, int compressionLevel)
+size_t ZSTD_compress(void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ int compressionLevel)
{
size_t result;
ZSTD_CCtx ctxBody;
- memset(&ctxBody, 0, sizeof(ctxBody));
- ctxBody.customMem = ZSTD_defaultCMem;
+ ZSTD_initCCtx(&ctxBody, ZSTD_defaultCMem);
result = ZSTD_compressCCtx(&ctxBody, dst, dstCapacity, src, srcSize, compressionLevel);
- ZSTD_free(ctxBody.workSpace, ZSTD_defaultCMem); /* can't free ctxBody itself, as it's on stack; free only heap content */
+ ZSTD_freeCCtxContent(&ctxBody); /* can't free ctxBody itself, as it's on stack; free only heap content */
return result;
}
@@ -2594,7 +3571,7 @@
size_t dictSize, ZSTD_compressionParameters cParams,
ZSTD_dictLoadMethod_e dictLoadMethod)
{
- DEBUGLOG(5, "sizeof(ZSTD_CDict) : %u", (U32)sizeof(ZSTD_CDict));
+ DEBUGLOG(5, "sizeof(ZSTD_CDict) : %u", (unsigned)sizeof(ZSTD_CDict));
return sizeof(ZSTD_CDict) + HUF_WORKSPACE_SIZE + ZSTD_sizeof_matchState(&cParams, /* forCCtx */ 0)
+ (dictLoadMethod == ZSTD_dlm_byRef ? 0 : dictSize);
}
@@ -2608,7 +3585,7 @@
size_t ZSTD_sizeof_CDict(const ZSTD_CDict* cdict)
{
if (cdict==NULL) return 0; /* support sizeof on NULL */
- DEBUGLOG(5, "sizeof(*cdict) : %u", (U32)sizeof(*cdict));
+ DEBUGLOG(5, "sizeof(*cdict) : %u", (unsigned)sizeof(*cdict));
return cdict->workspaceSize + (cdict->dictBuffer ? cdict->dictContentSize : 0) + sizeof(*cdict);
}
@@ -2619,9 +3596,9 @@
ZSTD_dictContentType_e dictContentType,
ZSTD_compressionParameters cParams)
{
- DEBUGLOG(3, "ZSTD_initCDict_internal, dictContentType %u", (U32)dictContentType);
+ DEBUGLOG(3, "ZSTD_initCDict_internal (dictContentType:%u)", (unsigned)dictContentType);
assert(!ZSTD_checkCParams(cParams));
- cdict->cParams = cParams;
+ cdict->matchState.cParams = cParams;
if ((dictLoadMethod == ZSTD_dlm_byRef) || (!dictBuffer) || (!dictSize)) {
cdict->dictBuffer = NULL;
cdict->dictContent = dictBuffer;
@@ -2629,17 +3606,17 @@
void* const internalBuffer = ZSTD_malloc(dictSize, cdict->customMem);
cdict->dictBuffer = internalBuffer;
cdict->dictContent = internalBuffer;
- if (!internalBuffer) return ERROR(memory_allocation);
+ RETURN_ERROR_IF(!internalBuffer, memory_allocation);
memcpy(internalBuffer, dictBuffer, dictSize);
}
cdict->dictContentSize = dictSize;
/* Reset the state to no dictionary */
ZSTD_reset_compressedBlockState(&cdict->cBlockState);
- { void* const end = ZSTD_reset_matchState(
- &cdict->matchState,
- (U32*)cdict->workspace + HUF_WORKSPACE_SIZE_U32,
- &cParams, ZSTDcrp_continue, /* forCCtx */ 0);
+ { void* const end = ZSTD_reset_matchState(&cdict->matchState,
+ (U32*)cdict->workspace + HUF_WORKSPACE_SIZE_U32,
+ &cParams,
+ ZSTDcrp_continue, ZSTD_resetTarget_CDict);
assert(end == (char*)cdict->workspace + cdict->workspaceSize);
(void)end;
}
@@ -2654,8 +3631,8 @@
{ size_t const dictID = ZSTD_compress_insertDictionary(
&cdict->cBlockState, &cdict->matchState, ¶ms,
cdict->dictContent, cdict->dictContentSize,
- dictContentType, cdict->workspace);
- if (ZSTD_isError(dictID)) return dictID;
+ dictContentType, ZSTD_dtlm_full, cdict->workspace);
+ FORWARD_IF_ERROR(dictID);
assert(dictID <= (size_t)(U32)-1);
cdict->dictID = (U32)dictID;
}
@@ -2669,7 +3646,7 @@
ZSTD_dictContentType_e dictContentType,
ZSTD_compressionParameters cParams, ZSTD_customMem customMem)
{
- DEBUGLOG(3, "ZSTD_createCDict_advanced, mode %u", (U32)dictContentType);
+ DEBUGLOG(3, "ZSTD_createCDict_advanced, mode %u", (unsigned)dictContentType);
if (!customMem.customAlloc ^ !customMem.customFree) return NULL;
{ ZSTD_CDict* const cdict = (ZSTD_CDict*)ZSTD_malloc(sizeof(ZSTD_CDict), customMem);
@@ -2750,7 +3727,7 @@
void* ptr;
if ((size_t)workspace & 7) return NULL; /* 8-aligned */
DEBUGLOG(4, "(workspaceSize < neededSize) : (%u < %u) => %u",
- (U32)workspaceSize, (U32)neededSize, (U32)(workspaceSize < neededSize));
+ (unsigned)workspaceSize, (unsigned)neededSize, (unsigned)(workspaceSize < neededSize));
if (workspaceSize < neededSize) return NULL;
if (dictLoadMethod == ZSTD_dlm_byCopy) {
@@ -2775,7 +3752,7 @@
ZSTD_compressionParameters ZSTD_getCParamsFromCDict(const ZSTD_CDict* cdict)
{
assert(cdict != NULL);
- return cdict->cParams;
+ return cdict->matchState.cParams;
}
/* ZSTD_compressBegin_usingCDict_advanced() :
@@ -2785,7 +3762,7 @@
ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize)
{
DEBUGLOG(4, "ZSTD_compressBegin_usingCDict_advanced");
- if (cdict==NULL) return ERROR(dictionary_wrong);
+ RETURN_ERROR_IF(cdict==NULL, dictionary_wrong);
{ ZSTD_CCtx_params params = cctx->requestedParams;
params.cParams = ZSTD_getCParamsFromCDict(cdict);
/* Increase window log to fit the entire dictionary and source if the
@@ -2799,7 +3776,7 @@
}
params.fParams = fParams;
return ZSTD_compressBegin_internal(cctx,
- NULL, 0, ZSTD_dct_auto,
+ NULL, 0, ZSTD_dct_auto, ZSTD_dtlm_fast,
cdict,
params, pledgedSrcSize,
ZSTDb_not_buffered);
@@ -2813,7 +3790,7 @@
{
ZSTD_frameParameters const fParams = { 0 /*content*/, 0 /*checksum*/, 0 /*noDictID*/ };
DEBUGLOG(4, "ZSTD_compressBegin_usingCDict : dictIDFlag == %u", !fParams.noDictIDFlag);
- return ZSTD_compressBegin_usingCDict_advanced(cctx, cdict, fParams, 0);
+ return ZSTD_compressBegin_usingCDict_advanced(cctx, cdict, fParams, ZSTD_CONTENTSIZE_UNKNOWN);
}
size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx,
@@ -2821,7 +3798,7 @@
const void* src, size_t srcSize,
const ZSTD_CDict* cdict, ZSTD_frameParameters fParams)
{
- CHECK_F (ZSTD_compressBegin_usingCDict_advanced(cctx, cdict, fParams, srcSize)); /* will check if cdict != NULL */
+ FORWARD_IF_ERROR(ZSTD_compressBegin_usingCDict_advanced(cctx, cdict, fParams, srcSize)); /* will check if cdict != NULL */
return ZSTD_compressEnd(cctx, dst, dstCapacity, src, srcSize);
}
@@ -2880,16 +3857,17 @@
static size_t ZSTD_resetCStream_internal(ZSTD_CStream* cctx,
const void* const dict, size_t const dictSize, ZSTD_dictContentType_e const dictContentType,
const ZSTD_CDict* const cdict,
- ZSTD_CCtx_params const params, unsigned long long const pledgedSrcSize)
+ ZSTD_CCtx_params params, unsigned long long const pledgedSrcSize)
{
- DEBUGLOG(4, "ZSTD_resetCStream_internal (disableLiteralCompression=%i)",
- params.disableLiteralCompression);
+ DEBUGLOG(4, "ZSTD_resetCStream_internal");
+ /* Finalize the compression parameters */
+ params.cParams = ZSTD_getCParamsFromCCtxParams(¶ms, pledgedSrcSize, dictSize);
/* params are supposed to be fully validated at this point */
assert(!ZSTD_isError(ZSTD_checkCParams(params.cParams)));
assert(!((dict) && (cdict))); /* either dict or cdict, not both */
- CHECK_F( ZSTD_compressBegin_internal(cctx,
- dict, dictSize, dictContentType,
+ FORWARD_IF_ERROR( ZSTD_compressBegin_internal(cctx,
+ dict, dictSize, dictContentType, ZSTD_dtlm_fast,
cdict,
params, pledgedSrcSize,
ZSTDb_buffered) );
@@ -2906,14 +3884,17 @@
/* ZSTD_resetCStream():
* pledgedSrcSize == 0 means "unknown" */
-size_t ZSTD_resetCStream(ZSTD_CStream* zcs, unsigned long long pledgedSrcSize)
+size_t ZSTD_resetCStream(ZSTD_CStream* zcs, unsigned long long pss)
{
- ZSTD_CCtx_params params = zcs->requestedParams;
- DEBUGLOG(4, "ZSTD_resetCStream: pledgedSrcSize = %u", (U32)pledgedSrcSize);
- if (pledgedSrcSize==0) pledgedSrcSize = ZSTD_CONTENTSIZE_UNKNOWN;
- params.fParams.contentSizeFlag = 1;
- params.cParams = ZSTD_getCParamsFromCCtxParams(¶ms, pledgedSrcSize, 0);
- return ZSTD_resetCStream_internal(zcs, NULL, 0, ZSTD_dct_auto, zcs->cdict, params, pledgedSrcSize);
+ /* temporary : 0 interpreted as "unknown" during transition period.
+ * Users willing to specify "unknown" **must** use ZSTD_CONTENTSIZE_UNKNOWN.
+ * 0 will be interpreted as "empty" in the future.
+ */
+ U64 const pledgedSrcSize = (pss==0) ? ZSTD_CONTENTSIZE_UNKNOWN : pss;
+ DEBUGLOG(4, "ZSTD_resetCStream: pledgedSrcSize = %u", (unsigned)pledgedSrcSize);
+ FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) );
+ return 0;
}
/*! ZSTD_initCStream_internal() :
@@ -2925,31 +3906,18 @@
ZSTD_CCtx_params params, unsigned long long pledgedSrcSize)
{
DEBUGLOG(4, "ZSTD_initCStream_internal");
+ FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) );
assert(!ZSTD_isError(ZSTD_checkCParams(params.cParams)));
+ zcs->requestedParams = params;
assert(!((dict) && (cdict))); /* either dict or cdict, not both */
-
- if (dict && dictSize >= 8) {
- DEBUGLOG(4, "loading dictionary of size %u", (U32)dictSize);
- if (zcs->staticSize) { /* static CCtx : never uses malloc */
- /* incompatible with internal cdict creation */
- return ERROR(memory_allocation);
- }
- ZSTD_freeCDict(zcs->cdictLocal);
- zcs->cdictLocal = ZSTD_createCDict_advanced(dict, dictSize,
- ZSTD_dlm_byCopy, ZSTD_dct_auto,
- params.cParams, zcs->customMem);
- zcs->cdict = zcs->cdictLocal;
- if (zcs->cdictLocal == NULL) return ERROR(memory_allocation);
+ if (dict) {
+ FORWARD_IF_ERROR( ZSTD_CCtx_loadDictionary(zcs, dict, dictSize) );
} else {
- if (cdict) {
- params.cParams = ZSTD_getCParamsFromCDict(cdict); /* cParams are enforced from cdict; it includes windowLog */
- }
- ZSTD_freeCDict(zcs->cdictLocal);
- zcs->cdictLocal = NULL;
- zcs->cdict = cdict;
+ /* Dictionary is cleared if !cdict */
+ FORWARD_IF_ERROR( ZSTD_CCtx_refCDict(zcs, cdict) );
}
-
- return ZSTD_resetCStream_internal(zcs, NULL, 0, ZSTD_dct_auto, zcs->cdict, params, pledgedSrcSize);
+ return 0;
}
/* ZSTD_initCStream_usingCDict_advanced() :
@@ -2960,22 +3928,20 @@
unsigned long long pledgedSrcSize)
{
DEBUGLOG(4, "ZSTD_initCStream_usingCDict_advanced");
- if (!cdict) return ERROR(dictionary_wrong); /* cannot handle NULL cdict (does not know what to do) */
- { ZSTD_CCtx_params params = zcs->requestedParams;
- params.cParams = ZSTD_getCParamsFromCDict(cdict);
- params.fParams = fParams;
- return ZSTD_initCStream_internal(zcs,
- NULL, 0, cdict,
- params, pledgedSrcSize);
- }
+ FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) );
+ zcs->requestedParams.fParams = fParams;
+ FORWARD_IF_ERROR( ZSTD_CCtx_refCDict(zcs, cdict) );
+ return 0;
}
/* note : cdict must outlive compression session */
size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict)
{
- ZSTD_frameParameters const fParams = { 0 /* contentSizeFlag */, 0 /* checksum */, 0 /* hideDictID */ };
DEBUGLOG(4, "ZSTD_initCStream_usingCDict");
- return ZSTD_initCStream_usingCDict_advanced(zcs, cdict, fParams, ZSTD_CONTENTSIZE_UNKNOWN); /* note : will check that cdict != NULL */
+ FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_refCDict(zcs, cdict) );
+ return 0;
}
@@ -2985,43 +3951,66 @@
* dict is loaded with default parameters ZSTD_dm_auto and ZSTD_dlm_byCopy. */
size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs,
const void* dict, size_t dictSize,
- ZSTD_parameters params, unsigned long long pledgedSrcSize)
+ ZSTD_parameters params, unsigned long long pss)
{
- DEBUGLOG(4, "ZSTD_initCStream_advanced: pledgedSrcSize=%u, flag=%u",
- (U32)pledgedSrcSize, params.fParams.contentSizeFlag);
- CHECK_F( ZSTD_checkCParams(params.cParams) );
- if ((pledgedSrcSize==0) && (params.fParams.contentSizeFlag==0)) pledgedSrcSize = ZSTD_CONTENTSIZE_UNKNOWN; /* for compatibility with older programs relying on this behavior. Users should now specify ZSTD_CONTENTSIZE_UNKNOWN. This line will be removed in the future. */
- { ZSTD_CCtx_params const cctxParams = ZSTD_assignParamsToCCtxParams(zcs->requestedParams, params);
- return ZSTD_initCStream_internal(zcs, dict, dictSize, NULL /*cdict*/, cctxParams, pledgedSrcSize);
- }
+ /* for compatibility with older programs relying on this behavior.
+ * Users should now specify ZSTD_CONTENTSIZE_UNKNOWN.
+ * This line will be removed in the future.
+ */
+ U64 const pledgedSrcSize = (pss==0 && params.fParams.contentSizeFlag==0) ? ZSTD_CONTENTSIZE_UNKNOWN : pss;
+ DEBUGLOG(4, "ZSTD_initCStream_advanced");
+ FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) );
+ FORWARD_IF_ERROR( ZSTD_checkCParams(params.cParams) );
+ zcs->requestedParams = ZSTD_assignParamsToCCtxParams(zcs->requestedParams, params);
+ FORWARD_IF_ERROR( ZSTD_CCtx_loadDictionary(zcs, dict, dictSize) );
+ return 0;
}
size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel)
{
- ZSTD_parameters const params = ZSTD_getParams(compressionLevel, 0, dictSize);
- ZSTD_CCtx_params const cctxParams =
- ZSTD_assignParamsToCCtxParams(zcs->requestedParams, params);
- return ZSTD_initCStream_internal(zcs, dict, dictSize, NULL, cctxParams, ZSTD_CONTENTSIZE_UNKNOWN);
+ DEBUGLOG(4, "ZSTD_initCStream_usingDict");
+ FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_loadDictionary(zcs, dict, dictSize) );
+ return 0;
}
size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, int compressionLevel, unsigned long long pss)
{
- U64 const pledgedSrcSize = (pss==0) ? ZSTD_CONTENTSIZE_UNKNOWN : pss; /* temporary : 0 interpreted as "unknown" during transition period. Users willing to specify "unknown" **must** use ZSTD_CONTENTSIZE_UNKNOWN. `0` will be interpreted as "empty" in the future */
- ZSTD_parameters const params = ZSTD_getParams(compressionLevel, pledgedSrcSize, 0);
- ZSTD_CCtx_params const cctxParams = ZSTD_assignParamsToCCtxParams(zcs->requestedParams, params);
- return ZSTD_initCStream_internal(zcs, NULL, 0, NULL, cctxParams, pledgedSrcSize);
+ /* temporary : 0 interpreted as "unknown" during transition period.
+ * Users willing to specify "unknown" **must** use ZSTD_CONTENTSIZE_UNKNOWN.
+ * 0 will be interpreted as "empty" in the future.
+ */
+ U64 const pledgedSrcSize = (pss==0) ? ZSTD_CONTENTSIZE_UNKNOWN : pss;
+ DEBUGLOG(4, "ZSTD_initCStream_srcSize");
+ FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_refCDict(zcs, NULL) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) );
+ return 0;
}
size_t ZSTD_initCStream(ZSTD_CStream* zcs, int compressionLevel)
{
DEBUGLOG(4, "ZSTD_initCStream");
- return ZSTD_initCStream_srcSize(zcs, compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN);
+ FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_refCDict(zcs, NULL) );
+ FORWARD_IF_ERROR( ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel) );
+ return 0;
}
/*====== Compression ======*/
-MEM_STATIC size_t ZSTD_limitCopy(void* dst, size_t dstCapacity,
- const void* src, size_t srcSize)
+static size_t ZSTD_nextInputSizeHint(const ZSTD_CCtx* cctx)
+{
+ size_t hintInSize = cctx->inBuffTarget - cctx->inBuffPos;
+ if (hintInSize==0) hintInSize = cctx->blockSize;
+ return hintInSize;
+}
+
+static size_t ZSTD_limitCopy(void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize)
{
size_t const length = MIN(dstCapacity, srcSize);
if (length) memcpy(dst, src, length);
@@ -3029,13 +4018,13 @@
}
/** ZSTD_compressStream_generic():
- * internal function for all *compressStream*() variants and *compress_generic()
+ * internal function for all *compressStream*() variants
* non-static, because can be called from zstdmt_compress.c
* @return : hint size for next input */
-size_t ZSTD_compressStream_generic(ZSTD_CStream* zcs,
- ZSTD_outBuffer* output,
- ZSTD_inBuffer* input,
- ZSTD_EndDirective const flushMode)
+static size_t ZSTD_compressStream_generic(ZSTD_CStream* zcs,
+ ZSTD_outBuffer* output,
+ ZSTD_inBuffer* input,
+ ZSTD_EndDirective const flushMode)
{
const char* const istart = (const char*)input->src;
const char* const iend = istart + input->size;
@@ -3046,7 +4035,7 @@
U32 someMoreWork = 1;
/* check expectations */
- DEBUGLOG(5, "ZSTD_compressStream_generic, flush=%u", (U32)flushMode);
+ DEBUGLOG(5, "ZSTD_compressStream_generic, flush=%u", (unsigned)flushMode);
assert(zcs->inBuff != NULL);
assert(zcs->inBuffSize > 0);
assert(zcs->outBuff != NULL);
@@ -3058,8 +4047,7 @@
switch(zcs->streamStage)
{
case zcss_init:
- /* call ZSTD_initCStream() first ! */
- return ERROR(init_missing);
+ RETURN_ERROR(init_missing, "call ZSTD_initCStream() first!");
case zcss_load:
if ( (flushMode == ZSTD_e_end)
@@ -3068,12 +4056,12 @@
/* shortcut to compression pass directly into output buffer */
size_t const cSize = ZSTD_compressEnd(zcs,
op, oend-op, ip, iend-ip);
- DEBUGLOG(4, "ZSTD_compressEnd : %u", (U32)cSize);
- if (ZSTD_isError(cSize)) return cSize;
+ DEBUGLOG(4, "ZSTD_compressEnd : cSize=%u", (unsigned)cSize);
+ FORWARD_IF_ERROR(cSize);
ip = iend;
op += cSize;
zcs->frameEnded = 1;
- ZSTD_startNewCompression(zcs);
+ ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
someMoreWork = 0; break;
}
/* complete loading into inBuffer */
@@ -3110,14 +4098,14 @@
zcs->inBuff + zcs->inToCompress, iSize) :
ZSTD_compressContinue(zcs, cDst, oSize,
zcs->inBuff + zcs->inToCompress, iSize);
- if (ZSTD_isError(cSize)) return cSize;
+ FORWARD_IF_ERROR(cSize);
zcs->frameEnded = lastBlock;
/* prepare next block */
zcs->inBuffTarget = zcs->inBuffPos + zcs->blockSize;
if (zcs->inBuffTarget > zcs->inBuffSize)
zcs->inBuffPos = 0, zcs->inBuffTarget = zcs->blockSize;
DEBUGLOG(5, "inBuffTarget:%u / inBuffSize:%u",
- (U32)zcs->inBuffTarget, (U32)zcs->inBuffSize);
+ (unsigned)zcs->inBuffTarget, (unsigned)zcs->inBuffSize);
if (!lastBlock)
assert(zcs->inBuffTarget <= zcs->inBuffSize);
zcs->inToCompress = zcs->inBuffPos;
@@ -3126,7 +4114,7 @@
if (zcs->frameEnded) {
DEBUGLOG(5, "Frame completed directly in outBuffer");
someMoreWork = 0;
- ZSTD_startNewCompression(zcs);
+ ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
}
break;
}
@@ -3138,10 +4126,10 @@
case zcss_flush:
DEBUGLOG(5, "flush stage");
{ size_t const toFlush = zcs->outBuffContentSize - zcs->outBuffFlushedSize;
- size_t const flushed = ZSTD_limitCopy(op, oend-op,
+ size_t const flushed = ZSTD_limitCopy(op, (size_t)(oend-op),
zcs->outBuff + zcs->outBuffFlushedSize, toFlush);
DEBUGLOG(5, "toFlush: %u into %u ==> flushed: %u",
- (U32)toFlush, (U32)(oend-op), (U32)flushed);
+ (unsigned)toFlush, (unsigned)(oend-op), (unsigned)flushed);
op += flushed;
zcs->outBuffFlushedSize += flushed;
if (toFlush!=flushed) {
@@ -3154,7 +4142,7 @@
if (zcs->frameEnded) {
DEBUGLOG(5, "Frame completed on flush");
someMoreWork = 0;
- ZSTD_startNewCompression(zcs);
+ ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
break;
}
zcs->streamStage = zcss_load;
@@ -3169,63 +4157,67 @@
input->pos = ip - istart;
output->pos = op - ostart;
if (zcs->frameEnded) return 0;
- { size_t hintInSize = zcs->inBuffTarget - zcs->inBuffPos;
- if (hintInSize==0) hintInSize = zcs->blockSize;
- return hintInSize;
+ return ZSTD_nextInputSizeHint(zcs);
+}
+
+static size_t ZSTD_nextInputSizeHint_MTorST(const ZSTD_CCtx* cctx)
+{
+#ifdef ZSTD_MULTITHREAD
+ if (cctx->appliedParams.nbWorkers >= 1) {
+ assert(cctx->mtctx != NULL);
+ return ZSTDMT_nextInputSizeHint(cctx->mtctx);
}
+#endif
+ return ZSTD_nextInputSizeHint(cctx);
+
}
size_t ZSTD_compressStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output, ZSTD_inBuffer* input)
{
- /* check conditions */
- if (output->pos > output->size) return ERROR(GENERIC);
- if (input->pos > input->size) return ERROR(GENERIC);
-
- return ZSTD_compressStream_generic(zcs, output, input, ZSTD_e_continue);
+ FORWARD_IF_ERROR( ZSTD_compressStream2(zcs, output, input, ZSTD_e_continue) );
+ return ZSTD_nextInputSizeHint_MTorST(zcs);
}
-size_t ZSTD_compress_generic (ZSTD_CCtx* cctx,
- ZSTD_outBuffer* output,
- ZSTD_inBuffer* input,
- ZSTD_EndDirective endOp)
+size_t ZSTD_compressStream2( ZSTD_CCtx* cctx,
+ ZSTD_outBuffer* output,
+ ZSTD_inBuffer* input,
+ ZSTD_EndDirective endOp)
{
- DEBUGLOG(5, "ZSTD_compress_generic, endOp=%u ", (U32)endOp);
+ DEBUGLOG(5, "ZSTD_compressStream2, endOp=%u ", (unsigned)endOp);
/* check conditions */
- if (output->pos > output->size) return ERROR(GENERIC);
- if (input->pos > input->size) return ERROR(GENERIC);
+ RETURN_ERROR_IF(output->pos > output->size, GENERIC);
+ RETURN_ERROR_IF(input->pos > input->size, GENERIC);
assert(cctx!=NULL);
/* transparent initialization stage */
if (cctx->streamStage == zcss_init) {
ZSTD_CCtx_params params = cctx->requestedParams;
ZSTD_prefixDict const prefixDict = cctx->prefixDict;
- memset(&cctx->prefixDict, 0, sizeof(cctx->prefixDict)); /* single usage */
- assert(prefixDict.dict==NULL || cctx->cdict==NULL); /* only one can be set */
- DEBUGLOG(4, "ZSTD_compress_generic : transparent init stage");
+ FORWARD_IF_ERROR( ZSTD_initLocalDict(cctx) ); /* Init the local dict if present. */
+ memset(&cctx->prefixDict, 0, sizeof(cctx->prefixDict)); /* single usage */
+ assert(prefixDict.dict==NULL || cctx->cdict==NULL); /* only one can be set */
+ DEBUGLOG(4, "ZSTD_compressStream2 : transparent init stage");
if (endOp == ZSTD_e_end) cctx->pledgedSrcSizePlusOne = input->size + 1; /* auto-fix pledgedSrcSize */
params.cParams = ZSTD_getCParamsFromCCtxParams(
&cctx->requestedParams, cctx->pledgedSrcSizePlusOne-1, 0 /*dictSize*/);
+
#ifdef ZSTD_MULTITHREAD
if ((cctx->pledgedSrcSizePlusOne-1) <= ZSTDMT_JOBSIZE_MIN) {
params.nbWorkers = 0; /* do not invoke multi-threading when src size is too small */
}
if (params.nbWorkers > 0) {
/* mt context creation */
- if (cctx->mtctx == NULL || (params.nbWorkers != ZSTDMT_getNbWorkers(cctx->mtctx))) {
- DEBUGLOG(4, "ZSTD_compress_generic: creating new mtctx for nbWorkers=%u",
+ if (cctx->mtctx == NULL) {
+ DEBUGLOG(4, "ZSTD_compressStream2: creating new mtctx for nbWorkers=%u",
params.nbWorkers);
- if (cctx->mtctx != NULL)
- DEBUGLOG(4, "ZSTD_compress_generic: previous nbWorkers was %u",
- ZSTDMT_getNbWorkers(cctx->mtctx));
- ZSTDMT_freeCCtx(cctx->mtctx);
cctx->mtctx = ZSTDMT_createCCtx_advanced(params.nbWorkers, cctx->customMem);
- if (cctx->mtctx == NULL) return ERROR(memory_allocation);
+ RETURN_ERROR_IF(cctx->mtctx == NULL, memory_allocation);
}
/* mt compression */
DEBUGLOG(4, "call ZSTDMT_initCStream_internal as nbWorkers=%u", params.nbWorkers);
- CHECK_F( ZSTDMT_initCStream_internal(
+ FORWARD_IF_ERROR( ZSTDMT_initCStream_internal(
cctx->mtctx,
prefixDict.dict, prefixDict.dictSize, ZSTD_dct_rawContent,
cctx->cdict, params, cctx->pledgedSrcSizePlusOne-1) );
@@ -3233,35 +4225,47 @@
cctx->appliedParams.nbWorkers = params.nbWorkers;
} else
#endif
- { CHECK_F( ZSTD_resetCStream_internal(cctx,
+ { FORWARD_IF_ERROR( ZSTD_resetCStream_internal(cctx,
prefixDict.dict, prefixDict.dictSize, prefixDict.dictContentType,
cctx->cdict,
params, cctx->pledgedSrcSizePlusOne-1) );
assert(cctx->streamStage == zcss_load);
assert(cctx->appliedParams.nbWorkers == 0);
} }
+ /* end of transparent initialization stage */
/* compression stage */
#ifdef ZSTD_MULTITHREAD
if (cctx->appliedParams.nbWorkers > 0) {
+ int const forceMaxProgress = (endOp == ZSTD_e_flush || endOp == ZSTD_e_end);
+ size_t flushMin;
+ assert(forceMaxProgress || endOp == ZSTD_e_continue /* Protection for a new flush type */);
if (cctx->cParamsChanged) {
ZSTDMT_updateCParams_whileCompressing(cctx->mtctx, &cctx->requestedParams);
cctx->cParamsChanged = 0;
}
- { size_t const flushMin = ZSTDMT_compressStream_generic(cctx->mtctx, output, input, endOp);
+ do {
+ flushMin = ZSTDMT_compressStream_generic(cctx->mtctx, output, input, endOp);
if ( ZSTD_isError(flushMin)
|| (endOp == ZSTD_e_end && flushMin == 0) ) { /* compression completed */
- ZSTD_startNewCompression(cctx);
+ ZSTD_CCtx_reset(cctx, ZSTD_reset_session_only);
}
- return flushMin;
- } }
+ FORWARD_IF_ERROR(flushMin);
+ } while (forceMaxProgress && flushMin != 0 && output->pos < output->size);
+ DEBUGLOG(5, "completed ZSTD_compressStream2 delegating to ZSTDMT_compressStream_generic");
+ /* Either we don't require maximum forward progress, we've finished the
+ * flush, or we are out of output space.
+ */
+ assert(!forceMaxProgress || flushMin == 0 || output->pos == output->size);
+ return flushMin;
+ }
#endif
- CHECK_F( ZSTD_compressStream_generic(cctx, output, input, endOp) );
- DEBUGLOG(5, "completed ZSTD_compress_generic");
+ FORWARD_IF_ERROR( ZSTD_compressStream_generic(cctx, output, input, endOp) );
+ DEBUGLOG(5, "completed ZSTD_compressStream2");
return cctx->outBuffContentSize - cctx->outBuffFlushedSize; /* remaining to flush */
}
-size_t ZSTD_compress_generic_simpleArgs (
+size_t ZSTD_compressStream2_simpleArgs (
ZSTD_CCtx* cctx,
void* dst, size_t dstCapacity, size_t* dstPos,
const void* src, size_t srcSize, size_t* srcPos,
@@ -3269,13 +4273,33 @@
{
ZSTD_outBuffer output = { dst, dstCapacity, *dstPos };
ZSTD_inBuffer input = { src, srcSize, *srcPos };
- /* ZSTD_compress_generic() will check validity of dstPos and srcPos */
- size_t const cErr = ZSTD_compress_generic(cctx, &output, &input, endOp);
+ /* ZSTD_compressStream2() will check validity of dstPos and srcPos */
+ size_t const cErr = ZSTD_compressStream2(cctx, &output, &input, endOp);
*dstPos = output.pos;
*srcPos = input.pos;
return cErr;
}
+size_t ZSTD_compress2(ZSTD_CCtx* cctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize)
+{
+ ZSTD_CCtx_reset(cctx, ZSTD_reset_session_only);
+ { size_t oPos = 0;
+ size_t iPos = 0;
+ size_t const result = ZSTD_compressStream2_simpleArgs(cctx,
+ dst, dstCapacity, &oPos,
+ src, srcSize, &iPos,
+ ZSTD_e_end);
+ FORWARD_IF_ERROR(result);
+ if (result != 0) { /* compression not completed, due to lack of output space */
+ assert(oPos == dstCapacity);
+ RETURN_ERROR(dstSize_tooSmall);
+ }
+ assert(iPos == srcSize); /* all input is expected consumed */
+ return oPos;
+ }
+}
/*====== Finalize ======*/
@@ -3284,21 +4308,21 @@
size_t ZSTD_flushStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output)
{
ZSTD_inBuffer input = { NULL, 0, 0 };
- if (output->pos > output->size) return ERROR(GENERIC);
- CHECK_F( ZSTD_compressStream_generic(zcs, output, &input, ZSTD_e_flush) );
- return zcs->outBuffContentSize - zcs->outBuffFlushedSize; /* remaining to flush */
+ return ZSTD_compressStream2(zcs, output, &input, ZSTD_e_flush);
}
size_t ZSTD_endStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output)
{
ZSTD_inBuffer input = { NULL, 0, 0 };
- if (output->pos > output->size) return ERROR(GENERIC);
- CHECK_F( ZSTD_compressStream_generic(zcs, output, &input, ZSTD_e_end) );
+ size_t const remainingToFlush = ZSTD_compressStream2(zcs, output, &input, ZSTD_e_end);
+ FORWARD_IF_ERROR( remainingToFlush );
+ if (zcs->appliedParams.nbWorkers > 0) return remainingToFlush; /* minimal estimation */
+ /* single thread mode : attempt to calculate remaining to flush more precisely */
{ size_t const lastBlockSize = zcs->frameEnded ? 0 : ZSTD_BLOCKHEADERSIZE;
- size_t const checksumSize = zcs->frameEnded ? 0 : zcs->appliedParams.fParams.checksumFlag * 4;
- size_t const toFlush = zcs->outBuffContentSize - zcs->outBuffFlushedSize + lastBlockSize + checksumSize;
- DEBUGLOG(4, "ZSTD_endStream : remaining to flush : %u", (U32)toFlush);
+ size_t const checksumSize = (size_t)(zcs->frameEnded ? 0 : zcs->appliedParams.fParams.checksumFlag * 4);
+ size_t const toFlush = remainingToFlush + lastBlockSize + checksumSize;
+ DEBUGLOG(4, "ZSTD_endStream : remaining to flush : %u", (unsigned)toFlush);
return toFlush;
}
}
@@ -3308,122 +4332,123 @@
#define ZSTD_MAX_CLEVEL 22
int ZSTD_maxCLevel(void) { return ZSTD_MAX_CLEVEL; }
+int ZSTD_minCLevel(void) { return (int)-ZSTD_TARGETLENGTH_MAX; }
static const ZSTD_compressionParameters ZSTD_defaultCParameters[4][ZSTD_MAX_CLEVEL+1] = {
-{ /* "default" - guarantees a monotonically increasing memory budget */
+{ /* "default" - for any srcSize > 256 KB */
/* W, C, H, S, L, TL, strat */
{ 19, 12, 13, 1, 6, 1, ZSTD_fast }, /* base for negative levels */
- { 19, 13, 14, 1, 7, 1, ZSTD_fast }, /* level 1 */
- { 19, 15, 16, 1, 6, 1, ZSTD_fast }, /* level 2 */
- { 20, 16, 17, 1, 5, 8, ZSTD_dfast }, /* level 3 */
- { 20, 17, 18, 1, 5, 8, ZSTD_dfast }, /* level 4 */
- { 20, 17, 18, 2, 5, 16, ZSTD_greedy }, /* level 5 */
- { 21, 17, 19, 2, 5, 16, ZSTD_lazy }, /* level 6 */
- { 21, 18, 19, 3, 5, 16, ZSTD_lazy }, /* level 7 */
- { 21, 18, 20, 3, 5, 16, ZSTD_lazy2 }, /* level 8 */
- { 21, 19, 20, 3, 5, 16, ZSTD_lazy2 }, /* level 9 */
- { 21, 19, 21, 4, 5, 16, ZSTD_lazy2 }, /* level 10 */
- { 22, 20, 22, 4, 5, 16, ZSTD_lazy2 }, /* level 11 */
- { 22, 20, 22, 5, 5, 16, ZSTD_lazy2 }, /* level 12 */
- { 22, 21, 22, 4, 5, 32, ZSTD_btlazy2 }, /* level 13 */
- { 22, 21, 22, 5, 5, 32, ZSTD_btlazy2 }, /* level 14 */
- { 22, 22, 22, 6, 5, 32, ZSTD_btlazy2 }, /* level 15 */
- { 22, 21, 22, 4, 5, 48, ZSTD_btopt }, /* level 16 */
- { 23, 22, 22, 4, 4, 48, ZSTD_btopt }, /* level 17 */
- { 23, 22, 22, 5, 3, 64, ZSTD_btopt }, /* level 18 */
- { 23, 23, 22, 7, 3,128, ZSTD_btopt }, /* level 19 */
- { 25, 25, 23, 7, 3,128, ZSTD_btultra }, /* level 20 */
- { 26, 26, 24, 7, 3,256, ZSTD_btultra }, /* level 21 */
- { 27, 27, 25, 9, 3,512, ZSTD_btultra }, /* level 22 */
+ { 19, 13, 14, 1, 7, 0, ZSTD_fast }, /* level 1 */
+ { 20, 15, 16, 1, 6, 0, ZSTD_fast }, /* level 2 */
+ { 21, 16, 17, 1, 5, 1, ZSTD_dfast }, /* level 3 */
+ { 21, 18, 18, 1, 5, 1, ZSTD_dfast }, /* level 4 */
+ { 21, 18, 19, 2, 5, 2, ZSTD_greedy }, /* level 5 */
+ { 21, 19, 19, 3, 5, 4, ZSTD_greedy }, /* level 6 */
+ { 21, 19, 19, 3, 5, 8, ZSTD_lazy }, /* level 7 */
+ { 21, 19, 19, 3, 5, 16, ZSTD_lazy2 }, /* level 8 */
+ { 21, 19, 20, 4, 5, 16, ZSTD_lazy2 }, /* level 9 */
+ { 22, 20, 21, 4, 5, 16, ZSTD_lazy2 }, /* level 10 */
+ { 22, 21, 22, 4, 5, 16, ZSTD_lazy2 }, /* level 11 */
+ { 22, 21, 22, 5, 5, 16, ZSTD_lazy2 }, /* level 12 */
+ { 22, 21, 22, 5, 5, 32, ZSTD_btlazy2 }, /* level 13 */
+ { 22, 22, 23, 5, 5, 32, ZSTD_btlazy2 }, /* level 14 */
+ { 22, 23, 23, 6, 5, 32, ZSTD_btlazy2 }, /* level 15 */
+ { 22, 22, 22, 5, 5, 48, ZSTD_btopt }, /* level 16 */
+ { 23, 23, 22, 5, 4, 64, ZSTD_btopt }, /* level 17 */
+ { 23, 23, 22, 6, 3, 64, ZSTD_btultra }, /* level 18 */
+ { 23, 24, 22, 7, 3,256, ZSTD_btultra2}, /* level 19 */
+ { 25, 25, 23, 7, 3,256, ZSTD_btultra2}, /* level 20 */
+ { 26, 26, 24, 7, 3,512, ZSTD_btultra2}, /* level 21 */
+ { 27, 27, 25, 9, 3,999, ZSTD_btultra2}, /* level 22 */
},
{ /* for srcSize <= 256 KB */
/* W, C, H, S, L, T, strat */
{ 18, 12, 13, 1, 5, 1, ZSTD_fast }, /* base for negative levels */
- { 18, 13, 14, 1, 6, 1, ZSTD_fast }, /* level 1 */
- { 18, 14, 13, 1, 5, 8, ZSTD_dfast }, /* level 2 */
- { 18, 16, 15, 1, 5, 8, ZSTD_dfast }, /* level 3 */
- { 18, 15, 17, 1, 5, 8, ZSTD_greedy }, /* level 4.*/
- { 18, 16, 17, 4, 5, 8, ZSTD_greedy }, /* level 5.*/
- { 18, 16, 17, 3, 5, 8, ZSTD_lazy }, /* level 6.*/
- { 18, 17, 17, 4, 4, 8, ZSTD_lazy }, /* level 7 */
- { 18, 17, 17, 4, 4, 8, ZSTD_lazy2 }, /* level 8 */
- { 18, 17, 17, 5, 4, 8, ZSTD_lazy2 }, /* level 9 */
- { 18, 17, 17, 6, 4, 8, ZSTD_lazy2 }, /* level 10 */
- { 18, 18, 17, 6, 4, 8, ZSTD_lazy2 }, /* level 11.*/
- { 18, 18, 17, 5, 4, 8, ZSTD_btlazy2 }, /* level 12.*/
- { 18, 19, 17, 7, 4, 8, ZSTD_btlazy2 }, /* level 13 */
- { 18, 18, 18, 4, 4, 16, ZSTD_btopt }, /* level 14.*/
- { 18, 18, 18, 4, 3, 16, ZSTD_btopt }, /* level 15.*/
- { 18, 19, 18, 6, 3, 32, ZSTD_btopt }, /* level 16.*/
- { 18, 19, 18, 8, 3, 64, ZSTD_btopt }, /* level 17.*/
- { 18, 19, 18, 9, 3,128, ZSTD_btopt }, /* level 18.*/
- { 18, 19, 18, 10, 3,256, ZSTD_btopt }, /* level 19.*/
- { 18, 19, 18, 11, 3,512, ZSTD_btultra }, /* level 20.*/
- { 18, 19, 18, 12, 3,512, ZSTD_btultra }, /* level 21.*/
- { 18, 19, 18, 13, 3,512, ZSTD_btultra }, /* level 22.*/
+ { 18, 13, 14, 1, 6, 0, ZSTD_fast }, /* level 1 */
+ { 18, 14, 14, 1, 5, 1, ZSTD_dfast }, /* level 2 */
+ { 18, 16, 16, 1, 4, 1, ZSTD_dfast }, /* level 3 */
+ { 18, 16, 17, 2, 5, 2, ZSTD_greedy }, /* level 4.*/
+ { 18, 18, 18, 3, 5, 2, ZSTD_greedy }, /* level 5.*/
+ { 18, 18, 19, 3, 5, 4, ZSTD_lazy }, /* level 6.*/
+ { 18, 18, 19, 4, 4, 4, ZSTD_lazy }, /* level 7 */
+ { 18, 18, 19, 4, 4, 8, ZSTD_lazy2 }, /* level 8 */
+ { 18, 18, 19, 5, 4, 8, ZSTD_lazy2 }, /* level 9 */
+ { 18, 18, 19, 6, 4, 8, ZSTD_lazy2 }, /* level 10 */
+ { 18, 18, 19, 5, 4, 12, ZSTD_btlazy2 }, /* level 11.*/
+ { 18, 19, 19, 7, 4, 12, ZSTD_btlazy2 }, /* level 12.*/
+ { 18, 18, 19, 4, 4, 16, ZSTD_btopt }, /* level 13 */
+ { 18, 18, 19, 4, 3, 32, ZSTD_btopt }, /* level 14.*/
+ { 18, 18, 19, 6, 3,128, ZSTD_btopt }, /* level 15.*/
+ { 18, 19, 19, 6, 3,128, ZSTD_btultra }, /* level 16.*/
+ { 18, 19, 19, 8, 3,256, ZSTD_btultra }, /* level 17.*/
+ { 18, 19, 19, 6, 3,128, ZSTD_btultra2}, /* level 18.*/
+ { 18, 19, 19, 8, 3,256, ZSTD_btultra2}, /* level 19.*/
+ { 18, 19, 19, 10, 3,512, ZSTD_btultra2}, /* level 20.*/
+ { 18, 19, 19, 12, 3,512, ZSTD_btultra2}, /* level 21.*/
+ { 18, 19, 19, 13, 3,999, ZSTD_btultra2}, /* level 22.*/
},
{ /* for srcSize <= 128 KB */
/* W, C, H, S, L, T, strat */
- { 17, 12, 12, 1, 5, 1, ZSTD_fast }, /* level 0 - not used */
- { 17, 12, 13, 1, 6, 1, ZSTD_fast }, /* level 1 */
- { 17, 13, 16, 1, 5, 1, ZSTD_fast }, /* level 2 */
- { 17, 16, 16, 2, 5, 8, ZSTD_dfast }, /* level 3 */
- { 17, 13, 15, 3, 4, 8, ZSTD_greedy }, /* level 4 */
- { 17, 15, 17, 4, 4, 8, ZSTD_greedy }, /* level 5 */
- { 17, 16, 17, 3, 4, 8, ZSTD_lazy }, /* level 6 */
- { 17, 15, 17, 4, 4, 8, ZSTD_lazy2 }, /* level 7 */
+ { 17, 12, 12, 1, 5, 1, ZSTD_fast }, /* base for negative levels */
+ { 17, 12, 13, 1, 6, 0, ZSTD_fast }, /* level 1 */
+ { 17, 13, 15, 1, 5, 0, ZSTD_fast }, /* level 2 */
+ { 17, 15, 16, 2, 5, 1, ZSTD_dfast }, /* level 3 */
+ { 17, 17, 17, 2, 4, 1, ZSTD_dfast }, /* level 4 */
+ { 17, 16, 17, 3, 4, 2, ZSTD_greedy }, /* level 5 */
+ { 17, 17, 17, 3, 4, 4, ZSTD_lazy }, /* level 6 */
+ { 17, 17, 17, 3, 4, 8, ZSTD_lazy2 }, /* level 7 */
{ 17, 17, 17, 4, 4, 8, ZSTD_lazy2 }, /* level 8 */
{ 17, 17, 17, 5, 4, 8, ZSTD_lazy2 }, /* level 9 */
{ 17, 17, 17, 6, 4, 8, ZSTD_lazy2 }, /* level 10 */
- { 17, 17, 17, 7, 4, 8, ZSTD_lazy2 }, /* level 11 */
- { 17, 17, 17, 8, 4, 8, ZSTD_lazy2 }, /* level 12 */
- { 17, 18, 17, 6, 4, 8, ZSTD_btlazy2 }, /* level 13.*/
- { 17, 17, 17, 7, 3, 8, ZSTD_btopt }, /* level 14.*/
- { 17, 17, 17, 7, 3, 16, ZSTD_btopt }, /* level 15.*/
- { 17, 18, 17, 7, 3, 32, ZSTD_btopt }, /* level 16.*/
- { 17, 18, 17, 7, 3, 64, ZSTD_btopt }, /* level 17.*/
- { 17, 18, 17, 7, 3,256, ZSTD_btopt }, /* level 18.*/
- { 17, 18, 17, 8, 3,256, ZSTD_btopt }, /* level 19.*/
- { 17, 18, 17, 9, 3,256, ZSTD_btultra }, /* level 20.*/
- { 17, 18, 17, 10, 3,256, ZSTD_btultra }, /* level 21.*/
- { 17, 18, 17, 11, 3,512, ZSTD_btultra }, /* level 22.*/
+ { 17, 17, 17, 5, 4, 8, ZSTD_btlazy2 }, /* level 11 */
+ { 17, 18, 17, 7, 4, 12, ZSTD_btlazy2 }, /* level 12 */
+ { 17, 18, 17, 3, 4, 12, ZSTD_btopt }, /* level 13.*/
+ { 17, 18, 17, 4, 3, 32, ZSTD_btopt }, /* level 14.*/
+ { 17, 18, 17, 6, 3,256, ZSTD_btopt }, /* level 15.*/
+ { 17, 18, 17, 6, 3,128, ZSTD_btultra }, /* level 16.*/
+ { 17, 18, 17, 8, 3,256, ZSTD_btultra }, /* level 17.*/
+ { 17, 18, 17, 10, 3,512, ZSTD_btultra }, /* level 18.*/
+ { 17, 18, 17, 5, 3,256, ZSTD_btultra2}, /* level 19.*/
+ { 17, 18, 17, 7, 3,512, ZSTD_btultra2}, /* level 20.*/
+ { 17, 18, 17, 9, 3,512, ZSTD_btultra2}, /* level 21.*/
+ { 17, 18, 17, 11, 3,999, ZSTD_btultra2}, /* level 22.*/
},
{ /* for srcSize <= 16 KB */
/* W, C, H, S, L, T, strat */
{ 14, 12, 13, 1, 5, 1, ZSTD_fast }, /* base for negative levels */
- { 14, 14, 14, 1, 6, 1, ZSTD_fast }, /* level 1 */
- { 14, 14, 14, 1, 4, 1, ZSTD_fast }, /* level 2 */
- { 14, 14, 14, 1, 4, 6, ZSTD_dfast }, /* level 3.*/
- { 14, 14, 14, 4, 4, 6, ZSTD_greedy }, /* level 4.*/
- { 14, 14, 14, 3, 4, 6, ZSTD_lazy }, /* level 5.*/
- { 14, 14, 14, 4, 4, 6, ZSTD_lazy2 }, /* level 6 */
- { 14, 14, 14, 5, 4, 6, ZSTD_lazy2 }, /* level 7 */
- { 14, 14, 14, 6, 4, 6, ZSTD_lazy2 }, /* level 8.*/
- { 14, 15, 14, 6, 4, 6, ZSTD_btlazy2 }, /* level 9.*/
- { 14, 15, 14, 3, 3, 6, ZSTD_btopt }, /* level 10.*/
- { 14, 15, 14, 6, 3, 8, ZSTD_btopt }, /* level 11.*/
- { 14, 15, 14, 6, 3, 16, ZSTD_btopt }, /* level 12.*/
- { 14, 15, 14, 6, 3, 24, ZSTD_btopt }, /* level 13.*/
- { 14, 15, 15, 6, 3, 48, ZSTD_btopt }, /* level 14.*/
- { 14, 15, 15, 6, 3, 64, ZSTD_btopt }, /* level 15.*/
- { 14, 15, 15, 6, 3, 96, ZSTD_btopt }, /* level 16.*/
- { 14, 15, 15, 6, 3,128, ZSTD_btopt }, /* level 17.*/
- { 14, 15, 15, 6, 3,256, ZSTD_btopt }, /* level 18.*/
- { 14, 15, 15, 7, 3,256, ZSTD_btopt }, /* level 19.*/
- { 14, 15, 15, 8, 3,256, ZSTD_btultra }, /* level 20.*/
- { 14, 15, 15, 9, 3,256, ZSTD_btultra }, /* level 21.*/
- { 14, 15, 15, 10, 3,256, ZSTD_btultra }, /* level 22.*/
+ { 14, 14, 15, 1, 5, 0, ZSTD_fast }, /* level 1 */
+ { 14, 14, 15, 1, 4, 0, ZSTD_fast }, /* level 2 */
+ { 14, 14, 15, 2, 4, 1, ZSTD_dfast }, /* level 3 */
+ { 14, 14, 14, 4, 4, 2, ZSTD_greedy }, /* level 4 */
+ { 14, 14, 14, 3, 4, 4, ZSTD_lazy }, /* level 5.*/
+ { 14, 14, 14, 4, 4, 8, ZSTD_lazy2 }, /* level 6 */
+ { 14, 14, 14, 6, 4, 8, ZSTD_lazy2 }, /* level 7 */
+ { 14, 14, 14, 8, 4, 8, ZSTD_lazy2 }, /* level 8.*/
+ { 14, 15, 14, 5, 4, 8, ZSTD_btlazy2 }, /* level 9.*/
+ { 14, 15, 14, 9, 4, 8, ZSTD_btlazy2 }, /* level 10.*/
+ { 14, 15, 14, 3, 4, 12, ZSTD_btopt }, /* level 11.*/
+ { 14, 15, 14, 4, 3, 24, ZSTD_btopt }, /* level 12.*/
+ { 14, 15, 14, 5, 3, 32, ZSTD_btultra }, /* level 13.*/
+ { 14, 15, 15, 6, 3, 64, ZSTD_btultra }, /* level 14.*/
+ { 14, 15, 15, 7, 3,256, ZSTD_btultra }, /* level 15.*/
+ { 14, 15, 15, 5, 3, 48, ZSTD_btultra2}, /* level 16.*/
+ { 14, 15, 15, 6, 3,128, ZSTD_btultra2}, /* level 17.*/
+ { 14, 15, 15, 7, 3,256, ZSTD_btultra2}, /* level 18.*/
+ { 14, 15, 15, 8, 3,256, ZSTD_btultra2}, /* level 19.*/
+ { 14, 15, 15, 8, 3,512, ZSTD_btultra2}, /* level 20.*/
+ { 14, 15, 15, 9, 3,512, ZSTD_btultra2}, /* level 21.*/
+ { 14, 15, 15, 10, 3,999, ZSTD_btultra2}, /* level 22.*/
},
};
/*! ZSTD_getCParams() :
-* @return ZSTD_compressionParameters structure for a selected compression level, srcSize and dictSize.
-* Size values are optional, provide 0 if not known or unused */
+ * @return ZSTD_compressionParameters structure for a selected compression level, srcSize and dictSize.
+ * Size values are optional, provide 0 if not known or unused */
ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize)
{
size_t const addedSize = srcSizeHint ? 0 : 500;
- U64 const rSize = srcSizeHint+dictSize ? srcSizeHint+dictSize+addedSize : (U64)-1;
- U32 const tableID = (rSize <= 256 KB) + (rSize <= 128 KB) + (rSize <= 16 KB); /* intentional underflow for srcSizeHint == 0 */
+ U64 const rSize = srcSizeHint+dictSize ? srcSizeHint+dictSize+addedSize : ZSTD_CONTENTSIZE_UNKNOWN; /* intentional overflow for srcSizeHint == ZSTD_CONTENTSIZE_UNKNOWN */
+ U32 const tableID = (rSize <= 256 KB) + (rSize <= 128 KB) + (rSize <= 16 KB);
int row = compressionLevel;
DEBUGLOG(5, "ZSTD_getCParams (cLevel=%i)", compressionLevel);
if (compressionLevel == 0) row = ZSTD_CLEVEL_DEFAULT; /* 0 == default */
@@ -3431,13 +4456,14 @@
if (compressionLevel > ZSTD_MAX_CLEVEL) row = ZSTD_MAX_CLEVEL;
{ ZSTD_compressionParameters cp = ZSTD_defaultCParameters[tableID][row];
if (compressionLevel < 0) cp.targetLength = (unsigned)(-compressionLevel); /* acceleration factor */
- return ZSTD_adjustCParams_internal(cp, srcSizeHint, dictSize); }
-
+ return ZSTD_adjustCParams_internal(cp, srcSizeHint, dictSize); /* refine parameters based on srcSize & dictSize */
+ }
}
/*! ZSTD_getParams() :
-* same as ZSTD_getCParams(), but @return a `ZSTD_parameters` object (instead of `ZSTD_compressionParameters`).
-* All fields of `ZSTD_frameParameters` are set to default (0) */
+ * same idea as ZSTD_getCParams()
+ * @return a `ZSTD_parameters` structure (instead of `ZSTD_compressionParameters`).
+ * Fields of `ZSTD_frameParameters` are set to default values */
ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize) {
ZSTD_parameters params;
ZSTD_compressionParameters const cParams = ZSTD_getCParams(compressionLevel, srcSizeHint, dictSize);
diff --git a/vendor/github.com/DataDog/zstd/zstd_compress_internal.h b/vendor/github.com/DataDog/zstd/zstd_compress_internal.h
index 81f12ca..5495899 100644
--- a/vendor/github.com/DataDog/zstd/zstd_compress_internal.h
+++ b/vendor/github.com/DataDog/zstd/zstd_compress_internal.h
@@ -27,17 +27,19 @@
extern "C" {
#endif
+
/*-*************************************
* Constants
***************************************/
#define kSearchStrength 8
#define HASH_READ_SIZE 8
-#define ZSTD_DUBT_UNSORTED_MARK 1 /* For btlazy2 strategy, index 1 now means "unsorted".
+#define ZSTD_DUBT_UNSORTED_MARK 1 /* For btlazy2 strategy, index ZSTD_DUBT_UNSORTED_MARK==1 means "unsorted".
It could be confused for a real successor at index "1", if sorted as larger than its predecessor.
It's not a big deal though : candidate will just be sorted again.
- Additionnally, candidate position 1 will be lost.
+ Additionally, candidate position 1 will be lost.
But candidate 1 cannot hide a large tree of candidates, so it's a minimal loss.
- The benefit is that ZSTD_DUBT_UNSORTED_MARK cannot be misdhandled after table re-use with a different strategy */
+ The benefit is that ZSTD_DUBT_UNSORTED_MARK cannot be mishandled after table re-use with a different strategy.
+ This constant is required by ZSTD_compressBlock_btlazy2() and ZSTD_reduceTable_internal() */
/*-*************************************
@@ -53,14 +55,30 @@
} ZSTD_prefixDict;
typedef struct {
- U32 hufCTable[HUF_CTABLE_SIZE_U32(255)];
+ void* dictBuffer;
+ void const* dict;
+ size_t dictSize;
+ ZSTD_dictContentType_e dictContentType;
+ ZSTD_CDict* cdict;
+} ZSTD_localDict;
+
+typedef struct {
+ U32 CTable[HUF_CTABLE_SIZE_U32(255)];
+ HUF_repeat repeatMode;
+} ZSTD_hufCTables_t;
+
+typedef struct {
FSE_CTable offcodeCTable[FSE_CTABLE_SIZE_U32(OffFSELog, MaxOff)];
FSE_CTable matchlengthCTable[FSE_CTABLE_SIZE_U32(MLFSELog, MaxML)];
FSE_CTable litlengthCTable[FSE_CTABLE_SIZE_U32(LLFSELog, MaxLL)];
- HUF_repeat hufCTable_repeatMode;
FSE_repeat offcode_repeatMode;
FSE_repeat matchlength_repeatMode;
FSE_repeat litlength_repeatMode;
+} ZSTD_fseCTables_t;
+
+typedef struct {
+ ZSTD_hufCTables_t huf;
+ ZSTD_fseCTables_t fse;
} ZSTD_entropyCTables_t;
typedef struct {
@@ -76,26 +94,28 @@
U32 rep[ZSTD_REP_NUM];
} ZSTD_optimal_t;
+typedef enum { zop_dynamic=0, zop_predef } ZSTD_OptPrice_e;
+
typedef struct {
/* All tables are allocated inside cctx->workspace by ZSTD_resetCCtx_internal() */
- U32* litFreq; /* table of literals statistics, of size 256 */
- U32* litLengthFreq; /* table of litLength statistics, of size (MaxLL+1) */
- U32* matchLengthFreq; /* table of matchLength statistics, of size (MaxML+1) */
- U32* offCodeFreq; /* table of offCode statistics, of size (MaxOff+1) */
- ZSTD_match_t* matchTable; /* list of found matches, of size ZSTD_OPT_NUM+1 */
- ZSTD_optimal_t* priceTable; /* All positions tracked by optimal parser, of size ZSTD_OPT_NUM+1 */
+ unsigned* litFreq; /* table of literals statistics, of size 256 */
+ unsigned* litLengthFreq; /* table of litLength statistics, of size (MaxLL+1) */
+ unsigned* matchLengthFreq; /* table of matchLength statistics, of size (MaxML+1) */
+ unsigned* offCodeFreq; /* table of offCode statistics, of size (MaxOff+1) */
+ ZSTD_match_t* matchTable; /* list of found matches, of size ZSTD_OPT_NUM+1 */
+ ZSTD_optimal_t* priceTable; /* All positions tracked by optimal parser, of size ZSTD_OPT_NUM+1 */
U32 litSum; /* nb of literals */
U32 litLengthSum; /* nb of litLength codes */
U32 matchLengthSum; /* nb of matchLength codes */
U32 offCodeSum; /* nb of offset codes */
- /* begin updated by ZSTD_setLog2Prices */
- U32 log2litSum; /* pow2 to compare log2(litfreq) to */
- U32 log2litLengthSum; /* pow2 to compare log2(llfreq) to */
- U32 log2matchLengthSum; /* pow2 to compare log2(mlfreq) to */
- U32 log2offCodeSum; /* pow2 to compare log2(offreq) to */
- /* end : updated by ZSTD_setLog2Prices */
- U32 staticPrices; /* prices follow a pre-defined cost structure, statistics are irrelevant */
+ U32 litSumBasePrice; /* to compare to log2(litfreq) */
+ U32 litLengthSumBasePrice; /* to compare to log2(llfreq) */
+ U32 matchLengthSumBasePrice;/* to compare to log2(mlfreq) */
+ U32 offCodeSumBasePrice; /* to compare to log2(offreq) */
+ ZSTD_OptPrice_e priceType; /* prices can be determined dynamically, or follow a pre-defined cost structure */
+ const ZSTD_entropyCTables_t* symbolCosts; /* pre-calculated dictionary statistics */
+ ZSTD_literalCompressionMode_e literalCompressionMode;
} optState_t;
typedef struct {
@@ -108,20 +128,22 @@
BYTE const* base; /* All regular indexes relative to this position */
BYTE const* dictBase; /* extDict indexes relative to this position */
U32 dictLimit; /* below that point, need extDict */
- U32 lowLimit; /* below that point, no more data */
+ U32 lowLimit; /* below that point, no more valid data */
} ZSTD_window_t;
-typedef struct {
- ZSTD_window_t window; /* State for window round buffer management */
- U32 loadedDictEnd; /* index of end of dictionary */
- U32 nextToUpdate; /* index from which to continue table update */
- U32 nextToUpdate3; /* index from which to continue table update */
- U32 hashLog3; /* dispatch table : larger == faster, more memory */
+typedef struct ZSTD_matchState_t ZSTD_matchState_t;
+struct ZSTD_matchState_t {
+ ZSTD_window_t window; /* State for window round buffer management */
+ U32 loadedDictEnd; /* index of end of dictionary, within context's referential. When dict referential is copied into active context (i.e. not attached), effectively same value as dictSize, since referential starts from zero */
+ U32 nextToUpdate; /* index from which to continue table update */
+ U32 hashLog3; /* dispatch table : larger == faster, more memory */
U32* hashTable;
U32* hashTable3;
U32* chainTable;
optState_t opt; /* optimal parser state */
-} ZSTD_matchState_t;
+ const ZSTD_matchState_t* dictMatchState;
+ ZSTD_compressionParameters cParams;
+};
typedef struct {
ZSTD_compressedBlockState_t* prevCBlock;
@@ -147,7 +169,7 @@
U32 hashLog; /* Log size of hashTable */
U32 bucketSizeLog; /* Log bucket size for collision resolution, at most 8 */
U32 minMatchLength; /* Minimum match length */
- U32 hashEveryLog; /* Log number of entries to skip */
+ U32 hashRateLog; /* Log number of entries to skip */
U32 windowLog; /* Window log for the LDM */
} ldmParams_t;
@@ -161,7 +183,7 @@
rawSeq* seq; /* The start of the sequences */
size_t pos; /* The position where reading stopped. <= size. */
size_t size; /* The number of sequences. <= capacity. */
- size_t capacity; /* The capacity of the `seq` pointer */
+ size_t capacity; /* The capacity starting from `seq` pointer */
} rawSeqStore_t;
struct ZSTD_CCtx_params_s {
@@ -170,14 +192,20 @@
ZSTD_frameParameters fParams;
int compressionLevel;
- int disableLiteralCompression;
int forceWindow; /* force back-references to respect limit of
* 1<<wLog, even for dictionary */
+ size_t targetCBlockSize; /* Tries to fit compressed block size to be around targetCBlockSize.
+ * No target when targetCBlockSize == 0.
+ * There is no guarantee on compressed block size */
+
+ ZSTD_dictAttachPref_e attachDictPref;
+ ZSTD_literalCompressionMode_e literalCompressionMode;
/* Multithreading: used to pass parameters to mtctx */
- unsigned nbWorkers;
- unsigned jobSize;
- unsigned overlapSizeLog;
+ int nbWorkers;
+ size_t jobSize;
+ int overlapLog;
+ int rsyncable;
/* Long distance matching parameters */
ldmParams_t ldmParams;
@@ -193,6 +221,8 @@
ZSTD_CCtx_params requestedParams;
ZSTD_CCtx_params appliedParams;
U32 dictID;
+
+ int workSpaceOversizedDuration;
void* workSpace;
size_t workSpaceSize;
size_t blockSize;
@@ -225,7 +255,7 @@
U32 frameEnded;
/* Dictionary */
- ZSTD_CDict* cdictLocal;
+ ZSTD_localDict localDict;
const ZSTD_CDict* cdict;
ZSTD_prefixDict prefixDict; /* single-usage dictionary */
@@ -235,11 +265,15 @@
#endif
};
+typedef enum { ZSTD_dtlm_fast, ZSTD_dtlm_full } ZSTD_dictTableLoadMethod_e;
+
+typedef enum { ZSTD_noDict = 0, ZSTD_extDict = 1, ZSTD_dictMatchState = 2 } ZSTD_dictMode_e;
+
typedef size_t (*ZSTD_blockCompressor) (
ZSTD_matchState_t* bs, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
-ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, int extDict);
+ void const* src, size_t srcSize);
+ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, ZSTD_dictMode_e dictMode);
MEM_STATIC U32 ZSTD_LLcode(U32 litLength)
@@ -280,17 +314,19 @@
*/
MEM_STATIC void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const void* literals, U32 offsetCode, size_t mlBase)
{
-#if defined(ZSTD_DEBUG) && (ZSTD_DEBUG >= 6)
+#if defined(DEBUGLEVEL) && (DEBUGLEVEL >= 6)
static const BYTE* g_start = NULL;
if (g_start==NULL) g_start = (const BYTE*)literals; /* note : index only works for compression within a single segment */
{ U32 const pos = (U32)((const BYTE*)literals - g_start);
- DEBUGLOG(6, "Cpos%7u :%3u literals, match%3u bytes at dist.code%7u",
+ DEBUGLOG(6, "Cpos%7u :%3u literals, match%4u bytes at offCode%7u",
pos, (U32)litLength, (U32)mlBase+MINMATCH, (U32)offsetCode);
}
#endif
+ assert((size_t)(seqStorePtr->sequences - seqStorePtr->sequencesStart) < seqStorePtr->maxNbSeq);
/* copy Literals */
- assert(seqStorePtr->lit + litLength <= seqStorePtr->litStart + 128 KB);
- ZSTD_wildcopy(seqStorePtr->lit, literals, litLength);
+ assert(seqStorePtr->maxNbLit <= 128 KB);
+ assert(seqStorePtr->lit + litLength <= seqStorePtr->litStart + seqStorePtr->maxNbLit);
+ ZSTD_wildcopy(seqStorePtr->lit, literals, litLength, ZSTD_no_overlap);
seqStorePtr->lit += litLength;
/* literal Length */
@@ -420,6 +456,11 @@
const BYTE* const vEnd = MIN( ip + (mEnd - match), iEnd);
size_t const matchLength = ZSTD_count(ip, match, vEnd);
if (match + matchLength != mEnd) return matchLength;
+ DEBUGLOG(7, "ZSTD_count_2segments: found a 2-parts match (current length==%zu)", matchLength);
+ DEBUGLOG(7, "distance from match beginning to end dictionary = %zi", mEnd - match);
+ DEBUGLOG(7, "distance from current pos to end buffer = %zi", iEnd - ip);
+ DEBUGLOG(7, "next byte : ip==%02X, istart==%02X", ip[matchLength], *iStart);
+ DEBUGLOG(7, "final match length = %zu", matchLength + ZSTD_count(ip+matchLength, iStart, iEnd));
return matchLength + ZSTD_count(ip+matchLength, iStart, iEnd);
}
@@ -464,9 +505,70 @@
}
}
+/** ZSTD_ipow() :
+ * Return base^exponent.
+ */
+static U64 ZSTD_ipow(U64 base, U64 exponent)
+{
+ U64 power = 1;
+ while (exponent) {
+ if (exponent & 1) power *= base;
+ exponent >>= 1;
+ base *= base;
+ }
+ return power;
+}
+
+#define ZSTD_ROLL_HASH_CHAR_OFFSET 10
+
+/** ZSTD_rollingHash_append() :
+ * Add the buffer to the hash value.
+ */
+static U64 ZSTD_rollingHash_append(U64 hash, void const* buf, size_t size)
+{
+ BYTE const* istart = (BYTE const*)buf;
+ size_t pos;
+ for (pos = 0; pos < size; ++pos) {
+ hash *= prime8bytes;
+ hash += istart[pos] + ZSTD_ROLL_HASH_CHAR_OFFSET;
+ }
+ return hash;
+}
+
+/** ZSTD_rollingHash_compute() :
+ * Compute the rolling hash value of the buffer.
+ */
+MEM_STATIC U64 ZSTD_rollingHash_compute(void const* buf, size_t size)
+{
+ return ZSTD_rollingHash_append(0, buf, size);
+}
+
+/** ZSTD_rollingHash_primePower() :
+ * Compute the primePower to be passed to ZSTD_rollingHash_rotate() for a hash
+ * over a window of length bytes.
+ */
+MEM_STATIC U64 ZSTD_rollingHash_primePower(U32 length)
+{
+ return ZSTD_ipow(prime8bytes, length - 1);
+}
+
+/** ZSTD_rollingHash_rotate() :
+ * Rotate the rolling hash by one byte.
+ */
+MEM_STATIC U64 ZSTD_rollingHash_rotate(U64 hash, BYTE toRemove, BYTE toAdd, U64 primePower)
+{
+ hash -= (toRemove + ZSTD_ROLL_HASH_CHAR_OFFSET) * primePower;
+ hash *= prime8bytes;
+ hash += toAdd + ZSTD_ROLL_HASH_CHAR_OFFSET;
+ return hash;
+}
+
/*-*************************************
* Round buffer management
***************************************/
+#if (ZSTD_WINDOWLOG_MAX_64 > 31)
+# error "ZSTD_WINDOWLOG_MAX is too large : would overflow ZSTD_CURRENT_MAX"
+#endif
/* Max current allowed */
#define ZSTD_CURRENT_MAX ((3U << 29) + (1U << ZSTD_WINDOWLOG_MAX))
/* Maximum chunk size before overflow correction needs to be called again */
@@ -497,6 +599,20 @@
}
/**
+ * ZSTD_matchState_dictMode():
+ * Inspects the provided matchState and figures out what dictMode should be
+ * passed to the compressor.
+ */
+MEM_STATIC ZSTD_dictMode_e ZSTD_matchState_dictMode(const ZSTD_matchState_t *ms)
+{
+ return ZSTD_window_hasExtDict(ms->window) ?
+ ZSTD_extDict :
+ ms->dictMatchState != NULL ?
+ ZSTD_dictMatchState :
+ ZSTD_noDict;
+}
+
+/**
* ZSTD_window_needOverflowCorrection():
* Returns non-zero if the indices are getting too large and need overflow
* protection.
@@ -563,31 +679,83 @@
* ZSTD_window_enforceMaxDist():
* Updates lowLimit so that:
* (srcEnd - base) - lowLimit == maxDist + loadedDictEnd
- * This allows a simple check that index >= lowLimit to see if index is valid.
- * This must be called before a block compression call, with srcEnd as the block
- * source end.
- * If loadedDictEndPtr is not NULL, we set it to zero once we update lowLimit.
- * This is because dictionaries are allowed to be referenced as long as the last
- * byte of the dictionary is in the window, but once they are out of range,
- * they cannot be referenced. If loadedDictEndPtr is NULL, we use
- * loadedDictEnd == 0.
+ *
+ * It ensures index is valid as long as index >= lowLimit.
+ * This must be called before a block compression call.
+ *
+ * loadedDictEnd is only defined if a dictionary is in use for current compression.
+ * As the name implies, loadedDictEnd represents the index at end of dictionary.
+ * The value lies within context's referential, it can be directly compared to blockEndIdx.
+ *
+ * If loadedDictEndPtr is NULL, no dictionary is in use, and we use loadedDictEnd == 0.
+ * If loadedDictEndPtr is not NULL, we set it to zero after updating lowLimit.
+ * This is because dictionaries are allowed to be referenced fully
+ * as long as the last byte of the dictionary is in the window.
+ * Once input has progressed beyond window size, dictionary cannot be referenced anymore.
+ *
+ * In normal dict mode, the dictionary lies between lowLimit and dictLimit.
+ * In dictMatchState mode, lowLimit and dictLimit are the same,
+ * and the dictionary is below them.
+ * forceWindow and dictMatchState are therefore incompatible.
*/
-MEM_STATIC void ZSTD_window_enforceMaxDist(ZSTD_window_t* window,
- void const* srcEnd, U32 maxDist,
- U32* loadedDictEndPtr)
+MEM_STATIC void
+ZSTD_window_enforceMaxDist(ZSTD_window_t* window,
+ const void* blockEnd,
+ U32 maxDist,
+ U32* loadedDictEndPtr,
+ const ZSTD_matchState_t** dictMatchStatePtr)
{
- U32 const current = (U32)((BYTE const*)srcEnd - window->base);
- U32 loadedDictEnd = loadedDictEndPtr != NULL ? *loadedDictEndPtr : 0;
- if (current > maxDist + loadedDictEnd) {
- U32 const newLowLimit = current - maxDist;
+ U32 const blockEndIdx = (U32)((BYTE const*)blockEnd - window->base);
+ U32 const loadedDictEnd = (loadedDictEndPtr != NULL) ? *loadedDictEndPtr : 0;
+ DEBUGLOG(5, "ZSTD_window_enforceMaxDist: blockEndIdx=%u, maxDist=%u, loadedDictEnd=%u",
+ (unsigned)blockEndIdx, (unsigned)maxDist, (unsigned)loadedDictEnd);
+
+ /* - When there is no dictionary : loadedDictEnd == 0.
+ In which case, the test (blockEndIdx > maxDist) is merely to avoid
+ overflowing next operation `newLowLimit = blockEndIdx - maxDist`.
+ - When there is a standard dictionary :
+ Index referential is copied from the dictionary,
+ which means it starts from 0.
+ In which case, loadedDictEnd == dictSize,
+ and it makes sense to compare `blockEndIdx > maxDist + dictSize`
+ since `blockEndIdx` also starts from zero.
+ - When there is an attached dictionary :
+ loadedDictEnd is expressed within the referential of the context,
+ so it can be directly compared against blockEndIdx.
+ */
+ if (blockEndIdx > maxDist + loadedDictEnd) {
+ U32 const newLowLimit = blockEndIdx - maxDist;
if (window->lowLimit < newLowLimit) window->lowLimit = newLowLimit;
if (window->dictLimit < window->lowLimit) {
- DEBUGLOG(5, "Update dictLimit from %u to %u", window->dictLimit,
- window->lowLimit);
+ DEBUGLOG(5, "Update dictLimit to match lowLimit, from %u to %u",
+ (unsigned)window->dictLimit, (unsigned)window->lowLimit);
window->dictLimit = window->lowLimit;
}
- if (loadedDictEndPtr)
- *loadedDictEndPtr = 0;
+ /* On reaching window size, dictionaries are invalidated */
+ if (loadedDictEndPtr) *loadedDictEndPtr = 0;
+ if (dictMatchStatePtr) *dictMatchStatePtr = NULL;
+ }
+}
+
+/* Similar to ZSTD_window_enforceMaxDist(),
+ * but only invalidates dictionary
+ * when input progresses beyond window size. */
+MEM_STATIC void
+ZSTD_checkDictValidity(ZSTD_window_t* window,
+ const void* blockEnd,
+ U32 maxDist,
+ U32* loadedDictEndPtr,
+ const ZSTD_matchState_t** dictMatchStatePtr)
+{
+ U32 const blockEndIdx = (U32)((BYTE const*)blockEnd - window->base);
+ U32 const loadedDictEnd = (loadedDictEndPtr != NULL) ? *loadedDictEndPtr : 0;
+ DEBUGLOG(5, "ZSTD_checkDictValidity: blockEndIdx=%u, maxDist=%u, loadedDictEnd=%u",
+ (unsigned)blockEndIdx, (unsigned)maxDist, (unsigned)loadedDictEnd);
+
+ if (loadedDictEnd && (blockEndIdx > maxDist + loadedDictEnd)) {
+ /* On reaching window size, dictionaries are invalidated */
+ if (loadedDictEndPtr) *loadedDictEndPtr = 0;
+ if (dictMatchStatePtr) *dictMatchStatePtr = NULL;
}
}
@@ -603,12 +771,12 @@
{
BYTE const* const ip = (BYTE const*)src;
U32 contiguous = 1;
+ DEBUGLOG(5, "ZSTD_window_update");
/* Check if blocks follow each other */
if (src != window->nextSrc) {
/* not contiguous */
size_t const distanceFromBase = (size_t)(window->nextSrc - window->base);
- DEBUGLOG(5, "Non contiguous blocks, new segment starts at %u",
- window->dictLimit);
+ DEBUGLOG(5, "Non contiguous blocks, new segment starts at %u", window->dictLimit);
window->lowLimit = window->dictLimit;
assert(distanceFromBase == (size_t)(U32)distanceFromBase); /* should never overflow */
window->dictLimit = (U32)distanceFromBase;
@@ -625,10 +793,44 @@
ptrdiff_t const highInputIdx = (ip + srcSize) - window->dictBase;
U32 const lowLimitMax = (highInputIdx > (ptrdiff_t)window->dictLimit) ? window->dictLimit : (U32)highInputIdx;
window->lowLimit = lowLimitMax;
+ DEBUGLOG(5, "Overlapping extDict and input : new lowLimit = %u", window->lowLimit);
}
return contiguous;
}
+
+/* debug functions */
+#if (DEBUGLEVEL>=2)
+
+MEM_STATIC double ZSTD_fWeight(U32 rawStat)
+{
+ U32 const fp_accuracy = 8;
+ U32 const fp_multiplier = (1 << fp_accuracy);
+ U32 const newStat = rawStat + 1;
+ U32 const hb = ZSTD_highbit32(newStat);
+ U32 const BWeight = hb * fp_multiplier;
+ U32 const FWeight = (newStat << fp_accuracy) >> hb;
+ U32 const weight = BWeight + FWeight;
+ assert(hb + fp_accuracy < 31);
+ return (double)weight / fp_multiplier;
+}
+
+/* display a table content,
+ * listing each element, its frequency, and its predicted bit cost */
+MEM_STATIC void ZSTD_debugTable(const U32* table, U32 max)
+{
+ unsigned u, sum;
+ for (u=0, sum=0; u<=max; u++) sum += table[u];
+ DEBUGLOG(2, "total nb elts: %u", sum);
+ for (u=0; u<=max; u++) {
+ DEBUGLOG(2, "%2u: %5u (%.2f)",
+ u, table[u], ZSTD_fWeight(sum) - ZSTD_fWeight(table[u]) );
+ }
+}
+
+#endif
+
+
#if defined (__cplusplus)
}
#endif
@@ -640,7 +842,7 @@
* ============================================================== */
/* ZSTD_getCParamsFromCCtxParams() :
- * cParams are built depending on compressionLevel, src size hints,
+ * cParams are built depending on compressionLevel, src size hints,
* LDM and manually set compression parameters.
*/
ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams(
@@ -656,12 +858,7 @@
const ZSTD_CDict* cdict,
ZSTD_CCtx_params params, unsigned long long pledgedSrcSize);
-/*! ZSTD_compressStream_generic() :
- * Private use only. To be called from zstdmt_compress.c in single-thread mode. */
-size_t ZSTD_compressStream_generic(ZSTD_CStream* zcs,
- ZSTD_outBuffer* output,
- ZSTD_inBuffer* input,
- ZSTD_EndDirective const flushMode);
+void ZSTD_resetSeqStore(seqStore_t* ssPtr);
/*! ZSTD_getCParamsFromCDict() :
* as the name implies */
@@ -672,6 +869,7 @@
size_t ZSTD_compressBegin_advanced_internal(ZSTD_CCtx* cctx,
const void* dict, size_t dictSize,
ZSTD_dictContentType_e dictContentType,
+ ZSTD_dictTableLoadMethod_e dtlm,
const ZSTD_CDict* cdict,
ZSTD_CCtx_params params,
unsigned long long pledgedSrcSize);
@@ -688,7 +886,7 @@
/* ZSTD_writeLastEmptyBlock() :
* output an empty Block with end-of-frame mark to complete a frame
* @return : size of data written into `dst` (== ZSTD_blockHeaderSize (defined in zstd_internal.h))
- * or an error code if `dstCapcity` is too small (<ZSTD_blockHeaderSize)
+ * or an error code if `dstCapacity` is too small (<ZSTD_blockHeaderSize)
*/
size_t ZSTD_writeLastEmptyBlock(void* dst, size_t dstCapacity);
diff --git a/vendor/github.com/DataDog/zstd/zstd_ddict.c b/vendor/github.com/DataDog/zstd/zstd_ddict.c
new file mode 100644
index 0000000..0af3d23
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/zstd_ddict.c
@@ -0,0 +1,240 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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.
+ */
+
+/* zstd_ddict.c :
+ * concentrates all logic that needs to know the internals of ZSTD_DDict object */
+
+/*-*******************************************************
+* Dependencies
+*********************************************************/
+#include <string.h> /* memcpy, memmove, memset */
+#include "cpu.h" /* bmi2 */
+#include "mem.h" /* low level memory routines */
+#define FSE_STATIC_LINKING_ONLY
+#include "fse.h"
+#define HUF_STATIC_LINKING_ONLY
+#include "huf.h"
+#include "zstd_decompress_internal.h"
+#include "zstd_ddict.h"
+
+#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1)
+# include "zstd_legacy.h"
+#endif
+
+
+
+/*-*******************************************************
+* Types
+*********************************************************/
+struct ZSTD_DDict_s {
+ void* dictBuffer;
+ const void* dictContent;
+ size_t dictSize;
+ ZSTD_entropyDTables_t entropy;
+ U32 dictID;
+ U32 entropyPresent;
+ ZSTD_customMem cMem;
+}; /* typedef'd to ZSTD_DDict within "zstd.h" */
+
+const void* ZSTD_DDict_dictContent(const ZSTD_DDict* ddict)
+{
+ assert(ddict != NULL);
+ return ddict->dictContent;
+}
+
+size_t ZSTD_DDict_dictSize(const ZSTD_DDict* ddict)
+{
+ assert(ddict != NULL);
+ return ddict->dictSize;
+}
+
+void ZSTD_copyDDictParameters(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict)
+{
+ DEBUGLOG(4, "ZSTD_copyDDictParameters");
+ assert(dctx != NULL);
+ assert(ddict != NULL);
+ dctx->dictID = ddict->dictID;
+ dctx->prefixStart = ddict->dictContent;
+ dctx->virtualStart = ddict->dictContent;
+ dctx->dictEnd = (const BYTE*)ddict->dictContent + ddict->dictSize;
+ dctx->previousDstEnd = dctx->dictEnd;
+ if (ddict->entropyPresent) {
+ dctx->litEntropy = 1;
+ dctx->fseEntropy = 1;
+ dctx->LLTptr = ddict->entropy.LLTable;
+ dctx->MLTptr = ddict->entropy.MLTable;
+ dctx->OFTptr = ddict->entropy.OFTable;
+ dctx->HUFptr = ddict->entropy.hufTable;
+ dctx->entropy.rep[0] = ddict->entropy.rep[0];
+ dctx->entropy.rep[1] = ddict->entropy.rep[1];
+ dctx->entropy.rep[2] = ddict->entropy.rep[2];
+ } else {
+ dctx->litEntropy = 0;
+ dctx->fseEntropy = 0;
+ }
+}
+
+
+static size_t
+ZSTD_loadEntropy_intoDDict(ZSTD_DDict* ddict,
+ ZSTD_dictContentType_e dictContentType)
+{
+ ddict->dictID = 0;
+ ddict->entropyPresent = 0;
+ if (dictContentType == ZSTD_dct_rawContent) return 0;
+
+ if (ddict->dictSize < 8) {
+ if (dictContentType == ZSTD_dct_fullDict)
+ return ERROR(dictionary_corrupted); /* only accept specified dictionaries */
+ return 0; /* pure content mode */
+ }
+ { U32 const magic = MEM_readLE32(ddict->dictContent);
+ if (magic != ZSTD_MAGIC_DICTIONARY) {
+ if (dictContentType == ZSTD_dct_fullDict)
+ return ERROR(dictionary_corrupted); /* only accept specified dictionaries */
+ return 0; /* pure content mode */
+ }
+ }
+ ddict->dictID = MEM_readLE32((const char*)ddict->dictContent + ZSTD_FRAMEIDSIZE);
+
+ /* load entropy tables */
+ RETURN_ERROR_IF(ZSTD_isError(ZSTD_loadDEntropy(
+ &ddict->entropy, ddict->dictContent, ddict->dictSize)),
+ dictionary_corrupted);
+ ddict->entropyPresent = 1;
+ return 0;
+}
+
+
+static size_t ZSTD_initDDict_internal(ZSTD_DDict* ddict,
+ const void* dict, size_t dictSize,
+ ZSTD_dictLoadMethod_e dictLoadMethod,
+ ZSTD_dictContentType_e dictContentType)
+{
+ if ((dictLoadMethod == ZSTD_dlm_byRef) || (!dict) || (!dictSize)) {
+ ddict->dictBuffer = NULL;
+ ddict->dictContent = dict;
+ if (!dict) dictSize = 0;
+ } else {
+ void* const internalBuffer = ZSTD_malloc(dictSize, ddict->cMem);
+ ddict->dictBuffer = internalBuffer;
+ ddict->dictContent = internalBuffer;
+ if (!internalBuffer) return ERROR(memory_allocation);
+ memcpy(internalBuffer, dict, dictSize);
+ }
+ ddict->dictSize = dictSize;
+ ddict->entropy.hufTable[0] = (HUF_DTable)((HufLog)*0x1000001); /* cover both little and big endian */
+
+ /* parse dictionary content */
+ FORWARD_IF_ERROR( ZSTD_loadEntropy_intoDDict(ddict, dictContentType) );
+
+ return 0;
+}
+
+ZSTD_DDict* ZSTD_createDDict_advanced(const void* dict, size_t dictSize,
+ ZSTD_dictLoadMethod_e dictLoadMethod,
+ ZSTD_dictContentType_e dictContentType,
+ ZSTD_customMem customMem)
+{
+ if (!customMem.customAlloc ^ !customMem.customFree) return NULL;
+
+ { ZSTD_DDict* const ddict = (ZSTD_DDict*) ZSTD_malloc(sizeof(ZSTD_DDict), customMem);
+ if (ddict == NULL) return NULL;
+ ddict->cMem = customMem;
+ { size_t const initResult = ZSTD_initDDict_internal(ddict,
+ dict, dictSize,
+ dictLoadMethod, dictContentType);
+ if (ZSTD_isError(initResult)) {
+ ZSTD_freeDDict(ddict);
+ return NULL;
+ } }
+ return ddict;
+ }
+}
+
+/*! ZSTD_createDDict() :
+* Create a digested dictionary, to start decompression without startup delay.
+* `dict` content is copied inside DDict.
+* Consequently, `dict` can be released after `ZSTD_DDict` creation */
+ZSTD_DDict* ZSTD_createDDict(const void* dict, size_t dictSize)
+{
+ ZSTD_customMem const allocator = { NULL, NULL, NULL };
+ return ZSTD_createDDict_advanced(dict, dictSize, ZSTD_dlm_byCopy, ZSTD_dct_auto, allocator);
+}
+
+/*! ZSTD_createDDict_byReference() :
+ * Create a digested dictionary, to start decompression without startup delay.
+ * Dictionary content is simply referenced, it will be accessed during decompression.
+ * Warning : dictBuffer must outlive DDict (DDict must be freed before dictBuffer) */
+ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize)
+{
+ ZSTD_customMem const allocator = { NULL, NULL, NULL };
+ return ZSTD_createDDict_advanced(dictBuffer, dictSize, ZSTD_dlm_byRef, ZSTD_dct_auto, allocator);
+}
+
+
+const ZSTD_DDict* ZSTD_initStaticDDict(
+ void* sBuffer, size_t sBufferSize,
+ const void* dict, size_t dictSize,
+ ZSTD_dictLoadMethod_e dictLoadMethod,
+ ZSTD_dictContentType_e dictContentType)
+{
+ size_t const neededSpace = sizeof(ZSTD_DDict)
+ + (dictLoadMethod == ZSTD_dlm_byRef ? 0 : dictSize);
+ ZSTD_DDict* const ddict = (ZSTD_DDict*)sBuffer;
+ assert(sBuffer != NULL);
+ assert(dict != NULL);
+ if ((size_t)sBuffer & 7) return NULL; /* 8-aligned */
+ if (sBufferSize < neededSpace) return NULL;
+ if (dictLoadMethod == ZSTD_dlm_byCopy) {
+ memcpy(ddict+1, dict, dictSize); /* local copy */
+ dict = ddict+1;
+ }
+ if (ZSTD_isError( ZSTD_initDDict_internal(ddict,
+ dict, dictSize,
+ ZSTD_dlm_byRef, dictContentType) ))
+ return NULL;
+ return ddict;
+}
+
+
+size_t ZSTD_freeDDict(ZSTD_DDict* ddict)
+{
+ if (ddict==NULL) return 0; /* support free on NULL */
+ { ZSTD_customMem const cMem = ddict->cMem;
+ ZSTD_free(ddict->dictBuffer, cMem);
+ ZSTD_free(ddict, cMem);
+ return 0;
+ }
+}
+
+/*! ZSTD_estimateDDictSize() :
+ * Estimate amount of memory that will be needed to create a dictionary for decompression.
+ * Note : dictionary created by reference using ZSTD_dlm_byRef are smaller */
+size_t ZSTD_estimateDDictSize(size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod)
+{
+ return sizeof(ZSTD_DDict) + (dictLoadMethod == ZSTD_dlm_byRef ? 0 : dictSize);
+}
+
+size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict)
+{
+ if (ddict==NULL) return 0; /* support sizeof on NULL */
+ return sizeof(*ddict) + (ddict->dictBuffer ? ddict->dictSize : 0) ;
+}
+
+/*! ZSTD_getDictID_fromDDict() :
+ * Provides the dictID of the dictionary loaded into `ddict`.
+ * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty.
+ * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */
+unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict)
+{
+ if (ddict==NULL) return 0;
+ return ZSTD_getDictID_fromDict(ddict->dictContent, ddict->dictSize);
+}
diff --git a/vendor/github.com/DataDog/zstd/zstd_ddict.h b/vendor/github.com/DataDog/zstd/zstd_ddict.h
new file mode 100644
index 0000000..0479d11
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/zstd_ddict.h
@@ -0,0 +1,44 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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_DDICT_H
+#define ZSTD_DDICT_H
+
+/*-*******************************************************
+ * Dependencies
+ *********************************************************/
+#include <stddef.h> /* size_t */
+#include "zstd.h" /* ZSTD_DDict, and several public functions */
+
+
+/*-*******************************************************
+ * Interface
+ *********************************************************/
+
+/* note: several prototypes are already published in `zstd.h` :
+ * ZSTD_createDDict()
+ * ZSTD_createDDict_byReference()
+ * ZSTD_createDDict_advanced()
+ * ZSTD_freeDDict()
+ * ZSTD_initStaticDDict()
+ * ZSTD_sizeof_DDict()
+ * ZSTD_estimateDDictSize()
+ * ZSTD_getDictID_fromDict()
+ */
+
+const void* ZSTD_DDict_dictContent(const ZSTD_DDict* ddict);
+size_t ZSTD_DDict_dictSize(const ZSTD_DDict* ddict);
+
+void ZSTD_copyDDictParameters(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict);
+
+
+
+#endif /* ZSTD_DDICT_H */
diff --git a/vendor/github.com/DataDog/zstd/zstd_decompress.c b/vendor/github.com/DataDog/zstd/zstd_decompress.c
index 3ec6a1c..e42872a 100644
--- a/vendor/github.com/DataDog/zstd/zstd_decompress.c
+++ b/vendor/github.com/DataDog/zstd/zstd_decompress.c
@@ -37,7 +37,18 @@
* It's possible to set a different limit using ZSTD_DCtx_setMaxWindowSize().
*/
#ifndef ZSTD_MAXWINDOWSIZE_DEFAULT
-# define ZSTD_MAXWINDOWSIZE_DEFAULT (((U32)1 << ZSTD_WINDOWLOG_DEFAULTMAX) + 1)
+# define ZSTD_MAXWINDOWSIZE_DEFAULT (((U32)1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT) + 1)
+#endif
+
+/*!
+ * NO_FORWARD_PROGRESS_MAX :
+ * maximum allowed nb of calls to ZSTD_decompressStream()
+ * without any forward progress
+ * (defined as: no byte read from input, and no byte flushed to output)
+ * before triggering an error.
+ */
+#ifndef ZSTD_NO_FORWARD_PROGRESS_MAX
+# define ZSTD_NO_FORWARD_PROGRESS_MAX 16
#endif
@@ -45,120 +56,25 @@
* Dependencies
*********************************************************/
#include <string.h> /* memcpy, memmove, memset */
-#include "cpu.h"
+#include "cpu.h" /* bmi2 */
#include "mem.h" /* low level memory routines */
#define FSE_STATIC_LINKING_ONLY
#include "fse.h"
#define HUF_STATIC_LINKING_ONLY
#include "huf.h"
-#include "zstd_internal.h"
+#include "zstd_internal.h" /* blockProperties_t */
+#include "zstd_decompress_internal.h" /* ZSTD_DCtx */
+#include "zstd_ddict.h" /* ZSTD_DDictDictContent */
+#include "zstd_decompress_block.h" /* ZSTD_decompressBlock_internal */
#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1)
# include "zstd_legacy.h"
#endif
-/*-*************************************
-* Errors
-***************************************/
-#define ZSTD_isError ERR_isError /* for inlining */
-#define FSE_isError ERR_isError
-#define HUF_isError ERR_isError
-
-
-/*_*******************************************************
-* Memory operations
-**********************************************************/
-static void ZSTD_copy4(void* dst, const void* src) { memcpy(dst, src, 4); }
-
-
/*-*************************************************************
* Context management
***************************************************************/
-typedef enum { ZSTDds_getFrameHeaderSize, ZSTDds_decodeFrameHeader,
- ZSTDds_decodeBlockHeader, ZSTDds_decompressBlock,
- ZSTDds_decompressLastBlock, ZSTDds_checkChecksum,
- ZSTDds_decodeSkippableHeader, ZSTDds_skipFrame } ZSTD_dStage;
-
-typedef enum { zdss_init=0, zdss_loadHeader,
- zdss_read, zdss_load, zdss_flush } ZSTD_dStreamStage;
-
-
-typedef struct {
- U32 fastMode;
- U32 tableLog;
-} ZSTD_seqSymbol_header;
-
-typedef struct {
- U16 nextState;
- BYTE nbAdditionalBits;
- BYTE nbBits;
- U32 baseValue;
-} ZSTD_seqSymbol;
-
-#define SEQSYMBOL_TABLE_SIZE(log) (1 + (1 << (log)))
-
-typedef struct {
- ZSTD_seqSymbol LLTable[SEQSYMBOL_TABLE_SIZE(LLFSELog)];
- ZSTD_seqSymbol OFTable[SEQSYMBOL_TABLE_SIZE(OffFSELog)];
- ZSTD_seqSymbol MLTable[SEQSYMBOL_TABLE_SIZE(MLFSELog)];
- HUF_DTable hufTable[HUF_DTABLE_SIZE(HufLog)]; /* can accommodate HUF_decompress4X */
- U32 workspace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32];
- U32 rep[ZSTD_REP_NUM];
-} ZSTD_entropyDTables_t;
-
-struct ZSTD_DCtx_s
-{
- const ZSTD_seqSymbol* LLTptr;
- const ZSTD_seqSymbol* MLTptr;
- const ZSTD_seqSymbol* OFTptr;
- const HUF_DTable* HUFptr;
- ZSTD_entropyDTables_t entropy;
- const void* previousDstEnd; /* detect continuity */
- const void* base; /* start of current segment */
- const void* vBase; /* virtual start of previous segment if it was just before current one */
- const void* dictEnd; /* end of previous segment */
- size_t expected;
- ZSTD_frameHeader fParams;
- U64 decodedSize;
- blockType_e bType; /* used in ZSTD_decompressContinue(), store blockType between block header decoding and block decompression stages */
- ZSTD_dStage stage;
- U32 litEntropy;
- U32 fseEntropy;
- XXH64_state_t xxhState;
- size_t headerSize;
- U32 dictID;
- ZSTD_format_e format;
- const BYTE* litPtr;
- ZSTD_customMem customMem;
- size_t litSize;
- size_t rleSize;
- size_t staticSize;
- int bmi2; /* == 1 if the CPU supports BMI2 and 0 otherwise. CPU support is determined dynamically once per context lifetime. */
-
- /* streaming */
- ZSTD_DDict* ddictLocal;
- const ZSTD_DDict* ddict;
- ZSTD_dStreamStage streamStage;
- char* inBuff;
- size_t inBuffSize;
- size_t inPos;
- size_t maxWindowSize;
- char* outBuff;
- size_t outBuffSize;
- size_t outStart;
- size_t outEnd;
- size_t lhSize;
- void* legacyContext;
- U32 previousLegacyVersion;
- U32 legacyVersion;
- U32 hostageByte;
-
- /* workspace */
- BYTE litBuffer[ZSTD_BLOCKSIZE_MAX + WILDCOPY_OVERLENGTH];
- BYTE headerBuffer[ZSTD_FRAMEHEADERSIZE_MAX];
-}; /* typedef'd to ZSTD_DCtx within "zstd.h" */
-
size_t ZSTD_sizeof_DCtx (const ZSTD_DCtx* dctx)
{
if (dctx==NULL) return 0; /* support sizeof NULL */
@@ -173,8 +89,8 @@
static size_t ZSTD_startingInputLength(ZSTD_format_e format)
{
size_t const startingInputLength = (format==ZSTD_f_zstd1_magicless) ?
- ZSTD_frameHeaderSize_prefix - ZSTD_frameIdSize :
- ZSTD_frameHeaderSize_prefix;
+ ZSTD_FRAMEHEADERSIZE_PREFIX - ZSTD_FRAMEIDSIZE :
+ ZSTD_FRAMEHEADERSIZE_PREFIX;
ZSTD_STATIC_ASSERT(ZSTD_FRAMEHEADERSIZE_PREFIX >= ZSTD_FRAMEIDSIZE);
/* only supports formats ZSTD_f_zstd1 and ZSTD_f_zstd1_magicless */
assert( (format == ZSTD_f_zstd1) || (format == ZSTD_f_zstd1_magicless) );
@@ -188,10 +104,16 @@
dctx->maxWindowSize = ZSTD_MAXWINDOWSIZE_DEFAULT;
dctx->ddict = NULL;
dctx->ddictLocal = NULL;
+ dctx->dictEnd = NULL;
+ dctx->ddictIsCold = 0;
+ dctx->dictUses = ZSTD_dont_use;
dctx->inBuff = NULL;
dctx->inBuffSize = 0;
dctx->outBuffSize = 0;
dctx->streamStage = zdss_init;
+ dctx->legacyContext = NULL;
+ dctx->previousLegacyVersion = 0;
+ dctx->noForwardProgress = 0;
dctx->bmi2 = ZSTD_cpuid_bmi2(ZSTD_cpuid());
}
@@ -215,8 +137,6 @@
{ ZSTD_DCtx* const dctx = (ZSTD_DCtx*)ZSTD_malloc(sizeof(*dctx), customMem);
if (!dctx) return NULL;
dctx->customMem = customMem;
- dctx->legacyContext = NULL;
- dctx->previousLegacyVersion = 0;
ZSTD_initDCtx_internal(dctx);
return dctx;
}
@@ -228,13 +148,20 @@
return ZSTD_createDCtx_advanced(ZSTD_defaultCMem);
}
+static void ZSTD_clearDict(ZSTD_DCtx* dctx)
+{
+ ZSTD_freeDDict(dctx->ddictLocal);
+ dctx->ddictLocal = NULL;
+ dctx->ddict = NULL;
+ dctx->dictUses = ZSTD_dont_use;
+}
+
size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx)
{
if (dctx==NULL) return 0; /* support free on NULL */
- if (dctx->staticSize) return ERROR(memory_allocation); /* not compatible with static DCtx */
+ RETURN_ERROR_IF(dctx->staticSize, memory_allocation, "not compatible with static DCtx");
{ ZSTD_customMem const cMem = dctx->customMem;
- ZSTD_freeDDict(dctx->ddictLocal);
- dctx->ddictLocal = NULL;
+ ZSTD_clearDict(dctx);
ZSTD_free(dctx->inBuff, cMem);
dctx->inBuff = NULL;
#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1)
@@ -265,10 +192,10 @@
* Note 3 : Skippable Frame Identifiers are considered valid. */
unsigned ZSTD_isFrame(const void* buffer, size_t size)
{
- if (size < ZSTD_frameIdSize) return 0;
+ if (size < ZSTD_FRAMEIDSIZE) return 0;
{ U32 const magic = MEM_readLE32(buffer);
if (magic == ZSTD_MAGICNUMBER) return 1;
- if ((magic & 0xFFFFFFF0U) == ZSTD_MAGIC_SKIPPABLE_START) return 1;
+ if ((magic & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) return 1;
}
#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1)
if (ZSTD_isLegacy(buffer, size)) return 1;
@@ -284,7 +211,7 @@
static size_t ZSTD_frameHeaderSize_internal(const void* src, size_t srcSize, ZSTD_format_e format)
{
size_t const minInputSize = ZSTD_startingInputLength(format);
- if (srcSize < minInputSize) return ERROR(srcSize_wrong);
+ RETURN_ERROR_IF(srcSize < minInputSize, srcSize_wrong);
{ BYTE const fhd = ((const BYTE*)src)[minInputSize-1];
U32 const dictID= fhd & 3;
@@ -298,38 +225,41 @@
/** ZSTD_frameHeaderSize() :
* srcSize must be >= ZSTD_frameHeaderSize_prefix.
- * @return : size of the Frame Header */
+ * @return : size of the Frame Header,
+ * or an error code (if srcSize is too small) */
size_t ZSTD_frameHeaderSize(const void* src, size_t srcSize)
{
return ZSTD_frameHeaderSize_internal(src, srcSize, ZSTD_f_zstd1);
}
-/** ZSTD_getFrameHeader_internal() :
+/** ZSTD_getFrameHeader_advanced() :
* decode Frame Header, or require larger `srcSize`.
* note : only works for formats ZSTD_f_zstd1 and ZSTD_f_zstd1_magicless
* @return : 0, `zfhPtr` is correctly filled,
* >0, `srcSize` is too small, value is wanted `srcSize` amount,
* or an error code, which can be tested using ZSTD_isError() */
-static size_t ZSTD_getFrameHeader_internal(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format)
+size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format)
{
const BYTE* ip = (const BYTE*)src;
size_t const minInputSize = ZSTD_startingInputLength(format);
+ memset(zfhPtr, 0, sizeof(*zfhPtr)); /* not strictly necessary, but static analyzer do not understand that zfhPtr is only going to be read only if return value is zero, since they are 2 different signals */
if (srcSize < minInputSize) return minInputSize;
+ RETURN_ERROR_IF(src==NULL, GENERIC, "invalid parameter");
if ( (format != ZSTD_f_zstd1_magicless)
&& (MEM_readLE32(src) != ZSTD_MAGICNUMBER) ) {
- if ((MEM_readLE32(src) & 0xFFFFFFF0U) == ZSTD_MAGIC_SKIPPABLE_START) {
+ if ((MEM_readLE32(src) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) {
/* skippable frame */
- if (srcSize < ZSTD_skippableHeaderSize)
- return ZSTD_skippableHeaderSize; /* magic number + frame length */
+ if (srcSize < ZSTD_SKIPPABLEHEADERSIZE)
+ return ZSTD_SKIPPABLEHEADERSIZE; /* magic number + frame length */
memset(zfhPtr, 0, sizeof(*zfhPtr));
- zfhPtr->frameContentSize = MEM_readLE32((const char *)src + ZSTD_frameIdSize);
+ zfhPtr->frameContentSize = MEM_readLE32((const char *)src + ZSTD_FRAMEIDSIZE);
zfhPtr->frameType = ZSTD_skippableFrame;
return 0;
}
- return ERROR(prefix_unknown);
+ RETURN_ERROR(prefix_unknown);
}
/* ensure there is enough `srcSize` to fully read/decode frame header */
@@ -347,14 +277,13 @@
U64 windowSize = 0;
U32 dictID = 0;
U64 frameContentSize = ZSTD_CONTENTSIZE_UNKNOWN;
- if ((fhdByte & 0x08) != 0)
- return ERROR(frameParameter_unsupported); /* reserved bits, must be zero */
+ RETURN_ERROR_IF((fhdByte & 0x08) != 0, frameParameter_unsupported,
+ "reserved bits, must be zero");
if (!singleSegment) {
BYTE const wlByte = ip[pos++];
U32 const windowLog = (wlByte >> 3) + ZSTD_WINDOWLOG_ABSOLUTEMIN;
- if (windowLog > ZSTD_WINDOWLOG_MAX)
- return ERROR(frameParameter_windowTooLarge);
+ RETURN_ERROR_IF(windowLog > ZSTD_WINDOWLOG_MAX, frameParameter_windowTooLarge);
windowSize = (1ULL << windowLog);
windowSize += (windowSize >> 3) * (wlByte&7);
}
@@ -394,7 +323,7 @@
* or an error code, which can be tested using ZSTD_isError() */
size_t ZSTD_getFrameHeader(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize)
{
- return ZSTD_getFrameHeader_internal(zfhPtr, src, srcSize, ZSTD_f_zstd1);
+ return ZSTD_getFrameHeader_advanced(zfhPtr, src, srcSize, ZSTD_f_zstd1);
}
@@ -421,6 +350,23 @@
} }
}
+static size_t readSkippableFrameSize(void const* src, size_t srcSize)
+{
+ size_t const skippableHeaderSize = ZSTD_SKIPPABLEHEADERSIZE;
+ U32 sizeU32;
+
+ RETURN_ERROR_IF(srcSize < ZSTD_SKIPPABLEHEADERSIZE, srcSize_wrong);
+
+ sizeU32 = MEM_readLE32((BYTE const*)src + ZSTD_FRAMEIDSIZE);
+ RETURN_ERROR_IF((U32)(sizeU32 + ZSTD_SKIPPABLEHEADERSIZE) < sizeU32,
+ frameParameter_unsupported);
+ {
+ size_t const skippableSize = skippableHeaderSize + sizeU32;
+ RETURN_ERROR_IF(skippableSize > srcSize, srcSize_wrong);
+ return skippableSize;
+ }
+}
+
/** ZSTD_findDecompressedSize() :
* compatible with legacy mode
* `srcSize` must be the exact length of some number of ZSTD compressed and/or
@@ -430,18 +376,15 @@
{
unsigned long long totalDstSize = 0;
- while (srcSize >= ZSTD_frameHeaderSize_prefix) {
+ while (srcSize >= ZSTD_FRAMEHEADERSIZE_PREFIX) {
U32 const magicNumber = MEM_readLE32(src);
- if ((magicNumber & 0xFFFFFFF0U) == ZSTD_MAGIC_SKIPPABLE_START) {
- size_t skippableSize;
- if (srcSize < ZSTD_skippableHeaderSize)
- return ERROR(srcSize_wrong);
- skippableSize = MEM_readLE32((const BYTE *)src + ZSTD_frameIdSize)
- + ZSTD_skippableHeaderSize;
- if (srcSize < skippableSize) {
+ if ((magicNumber & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) {
+ size_t const skippableSize = readSkippableFrameSize(src, srcSize);
+ if (ZSTD_isError(skippableSize)) {
return ZSTD_CONTENTSIZE_ERROR;
}
+ assert(skippableSize <= srcSize);
src = (const BYTE *)src + skippableSize;
srcSize -= skippableSize;
@@ -471,9 +414,9 @@
}
/** ZSTD_getDecompressedSize() :
-* compatible with legacy mode
-* @return : decompressed size if known, 0 otherwise
- note : 0 can mean any of the following :
+ * compatible with legacy mode
+ * @return : decompressed size if known, 0 otherwise
+ note : 0 can mean any of the following :
- frame content is empty
- decompressed size field is not present in frame header
- frame header unknown / not supported
@@ -487,1255 +430,98 @@
/** ZSTD_decodeFrameHeader() :
-* `headerSize` must be the size provided by ZSTD_frameHeaderSize().
-* @return : 0 if success, or an error code, which can be tested using ZSTD_isError() */
+ * `headerSize` must be the size provided by ZSTD_frameHeaderSize().
+ * @return : 0 if success, or an error code, which can be tested using ZSTD_isError() */
static size_t ZSTD_decodeFrameHeader(ZSTD_DCtx* dctx, const void* src, size_t headerSize)
{
- size_t const result = ZSTD_getFrameHeader_internal(&(dctx->fParams), src, headerSize, dctx->format);
+ size_t const result = ZSTD_getFrameHeader_advanced(&(dctx->fParams), src, headerSize, dctx->format);
if (ZSTD_isError(result)) return result; /* invalid header */
- if (result>0) return ERROR(srcSize_wrong); /* headerSize too small */
- if (dctx->fParams.dictID && (dctx->dictID != dctx->fParams.dictID))
- return ERROR(dictionary_wrong);
+ RETURN_ERROR_IF(result>0, srcSize_wrong, "headerSize too small");
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
+ /* Skip the dictID check in fuzzing mode, because it makes the search
+ * harder.
+ */
+ RETURN_ERROR_IF(dctx->fParams.dictID && (dctx->dictID != dctx->fParams.dictID),
+ dictionary_wrong);
+#endif
if (dctx->fParams.checksumFlag) XXH64_reset(&dctx->xxhState, 0);
return 0;
}
-
-/*-*************************************************************
- * Block decoding
- ***************************************************************/
-
-/*! ZSTD_getcBlockSize() :
-* Provides the size of compressed block from block header `src` */
-size_t ZSTD_getcBlockSize(const void* src, size_t srcSize,
- blockProperties_t* bpPtr)
+static ZSTD_frameSizeInfo ZSTD_errorFrameSizeInfo(size_t ret)
{
- if (srcSize < ZSTD_blockHeaderSize) return ERROR(srcSize_wrong);
- { U32 const cBlockHeader = MEM_readLE24(src);
- U32 const cSize = cBlockHeader >> 3;
- bpPtr->lastBlock = cBlockHeader & 1;
- bpPtr->blockType = (blockType_e)((cBlockHeader >> 1) & 3);
- bpPtr->origSize = cSize; /* only useful for RLE */
- if (bpPtr->blockType == bt_rle) return 1;
- if (bpPtr->blockType == bt_reserved) return ERROR(corruption_detected);
- return cSize;
- }
+ ZSTD_frameSizeInfo frameSizeInfo;
+ frameSizeInfo.compressedSize = ret;
+ frameSizeInfo.decompressedBound = ZSTD_CONTENTSIZE_ERROR;
+ return frameSizeInfo;
}
-
-static size_t ZSTD_copyRawBlock(void* dst, size_t dstCapacity,
- const void* src, size_t srcSize)
+static ZSTD_frameSizeInfo ZSTD_findFrameSizeInfo(const void* src, size_t srcSize)
{
- if (srcSize > dstCapacity) return ERROR(dstSize_tooSmall);
- memcpy(dst, src, srcSize);
- return srcSize;
-}
+ ZSTD_frameSizeInfo frameSizeInfo;
+ memset(&frameSizeInfo, 0, sizeof(ZSTD_frameSizeInfo));
-
-static size_t ZSTD_setRleBlock(void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- size_t regenSize)
-{
- if (srcSize != 1) return ERROR(srcSize_wrong);
- if (regenSize > dstCapacity) return ERROR(dstSize_tooSmall);
- memset(dst, *(const BYTE*)src, regenSize);
- return regenSize;
-}
-
-/*! ZSTD_decodeLiteralsBlock() :
- * @return : nb of bytes read from src (< srcSize )
- * note : symbol not declared but exposed for fullbench */
-size_t ZSTD_decodeLiteralsBlock(ZSTD_DCtx* dctx,
- const void* src, size_t srcSize) /* note : srcSize < BLOCKSIZE */
-{
- if (srcSize < MIN_CBLOCK_SIZE) return ERROR(corruption_detected);
-
- { const BYTE* const istart = (const BYTE*) src;
- symbolEncodingType_e const litEncType = (symbolEncodingType_e)(istart[0] & 3);
-
- switch(litEncType)
- {
- case set_repeat:
- if (dctx->litEntropy==0) return ERROR(dictionary_corrupted);
- /* fall-through */
- case set_compressed:
- if (srcSize < 5) return ERROR(corruption_detected); /* srcSize >= MIN_CBLOCK_SIZE == 3; here we need up to 5 for case 3 */
- { size_t lhSize, litSize, litCSize;
- U32 singleStream=0;
- U32 const lhlCode = (istart[0] >> 2) & 3;
- U32 const lhc = MEM_readLE32(istart);
- switch(lhlCode)
- {
- case 0: case 1: default: /* note : default is impossible, since lhlCode into [0..3] */
- /* 2 - 2 - 10 - 10 */
- singleStream = !lhlCode;
- lhSize = 3;
- litSize = (lhc >> 4) & 0x3FF;
- litCSize = (lhc >> 14) & 0x3FF;
- break;
- case 2:
- /* 2 - 2 - 14 - 14 */
- lhSize = 4;
- litSize = (lhc >> 4) & 0x3FFF;
- litCSize = lhc >> 18;
- break;
- case 3:
- /* 2 - 2 - 18 - 18 */
- lhSize = 5;
- litSize = (lhc >> 4) & 0x3FFFF;
- litCSize = (lhc >> 22) + (istart[4] << 10);
- break;
- }
- if (litSize > ZSTD_BLOCKSIZE_MAX) return ERROR(corruption_detected);
- if (litCSize + lhSize > srcSize) return ERROR(corruption_detected);
-
- if (HUF_isError((litEncType==set_repeat) ?
- ( singleStream ?
- HUF_decompress1X_usingDTable_bmi2(dctx->litBuffer, litSize, istart+lhSize, litCSize, dctx->HUFptr, dctx->bmi2) :
- HUF_decompress4X_usingDTable_bmi2(dctx->litBuffer, litSize, istart+lhSize, litCSize, dctx->HUFptr, dctx->bmi2) ) :
- ( singleStream ?
- HUF_decompress1X2_DCtx_wksp_bmi2(dctx->entropy.hufTable, dctx->litBuffer, litSize, istart+lhSize, litCSize,
- dctx->entropy.workspace, sizeof(dctx->entropy.workspace), dctx->bmi2) :
- HUF_decompress4X_hufOnly_wksp_bmi2(dctx->entropy.hufTable, dctx->litBuffer, litSize, istart+lhSize, litCSize,
- dctx->entropy.workspace, sizeof(dctx->entropy.workspace), dctx->bmi2))))
- return ERROR(corruption_detected);
-
- dctx->litPtr = dctx->litBuffer;
- dctx->litSize = litSize;
- dctx->litEntropy = 1;
- if (litEncType==set_compressed) dctx->HUFptr = dctx->entropy.hufTable;
- memset(dctx->litBuffer + dctx->litSize, 0, WILDCOPY_OVERLENGTH);
- return litCSize + lhSize;
- }
-
- case set_basic:
- { size_t litSize, lhSize;
- U32 const lhlCode = ((istart[0]) >> 2) & 3;
- switch(lhlCode)
- {
- case 0: case 2: default: /* note : default is impossible, since lhlCode into [0..3] */
- lhSize = 1;
- litSize = istart[0] >> 3;
- break;
- case 1:
- lhSize = 2;
- litSize = MEM_readLE16(istart) >> 4;
- break;
- case 3:
- lhSize = 3;
- litSize = MEM_readLE24(istart) >> 4;
- break;
- }
-
- if (lhSize+litSize+WILDCOPY_OVERLENGTH > srcSize) { /* risk reading beyond src buffer with wildcopy */
- if (litSize+lhSize > srcSize) return ERROR(corruption_detected);
- memcpy(dctx->litBuffer, istart+lhSize, litSize);
- dctx->litPtr = dctx->litBuffer;
- dctx->litSize = litSize;
- memset(dctx->litBuffer + dctx->litSize, 0, WILDCOPY_OVERLENGTH);
- return lhSize+litSize;
- }
- /* direct reference into compressed stream */
- dctx->litPtr = istart+lhSize;
- dctx->litSize = litSize;
- return lhSize+litSize;
- }
-
- case set_rle:
- { U32 const lhlCode = ((istart[0]) >> 2) & 3;
- size_t litSize, lhSize;
- switch(lhlCode)
- {
- case 0: case 2: default: /* note : default is impossible, since lhlCode into [0..3] */
- lhSize = 1;
- litSize = istart[0] >> 3;
- break;
- case 1:
- lhSize = 2;
- litSize = MEM_readLE16(istart) >> 4;
- break;
- case 3:
- lhSize = 3;
- litSize = MEM_readLE24(istart) >> 4;
- if (srcSize<4) return ERROR(corruption_detected); /* srcSize >= MIN_CBLOCK_SIZE == 3; here we need lhSize+1 = 4 */
- break;
- }
- if (litSize > ZSTD_BLOCKSIZE_MAX) return ERROR(corruption_detected);
- memset(dctx->litBuffer, istart[lhSize], litSize + WILDCOPY_OVERLENGTH);
- dctx->litPtr = dctx->litBuffer;
- dctx->litSize = litSize;
- return lhSize+1;
- }
- default:
- return ERROR(corruption_detected); /* impossible */
- }
- }
-}
-
-/* Default FSE distribution tables.
- * These are pre-calculated FSE decoding tables using default distributions as defined in specification :
- * https://github.com/facebook/zstd/blob/master/doc/zstd_compression_format.md#default-distributions
- * They were generated programmatically with following method :
- * - start from default distributions, present in /lib/common/zstd_internal.h
- * - generate tables normally, using ZSTD_buildFSETable()
- * - printout the content of tables
- * - pretify output, report below, test with fuzzer to ensure it's correct */
-
-/* Default FSE distribution table for Literal Lengths */
-static const ZSTD_seqSymbol LL_defaultDTable[(1<<LL_DEFAULTNORMLOG)+1] = {
- { 1, 1, 1, LL_DEFAULTNORMLOG}, /* header : fastMode, tableLog */
- /* nextState, nbAddBits, nbBits, baseVal */
- { 0, 0, 4, 0}, { 16, 0, 4, 0},
- { 32, 0, 5, 1}, { 0, 0, 5, 3},
- { 0, 0, 5, 4}, { 0, 0, 5, 6},
- { 0, 0, 5, 7}, { 0, 0, 5, 9},
- { 0, 0, 5, 10}, { 0, 0, 5, 12},
- { 0, 0, 6, 14}, { 0, 1, 5, 16},
- { 0, 1, 5, 20}, { 0, 1, 5, 22},
- { 0, 2, 5, 28}, { 0, 3, 5, 32},
- { 0, 4, 5, 48}, { 32, 6, 5, 64},
- { 0, 7, 5, 128}, { 0, 8, 6, 256},
- { 0, 10, 6, 1024}, { 0, 12, 6, 4096},
- { 32, 0, 4, 0}, { 0, 0, 4, 1},
- { 0, 0, 5, 2}, { 32, 0, 5, 4},
- { 0, 0, 5, 5}, { 32, 0, 5, 7},
- { 0, 0, 5, 8}, { 32, 0, 5, 10},
- { 0, 0, 5, 11}, { 0, 0, 6, 13},
- { 32, 1, 5, 16}, { 0, 1, 5, 18},
- { 32, 1, 5, 22}, { 0, 2, 5, 24},
- { 32, 3, 5, 32}, { 0, 3, 5, 40},
- { 0, 6, 4, 64}, { 16, 6, 4, 64},
- { 32, 7, 5, 128}, { 0, 9, 6, 512},
- { 0, 11, 6, 2048}, { 48, 0, 4, 0},
- { 16, 0, 4, 1}, { 32, 0, 5, 2},
- { 32, 0, 5, 3}, { 32, 0, 5, 5},
- { 32, 0, 5, 6}, { 32, 0, 5, 8},
- { 32, 0, 5, 9}, { 32, 0, 5, 11},
- { 32, 0, 5, 12}, { 0, 0, 6, 15},
- { 32, 1, 5, 18}, { 32, 1, 5, 20},
- { 32, 2, 5, 24}, { 32, 2, 5, 28},
- { 32, 3, 5, 40}, { 32, 4, 5, 48},
- { 0, 16, 6,65536}, { 0, 15, 6,32768},
- { 0, 14, 6,16384}, { 0, 13, 6, 8192},
-}; /* LL_defaultDTable */
-
-/* Default FSE distribution table for Offset Codes */
-static const ZSTD_seqSymbol OF_defaultDTable[(1<<OF_DEFAULTNORMLOG)+1] = {
- { 1, 1, 1, OF_DEFAULTNORMLOG}, /* header : fastMode, tableLog */
- /* nextState, nbAddBits, nbBits, baseVal */
- { 0, 0, 5, 0}, { 0, 6, 4, 61},
- { 0, 9, 5, 509}, { 0, 15, 5,32765},
- { 0, 21, 5,2097149}, { 0, 3, 5, 5},
- { 0, 7, 4, 125}, { 0, 12, 5, 4093},
- { 0, 18, 5,262141}, { 0, 23, 5,8388605},
- { 0, 5, 5, 29}, { 0, 8, 4, 253},
- { 0, 14, 5,16381}, { 0, 20, 5,1048573},
- { 0, 2, 5, 1}, { 16, 7, 4, 125},
- { 0, 11, 5, 2045}, { 0, 17, 5,131069},
- { 0, 22, 5,4194301}, { 0, 4, 5, 13},
- { 16, 8, 4, 253}, { 0, 13, 5, 8189},
- { 0, 19, 5,524285}, { 0, 1, 5, 1},
- { 16, 6, 4, 61}, { 0, 10, 5, 1021},
- { 0, 16, 5,65533}, { 0, 28, 5,268435453},
- { 0, 27, 5,134217725}, { 0, 26, 5,67108861},
- { 0, 25, 5,33554429}, { 0, 24, 5,16777213},
-}; /* OF_defaultDTable */
-
-
-/* Default FSE distribution table for Match Lengths */
-static const ZSTD_seqSymbol ML_defaultDTable[(1<<ML_DEFAULTNORMLOG)+1] = {
- { 1, 1, 1, ML_DEFAULTNORMLOG}, /* header : fastMode, tableLog */
- /* nextState, nbAddBits, nbBits, baseVal */
- { 0, 0, 6, 3}, { 0, 0, 4, 4},
- { 32, 0, 5, 5}, { 0, 0, 5, 6},
- { 0, 0, 5, 8}, { 0, 0, 5, 9},
- { 0, 0, 5, 11}, { 0, 0, 6, 13},
- { 0, 0, 6, 16}, { 0, 0, 6, 19},
- { 0, 0, 6, 22}, { 0, 0, 6, 25},
- { 0, 0, 6, 28}, { 0, 0, 6, 31},
- { 0, 0, 6, 34}, { 0, 1, 6, 37},
- { 0, 1, 6, 41}, { 0, 2, 6, 47},
- { 0, 3, 6, 59}, { 0, 4, 6, 83},
- { 0, 7, 6, 131}, { 0, 9, 6, 515},
- { 16, 0, 4, 4}, { 0, 0, 4, 5},
- { 32, 0, 5, 6}, { 0, 0, 5, 7},
- { 32, 0, 5, 9}, { 0, 0, 5, 10},
- { 0, 0, 6, 12}, { 0, 0, 6, 15},
- { 0, 0, 6, 18}, { 0, 0, 6, 21},
- { 0, 0, 6, 24}, { 0, 0, 6, 27},
- { 0, 0, 6, 30}, { 0, 0, 6, 33},
- { 0, 1, 6, 35}, { 0, 1, 6, 39},
- { 0, 2, 6, 43}, { 0, 3, 6, 51},
- { 0, 4, 6, 67}, { 0, 5, 6, 99},
- { 0, 8, 6, 259}, { 32, 0, 4, 4},
- { 48, 0, 4, 4}, { 16, 0, 4, 5},
- { 32, 0, 5, 7}, { 32, 0, 5, 8},
- { 32, 0, 5, 10}, { 32, 0, 5, 11},
- { 0, 0, 6, 14}, { 0, 0, 6, 17},
- { 0, 0, 6, 20}, { 0, 0, 6, 23},
- { 0, 0, 6, 26}, { 0, 0, 6, 29},
- { 0, 0, 6, 32}, { 0, 16, 6,65539},
- { 0, 15, 6,32771}, { 0, 14, 6,16387},
- { 0, 13, 6, 8195}, { 0, 12, 6, 4099},
- { 0, 11, 6, 2051}, { 0, 10, 6, 1027},
-}; /* ML_defaultDTable */
-
-
-static void ZSTD_buildSeqTable_rle(ZSTD_seqSymbol* dt, U32 baseValue, U32 nbAddBits)
-{
- void* ptr = dt;
- ZSTD_seqSymbol_header* const DTableH = (ZSTD_seqSymbol_header*)ptr;
- ZSTD_seqSymbol* const cell = dt + 1;
-
- DTableH->tableLog = 0;
- DTableH->fastMode = 0;
-
- cell->nbBits = 0;
- cell->nextState = 0;
- assert(nbAddBits < 255);
- cell->nbAdditionalBits = (BYTE)nbAddBits;
- cell->baseValue = baseValue;
-}
-
-
-/* ZSTD_buildFSETable() :
- * generate FSE decoding table for one symbol (ll, ml or off) */
-static void
-ZSTD_buildFSETable(ZSTD_seqSymbol* dt,
- const short* normalizedCounter, unsigned maxSymbolValue,
- const U32* baseValue, const U32* nbAdditionalBits,
- unsigned tableLog)
-{
- ZSTD_seqSymbol* const tableDecode = dt+1;
- U16 symbolNext[MaxSeq+1];
-
- U32 const maxSV1 = maxSymbolValue + 1;
- U32 const tableSize = 1 << tableLog;
- U32 highThreshold = tableSize-1;
-
- /* Sanity Checks */
- assert(maxSymbolValue <= MaxSeq);
- assert(tableLog <= MaxFSELog);
-
- /* Init, lay down lowprob symbols */
- { ZSTD_seqSymbol_header DTableH;
- DTableH.tableLog = tableLog;
- DTableH.fastMode = 1;
- { S16 const largeLimit= (S16)(1 << (tableLog-1));
- U32 s;
- for (s=0; s<maxSV1; s++) {
- if (normalizedCounter[s]==-1) {
- tableDecode[highThreshold--].baseValue = s;
- symbolNext[s] = 1;
- } else {
- if (normalizedCounter[s] >= largeLimit) DTableH.fastMode=0;
- symbolNext[s] = normalizedCounter[s];
- } } }
- memcpy(dt, &DTableH, sizeof(DTableH));
- }
-
- /* Spread symbols */
- { U32 const tableMask = tableSize-1;
- U32 const step = FSE_TABLESTEP(tableSize);
- U32 s, position = 0;
- for (s=0; s<maxSV1; s++) {
- int i;
- for (i=0; i<normalizedCounter[s]; i++) {
- tableDecode[position].baseValue = s;
- position = (position + step) & tableMask;
- while (position > highThreshold) position = (position + step) & tableMask; /* lowprob area */
- } }
- assert(position == 0); /* position must reach all cells once, otherwise normalizedCounter is incorrect */
- }
-
- /* Build Decoding table */
- { U32 u;
- for (u=0; u<tableSize; u++) {
- U32 const symbol = tableDecode[u].baseValue;
- U32 const nextState = symbolNext[symbol]++;
- tableDecode[u].nbBits = (BYTE) (tableLog - BIT_highbit32(nextState) );
- tableDecode[u].nextState = (U16) ( (nextState << tableDecode[u].nbBits) - tableSize);
- assert(nbAdditionalBits[symbol] < 255);
- tableDecode[u].nbAdditionalBits = (BYTE)nbAdditionalBits[symbol];
- tableDecode[u].baseValue = baseValue[symbol];
- } }
-}
-
-
-/*! ZSTD_buildSeqTable() :
- * @return : nb bytes read from src,
- * or an error code if it fails */
-static size_t ZSTD_buildSeqTable(ZSTD_seqSymbol* DTableSpace, const ZSTD_seqSymbol** DTablePtr,
- symbolEncodingType_e type, U32 max, U32 maxLog,
- const void* src, size_t srcSize,
- const U32* baseValue, const U32* nbAdditionalBits,
- const ZSTD_seqSymbol* defaultTable, U32 flagRepeatTable)
-{
- switch(type)
- {
- case set_rle :
- if (!srcSize) return ERROR(srcSize_wrong);
- if ( (*(const BYTE*)src) > max) return ERROR(corruption_detected);
- { U32 const symbol = *(const BYTE*)src;
- U32 const baseline = baseValue[symbol];
- U32 const nbBits = nbAdditionalBits[symbol];
- ZSTD_buildSeqTable_rle(DTableSpace, baseline, nbBits);
- }
- *DTablePtr = DTableSpace;
- return 1;
- case set_basic :
- *DTablePtr = defaultTable;
- return 0;
- case set_repeat:
- if (!flagRepeatTable) return ERROR(corruption_detected);
- return 0;
- case set_compressed :
- { U32 tableLog;
- S16 norm[MaxSeq+1];
- size_t const headerSize = FSE_readNCount(norm, &max, &tableLog, src, srcSize);
- if (FSE_isError(headerSize)) return ERROR(corruption_detected);
- if (tableLog > maxLog) return ERROR(corruption_detected);
- ZSTD_buildFSETable(DTableSpace, norm, max, baseValue, nbAdditionalBits, tableLog);
- *DTablePtr = DTableSpace;
- return headerSize;
- }
- default : /* impossible */
- assert(0);
- return ERROR(GENERIC);
- }
-}
-
-static const U32 LL_base[MaxLL+1] = {
- 0, 1, 2, 3, 4, 5, 6, 7,
- 8, 9, 10, 11, 12, 13, 14, 15,
- 16, 18, 20, 22, 24, 28, 32, 40,
- 48, 64, 0x80, 0x100, 0x200, 0x400, 0x800, 0x1000,
- 0x2000, 0x4000, 0x8000, 0x10000 };
-
-static const U32 OF_base[MaxOff+1] = {
- 0, 1, 1, 5, 0xD, 0x1D, 0x3D, 0x7D,
- 0xFD, 0x1FD, 0x3FD, 0x7FD, 0xFFD, 0x1FFD, 0x3FFD, 0x7FFD,
- 0xFFFD, 0x1FFFD, 0x3FFFD, 0x7FFFD, 0xFFFFD, 0x1FFFFD, 0x3FFFFD, 0x7FFFFD,
- 0xFFFFFD, 0x1FFFFFD, 0x3FFFFFD, 0x7FFFFFD, 0xFFFFFFD, 0x1FFFFFFD, 0x3FFFFFFD, 0x7FFFFFFD };
-
-static const U32 OF_bits[MaxOff+1] = {
- 0, 1, 2, 3, 4, 5, 6, 7,
- 8, 9, 10, 11, 12, 13, 14, 15,
- 16, 17, 18, 19, 20, 21, 22, 23,
- 24, 25, 26, 27, 28, 29, 30, 31 };
-
-static const U32 ML_base[MaxML+1] = {
- 3, 4, 5, 6, 7, 8, 9, 10,
- 11, 12, 13, 14, 15, 16, 17, 18,
- 19, 20, 21, 22, 23, 24, 25, 26,
- 27, 28, 29, 30, 31, 32, 33, 34,
- 35, 37, 39, 41, 43, 47, 51, 59,
- 67, 83, 99, 0x83, 0x103, 0x203, 0x403, 0x803,
- 0x1003, 0x2003, 0x4003, 0x8003, 0x10003 };
-
-
-size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr,
- const void* src, size_t srcSize)
-{
- const BYTE* const istart = (const BYTE* const)src;
- const BYTE* const iend = istart + srcSize;
- const BYTE* ip = istart;
- DEBUGLOG(5, "ZSTD_decodeSeqHeaders");
-
- /* check */
- if (srcSize < MIN_SEQUENCES_SIZE) return ERROR(srcSize_wrong);
-
- /* SeqHead */
- { int nbSeq = *ip++;
- if (!nbSeq) { *nbSeqPtr=0; return 1; }
- if (nbSeq > 0x7F) {
- if (nbSeq == 0xFF) {
- if (ip+2 > iend) return ERROR(srcSize_wrong);
- nbSeq = MEM_readLE16(ip) + LONGNBSEQ, ip+=2;
- } else {
- if (ip >= iend) return ERROR(srcSize_wrong);
- nbSeq = ((nbSeq-0x80)<<8) + *ip++;
- }
- }
- *nbSeqPtr = nbSeq;
- }
-
- /* FSE table descriptors */
- if (ip+4 > iend) return ERROR(srcSize_wrong); /* minimum possible size */
- { symbolEncodingType_e const LLtype = (symbolEncodingType_e)(*ip >> 6);
- symbolEncodingType_e const OFtype = (symbolEncodingType_e)((*ip >> 4) & 3);
- symbolEncodingType_e const MLtype = (symbolEncodingType_e)((*ip >> 2) & 3);
- ip++;
-
- /* Build DTables */
- { size_t const llhSize = ZSTD_buildSeqTable(dctx->entropy.LLTable, &dctx->LLTptr,
- LLtype, MaxLL, LLFSELog,
- ip, iend-ip,
- LL_base, LL_bits,
- LL_defaultDTable, dctx->fseEntropy);
- if (ZSTD_isError(llhSize)) return ERROR(corruption_detected);
- ip += llhSize;
- }
-
- { size_t const ofhSize = ZSTD_buildSeqTable(dctx->entropy.OFTable, &dctx->OFTptr,
- OFtype, MaxOff, OffFSELog,
- ip, iend-ip,
- OF_base, OF_bits,
- OF_defaultDTable, dctx->fseEntropy);
- if (ZSTD_isError(ofhSize)) return ERROR(corruption_detected);
- ip += ofhSize;
- }
-
- { size_t const mlhSize = ZSTD_buildSeqTable(dctx->entropy.MLTable, &dctx->MLTptr,
- MLtype, MaxML, MLFSELog,
- ip, iend-ip,
- ML_base, ML_bits,
- ML_defaultDTable, dctx->fseEntropy);
- if (ZSTD_isError(mlhSize)) return ERROR(corruption_detected);
- ip += mlhSize;
- }
- }
-
- return ip-istart;
-}
-
-
-typedef struct {
- size_t litLength;
- size_t matchLength;
- size_t offset;
- const BYTE* match;
-} seq_t;
-
-typedef struct {
- size_t state;
- const ZSTD_seqSymbol* table;
-} ZSTD_fseState;
-
-typedef struct {
- BIT_DStream_t DStream;
- ZSTD_fseState stateLL;
- ZSTD_fseState stateOffb;
- ZSTD_fseState stateML;
- size_t prevOffset[ZSTD_REP_NUM];
- const BYTE* prefixStart;
- const BYTE* dictEnd;
- size_t pos;
-} seqState_t;
-
-
-FORCE_NOINLINE
-size_t ZSTD_execSequenceLast7(BYTE* op,
- BYTE* const oend, seq_t sequence,
- const BYTE** litPtr, const BYTE* const litLimit,
- const BYTE* const base, const BYTE* const vBase, const BYTE* const dictEnd)
-{
- BYTE* const oLitEnd = op + sequence.litLength;
- size_t const sequenceLength = sequence.litLength + sequence.matchLength;
- BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */
- BYTE* const oend_w = oend - WILDCOPY_OVERLENGTH;
- const BYTE* const iLitEnd = *litPtr + sequence.litLength;
- const BYTE* match = oLitEnd - sequence.offset;
-
- /* check */
- if (oMatchEnd>oend) return ERROR(dstSize_tooSmall); /* last match must start at a minimum distance of WILDCOPY_OVERLENGTH from oend */
- if (iLitEnd > litLimit) return ERROR(corruption_detected); /* over-read beyond lit buffer */
- if (oLitEnd <= oend_w) return ERROR(GENERIC); /* Precondition */
-
- /* copy literals */
- if (op < oend_w) {
- ZSTD_wildcopy(op, *litPtr, oend_w - op);
- *litPtr += oend_w - op;
- op = oend_w;
- }
- while (op < oLitEnd) *op++ = *(*litPtr)++;
-
- /* copy Match */
- if (sequence.offset > (size_t)(oLitEnd - base)) {
- /* offset beyond prefix */
- if (sequence.offset > (size_t)(oLitEnd - vBase)) return ERROR(corruption_detected);
- match = dictEnd - (base-match);
- if (match + sequence.matchLength <= dictEnd) {
- memmove(oLitEnd, match, sequence.matchLength);
- return sequenceLength;
- }
- /* span extDict & currentPrefixSegment */
- { size_t const length1 = dictEnd - match;
- memmove(oLitEnd, match, length1);
- op = oLitEnd + length1;
- sequence.matchLength -= length1;
- match = base;
- } }
- while (op < oMatchEnd) *op++ = *match++;
- return sequenceLength;
-}
-
-
-HINT_INLINE
-size_t ZSTD_execSequence(BYTE* op,
- BYTE* const oend, seq_t sequence,
- const BYTE** litPtr, const BYTE* const litLimit,
- const BYTE* const base, const BYTE* const vBase, const BYTE* const dictEnd)
-{
- BYTE* const oLitEnd = op + sequence.litLength;
- size_t const sequenceLength = sequence.litLength + sequence.matchLength;
- BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */
- BYTE* const oend_w = oend - WILDCOPY_OVERLENGTH;
- const BYTE* const iLitEnd = *litPtr + sequence.litLength;
- const BYTE* match = oLitEnd - sequence.offset;
-
- /* check */
- if (oMatchEnd>oend) return ERROR(dstSize_tooSmall); /* last match must start at a minimum distance of WILDCOPY_OVERLENGTH from oend */
- if (iLitEnd > litLimit) return ERROR(corruption_detected); /* over-read beyond lit buffer */
- if (oLitEnd>oend_w) return ZSTD_execSequenceLast7(op, oend, sequence, litPtr, litLimit, base, vBase, dictEnd);
-
- /* copy Literals */
- ZSTD_copy8(op, *litPtr);
- if (sequence.litLength > 8)
- ZSTD_wildcopy(op+8, (*litPtr)+8, sequence.litLength - 8); /* note : since oLitEnd <= oend-WILDCOPY_OVERLENGTH, no risk of overwrite beyond oend */
- op = oLitEnd;
- *litPtr = iLitEnd; /* update for next sequence */
-
- /* copy Match */
- if (sequence.offset > (size_t)(oLitEnd - base)) {
- /* offset beyond prefix -> go into extDict */
- if (sequence.offset > (size_t)(oLitEnd - vBase))
- return ERROR(corruption_detected);
- match = dictEnd + (match - base);
- if (match + sequence.matchLength <= dictEnd) {
- memmove(oLitEnd, match, sequence.matchLength);
- return sequenceLength;
- }
- /* span extDict & currentPrefixSegment */
- { size_t const length1 = dictEnd - match;
- memmove(oLitEnd, match, length1);
- op = oLitEnd + length1;
- sequence.matchLength -= length1;
- match = base;
- if (op > oend_w || sequence.matchLength < MINMATCH) {
- U32 i;
- for (i = 0; i < sequence.matchLength; ++i) op[i] = match[i];
- return sequenceLength;
- }
- } }
- /* Requirement: op <= oend_w && sequence.matchLength >= MINMATCH */
-
- /* match within prefix */
- if (sequence.offset < 8) {
- /* close range match, overlap */
- static const U32 dec32table[] = { 0, 1, 2, 1, 4, 4, 4, 4 }; /* added */
- static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* subtracted */
- int const sub2 = dec64table[sequence.offset];
- op[0] = match[0];
- op[1] = match[1];
- op[2] = match[2];
- op[3] = match[3];
- match += dec32table[sequence.offset];
- ZSTD_copy4(op+4, match);
- match -= sub2;
- } else {
- ZSTD_copy8(op, match);
- }
- op += 8; match += 8;
-
- if (oMatchEnd > oend-(16-MINMATCH)) {
- if (op < oend_w) {
- ZSTD_wildcopy(op, match, oend_w - op);
- match += oend_w - op;
- op = oend_w;
- }
- while (op < oMatchEnd) *op++ = *match++;
- } else {
- ZSTD_wildcopy(op, match, (ptrdiff_t)sequence.matchLength-8); /* works even if matchLength < 8 */
- }
- return sequenceLength;
-}
-
-
-HINT_INLINE
-size_t ZSTD_execSequenceLong(BYTE* op,
- BYTE* const oend, seq_t sequence,
- const BYTE** litPtr, const BYTE* const litLimit,
- const BYTE* const prefixStart, const BYTE* const dictStart, const BYTE* const dictEnd)
-{
- BYTE* const oLitEnd = op + sequence.litLength;
- size_t const sequenceLength = sequence.litLength + sequence.matchLength;
- BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */
- BYTE* const oend_w = oend - WILDCOPY_OVERLENGTH;
- const BYTE* const iLitEnd = *litPtr + sequence.litLength;
- const BYTE* match = sequence.match;
-
- /* check */
- if (oMatchEnd > oend) return ERROR(dstSize_tooSmall); /* last match must start at a minimum distance of WILDCOPY_OVERLENGTH from oend */
- if (iLitEnd > litLimit) return ERROR(corruption_detected); /* over-read beyond lit buffer */
- if (oLitEnd > oend_w) return ZSTD_execSequenceLast7(op, oend, sequence, litPtr, litLimit, prefixStart, dictStart, dictEnd);
-
- /* copy Literals */
- ZSTD_copy8(op, *litPtr); /* note : op <= oLitEnd <= oend_w == oend - 8 */
- if (sequence.litLength > 8)
- ZSTD_wildcopy(op+8, (*litPtr)+8, sequence.litLength - 8); /* note : since oLitEnd <= oend-WILDCOPY_OVERLENGTH, no risk of overwrite beyond oend */
- op = oLitEnd;
- *litPtr = iLitEnd; /* update for next sequence */
-
- /* copy Match */
- if (sequence.offset > (size_t)(oLitEnd - prefixStart)) {
- /* offset beyond prefix */
- if (sequence.offset > (size_t)(oLitEnd - dictStart)) return ERROR(corruption_detected);
- if (match + sequence.matchLength <= dictEnd) {
- memmove(oLitEnd, match, sequence.matchLength);
- return sequenceLength;
- }
- /* span extDict & currentPrefixSegment */
- { size_t const length1 = dictEnd - match;
- memmove(oLitEnd, match, length1);
- op = oLitEnd + length1;
- sequence.matchLength -= length1;
- match = prefixStart;
- if (op > oend_w || sequence.matchLength < MINMATCH) {
- U32 i;
- for (i = 0; i < sequence.matchLength; ++i) op[i] = match[i];
- return sequenceLength;
- }
- } }
- assert(op <= oend_w);
- assert(sequence.matchLength >= MINMATCH);
-
- /* match within prefix */
- if (sequence.offset < 8) {
- /* close range match, overlap */
- static const U32 dec32table[] = { 0, 1, 2, 1, 4, 4, 4, 4 }; /* added */
- static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* subtracted */
- int const sub2 = dec64table[sequence.offset];
- op[0] = match[0];
- op[1] = match[1];
- op[2] = match[2];
- op[3] = match[3];
- match += dec32table[sequence.offset];
- ZSTD_copy4(op+4, match);
- match -= sub2;
- } else {
- ZSTD_copy8(op, match);
- }
- op += 8; match += 8;
-
- if (oMatchEnd > oend-(16-MINMATCH)) {
- if (op < oend_w) {
- ZSTD_wildcopy(op, match, oend_w - op);
- match += oend_w - op;
- op = oend_w;
- }
- while (op < oMatchEnd) *op++ = *match++;
- } else {
- ZSTD_wildcopy(op, match, (ptrdiff_t)sequence.matchLength-8); /* works even if matchLength < 8 */
- }
- return sequenceLength;
-}
-
-static void
-ZSTD_initFseState(ZSTD_fseState* DStatePtr, BIT_DStream_t* bitD, const ZSTD_seqSymbol* dt)
-{
- const void* ptr = dt;
- const ZSTD_seqSymbol_header* const DTableH = (const ZSTD_seqSymbol_header*)ptr;
- DStatePtr->state = BIT_readBits(bitD, DTableH->tableLog);
- DEBUGLOG(6, "ZSTD_initFseState : val=%u using %u bits",
- (U32)DStatePtr->state, DTableH->tableLog);
- BIT_reloadDStream(bitD);
- DStatePtr->table = dt + 1;
-}
-
-FORCE_INLINE_TEMPLATE void
-ZSTD_updateFseState(ZSTD_fseState* DStatePtr, BIT_DStream_t* bitD)
-{
- ZSTD_seqSymbol const DInfo = DStatePtr->table[DStatePtr->state];
- U32 const nbBits = DInfo.nbBits;
- size_t const lowBits = BIT_readBits(bitD, nbBits);
- DStatePtr->state = DInfo.nextState + lowBits;
-}
-
-/* We need to add at most (ZSTD_WINDOWLOG_MAX_32 - 1) bits to read the maximum
- * offset bits. But we can only read at most (STREAM_ACCUMULATOR_MIN_32 - 1)
- * bits before reloading. This value is the maximum number of bytes we read
- * after reloading when we are decoding long offets.
- */
-#define LONG_OFFSETS_MAX_EXTRA_BITS_32 \
- (ZSTD_WINDOWLOG_MAX_32 > STREAM_ACCUMULATOR_MIN_32 \
- ? ZSTD_WINDOWLOG_MAX_32 - STREAM_ACCUMULATOR_MIN_32 \
- : 0)
-
-typedef enum { ZSTD_lo_isRegularOffset, ZSTD_lo_isLongOffset=1 } ZSTD_longOffset_e;
-
-FORCE_INLINE_TEMPLATE seq_t
-ZSTD_decodeSequence(seqState_t* seqState, const ZSTD_longOffset_e longOffsets)
-{
- seq_t seq;
- U32 const llBits = seqState->stateLL.table[seqState->stateLL.state].nbAdditionalBits;
- U32 const mlBits = seqState->stateML.table[seqState->stateML.state].nbAdditionalBits;
- U32 const ofBits = seqState->stateOffb.table[seqState->stateOffb.state].nbAdditionalBits;
- U32 const totalBits = llBits+mlBits+ofBits;
- U32 const llBase = seqState->stateLL.table[seqState->stateLL.state].baseValue;
- U32 const mlBase = seqState->stateML.table[seqState->stateML.state].baseValue;
- U32 const ofBase = seqState->stateOffb.table[seqState->stateOffb.state].baseValue;
-
- /* sequence */
- { size_t offset;
- if (!ofBits)
- offset = 0;
- else {
- ZSTD_STATIC_ASSERT(ZSTD_lo_isLongOffset == 1);
- ZSTD_STATIC_ASSERT(LONG_OFFSETS_MAX_EXTRA_BITS_32 == 5);
- assert(ofBits <= MaxOff);
- if (MEM_32bits() && longOffsets && (ofBits >= STREAM_ACCUMULATOR_MIN_32)) {
- U32 const extraBits = ofBits - MIN(ofBits, 32 - seqState->DStream.bitsConsumed);
- offset = ofBase + (BIT_readBitsFast(&seqState->DStream, ofBits - extraBits) << extraBits);
- BIT_reloadDStream(&seqState->DStream);
- if (extraBits) offset += BIT_readBitsFast(&seqState->DStream, extraBits);
- assert(extraBits <= LONG_OFFSETS_MAX_EXTRA_BITS_32); /* to avoid another reload */
- } else {
- offset = ofBase + BIT_readBitsFast(&seqState->DStream, ofBits/*>0*/); /* <= (ZSTD_WINDOWLOG_MAX-1) bits */
- if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream);
- }
- }
-
- if (ofBits <= 1) {
- offset += (llBase==0);
- if (offset) {
- size_t temp = (offset==3) ? seqState->prevOffset[0] - 1 : seqState->prevOffset[offset];
- temp += !temp; /* 0 is not valid; input is corrupted; force offset to 1 */
- if (offset != 1) seqState->prevOffset[2] = seqState->prevOffset[1];
- seqState->prevOffset[1] = seqState->prevOffset[0];
- seqState->prevOffset[0] = offset = temp;
- } else { /* offset == 0 */
- offset = seqState->prevOffset[0];
- }
- } else {
- seqState->prevOffset[2] = seqState->prevOffset[1];
- seqState->prevOffset[1] = seqState->prevOffset[0];
- seqState->prevOffset[0] = offset;
- }
- seq.offset = offset;
- }
-
- seq.matchLength = mlBase
- + ((mlBits>0) ? BIT_readBitsFast(&seqState->DStream, mlBits/*>0*/) : 0); /* <= 16 bits */
- if (MEM_32bits() && (mlBits+llBits >= STREAM_ACCUMULATOR_MIN_32-LONG_OFFSETS_MAX_EXTRA_BITS_32))
- BIT_reloadDStream(&seqState->DStream);
- if (MEM_64bits() && (totalBits >= STREAM_ACCUMULATOR_MIN_64-(LLFSELog+MLFSELog+OffFSELog)))
- BIT_reloadDStream(&seqState->DStream);
- /* Ensure there are enough bits to read the rest of data in 64-bit mode. */
- ZSTD_STATIC_ASSERT(16+LLFSELog+MLFSELog+OffFSELog < STREAM_ACCUMULATOR_MIN_64);
-
- seq.litLength = llBase
- + ((llBits>0) ? BIT_readBitsFast(&seqState->DStream, llBits/*>0*/) : 0); /* <= 16 bits */
- if (MEM_32bits())
- BIT_reloadDStream(&seqState->DStream);
-
- DEBUGLOG(6, "seq: litL=%u, matchL=%u, offset=%u",
- (U32)seq.litLength, (U32)seq.matchLength, (U32)seq.offset);
-
- /* ANS state update */
- ZSTD_updateFseState(&seqState->stateLL, &seqState->DStream); /* <= 9 bits */
- ZSTD_updateFseState(&seqState->stateML, &seqState->DStream); /* <= 9 bits */
- if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream); /* <= 18 bits */
- ZSTD_updateFseState(&seqState->stateOffb, &seqState->DStream); /* <= 8 bits */
-
- return seq;
-}
-
-FORCE_INLINE_TEMPLATE size_t
-ZSTD_decompressSequences_body( ZSTD_DCtx* dctx,
- void* dst, size_t maxDstSize,
- const void* seqStart, size_t seqSize, int nbSeq,
- const ZSTD_longOffset_e isLongOffset)
-{
- const BYTE* ip = (const BYTE*)seqStart;
- const BYTE* const iend = ip + seqSize;
- BYTE* const ostart = (BYTE* const)dst;
- BYTE* const oend = ostart + maxDstSize;
- BYTE* op = ostart;
- const BYTE* litPtr = dctx->litPtr;
- const BYTE* const litEnd = litPtr + dctx->litSize;
- const BYTE* const base = (const BYTE*) (dctx->base);
- const BYTE* const vBase = (const BYTE*) (dctx->vBase);
- const BYTE* const dictEnd = (const BYTE*) (dctx->dictEnd);
- DEBUGLOG(5, "ZSTD_decompressSequences");
-
- /* Regen sequences */
- if (nbSeq) {
- seqState_t seqState;
- dctx->fseEntropy = 1;
- { U32 i; for (i=0; i<ZSTD_REP_NUM; i++) seqState.prevOffset[i] = dctx->entropy.rep[i]; }
- CHECK_E(BIT_initDStream(&seqState.DStream, ip, iend-ip), corruption_detected);
- ZSTD_initFseState(&seqState.stateLL, &seqState.DStream, dctx->LLTptr);
- ZSTD_initFseState(&seqState.stateOffb, &seqState.DStream, dctx->OFTptr);
- ZSTD_initFseState(&seqState.stateML, &seqState.DStream, dctx->MLTptr);
-
- for ( ; (BIT_reloadDStream(&(seqState.DStream)) <= BIT_DStream_completed) && nbSeq ; ) {
- nbSeq--;
- { seq_t const sequence = ZSTD_decodeSequence(&seqState, isLongOffset);
- size_t const oneSeqSize = ZSTD_execSequence(op, oend, sequence, &litPtr, litEnd, base, vBase, dictEnd);
- DEBUGLOG(6, "regenerated sequence size : %u", (U32)oneSeqSize);
- if (ZSTD_isError(oneSeqSize)) return oneSeqSize;
- op += oneSeqSize;
- } }
-
- /* check if reached exact end */
- DEBUGLOG(5, "ZSTD_decompressSequences: after decode loop, remaining nbSeq : %i", nbSeq);
- if (nbSeq) return ERROR(corruption_detected);
- /* save reps for next block */
- { U32 i; for (i=0; i<ZSTD_REP_NUM; i++) dctx->entropy.rep[i] = (U32)(seqState.prevOffset[i]); }
- }
-
- /* last literal segment */
- { size_t const lastLLSize = litEnd - litPtr;
- if (lastLLSize > (size_t)(oend-op)) return ERROR(dstSize_tooSmall);
- memcpy(op, litPtr, lastLLSize);
- op += lastLLSize;
- }
-
- return op-ostart;
-}
-
-static size_t
-ZSTD_decompressSequences_default(ZSTD_DCtx* dctx,
- void* dst, size_t maxDstSize,
- const void* seqStart, size_t seqSize, int nbSeq,
- const ZSTD_longOffset_e isLongOffset)
-{
- return ZSTD_decompressSequences_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
-}
-
-
-
-FORCE_INLINE_TEMPLATE seq_t
-ZSTD_decodeSequenceLong(seqState_t* seqState, ZSTD_longOffset_e const longOffsets)
-{
- seq_t seq;
- U32 const llBits = seqState->stateLL.table[seqState->stateLL.state].nbAdditionalBits;
- U32 const mlBits = seqState->stateML.table[seqState->stateML.state].nbAdditionalBits;
- U32 const ofBits = seqState->stateOffb.table[seqState->stateOffb.state].nbAdditionalBits;
- U32 const totalBits = llBits+mlBits+ofBits;
- U32 const llBase = seqState->stateLL.table[seqState->stateLL.state].baseValue;
- U32 const mlBase = seqState->stateML.table[seqState->stateML.state].baseValue;
- U32 const ofBase = seqState->stateOffb.table[seqState->stateOffb.state].baseValue;
-
- /* sequence */
- { size_t offset;
- if (!ofBits)
- offset = 0;
- else {
- ZSTD_STATIC_ASSERT(ZSTD_lo_isLongOffset == 1);
- ZSTD_STATIC_ASSERT(LONG_OFFSETS_MAX_EXTRA_BITS_32 == 5);
- assert(ofBits <= MaxOff);
- if (MEM_32bits() && longOffsets) {
- U32 const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN_32-1);
- offset = ofBase + (BIT_readBitsFast(&seqState->DStream, ofBits - extraBits) << extraBits);
- if (MEM_32bits() || extraBits) BIT_reloadDStream(&seqState->DStream);
- if (extraBits) offset += BIT_readBitsFast(&seqState->DStream, extraBits);
- } else {
- offset = ofBase + BIT_readBitsFast(&seqState->DStream, ofBits); /* <= (ZSTD_WINDOWLOG_MAX-1) bits */
- if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream);
- }
- }
-
- if (ofBits <= 1) {
- offset += (llBase==0);
- if (offset) {
- size_t temp = (offset==3) ? seqState->prevOffset[0] - 1 : seqState->prevOffset[offset];
- temp += !temp; /* 0 is not valid; input is corrupted; force offset to 1 */
- if (offset != 1) seqState->prevOffset[2] = seqState->prevOffset[1];
- seqState->prevOffset[1] = seqState->prevOffset[0];
- seqState->prevOffset[0] = offset = temp;
- } else {
- offset = seqState->prevOffset[0];
- }
- } else {
- seqState->prevOffset[2] = seqState->prevOffset[1];
- seqState->prevOffset[1] = seqState->prevOffset[0];
- seqState->prevOffset[0] = offset;
- }
- seq.offset = offset;
- }
-
- seq.matchLength = mlBase + ((mlBits>0) ? BIT_readBitsFast(&seqState->DStream, mlBits) : 0); /* <= 16 bits */
- if (MEM_32bits() && (mlBits+llBits >= STREAM_ACCUMULATOR_MIN_32-LONG_OFFSETS_MAX_EXTRA_BITS_32))
- BIT_reloadDStream(&seqState->DStream);
- if (MEM_64bits() && (totalBits >= STREAM_ACCUMULATOR_MIN_64-(LLFSELog+MLFSELog+OffFSELog)))
- BIT_reloadDStream(&seqState->DStream);
- /* Verify that there is enough bits to read the rest of the data in 64-bit mode. */
- ZSTD_STATIC_ASSERT(16+LLFSELog+MLFSELog+OffFSELog < STREAM_ACCUMULATOR_MIN_64);
-
- seq.litLength = llBase + ((llBits>0) ? BIT_readBitsFast(&seqState->DStream, llBits) : 0); /* <= 16 bits */
- if (MEM_32bits())
- BIT_reloadDStream(&seqState->DStream);
-
- { size_t const pos = seqState->pos + seq.litLength;
- const BYTE* const matchBase = (seq.offset > pos) ? seqState->dictEnd : seqState->prefixStart;
- seq.match = matchBase + pos - seq.offset; /* note : this operation can overflow when seq.offset is really too large, which can only happen when input is corrupted.
- * No consequence though : no memory access will occur, overly large offset will be detected in ZSTD_execSequenceLong() */
- seqState->pos = pos + seq.matchLength;
- }
-
- /* ANS state update */
- ZSTD_updateFseState(&seqState->stateLL, &seqState->DStream); /* <= 9 bits */
- ZSTD_updateFseState(&seqState->stateML, &seqState->DStream); /* <= 9 bits */
- if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream); /* <= 18 bits */
- ZSTD_updateFseState(&seqState->stateOffb, &seqState->DStream); /* <= 8 bits */
-
- return seq;
-}
-
-FORCE_INLINE_TEMPLATE size_t
-ZSTD_decompressSequencesLong_body(
- ZSTD_DCtx* dctx,
- void* dst, size_t maxDstSize,
- const void* seqStart, size_t seqSize, int nbSeq,
- const ZSTD_longOffset_e isLongOffset)
-{
- const BYTE* ip = (const BYTE*)seqStart;
- const BYTE* const iend = ip + seqSize;
- BYTE* const ostart = (BYTE* const)dst;
- BYTE* const oend = ostart + maxDstSize;
- BYTE* op = ostart;
- const BYTE* litPtr = dctx->litPtr;
- const BYTE* const litEnd = litPtr + dctx->litSize;
- const BYTE* const prefixStart = (const BYTE*) (dctx->base);
- const BYTE* const dictStart = (const BYTE*) (dctx->vBase);
- const BYTE* const dictEnd = (const BYTE*) (dctx->dictEnd);
-
- /* Regen sequences */
- if (nbSeq) {
-#define STORED_SEQS 4
-#define STOSEQ_MASK (STORED_SEQS-1)
-#define ADVANCED_SEQS 4
- seq_t sequences[STORED_SEQS];
- int const seqAdvance = MIN(nbSeq, ADVANCED_SEQS);
- seqState_t seqState;
- int seqNb;
- dctx->fseEntropy = 1;
- { U32 i; for (i=0; i<ZSTD_REP_NUM; i++) seqState.prevOffset[i] = dctx->entropy.rep[i]; }
- seqState.prefixStart = prefixStart;
- seqState.pos = (size_t)(op-prefixStart);
- seqState.dictEnd = dictEnd;
- CHECK_E(BIT_initDStream(&seqState.DStream, ip, iend-ip), corruption_detected);
- ZSTD_initFseState(&seqState.stateLL, &seqState.DStream, dctx->LLTptr);
- ZSTD_initFseState(&seqState.stateOffb, &seqState.DStream, dctx->OFTptr);
- ZSTD_initFseState(&seqState.stateML, &seqState.DStream, dctx->MLTptr);
-
- /* prepare in advance */
- for (seqNb=0; (BIT_reloadDStream(&seqState.DStream) <= BIT_DStream_completed) && (seqNb<seqAdvance); seqNb++) {
- sequences[seqNb] = ZSTD_decodeSequenceLong(&seqState, isLongOffset);
- }
- if (seqNb<seqAdvance) return ERROR(corruption_detected);
-
- /* decode and decompress */
- for ( ; (BIT_reloadDStream(&(seqState.DStream)) <= BIT_DStream_completed) && (seqNb<nbSeq) ; seqNb++) {
- seq_t const sequence = ZSTD_decodeSequenceLong(&seqState, isLongOffset);
- size_t const oneSeqSize = ZSTD_execSequenceLong(op, oend, sequences[(seqNb-ADVANCED_SEQS) & STOSEQ_MASK], &litPtr, litEnd, prefixStart, dictStart, dictEnd);
- if (ZSTD_isError(oneSeqSize)) return oneSeqSize;
- PREFETCH(sequence.match); /* note : it's safe to invoke PREFETCH() on any memory address, including invalid ones */
- sequences[seqNb&STOSEQ_MASK] = sequence;
- op += oneSeqSize;
- }
- if (seqNb<nbSeq) return ERROR(corruption_detected);
-
- /* finish queue */
- seqNb -= seqAdvance;
- for ( ; seqNb<nbSeq ; seqNb++) {
- size_t const oneSeqSize = ZSTD_execSequenceLong(op, oend, sequences[seqNb&STOSEQ_MASK], &litPtr, litEnd, prefixStart, dictStart, dictEnd);
- if (ZSTD_isError(oneSeqSize)) return oneSeqSize;
- op += oneSeqSize;
- }
-
- /* save reps for next block */
- { U32 i; for (i=0; i<ZSTD_REP_NUM; i++) dctx->entropy.rep[i] = (U32)(seqState.prevOffset[i]); }
-#undef STORED_SEQS
-#undef STOSEQ_MASK
-#undef ADVANCED_SEQS
- }
-
- /* last literal segment */
- { size_t const lastLLSize = litEnd - litPtr;
- if (lastLLSize > (size_t)(oend-op)) return ERROR(dstSize_tooSmall);
- memcpy(op, litPtr, lastLLSize);
- op += lastLLSize;
- }
-
- return op-ostart;
-}
-
-static size_t
-ZSTD_decompressSequencesLong_default(ZSTD_DCtx* dctx,
- void* dst, size_t maxDstSize,
- const void* seqStart, size_t seqSize, int nbSeq,
- const ZSTD_longOffset_e isLongOffset)
-{
- return ZSTD_decompressSequencesLong_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
-}
-
-
-
-#if DYNAMIC_BMI2
-
-static TARGET_ATTRIBUTE("bmi2") size_t
-ZSTD_decompressSequences_bmi2(ZSTD_DCtx* dctx,
- void* dst, size_t maxDstSize,
- const void* seqStart, size_t seqSize, int nbSeq,
- const ZSTD_longOffset_e isLongOffset)
-{
- return ZSTD_decompressSequences_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
-}
-
-static TARGET_ATTRIBUTE("bmi2") size_t
-ZSTD_decompressSequencesLong_bmi2(ZSTD_DCtx* dctx,
- void* dst, size_t maxDstSize,
- const void* seqStart, size_t seqSize, int nbSeq,
- const ZSTD_longOffset_e isLongOffset)
-{
- return ZSTD_decompressSequencesLong_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
-}
-
+#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1)
+ if (ZSTD_isLegacy(src, srcSize))
+ return ZSTD_findFrameSizeInfoLegacy(src, srcSize);
#endif
-typedef size_t (*ZSTD_decompressSequences_t)(
- ZSTD_DCtx *dctx, void *dst, size_t maxDstSize,
- const void *seqStart, size_t seqSize, int nbSeq,
- const ZSTD_longOffset_e isLongOffset);
+ if ((srcSize >= ZSTD_SKIPPABLEHEADERSIZE)
+ && (MEM_readLE32(src) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) {
+ frameSizeInfo.compressedSize = readSkippableFrameSize(src, srcSize);
+ assert(ZSTD_isError(frameSizeInfo.compressedSize) ||
+ frameSizeInfo.compressedSize <= srcSize);
+ return frameSizeInfo;
+ } else {
+ const BYTE* ip = (const BYTE*)src;
+ const BYTE* const ipstart = ip;
+ size_t remainingSize = srcSize;
+ size_t nbBlocks = 0;
+ ZSTD_frameHeader zfh;
-static size_t ZSTD_decompressSequences(ZSTD_DCtx* dctx, void* dst, size_t maxDstSize,
- const void* seqStart, size_t seqSize, int nbSeq,
- const ZSTD_longOffset_e isLongOffset)
-{
- DEBUGLOG(5, "ZSTD_decompressSequences");
-#if DYNAMIC_BMI2
- if (dctx->bmi2) {
- return ZSTD_decompressSequences_bmi2(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
- }
-#endif
- return ZSTD_decompressSequences_default(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
-}
-
-static size_t ZSTD_decompressSequencesLong(ZSTD_DCtx* dctx,
- void* dst, size_t maxDstSize,
- const void* seqStart, size_t seqSize, int nbSeq,
- const ZSTD_longOffset_e isLongOffset)
-{
- DEBUGLOG(5, "ZSTD_decompressSequencesLong");
-#if DYNAMIC_BMI2
- if (dctx->bmi2) {
- return ZSTD_decompressSequencesLong_bmi2(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
- }
-#endif
- return ZSTD_decompressSequencesLong_default(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
-}
-
-/* ZSTD_getLongOffsetsShare() :
- * condition : offTable must be valid
- * @return : "share" of long offsets (arbitrarily defined as > (1<<23))
- * compared to maximum possible of (1<<OffFSELog) */
-static unsigned
-ZSTD_getLongOffsetsShare(const ZSTD_seqSymbol* offTable)
-{
- const void* ptr = offTable;
- U32 const tableLog = ((const ZSTD_seqSymbol_header*)ptr)[0].tableLog;
- const ZSTD_seqSymbol* table = offTable + 1;
- U32 const max = 1 << tableLog;
- U32 u, total = 0;
- DEBUGLOG(5, "ZSTD_getLongOffsetsShare: (tableLog=%u)", tableLog);
-
- assert(max <= (1 << OffFSELog)); /* max not too large */
- for (u=0; u<max; u++) {
- if (table[u].nbAdditionalBits > 22) total += 1;
- }
-
- assert(tableLog <= OffFSELog);
- total <<= (OffFSELog - tableLog); /* scale to OffFSELog */
-
- return total;
-}
-
-
-static size_t ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize, const int frame)
-{ /* blockType == blockCompressed */
- const BYTE* ip = (const BYTE*)src;
- /* isLongOffset must be true if there are long offsets.
- * Offsets are long if they are larger than 2^STREAM_ACCUMULATOR_MIN.
- * We don't expect that to be the case in 64-bit mode.
- * In block mode, window size is not known, so we have to be conservative. (note: but it could be evaluated from current-lowLimit)
- */
- ZSTD_longOffset_e const isLongOffset = (ZSTD_longOffset_e)(MEM_32bits() && (!frame || dctx->fParams.windowSize > (1ULL << STREAM_ACCUMULATOR_MIN)));
- DEBUGLOG(5, "ZSTD_decompressBlock_internal (size : %u)", (U32)srcSize);
-
- if (srcSize >= ZSTD_BLOCKSIZE_MAX) return ERROR(srcSize_wrong);
-
- /* Decode literals section */
- { size_t const litCSize = ZSTD_decodeLiteralsBlock(dctx, src, srcSize);
- DEBUGLOG(5, "ZSTD_decodeLiteralsBlock : %u", (U32)litCSize);
- if (ZSTD_isError(litCSize)) return litCSize;
- ip += litCSize;
- srcSize -= litCSize;
- }
-
- /* Build Decoding Tables */
- { int nbSeq;
- size_t const seqHSize = ZSTD_decodeSeqHeaders(dctx, &nbSeq, ip, srcSize);
- if (ZSTD_isError(seqHSize)) return seqHSize;
- ip += seqHSize;
- srcSize -= seqHSize;
-
- if ( (!frame || dctx->fParams.windowSize > (1<<24))
- && (nbSeq>0) ) { /* could probably use a larger nbSeq limit */
- U32 const shareLongOffsets = ZSTD_getLongOffsetsShare(dctx->OFTptr);
- U32 const minShare = MEM_64bits() ? 7 : 20; /* heuristic values, correspond to 2.73% and 7.81% */
- if (shareLongOffsets >= minShare)
- return ZSTD_decompressSequencesLong(dctx, dst, dstCapacity, ip, srcSize, nbSeq, isLongOffset);
+ /* Extract Frame Header */
+ { size_t const ret = ZSTD_getFrameHeader(&zfh, src, srcSize);
+ if (ZSTD_isError(ret))
+ return ZSTD_errorFrameSizeInfo(ret);
+ if (ret > 0)
+ return ZSTD_errorFrameSizeInfo(ERROR(srcSize_wrong));
}
- return ZSTD_decompressSequences(dctx, dst, dstCapacity, ip, srcSize, nbSeq, isLongOffset);
+ ip += zfh.headerSize;
+ remainingSize -= zfh.headerSize;
+
+ /* Iterate over each block */
+ while (1) {
+ blockProperties_t blockProperties;
+ size_t const cBlockSize = ZSTD_getcBlockSize(ip, remainingSize, &blockProperties);
+ if (ZSTD_isError(cBlockSize))
+ return ZSTD_errorFrameSizeInfo(cBlockSize);
+
+ if (ZSTD_blockHeaderSize + cBlockSize > remainingSize)
+ return ZSTD_errorFrameSizeInfo(ERROR(srcSize_wrong));
+
+ ip += ZSTD_blockHeaderSize + cBlockSize;
+ remainingSize -= ZSTD_blockHeaderSize + cBlockSize;
+ nbBlocks++;
+
+ if (blockProperties.lastBlock) break;
+ }
+
+ /* Final frame content checksum */
+ if (zfh.checksumFlag) {
+ if (remainingSize < 4)
+ return ZSTD_errorFrameSizeInfo(ERROR(srcSize_wrong));
+ ip += 4;
+ }
+
+ frameSizeInfo.compressedSize = ip - ipstart;
+ frameSizeInfo.decompressedBound = (zfh.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN)
+ ? zfh.frameContentSize
+ : nbBlocks * zfh.blockSizeMax;
+ return frameSizeInfo;
}
}
-
-static void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst)
-{
- if (dst != dctx->previousDstEnd) { /* not contiguous */
- dctx->dictEnd = dctx->previousDstEnd;
- dctx->vBase = (const char*)dst - ((const char*)(dctx->previousDstEnd) - (const char*)(dctx->base));
- dctx->base = dst;
- dctx->previousDstEnd = dst;
- }
-}
-
-size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize)
-{
- size_t dSize;
- ZSTD_checkContinuity(dctx, dst);
- dSize = ZSTD_decompressBlock_internal(dctx, dst, dstCapacity, src, srcSize, /* frame */ 0);
- dctx->previousDstEnd = (char*)dst + dSize;
- return dSize;
-}
-
-
-/** ZSTD_insertBlock() :
- insert `src` block into `dctx` history. Useful to track uncompressed blocks. */
-ZSTDLIB_API size_t ZSTD_insertBlock(ZSTD_DCtx* dctx, const void* blockStart, size_t blockSize)
-{
- ZSTD_checkContinuity(dctx, blockStart);
- dctx->previousDstEnd = (const char*)blockStart + blockSize;
- return blockSize;
-}
-
-
-static size_t ZSTD_generateNxBytes(void* dst, size_t dstCapacity, BYTE byte, size_t length)
-{
- if (length > dstCapacity) return ERROR(dstSize_tooSmall);
- memset(dst, byte, length);
- return length;
-}
-
/** ZSTD_findFrameCompressedSize() :
* compatible with legacy mode
* `src` must point to the start of a ZSTD frame, ZSTD legacy frame, or skippable frame
@@ -1743,55 +529,91 @@
* @return : the compressed size of the frame starting at `src` */
size_t ZSTD_findFrameCompressedSize(const void *src, size_t srcSize)
{
-#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1)
- if (ZSTD_isLegacy(src, srcSize))
- return ZSTD_findFrameCompressedSizeLegacy(src, srcSize);
-#endif
- if ( (srcSize >= ZSTD_skippableHeaderSize)
- && (MEM_readLE32(src) & 0xFFFFFFF0U) == ZSTD_MAGIC_SKIPPABLE_START ) {
- return ZSTD_skippableHeaderSize + MEM_readLE32((const BYTE*)src + ZSTD_frameIdSize);
- } else {
- const BYTE* ip = (const BYTE*)src;
- const BYTE* const ipstart = ip;
- size_t remainingSize = srcSize;
- ZSTD_frameHeader zfh;
+ ZSTD_frameSizeInfo const frameSizeInfo = ZSTD_findFrameSizeInfo(src, srcSize);
+ return frameSizeInfo.compressedSize;
+}
- /* Extract Frame Header */
- { size_t const ret = ZSTD_getFrameHeader(&zfh, src, srcSize);
- if (ZSTD_isError(ret)) return ret;
- if (ret > 0) return ERROR(srcSize_wrong);
- }
+/** ZSTD_decompressBound() :
+ * compatible with legacy mode
+ * `src` must point to the start of a ZSTD frame or a skippeable frame
+ * `srcSize` must be at least as large as the frame contained
+ * @return : the maximum decompressed size of the compressed source
+ */
+unsigned long long ZSTD_decompressBound(const void* src, size_t srcSize)
+{
+ unsigned long long bound = 0;
+ /* Iterate over each frame */
+ while (srcSize > 0) {
+ ZSTD_frameSizeInfo const frameSizeInfo = ZSTD_findFrameSizeInfo(src, srcSize);
+ size_t const compressedSize = frameSizeInfo.compressedSize;
+ unsigned long long const decompressedBound = frameSizeInfo.decompressedBound;
+ if (ZSTD_isError(compressedSize) || decompressedBound == ZSTD_CONTENTSIZE_ERROR)
+ return ZSTD_CONTENTSIZE_ERROR;
+ assert(srcSize >= compressedSize);
+ src = (const BYTE*)src + compressedSize;
+ srcSize -= compressedSize;
+ bound += decompressedBound;
+ }
+ return bound;
+}
- ip += zfh.headerSize;
- remainingSize -= zfh.headerSize;
- /* Loop on each block */
- while (1) {
- blockProperties_t blockProperties;
- size_t const cBlockSize = ZSTD_getcBlockSize(ip, remainingSize, &blockProperties);
- if (ZSTD_isError(cBlockSize)) return cBlockSize;
+/*-*************************************************************
+ * Frame decoding
+ ***************************************************************/
- if (ZSTD_blockHeaderSize + cBlockSize > remainingSize)
- return ERROR(srcSize_wrong);
- ip += ZSTD_blockHeaderSize + cBlockSize;
- remainingSize -= ZSTD_blockHeaderSize + cBlockSize;
-
- if (blockProperties.lastBlock) break;
- }
-
- if (zfh.checksumFlag) { /* Final frame content checksum */
- if (remainingSize < 4) return ERROR(srcSize_wrong);
- ip += 4;
- remainingSize -= 4;
- }
-
- return ip - ipstart;
+void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst)
+{
+ if (dst != dctx->previousDstEnd) { /* not contiguous */
+ dctx->dictEnd = dctx->previousDstEnd;
+ dctx->virtualStart = (const char*)dst - ((const char*)(dctx->previousDstEnd) - (const char*)(dctx->prefixStart));
+ dctx->prefixStart = dst;
+ dctx->previousDstEnd = dst;
}
}
+/** ZSTD_insertBlock() :
+ insert `src` block into `dctx` history. Useful to track uncompressed blocks. */
+size_t ZSTD_insertBlock(ZSTD_DCtx* dctx, const void* blockStart, size_t blockSize)
+{
+ ZSTD_checkContinuity(dctx, blockStart);
+ dctx->previousDstEnd = (const char*)blockStart + blockSize;
+ return blockSize;
+}
+
+
+static size_t ZSTD_copyRawBlock(void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize)
+{
+ DEBUGLOG(5, "ZSTD_copyRawBlock");
+ if (dst == NULL) {
+ if (srcSize == 0) return 0;
+ RETURN_ERROR(dstBuffer_null);
+ }
+ RETURN_ERROR_IF(srcSize > dstCapacity, dstSize_tooSmall);
+ memcpy(dst, src, srcSize);
+ return srcSize;
+}
+
+static size_t ZSTD_setRleBlock(void* dst, size_t dstCapacity,
+ BYTE b,
+ size_t regenSize)
+{
+ if (dst == NULL) {
+ if (regenSize == 0) return 0;
+ RETURN_ERROR(dstBuffer_null);
+ }
+ RETURN_ERROR_IF(regenSize > dstCapacity, dstSize_tooSmall);
+ memset(dst, b, regenSize);
+ return regenSize;
+}
+
+
/*! ZSTD_decompressFrame() :
-* @dctx must be properly initialized */
+ * @dctx must be properly initialized
+ * will update *srcPtr and *srcSizePtr,
+ * to make *srcPtr progress by one frame. */
static size_t ZSTD_decompressFrame(ZSTD_DCtx* dctx,
void* dst, size_t dstCapacity,
const void** srcPtr, size_t *srcSizePtr)
@@ -1800,31 +622,34 @@
BYTE* const ostart = (BYTE* const)dst;
BYTE* const oend = ostart + dstCapacity;
BYTE* op = ostart;
- size_t remainingSize = *srcSizePtr;
+ size_t remainingSrcSize = *srcSizePtr;
+
+ DEBUGLOG(4, "ZSTD_decompressFrame (srcSize:%i)", (int)*srcSizePtr);
/* check */
- if (remainingSize < ZSTD_frameHeaderSize_min+ZSTD_blockHeaderSize)
- return ERROR(srcSize_wrong);
+ RETURN_ERROR_IF(
+ remainingSrcSize < ZSTD_FRAMEHEADERSIZE_MIN+ZSTD_blockHeaderSize,
+ srcSize_wrong);
/* Frame Header */
- { size_t const frameHeaderSize = ZSTD_frameHeaderSize(ip, ZSTD_frameHeaderSize_prefix);
+ { size_t const frameHeaderSize = ZSTD_frameHeaderSize(ip, ZSTD_FRAMEHEADERSIZE_PREFIX);
if (ZSTD_isError(frameHeaderSize)) return frameHeaderSize;
- if (remainingSize < frameHeaderSize+ZSTD_blockHeaderSize)
- return ERROR(srcSize_wrong);
- CHECK_F( ZSTD_decodeFrameHeader(dctx, ip, frameHeaderSize) );
- ip += frameHeaderSize; remainingSize -= frameHeaderSize;
+ RETURN_ERROR_IF(remainingSrcSize < frameHeaderSize+ZSTD_blockHeaderSize,
+ srcSize_wrong);
+ FORWARD_IF_ERROR( ZSTD_decodeFrameHeader(dctx, ip, frameHeaderSize) );
+ ip += frameHeaderSize; remainingSrcSize -= frameHeaderSize;
}
/* Loop on each block */
while (1) {
size_t decodedSize;
blockProperties_t blockProperties;
- size_t const cBlockSize = ZSTD_getcBlockSize(ip, remainingSize, &blockProperties);
+ size_t const cBlockSize = ZSTD_getcBlockSize(ip, remainingSrcSize, &blockProperties);
if (ZSTD_isError(cBlockSize)) return cBlockSize;
ip += ZSTD_blockHeaderSize;
- remainingSize -= ZSTD_blockHeaderSize;
- if (cBlockSize > remainingSize) return ERROR(srcSize_wrong);
+ remainingSrcSize -= ZSTD_blockHeaderSize;
+ RETURN_ERROR_IF(cBlockSize > remainingSrcSize, srcSize_wrong);
switch(blockProperties.blockType)
{
@@ -1835,11 +660,11 @@
decodedSize = ZSTD_copyRawBlock(op, oend-op, ip, cBlockSize);
break;
case bt_rle :
- decodedSize = ZSTD_generateNxBytes(op, oend-op, *ip, blockProperties.origSize);
+ decodedSize = ZSTD_setRleBlock(op, oend-op, *ip, blockProperties.origSize);
break;
case bt_reserved :
default:
- return ERROR(corruption_detected);
+ RETURN_ERROR(corruption_detected);
}
if (ZSTD_isError(decodedSize)) return decodedSize;
@@ -1847,33 +672,30 @@
XXH64_update(&dctx->xxhState, op, decodedSize);
op += decodedSize;
ip += cBlockSize;
- remainingSize -= cBlockSize;
+ remainingSrcSize -= cBlockSize;
if (blockProperties.lastBlock) break;
}
if (dctx->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN) {
- if ((U64)(op-ostart) != dctx->fParams.frameContentSize) {
- return ERROR(corruption_detected);
- } }
+ RETURN_ERROR_IF((U64)(op-ostart) != dctx->fParams.frameContentSize,
+ corruption_detected);
+ }
if (dctx->fParams.checksumFlag) { /* Frame content checksum verification */
U32 const checkCalc = (U32)XXH64_digest(&dctx->xxhState);
U32 checkRead;
- if (remainingSize<4) return ERROR(checksum_wrong);
+ RETURN_ERROR_IF(remainingSrcSize<4, checksum_wrong);
checkRead = MEM_readLE32(ip);
- if (checkRead != checkCalc) return ERROR(checksum_wrong);
+ RETURN_ERROR_IF(checkRead != checkCalc, checksum_wrong);
ip += 4;
- remainingSize -= 4;
+ remainingSrcSize -= 4;
}
/* Allow caller to get size read */
*srcPtr = ip;
- *srcSizePtr = remainingSize;
+ *srcSizePtr = remainingSrcSize;
return op-ostart;
}
-static const void* ZSTD_DDictDictContent(const ZSTD_DDict* ddict);
-static size_t ZSTD_DDictDictSize(const ZSTD_DDict* ddict);
-
static size_t ZSTD_decompressMultiFrame(ZSTD_DCtx* dctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
@@ -1881,26 +703,30 @@
const ZSTD_DDict* ddict)
{
void* const dststart = dst;
+ int moreThan1Frame = 0;
+
+ DEBUGLOG(5, "ZSTD_decompressMultiFrame");
assert(dict==NULL || ddict==NULL); /* either dict or ddict set, not both */
if (ddict) {
- dict = ZSTD_DDictDictContent(ddict);
- dictSize = ZSTD_DDictDictSize(ddict);
+ dict = ZSTD_DDict_dictContent(ddict);
+ dictSize = ZSTD_DDict_dictSize(ddict);
}
- while (srcSize >= ZSTD_frameHeaderSize_prefix) {
- U32 magicNumber;
+ while (srcSize >= ZSTD_FRAMEHEADERSIZE_PREFIX) {
#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1)
if (ZSTD_isLegacy(src, srcSize)) {
size_t decodedSize;
size_t const frameSize = ZSTD_findFrameCompressedSizeLegacy(src, srcSize);
if (ZSTD_isError(frameSize)) return frameSize;
- /* legacy support is not compatible with static dctx */
- if (dctx->staticSize) return ERROR(memory_allocation);
+ RETURN_ERROR_IF(dctx->staticSize, memory_allocation,
+ "legacy support is not compatible with static dctx");
decodedSize = ZSTD_decompressLegacy(dst, dstCapacity, src, frameSize, dict, dictSize);
+ if (ZSTD_isError(decodedSize)) return decodedSize;
+ assert(decodedSize <=- dstCapacity);
dst = (BYTE*)dst + decodedSize;
dstCapacity -= decodedSize;
@@ -1911,45 +737,53 @@
}
#endif
- magicNumber = MEM_readLE32(src);
- DEBUGLOG(4, "reading magic number %08X (expecting %08X)",
- (U32)magicNumber, (U32)ZSTD_MAGICNUMBER);
- if (magicNumber != ZSTD_MAGICNUMBER) {
- if ((magicNumber & 0xFFFFFFF0U) == ZSTD_MAGIC_SKIPPABLE_START) {
- size_t skippableSize;
- if (srcSize < ZSTD_skippableHeaderSize)
- return ERROR(srcSize_wrong);
- skippableSize = MEM_readLE32((const BYTE*)src + ZSTD_frameIdSize)
- + ZSTD_skippableHeaderSize;
- if (srcSize < skippableSize) return ERROR(srcSize_wrong);
+ { U32 const magicNumber = MEM_readLE32(src);
+ DEBUGLOG(4, "reading magic number %08X (expecting %08X)",
+ (unsigned)magicNumber, ZSTD_MAGICNUMBER);
+ if ((magicNumber & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) {
+ size_t const skippableSize = readSkippableFrameSize(src, srcSize);
+ FORWARD_IF_ERROR(skippableSize);
+ assert(skippableSize <= srcSize);
src = (const BYTE *)src + skippableSize;
srcSize -= skippableSize;
continue;
- }
- return ERROR(prefix_unknown);
- }
+ } }
if (ddict) {
/* we were called from ZSTD_decompress_usingDDict */
- CHECK_F(ZSTD_decompressBegin_usingDDict(dctx, ddict));
+ FORWARD_IF_ERROR(ZSTD_decompressBegin_usingDDict(dctx, ddict));
} else {
/* this will initialize correctly with no dict if dict == NULL, so
* use this in all cases but ddict */
- CHECK_F(ZSTD_decompressBegin_usingDict(dctx, dict, dictSize));
+ FORWARD_IF_ERROR(ZSTD_decompressBegin_usingDict(dctx, dict, dictSize));
}
ZSTD_checkContinuity(dctx, dst);
{ const size_t res = ZSTD_decompressFrame(dctx, dst, dstCapacity,
&src, &srcSize);
+ RETURN_ERROR_IF(
+ (ZSTD_getErrorCode(res) == ZSTD_error_prefix_unknown)
+ && (moreThan1Frame==1),
+ srcSize_wrong,
+ "at least one frame successfully completed, but following "
+ "bytes are garbage: it's more likely to be a srcSize error, "
+ "specifying more bytes than compressed size of frame(s). This "
+ "error message replaces ERROR(prefix_unknown), which would be "
+ "confusing, as the first header is actually correct. Note that "
+ "one could be unlucky, it might be a corruption error instead, "
+ "happening right at the place where we expect zstd magic "
+ "bytes. But this is _much_ less likely than a srcSize field "
+ "error.");
if (ZSTD_isError(res)) return res;
- /* no need to bound check, ZSTD_decompressFrame already has */
+ assert(res <= dstCapacity);
dst = (BYTE*)dst + res;
dstCapacity -= res;
}
+ moreThan1Frame = 1;
} /* while (srcSize >= ZSTD_frameHeaderSize_prefix) */
- if (srcSize) return ERROR(srcSize_wrong); /* input not entirely consumed */
+ RETURN_ERROR_IF(srcSize, srcSize_wrong, "input not entirely consumed");
return (BYTE*)dst - (BYTE*)dststart;
}
@@ -1963,9 +797,26 @@
}
+static ZSTD_DDict const* ZSTD_getDDict(ZSTD_DCtx* dctx)
+{
+ switch (dctx->dictUses) {
+ default:
+ assert(0 /* Impossible */);
+ /* fall-through */
+ case ZSTD_dont_use:
+ ZSTD_clearDict(dctx);
+ return NULL;
+ case ZSTD_use_indefinitely:
+ return dctx->ddict;
+ case ZSTD_use_once:
+ dctx->dictUses = ZSTD_dont_use;
+ return dctx->ddict;
+ }
+}
+
size_t ZSTD_decompressDCtx(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
- return ZSTD_decompress_usingDict(dctx, dst, dstCapacity, src, srcSize, NULL, 0);
+ return ZSTD_decompress_usingDDict(dctx, dst, dstCapacity, src, srcSize, ZSTD_getDDict(dctx));
}
@@ -1974,12 +825,13 @@
#if defined(ZSTD_HEAPMODE) && (ZSTD_HEAPMODE>=1)
size_t regenSize;
ZSTD_DCtx* const dctx = ZSTD_createDCtx();
- if (dctx==NULL) return ERROR(memory_allocation);
+ RETURN_ERROR_IF(dctx==NULL, memory_allocation);
regenSize = ZSTD_decompressDCtx(dctx, dst, dstCapacity, src, srcSize);
ZSTD_freeDCtx(dctx);
return regenSize;
#else /* stack mode */
ZSTD_DCtx dctx;
+ ZSTD_initDCtx_internal(&dctx);
return ZSTD_decompressDCtx(&dctx, dst, dstCapacity, src, srcSize);
#endif
}
@@ -2021,9 +873,9 @@
* or an error code, which can be tested using ZSTD_isError() */
size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
- DEBUGLOG(5, "ZSTD_decompressContinue (srcSize:%u)", (U32)srcSize);
+ DEBUGLOG(5, "ZSTD_decompressContinue (srcSize:%u)", (unsigned)srcSize);
/* Sanity check */
- if (srcSize != dctx->expected) return ERROR(srcSize_wrong); /* not allowed */
+ RETURN_ERROR_IF(srcSize != dctx->expected, srcSize_wrong, "not allowed");
if (dstCapacity) ZSTD_checkContinuity(dctx, dst);
switch (dctx->stage)
@@ -2031,10 +883,10 @@
case ZSTDds_getFrameHeaderSize :
assert(src != NULL);
if (dctx->format == ZSTD_f_zstd1) { /* allows header */
- assert(srcSize >= ZSTD_frameIdSize); /* to read skippable magic number */
- if ((MEM_readLE32(src) & 0xFFFFFFF0U) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame */
+ assert(srcSize >= ZSTD_FRAMEIDSIZE); /* to read skippable magic number */
+ if ((MEM_readLE32(src) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame */
memcpy(dctx->headerBuffer, src, srcSize);
- dctx->expected = ZSTD_skippableHeaderSize - srcSize; /* remaining to load to get full skippable frame header */
+ dctx->expected = ZSTD_SKIPPABLEHEADERSIZE - srcSize; /* remaining to load to get full skippable frame header */
dctx->stage = ZSTDds_decodeSkippableHeader;
return 0;
} }
@@ -2048,7 +900,7 @@
case ZSTDds_decodeFrameHeader:
assert(src != NULL);
memcpy(dctx->headerBuffer + (dctx->headerSize - srcSize), src, srcSize);
- CHECK_F(ZSTD_decodeFrameHeader(dctx, dctx->headerBuffer, dctx->headerSize));
+ FORWARD_IF_ERROR(ZSTD_decodeFrameHeader(dctx, dctx->headerBuffer, dctx->headerSize));
dctx->expected = ZSTD_blockHeaderSize;
dctx->stage = ZSTDds_decodeBlockHeader;
return 0;
@@ -2094,23 +946,23 @@
rSize = ZSTD_copyRawBlock(dst, dstCapacity, src, srcSize);
break;
case bt_rle :
- rSize = ZSTD_setRleBlock(dst, dstCapacity, src, srcSize, dctx->rleSize);
+ rSize = ZSTD_setRleBlock(dst, dstCapacity, *(const BYTE*)src, dctx->rleSize);
break;
case bt_reserved : /* should never happen */
default:
- return ERROR(corruption_detected);
+ RETURN_ERROR(corruption_detected);
}
if (ZSTD_isError(rSize)) return rSize;
- DEBUGLOG(5, "ZSTD_decompressContinue: decoded size from block : %u", (U32)rSize);
+ DEBUGLOG(5, "ZSTD_decompressContinue: decoded size from block : %u", (unsigned)rSize);
dctx->decodedSize += rSize;
if (dctx->fParams.checksumFlag) XXH64_update(&dctx->xxhState, dst, rSize);
if (dctx->stage == ZSTDds_decompressLastBlock) { /* end of frame */
- DEBUGLOG(4, "ZSTD_decompressContinue: decoded size from frame : %u", (U32)dctx->decodedSize);
- if (dctx->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN) {
- if (dctx->decodedSize != dctx->fParams.frameContentSize) {
- return ERROR(corruption_detected);
- } }
+ DEBUGLOG(4, "ZSTD_decompressContinue: decoded size from frame : %u", (unsigned)dctx->decodedSize);
+ RETURN_ERROR_IF(
+ dctx->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN
+ && dctx->decodedSize != dctx->fParams.frameContentSize,
+ corruption_detected);
if (dctx->fParams.checksumFlag) { /* another round for frame checksum */
dctx->expected = 4;
dctx->stage = ZSTDds_checkChecksum;
@@ -2130,8 +982,8 @@
assert(srcSize == 4); /* guaranteed by dctx->expected */
{ U32 const h32 = (U32)XXH64_digest(&dctx->xxhState);
U32 const check32 = MEM_readLE32(src);
- DEBUGLOG(4, "ZSTD_decompressContinue: checksum : calculated %08X :: %08X read", h32, check32);
- if (check32 != h32) return ERROR(checksum_wrong);
+ DEBUGLOG(4, "ZSTD_decompressContinue: checksum : calculated %08X :: %08X read", (unsigned)h32, (unsigned)check32);
+ RETURN_ERROR_IF(check32 != h32, checksum_wrong);
dctx->expected = 0;
dctx->stage = ZSTDds_getFrameHeaderSize;
return 0;
@@ -2139,9 +991,9 @@
case ZSTDds_decodeSkippableHeader:
assert(src != NULL);
- assert(srcSize <= ZSTD_skippableHeaderSize);
- memcpy(dctx->headerBuffer + (ZSTD_skippableHeaderSize - srcSize), src, srcSize); /* complete skippable header */
- dctx->expected = MEM_readLE32(dctx->headerBuffer + ZSTD_frameIdSize); /* note : dctx->expected can grow seriously large, beyond local buffer size */
+ assert(srcSize <= ZSTD_SKIPPABLEHEADERSIZE);
+ memcpy(dctx->headerBuffer + (ZSTD_SKIPPABLEHEADERSIZE - srcSize), src, srcSize); /* complete skippable header */
+ dctx->expected = MEM_readLE32(dctx->headerBuffer + ZSTD_FRAMEIDSIZE); /* note : dctx->expected can grow seriously large, beyond local buffer size */
dctx->stage = ZSTDds_skipFrame;
return 0;
@@ -2151,7 +1003,8 @@
return 0;
default:
- return ERROR(GENERIC); /* impossible */
+ assert(0); /* impossible */
+ RETURN_ERROR(GENERIC); /* some compiler require default to do something */
}
}
@@ -2159,38 +1012,52 @@
static size_t ZSTD_refDictContent(ZSTD_DCtx* dctx, const void* dict, size_t dictSize)
{
dctx->dictEnd = dctx->previousDstEnd;
- dctx->vBase = (const char*)dict - ((const char*)(dctx->previousDstEnd) - (const char*)(dctx->base));
- dctx->base = dict;
+ dctx->virtualStart = (const char*)dict - ((const char*)(dctx->previousDstEnd) - (const char*)(dctx->prefixStart));
+ dctx->prefixStart = dict;
dctx->previousDstEnd = (const char*)dict + dictSize;
return 0;
}
-/* ZSTD_loadEntropy() :
- * dict : must point at beginning of a valid zstd dictionary
+/*! ZSTD_loadDEntropy() :
+ * dict : must point at beginning of a valid zstd dictionary.
* @return : size of entropy tables read */
-static size_t ZSTD_loadEntropy(ZSTD_entropyDTables_t* entropy, const void* const dict, size_t const dictSize)
+size_t
+ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy,
+ const void* const dict, size_t const dictSize)
{
const BYTE* dictPtr = (const BYTE*)dict;
const BYTE* const dictEnd = dictPtr + dictSize;
- if (dictSize <= 8) return ERROR(dictionary_corrupted);
+ RETURN_ERROR_IF(dictSize <= 8, dictionary_corrupted);
+ assert(MEM_readLE32(dict) == ZSTD_MAGIC_DICTIONARY); /* dict must be valid */
dictPtr += 8; /* skip header = magic + dictID */
-
- { size_t const hSize = HUF_readDTableX4_wksp(
- entropy->hufTable, dictPtr, dictEnd - dictPtr,
- entropy->workspace, sizeof(entropy->workspace));
- if (HUF_isError(hSize)) return ERROR(dictionary_corrupted);
+ ZSTD_STATIC_ASSERT(offsetof(ZSTD_entropyDTables_t, OFTable) == offsetof(ZSTD_entropyDTables_t, LLTable) + sizeof(entropy->LLTable));
+ ZSTD_STATIC_ASSERT(offsetof(ZSTD_entropyDTables_t, MLTable) == offsetof(ZSTD_entropyDTables_t, OFTable) + sizeof(entropy->OFTable));
+ ZSTD_STATIC_ASSERT(sizeof(entropy->LLTable) + sizeof(entropy->OFTable) + sizeof(entropy->MLTable) >= HUF_DECOMPRESS_WORKSPACE_SIZE);
+ { void* const workspace = &entropy->LLTable; /* use fse tables as temporary workspace; implies fse tables are grouped together */
+ size_t const workspaceSize = sizeof(entropy->LLTable) + sizeof(entropy->OFTable) + sizeof(entropy->MLTable);
+#ifdef HUF_FORCE_DECOMPRESS_X1
+ /* in minimal huffman, we always use X1 variants */
+ size_t const hSize = HUF_readDTableX1_wksp(entropy->hufTable,
+ dictPtr, dictEnd - dictPtr,
+ workspace, workspaceSize);
+#else
+ size_t const hSize = HUF_readDTableX2_wksp(entropy->hufTable,
+ dictPtr, dictEnd - dictPtr,
+ workspace, workspaceSize);
+#endif
+ RETURN_ERROR_IF(HUF_isError(hSize), dictionary_corrupted);
dictPtr += hSize;
}
{ short offcodeNCount[MaxOff+1];
- U32 offcodeMaxValue = MaxOff, offcodeLog;
+ unsigned offcodeMaxValue = MaxOff, offcodeLog;
size_t const offcodeHeaderSize = FSE_readNCount(offcodeNCount, &offcodeMaxValue, &offcodeLog, dictPtr, dictEnd-dictPtr);
- if (FSE_isError(offcodeHeaderSize)) return ERROR(dictionary_corrupted);
- if (offcodeMaxValue > MaxOff) return ERROR(dictionary_corrupted);
- if (offcodeLog > OffFSELog) return ERROR(dictionary_corrupted);
- ZSTD_buildFSETable(entropy->OFTable,
+ RETURN_ERROR_IF(FSE_isError(offcodeHeaderSize), dictionary_corrupted);
+ RETURN_ERROR_IF(offcodeMaxValue > MaxOff, dictionary_corrupted);
+ RETURN_ERROR_IF(offcodeLog > OffFSELog, dictionary_corrupted);
+ ZSTD_buildFSETable( entropy->OFTable,
offcodeNCount, offcodeMaxValue,
OF_base, OF_bits,
offcodeLog);
@@ -2200,10 +1067,10 @@
{ short matchlengthNCount[MaxML+1];
unsigned matchlengthMaxValue = MaxML, matchlengthLog;
size_t const matchlengthHeaderSize = FSE_readNCount(matchlengthNCount, &matchlengthMaxValue, &matchlengthLog, dictPtr, dictEnd-dictPtr);
- if (FSE_isError(matchlengthHeaderSize)) return ERROR(dictionary_corrupted);
- if (matchlengthMaxValue > MaxML) return ERROR(dictionary_corrupted);
- if (matchlengthLog > MLFSELog) return ERROR(dictionary_corrupted);
- ZSTD_buildFSETable(entropy->MLTable,
+ RETURN_ERROR_IF(FSE_isError(matchlengthHeaderSize), dictionary_corrupted);
+ RETURN_ERROR_IF(matchlengthMaxValue > MaxML, dictionary_corrupted);
+ RETURN_ERROR_IF(matchlengthLog > MLFSELog, dictionary_corrupted);
+ ZSTD_buildFSETable( entropy->MLTable,
matchlengthNCount, matchlengthMaxValue,
ML_base, ML_bits,
matchlengthLog);
@@ -2213,22 +1080,23 @@
{ short litlengthNCount[MaxLL+1];
unsigned litlengthMaxValue = MaxLL, litlengthLog;
size_t const litlengthHeaderSize = FSE_readNCount(litlengthNCount, &litlengthMaxValue, &litlengthLog, dictPtr, dictEnd-dictPtr);
- if (FSE_isError(litlengthHeaderSize)) return ERROR(dictionary_corrupted);
- if (litlengthMaxValue > MaxLL) return ERROR(dictionary_corrupted);
- if (litlengthLog > LLFSELog) return ERROR(dictionary_corrupted);
- ZSTD_buildFSETable(entropy->LLTable,
+ RETURN_ERROR_IF(FSE_isError(litlengthHeaderSize), dictionary_corrupted);
+ RETURN_ERROR_IF(litlengthMaxValue > MaxLL, dictionary_corrupted);
+ RETURN_ERROR_IF(litlengthLog > LLFSELog, dictionary_corrupted);
+ ZSTD_buildFSETable( entropy->LLTable,
litlengthNCount, litlengthMaxValue,
LL_base, LL_bits,
litlengthLog);
dictPtr += litlengthHeaderSize;
}
- if (dictPtr+12 > dictEnd) return ERROR(dictionary_corrupted);
+ RETURN_ERROR_IF(dictPtr+12 > dictEnd, dictionary_corrupted);
{ int i;
size_t const dictContentSize = (size_t)(dictEnd - (dictPtr+12));
for (i=0; i<3; i++) {
U32 const rep = MEM_readLE32(dictPtr); dictPtr += 4;
- if (rep==0 || rep >= dictContentSize) return ERROR(dictionary_corrupted);
+ RETURN_ERROR_IF(rep==0 || rep >= dictContentSize,
+ dictionary_corrupted);
entropy->rep[i] = rep;
} }
@@ -2242,11 +1110,11 @@
if (magic != ZSTD_MAGIC_DICTIONARY) {
return ZSTD_refDictContent(dctx, dict, dictSize); /* pure content mode */
} }
- dctx->dictID = MEM_readLE32((const char*)dict + ZSTD_frameIdSize);
+ dctx->dictID = MEM_readLE32((const char*)dict + ZSTD_FRAMEIDSIZE);
/* load entropy tables */
- { size_t const eSize = ZSTD_loadEntropy(&dctx->entropy, dict, dictSize);
- if (ZSTD_isError(eSize)) return ERROR(dictionary_corrupted);
+ { size_t const eSize = ZSTD_loadDEntropy(&dctx->entropy, dict, dictSize);
+ RETURN_ERROR_IF(ZSTD_isError(eSize), dictionary_corrupted);
dict = (const char*)dict + eSize;
dictSize -= eSize;
}
@@ -2256,7 +1124,6 @@
return ZSTD_refDictContent(dctx, dict, dictSize);
}
-/* Note : this function cannot fail */
size_t ZSTD_decompressBegin(ZSTD_DCtx* dctx)
{
assert(dctx != NULL);
@@ -2264,8 +1131,8 @@
dctx->stage = ZSTDds_getFrameHeaderSize;
dctx->decodedSize = 0;
dctx->previousDstEnd = NULL;
- dctx->base = NULL;
- dctx->vBase = NULL;
+ dctx->prefixStart = NULL;
+ dctx->virtualStart = NULL;
dctx->dictEnd = NULL;
dctx->entropy.hufTable[0] = (HUF_DTable)((HufLog)*0x1000001); /* cover both little and big endian */
dctx->litEntropy = dctx->fseEntropy = 0;
@@ -2281,201 +1148,36 @@
size_t ZSTD_decompressBegin_usingDict(ZSTD_DCtx* dctx, const void* dict, size_t dictSize)
{
- CHECK_F( ZSTD_decompressBegin(dctx) );
+ FORWARD_IF_ERROR( ZSTD_decompressBegin(dctx) );
if (dict && dictSize)
- CHECK_E(ZSTD_decompress_insertDictionary(dctx, dict, dictSize), dictionary_corrupted);
+ RETURN_ERROR_IF(
+ ZSTD_isError(ZSTD_decompress_insertDictionary(dctx, dict, dictSize)),
+ dictionary_corrupted);
return 0;
}
/* ====== ZSTD_DDict ====== */
-struct ZSTD_DDict_s {
- void* dictBuffer;
- const void* dictContent;
- size_t dictSize;
- ZSTD_entropyDTables_t entropy;
- U32 dictID;
- U32 entropyPresent;
- ZSTD_customMem cMem;
-}; /* typedef'd to ZSTD_DDict within "zstd.h" */
-
-static const void* ZSTD_DDictDictContent(const ZSTD_DDict* ddict)
+size_t ZSTD_decompressBegin_usingDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict)
{
- return ddict->dictContent;
-}
-
-static size_t ZSTD_DDictDictSize(const ZSTD_DDict* ddict)
-{
- return ddict->dictSize;
-}
-
-size_t ZSTD_decompressBegin_usingDDict(ZSTD_DCtx* dstDCtx, const ZSTD_DDict* ddict)
-{
- CHECK_F( ZSTD_decompressBegin(dstDCtx) );
- if (ddict) { /* support begin on NULL */
- dstDCtx->dictID = ddict->dictID;
- dstDCtx->base = ddict->dictContent;
- dstDCtx->vBase = ddict->dictContent;
- dstDCtx->dictEnd = (const BYTE*)ddict->dictContent + ddict->dictSize;
- dstDCtx->previousDstEnd = dstDCtx->dictEnd;
- if (ddict->entropyPresent) {
- dstDCtx->litEntropy = 1;
- dstDCtx->fseEntropy = 1;
- dstDCtx->LLTptr = ddict->entropy.LLTable;
- dstDCtx->MLTptr = ddict->entropy.MLTable;
- dstDCtx->OFTptr = ddict->entropy.OFTable;
- dstDCtx->HUFptr = ddict->entropy.hufTable;
- dstDCtx->entropy.rep[0] = ddict->entropy.rep[0];
- dstDCtx->entropy.rep[1] = ddict->entropy.rep[1];
- dstDCtx->entropy.rep[2] = ddict->entropy.rep[2];
- } else {
- dstDCtx->litEntropy = 0;
- dstDCtx->fseEntropy = 0;
- }
+ DEBUGLOG(4, "ZSTD_decompressBegin_usingDDict");
+ assert(dctx != NULL);
+ if (ddict) {
+ const char* const dictStart = (const char*)ZSTD_DDict_dictContent(ddict);
+ size_t const dictSize = ZSTD_DDict_dictSize(ddict);
+ const void* const dictEnd = dictStart + dictSize;
+ dctx->ddictIsCold = (dctx->dictEnd != dictEnd);
+ DEBUGLOG(4, "DDict is %s",
+ dctx->ddictIsCold ? "~cold~" : "hot!");
+ }
+ FORWARD_IF_ERROR( ZSTD_decompressBegin(dctx) );
+ if (ddict) { /* NULL ddict is equivalent to no dictionary */
+ ZSTD_copyDDictParameters(dctx, ddict);
}
return 0;
}
-static size_t ZSTD_loadEntropy_inDDict(ZSTD_DDict* ddict, ZSTD_dictContentType_e dictContentType)
-{
- ddict->dictID = 0;
- ddict->entropyPresent = 0;
- if (dictContentType == ZSTD_dct_rawContent) return 0;
-
- if (ddict->dictSize < 8) {
- if (dictContentType == ZSTD_dct_fullDict)
- return ERROR(dictionary_corrupted); /* only accept specified dictionaries */
- return 0; /* pure content mode */
- }
- { U32 const magic = MEM_readLE32(ddict->dictContent);
- if (magic != ZSTD_MAGIC_DICTIONARY) {
- if (dictContentType == ZSTD_dct_fullDict)
- return ERROR(dictionary_corrupted); /* only accept specified dictionaries */
- return 0; /* pure content mode */
- }
- }
- ddict->dictID = MEM_readLE32((const char*)ddict->dictContent + ZSTD_frameIdSize);
-
- /* load entropy tables */
- CHECK_E( ZSTD_loadEntropy(&ddict->entropy, ddict->dictContent, ddict->dictSize), dictionary_corrupted );
- ddict->entropyPresent = 1;
- return 0;
-}
-
-
-static size_t ZSTD_initDDict_internal(ZSTD_DDict* ddict,
- const void* dict, size_t dictSize,
- ZSTD_dictLoadMethod_e dictLoadMethod,
- ZSTD_dictContentType_e dictContentType)
-{
- if ((dictLoadMethod == ZSTD_dlm_byRef) || (!dict) || (!dictSize)) {
- ddict->dictBuffer = NULL;
- ddict->dictContent = dict;
- } else {
- void* const internalBuffer = ZSTD_malloc(dictSize, ddict->cMem);
- ddict->dictBuffer = internalBuffer;
- ddict->dictContent = internalBuffer;
- if (!internalBuffer) return ERROR(memory_allocation);
- memcpy(internalBuffer, dict, dictSize);
- }
- ddict->dictSize = dictSize;
- ddict->entropy.hufTable[0] = (HUF_DTable)((HufLog)*0x1000001); /* cover both little and big endian */
-
- /* parse dictionary content */
- CHECK_F( ZSTD_loadEntropy_inDDict(ddict, dictContentType) );
-
- return 0;
-}
-
-ZSTD_DDict* ZSTD_createDDict_advanced(const void* dict, size_t dictSize,
- ZSTD_dictLoadMethod_e dictLoadMethod,
- ZSTD_dictContentType_e dictContentType,
- ZSTD_customMem customMem)
-{
- if (!customMem.customAlloc ^ !customMem.customFree) return NULL;
-
- { ZSTD_DDict* const ddict = (ZSTD_DDict*) ZSTD_malloc(sizeof(ZSTD_DDict), customMem);
- if (!ddict) return NULL;
- ddict->cMem = customMem;
-
- if (ZSTD_isError( ZSTD_initDDict_internal(ddict, dict, dictSize, dictLoadMethod, dictContentType) )) {
- ZSTD_freeDDict(ddict);
- return NULL;
- }
-
- return ddict;
- }
-}
-
-/*! ZSTD_createDDict() :
-* Create a digested dictionary, to start decompression without startup delay.
-* `dict` content is copied inside DDict.
-* Consequently, `dict` can be released after `ZSTD_DDict` creation */
-ZSTD_DDict* ZSTD_createDDict(const void* dict, size_t dictSize)
-{
- ZSTD_customMem const allocator = { NULL, NULL, NULL };
- return ZSTD_createDDict_advanced(dict, dictSize, ZSTD_dlm_byCopy, ZSTD_dct_auto, allocator);
-}
-
-/*! ZSTD_createDDict_byReference() :
- * Create a digested dictionary, to start decompression without startup delay.
- * Dictionary content is simply referenced, it will be accessed during decompression.
- * Warning : dictBuffer must outlive DDict (DDict must be freed before dictBuffer) */
-ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize)
-{
- ZSTD_customMem const allocator = { NULL, NULL, NULL };
- return ZSTD_createDDict_advanced(dictBuffer, dictSize, ZSTD_dlm_byRef, ZSTD_dct_auto, allocator);
-}
-
-
-const ZSTD_DDict* ZSTD_initStaticDDict(
- void* workspace, size_t workspaceSize,
- const void* dict, size_t dictSize,
- ZSTD_dictLoadMethod_e dictLoadMethod,
- ZSTD_dictContentType_e dictContentType)
-{
- size_t const neededSpace =
- sizeof(ZSTD_DDict) + (dictLoadMethod == ZSTD_dlm_byRef ? 0 : dictSize);
- ZSTD_DDict* const ddict = (ZSTD_DDict*)workspace;
- assert(workspace != NULL);
- assert(dict != NULL);
- if ((size_t)workspace & 7) return NULL; /* 8-aligned */
- if (workspaceSize < neededSpace) return NULL;
- if (dictLoadMethod == ZSTD_dlm_byCopy) {
- memcpy(ddict+1, dict, dictSize); /* local copy */
- dict = ddict+1;
- }
- if (ZSTD_isError( ZSTD_initDDict_internal(ddict, dict, dictSize, ZSTD_dlm_byRef, dictContentType) ))
- return NULL;
- return ddict;
-}
-
-
-size_t ZSTD_freeDDict(ZSTD_DDict* ddict)
-{
- if (ddict==NULL) return 0; /* support free on NULL */
- { ZSTD_customMem const cMem = ddict->cMem;
- ZSTD_free(ddict->dictBuffer, cMem);
- ZSTD_free(ddict, cMem);
- return 0;
- }
-}
-
-/*! ZSTD_estimateDDictSize() :
- * Estimate amount of memory that will be needed to create a dictionary for decompression.
- * Note : dictionary created by reference using ZSTD_dlm_byRef are smaller */
-size_t ZSTD_estimateDDictSize(size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod)
-{
- return sizeof(ZSTD_DDict) + (dictLoadMethod == ZSTD_dlm_byRef ? 0 : dictSize);
-}
-
-size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict)
-{
- if (ddict==NULL) return 0; /* support sizeof on NULL */
- return sizeof(*ddict) + (ddict->dictBuffer ? ddict->dictSize : 0) ;
-}
-
/*! ZSTD_getDictID_fromDict() :
* Provides the dictID stored within dictionary.
* if @return == 0, the dictionary is not conformant with Zstandard specification.
@@ -2484,21 +1186,11 @@
{
if (dictSize < 8) return 0;
if (MEM_readLE32(dict) != ZSTD_MAGIC_DICTIONARY) return 0;
- return MEM_readLE32((const char*)dict + ZSTD_frameIdSize);
-}
-
-/*! ZSTD_getDictID_fromDDict() :
- * Provides the dictID of the dictionary loaded into `ddict`.
- * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty.
- * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */
-unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict)
-{
- if (ddict==NULL) return 0;
- return ZSTD_getDictID_fromDict(ddict->dictContent, ddict->dictSize);
+ return MEM_readLE32((const char*)dict + ZSTD_FRAMEIDSIZE);
}
/*! ZSTD_getDictID_fromFrame() :
- * Provides the dictID required to decompresse frame stored within `src`.
+ * Provides the dictID required to decompress frame stored within `src`.
* If @return == 0, the dictID could not be decoded.
* This could for one of the following reasons :
* - The frame does not require a dictionary (most common case).
@@ -2560,22 +1252,24 @@
}
-/* *** Initialization *** */
+/* *** Initialization *** */
size_t ZSTD_DStreamInSize(void) { return ZSTD_BLOCKSIZE_MAX + ZSTD_blockHeaderSize; }
size_t ZSTD_DStreamOutSize(void) { return ZSTD_BLOCKSIZE_MAX; }
-size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType)
+size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx,
+ const void* dict, size_t dictSize,
+ ZSTD_dictLoadMethod_e dictLoadMethod,
+ ZSTD_dictContentType_e dictContentType)
{
- if (dctx->streamStage != zdss_init) return ERROR(stage_wrong);
- ZSTD_freeDDict(dctx->ddictLocal);
+ RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong);
+ ZSTD_clearDict(dctx);
if (dict && dictSize >= 8) {
dctx->ddictLocal = ZSTD_createDDict_advanced(dict, dictSize, dictLoadMethod, dictContentType, dctx->customMem);
- if (dctx->ddictLocal == NULL) return ERROR(memory_allocation);
- } else {
- dctx->ddictLocal = NULL;
+ RETURN_ERROR_IF(dctx->ddictLocal == NULL, memory_allocation);
+ dctx->ddict = dctx->ddictLocal;
+ dctx->dictUses = ZSTD_use_indefinitely;
}
- dctx->ddict = dctx->ddictLocal;
return 0;
}
@@ -2591,7 +1285,9 @@
size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType)
{
- return ZSTD_DCtx_loadDictionary_advanced(dctx, prefix, prefixSize, ZSTD_dlm_byRef, dictContentType);
+ FORWARD_IF_ERROR(ZSTD_DCtx_loadDictionary_advanced(dctx, prefix, prefixSize, ZSTD_dlm_byRef, dictContentType));
+ dctx->dictUses = ZSTD_use_once;
+ return 0;
}
size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize)
@@ -2601,28 +1297,21 @@
/* ZSTD_initDStream_usingDict() :
- * return : expected size, aka ZSTD_frameHeaderSize_prefix.
+ * return : expected size, aka ZSTD_FRAMEHEADERSIZE_PREFIX.
* this function cannot fail */
size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize)
{
DEBUGLOG(4, "ZSTD_initDStream_usingDict");
- zds->streamStage = zdss_init;
- CHECK_F( ZSTD_DCtx_loadDictionary(zds, dict, dictSize) );
- return ZSTD_frameHeaderSize_prefix;
+ FORWARD_IF_ERROR( ZSTD_DCtx_reset(zds, ZSTD_reset_session_only) );
+ FORWARD_IF_ERROR( ZSTD_DCtx_loadDictionary(zds, dict, dictSize) );
+ return ZSTD_FRAMEHEADERSIZE_PREFIX;
}
/* note : this variant can't fail */
size_t ZSTD_initDStream(ZSTD_DStream* zds)
{
DEBUGLOG(4, "ZSTD_initDStream");
- return ZSTD_initDStream_usingDict(zds, NULL, 0);
-}
-
-size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict)
-{
- if (dctx->streamStage != zdss_init) return ERROR(stage_wrong);
- dctx->ddict = ddict;
- return 0;
+ return ZSTD_initDStream_usingDDict(zds, NULL);
}
/* ZSTD_initDStream_usingDDict() :
@@ -2630,51 +1319,119 @@
* this function cannot fail */
size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* dctx, const ZSTD_DDict* ddict)
{
- size_t const initResult = ZSTD_initDStream(dctx);
- dctx->ddict = ddict;
- return initResult;
+ FORWARD_IF_ERROR( ZSTD_DCtx_reset(dctx, ZSTD_reset_session_only) );
+ FORWARD_IF_ERROR( ZSTD_DCtx_refDDict(dctx, ddict) );
+ return ZSTD_FRAMEHEADERSIZE_PREFIX;
}
/* ZSTD_resetDStream() :
- * return : expected size, aka ZSTD_frameHeaderSize_prefix.
+ * return : expected size, aka ZSTD_FRAMEHEADERSIZE_PREFIX.
* this function cannot fail */
size_t ZSTD_resetDStream(ZSTD_DStream* dctx)
{
- DEBUGLOG(4, "ZSTD_resetDStream");
- dctx->streamStage = zdss_loadHeader;
- dctx->lhSize = dctx->inPos = dctx->outStart = dctx->outEnd = 0;
- dctx->legacyVersion = 0;
- dctx->hostageByte = 0;
- return ZSTD_frameHeaderSize_prefix;
+ FORWARD_IF_ERROR(ZSTD_DCtx_reset(dctx, ZSTD_reset_session_only));
+ return ZSTD_FRAMEHEADERSIZE_PREFIX;
}
-size_t ZSTD_setDStreamParameter(ZSTD_DStream* dctx,
- ZSTD_DStreamParameter_e paramType, unsigned paramValue)
+
+size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict)
{
- if (dctx->streamStage != zdss_init) return ERROR(stage_wrong);
- switch(paramType)
- {
- default : return ERROR(parameter_unsupported);
- case DStream_p_maxWindowSize :
- DEBUGLOG(4, "setting maxWindowSize = %u KB", paramValue >> 10);
- dctx->maxWindowSize = paramValue ? paramValue : (U32)(-1);
- break;
+ RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong);
+ ZSTD_clearDict(dctx);
+ if (ddict) {
+ dctx->ddict = ddict;
+ dctx->dictUses = ZSTD_use_indefinitely;
}
return 0;
}
+/* ZSTD_DCtx_setMaxWindowSize() :
+ * note : no direct equivalence in ZSTD_DCtx_setParameter,
+ * since this version sets windowSize, and the other sets windowLog */
size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize)
{
- if (dctx->streamStage != zdss_init) return ERROR(stage_wrong);
+ ZSTD_bounds const bounds = ZSTD_dParam_getBounds(ZSTD_d_windowLogMax);
+ size_t const min = (size_t)1 << bounds.lowerBound;
+ size_t const max = (size_t)1 << bounds.upperBound;
+ RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong);
+ RETURN_ERROR_IF(maxWindowSize < min, parameter_outOfBound);
+ RETURN_ERROR_IF(maxWindowSize > max, parameter_outOfBound);
dctx->maxWindowSize = maxWindowSize;
return 0;
}
size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format)
{
- DEBUGLOG(4, "ZSTD_DCtx_setFormat : %u", (unsigned)format);
- if (dctx->streamStage != zdss_init) return ERROR(stage_wrong);
- dctx->format = format;
+ return ZSTD_DCtx_setParameter(dctx, ZSTD_d_format, format);
+}
+
+ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam)
+{
+ ZSTD_bounds bounds = { 0, 0, 0 };
+ switch(dParam) {
+ case ZSTD_d_windowLogMax:
+ bounds.lowerBound = ZSTD_WINDOWLOG_ABSOLUTEMIN;
+ bounds.upperBound = ZSTD_WINDOWLOG_MAX;
+ return bounds;
+ case ZSTD_d_format:
+ bounds.lowerBound = (int)ZSTD_f_zstd1;
+ bounds.upperBound = (int)ZSTD_f_zstd1_magicless;
+ ZSTD_STATIC_ASSERT(ZSTD_f_zstd1 < ZSTD_f_zstd1_magicless);
+ return bounds;
+ default:;
+ }
+ bounds.error = ERROR(parameter_unsupported);
+ return bounds;
+}
+
+/* ZSTD_dParam_withinBounds:
+ * @return 1 if value is within dParam bounds,
+ * 0 otherwise */
+static int ZSTD_dParam_withinBounds(ZSTD_dParameter dParam, int value)
+{
+ ZSTD_bounds const bounds = ZSTD_dParam_getBounds(dParam);
+ if (ZSTD_isError(bounds.error)) return 0;
+ if (value < bounds.lowerBound) return 0;
+ if (value > bounds.upperBound) return 0;
+ return 1;
+}
+
+#define CHECK_DBOUNDS(p,v) { \
+ RETURN_ERROR_IF(!ZSTD_dParam_withinBounds(p, v), parameter_outOfBound); \
+}
+
+size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter dParam, int value)
+{
+ RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong);
+ switch(dParam) {
+ case ZSTD_d_windowLogMax:
+ if (value == 0) value = ZSTD_WINDOWLOG_LIMIT_DEFAULT;
+ CHECK_DBOUNDS(ZSTD_d_windowLogMax, value);
+ dctx->maxWindowSize = ((size_t)1) << value;
+ return 0;
+ case ZSTD_d_format:
+ CHECK_DBOUNDS(ZSTD_d_format, value);
+ dctx->format = (ZSTD_format_e)value;
+ return 0;
+ default:;
+ }
+ RETURN_ERROR(parameter_unsupported);
+}
+
+size_t ZSTD_DCtx_reset(ZSTD_DCtx* dctx, ZSTD_ResetDirective reset)
+{
+ if ( (reset == ZSTD_reset_session_only)
+ || (reset == ZSTD_reset_session_and_parameters) ) {
+ dctx->streamStage = zdss_init;
+ dctx->noForwardProgress = 0;
+ }
+ if ( (reset == ZSTD_reset_parameters)
+ || (reset == ZSTD_reset_session_and_parameters) ) {
+ RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong);
+ ZSTD_clearDict(dctx);
+ dctx->format = ZSTD_f_zstd1;
+ dctx->maxWindowSize = ZSTD_MAXWINDOWSIZE_DEFAULT;
+ }
return 0;
}
@@ -2690,7 +1447,8 @@
unsigned long long const neededRBSize = windowSize + blockSize + (WILDCOPY_OVERLENGTH * 2);
unsigned long long const neededSize = MIN(frameContentSize, neededRBSize);
size_t const minRBSize = (size_t) neededSize;
- if ((unsigned long long)minRBSize != neededSize) return ERROR(frameParameter_windowTooLarge);
+ RETURN_ERROR_IF((unsigned long long)minRBSize != neededSize,
+ frameParameter_windowTooLarge);
return minRBSize;
}
@@ -2704,13 +1462,13 @@
size_t ZSTD_estimateDStreamSize_fromFrame(const void* src, size_t srcSize)
{
- U32 const windowSizeMax = 1U << ZSTD_WINDOWLOG_MAX; /* note : should be user-selectable */
+ U32 const windowSizeMax = 1U << ZSTD_WINDOWLOG_MAX; /* note : should be user-selectable, but requires an additional parameter (or a dctx) */
ZSTD_frameHeader zfh;
size_t const err = ZSTD_getFrameHeader(&zfh, src, srcSize);
if (ZSTD_isError(err)) return err;
- if (err>0) return ERROR(srcSize_wrong);
- if (zfh.windowSize > windowSizeMax)
- return ERROR(frameParameter_windowTooLarge);
+ RETURN_ERROR_IF(err>0, srcSize_wrong);
+ RETURN_ERROR_IF(zfh.windowSize > windowSizeMax,
+ frameParameter_windowTooLarge);
return ZSTD_estimateDStreamSize((size_t)zfh.windowSize);
}
@@ -2736,16 +1494,16 @@
U32 someMoreWork = 1;
DEBUGLOG(5, "ZSTD_decompressStream");
- if (input->pos > input->size) { /* forbidden */
- DEBUGLOG(5, "in: pos: %u vs size: %u",
- (U32)input->pos, (U32)input->size);
- return ERROR(srcSize_wrong);
- }
- if (output->pos > output->size) { /* forbidden */
- DEBUGLOG(5, "out: pos: %u vs size: %u",
- (U32)output->pos, (U32)output->size);
- return ERROR(dstSize_tooSmall);
- }
+ RETURN_ERROR_IF(
+ input->pos > input->size,
+ srcSize_wrong,
+ "forbidden. in: pos: %u vs size: %u",
+ (U32)input->pos, (U32)input->size);
+ RETURN_ERROR_IF(
+ output->pos > output->size,
+ dstSize_tooSmall,
+ "forbidden. out: pos: %u vs size: %u",
+ (U32)output->pos, (U32)output->size);
DEBUGLOG(5, "input size : %u", (U32)(input->size - input->pos));
while (someMoreWork) {
@@ -2753,32 +1511,36 @@
{
case zdss_init :
DEBUGLOG(5, "stage zdss_init => transparent reset ");
- ZSTD_resetDStream(zds); /* transparent reset on starting decoding a new frame */
+ zds->streamStage = zdss_loadHeader;
+ zds->lhSize = zds->inPos = zds->outStart = zds->outEnd = 0;
+ zds->legacyVersion = 0;
+ zds->hostageByte = 0;
/* fall-through */
case zdss_loadHeader :
DEBUGLOG(5, "stage zdss_loadHeader (srcSize : %u)", (U32)(iend - ip));
#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1)
if (zds->legacyVersion) {
- /* legacy support is incompatible with static dctx */
- if (zds->staticSize) return ERROR(memory_allocation);
+ RETURN_ERROR_IF(zds->staticSize, memory_allocation,
+ "legacy support is incompatible with static dctx");
{ size_t const hint = ZSTD_decompressLegacyStream(zds->legacyContext, zds->legacyVersion, output, input);
if (hint==0) zds->streamStage = zdss_init;
return hint;
} }
#endif
- { size_t const hSize = ZSTD_getFrameHeader_internal(&zds->fParams, zds->headerBuffer, zds->lhSize, zds->format);
+ { size_t const hSize = ZSTD_getFrameHeader_advanced(&zds->fParams, zds->headerBuffer, zds->lhSize, zds->format);
DEBUGLOG(5, "header size : %u", (U32)hSize);
if (ZSTD_isError(hSize)) {
#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1)
U32 const legacyVersion = ZSTD_isLegacy(istart, iend-istart);
if (legacyVersion) {
- const void* const dict = zds->ddict ? zds->ddict->dictContent : NULL;
- size_t const dictSize = zds->ddict ? zds->ddict->dictSize : 0;
+ ZSTD_DDict const* const ddict = ZSTD_getDDict(zds);
+ const void* const dict = ddict ? ZSTD_DDict_dictContent(ddict) : NULL;
+ size_t const dictSize = ddict ? ZSTD_DDict_dictSize(ddict) : 0;
DEBUGLOG(5, "ZSTD_decompressStream: detected legacy version v0.%u", legacyVersion);
- /* legacy support is incompatible with static dctx */
- if (zds->staticSize) return ERROR(memory_allocation);
- CHECK_F(ZSTD_initLegacyStream(&zds->legacyContext,
+ RETURN_ERROR_IF(zds->staticSize, memory_allocation,
+ "legacy support is incompatible with static dctx");
+ FORWARD_IF_ERROR(ZSTD_initLegacyStream(&zds->legacyContext,
zds->previousLegacyVersion, legacyVersion,
dict, dictSize));
zds->legacyVersion = zds->previousLegacyVersion = legacyVersion;
@@ -2799,7 +1561,7 @@
zds->lhSize += remainingInput;
}
input->pos = input->size;
- return (MAX(ZSTD_frameHeaderSize_min, hSize) - zds->lhSize) + ZSTD_blockHeaderSize; /* remaining header bytes + next block header */
+ return (MAX(ZSTD_FRAMEHEADERSIZE_MIN, hSize) - zds->lhSize) + ZSTD_blockHeaderSize; /* remaining header bytes + next block header */
}
assert(ip != NULL);
memcpy(zds->headerBuffer + zds->lhSize, ip, toLoad); zds->lhSize = hSize; ip += toLoad;
@@ -2812,7 +1574,7 @@
size_t const cSize = ZSTD_findFrameCompressedSize(istart, iend-istart);
if (cSize <= (size_t)(iend-istart)) {
/* shortcut : using single-pass mode */
- size_t const decompressedSize = ZSTD_decompress_usingDDict(zds, op, oend-op, istart, cSize, zds->ddict);
+ size_t const decompressedSize = ZSTD_decompress_usingDDict(zds, op, oend-op, istart, cSize, ZSTD_getDDict(zds));
if (ZSTD_isError(decompressedSize)) return decompressedSize;
DEBUGLOG(4, "shortcut to single-pass ZSTD_decompress_usingDDict()")
ip = istart + cSize;
@@ -2825,13 +1587,13 @@
/* Consume header (see ZSTDds_decodeFrameHeader) */
DEBUGLOG(4, "Consume header");
- CHECK_F(ZSTD_decompressBegin_usingDDict(zds, zds->ddict));
+ FORWARD_IF_ERROR(ZSTD_decompressBegin_usingDDict(zds, ZSTD_getDDict(zds)));
- if ((MEM_readLE32(zds->headerBuffer) & 0xFFFFFFF0U) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame */
- zds->expected = MEM_readLE32(zds->headerBuffer + ZSTD_frameIdSize);
+ if ((MEM_readLE32(zds->headerBuffer) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame */
+ zds->expected = MEM_readLE32(zds->headerBuffer + ZSTD_FRAMEIDSIZE);
zds->stage = ZSTDds_skipFrame;
} else {
- CHECK_F(ZSTD_decodeFrameHeader(zds, zds->headerBuffer, zds->lhSize));
+ FORWARD_IF_ERROR(ZSTD_decodeFrameHeader(zds, zds->headerBuffer, zds->lhSize));
zds->expected = ZSTD_blockHeaderSize;
zds->stage = ZSTDds_decodeBlockHeader;
}
@@ -2841,7 +1603,8 @@
(U32)(zds->fParams.windowSize >>10),
(U32)(zds->maxWindowSize >> 10) );
zds->fParams.windowSize = MAX(zds->fParams.windowSize, 1U << ZSTD_WINDOWLOG_ABSOLUTEMIN);
- if (zds->fParams.windowSize > zds->maxWindowSize) return ERROR(frameParameter_windowTooLarge);
+ RETURN_ERROR_IF(zds->fParams.windowSize > zds->maxWindowSize,
+ frameParameter_windowTooLarge);
/* Adapt buffer sizes to frame header instructions */
{ size_t const neededInBuffSize = MAX(zds->fParams.blockSizeMax, 4 /* frame checksum */);
@@ -2855,14 +1618,15 @@
if (zds->staticSize) { /* static DCtx */
DEBUGLOG(4, "staticSize : %u", (U32)zds->staticSize);
assert(zds->staticSize >= sizeof(ZSTD_DCtx)); /* controlled at init */
- if (bufferSize > zds->staticSize - sizeof(ZSTD_DCtx))
- return ERROR(memory_allocation);
+ RETURN_ERROR_IF(
+ bufferSize > zds->staticSize - sizeof(ZSTD_DCtx),
+ memory_allocation);
} else {
ZSTD_free(zds->inBuff, zds->customMem);
zds->inBuffSize = 0;
zds->outBuffSize = 0;
zds->inBuff = (char*)ZSTD_malloc(bufferSize, zds->customMem);
- if (zds->inBuff == NULL) return ERROR(memory_allocation);
+ RETURN_ERROR_IF(zds->inBuff == NULL, memory_allocation);
}
zds->inBuffSize = neededInBuffSize;
zds->outBuff = zds->inBuff + zds->inBuffSize;
@@ -2904,7 +1668,9 @@
if (isSkipFrame) {
loadedSize = MIN(toLoad, (size_t)(iend-ip));
} else {
- if (toLoad > zds->inBuffSize - zds->inPos) return ERROR(corruption_detected); /* should never happen */
+ RETURN_ERROR_IF(toLoad > zds->inBuffSize - zds->inPos,
+ corruption_detected,
+ "should never happen");
loadedSize = ZSTD_limitCopy(zds->inBuff + zds->inPos, toLoad, ip, iend-ip);
}
ip += loadedSize;
@@ -2943,12 +1709,24 @@
someMoreWork = 0;
break;
- default: return ERROR(GENERIC); /* impossible */
+ default:
+ assert(0); /* impossible */
+ RETURN_ERROR(GENERIC); /* some compiler require default to do something */
} }
/* result */
- input->pos += (size_t)(ip-istart);
- output->pos += (size_t)(op-ostart);
+ input->pos = (size_t)(ip - (const char*)(input->src));
+ output->pos = (size_t)(op - (char*)(output->dst));
+ if ((ip==istart) && (op==ostart)) { /* no forward progress */
+ zds->noForwardProgress ++;
+ if (zds->noForwardProgress >= ZSTD_NO_FORWARD_PROGRESS_MAX) {
+ RETURN_ERROR_IF(op==oend, dstSize_tooSmall);
+ RETURN_ERROR_IF(ip==iend, srcSize_wrong);
+ assert(0);
+ }
+ } else {
+ zds->noForwardProgress = 0;
+ }
{ size_t nextSrcSizeHint = ZSTD_nextSrcSizeToDecompress(zds);
if (!nextSrcSizeHint) { /* frame fully decoded */
if (zds->outEnd == zds->outStart) { /* output fully flushed */
@@ -2975,13 +1753,7 @@
}
}
-
-size_t ZSTD_decompress_generic(ZSTD_DCtx* dctx, ZSTD_outBuffer* output, ZSTD_inBuffer* input)
-{
- return ZSTD_decompressStream(dctx, output, input);
-}
-
-size_t ZSTD_decompress_generic_simpleArgs (
+size_t ZSTD_decompressStream_simpleArgs (
ZSTD_DCtx* dctx,
void* dst, size_t dstCapacity, size_t* dstPos,
const void* src, size_t srcSize, size_t* srcPos)
@@ -2989,15 +1761,8 @@
ZSTD_outBuffer output = { dst, dstCapacity, *dstPos };
ZSTD_inBuffer input = { src, srcSize, *srcPos };
/* ZSTD_compress_generic() will check validity of dstPos and srcPos */
- size_t const cErr = ZSTD_decompress_generic(dctx, &output, &input);
+ size_t const cErr = ZSTD_decompressStream(dctx, &output, &input);
*dstPos = output.pos;
*srcPos = input.pos;
return cErr;
}
-
-void ZSTD_DCtx_reset(ZSTD_DCtx* dctx)
-{
- (void)ZSTD_initDStream(dctx);
- dctx->format = ZSTD_f_zstd1;
- dctx->maxWindowSize = ZSTD_MAXWINDOWSIZE_DEFAULT;
-}
diff --git a/vendor/github.com/DataDog/zstd/zstd_decompress_block.c b/vendor/github.com/DataDog/zstd/zstd_decompress_block.c
new file mode 100644
index 0000000..24f4859
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/zstd_decompress_block.c
@@ -0,0 +1,1322 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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.
+ */
+
+/* zstd_decompress_block :
+ * this module takes care of decompressing _compressed_ block */
+
+/*-*******************************************************
+* Dependencies
+*********************************************************/
+#include <string.h> /* memcpy, memmove, memset */
+#include "compiler.h" /* prefetch */
+#include "cpu.h" /* bmi2 */
+#include "mem.h" /* low level memory routines */
+#define FSE_STATIC_LINKING_ONLY
+#include "fse.h"
+#define HUF_STATIC_LINKING_ONLY
+#include "huf.h"
+#include "zstd_internal.h"
+#include "zstd_decompress_internal.h" /* ZSTD_DCtx */
+#include "zstd_ddict.h" /* ZSTD_DDictDictContent */
+#include "zstd_decompress_block.h"
+
+/*_*******************************************************
+* Macros
+**********************************************************/
+
+/* These two optional macros force the use one way or another of the two
+ * ZSTD_decompressSequences implementations. You can't force in both directions
+ * at the same time.
+ */
+#if defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT) && \
+ defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG)
+#error "Cannot force the use of the short and the long ZSTD_decompressSequences variants!"
+#endif
+
+
+/*_*******************************************************
+* Memory operations
+**********************************************************/
+static void ZSTD_copy4(void* dst, const void* src) { memcpy(dst, src, 4); }
+
+
+/*-*************************************************************
+ * Block decoding
+ ***************************************************************/
+
+/*! ZSTD_getcBlockSize() :
+ * Provides the size of compressed block from block header `src` */
+size_t ZSTD_getcBlockSize(const void* src, size_t srcSize,
+ blockProperties_t* bpPtr)
+{
+ RETURN_ERROR_IF(srcSize < ZSTD_blockHeaderSize, srcSize_wrong);
+
+ { U32 const cBlockHeader = MEM_readLE24(src);
+ U32 const cSize = cBlockHeader >> 3;
+ bpPtr->lastBlock = cBlockHeader & 1;
+ bpPtr->blockType = (blockType_e)((cBlockHeader >> 1) & 3);
+ bpPtr->origSize = cSize; /* only useful for RLE */
+ if (bpPtr->blockType == bt_rle) return 1;
+ RETURN_ERROR_IF(bpPtr->blockType == bt_reserved, corruption_detected);
+ return cSize;
+ }
+}
+
+
+/* Hidden declaration for fullbench */
+size_t ZSTD_decodeLiteralsBlock(ZSTD_DCtx* dctx,
+ const void* src, size_t srcSize);
+/*! ZSTD_decodeLiteralsBlock() :
+ * @return : nb of bytes read from src (< srcSize )
+ * note : symbol not declared but exposed for fullbench */
+size_t ZSTD_decodeLiteralsBlock(ZSTD_DCtx* dctx,
+ const void* src, size_t srcSize) /* note : srcSize < BLOCKSIZE */
+{
+ RETURN_ERROR_IF(srcSize < MIN_CBLOCK_SIZE, corruption_detected);
+
+ { const BYTE* const istart = (const BYTE*) src;
+ symbolEncodingType_e const litEncType = (symbolEncodingType_e)(istart[0] & 3);
+
+ switch(litEncType)
+ {
+ case set_repeat:
+ RETURN_ERROR_IF(dctx->litEntropy==0, dictionary_corrupted);
+ /* fall-through */
+
+ case set_compressed:
+ RETURN_ERROR_IF(srcSize < 5, corruption_detected, "srcSize >= MIN_CBLOCK_SIZE == 3; here we need up to 5 for case 3");
+ { size_t lhSize, litSize, litCSize;
+ U32 singleStream=0;
+ U32 const lhlCode = (istart[0] >> 2) & 3;
+ U32 const lhc = MEM_readLE32(istart);
+ size_t hufSuccess;
+ switch(lhlCode)
+ {
+ case 0: case 1: default: /* note : default is impossible, since lhlCode into [0..3] */
+ /* 2 - 2 - 10 - 10 */
+ singleStream = !lhlCode;
+ lhSize = 3;
+ litSize = (lhc >> 4) & 0x3FF;
+ litCSize = (lhc >> 14) & 0x3FF;
+ break;
+ case 2:
+ /* 2 - 2 - 14 - 14 */
+ lhSize = 4;
+ litSize = (lhc >> 4) & 0x3FFF;
+ litCSize = lhc >> 18;
+ break;
+ case 3:
+ /* 2 - 2 - 18 - 18 */
+ lhSize = 5;
+ litSize = (lhc >> 4) & 0x3FFFF;
+ litCSize = (lhc >> 22) + (istart[4] << 10);
+ break;
+ }
+ RETURN_ERROR_IF(litSize > ZSTD_BLOCKSIZE_MAX, corruption_detected);
+ RETURN_ERROR_IF(litCSize + lhSize > srcSize, corruption_detected);
+
+ /* prefetch huffman table if cold */
+ if (dctx->ddictIsCold && (litSize > 768 /* heuristic */)) {
+ PREFETCH_AREA(dctx->HUFptr, sizeof(dctx->entropy.hufTable));
+ }
+
+ if (litEncType==set_repeat) {
+ if (singleStream) {
+ hufSuccess = HUF_decompress1X_usingDTable_bmi2(
+ dctx->litBuffer, litSize, istart+lhSize, litCSize,
+ dctx->HUFptr, dctx->bmi2);
+ } else {
+ hufSuccess = HUF_decompress4X_usingDTable_bmi2(
+ dctx->litBuffer, litSize, istart+lhSize, litCSize,
+ dctx->HUFptr, dctx->bmi2);
+ }
+ } else {
+ if (singleStream) {
+#if defined(HUF_FORCE_DECOMPRESS_X2)
+ hufSuccess = HUF_decompress1X_DCtx_wksp(
+ dctx->entropy.hufTable, dctx->litBuffer, litSize,
+ istart+lhSize, litCSize, dctx->workspace,
+ sizeof(dctx->workspace));
+#else
+ hufSuccess = HUF_decompress1X1_DCtx_wksp_bmi2(
+ dctx->entropy.hufTable, dctx->litBuffer, litSize,
+ istart+lhSize, litCSize, dctx->workspace,
+ sizeof(dctx->workspace), dctx->bmi2);
+#endif
+ } else {
+ hufSuccess = HUF_decompress4X_hufOnly_wksp_bmi2(
+ dctx->entropy.hufTable, dctx->litBuffer, litSize,
+ istart+lhSize, litCSize, dctx->workspace,
+ sizeof(dctx->workspace), dctx->bmi2);
+ }
+ }
+
+ RETURN_ERROR_IF(HUF_isError(hufSuccess), corruption_detected);
+
+ dctx->litPtr = dctx->litBuffer;
+ dctx->litSize = litSize;
+ dctx->litEntropy = 1;
+ if (litEncType==set_compressed) dctx->HUFptr = dctx->entropy.hufTable;
+ memset(dctx->litBuffer + dctx->litSize, 0, WILDCOPY_OVERLENGTH);
+ return litCSize + lhSize;
+ }
+
+ case set_basic:
+ { size_t litSize, lhSize;
+ U32 const lhlCode = ((istart[0]) >> 2) & 3;
+ switch(lhlCode)
+ {
+ case 0: case 2: default: /* note : default is impossible, since lhlCode into [0..3] */
+ lhSize = 1;
+ litSize = istart[0] >> 3;
+ break;
+ case 1:
+ lhSize = 2;
+ litSize = MEM_readLE16(istart) >> 4;
+ break;
+ case 3:
+ lhSize = 3;
+ litSize = MEM_readLE24(istart) >> 4;
+ break;
+ }
+
+ if (lhSize+litSize+WILDCOPY_OVERLENGTH > srcSize) { /* risk reading beyond src buffer with wildcopy */
+ RETURN_ERROR_IF(litSize+lhSize > srcSize, corruption_detected);
+ memcpy(dctx->litBuffer, istart+lhSize, litSize);
+ dctx->litPtr = dctx->litBuffer;
+ dctx->litSize = litSize;
+ memset(dctx->litBuffer + dctx->litSize, 0, WILDCOPY_OVERLENGTH);
+ return lhSize+litSize;
+ }
+ /* direct reference into compressed stream */
+ dctx->litPtr = istart+lhSize;
+ dctx->litSize = litSize;
+ return lhSize+litSize;
+ }
+
+ case set_rle:
+ { U32 const lhlCode = ((istart[0]) >> 2) & 3;
+ size_t litSize, lhSize;
+ switch(lhlCode)
+ {
+ case 0: case 2: default: /* note : default is impossible, since lhlCode into [0..3] */
+ lhSize = 1;
+ litSize = istart[0] >> 3;
+ break;
+ case 1:
+ lhSize = 2;
+ litSize = MEM_readLE16(istart) >> 4;
+ break;
+ case 3:
+ lhSize = 3;
+ litSize = MEM_readLE24(istart) >> 4;
+ RETURN_ERROR_IF(srcSize<4, corruption_detected, "srcSize >= MIN_CBLOCK_SIZE == 3; here we need lhSize+1 = 4");
+ break;
+ }
+ RETURN_ERROR_IF(litSize > ZSTD_BLOCKSIZE_MAX, corruption_detected);
+ memset(dctx->litBuffer, istart[lhSize], litSize + WILDCOPY_OVERLENGTH);
+ dctx->litPtr = dctx->litBuffer;
+ dctx->litSize = litSize;
+ return lhSize+1;
+ }
+ default:
+ RETURN_ERROR(corruption_detected, "impossible");
+ }
+ }
+}
+
+/* Default FSE distribution tables.
+ * These are pre-calculated FSE decoding tables using default distributions as defined in specification :
+ * https://github.com/facebook/zstd/blob/master/doc/zstd_compression_format.md#default-distributions
+ * They were generated programmatically with following method :
+ * - start from default distributions, present in /lib/common/zstd_internal.h
+ * - generate tables normally, using ZSTD_buildFSETable()
+ * - printout the content of tables
+ * - pretify output, report below, test with fuzzer to ensure it's correct */
+
+/* Default FSE distribution table for Literal Lengths */
+static const ZSTD_seqSymbol LL_defaultDTable[(1<<LL_DEFAULTNORMLOG)+1] = {
+ { 1, 1, 1, LL_DEFAULTNORMLOG}, /* header : fastMode, tableLog */
+ /* nextState, nbAddBits, nbBits, baseVal */
+ { 0, 0, 4, 0}, { 16, 0, 4, 0},
+ { 32, 0, 5, 1}, { 0, 0, 5, 3},
+ { 0, 0, 5, 4}, { 0, 0, 5, 6},
+ { 0, 0, 5, 7}, { 0, 0, 5, 9},
+ { 0, 0, 5, 10}, { 0, 0, 5, 12},
+ { 0, 0, 6, 14}, { 0, 1, 5, 16},
+ { 0, 1, 5, 20}, { 0, 1, 5, 22},
+ { 0, 2, 5, 28}, { 0, 3, 5, 32},
+ { 0, 4, 5, 48}, { 32, 6, 5, 64},
+ { 0, 7, 5, 128}, { 0, 8, 6, 256},
+ { 0, 10, 6, 1024}, { 0, 12, 6, 4096},
+ { 32, 0, 4, 0}, { 0, 0, 4, 1},
+ { 0, 0, 5, 2}, { 32, 0, 5, 4},
+ { 0, 0, 5, 5}, { 32, 0, 5, 7},
+ { 0, 0, 5, 8}, { 32, 0, 5, 10},
+ { 0, 0, 5, 11}, { 0, 0, 6, 13},
+ { 32, 1, 5, 16}, { 0, 1, 5, 18},
+ { 32, 1, 5, 22}, { 0, 2, 5, 24},
+ { 32, 3, 5, 32}, { 0, 3, 5, 40},
+ { 0, 6, 4, 64}, { 16, 6, 4, 64},
+ { 32, 7, 5, 128}, { 0, 9, 6, 512},
+ { 0, 11, 6, 2048}, { 48, 0, 4, 0},
+ { 16, 0, 4, 1}, { 32, 0, 5, 2},
+ { 32, 0, 5, 3}, { 32, 0, 5, 5},
+ { 32, 0, 5, 6}, { 32, 0, 5, 8},
+ { 32, 0, 5, 9}, { 32, 0, 5, 11},
+ { 32, 0, 5, 12}, { 0, 0, 6, 15},
+ { 32, 1, 5, 18}, { 32, 1, 5, 20},
+ { 32, 2, 5, 24}, { 32, 2, 5, 28},
+ { 32, 3, 5, 40}, { 32, 4, 5, 48},
+ { 0, 16, 6,65536}, { 0, 15, 6,32768},
+ { 0, 14, 6,16384}, { 0, 13, 6, 8192},
+}; /* LL_defaultDTable */
+
+/* Default FSE distribution table for Offset Codes */
+static const ZSTD_seqSymbol OF_defaultDTable[(1<<OF_DEFAULTNORMLOG)+1] = {
+ { 1, 1, 1, OF_DEFAULTNORMLOG}, /* header : fastMode, tableLog */
+ /* nextState, nbAddBits, nbBits, baseVal */
+ { 0, 0, 5, 0}, { 0, 6, 4, 61},
+ { 0, 9, 5, 509}, { 0, 15, 5,32765},
+ { 0, 21, 5,2097149}, { 0, 3, 5, 5},
+ { 0, 7, 4, 125}, { 0, 12, 5, 4093},
+ { 0, 18, 5,262141}, { 0, 23, 5,8388605},
+ { 0, 5, 5, 29}, { 0, 8, 4, 253},
+ { 0, 14, 5,16381}, { 0, 20, 5,1048573},
+ { 0, 2, 5, 1}, { 16, 7, 4, 125},
+ { 0, 11, 5, 2045}, { 0, 17, 5,131069},
+ { 0, 22, 5,4194301}, { 0, 4, 5, 13},
+ { 16, 8, 4, 253}, { 0, 13, 5, 8189},
+ { 0, 19, 5,524285}, { 0, 1, 5, 1},
+ { 16, 6, 4, 61}, { 0, 10, 5, 1021},
+ { 0, 16, 5,65533}, { 0, 28, 5,268435453},
+ { 0, 27, 5,134217725}, { 0, 26, 5,67108861},
+ { 0, 25, 5,33554429}, { 0, 24, 5,16777213},
+}; /* OF_defaultDTable */
+
+
+/* Default FSE distribution table for Match Lengths */
+static const ZSTD_seqSymbol ML_defaultDTable[(1<<ML_DEFAULTNORMLOG)+1] = {
+ { 1, 1, 1, ML_DEFAULTNORMLOG}, /* header : fastMode, tableLog */
+ /* nextState, nbAddBits, nbBits, baseVal */
+ { 0, 0, 6, 3}, { 0, 0, 4, 4},
+ { 32, 0, 5, 5}, { 0, 0, 5, 6},
+ { 0, 0, 5, 8}, { 0, 0, 5, 9},
+ { 0, 0, 5, 11}, { 0, 0, 6, 13},
+ { 0, 0, 6, 16}, { 0, 0, 6, 19},
+ { 0, 0, 6, 22}, { 0, 0, 6, 25},
+ { 0, 0, 6, 28}, { 0, 0, 6, 31},
+ { 0, 0, 6, 34}, { 0, 1, 6, 37},
+ { 0, 1, 6, 41}, { 0, 2, 6, 47},
+ { 0, 3, 6, 59}, { 0, 4, 6, 83},
+ { 0, 7, 6, 131}, { 0, 9, 6, 515},
+ { 16, 0, 4, 4}, { 0, 0, 4, 5},
+ { 32, 0, 5, 6}, { 0, 0, 5, 7},
+ { 32, 0, 5, 9}, { 0, 0, 5, 10},
+ { 0, 0, 6, 12}, { 0, 0, 6, 15},
+ { 0, 0, 6, 18}, { 0, 0, 6, 21},
+ { 0, 0, 6, 24}, { 0, 0, 6, 27},
+ { 0, 0, 6, 30}, { 0, 0, 6, 33},
+ { 0, 1, 6, 35}, { 0, 1, 6, 39},
+ { 0, 2, 6, 43}, { 0, 3, 6, 51},
+ { 0, 4, 6, 67}, { 0, 5, 6, 99},
+ { 0, 8, 6, 259}, { 32, 0, 4, 4},
+ { 48, 0, 4, 4}, { 16, 0, 4, 5},
+ { 32, 0, 5, 7}, { 32, 0, 5, 8},
+ { 32, 0, 5, 10}, { 32, 0, 5, 11},
+ { 0, 0, 6, 14}, { 0, 0, 6, 17},
+ { 0, 0, 6, 20}, { 0, 0, 6, 23},
+ { 0, 0, 6, 26}, { 0, 0, 6, 29},
+ { 0, 0, 6, 32}, { 0, 16, 6,65539},
+ { 0, 15, 6,32771}, { 0, 14, 6,16387},
+ { 0, 13, 6, 8195}, { 0, 12, 6, 4099},
+ { 0, 11, 6, 2051}, { 0, 10, 6, 1027},
+}; /* ML_defaultDTable */
+
+
+static void ZSTD_buildSeqTable_rle(ZSTD_seqSymbol* dt, U32 baseValue, U32 nbAddBits)
+{
+ void* ptr = dt;
+ ZSTD_seqSymbol_header* const DTableH = (ZSTD_seqSymbol_header*)ptr;
+ ZSTD_seqSymbol* const cell = dt + 1;
+
+ DTableH->tableLog = 0;
+ DTableH->fastMode = 0;
+
+ cell->nbBits = 0;
+ cell->nextState = 0;
+ assert(nbAddBits < 255);
+ cell->nbAdditionalBits = (BYTE)nbAddBits;
+ cell->baseValue = baseValue;
+}
+
+
+/* ZSTD_buildFSETable() :
+ * generate FSE decoding table for one symbol (ll, ml or off)
+ * cannot fail if input is valid =>
+ * all inputs are presumed validated at this stage */
+void
+ZSTD_buildFSETable(ZSTD_seqSymbol* dt,
+ const short* normalizedCounter, unsigned maxSymbolValue,
+ const U32* baseValue, const U32* nbAdditionalBits,
+ unsigned tableLog)
+{
+ ZSTD_seqSymbol* const tableDecode = dt+1;
+ U16 symbolNext[MaxSeq+1];
+
+ U32 const maxSV1 = maxSymbolValue + 1;
+ U32 const tableSize = 1 << tableLog;
+ U32 highThreshold = tableSize-1;
+
+ /* Sanity Checks */
+ assert(maxSymbolValue <= MaxSeq);
+ assert(tableLog <= MaxFSELog);
+
+ /* Init, lay down lowprob symbols */
+ { ZSTD_seqSymbol_header DTableH;
+ DTableH.tableLog = tableLog;
+ DTableH.fastMode = 1;
+ { S16 const largeLimit= (S16)(1 << (tableLog-1));
+ U32 s;
+ for (s=0; s<maxSV1; s++) {
+ if (normalizedCounter[s]==-1) {
+ tableDecode[highThreshold--].baseValue = s;
+ symbolNext[s] = 1;
+ } else {
+ if (normalizedCounter[s] >= largeLimit) DTableH.fastMode=0;
+ symbolNext[s] = normalizedCounter[s];
+ } } }
+ memcpy(dt, &DTableH, sizeof(DTableH));
+ }
+
+ /* Spread symbols */
+ { U32 const tableMask = tableSize-1;
+ U32 const step = FSE_TABLESTEP(tableSize);
+ U32 s, position = 0;
+ for (s=0; s<maxSV1; s++) {
+ int i;
+ for (i=0; i<normalizedCounter[s]; i++) {
+ tableDecode[position].baseValue = s;
+ position = (position + step) & tableMask;
+ while (position > highThreshold) position = (position + step) & tableMask; /* lowprob area */
+ } }
+ assert(position == 0); /* position must reach all cells once, otherwise normalizedCounter is incorrect */
+ }
+
+ /* Build Decoding table */
+ { U32 u;
+ for (u=0; u<tableSize; u++) {
+ U32 const symbol = tableDecode[u].baseValue;
+ U32 const nextState = symbolNext[symbol]++;
+ tableDecode[u].nbBits = (BYTE) (tableLog - BIT_highbit32(nextState) );
+ tableDecode[u].nextState = (U16) ( (nextState << tableDecode[u].nbBits) - tableSize);
+ assert(nbAdditionalBits[symbol] < 255);
+ tableDecode[u].nbAdditionalBits = (BYTE)nbAdditionalBits[symbol];
+ tableDecode[u].baseValue = baseValue[symbol];
+ } }
+}
+
+
+/*! ZSTD_buildSeqTable() :
+ * @return : nb bytes read from src,
+ * or an error code if it fails */
+static size_t ZSTD_buildSeqTable(ZSTD_seqSymbol* DTableSpace, const ZSTD_seqSymbol** DTablePtr,
+ symbolEncodingType_e type, unsigned max, U32 maxLog,
+ const void* src, size_t srcSize,
+ const U32* baseValue, const U32* nbAdditionalBits,
+ const ZSTD_seqSymbol* defaultTable, U32 flagRepeatTable,
+ int ddictIsCold, int nbSeq)
+{
+ switch(type)
+ {
+ case set_rle :
+ RETURN_ERROR_IF(!srcSize, srcSize_wrong);
+ RETURN_ERROR_IF((*(const BYTE*)src) > max, corruption_detected);
+ { U32 const symbol = *(const BYTE*)src;
+ U32 const baseline = baseValue[symbol];
+ U32 const nbBits = nbAdditionalBits[symbol];
+ ZSTD_buildSeqTable_rle(DTableSpace, baseline, nbBits);
+ }
+ *DTablePtr = DTableSpace;
+ return 1;
+ case set_basic :
+ *DTablePtr = defaultTable;
+ return 0;
+ case set_repeat:
+ RETURN_ERROR_IF(!flagRepeatTable, corruption_detected);
+ /* prefetch FSE table if used */
+ if (ddictIsCold && (nbSeq > 24 /* heuristic */)) {
+ const void* const pStart = *DTablePtr;
+ size_t const pSize = sizeof(ZSTD_seqSymbol) * (SEQSYMBOL_TABLE_SIZE(maxLog));
+ PREFETCH_AREA(pStart, pSize);
+ }
+ return 0;
+ case set_compressed :
+ { unsigned tableLog;
+ S16 norm[MaxSeq+1];
+ size_t const headerSize = FSE_readNCount(norm, &max, &tableLog, src, srcSize);
+ RETURN_ERROR_IF(FSE_isError(headerSize), corruption_detected);
+ RETURN_ERROR_IF(tableLog > maxLog, corruption_detected);
+ ZSTD_buildFSETable(DTableSpace, norm, max, baseValue, nbAdditionalBits, tableLog);
+ *DTablePtr = DTableSpace;
+ return headerSize;
+ }
+ default :
+ assert(0);
+ RETURN_ERROR(GENERIC, "impossible");
+ }
+}
+
+size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr,
+ const void* src, size_t srcSize)
+{
+ const BYTE* const istart = (const BYTE* const)src;
+ const BYTE* const iend = istart + srcSize;
+ const BYTE* ip = istart;
+ int nbSeq;
+ DEBUGLOG(5, "ZSTD_decodeSeqHeaders");
+
+ /* check */
+ RETURN_ERROR_IF(srcSize < MIN_SEQUENCES_SIZE, srcSize_wrong);
+
+ /* SeqHead */
+ nbSeq = *ip++;
+ if (!nbSeq) {
+ *nbSeqPtr=0;
+ RETURN_ERROR_IF(srcSize != 1, srcSize_wrong);
+ return 1;
+ }
+ if (nbSeq > 0x7F) {
+ if (nbSeq == 0xFF) {
+ RETURN_ERROR_IF(ip+2 > iend, srcSize_wrong);
+ nbSeq = MEM_readLE16(ip) + LONGNBSEQ, ip+=2;
+ } else {
+ RETURN_ERROR_IF(ip >= iend, srcSize_wrong);
+ nbSeq = ((nbSeq-0x80)<<8) + *ip++;
+ }
+ }
+ *nbSeqPtr = nbSeq;
+
+ /* FSE table descriptors */
+ RETURN_ERROR_IF(ip+1 > iend, srcSize_wrong); /* minimum possible size: 1 byte for symbol encoding types */
+ { symbolEncodingType_e const LLtype = (symbolEncodingType_e)(*ip >> 6);
+ symbolEncodingType_e const OFtype = (symbolEncodingType_e)((*ip >> 4) & 3);
+ symbolEncodingType_e const MLtype = (symbolEncodingType_e)((*ip >> 2) & 3);
+ ip++;
+
+ /* Build DTables */
+ { size_t const llhSize = ZSTD_buildSeqTable(dctx->entropy.LLTable, &dctx->LLTptr,
+ LLtype, MaxLL, LLFSELog,
+ ip, iend-ip,
+ LL_base, LL_bits,
+ LL_defaultDTable, dctx->fseEntropy,
+ dctx->ddictIsCold, nbSeq);
+ RETURN_ERROR_IF(ZSTD_isError(llhSize), corruption_detected);
+ ip += llhSize;
+ }
+
+ { size_t const ofhSize = ZSTD_buildSeqTable(dctx->entropy.OFTable, &dctx->OFTptr,
+ OFtype, MaxOff, OffFSELog,
+ ip, iend-ip,
+ OF_base, OF_bits,
+ OF_defaultDTable, dctx->fseEntropy,
+ dctx->ddictIsCold, nbSeq);
+ RETURN_ERROR_IF(ZSTD_isError(ofhSize), corruption_detected);
+ ip += ofhSize;
+ }
+
+ { size_t const mlhSize = ZSTD_buildSeqTable(dctx->entropy.MLTable, &dctx->MLTptr,
+ MLtype, MaxML, MLFSELog,
+ ip, iend-ip,
+ ML_base, ML_bits,
+ ML_defaultDTable, dctx->fseEntropy,
+ dctx->ddictIsCold, nbSeq);
+ RETURN_ERROR_IF(ZSTD_isError(mlhSize), corruption_detected);
+ ip += mlhSize;
+ }
+ }
+
+ return ip-istart;
+}
+
+
+typedef struct {
+ size_t litLength;
+ size_t matchLength;
+ size_t offset;
+ const BYTE* match;
+} seq_t;
+
+typedef struct {
+ size_t state;
+ const ZSTD_seqSymbol* table;
+} ZSTD_fseState;
+
+typedef struct {
+ BIT_DStream_t DStream;
+ ZSTD_fseState stateLL;
+ ZSTD_fseState stateOffb;
+ ZSTD_fseState stateML;
+ size_t prevOffset[ZSTD_REP_NUM];
+ const BYTE* prefixStart;
+ const BYTE* dictEnd;
+ size_t pos;
+} seqState_t;
+
+
+/* ZSTD_execSequenceLast7():
+ * exceptional case : decompress a match starting within last 7 bytes of output buffer.
+ * requires more careful checks, to ensure there is no overflow.
+ * performance does not matter though.
+ * note : this case is supposed to be never generated "naturally" by reference encoder,
+ * since in most cases it needs at least 8 bytes to look for a match.
+ * but it's allowed by the specification. */
+FORCE_NOINLINE
+size_t ZSTD_execSequenceLast7(BYTE* op,
+ BYTE* const oend, seq_t sequence,
+ const BYTE** litPtr, const BYTE* const litLimit,
+ const BYTE* const base, const BYTE* const vBase, const BYTE* const dictEnd)
+{
+ BYTE* const oLitEnd = op + sequence.litLength;
+ size_t const sequenceLength = sequence.litLength + sequence.matchLength;
+ BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */
+ const BYTE* const iLitEnd = *litPtr + sequence.litLength;
+ const BYTE* match = oLitEnd - sequence.offset;
+
+ /* check */
+ RETURN_ERROR_IF(oMatchEnd>oend, dstSize_tooSmall, "last match must fit within dstBuffer");
+ RETURN_ERROR_IF(iLitEnd > litLimit, corruption_detected, "try to read beyond literal buffer");
+
+ /* copy literals */
+ while (op < oLitEnd) *op++ = *(*litPtr)++;
+
+ /* copy Match */
+ if (sequence.offset > (size_t)(oLitEnd - base)) {
+ /* offset beyond prefix */
+ RETURN_ERROR_IF(sequence.offset > (size_t)(oLitEnd - vBase),corruption_detected);
+ match = dictEnd - (base-match);
+ if (match + sequence.matchLength <= dictEnd) {
+ memmove(oLitEnd, match, sequence.matchLength);
+ return sequenceLength;
+ }
+ /* span extDict & currentPrefixSegment */
+ { size_t const length1 = dictEnd - match;
+ memmove(oLitEnd, match, length1);
+ op = oLitEnd + length1;
+ sequence.matchLength -= length1;
+ match = base;
+ } }
+ while (op < oMatchEnd) *op++ = *match++;
+ return sequenceLength;
+}
+
+
+HINT_INLINE
+size_t ZSTD_execSequence(BYTE* op,
+ BYTE* const oend, seq_t sequence,
+ const BYTE** litPtr, const BYTE* const litLimit,
+ const BYTE* const prefixStart, const BYTE* const virtualStart, const BYTE* const dictEnd)
+{
+ BYTE* const oLitEnd = op + sequence.litLength;
+ size_t const sequenceLength = sequence.litLength + sequence.matchLength;
+ BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */
+ BYTE* const oend_w = oend - WILDCOPY_OVERLENGTH;
+ const BYTE* const iLitEnd = *litPtr + sequence.litLength;
+ const BYTE* match = oLitEnd - sequence.offset;
+
+ /* check */
+ RETURN_ERROR_IF(oMatchEnd>oend, dstSize_tooSmall, "last match must start at a minimum distance of WILDCOPY_OVERLENGTH from oend");
+ RETURN_ERROR_IF(iLitEnd > litLimit, corruption_detected, "over-read beyond lit buffer");
+ if (oLitEnd>oend_w) return ZSTD_execSequenceLast7(op, oend, sequence, litPtr, litLimit, prefixStart, virtualStart, dictEnd);
+
+ /* copy Literals */
+ if (sequence.litLength > 8)
+ ZSTD_wildcopy_16min(op, (*litPtr), sequence.litLength, ZSTD_no_overlap); /* note : since oLitEnd <= oend-WILDCOPY_OVERLENGTH, no risk of overwrite beyond oend */
+ else
+ ZSTD_copy8(op, *litPtr);
+ op = oLitEnd;
+ *litPtr = iLitEnd; /* update for next sequence */
+
+ /* copy Match */
+ if (sequence.offset > (size_t)(oLitEnd - prefixStart)) {
+ /* offset beyond prefix -> go into extDict */
+ RETURN_ERROR_IF(sequence.offset > (size_t)(oLitEnd - virtualStart), corruption_detected);
+ match = dictEnd + (match - prefixStart);
+ if (match + sequence.matchLength <= dictEnd) {
+ memmove(oLitEnd, match, sequence.matchLength);
+ return sequenceLength;
+ }
+ /* span extDict & currentPrefixSegment */
+ { size_t const length1 = dictEnd - match;
+ memmove(oLitEnd, match, length1);
+ op = oLitEnd + length1;
+ sequence.matchLength -= length1;
+ match = prefixStart;
+ if (op > oend_w || sequence.matchLength < MINMATCH) {
+ U32 i;
+ for (i = 0; i < sequence.matchLength; ++i) op[i] = match[i];
+ return sequenceLength;
+ }
+ } }
+ /* Requirement: op <= oend_w && sequence.matchLength >= MINMATCH */
+
+ /* match within prefix */
+ if (sequence.offset < 8) {
+ /* close range match, overlap */
+ static const U32 dec32table[] = { 0, 1, 2, 1, 4, 4, 4, 4 }; /* added */
+ static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* subtracted */
+ int const sub2 = dec64table[sequence.offset];
+ op[0] = match[0];
+ op[1] = match[1];
+ op[2] = match[2];
+ op[3] = match[3];
+ match += dec32table[sequence.offset];
+ ZSTD_copy4(op+4, match);
+ match -= sub2;
+ } else {
+ ZSTD_copy8(op, match);
+ }
+ op += 8; match += 8;
+
+ if (oMatchEnd > oend-(16-MINMATCH)) {
+ if (op < oend_w) {
+ ZSTD_wildcopy(op, match, oend_w - op, ZSTD_overlap_src_before_dst);
+ match += oend_w - op;
+ op = oend_w;
+ }
+ while (op < oMatchEnd) *op++ = *match++;
+ } else {
+ ZSTD_wildcopy(op, match, (ptrdiff_t)sequence.matchLength-8, ZSTD_overlap_src_before_dst); /* works even if matchLength < 8 */
+ }
+ return sequenceLength;
+}
+
+
+HINT_INLINE
+size_t ZSTD_execSequenceLong(BYTE* op,
+ BYTE* const oend, seq_t sequence,
+ const BYTE** litPtr, const BYTE* const litLimit,
+ const BYTE* const prefixStart, const BYTE* const dictStart, const BYTE* const dictEnd)
+{
+ BYTE* const oLitEnd = op + sequence.litLength;
+ size_t const sequenceLength = sequence.litLength + sequence.matchLength;
+ BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */
+ BYTE* const oend_w = oend - WILDCOPY_OVERLENGTH;
+ const BYTE* const iLitEnd = *litPtr + sequence.litLength;
+ const BYTE* match = sequence.match;
+
+ /* check */
+ RETURN_ERROR_IF(oMatchEnd > oend, dstSize_tooSmall, "last match must start at a minimum distance of WILDCOPY_OVERLENGTH from oend");
+ RETURN_ERROR_IF(iLitEnd > litLimit, corruption_detected, "over-read beyond lit buffer");
+ if (oLitEnd > oend_w) return ZSTD_execSequenceLast7(op, oend, sequence, litPtr, litLimit, prefixStart, dictStart, dictEnd);
+
+ /* copy Literals */
+ if (sequence.litLength > 8)
+ ZSTD_wildcopy_16min(op, *litPtr, sequence.litLength, ZSTD_no_overlap); /* note : since oLitEnd <= oend-WILDCOPY_OVERLENGTH, no risk of overwrite beyond oend */
+ else
+ ZSTD_copy8(op, *litPtr); /* note : op <= oLitEnd <= oend_w == oend - 8 */
+
+ op = oLitEnd;
+ *litPtr = iLitEnd; /* update for next sequence */
+
+ /* copy Match */
+ if (sequence.offset > (size_t)(oLitEnd - prefixStart)) {
+ /* offset beyond prefix */
+ RETURN_ERROR_IF(sequence.offset > (size_t)(oLitEnd - dictStart), corruption_detected);
+ if (match + sequence.matchLength <= dictEnd) {
+ memmove(oLitEnd, match, sequence.matchLength);
+ return sequenceLength;
+ }
+ /* span extDict & currentPrefixSegment */
+ { size_t const length1 = dictEnd - match;
+ memmove(oLitEnd, match, length1);
+ op = oLitEnd + length1;
+ sequence.matchLength -= length1;
+ match = prefixStart;
+ if (op > oend_w || sequence.matchLength < MINMATCH) {
+ U32 i;
+ for (i = 0; i < sequence.matchLength; ++i) op[i] = match[i];
+ return sequenceLength;
+ }
+ } }
+ assert(op <= oend_w);
+ assert(sequence.matchLength >= MINMATCH);
+
+ /* match within prefix */
+ if (sequence.offset < 8) {
+ /* close range match, overlap */
+ static const U32 dec32table[] = { 0, 1, 2, 1, 4, 4, 4, 4 }; /* added */
+ static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* subtracted */
+ int const sub2 = dec64table[sequence.offset];
+ op[0] = match[0];
+ op[1] = match[1];
+ op[2] = match[2];
+ op[3] = match[3];
+ match += dec32table[sequence.offset];
+ ZSTD_copy4(op+4, match);
+ match -= sub2;
+ } else {
+ ZSTD_copy8(op, match);
+ }
+ op += 8; match += 8;
+
+ if (oMatchEnd > oend-(16-MINMATCH)) {
+ if (op < oend_w) {
+ ZSTD_wildcopy(op, match, oend_w - op, ZSTD_overlap_src_before_dst);
+ match += oend_w - op;
+ op = oend_w;
+ }
+ while (op < oMatchEnd) *op++ = *match++;
+ } else {
+ ZSTD_wildcopy(op, match, (ptrdiff_t)sequence.matchLength-8, ZSTD_overlap_src_before_dst); /* works even if matchLength < 8 */
+ }
+ return sequenceLength;
+}
+
+static void
+ZSTD_initFseState(ZSTD_fseState* DStatePtr, BIT_DStream_t* bitD, const ZSTD_seqSymbol* dt)
+{
+ const void* ptr = dt;
+ const ZSTD_seqSymbol_header* const DTableH = (const ZSTD_seqSymbol_header*)ptr;
+ DStatePtr->state = BIT_readBits(bitD, DTableH->tableLog);
+ DEBUGLOG(6, "ZSTD_initFseState : val=%u using %u bits",
+ (U32)DStatePtr->state, DTableH->tableLog);
+ BIT_reloadDStream(bitD);
+ DStatePtr->table = dt + 1;
+}
+
+FORCE_INLINE_TEMPLATE void
+ZSTD_updateFseState(ZSTD_fseState* DStatePtr, BIT_DStream_t* bitD)
+{
+ ZSTD_seqSymbol const DInfo = DStatePtr->table[DStatePtr->state];
+ U32 const nbBits = DInfo.nbBits;
+ size_t const lowBits = BIT_readBits(bitD, nbBits);
+ DStatePtr->state = DInfo.nextState + lowBits;
+}
+
+/* We need to add at most (ZSTD_WINDOWLOG_MAX_32 - 1) bits to read the maximum
+ * offset bits. But we can only read at most (STREAM_ACCUMULATOR_MIN_32 - 1)
+ * bits before reloading. This value is the maximum number of bytes we read
+ * after reloading when we are decoding long offsets.
+ */
+#define LONG_OFFSETS_MAX_EXTRA_BITS_32 \
+ (ZSTD_WINDOWLOG_MAX_32 > STREAM_ACCUMULATOR_MIN_32 \
+ ? ZSTD_WINDOWLOG_MAX_32 - STREAM_ACCUMULATOR_MIN_32 \
+ : 0)
+
+typedef enum { ZSTD_lo_isRegularOffset, ZSTD_lo_isLongOffset=1 } ZSTD_longOffset_e;
+
+#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG
+FORCE_INLINE_TEMPLATE seq_t
+ZSTD_decodeSequence(seqState_t* seqState, const ZSTD_longOffset_e longOffsets)
+{
+ seq_t seq;
+ U32 const llBits = seqState->stateLL.table[seqState->stateLL.state].nbAdditionalBits;
+ U32 const mlBits = seqState->stateML.table[seqState->stateML.state].nbAdditionalBits;
+ U32 const ofBits = seqState->stateOffb.table[seqState->stateOffb.state].nbAdditionalBits;
+ U32 const totalBits = llBits+mlBits+ofBits;
+ U32 const llBase = seqState->stateLL.table[seqState->stateLL.state].baseValue;
+ U32 const mlBase = seqState->stateML.table[seqState->stateML.state].baseValue;
+ U32 const ofBase = seqState->stateOffb.table[seqState->stateOffb.state].baseValue;
+
+ /* sequence */
+ { size_t offset;
+ if (!ofBits)
+ offset = 0;
+ else {
+ ZSTD_STATIC_ASSERT(ZSTD_lo_isLongOffset == 1);
+ ZSTD_STATIC_ASSERT(LONG_OFFSETS_MAX_EXTRA_BITS_32 == 5);
+ assert(ofBits <= MaxOff);
+ if (MEM_32bits() && longOffsets && (ofBits >= STREAM_ACCUMULATOR_MIN_32)) {
+ U32 const extraBits = ofBits - MIN(ofBits, 32 - seqState->DStream.bitsConsumed);
+ offset = ofBase + (BIT_readBitsFast(&seqState->DStream, ofBits - extraBits) << extraBits);
+ BIT_reloadDStream(&seqState->DStream);
+ if (extraBits) offset += BIT_readBitsFast(&seqState->DStream, extraBits);
+ assert(extraBits <= LONG_OFFSETS_MAX_EXTRA_BITS_32); /* to avoid another reload */
+ } else {
+ offset = ofBase + BIT_readBitsFast(&seqState->DStream, ofBits/*>0*/); /* <= (ZSTD_WINDOWLOG_MAX-1) bits */
+ if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream);
+ }
+ }
+
+ if (ofBits <= 1) {
+ offset += (llBase==0);
+ if (offset) {
+ size_t temp = (offset==3) ? seqState->prevOffset[0] - 1 : seqState->prevOffset[offset];
+ temp += !temp; /* 0 is not valid; input is corrupted; force offset to 1 */
+ if (offset != 1) seqState->prevOffset[2] = seqState->prevOffset[1];
+ seqState->prevOffset[1] = seqState->prevOffset[0];
+ seqState->prevOffset[0] = offset = temp;
+ } else { /* offset == 0 */
+ offset = seqState->prevOffset[0];
+ }
+ } else {
+ seqState->prevOffset[2] = seqState->prevOffset[1];
+ seqState->prevOffset[1] = seqState->prevOffset[0];
+ seqState->prevOffset[0] = offset;
+ }
+ seq.offset = offset;
+ }
+
+ seq.matchLength = mlBase
+ + ((mlBits>0) ? BIT_readBitsFast(&seqState->DStream, mlBits/*>0*/) : 0); /* <= 16 bits */
+ if (MEM_32bits() && (mlBits+llBits >= STREAM_ACCUMULATOR_MIN_32-LONG_OFFSETS_MAX_EXTRA_BITS_32))
+ BIT_reloadDStream(&seqState->DStream);
+ if (MEM_64bits() && (totalBits >= STREAM_ACCUMULATOR_MIN_64-(LLFSELog+MLFSELog+OffFSELog)))
+ BIT_reloadDStream(&seqState->DStream);
+ /* Ensure there are enough bits to read the rest of data in 64-bit mode. */
+ ZSTD_STATIC_ASSERT(16+LLFSELog+MLFSELog+OffFSELog < STREAM_ACCUMULATOR_MIN_64);
+
+ seq.litLength = llBase
+ + ((llBits>0) ? BIT_readBitsFast(&seqState->DStream, llBits/*>0*/) : 0); /* <= 16 bits */
+ if (MEM_32bits())
+ BIT_reloadDStream(&seqState->DStream);
+
+ DEBUGLOG(6, "seq: litL=%u, matchL=%u, offset=%u",
+ (U32)seq.litLength, (U32)seq.matchLength, (U32)seq.offset);
+
+ /* ANS state update */
+ ZSTD_updateFseState(&seqState->stateLL, &seqState->DStream); /* <= 9 bits */
+ ZSTD_updateFseState(&seqState->stateML, &seqState->DStream); /* <= 9 bits */
+ if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream); /* <= 18 bits */
+ ZSTD_updateFseState(&seqState->stateOffb, &seqState->DStream); /* <= 8 bits */
+
+ return seq;
+}
+
+FORCE_INLINE_TEMPLATE size_t
+DONT_VECTORIZE
+ZSTD_decompressSequences_body( ZSTD_DCtx* dctx,
+ void* dst, size_t maxDstSize,
+ const void* seqStart, size_t seqSize, int nbSeq,
+ const ZSTD_longOffset_e isLongOffset)
+{
+ const BYTE* ip = (const BYTE*)seqStart;
+ const BYTE* const iend = ip + seqSize;
+ BYTE* const ostart = (BYTE* const)dst;
+ BYTE* const oend = ostart + maxDstSize;
+ BYTE* op = ostart;
+ const BYTE* litPtr = dctx->litPtr;
+ const BYTE* const litEnd = litPtr + dctx->litSize;
+ const BYTE* const prefixStart = (const BYTE*) (dctx->prefixStart);
+ const BYTE* const vBase = (const BYTE*) (dctx->virtualStart);
+ const BYTE* const dictEnd = (const BYTE*) (dctx->dictEnd);
+ DEBUGLOG(5, "ZSTD_decompressSequences_body");
+
+ /* Regen sequences */
+ if (nbSeq) {
+ seqState_t seqState;
+ dctx->fseEntropy = 1;
+ { U32 i; for (i=0; i<ZSTD_REP_NUM; i++) seqState.prevOffset[i] = dctx->entropy.rep[i]; }
+ RETURN_ERROR_IF(
+ ERR_isError(BIT_initDStream(&seqState.DStream, ip, iend-ip)),
+ corruption_detected);
+ ZSTD_initFseState(&seqState.stateLL, &seqState.DStream, dctx->LLTptr);
+ ZSTD_initFseState(&seqState.stateOffb, &seqState.DStream, dctx->OFTptr);
+ ZSTD_initFseState(&seqState.stateML, &seqState.DStream, dctx->MLTptr);
+
+ ZSTD_STATIC_ASSERT(
+ BIT_DStream_unfinished < BIT_DStream_completed &&
+ BIT_DStream_endOfBuffer < BIT_DStream_completed &&
+ BIT_DStream_completed < BIT_DStream_overflow);
+
+ for ( ; (BIT_reloadDStream(&(seqState.DStream)) <= BIT_DStream_completed) && nbSeq ; ) {
+ nbSeq--;
+ { seq_t const sequence = ZSTD_decodeSequence(&seqState, isLongOffset);
+ size_t const oneSeqSize = ZSTD_execSequence(op, oend, sequence, &litPtr, litEnd, prefixStart, vBase, dictEnd);
+ DEBUGLOG(6, "regenerated sequence size : %u", (U32)oneSeqSize);
+ if (ZSTD_isError(oneSeqSize)) return oneSeqSize;
+ op += oneSeqSize;
+ } }
+
+ /* check if reached exact end */
+ DEBUGLOG(5, "ZSTD_decompressSequences_body: after decode loop, remaining nbSeq : %i", nbSeq);
+ RETURN_ERROR_IF(nbSeq, corruption_detected);
+ RETURN_ERROR_IF(BIT_reloadDStream(&seqState.DStream) < BIT_DStream_completed, corruption_detected);
+ /* save reps for next block */
+ { U32 i; for (i=0; i<ZSTD_REP_NUM; i++) dctx->entropy.rep[i] = (U32)(seqState.prevOffset[i]); }
+ }
+
+ /* last literal segment */
+ { size_t const lastLLSize = litEnd - litPtr;
+ RETURN_ERROR_IF(lastLLSize > (size_t)(oend-op), dstSize_tooSmall);
+ memcpy(op, litPtr, lastLLSize);
+ op += lastLLSize;
+ }
+
+ return op-ostart;
+}
+
+static size_t
+ZSTD_decompressSequences_default(ZSTD_DCtx* dctx,
+ void* dst, size_t maxDstSize,
+ const void* seqStart, size_t seqSize, int nbSeq,
+ const ZSTD_longOffset_e isLongOffset)
+{
+ return ZSTD_decompressSequences_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
+}
+#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG */
+
+
+
+#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT
+FORCE_INLINE_TEMPLATE seq_t
+ZSTD_decodeSequenceLong(seqState_t* seqState, ZSTD_longOffset_e const longOffsets)
+{
+ seq_t seq;
+ U32 const llBits = seqState->stateLL.table[seqState->stateLL.state].nbAdditionalBits;
+ U32 const mlBits = seqState->stateML.table[seqState->stateML.state].nbAdditionalBits;
+ U32 const ofBits = seqState->stateOffb.table[seqState->stateOffb.state].nbAdditionalBits;
+ U32 const totalBits = llBits+mlBits+ofBits;
+ U32 const llBase = seqState->stateLL.table[seqState->stateLL.state].baseValue;
+ U32 const mlBase = seqState->stateML.table[seqState->stateML.state].baseValue;
+ U32 const ofBase = seqState->stateOffb.table[seqState->stateOffb.state].baseValue;
+
+ /* sequence */
+ { size_t offset;
+ if (!ofBits)
+ offset = 0;
+ else {
+ ZSTD_STATIC_ASSERT(ZSTD_lo_isLongOffset == 1);
+ ZSTD_STATIC_ASSERT(LONG_OFFSETS_MAX_EXTRA_BITS_32 == 5);
+ assert(ofBits <= MaxOff);
+ if (MEM_32bits() && longOffsets) {
+ U32 const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN_32-1);
+ offset = ofBase + (BIT_readBitsFast(&seqState->DStream, ofBits - extraBits) << extraBits);
+ if (MEM_32bits() || extraBits) BIT_reloadDStream(&seqState->DStream);
+ if (extraBits) offset += BIT_readBitsFast(&seqState->DStream, extraBits);
+ } else {
+ offset = ofBase + BIT_readBitsFast(&seqState->DStream, ofBits); /* <= (ZSTD_WINDOWLOG_MAX-1) bits */
+ if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream);
+ }
+ }
+
+ if (ofBits <= 1) {
+ offset += (llBase==0);
+ if (offset) {
+ size_t temp = (offset==3) ? seqState->prevOffset[0] - 1 : seqState->prevOffset[offset];
+ temp += !temp; /* 0 is not valid; input is corrupted; force offset to 1 */
+ if (offset != 1) seqState->prevOffset[2] = seqState->prevOffset[1];
+ seqState->prevOffset[1] = seqState->prevOffset[0];
+ seqState->prevOffset[0] = offset = temp;
+ } else {
+ offset = seqState->prevOffset[0];
+ }
+ } else {
+ seqState->prevOffset[2] = seqState->prevOffset[1];
+ seqState->prevOffset[1] = seqState->prevOffset[0];
+ seqState->prevOffset[0] = offset;
+ }
+ seq.offset = offset;
+ }
+
+ seq.matchLength = mlBase + ((mlBits>0) ? BIT_readBitsFast(&seqState->DStream, mlBits) : 0); /* <= 16 bits */
+ if (MEM_32bits() && (mlBits+llBits >= STREAM_ACCUMULATOR_MIN_32-LONG_OFFSETS_MAX_EXTRA_BITS_32))
+ BIT_reloadDStream(&seqState->DStream);
+ if (MEM_64bits() && (totalBits >= STREAM_ACCUMULATOR_MIN_64-(LLFSELog+MLFSELog+OffFSELog)))
+ BIT_reloadDStream(&seqState->DStream);
+ /* Verify that there is enough bits to read the rest of the data in 64-bit mode. */
+ ZSTD_STATIC_ASSERT(16+LLFSELog+MLFSELog+OffFSELog < STREAM_ACCUMULATOR_MIN_64);
+
+ seq.litLength = llBase + ((llBits>0) ? BIT_readBitsFast(&seqState->DStream, llBits) : 0); /* <= 16 bits */
+ if (MEM_32bits())
+ BIT_reloadDStream(&seqState->DStream);
+
+ { size_t const pos = seqState->pos + seq.litLength;
+ const BYTE* const matchBase = (seq.offset > pos) ? seqState->dictEnd : seqState->prefixStart;
+ seq.match = matchBase + pos - seq.offset; /* note : this operation can overflow when seq.offset is really too large, which can only happen when input is corrupted.
+ * No consequence though : no memory access will occur, overly large offset will be detected in ZSTD_execSequenceLong() */
+ seqState->pos = pos + seq.matchLength;
+ }
+
+ /* ANS state update */
+ ZSTD_updateFseState(&seqState->stateLL, &seqState->DStream); /* <= 9 bits */
+ ZSTD_updateFseState(&seqState->stateML, &seqState->DStream); /* <= 9 bits */
+ if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream); /* <= 18 bits */
+ ZSTD_updateFseState(&seqState->stateOffb, &seqState->DStream); /* <= 8 bits */
+
+ return seq;
+}
+
+FORCE_INLINE_TEMPLATE size_t
+ZSTD_decompressSequencesLong_body(
+ ZSTD_DCtx* dctx,
+ void* dst, size_t maxDstSize,
+ const void* seqStart, size_t seqSize, int nbSeq,
+ const ZSTD_longOffset_e isLongOffset)
+{
+ const BYTE* ip = (const BYTE*)seqStart;
+ const BYTE* const iend = ip + seqSize;
+ BYTE* const ostart = (BYTE* const)dst;
+ BYTE* const oend = ostart + maxDstSize;
+ BYTE* op = ostart;
+ const BYTE* litPtr = dctx->litPtr;
+ const BYTE* const litEnd = litPtr + dctx->litSize;
+ const BYTE* const prefixStart = (const BYTE*) (dctx->prefixStart);
+ const BYTE* const dictStart = (const BYTE*) (dctx->virtualStart);
+ const BYTE* const dictEnd = (const BYTE*) (dctx->dictEnd);
+
+ /* Regen sequences */
+ if (nbSeq) {
+#define STORED_SEQS 4
+#define STORED_SEQS_MASK (STORED_SEQS-1)
+#define ADVANCED_SEQS 4
+ seq_t sequences[STORED_SEQS];
+ int const seqAdvance = MIN(nbSeq, ADVANCED_SEQS);
+ seqState_t seqState;
+ int seqNb;
+ dctx->fseEntropy = 1;
+ { int i; for (i=0; i<ZSTD_REP_NUM; i++) seqState.prevOffset[i] = dctx->entropy.rep[i]; }
+ seqState.prefixStart = prefixStart;
+ seqState.pos = (size_t)(op-prefixStart);
+ seqState.dictEnd = dictEnd;
+ assert(iend >= ip);
+ RETURN_ERROR_IF(
+ ERR_isError(BIT_initDStream(&seqState.DStream, ip, iend-ip)),
+ corruption_detected);
+ ZSTD_initFseState(&seqState.stateLL, &seqState.DStream, dctx->LLTptr);
+ ZSTD_initFseState(&seqState.stateOffb, &seqState.DStream, dctx->OFTptr);
+ ZSTD_initFseState(&seqState.stateML, &seqState.DStream, dctx->MLTptr);
+
+ /* prepare in advance */
+ for (seqNb=0; (BIT_reloadDStream(&seqState.DStream) <= BIT_DStream_completed) && (seqNb<seqAdvance); seqNb++) {
+ sequences[seqNb] = ZSTD_decodeSequenceLong(&seqState, isLongOffset);
+ PREFETCH_L1(sequences[seqNb].match); PREFETCH_L1(sequences[seqNb].match + sequences[seqNb].matchLength - 1); /* note : it's safe to invoke PREFETCH() on any memory address, including invalid ones */
+ }
+ RETURN_ERROR_IF(seqNb<seqAdvance, corruption_detected);
+
+ /* decode and decompress */
+ for ( ; (BIT_reloadDStream(&(seqState.DStream)) <= BIT_DStream_completed) && (seqNb<nbSeq) ; seqNb++) {
+ seq_t const sequence = ZSTD_decodeSequenceLong(&seqState, isLongOffset);
+ size_t const oneSeqSize = ZSTD_execSequenceLong(op, oend, sequences[(seqNb-ADVANCED_SEQS) & STORED_SEQS_MASK], &litPtr, litEnd, prefixStart, dictStart, dictEnd);
+ if (ZSTD_isError(oneSeqSize)) return oneSeqSize;
+ PREFETCH_L1(sequence.match); PREFETCH_L1(sequence.match + sequence.matchLength - 1); /* note : it's safe to invoke PREFETCH() on any memory address, including invalid ones */
+ sequences[seqNb & STORED_SEQS_MASK] = sequence;
+ op += oneSeqSize;
+ }
+ RETURN_ERROR_IF(seqNb<nbSeq, corruption_detected);
+
+ /* finish queue */
+ seqNb -= seqAdvance;
+ for ( ; seqNb<nbSeq ; seqNb++) {
+ size_t const oneSeqSize = ZSTD_execSequenceLong(op, oend, sequences[seqNb&STORED_SEQS_MASK], &litPtr, litEnd, prefixStart, dictStart, dictEnd);
+ if (ZSTD_isError(oneSeqSize)) return oneSeqSize;
+ op += oneSeqSize;
+ }
+
+ /* save reps for next block */
+ { U32 i; for (i=0; i<ZSTD_REP_NUM; i++) dctx->entropy.rep[i] = (U32)(seqState.prevOffset[i]); }
+ }
+
+ /* last literal segment */
+ { size_t const lastLLSize = litEnd - litPtr;
+ RETURN_ERROR_IF(lastLLSize > (size_t)(oend-op), dstSize_tooSmall);
+ memcpy(op, litPtr, lastLLSize);
+ op += lastLLSize;
+ }
+
+ return op-ostart;
+}
+
+static size_t
+ZSTD_decompressSequencesLong_default(ZSTD_DCtx* dctx,
+ void* dst, size_t maxDstSize,
+ const void* seqStart, size_t seqSize, int nbSeq,
+ const ZSTD_longOffset_e isLongOffset)
+{
+ return ZSTD_decompressSequencesLong_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
+}
+#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT */
+
+
+
+#if DYNAMIC_BMI2
+
+#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG
+static TARGET_ATTRIBUTE("bmi2") size_t
+DONT_VECTORIZE
+ZSTD_decompressSequences_bmi2(ZSTD_DCtx* dctx,
+ void* dst, size_t maxDstSize,
+ const void* seqStart, size_t seqSize, int nbSeq,
+ const ZSTD_longOffset_e isLongOffset)
+{
+ return ZSTD_decompressSequences_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
+}
+#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG */
+
+#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT
+static TARGET_ATTRIBUTE("bmi2") size_t
+ZSTD_decompressSequencesLong_bmi2(ZSTD_DCtx* dctx,
+ void* dst, size_t maxDstSize,
+ const void* seqStart, size_t seqSize, int nbSeq,
+ const ZSTD_longOffset_e isLongOffset)
+{
+ return ZSTD_decompressSequencesLong_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
+}
+#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT */
+
+#endif /* DYNAMIC_BMI2 */
+
+typedef size_t (*ZSTD_decompressSequences_t)(
+ ZSTD_DCtx* dctx,
+ void* dst, size_t maxDstSize,
+ const void* seqStart, size_t seqSize, int nbSeq,
+ const ZSTD_longOffset_e isLongOffset);
+
+#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG
+static size_t
+ZSTD_decompressSequences(ZSTD_DCtx* dctx, void* dst, size_t maxDstSize,
+ const void* seqStart, size_t seqSize, int nbSeq,
+ const ZSTD_longOffset_e isLongOffset)
+{
+ DEBUGLOG(5, "ZSTD_decompressSequences");
+#if DYNAMIC_BMI2
+ if (dctx->bmi2) {
+ return ZSTD_decompressSequences_bmi2(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
+ }
+#endif
+ return ZSTD_decompressSequences_default(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
+}
+#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG */
+
+
+#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT
+/* ZSTD_decompressSequencesLong() :
+ * decompression function triggered when a minimum share of offsets is considered "long",
+ * aka out of cache.
+ * note : "long" definition seems overloaded here, sometimes meaning "wider than bitstream register", and sometimes meaning "farther than memory cache distance".
+ * This function will try to mitigate main memory latency through the use of prefetching */
+static size_t
+ZSTD_decompressSequencesLong(ZSTD_DCtx* dctx,
+ void* dst, size_t maxDstSize,
+ const void* seqStart, size_t seqSize, int nbSeq,
+ const ZSTD_longOffset_e isLongOffset)
+{
+ DEBUGLOG(5, "ZSTD_decompressSequencesLong");
+#if DYNAMIC_BMI2
+ if (dctx->bmi2) {
+ return ZSTD_decompressSequencesLong_bmi2(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
+ }
+#endif
+ return ZSTD_decompressSequencesLong_default(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset);
+}
+#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT */
+
+
+
+#if !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT) && \
+ !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG)
+/* ZSTD_getLongOffsetsShare() :
+ * condition : offTable must be valid
+ * @return : "share" of long offsets (arbitrarily defined as > (1<<23))
+ * compared to maximum possible of (1<<OffFSELog) */
+static unsigned
+ZSTD_getLongOffsetsShare(const ZSTD_seqSymbol* offTable)
+{
+ const void* ptr = offTable;
+ U32 const tableLog = ((const ZSTD_seqSymbol_header*)ptr)[0].tableLog;
+ const ZSTD_seqSymbol* table = offTable + 1;
+ U32 const max = 1 << tableLog;
+ U32 u, total = 0;
+ DEBUGLOG(5, "ZSTD_getLongOffsetsShare: (tableLog=%u)", tableLog);
+
+ assert(max <= (1 << OffFSELog)); /* max not too large */
+ for (u=0; u<max; u++) {
+ if (table[u].nbAdditionalBits > 22) total += 1;
+ }
+
+ assert(tableLog <= OffFSELog);
+ total <<= (OffFSELog - tableLog); /* scale to OffFSELog */
+
+ return total;
+}
+#endif
+
+
+size_t
+ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize, const int frame)
+{ /* blockType == blockCompressed */
+ const BYTE* ip = (const BYTE*)src;
+ /* isLongOffset must be true if there are long offsets.
+ * Offsets are long if they are larger than 2^STREAM_ACCUMULATOR_MIN.
+ * We don't expect that to be the case in 64-bit mode.
+ * In block mode, window size is not known, so we have to be conservative.
+ * (note: but it could be evaluated from current-lowLimit)
+ */
+ ZSTD_longOffset_e const isLongOffset = (ZSTD_longOffset_e)(MEM_32bits() && (!frame || (dctx->fParams.windowSize > (1ULL << STREAM_ACCUMULATOR_MIN))));
+ DEBUGLOG(5, "ZSTD_decompressBlock_internal (size : %u)", (U32)srcSize);
+
+ RETURN_ERROR_IF(srcSize >= ZSTD_BLOCKSIZE_MAX, srcSize_wrong);
+
+ /* Decode literals section */
+ { size_t const litCSize = ZSTD_decodeLiteralsBlock(dctx, src, srcSize);
+ DEBUGLOG(5, "ZSTD_decodeLiteralsBlock : %u", (U32)litCSize);
+ if (ZSTD_isError(litCSize)) return litCSize;
+ ip += litCSize;
+ srcSize -= litCSize;
+ }
+
+ /* Build Decoding Tables */
+ {
+ /* These macros control at build-time which decompressor implementation
+ * we use. If neither is defined, we do some inspection and dispatch at
+ * runtime.
+ */
+#if !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT) && \
+ !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG)
+ int usePrefetchDecoder = dctx->ddictIsCold;
+#endif
+ int nbSeq;
+ size_t const seqHSize = ZSTD_decodeSeqHeaders(dctx, &nbSeq, ip, srcSize);
+ if (ZSTD_isError(seqHSize)) return seqHSize;
+ ip += seqHSize;
+ srcSize -= seqHSize;
+
+#if !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT) && \
+ !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG)
+ if ( !usePrefetchDecoder
+ && (!frame || (dctx->fParams.windowSize > (1<<24)))
+ && (nbSeq>ADVANCED_SEQS) ) { /* could probably use a larger nbSeq limit */
+ U32 const shareLongOffsets = ZSTD_getLongOffsetsShare(dctx->OFTptr);
+ U32 const minShare = MEM_64bits() ? 7 : 20; /* heuristic values, correspond to 2.73% and 7.81% */
+ usePrefetchDecoder = (shareLongOffsets >= minShare);
+ }
+#endif
+
+ dctx->ddictIsCold = 0;
+
+#if !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT) && \
+ !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG)
+ if (usePrefetchDecoder)
+#endif
+#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT
+ return ZSTD_decompressSequencesLong(dctx, dst, dstCapacity, ip, srcSize, nbSeq, isLongOffset);
+#endif
+
+#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG
+ /* else */
+ return ZSTD_decompressSequences(dctx, dst, dstCapacity, ip, srcSize, nbSeq, isLongOffset);
+#endif
+ }
+}
+
+
+size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize)
+{
+ size_t dSize;
+ ZSTD_checkContinuity(dctx, dst);
+ dSize = ZSTD_decompressBlock_internal(dctx, dst, dstCapacity, src, srcSize, /* frame */ 0);
+ dctx->previousDstEnd = (char*)dst + dSize;
+ return dSize;
+}
diff --git a/vendor/github.com/DataDog/zstd/zstd_decompress_block.h b/vendor/github.com/DataDog/zstd/zstd_decompress_block.h
new file mode 100644
index 0000000..7e92960
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/zstd_decompress_block.h
@@ -0,0 +1,59 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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_DEC_BLOCK_H
+#define ZSTD_DEC_BLOCK_H
+
+/*-*******************************************************
+ * Dependencies
+ *********************************************************/
+#include <stddef.h> /* size_t */
+#include "zstd.h" /* DCtx, and some public functions */
+#include "zstd_internal.h" /* blockProperties_t, and some public functions */
+#include "zstd_decompress_internal.h" /* ZSTD_seqSymbol */
+
+
+/* === Prototypes === */
+
+/* note: prototypes already published within `zstd.h` :
+ * ZSTD_decompressBlock()
+ */
+
+/* note: prototypes already published within `zstd_internal.h` :
+ * ZSTD_getcBlockSize()
+ * ZSTD_decodeSeqHeaders()
+ */
+
+
+/* ZSTD_decompressBlock_internal() :
+ * decompress block, starting at `src`,
+ * into destination buffer `dst`.
+ * @return : decompressed block size,
+ * or an error code (which can be tested using ZSTD_isError())
+ */
+size_t ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize, const int frame);
+
+/* ZSTD_buildFSETable() :
+ * generate FSE decoding table for one symbol (ll, ml or off)
+ * this function must be called with valid parameters only
+ * (dt is large enough, normalizedCounter distribution total is a power of 2, max is within range, etc.)
+ * in which case it cannot fail.
+ * Internal use only.
+ */
+void ZSTD_buildFSETable(ZSTD_seqSymbol* dt,
+ const short* normalizedCounter, unsigned maxSymbolValue,
+ const U32* baseValue, const U32* nbAdditionalBits,
+ unsigned tableLog);
+
+
+#endif /* ZSTD_DEC_BLOCK_H */
diff --git a/vendor/github.com/DataDog/zstd/zstd_decompress_internal.h b/vendor/github.com/DataDog/zstd/zstd_decompress_internal.h
new file mode 100644
index 0000000..ccbdfa0
--- /dev/null
+++ b/vendor/github.com/DataDog/zstd/zstd_decompress_internal.h
@@ -0,0 +1,175 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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.
+ */
+
+
+/* zstd_decompress_internal:
+ * objects and definitions shared within lib/decompress modules */
+
+ #ifndef ZSTD_DECOMPRESS_INTERNAL_H
+ #define ZSTD_DECOMPRESS_INTERNAL_H
+
+
+/*-*******************************************************
+ * Dependencies
+ *********************************************************/
+#include "mem.h" /* BYTE, U16, U32 */
+#include "zstd_internal.h" /* ZSTD_seqSymbol */
+
+
+
+/*-*******************************************************
+ * Constants
+ *********************************************************/
+static const U32 LL_base[MaxLL+1] = {
+ 0, 1, 2, 3, 4, 5, 6, 7,
+ 8, 9, 10, 11, 12, 13, 14, 15,
+ 16, 18, 20, 22, 24, 28, 32, 40,
+ 48, 64, 0x80, 0x100, 0x200, 0x400, 0x800, 0x1000,
+ 0x2000, 0x4000, 0x8000, 0x10000 };
+
+static const U32 OF_base[MaxOff+1] = {
+ 0, 1, 1, 5, 0xD, 0x1D, 0x3D, 0x7D,
+ 0xFD, 0x1FD, 0x3FD, 0x7FD, 0xFFD, 0x1FFD, 0x3FFD, 0x7FFD,
+ 0xFFFD, 0x1FFFD, 0x3FFFD, 0x7FFFD, 0xFFFFD, 0x1FFFFD, 0x3FFFFD, 0x7FFFFD,
+ 0xFFFFFD, 0x1FFFFFD, 0x3FFFFFD, 0x7FFFFFD, 0xFFFFFFD, 0x1FFFFFFD, 0x3FFFFFFD, 0x7FFFFFFD };
+
+static const U32 OF_bits[MaxOff+1] = {
+ 0, 1, 2, 3, 4, 5, 6, 7,
+ 8, 9, 10, 11, 12, 13, 14, 15,
+ 16, 17, 18, 19, 20, 21, 22, 23,
+ 24, 25, 26, 27, 28, 29, 30, 31 };
+
+static const U32 ML_base[MaxML+1] = {
+ 3, 4, 5, 6, 7, 8, 9, 10,
+ 11, 12, 13, 14, 15, 16, 17, 18,
+ 19, 20, 21, 22, 23, 24, 25, 26,
+ 27, 28, 29, 30, 31, 32, 33, 34,
+ 35, 37, 39, 41, 43, 47, 51, 59,
+ 67, 83, 99, 0x83, 0x103, 0x203, 0x403, 0x803,
+ 0x1003, 0x2003, 0x4003, 0x8003, 0x10003 };
+
+
+/*-*******************************************************
+ * Decompression types
+ *********************************************************/
+ typedef struct {
+ U32 fastMode;
+ U32 tableLog;
+ } ZSTD_seqSymbol_header;
+
+ typedef struct {
+ U16 nextState;
+ BYTE nbAdditionalBits;
+ BYTE nbBits;
+ U32 baseValue;
+ } ZSTD_seqSymbol;
+
+ #define SEQSYMBOL_TABLE_SIZE(log) (1 + (1 << (log)))
+
+typedef struct {
+ ZSTD_seqSymbol LLTable[SEQSYMBOL_TABLE_SIZE(LLFSELog)]; /* Note : Space reserved for FSE Tables */
+ ZSTD_seqSymbol OFTable[SEQSYMBOL_TABLE_SIZE(OffFSELog)]; /* is also used as temporary workspace while building hufTable during DDict creation */
+ ZSTD_seqSymbol MLTable[SEQSYMBOL_TABLE_SIZE(MLFSELog)]; /* and therefore must be at least HUF_DECOMPRESS_WORKSPACE_SIZE large */
+ HUF_DTable hufTable[HUF_DTABLE_SIZE(HufLog)]; /* can accommodate HUF_decompress4X */
+ U32 rep[ZSTD_REP_NUM];
+} ZSTD_entropyDTables_t;
+
+typedef enum { ZSTDds_getFrameHeaderSize, ZSTDds_decodeFrameHeader,
+ ZSTDds_decodeBlockHeader, ZSTDds_decompressBlock,
+ ZSTDds_decompressLastBlock, ZSTDds_checkChecksum,
+ ZSTDds_decodeSkippableHeader, ZSTDds_skipFrame } ZSTD_dStage;
+
+typedef enum { zdss_init=0, zdss_loadHeader,
+ zdss_read, zdss_load, zdss_flush } ZSTD_dStreamStage;
+
+typedef enum {
+ ZSTD_use_indefinitely = -1, /* Use the dictionary indefinitely */
+ ZSTD_dont_use = 0, /* Do not use the dictionary (if one exists free it) */
+ ZSTD_use_once = 1 /* Use the dictionary once and set to ZSTD_dont_use */
+} ZSTD_dictUses_e;
+
+struct ZSTD_DCtx_s
+{
+ const ZSTD_seqSymbol* LLTptr;
+ const ZSTD_seqSymbol* MLTptr;
+ const ZSTD_seqSymbol* OFTptr;
+ const HUF_DTable* HUFptr;
+ ZSTD_entropyDTables_t entropy;
+ U32 workspace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; /* space needed when building huffman tables */
+ const void* previousDstEnd; /* detect continuity */
+ const void* prefixStart; /* start of current segment */
+ const void* virtualStart; /* virtual start of previous segment if it was just before current one */
+ const void* dictEnd; /* end of previous segment */
+ size_t expected;
+ ZSTD_frameHeader fParams;
+ U64 decodedSize;
+ blockType_e bType; /* used in ZSTD_decompressContinue(), store blockType between block header decoding and block decompression stages */
+ ZSTD_dStage stage;
+ U32 litEntropy;
+ U32 fseEntropy;
+ XXH64_state_t xxhState;
+ size_t headerSize;
+ ZSTD_format_e format;
+ const BYTE* litPtr;
+ ZSTD_customMem customMem;
+ size_t litSize;
+ size_t rleSize;
+ size_t staticSize;
+ int bmi2; /* == 1 if the CPU supports BMI2 and 0 otherwise. CPU support is determined dynamically once per context lifetime. */
+
+ /* dictionary */
+ ZSTD_DDict* ddictLocal;
+ const ZSTD_DDict* ddict; /* set by ZSTD_initDStream_usingDDict(), or ZSTD_DCtx_refDDict() */
+ U32 dictID;
+ int ddictIsCold; /* if == 1 : dictionary is "new" for working context, and presumed "cold" (not in cpu cache) */
+ ZSTD_dictUses_e dictUses;
+
+ /* streaming */
+ ZSTD_dStreamStage streamStage;
+ char* inBuff;
+ size_t inBuffSize;
+ size_t inPos;
+ size_t maxWindowSize;
+ char* outBuff;
+ size_t outBuffSize;
+ size_t outStart;
+ size_t outEnd;
+ size_t lhSize;
+ void* legacyContext;
+ U32 previousLegacyVersion;
+ U32 legacyVersion;
+ U32 hostageByte;
+ int noForwardProgress;
+
+ /* workspace */
+ BYTE litBuffer[ZSTD_BLOCKSIZE_MAX + WILDCOPY_OVERLENGTH];
+ BYTE headerBuffer[ZSTD_FRAMEHEADERSIZE_MAX];
+}; /* typedef'd to ZSTD_DCtx within "zstd.h" */
+
+
+/*-*******************************************************
+ * Shared internal functions
+ *********************************************************/
+
+/*! ZSTD_loadDEntropy() :
+ * dict : must point at beginning of a valid zstd dictionary.
+ * @return : size of entropy tables read */
+size_t ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy,
+ const void* const dict, size_t const dictSize);
+
+/*! ZSTD_checkContinuity() :
+ * check if next `dst` follows previous position, where decompression ended.
+ * If yes, do nothing (continue on current segment).
+ * If not, classify previous segment as "external dictionary", and start a new segment.
+ * This function cannot fail. */
+void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst);
+
+
+#endif /* ZSTD_DECOMPRESS_INTERNAL_H */
diff --git a/vendor/github.com/DataDog/zstd/zstd_double_fast.c b/vendor/github.com/DataDog/zstd/zstd_double_fast.c
index 86e6b39..5957255 100644
--- a/vendor/github.com/DataDog/zstd/zstd_double_fast.c
+++ b/vendor/github.com/DataDog/zstd/zstd_double_fast.c
@@ -13,12 +13,12 @@
void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
- ZSTD_compressionParameters const* cParams,
- void const* end)
+ void const* end, ZSTD_dictTableLoadMethod_e dtlm)
{
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashLarge = ms->hashTable;
U32 const hBitsL = cParams->hashLog;
- U32 const mls = cParams->searchLength;
+ U32 const mls = cParams->minMatch;
U32* const hashSmall = ms->chainTable;
U32 const hBitsS = cParams->chainLog;
const BYTE* const base = ms->window.base;
@@ -40,17 +40,20 @@
hashSmall[smHash] = current + i;
if (i == 0 || hashLarge[lgHash] == 0)
hashLarge[lgHash] = current + i;
- }
- }
+ /* Only load extra positions for ZSTD_dtlm_full */
+ if (dtlm == ZSTD_dtlm_fast)
+ break;
+ } }
}
FORCE_INLINE_TEMPLATE
size_t ZSTD_compressBlock_doubleFast_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize,
- U32 const mls /* template */)
+ void const* src, size_t srcSize,
+ U32 const mls /* template */, ZSTD_dictMode_e const dictMode)
{
+ ZSTD_compressionParameters const* cParams = &ms->cParams;
U32* const hashLong = ms->hashTable;
const U32 hBitsL = cParams->hashLog;
U32* const hashSmall = ms->chainTable;
@@ -59,235 +62,424 @@
const BYTE* const istart = (const BYTE*)src;
const BYTE* ip = istart;
const BYTE* anchor = istart;
- const U32 lowestIndex = ms->window.dictLimit;
- const BYTE* const lowest = base + lowestIndex;
+ const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
+ const U32 lowestValid = ms->window.dictLimit;
+ const U32 maxDistance = 1U << cParams->windowLog;
+ const U32 prefixLowestIndex = (endIndex - lowestValid > maxDistance) ? endIndex - maxDistance : lowestValid;
+ const BYTE* const prefixLowest = base + prefixLowestIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - HASH_READ_SIZE;
U32 offset_1=rep[0], offset_2=rep[1];
U32 offsetSaved = 0;
+ const ZSTD_matchState_t* const dms = ms->dictMatchState;
+ const ZSTD_compressionParameters* const dictCParams =
+ dictMode == ZSTD_dictMatchState ?
+ &dms->cParams : NULL;
+ const U32* const dictHashLong = dictMode == ZSTD_dictMatchState ?
+ dms->hashTable : NULL;
+ const U32* const dictHashSmall = dictMode == ZSTD_dictMatchState ?
+ dms->chainTable : NULL;
+ const U32 dictStartIndex = dictMode == ZSTD_dictMatchState ?
+ dms->window.dictLimit : 0;
+ const BYTE* const dictBase = dictMode == ZSTD_dictMatchState ?
+ dms->window.base : NULL;
+ const BYTE* const dictStart = dictMode == ZSTD_dictMatchState ?
+ dictBase + dictStartIndex : NULL;
+ const BYTE* const dictEnd = dictMode == ZSTD_dictMatchState ?
+ dms->window.nextSrc : NULL;
+ const U32 dictIndexDelta = dictMode == ZSTD_dictMatchState ?
+ prefixLowestIndex - (U32)(dictEnd - dictBase) :
+ 0;
+ const U32 dictHBitsL = dictMode == ZSTD_dictMatchState ?
+ dictCParams->hashLog : hBitsL;
+ const U32 dictHBitsS = dictMode == ZSTD_dictMatchState ?
+ dictCParams->chainLog : hBitsS;
+ const U32 dictAndPrefixLength = (U32)(ip - prefixLowest + dictEnd - dictStart);
+
+ DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_generic");
+
+ assert(dictMode == ZSTD_noDict || dictMode == ZSTD_dictMatchState);
+
+ /* if a dictionary is attached, it must be within window range */
+ if (dictMode == ZSTD_dictMatchState) {
+ assert(lowestValid + maxDistance >= endIndex);
+ }
+
/* init */
- ip += (ip==lowest);
- { U32 const maxRep = (U32)(ip-lowest);
+ ip += (dictAndPrefixLength == 0);
+ if (dictMode == ZSTD_noDict) {
+ U32 const maxRep = (U32)(ip - prefixLowest);
if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0;
if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0;
}
+ if (dictMode == ZSTD_dictMatchState) {
+ /* dictMatchState repCode checks don't currently handle repCode == 0
+ * disabling. */
+ assert(offset_1 <= dictAndPrefixLength);
+ assert(offset_2 <= dictAndPrefixLength);
+ }
/* Main Search Loop */
while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */
size_t mLength;
+ U32 offset;
size_t const h2 = ZSTD_hashPtr(ip, hBitsL, 8);
size_t const h = ZSTD_hashPtr(ip, hBitsS, mls);
+ size_t const dictHL = ZSTD_hashPtr(ip, dictHBitsL, 8);
+ size_t const dictHS = ZSTD_hashPtr(ip, dictHBitsS, mls);
U32 const current = (U32)(ip-base);
U32 const matchIndexL = hashLong[h2];
- U32 const matchIndexS = hashSmall[h];
+ U32 matchIndexS = hashSmall[h];
const BYTE* matchLong = base + matchIndexL;
const BYTE* match = base + matchIndexS;
+ const U32 repIndex = current + 1 - offset_1;
+ const BYTE* repMatch = (dictMode == ZSTD_dictMatchState
+ && repIndex < prefixLowestIndex) ?
+ dictBase + (repIndex - dictIndexDelta) :
+ base + repIndex;
hashLong[h2] = hashSmall[h] = current; /* update hash tables */
- assert(offset_1 <= current); /* supposed guaranteed by construction */
- if ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1))) {
- /* favor repcode */
+ /* check dictMatchState repcode */
+ if (dictMode == ZSTD_dictMatchState
+ && ((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */)
+ && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
+ const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
+ mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
+ ip++;
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, 0, mLength-MINMATCH);
+ goto _match_stored;
+ }
+
+ /* check noDict repcode */
+ if ( dictMode == ZSTD_noDict
+ && ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1)))) {
mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
ip++;
- ZSTD_storeSeq(seqStore, ip-anchor, anchor, 0, mLength-MINMATCH);
- } else {
- U32 offset;
- if ( (matchIndexL > lowestIndex) && (MEM_read64(matchLong) == MEM_read64(ip)) ) {
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, 0, mLength-MINMATCH);
+ goto _match_stored;
+ }
+
+ if (matchIndexL > prefixLowestIndex) {
+ /* check prefix long match */
+ if (MEM_read64(matchLong) == MEM_read64(ip)) {
mLength = ZSTD_count(ip+8, matchLong+8, iend) + 8;
offset = (U32)(ip-matchLong);
- while (((ip>anchor) & (matchLong>lowest)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
- } else if ( (matchIndexS > lowestIndex) && (MEM_read32(match) == MEM_read32(ip)) ) {
- size_t const hl3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
- U32 const matchIndexL3 = hashLong[hl3];
- const BYTE* matchL3 = base + matchIndexL3;
- hashLong[hl3] = current + 1;
- if ( (matchIndexL3 > lowestIndex) && (MEM_read64(matchL3) == MEM_read64(ip+1)) ) {
+ while (((ip>anchor) & (matchLong>prefixLowest)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
+ goto _match_found;
+ }
+ } else if (dictMode == ZSTD_dictMatchState) {
+ /* check dictMatchState long match */
+ U32 const dictMatchIndexL = dictHashLong[dictHL];
+ const BYTE* dictMatchL = dictBase + dictMatchIndexL;
+ assert(dictMatchL < dictEnd);
+
+ if (dictMatchL > dictStart && MEM_read64(dictMatchL) == MEM_read64(ip)) {
+ mLength = ZSTD_count_2segments(ip+8, dictMatchL+8, iend, dictEnd, prefixLowest) + 8;
+ offset = (U32)(current - dictMatchIndexL - dictIndexDelta);
+ while (((ip>anchor) & (dictMatchL>dictStart)) && (ip[-1] == dictMatchL[-1])) { ip--; dictMatchL--; mLength++; } /* catch up */
+ goto _match_found;
+ } }
+
+ if (matchIndexS > prefixLowestIndex) {
+ /* check prefix short match */
+ if (MEM_read32(match) == MEM_read32(ip)) {
+ goto _search_next_long;
+ }
+ } else if (dictMode == ZSTD_dictMatchState) {
+ /* check dictMatchState short match */
+ U32 const dictMatchIndexS = dictHashSmall[dictHS];
+ match = dictBase + dictMatchIndexS;
+ matchIndexS = dictMatchIndexS + dictIndexDelta;
+
+ if (match > dictStart && MEM_read32(match) == MEM_read32(ip)) {
+ goto _search_next_long;
+ } }
+
+ ip += ((ip-anchor) >> kSearchStrength) + 1;
+ continue;
+
+_search_next_long:
+
+ { size_t const hl3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
+ size_t const dictHLNext = ZSTD_hashPtr(ip+1, dictHBitsL, 8);
+ U32 const matchIndexL3 = hashLong[hl3];
+ const BYTE* matchL3 = base + matchIndexL3;
+ hashLong[hl3] = current + 1;
+
+ /* check prefix long +1 match */
+ if (matchIndexL3 > prefixLowestIndex) {
+ if (MEM_read64(matchL3) == MEM_read64(ip+1)) {
mLength = ZSTD_count(ip+9, matchL3+8, iend) + 8;
ip++;
offset = (U32)(ip-matchL3);
- while (((ip>anchor) & (matchL3>lowest)) && (ip[-1] == matchL3[-1])) { ip--; matchL3--; mLength++; } /* catch up */
- } else {
- mLength = ZSTD_count(ip+4, match+4, iend) + 4;
- offset = (U32)(ip-match);
- while (((ip>anchor) & (match>lowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
+ while (((ip>anchor) & (matchL3>prefixLowest)) && (ip[-1] == matchL3[-1])) { ip--; matchL3--; mLength++; } /* catch up */
+ goto _match_found;
}
- } else {
- ip += ((ip-anchor) >> kSearchStrength) + 1;
- continue;
- }
+ } else if (dictMode == ZSTD_dictMatchState) {
+ /* check dict long +1 match */
+ U32 const dictMatchIndexL3 = dictHashLong[dictHLNext];
+ const BYTE* dictMatchL3 = dictBase + dictMatchIndexL3;
+ assert(dictMatchL3 < dictEnd);
+ if (dictMatchL3 > dictStart && MEM_read64(dictMatchL3) == MEM_read64(ip+1)) {
+ mLength = ZSTD_count_2segments(ip+1+8, dictMatchL3+8, iend, dictEnd, prefixLowest) + 8;
+ ip++;
+ offset = (U32)(current + 1 - dictMatchIndexL3 - dictIndexDelta);
+ while (((ip>anchor) & (dictMatchL3>dictStart)) && (ip[-1] == dictMatchL3[-1])) { ip--; dictMatchL3--; mLength++; } /* catch up */
+ goto _match_found;
+ } } }
- offset_2 = offset_1;
- offset_1 = offset;
-
- ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ /* if no long +1 match, explore the short match we found */
+ if (dictMode == ZSTD_dictMatchState && matchIndexS < prefixLowestIndex) {
+ mLength = ZSTD_count_2segments(ip+4, match+4, iend, dictEnd, prefixLowest) + 4;
+ offset = (U32)(current - matchIndexS);
+ while (((ip>anchor) & (match>dictStart)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
+ } else {
+ mLength = ZSTD_count(ip+4, match+4, iend) + 4;
+ offset = (U32)(ip - match);
+ while (((ip>anchor) & (match>prefixLowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
}
+ /* fall-through */
+
+_match_found:
+ offset_2 = offset_1;
+ offset_1 = offset;
+
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+
+_match_stored:
/* match found */
ip += mLength;
anchor = ip;
if (ip <= ilimit) {
- /* Fill Table */
- hashLong[ZSTD_hashPtr(base+current+2, hBitsL, 8)] =
- hashSmall[ZSTD_hashPtr(base+current+2, hBitsS, mls)] = current+2; /* here because current+2 could be > iend-8 */
- hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] =
- hashSmall[ZSTD_hashPtr(ip-2, hBitsS, mls)] = (U32)(ip-2-base);
+ /* Complementary insertion */
+ /* done after iLimit test, as candidates could be > iend-8 */
+ { U32 const indexToInsert = current+2;
+ hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
+ hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
+ hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
+ hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
+ }
/* check immediate repcode */
- while ( (ip <= ilimit)
- && ( (offset_2>0)
- & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
- /* store sequence */
- size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
- { U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; } /* swap offset_2 <=> offset_1 */
- hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip-base);
- hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip-base);
- ZSTD_storeSeq(seqStore, 0, anchor, 0, rLength-MINMATCH);
- ip += rLength;
- anchor = ip;
- continue; /* faster when present ... (?) */
- } } }
+ if (dictMode == ZSTD_dictMatchState) {
+ while (ip <= ilimit) {
+ U32 const current2 = (U32)(ip-base);
+ U32 const repIndex2 = current2 - offset_2;
+ const BYTE* repMatch2 = dictMode == ZSTD_dictMatchState
+ && repIndex2 < prefixLowestIndex ?
+ dictBase - dictIndexDelta + repIndex2 :
+ base + repIndex2;
+ if ( ((U32)((prefixLowestIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */)
+ && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
+ const BYTE* const repEnd2 = repIndex2 < prefixLowestIndex ? dictEnd : iend;
+ size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixLowest) + 4;
+ U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
+ ZSTD_storeSeq(seqStore, 0, anchor, 0, repLength2-MINMATCH);
+ hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
+ hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
+ ip += repLength2;
+ anchor = ip;
+ continue;
+ }
+ break;
+ } }
+
+ if (dictMode == ZSTD_noDict) {
+ while ( (ip <= ilimit)
+ && ( (offset_2>0)
+ & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
+ /* store sequence */
+ size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
+ U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; /* swap offset_2 <=> offset_1 */
+ hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip-base);
+ hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip-base);
+ ZSTD_storeSeq(seqStore, 0, anchor, 0, rLength-MINMATCH);
+ ip += rLength;
+ anchor = ip;
+ continue; /* faster when present ... (?) */
+ } } }
+ } /* while (ip < ilimit) */
/* save reps for next block */
rep[0] = offset_1 ? offset_1 : offsetSaved;
rep[1] = offset_2 ? offset_2 : offsetSaved;
/* Return the last literals size */
- return iend - anchor;
+ return (size_t)(iend - anchor);
}
size_t ZSTD_compressBlock_doubleFast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- const U32 mls = cParams->searchLength;
+ const U32 mls = ms->cParams.minMatch;
switch(mls)
{
default: /* includes case 3 */
case 4 :
- return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, cParams, src, srcSize, 4);
+ return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 4, ZSTD_noDict);
case 5 :
- return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, cParams, src, srcSize, 5);
+ return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 5, ZSTD_noDict);
case 6 :
- return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, cParams, src, srcSize, 6);
+ return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 6, ZSTD_noDict);
case 7 :
- return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, cParams, src, srcSize, 7);
+ return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 7, ZSTD_noDict);
+ }
+}
+
+
+size_t ZSTD_compressBlock_doubleFast_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize)
+{
+ const U32 mls = ms->cParams.minMatch;
+ switch(mls)
+ {
+ default: /* includes case 3 */
+ case 4 :
+ return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 4, ZSTD_dictMatchState);
+ case 5 :
+ return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 5, ZSTD_dictMatchState);
+ case 6 :
+ return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 6, ZSTD_dictMatchState);
+ case 7 :
+ return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 7, ZSTD_dictMatchState);
}
}
static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize,
+ void const* src, size_t srcSize,
U32 const mls /* template */)
{
+ ZSTD_compressionParameters const* cParams = &ms->cParams;
U32* const hashLong = ms->hashTable;
U32 const hBitsL = cParams->hashLog;
U32* const hashSmall = ms->chainTable;
U32 const hBitsS = cParams->chainLog;
- const BYTE* const base = ms->window.base;
- const BYTE* const dictBase = ms->window.dictBase;
const BYTE* const istart = (const BYTE*)src;
const BYTE* ip = istart;
const BYTE* anchor = istart;
- const U32 lowestIndex = ms->window.lowLimit;
- const BYTE* const dictStart = dictBase + lowestIndex;
- const U32 dictLimit = ms->window.dictLimit;
- const BYTE* const lowPrefixPtr = base + dictLimit;
- const BYTE* const dictEnd = dictBase + dictLimit;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - 8;
+ const BYTE* const base = ms->window.base;
+ const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
+ const U32 maxDistance = 1U << cParams->windowLog;
+ const U32 lowestValid = ms->window.lowLimit;
+ const U32 lowLimit = (endIndex - lowestValid > maxDistance) ? endIndex - maxDistance : lowestValid;
+ const U32 dictStartIndex = lowLimit;
+ const U32 dictLimit = ms->window.dictLimit;
+ const U32 prefixStartIndex = (dictLimit > lowLimit) ? dictLimit : lowLimit;
+ const BYTE* const prefixStart = base + prefixStartIndex;
+ const BYTE* const dictBase = ms->window.dictBase;
+ const BYTE* const dictStart = dictBase + dictStartIndex;
+ const BYTE* const dictEnd = dictBase + prefixStartIndex;
U32 offset_1=rep[0], offset_2=rep[1];
+ DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_extDict_generic (srcSize=%zu)", srcSize);
+
+ /* if extDict is invalidated due to maxDistance, switch to "regular" variant */
+ if (prefixStartIndex == dictStartIndex)
+ return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, mls, ZSTD_noDict);
+
/* Search Loop */
while (ip < ilimit) { /* < instead of <=, because (ip+1) */
const size_t hSmall = ZSTD_hashPtr(ip, hBitsS, mls);
const U32 matchIndex = hashSmall[hSmall];
- const BYTE* matchBase = matchIndex < dictLimit ? dictBase : base;
+ const BYTE* const matchBase = matchIndex < prefixStartIndex ? dictBase : base;
const BYTE* match = matchBase + matchIndex;
const size_t hLong = ZSTD_hashPtr(ip, hBitsL, 8);
const U32 matchLongIndex = hashLong[hLong];
- const BYTE* matchLongBase = matchLongIndex < dictLimit ? dictBase : base;
+ const BYTE* const matchLongBase = matchLongIndex < prefixStartIndex ? dictBase : base;
const BYTE* matchLong = matchLongBase + matchLongIndex;
const U32 current = (U32)(ip-base);
const U32 repIndex = current + 1 - offset_1; /* offset_1 expected <= current +1 */
- const BYTE* repBase = repIndex < dictLimit ? dictBase : base;
- const BYTE* repMatch = repBase + repIndex;
+ const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base;
+ const BYTE* const repMatch = repBase + repIndex;
size_t mLength;
hashSmall[hSmall] = hashLong[hLong] = current; /* update hash table */
- if ( (((U32)((dictLimit-1) - repIndex) >= 3) /* intentional underflow */ & (repIndex > lowestIndex))
- && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
- const BYTE* repMatchEnd = repIndex < dictLimit ? dictEnd : iend;
- mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, lowPrefixPtr) + 4;
+ if ((((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex doesn't overlap dict + prefix */
+ & (repIndex > dictStartIndex))
+ && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
+ const BYTE* repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
+ mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
ip++;
- ZSTD_storeSeq(seqStore, ip-anchor, anchor, 0, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, 0, mLength-MINMATCH);
} else {
- if ((matchLongIndex > lowestIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) {
- const BYTE* matchEnd = matchLongIndex < dictLimit ? dictEnd : iend;
- const BYTE* lowMatchPtr = matchLongIndex < dictLimit ? dictStart : lowPrefixPtr;
+ if ((matchLongIndex > dictStartIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) {
+ const BYTE* const matchEnd = matchLongIndex < prefixStartIndex ? dictEnd : iend;
+ const BYTE* const lowMatchPtr = matchLongIndex < prefixStartIndex ? dictStart : prefixStart;
U32 offset;
- mLength = ZSTD_count_2segments(ip+8, matchLong+8, iend, matchEnd, lowPrefixPtr) + 8;
+ mLength = ZSTD_count_2segments(ip+8, matchLong+8, iend, matchEnd, prefixStart) + 8;
offset = current - matchLongIndex;
while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
offset_2 = offset_1;
offset_1 = offset;
- ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
- } else if ((matchIndex > lowestIndex) && (MEM_read32(match) == MEM_read32(ip))) {
+ } else if ((matchIndex > dictStartIndex) && (MEM_read32(match) == MEM_read32(ip))) {
size_t const h3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
U32 const matchIndex3 = hashLong[h3];
- const BYTE* const match3Base = matchIndex3 < dictLimit ? dictBase : base;
+ const BYTE* const match3Base = matchIndex3 < prefixStartIndex ? dictBase : base;
const BYTE* match3 = match3Base + matchIndex3;
U32 offset;
hashLong[h3] = current + 1;
- if ( (matchIndex3 > lowestIndex) && (MEM_read64(match3) == MEM_read64(ip+1)) ) {
- const BYTE* matchEnd = matchIndex3 < dictLimit ? dictEnd : iend;
- const BYTE* lowMatchPtr = matchIndex3 < dictLimit ? dictStart : lowPrefixPtr;
- mLength = ZSTD_count_2segments(ip+9, match3+8, iend, matchEnd, lowPrefixPtr) + 8;
+ if ( (matchIndex3 > dictStartIndex) && (MEM_read64(match3) == MEM_read64(ip+1)) ) {
+ const BYTE* const matchEnd = matchIndex3 < prefixStartIndex ? dictEnd : iend;
+ const BYTE* const lowMatchPtr = matchIndex3 < prefixStartIndex ? dictStart : prefixStart;
+ mLength = ZSTD_count_2segments(ip+9, match3+8, iend, matchEnd, prefixStart) + 8;
ip++;
offset = current+1 - matchIndex3;
while (((ip>anchor) & (match3>lowMatchPtr)) && (ip[-1] == match3[-1])) { ip--; match3--; mLength++; } /* catch up */
} else {
- const BYTE* matchEnd = matchIndex < dictLimit ? dictEnd : iend;
- const BYTE* lowMatchPtr = matchIndex < dictLimit ? dictStart : lowPrefixPtr;
- mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, lowPrefixPtr) + 4;
+ const BYTE* const matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend;
+ const BYTE* const lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart;
+ mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4;
offset = current - matchIndex;
while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
}
offset_2 = offset_1;
offset_1 = offset;
- ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
} else {
ip += ((ip-anchor) >> kSearchStrength) + 1;
continue;
} }
- /* found a match : store it */
+ /* move to next sequence start */
ip += mLength;
anchor = ip;
if (ip <= ilimit) {
- /* Fill Table */
- hashSmall[ZSTD_hashPtr(base+current+2, hBitsS, mls)] = current+2;
- hashLong[ZSTD_hashPtr(base+current+2, hBitsL, 8)] = current+2;
- hashSmall[ZSTD_hashPtr(ip-2, hBitsS, mls)] = (U32)(ip-2-base);
- hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
+ /* Complementary insertion */
+ /* done after iLimit test, as candidates could be > iend-8 */
+ { U32 const indexToInsert = current+2;
+ hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
+ hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
+ hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
+ hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
+ }
+
/* check immediate repcode */
while (ip <= ilimit) {
U32 const current2 = (U32)(ip-base);
U32 const repIndex2 = current2 - offset_2;
- const BYTE* repMatch2 = repIndex2 < dictLimit ? dictBase + repIndex2 : base + repIndex2;
- if ( (((U32)((dictLimit-1) - repIndex2) >= 3) & (repIndex2 > lowestIndex)) /* intentional overflow */
- && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
- const BYTE* const repEnd2 = repIndex2 < dictLimit ? dictEnd : iend;
- size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, lowPrefixPtr) + 4;
- U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
+ const BYTE* repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
+ if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) /* intentional overflow : ensure repIndex2 doesn't overlap dict + prefix */
+ & (repIndex2 > dictStartIndex))
+ && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
+ const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
+ size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
+ U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
ZSTD_storeSeq(seqStore, 0, anchor, 0, repLength2-MINMATCH);
hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
@@ -303,25 +495,25 @@
rep[1] = offset_2;
/* Return the last literals size */
- return iend - anchor;
+ return (size_t)(iend - anchor);
}
size_t ZSTD_compressBlock_doubleFast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- U32 const mls = cParams->searchLength;
+ U32 const mls = ms->cParams.minMatch;
switch(mls)
{
default: /* includes case 3 */
case 4 :
- return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, cParams, src, srcSize, 4);
+ return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 4);
case 5 :
- return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, cParams, src, srcSize, 5);
+ return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 5);
case 6 :
- return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, cParams, src, srcSize, 6);
+ return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 6);
case 7 :
- return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, cParams, src, srcSize, 7);
+ return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 7);
}
}
diff --git a/vendor/github.com/DataDog/zstd/zstd_double_fast.h b/vendor/github.com/DataDog/zstd/zstd_double_fast.h
index 6d80b27..4fa31ac 100644
--- a/vendor/github.com/DataDog/zstd/zstd_double_fast.h
+++ b/vendor/github.com/DataDog/zstd/zstd_double_fast.h
@@ -19,14 +19,16 @@
#include "zstd_compress_internal.h" /* ZSTD_CCtx, size_t */
void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
- ZSTD_compressionParameters const* cParams,
- void const* end);
+ void const* end, ZSTD_dictTableLoadMethod_e dtlm);
size_t ZSTD_compressBlock_doubleFast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
+size_t ZSTD_compressBlock_doubleFast_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_doubleFast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
#if defined (__cplusplus)
diff --git a/vendor/github.com/DataDog/zstd/zstd_errors.h b/vendor/github.com/DataDog/zstd/zstd_errors.h
index 57533f2..92a3433 100644
--- a/vendor/github.com/DataDog/zstd/zstd_errors.h
+++ b/vendor/github.com/DataDog/zstd/zstd_errors.h
@@ -72,6 +72,7 @@
ZSTD_error_workSpace_tooSmall= 66,
ZSTD_error_dstSize_tooSmall = 70,
ZSTD_error_srcSize_wrong = 72,
+ ZSTD_error_dstBuffer_null = 74,
/* following error codes are __NOT STABLE__, they can be removed or changed in future versions */
ZSTD_error_frameIndex_tooLarge = 100,
ZSTD_error_seekableIO = 102,
diff --git a/vendor/github.com/DataDog/zstd/zstd_fast.c b/vendor/github.com/DataDog/zstd/zstd_fast.c
index df4d28b..a05b8a4 100644
--- a/vendor/github.com/DataDog/zstd/zstd_fast.c
+++ b/vendor/github.com/DataDog/zstd/zstd_fast.c
@@ -13,12 +13,13 @@
void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
- ZSTD_compressionParameters const* cParams,
- void const* end)
+ const void* const end,
+ ZSTD_dictTableLoadMethod_e dtlm)
{
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hBits = cParams->hashLog;
- U32 const mls = cParams->searchLength;
+ U32 const mls = cParams->minMatch;
const BYTE* const base = ms->window.base;
const BYTE* ip = base + ms->nextToUpdate;
const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
@@ -27,69 +28,278 @@
/* Always insert every fastHashFillStep position into the hash table.
* Insert the other positions if their hash entry is empty.
*/
- for (; ip + fastHashFillStep - 1 <= iend; ip += fastHashFillStep) {
+ for ( ; ip + fastHashFillStep < iend + 2; ip += fastHashFillStep) {
U32 const current = (U32)(ip - base);
- U32 i;
- for (i = 0; i < fastHashFillStep; ++i) {
- size_t const hash = ZSTD_hashPtr(ip + i, hBits, mls);
- if (i == 0 || hashTable[hash] == 0)
- hashTable[hash] = current + i;
- }
- }
+ size_t const hash0 = ZSTD_hashPtr(ip, hBits, mls);
+ hashTable[hash0] = current;
+ if (dtlm == ZSTD_dtlm_fast) continue;
+ /* Only load extra positions for ZSTD_dtlm_full */
+ { U32 p;
+ for (p = 1; p < fastHashFillStep; ++p) {
+ size_t const hash = ZSTD_hashPtr(ip + p, hBits, mls);
+ if (hashTable[hash] == 0) { /* not yet filled */
+ hashTable[hash] = current + p;
+ } } } }
}
+
FORCE_INLINE_TEMPLATE
size_t ZSTD_compressBlock_fast_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize,
- U32 const hlog, U32 const stepSize, U32 const mls)
+ U32 const mls)
{
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
+ U32 const hlog = cParams->hashLog;
+ /* support stepSize of 0 */
+ size_t const stepSize = cParams->targetLength + !(cParams->targetLength) + 1;
const BYTE* const base = ms->window.base;
const BYTE* const istart = (const BYTE*)src;
- const BYTE* ip = istart;
+ /* We check ip0 (ip + 0) and ip1 (ip + 1) each loop */
+ const BYTE* ip0 = istart;
+ const BYTE* ip1;
const BYTE* anchor = istart;
- const U32 lowestIndex = ms->window.dictLimit;
- const BYTE* const lowest = base + lowestIndex;
+ const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
+ const U32 maxDistance = 1U << cParams->windowLog;
+ const U32 validStartIndex = ms->window.dictLimit;
+ const U32 prefixStartIndex = (endIndex - validStartIndex > maxDistance) ? endIndex - maxDistance : validStartIndex;
+ const BYTE* const prefixStart = base + prefixStartIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - HASH_READ_SIZE;
U32 offset_1=rep[0], offset_2=rep[1];
U32 offsetSaved = 0;
/* init */
- ip += (ip==lowest);
- { U32 const maxRep = (U32)(ip-lowest);
+ ip0 += (ip0 == prefixStart);
+ ip1 = ip0 + 1;
+ {
+ U32 const maxRep = (U32)(ip0 - prefixStart);
if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0;
if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0;
}
/* Main Search Loop */
+ while (ip1 < ilimit) { /* < instead of <=, because check at ip0+2 */
+ size_t mLength;
+ BYTE const* ip2 = ip0 + 2;
+ size_t const h0 = ZSTD_hashPtr(ip0, hlog, mls);
+ U32 const val0 = MEM_read32(ip0);
+ size_t const h1 = ZSTD_hashPtr(ip1, hlog, mls);
+ U32 const val1 = MEM_read32(ip1);
+ U32 const current0 = (U32)(ip0-base);
+ U32 const current1 = (U32)(ip1-base);
+ U32 const matchIndex0 = hashTable[h0];
+ U32 const matchIndex1 = hashTable[h1];
+ BYTE const* repMatch = ip2-offset_1;
+ const BYTE* match0 = base + matchIndex0;
+ const BYTE* match1 = base + matchIndex1;
+ U32 offcode;
+ hashTable[h0] = current0; /* update hash table */
+ hashTable[h1] = current1; /* update hash table */
+
+ assert(ip0 + 1 == ip1);
+
+ if ((offset_1 > 0) & (MEM_read32(repMatch) == MEM_read32(ip2))) {
+ mLength = ip2[-1] == repMatch[-1] ? 1 : 0;
+ ip0 = ip2 - mLength;
+ match0 = repMatch - mLength;
+ offcode = 0;
+ goto _match;
+ }
+ if ((matchIndex0 > prefixStartIndex) && MEM_read32(match0) == val0) {
+ /* found a regular match */
+ goto _offset;
+ }
+ if ((matchIndex1 > prefixStartIndex) && MEM_read32(match1) == val1) {
+ /* found a regular match after one literal */
+ ip0 = ip1;
+ match0 = match1;
+ goto _offset;
+ }
+ {
+ size_t const step = ((ip0-anchor) >> (kSearchStrength - 1)) + stepSize;
+ assert(step >= 2);
+ ip0 += step;
+ ip1 += step;
+ continue;
+ }
+_offset: /* Requires: ip0, match0 */
+ /* Compute the offset code */
+ offset_2 = offset_1;
+ offset_1 = (U32)(ip0-match0);
+ offcode = offset_1 + ZSTD_REP_MOVE;
+ mLength = 0;
+ /* Count the backwards match length */
+ while (((ip0>anchor) & (match0>prefixStart))
+ && (ip0[-1] == match0[-1])) { ip0--; match0--; mLength++; } /* catch up */
+
+_match: /* Requires: ip0, match0, offcode */
+ /* Count the forward length */
+ mLength += ZSTD_count(ip0+mLength+4, match0+mLength+4, iend) + 4;
+ ZSTD_storeSeq(seqStore, ip0-anchor, anchor, offcode, mLength-MINMATCH);
+ /* match found */
+ ip0 += mLength;
+ anchor = ip0;
+ ip1 = ip0 + 1;
+
+ if (ip0 <= ilimit) {
+ /* Fill Table */
+ assert(base+current0+2 > istart); /* check base overflow */
+ hashTable[ZSTD_hashPtr(base+current0+2, hlog, mls)] = current0+2; /* here because current+2 could be > iend-8 */
+ hashTable[ZSTD_hashPtr(ip0-2, hlog, mls)] = (U32)(ip0-2-base);
+
+ while ( (ip0 <= ilimit)
+ && ( (offset_2>0)
+ & (MEM_read32(ip0) == MEM_read32(ip0 - offset_2)) )) {
+ /* store sequence */
+ size_t const rLength = ZSTD_count(ip0+4, ip0+4-offset_2, iend) + 4;
+ U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; /* swap offset_2 <=> offset_1 */
+ hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = (U32)(ip0-base);
+ ip0 += rLength;
+ ip1 = ip0 + 1;
+ ZSTD_storeSeq(seqStore, 0, anchor, 0, rLength-MINMATCH);
+ anchor = ip0;
+ continue; /* faster when present (confirmed on gcc-8) ... (?) */
+ }
+ }
+ }
+
+ /* save reps for next block */
+ rep[0] = offset_1 ? offset_1 : offsetSaved;
+ rep[1] = offset_2 ? offset_2 : offsetSaved;
+
+ /* Return the last literals size */
+ return (size_t)(iend - anchor);
+}
+
+
+size_t ZSTD_compressBlock_fast(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize)
+{
+ ZSTD_compressionParameters const* cParams = &ms->cParams;
+ U32 const mls = cParams->minMatch;
+ assert(ms->dictMatchState == NULL);
+ switch(mls)
+ {
+ default: /* includes case 3 */
+ case 4 :
+ return ZSTD_compressBlock_fast_generic(ms, seqStore, rep, src, srcSize, 4);
+ case 5 :
+ return ZSTD_compressBlock_fast_generic(ms, seqStore, rep, src, srcSize, 5);
+ case 6 :
+ return ZSTD_compressBlock_fast_generic(ms, seqStore, rep, src, srcSize, 6);
+ case 7 :
+ return ZSTD_compressBlock_fast_generic(ms, seqStore, rep, src, srcSize, 7);
+ }
+}
+
+FORCE_INLINE_TEMPLATE
+size_t ZSTD_compressBlock_fast_dictMatchState_generic(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize, U32 const mls)
+{
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
+ U32* const hashTable = ms->hashTable;
+ U32 const hlog = cParams->hashLog;
+ /* support stepSize of 0 */
+ U32 const stepSize = cParams->targetLength + !(cParams->targetLength);
+ const BYTE* const base = ms->window.base;
+ const BYTE* const istart = (const BYTE*)src;
+ const BYTE* ip = istart;
+ const BYTE* anchor = istart;
+ const U32 prefixStartIndex = ms->window.dictLimit;
+ const BYTE* const prefixStart = base + prefixStartIndex;
+ const BYTE* const iend = istart + srcSize;
+ const BYTE* const ilimit = iend - HASH_READ_SIZE;
+ U32 offset_1=rep[0], offset_2=rep[1];
+ U32 offsetSaved = 0;
+
+ const ZSTD_matchState_t* const dms = ms->dictMatchState;
+ const ZSTD_compressionParameters* const dictCParams = &dms->cParams ;
+ const U32* const dictHashTable = dms->hashTable;
+ const U32 dictStartIndex = dms->window.dictLimit;
+ const BYTE* const dictBase = dms->window.base;
+ const BYTE* const dictStart = dictBase + dictStartIndex;
+ const BYTE* const dictEnd = dms->window.nextSrc;
+ const U32 dictIndexDelta = prefixStartIndex - (U32)(dictEnd - dictBase);
+ const U32 dictAndPrefixLength = (U32)(ip - prefixStart + dictEnd - dictStart);
+ const U32 dictHLog = dictCParams->hashLog;
+
+ /* if a dictionary is still attached, it necessarily means that
+ * it is within window size. So we just check it. */
+ const U32 maxDistance = 1U << cParams->windowLog;
+ const U32 endIndex = (U32)((size_t)(ip - base) + srcSize);
+ assert(endIndex - prefixStartIndex <= maxDistance);
+ (void)maxDistance; (void)endIndex; /* these variables are not used when assert() is disabled */
+
+ /* ensure there will be no no underflow
+ * when translating a dict index into a local index */
+ assert(prefixStartIndex >= (U32)(dictEnd - dictBase));
+
+ /* init */
+ ip += (dictAndPrefixLength == 0);
+ /* dictMatchState repCode checks don't currently handle repCode == 0
+ * disabling. */
+ assert(offset_1 <= dictAndPrefixLength);
+ assert(offset_2 <= dictAndPrefixLength);
+
+ /* Main Search Loop */
while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */
size_t mLength;
size_t const h = ZSTD_hashPtr(ip, hlog, mls);
U32 const current = (U32)(ip-base);
U32 const matchIndex = hashTable[h];
const BYTE* match = base + matchIndex;
+ const U32 repIndex = current + 1 - offset_1;
+ const BYTE* repMatch = (repIndex < prefixStartIndex) ?
+ dictBase + (repIndex - dictIndexDelta) :
+ base + repIndex;
hashTable[h] = current; /* update hash table */
- if ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1))) {
- mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
+ if ( ((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex isn't overlapping dict + prefix */
+ && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
+ const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
+ mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
ip++;
- ZSTD_storeSeq(seqStore, ip-anchor, anchor, 0, mLength-MINMATCH);
- } else {
- if ( (matchIndex <= lowestIndex)
- || (MEM_read32(match) != MEM_read32(ip)) ) {
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, 0, mLength-MINMATCH);
+ } else if ( (matchIndex <= prefixStartIndex) ) {
+ size_t const dictHash = ZSTD_hashPtr(ip, dictHLog, mls);
+ U32 const dictMatchIndex = dictHashTable[dictHash];
+ const BYTE* dictMatch = dictBase + dictMatchIndex;
+ if (dictMatchIndex <= dictStartIndex ||
+ MEM_read32(dictMatch) != MEM_read32(ip)) {
assert(stepSize >= 1);
ip += ((ip-anchor) >> kSearchStrength) + stepSize;
continue;
- }
- mLength = ZSTD_count(ip+4, match+4, iend) + 4;
- { U32 const offset = (U32)(ip-match);
- while (((ip>anchor) & (match>lowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
+ } else {
+ /* found a dict match */
+ U32 const offset = (U32)(current-dictMatchIndex-dictIndexDelta);
+ mLength = ZSTD_count_2segments(ip+4, dictMatch+4, iend, dictEnd, prefixStart) + 4;
+ while (((ip>anchor) & (dictMatch>dictStart))
+ && (ip[-1] == dictMatch[-1])) {
+ ip--; dictMatch--; mLength++;
+ } /* catch up */
offset_2 = offset_1;
offset_1 = offset;
- ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
- } }
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ }
+ } else if (MEM_read32(match) != MEM_read32(ip)) {
+ /* it's not a match, and we're not going to check the dictionary */
+ assert(stepSize >= 1);
+ ip += ((ip-anchor) >> kSearchStrength) + stepSize;
+ continue;
+ } else {
+ /* found a regular match */
+ U32 const offset = (U32)(ip-match);
+ mLength = ZSTD_count(ip+4, match+4, iend) + 4;
+ while (((ip>anchor) & (match>prefixStart))
+ && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
+ offset_2 = offset_1;
+ offset_1 = offset;
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ }
/* match found */
ip += mLength;
@@ -97,108 +307,131 @@
if (ip <= ilimit) {
/* Fill Table */
+ assert(base+current+2 > istart); /* check base overflow */
hashTable[ZSTD_hashPtr(base+current+2, hlog, mls)] = current+2; /* here because current+2 could be > iend-8 */
hashTable[ZSTD_hashPtr(ip-2, hlog, mls)] = (U32)(ip-2-base);
+
/* check immediate repcode */
- while ( (ip <= ilimit)
- && ( (offset_2>0)
- & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
- /* store sequence */
- size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
- { U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; } /* swap offset_2 <=> offset_1 */
- hashTable[ZSTD_hashPtr(ip, hlog, mls)] = (U32)(ip-base);
- ZSTD_storeSeq(seqStore, 0, anchor, 0, rLength-MINMATCH);
- ip += rLength;
- anchor = ip;
- continue; /* faster when present ... (?) */
- } } }
+ while (ip <= ilimit) {
+ U32 const current2 = (U32)(ip-base);
+ U32 const repIndex2 = current2 - offset_2;
+ const BYTE* repMatch2 = repIndex2 < prefixStartIndex ?
+ dictBase - dictIndexDelta + repIndex2 :
+ base + repIndex2;
+ if ( ((U32)((prefixStartIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */)
+ && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
+ const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
+ size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
+ U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
+ ZSTD_storeSeq(seqStore, 0, anchor, 0, repLength2-MINMATCH);
+ hashTable[ZSTD_hashPtr(ip, hlog, mls)] = current2;
+ ip += repLength2;
+ anchor = ip;
+ continue;
+ }
+ break;
+ }
+ }
+ }
/* save reps for next block */
rep[0] = offset_1 ? offset_1 : offsetSaved;
rep[1] = offset_2 ? offset_2 : offsetSaved;
/* Return the last literals size */
- return iend - anchor;
+ return (size_t)(iend - anchor);
}
-
-size_t ZSTD_compressBlock_fast(
+size_t ZSTD_compressBlock_fast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- U32 const hlog = cParams->hashLog;
- U32 const mls = cParams->searchLength;
- U32 const stepSize = cParams->targetLength;
+ ZSTD_compressionParameters const* cParams = &ms->cParams;
+ U32 const mls = cParams->minMatch;
+ assert(ms->dictMatchState != NULL);
switch(mls)
{
default: /* includes case 3 */
case 4 :
- return ZSTD_compressBlock_fast_generic(ms, seqStore, rep, src, srcSize, hlog, stepSize, 4);
+ return ZSTD_compressBlock_fast_dictMatchState_generic(ms, seqStore, rep, src, srcSize, 4);
case 5 :
- return ZSTD_compressBlock_fast_generic(ms, seqStore, rep, src, srcSize, hlog, stepSize, 5);
+ return ZSTD_compressBlock_fast_dictMatchState_generic(ms, seqStore, rep, src, srcSize, 5);
case 6 :
- return ZSTD_compressBlock_fast_generic(ms, seqStore, rep, src, srcSize, hlog, stepSize, 6);
+ return ZSTD_compressBlock_fast_dictMatchState_generic(ms, seqStore, rep, src, srcSize, 6);
case 7 :
- return ZSTD_compressBlock_fast_generic(ms, seqStore, rep, src, srcSize, hlog, stepSize, 7);
+ return ZSTD_compressBlock_fast_dictMatchState_generic(ms, seqStore, rep, src, srcSize, 7);
}
}
static size_t ZSTD_compressBlock_fast_extDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- void const* src, size_t srcSize,
- U32 const hlog, U32 const stepSize, U32 const mls)
+ void const* src, size_t srcSize, U32 const mls)
{
- U32* hashTable = ms->hashTable;
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
+ U32* const hashTable = ms->hashTable;
+ U32 const hlog = cParams->hashLog;
+ /* support stepSize of 0 */
+ U32 const stepSize = cParams->targetLength + !(cParams->targetLength);
const BYTE* const base = ms->window.base;
const BYTE* const dictBase = ms->window.dictBase;
const BYTE* const istart = (const BYTE*)src;
const BYTE* ip = istart;
const BYTE* anchor = istart;
- const U32 lowestIndex = ms->window.lowLimit;
- const BYTE* const dictStart = dictBase + lowestIndex;
+ const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
+ const U32 maxDistance = 1U << cParams->windowLog;
+ const U32 validLow = ms->window.lowLimit;
+ const U32 lowLimit = (endIndex - validLow > maxDistance) ? endIndex - maxDistance : validLow;
+ const U32 dictStartIndex = lowLimit;
+ const BYTE* const dictStart = dictBase + dictStartIndex;
const U32 dictLimit = ms->window.dictLimit;
- const BYTE* const lowPrefixPtr = base + dictLimit;
- const BYTE* const dictEnd = dictBase + dictLimit;
+ const U32 prefixStartIndex = dictLimit < lowLimit ? lowLimit : dictLimit;
+ const BYTE* const prefixStart = base + prefixStartIndex;
+ const BYTE* const dictEnd = dictBase + prefixStartIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - 8;
U32 offset_1=rep[0], offset_2=rep[1];
+ /* switch to "regular" variant if extDict is invalidated due to maxDistance */
+ if (prefixStartIndex == dictStartIndex)
+ return ZSTD_compressBlock_fast_generic(ms, seqStore, rep, src, srcSize, mls);
+
/* Search Loop */
while (ip < ilimit) { /* < instead of <=, because (ip+1) */
const size_t h = ZSTD_hashPtr(ip, hlog, mls);
- const U32 matchIndex = hashTable[h];
- const BYTE* matchBase = matchIndex < dictLimit ? dictBase : base;
- const BYTE* match = matchBase + matchIndex;
- const U32 current = (U32)(ip-base);
- const U32 repIndex = current + 1 - offset_1; /* offset_1 expected <= current +1 */
- const BYTE* repBase = repIndex < dictLimit ? dictBase : base;
- const BYTE* repMatch = repBase + repIndex;
+ const U32 matchIndex = hashTable[h];
+ const BYTE* const matchBase = matchIndex < prefixStartIndex ? dictBase : base;
+ const BYTE* match = matchBase + matchIndex;
+ const U32 current = (U32)(ip-base);
+ const U32 repIndex = current + 1 - offset_1;
+ const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base;
+ const BYTE* const repMatch = repBase + repIndex;
size_t mLength;
hashTable[h] = current; /* update hash table */
+ assert(offset_1 <= current +1); /* check repIndex */
- if ( (((U32)((dictLimit-1) - repIndex) >= 3) /* intentional underflow */ & (repIndex > lowestIndex))
+ if ( (((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow */ & (repIndex > dictStartIndex))
&& (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
- const BYTE* repMatchEnd = repIndex < dictLimit ? dictEnd : iend;
- mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, lowPrefixPtr) + 4;
+ const BYTE* repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
+ mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
ip++;
- ZSTD_storeSeq(seqStore, ip-anchor, anchor, 0, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, 0, mLength-MINMATCH);
} else {
- if ( (matchIndex < lowestIndex) ||
+ if ( (matchIndex < dictStartIndex) ||
(MEM_read32(match) != MEM_read32(ip)) ) {
assert(stepSize >= 1);
ip += ((ip-anchor) >> kSearchStrength) + stepSize;
continue;
}
- { const BYTE* matchEnd = matchIndex < dictLimit ? dictEnd : iend;
- const BYTE* lowMatchPtr = matchIndex < dictLimit ? dictStart : lowPrefixPtr;
+ { const BYTE* matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend;
+ const BYTE* lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart;
U32 offset;
- mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, lowPrefixPtr) + 4;
+ mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4;
while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
offset = current - matchIndex;
offset_2 = offset_1;
offset_1 = offset;
- ZSTD_storeSeq(seqStore, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
} }
/* found a match : store it */
@@ -213,11 +446,11 @@
while (ip <= ilimit) {
U32 const current2 = (U32)(ip-base);
U32 const repIndex2 = current2 - offset_2;
- const BYTE* repMatch2 = repIndex2 < dictLimit ? dictBase + repIndex2 : base + repIndex2;
- if ( (((U32)((dictLimit-1) - repIndex2) >= 3) & (repIndex2 > lowestIndex)) /* intentional overflow */
+ const BYTE* repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
+ if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) & (repIndex2 > dictStartIndex)) /* intentional overflow */
&& (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
- const BYTE* const repEnd2 = repIndex2 < dictLimit ? dictEnd : iend;
- size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, lowPrefixPtr) + 4;
+ const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
+ size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
ZSTD_storeSeq(seqStore, 0, anchor, 0, repLength2-MINMATCH);
hashTable[ZSTD_hashPtr(ip, hlog, mls)] = current2;
@@ -233,27 +466,26 @@
rep[1] = offset_2;
/* Return the last literals size */
- return iend - anchor;
+ return (size_t)(iend - anchor);
}
size_t ZSTD_compressBlock_fast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- U32 const hlog = cParams->hashLog;
- U32 const mls = cParams->searchLength;
- U32 const stepSize = cParams->targetLength;
+ ZSTD_compressionParameters const* cParams = &ms->cParams;
+ U32 const mls = cParams->minMatch;
switch(mls)
{
default: /* includes case 3 */
case 4 :
- return ZSTD_compressBlock_fast_extDict_generic(ms, seqStore, rep, src, srcSize, hlog, stepSize, 4);
+ return ZSTD_compressBlock_fast_extDict_generic(ms, seqStore, rep, src, srcSize, 4);
case 5 :
- return ZSTD_compressBlock_fast_extDict_generic(ms, seqStore, rep, src, srcSize, hlog, stepSize, 5);
+ return ZSTD_compressBlock_fast_extDict_generic(ms, seqStore, rep, src, srcSize, 5);
case 6 :
- return ZSTD_compressBlock_fast_extDict_generic(ms, seqStore, rep, src, srcSize, hlog, stepSize, 6);
+ return ZSTD_compressBlock_fast_extDict_generic(ms, seqStore, rep, src, srcSize, 6);
case 7 :
- return ZSTD_compressBlock_fast_extDict_generic(ms, seqStore, rep, src, srcSize, hlog, stepSize, 7);
+ return ZSTD_compressBlock_fast_extDict_generic(ms, seqStore, rep, src, srcSize, 7);
}
}
diff --git a/vendor/github.com/DataDog/zstd/zstd_fast.h b/vendor/github.com/DataDog/zstd/zstd_fast.h
index f0438ad..b74a88c 100644
--- a/vendor/github.com/DataDog/zstd/zstd_fast.h
+++ b/vendor/github.com/DataDog/zstd/zstd_fast.h
@@ -19,14 +19,16 @@
#include "zstd_compress_internal.h"
void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
- ZSTD_compressionParameters const* cParams,
- void const* end);
+ void const* end, ZSTD_dictTableLoadMethod_e dtlm);
size_t ZSTD_compressBlock_fast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
+size_t ZSTD_compressBlock_fast_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_fast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
#if defined (__cplusplus)
}
diff --git a/vendor/github.com/DataDog/zstd/zstd_internal.h b/vendor/github.com/DataDog/zstd/zstd_internal.h
index 65c08a8..81b16ea 100644
--- a/vendor/github.com/DataDog/zstd/zstd_internal.h
+++ b/vendor/github.com/DataDog/zstd/zstd_internal.h
@@ -21,6 +21,7 @@
***************************************/
#include "compiler.h"
#include "mem.h"
+#include "debug.h" /* assert, DEBUGLOG, RAWLOG, g_debuglevel */
#include "error_private.h"
#define ZSTD_STATIC_LINKING_ONLY
#include "zstd.h"
@@ -33,48 +34,15 @@
#endif
#include "xxhash.h" /* XXH_reset, update, digest */
-
#if defined (__cplusplus)
extern "C" {
#endif
-
-/*-*************************************
-* Debug
-***************************************/
-#if defined(ZSTD_DEBUG) && (ZSTD_DEBUG>=1)
-# include <assert.h>
-#else
-# ifndef assert
-# define assert(condition) ((void)0)
-# endif
-#endif
-
-#define ZSTD_STATIC_ASSERT(c) { enum { ZSTD_static_assert = 1/(int)(!!(c)) }; }
-
-#if defined(ZSTD_DEBUG) && (ZSTD_DEBUG>=2)
-# include <stdio.h>
-extern int g_debuglog_enable;
-/* recommended values for ZSTD_DEBUG display levels :
- * 1 : no display, enables assert() only
- * 2 : reserved for currently active debug path
- * 3 : events once per object lifetime (CCtx, CDict, etc.)
- * 4 : events once per frame
- * 5 : events once per block
- * 6 : events once per sequence (*very* verbose) */
-# define RAWLOG(l, ...) { \
- if ((g_debuglog_enable) & (l<=ZSTD_DEBUG)) { \
- fprintf(stderr, __VA_ARGS__); \
- } }
-# define DEBUGLOG(l, ...) { \
- if ((g_debuglog_enable) & (l<=ZSTD_DEBUG)) { \
- fprintf(stderr, __FILE__ ": " __VA_ARGS__); \
- fprintf(stderr, " \n"); \
- } }
-#else
-# define RAWLOG(l, ...) {} /* disabled */
-# define DEBUGLOG(l, ...) {} /* disabled */
-#endif
+/* ---- static assert (debug) --- */
+#define ZSTD_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c)
+#define ZSTD_isError ERR_isError /* for inlining */
+#define FSE_isError ERR_isError
+#define HUF_isError ERR_isError
/*-*************************************
@@ -84,8 +52,50 @@
#undef MAX
#define MIN(a,b) ((a)<(b) ? (a) : (b))
#define MAX(a,b) ((a)>(b) ? (a) : (b))
-#define CHECK_F(f) { size_t const errcod = f; if (ERR_isError(errcod)) return errcod; } /* check and Forward error code */
-#define CHECK_E(f, e) { size_t const errcod = f; if (ERR_isError(errcod)) return ERROR(e); } /* check and send Error code */
+
+/**
+ * Return the specified error if the condition evaluates to true.
+ *
+ * In debug modes, prints additional information. In order to do that
+ * (particularly, printing the conditional that failed), this can't just wrap
+ * RETURN_ERROR().
+ */
+#define RETURN_ERROR_IF(cond, err, ...) \
+ if (cond) { \
+ RAWLOG(3, "%s:%d: ERROR!: check %s failed, returning %s", __FILE__, __LINE__, ZSTD_QUOTE(cond), ZSTD_QUOTE(ERROR(err))); \
+ RAWLOG(3, ": " __VA_ARGS__); \
+ RAWLOG(3, "\n"); \
+ return ERROR(err); \
+ }
+
+/**
+ * Unconditionally return the specified error.
+ *
+ * In debug modes, prints additional information.
+ */
+#define RETURN_ERROR(err, ...) \
+ do { \
+ RAWLOG(3, "%s:%d: ERROR!: unconditional check failed, returning %s", __FILE__, __LINE__, ZSTD_QUOTE(ERROR(err))); \
+ RAWLOG(3, ": " __VA_ARGS__); \
+ RAWLOG(3, "\n"); \
+ return ERROR(err); \
+ } while(0);
+
+/**
+ * If the provided expression evaluates to an error code, returns that error code.
+ *
+ * In debug modes, prints additional information.
+ */
+#define FORWARD_IF_ERROR(err, ...) \
+ do { \
+ size_t const err_code = (err); \
+ if (ERR_isError(err_code)) { \
+ RAWLOG(3, "%s:%d: ERROR!: forwarding error in %s: %s", __FILE__, __LINE__, ZSTD_QUOTE(err), ERR_getErrorName(err_code)); \
+ RAWLOG(3, ": " __VA_ARGS__); \
+ RAWLOG(3, "\n"); \
+ return err_code; \
+ } \
+ } while(0);
/*-*************************************
@@ -109,12 +119,10 @@
#define BIT0 1
#define ZSTD_WINDOWLOG_ABSOLUTEMIN 10
-#define ZSTD_WINDOWLOG_DEFAULTMAX 27 /* Default maximum allowed window log */
static const size_t ZSTD_fcs_fieldSize[4] = { 0, 2, 4, 8 };
static const size_t ZSTD_did_fieldSize[4] = { 0, 1, 2, 4 };
-#define ZSTD_FRAMEIDSIZE 4
-static const size_t ZSTD_frameIdSize = ZSTD_FRAMEIDSIZE; /* magic number size */
+#define ZSTD_FRAMEIDSIZE 4 /* magic number size */
#define ZSTD_BLOCKHEADERSIZE 3 /* C standard doesn't allow `static const` variable to be init using another `static const` variable */
static const size_t ZSTD_blockHeaderSize = ZSTD_BLOCKHEADERSIZE;
@@ -184,19 +192,72 @@
* Shared functions to include for inlining
*********************************************/
static void ZSTD_copy8(void* dst, const void* src) { memcpy(dst, src, 8); }
+
#define COPY8(d,s) { ZSTD_copy8(d,s); d+=8; s+=8; }
+static void ZSTD_copy16(void* dst, const void* src) { memcpy(dst, src, 16); }
+#define COPY16(d,s) { ZSTD_copy16(d,s); d+=16; s+=16; }
+
+#define WILDCOPY_OVERLENGTH 8
+#define VECLEN 16
+
+typedef enum {
+ ZSTD_no_overlap,
+ ZSTD_overlap_src_before_dst,
+ /* ZSTD_overlap_dst_before_src, */
+} ZSTD_overlap_e;
/*! ZSTD_wildcopy() :
* custom version of memcpy(), can overwrite up to WILDCOPY_OVERLENGTH bytes (if length==0) */
-#define WILDCOPY_OVERLENGTH 8
-MEM_STATIC void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length)
+MEM_STATIC FORCE_INLINE_ATTR DONT_VECTORIZE
+void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e ovtype)
{
+ ptrdiff_t diff = (BYTE*)dst - (const BYTE*)src;
const BYTE* ip = (const BYTE*)src;
BYTE* op = (BYTE*)dst;
BYTE* const oend = op + length;
- do
- COPY8(op, ip)
- while (op < oend);
+
+ assert(diff >= 8 || (ovtype == ZSTD_no_overlap && diff < -8));
+ if (length < VECLEN || (ovtype == ZSTD_overlap_src_before_dst && diff < VECLEN)) {
+ do
+ COPY8(op, ip)
+ while (op < oend);
+ }
+ else {
+ if ((length & 8) == 0)
+ COPY8(op, ip);
+ do {
+ COPY16(op, ip);
+ }
+ while (op < oend);
+ }
+}
+
+/*! ZSTD_wildcopy_16min() :
+ * same semantics as ZSTD_wilcopy() except guaranteed to be able to copy 16 bytes at the start */
+MEM_STATIC FORCE_INLINE_ATTR DONT_VECTORIZE
+void ZSTD_wildcopy_16min(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e ovtype)
+{
+ ptrdiff_t diff = (BYTE*)dst - (const BYTE*)src;
+ const BYTE* ip = (const BYTE*)src;
+ BYTE* op = (BYTE*)dst;
+ BYTE* const oend = op + length;
+
+ assert(length >= 8);
+ assert(diff >= 8 || (ovtype == ZSTD_no_overlap && diff < -8));
+
+ if (ovtype == ZSTD_overlap_src_before_dst && diff < VECLEN) {
+ do
+ COPY8(op, ip)
+ while (op < oend);
+ }
+ else {
+ if ((length & 8) == 0)
+ COPY8(op, ip);
+ do {
+ COPY16(op, ip);
+ }
+ while (op < oend);
+ }
}
MEM_STATIC void ZSTD_wildcopy_e(void* dst, const void* src, void* dstEnd) /* should be faster for decoding, but strangely, not verified on all platform */
@@ -227,10 +288,23 @@
BYTE* llCode;
BYTE* mlCode;
BYTE* ofCode;
+ size_t maxNbSeq;
+ size_t maxNbLit;
U32 longLengthID; /* 0 == no longLength; 1 == Lit.longLength; 2 == Match.longLength; */
U32 longLengthPos;
} seqStore_t;
+/**
+ * Contains the compressed frame size and an upper-bound for the decompressed frame size.
+ * Note: before using `compressedSize`, check for errors using ZSTD_isError().
+ * similarly, before using `decompressedBound`, check for errors using:
+ * `decompressedBound != ZSTD_CONTENTSIZE_ERROR`
+ */
+typedef struct {
+ 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) */
@@ -275,7 +349,7 @@
blockType_e blockType;
U32 lastBlock;
U32 origSize;
-} blockProperties_t;
+} blockProperties_t; /* declared here for decompress and fullbench */
/*! ZSTD_getcBlockSize() :
* Provides the size of compressed block from block header `src` */
@@ -283,6 +357,13 @@
size_t ZSTD_getcBlockSize(const void* src, size_t srcSize,
blockProperties_t* bpPtr);
+/*! ZSTD_decodeSeqHeaders() :
+ * decode sequence header from src */
+/* Used by: decompress, fullbench (does not get its definition from here) */
+size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr,
+ const void* src, size_t srcSize);
+
+
#if defined (__cplusplus)
}
#endif
diff --git a/vendor/github.com/DataDog/zstd/zstd_lazy.c b/vendor/github.com/DataDog/zstd/zstd_lazy.c
index 9f15812..94d906c 100644
--- a/vendor/github.com/DataDog/zstd/zstd_lazy.c
+++ b/vendor/github.com/DataDog/zstd/zstd_lazy.c
@@ -16,11 +16,12 @@
* Binary Tree search
***************************************/
-void ZSTD_updateDUBT(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+static void
+ZSTD_updateDUBT(ZSTD_matchState_t* ms,
const BYTE* ip, const BYTE* iend,
U32 mls)
{
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hashLog = cParams->hashLog;
@@ -59,14 +60,16 @@
* sort one already inserted but unsorted position
* assumption : current >= btlow == (current - btmask)
* doesn't fail */
-static void ZSTD_insertDUBT1(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+static void
+ZSTD_insertDUBT1(ZSTD_matchState_t* ms,
U32 current, const BYTE* inputEnd,
- U32 nbCompares, U32 btLow, int extDict)
+ U32 nbCompares, U32 btLow,
+ const ZSTD_dictMode_e dictMode)
{
- U32* const bt = ms->chainTable;
- U32 const btLog = cParams->chainLog - 1;
- U32 const btMask = (1 << btLog) - 1;
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
+ U32* const bt = ms->chainTable;
+ U32 const btLog = cParams->chainLog - 1;
+ U32 const btMask = (1 << btLog) - 1;
size_t commonLengthSmaller=0, commonLengthLarger=0;
const BYTE* const base = ms->window.base;
const BYTE* const dictBase = ms->window.dictBase;
@@ -78,9 +81,12 @@
const BYTE* match;
U32* smallerPtr = bt + 2*(current&btMask);
U32* largerPtr = smallerPtr + 1;
- U32 matchIndex = *smallerPtr;
+ U32 matchIndex = *smallerPtr; /* this candidate is unsorted : next sorted candidate is reached through *smallerPtr, while *largerPtr contains previous unsorted candidate (which is already saved and can be overwritten) */
U32 dummy32; /* to be nullified at the end */
- U32 const windowLow = ms->window.lowLimit;
+ U32 const windowValid = ms->window.lowLimit;
+ U32 const maxDistance = 1U << cParams->windowLog;
+ U32 const windowLow = (current - windowValid > maxDistance) ? current - maxDistance : windowValid;
+
DEBUGLOG(8, "ZSTD_insertDUBT1(%u) (dictLimit=%u, lowLimit=%u)",
current, dictLimit, windowLow);
@@ -91,11 +97,16 @@
U32* const nextPtr = bt + 2*(matchIndex & btMask);
size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
assert(matchIndex < current);
+ /* note : all candidates are now supposed sorted,
+ * but it's still possible to have nextPtr[1] == ZSTD_DUBT_UNSORTED_MARK
+ * when a real index has the same value as ZSTD_DUBT_UNSORTED_MARK */
- if ( (!extDict)
+ if ( (dictMode != ZSTD_extDict)
|| (matchIndex+matchLength >= dictLimit) /* both in current segment*/
|| (current < dictLimit) /* both in extDict */) {
- const BYTE* const mBase = !extDict || ((matchIndex+matchLength) >= dictLimit) ? base : dictBase;
+ const BYTE* const mBase = ( (dictMode != ZSTD_extDict)
+ || (matchIndex+matchLength >= dictLimit)) ?
+ base : dictBase;
assert( (matchIndex+matchLength >= dictLimit) /* might be wrong if extDict is incorrectly set to 0 */
|| (current < dictLimit) );
match = mBase + matchIndex;
@@ -104,7 +115,7 @@
match = dictBase + matchIndex;
matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iend, dictEnd, prefixStart);
if (matchIndex+matchLength >= dictLimit)
- match = base + matchIndex; /* to prepare for next usage of match[matchLength] */
+ match = base + matchIndex; /* preparation for next read of match[matchLength] */
}
DEBUGLOG(8, "ZSTD_insertDUBT1: comparing %u with %u : found %u common bytes ",
@@ -138,13 +149,92 @@
}
-static size_t ZSTD_DUBT_findBestMatch (
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
- const BYTE* const ip, const BYTE* const iend,
- size_t* offsetPtr,
- U32 const mls,
- U32 const extDict)
+static size_t
+ZSTD_DUBT_findBetterDictMatch (
+ ZSTD_matchState_t* ms,
+ const BYTE* const ip, const BYTE* const iend,
+ size_t* offsetPtr,
+ size_t bestLength,
+ U32 nbCompares,
+ U32 const mls,
+ const ZSTD_dictMode_e dictMode)
{
+ const ZSTD_matchState_t * const dms = ms->dictMatchState;
+ const ZSTD_compressionParameters* const dmsCParams = &dms->cParams;
+ const U32 * const dictHashTable = dms->hashTable;
+ U32 const hashLog = dmsCParams->hashLog;
+ size_t const h = ZSTD_hashPtr(ip, hashLog, mls);
+ U32 dictMatchIndex = dictHashTable[h];
+
+ const BYTE* const base = ms->window.base;
+ const BYTE* const prefixStart = base + ms->window.dictLimit;
+ U32 const current = (U32)(ip-base);
+ const BYTE* const dictBase = dms->window.base;
+ const BYTE* const dictEnd = dms->window.nextSrc;
+ U32 const dictHighLimit = (U32)(dms->window.nextSrc - dms->window.base);
+ U32 const dictLowLimit = dms->window.lowLimit;
+ U32 const dictIndexDelta = ms->window.lowLimit - dictHighLimit;
+
+ U32* const dictBt = dms->chainTable;
+ U32 const btLog = dmsCParams->chainLog - 1;
+ U32 const btMask = (1 << btLog) - 1;
+ U32 const btLow = (btMask >= dictHighLimit - dictLowLimit) ? dictLowLimit : dictHighLimit - btMask;
+
+ size_t commonLengthSmaller=0, commonLengthLarger=0;
+
+ (void)dictMode;
+ assert(dictMode == ZSTD_dictMatchState);
+
+ while (nbCompares-- && (dictMatchIndex > dictLowLimit)) {
+ U32* const nextPtr = dictBt + 2*(dictMatchIndex & btMask);
+ size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
+ const BYTE* match = dictBase + dictMatchIndex;
+ matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iend, dictEnd, prefixStart);
+ if (dictMatchIndex+matchLength >= dictHighLimit)
+ match = base + dictMatchIndex + dictIndexDelta; /* to prepare for next usage of match[matchLength] */
+
+ if (matchLength > bestLength) {
+ U32 matchIndex = dictMatchIndex + dictIndexDelta;
+ if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(current-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) ) {
+ DEBUGLOG(9, "ZSTD_DUBT_findBetterDictMatch(%u) : found better match length %u -> %u and offsetCode %u -> %u (dictMatchIndex %u, matchIndex %u)",
+ current, (U32)bestLength, (U32)matchLength, (U32)*offsetPtr, ZSTD_REP_MOVE + current - matchIndex, dictMatchIndex, matchIndex);
+ bestLength = matchLength, *offsetPtr = ZSTD_REP_MOVE + current - matchIndex;
+ }
+ if (ip+matchLength == iend) { /* reached end of input : ip[matchLength] is not valid, no way to know if it's larger or smaller than match */
+ break; /* drop, to guarantee consistency (miss a little bit of compression) */
+ }
+ }
+
+ if (match[matchLength] < ip[matchLength]) {
+ if (dictMatchIndex <= btLow) { break; } /* beyond tree size, stop the search */
+ commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */
+ dictMatchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */
+ } else {
+ /* match is larger than current */
+ if (dictMatchIndex <= btLow) { break; } /* beyond tree size, stop the search */
+ commonLengthLarger = matchLength;
+ dictMatchIndex = nextPtr[0];
+ }
+ }
+
+ if (bestLength >= MINMATCH) {
+ U32 const mIndex = current - ((U32)*offsetPtr - ZSTD_REP_MOVE); (void)mIndex;
+ DEBUGLOG(8, "ZSTD_DUBT_findBetterDictMatch(%u) : found match of length %u and offsetCode %u (pos %u)",
+ current, (U32)bestLength, (U32)*offsetPtr, mIndex);
+ }
+ return bestLength;
+
+}
+
+
+static size_t
+ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms,
+ const BYTE* const ip, const BYTE* const iend,
+ size_t* offsetPtr,
+ U32 const mls,
+ const ZSTD_dictMode_e dictMode)
+{
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hashLog = cParams->hashLog;
size_t const h = ZSTD_hashPtr(ip, hashLog, mls);
@@ -152,7 +242,9 @@
const BYTE* const base = ms->window.base;
U32 const current = (U32)(ip-base);
- U32 const windowLow = ms->window.lowLimit;
+ U32 const maxDistance = 1U << cParams->windowLog;
+ U32 const windowValid = ms->window.lowLimit;
+ U32 const windowLow = (current - windowValid > maxDistance) ? current - maxDistance : windowValid;
U32* const bt = ms->chainTable;
U32 const btLog = cParams->chainLog - 1;
@@ -175,7 +267,7 @@
&& (nbCandidates > 1) ) {
DEBUGLOG(8, "ZSTD_DUBT_findBestMatch: candidate %u is unsorted",
matchIndex);
- *unsortedMark = previousCandidate;
+ *unsortedMark = previousCandidate; /* the unsortedMark becomes a reversed chain, to move up back to original position */
previousCandidate = matchIndex;
matchIndex = *nextCandidate;
nextCandidate = bt + 2*(matchIndex&btMask);
@@ -183,11 +275,13 @@
nbCandidates --;
}
+ /* nullify last candidate if it's still unsorted
+ * simplification, detrimental to compression ratio, beneficial for speed */
if ( (matchIndex > unsortLimit)
&& (*unsortedMark==ZSTD_DUBT_UNSORTED_MARK) ) {
DEBUGLOG(7, "ZSTD_DUBT_findBestMatch: nullify last unsorted candidate %u",
matchIndex);
- *nextCandidate = *unsortedMark = 0; /* nullify next candidate if it's still unsorted (note : simplification, detrimental to compression ratio, beneficial for speed) */
+ *nextCandidate = *unsortedMark = 0;
}
/* batch sort stacked candidates */
@@ -195,21 +289,21 @@
while (matchIndex) { /* will end on matchIndex == 0 */
U32* const nextCandidateIdxPtr = bt + 2*(matchIndex&btMask) + 1;
U32 const nextCandidateIdx = *nextCandidateIdxPtr;
- ZSTD_insertDUBT1(ms, cParams, matchIndex, iend,
- nbCandidates, unsortLimit, extDict);
+ ZSTD_insertDUBT1(ms, matchIndex, iend,
+ nbCandidates, unsortLimit, dictMode);
matchIndex = nextCandidateIdx;
nbCandidates++;
}
/* find longest match */
- { size_t commonLengthSmaller=0, commonLengthLarger=0;
+ { size_t commonLengthSmaller = 0, commonLengthLarger = 0;
const BYTE* const dictBase = ms->window.dictBase;
const U32 dictLimit = ms->window.dictLimit;
const BYTE* const dictEnd = dictBase + dictLimit;
const BYTE* const prefixStart = base + dictLimit;
U32* smallerPtr = bt + 2*(current&btMask);
U32* largerPtr = bt + 2*(current&btMask) + 1;
- U32 matchEndIdx = current+8+1;
+ U32 matchEndIdx = current + 8 + 1;
U32 dummy32; /* to be nullified at the end */
size_t bestLength = 0;
@@ -221,7 +315,7 @@
size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
const BYTE* match;
- if ((!extDict) || (matchIndex+matchLength >= dictLimit)) {
+ if ((dictMode != ZSTD_extDict) || (matchIndex+matchLength >= dictLimit)) {
match = base + matchIndex;
matchLength += ZSTD_count(ip+matchLength, match+matchLength, iend);
} else {
@@ -237,6 +331,11 @@
if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(current-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) )
bestLength = matchLength, *offsetPtr = ZSTD_REP_MOVE + current - matchIndex;
if (ip+matchLength == iend) { /* equal : no way to know if inf or sup */
+ if (dictMode == ZSTD_dictMatchState) {
+ nbCompares = 0; /* in addition to avoiding checking any
+ * further in this loop, make sure we
+ * skip checking in the dictionary. */
+ }
break; /* drop, to guarantee consistency (miss a little bit of compression) */
}
}
@@ -259,6 +358,13 @@
*smallerPtr = *largerPtr = 0;
+ if (dictMode == ZSTD_dictMatchState && nbCompares) {
+ bestLength = ZSTD_DUBT_findBetterDictMatch(
+ ms, ip, iend,
+ offsetPtr, bestLength, nbCompares,
+ mls, dictMode);
+ }
+
assert(matchEndIdx > current+8); /* ensure nextToUpdate is increased */
ms->nextToUpdate = matchEndIdx - 8; /* skip repetitive patterns */
if (bestLength >= MINMATCH) {
@@ -272,61 +378,64 @@
/** ZSTD_BtFindBestMatch() : Tree updater, providing best match */
-static size_t ZSTD_BtFindBestMatch (
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
- const BYTE* const ip, const BYTE* const iLimit,
- size_t* offsetPtr,
- const U32 mls /* template */)
+FORCE_INLINE_TEMPLATE size_t
+ZSTD_BtFindBestMatch( ZSTD_matchState_t* ms,
+ const BYTE* const ip, const BYTE* const iLimit,
+ size_t* offsetPtr,
+ const U32 mls /* template */,
+ const ZSTD_dictMode_e dictMode)
{
DEBUGLOG(7, "ZSTD_BtFindBestMatch");
if (ip < ms->window.base + ms->nextToUpdate) return 0; /* skipped area */
- ZSTD_updateDUBT(ms, cParams, ip, iLimit, mls);
- return ZSTD_DUBT_findBestMatch(ms, cParams, ip, iLimit, offsetPtr, mls, 0);
+ ZSTD_updateDUBT(ms, ip, iLimit, mls);
+ return ZSTD_DUBT_findBestMatch(ms, ip, iLimit, offsetPtr, mls, dictMode);
}
-static size_t ZSTD_BtFindBestMatch_selectMLS (
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
- const BYTE* ip, const BYTE* const iLimit,
- size_t* offsetPtr)
+static size_t
+ZSTD_BtFindBestMatch_selectMLS ( ZSTD_matchState_t* ms,
+ const BYTE* ip, const BYTE* const iLimit,
+ size_t* offsetPtr)
{
- switch(cParams->searchLength)
+ switch(ms->cParams.minMatch)
{
default : /* includes case 3 */
- case 4 : return ZSTD_BtFindBestMatch(ms, cParams, ip, iLimit, offsetPtr, 4);
- case 5 : return ZSTD_BtFindBestMatch(ms, cParams, ip, iLimit, offsetPtr, 5);
+ case 4 : return ZSTD_BtFindBestMatch(ms, ip, iLimit, offsetPtr, 4, ZSTD_noDict);
+ case 5 : return ZSTD_BtFindBestMatch(ms, ip, iLimit, offsetPtr, 5, ZSTD_noDict);
case 7 :
- case 6 : return ZSTD_BtFindBestMatch(ms, cParams, ip, iLimit, offsetPtr, 6);
+ case 6 : return ZSTD_BtFindBestMatch(ms, ip, iLimit, offsetPtr, 6, ZSTD_noDict);
}
}
-/** Tree updater, providing best match */
-static size_t ZSTD_BtFindBestMatch_extDict (
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
- const BYTE* const ip, const BYTE* const iLimit,
- size_t* offsetPtr,
- const U32 mls)
-{
- DEBUGLOG(7, "ZSTD_BtFindBestMatch_extDict");
- if (ip < ms->window.base + ms->nextToUpdate) return 0; /* skipped area */
- ZSTD_updateDUBT(ms, cParams, ip, iLimit, mls);
- return ZSTD_DUBT_findBestMatch(ms, cParams, ip, iLimit, offsetPtr, mls, 1);
-}
-
-
-static size_t ZSTD_BtFindBestMatch_selectMLS_extDict (
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+static size_t ZSTD_BtFindBestMatch_dictMatchState_selectMLS (
+ ZSTD_matchState_t* ms,
const BYTE* ip, const BYTE* const iLimit,
size_t* offsetPtr)
{
- switch(cParams->searchLength)
+ switch(ms->cParams.minMatch)
{
default : /* includes case 3 */
- case 4 : return ZSTD_BtFindBestMatch_extDict(ms, cParams, ip, iLimit, offsetPtr, 4);
- case 5 : return ZSTD_BtFindBestMatch_extDict(ms, cParams, ip, iLimit, offsetPtr, 5);
+ case 4 : return ZSTD_BtFindBestMatch(ms, ip, iLimit, offsetPtr, 4, ZSTD_dictMatchState);
+ case 5 : return ZSTD_BtFindBestMatch(ms, ip, iLimit, offsetPtr, 5, ZSTD_dictMatchState);
case 7 :
- case 6 : return ZSTD_BtFindBestMatch_extDict(ms, cParams, ip, iLimit, offsetPtr, 6);
+ case 6 : return ZSTD_BtFindBestMatch(ms, ip, iLimit, offsetPtr, 6, ZSTD_dictMatchState);
+ }
+}
+
+
+static size_t ZSTD_BtFindBestMatch_extDict_selectMLS (
+ ZSTD_matchState_t* ms,
+ const BYTE* ip, const BYTE* const iLimit,
+ size_t* offsetPtr)
+{
+ switch(ms->cParams.minMatch)
+ {
+ default : /* includes case 3 */
+ case 4 : return ZSTD_BtFindBestMatch(ms, ip, iLimit, offsetPtr, 4, ZSTD_extDict);
+ case 5 : return ZSTD_BtFindBestMatch(ms, ip, iLimit, offsetPtr, 5, ZSTD_extDict);
+ case 7 :
+ case 6 : return ZSTD_BtFindBestMatch(ms, ip, iLimit, offsetPtr, 6, ZSTD_extDict);
}
}
@@ -335,12 +444,13 @@
/* *********************************
* Hash Chain
***********************************/
-#define NEXT_IN_CHAIN(d, mask) chainTable[(d) & mask]
+#define NEXT_IN_CHAIN(d, mask) chainTable[(d) & (mask)]
/* Update chains up to ip (excluded)
Assumption : always within prefix (i.e. not within extDict) */
static U32 ZSTD_insertAndFindFirstIndex_internal(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+ ZSTD_matchState_t* ms,
+ const ZSTD_compressionParameters* const cParams,
const BYTE* ip, U32 const mls)
{
U32* const hashTable = ms->hashTable;
@@ -362,22 +472,21 @@
return hashTable[ZSTD_hashPtr(ip, hashLog, mls)];
}
-U32 ZSTD_insertAndFindFirstIndex(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
- const BYTE* ip)
-{
- return ZSTD_insertAndFindFirstIndex_internal(ms, cParams, ip, cParams->searchLength);
+U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip) {
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
+ return ZSTD_insertAndFindFirstIndex_internal(ms, cParams, ip, ms->cParams.minMatch);
}
/* inlining is important to hardwire a hot branch (template emulation) */
FORCE_INLINE_TEMPLATE
size_t ZSTD_HcFindBestMatch_generic (
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+ ZSTD_matchState_t* ms,
const BYTE* const ip, const BYTE* const iLimit,
size_t* offsetPtr,
- const U32 mls, const U32 extDict)
+ const U32 mls, const ZSTD_dictMode_e dictMode)
{
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const chainTable = ms->chainTable;
const U32 chainSize = (1 << cParams->chainLog);
const U32 chainMask = chainSize-1;
@@ -386,8 +495,10 @@
const U32 dictLimit = ms->window.dictLimit;
const BYTE* const prefixStart = base + dictLimit;
const BYTE* const dictEnd = dictBase + dictLimit;
- const U32 lowLimit = ms->window.lowLimit;
const U32 current = (U32)(ip-base);
+ const U32 maxDistance = 1U << cParams->windowLog;
+ const U32 lowValid = ms->window.lowLimit;
+ const U32 lowLimit = (current - lowValid > maxDistance) ? current - maxDistance : lowValid;
const U32 minChain = current > chainSize ? current - chainSize : 0;
U32 nbAttempts = 1U << cParams->searchLog;
size_t ml=4-1;
@@ -397,8 +508,9 @@
for ( ; (matchIndex>lowLimit) & (nbAttempts>0) ; nbAttempts--) {
size_t currentMl=0;
- if ((!extDict) || matchIndex >= dictLimit) {
+ if ((dictMode != ZSTD_extDict) || matchIndex >= dictLimit) {
const BYTE* const match = base + matchIndex;
+ assert(matchIndex >= dictLimit); /* ensures this is true if dictMode != ZSTD_extDict */
if (match[ml] == ip[ml]) /* potentially better */
currentMl = ZSTD_count(ip, match, iLimit);
} else {
@@ -419,38 +531,87 @@
matchIndex = NEXT_IN_CHAIN(matchIndex, chainMask);
}
+ if (dictMode == ZSTD_dictMatchState) {
+ const ZSTD_matchState_t* const dms = ms->dictMatchState;
+ const U32* const dmsChainTable = dms->chainTable;
+ const U32 dmsChainSize = (1 << dms->cParams.chainLog);
+ const U32 dmsChainMask = dmsChainSize - 1;
+ const U32 dmsLowestIndex = dms->window.dictLimit;
+ const BYTE* const dmsBase = dms->window.base;
+ const BYTE* const dmsEnd = dms->window.nextSrc;
+ const U32 dmsSize = (U32)(dmsEnd - dmsBase);
+ const U32 dmsIndexDelta = dictLimit - dmsSize;
+ const U32 dmsMinChain = dmsSize > dmsChainSize ? dmsSize - dmsChainSize : 0;
+
+ matchIndex = dms->hashTable[ZSTD_hashPtr(ip, dms->cParams.hashLog, mls)];
+
+ for ( ; (matchIndex>dmsLowestIndex) & (nbAttempts>0) ; nbAttempts--) {
+ size_t currentMl=0;
+ const BYTE* const match = dmsBase + matchIndex;
+ assert(match+4 <= dmsEnd);
+ if (MEM_read32(match) == MEM_read32(ip)) /* assumption : matchIndex <= dictLimit-4 (by table construction) */
+ currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dmsEnd, prefixStart) + 4;
+
+ /* save best solution */
+ if (currentMl > ml) {
+ ml = currentMl;
+ *offsetPtr = current - (matchIndex + dmsIndexDelta) + ZSTD_REP_MOVE;
+ if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */
+ }
+
+ if (matchIndex <= dmsMinChain) break;
+ matchIndex = dmsChainTable[matchIndex & dmsChainMask];
+ }
+ }
+
return ml;
}
FORCE_INLINE_TEMPLATE size_t ZSTD_HcFindBestMatch_selectMLS (
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+ ZSTD_matchState_t* ms,
const BYTE* ip, const BYTE* const iLimit,
size_t* offsetPtr)
{
- switch(cParams->searchLength)
+ switch(ms->cParams.minMatch)
{
default : /* includes case 3 */
- case 4 : return ZSTD_HcFindBestMatch_generic(ms, cParams, ip, iLimit, offsetPtr, 4, 0);
- case 5 : return ZSTD_HcFindBestMatch_generic(ms, cParams, ip, iLimit, offsetPtr, 5, 0);
+ case 4 : return ZSTD_HcFindBestMatch_generic(ms, ip, iLimit, offsetPtr, 4, ZSTD_noDict);
+ case 5 : return ZSTD_HcFindBestMatch_generic(ms, ip, iLimit, offsetPtr, 5, ZSTD_noDict);
case 7 :
- case 6 : return ZSTD_HcFindBestMatch_generic(ms, cParams, ip, iLimit, offsetPtr, 6, 0);
+ case 6 : return ZSTD_HcFindBestMatch_generic(ms, ip, iLimit, offsetPtr, 6, ZSTD_noDict);
+ }
+}
+
+
+static size_t ZSTD_HcFindBestMatch_dictMatchState_selectMLS (
+ ZSTD_matchState_t* ms,
+ const BYTE* ip, const BYTE* const iLimit,
+ size_t* offsetPtr)
+{
+ switch(ms->cParams.minMatch)
+ {
+ default : /* includes case 3 */
+ case 4 : return ZSTD_HcFindBestMatch_generic(ms, ip, iLimit, offsetPtr, 4, ZSTD_dictMatchState);
+ case 5 : return ZSTD_HcFindBestMatch_generic(ms, ip, iLimit, offsetPtr, 5, ZSTD_dictMatchState);
+ case 7 :
+ case 6 : return ZSTD_HcFindBestMatch_generic(ms, ip, iLimit, offsetPtr, 6, ZSTD_dictMatchState);
}
}
FORCE_INLINE_TEMPLATE size_t ZSTD_HcFindBestMatch_extDict_selectMLS (
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+ ZSTD_matchState_t* ms,
const BYTE* ip, const BYTE* const iLimit,
- size_t* const offsetPtr)
+ size_t* offsetPtr)
{
- switch(cParams->searchLength)
+ switch(ms->cParams.minMatch)
{
default : /* includes case 3 */
- case 4 : return ZSTD_HcFindBestMatch_generic(ms, cParams, ip, iLimit, offsetPtr, 4, 1);
- case 5 : return ZSTD_HcFindBestMatch_generic(ms, cParams, ip, iLimit, offsetPtr, 5, 1);
+ case 4 : return ZSTD_HcFindBestMatch_generic(ms, ip, iLimit, offsetPtr, 4, ZSTD_extDict);
+ case 5 : return ZSTD_HcFindBestMatch_generic(ms, ip, iLimit, offsetPtr, 5, ZSTD_extDict);
case 7 :
- case 6 : return ZSTD_HcFindBestMatch_generic(ms, cParams, ip, iLimit, offsetPtr, 6, 1);
+ case 6 : return ZSTD_HcFindBestMatch_generic(ms, ip, iLimit, offsetPtr, 6, ZSTD_extDict);
}
}
@@ -462,30 +623,54 @@
size_t ZSTD_compressBlock_lazy_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore,
U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams,
const void* src, size_t srcSize,
- const U32 searchMethod, const U32 depth)
+ const U32 searchMethod, const U32 depth,
+ ZSTD_dictMode_e const dictMode)
{
const BYTE* const istart = (const BYTE*)src;
const BYTE* ip = istart;
const BYTE* anchor = istart;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - 8;
- const BYTE* const base = ms->window.base + ms->window.dictLimit;
+ const BYTE* const base = ms->window.base;
+ const U32 prefixLowestIndex = ms->window.dictLimit;
+ const BYTE* const prefixLowest = base + prefixLowestIndex;
typedef size_t (*searchMax_f)(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+ ZSTD_matchState_t* ms,
const BYTE* ip, const BYTE* iLimit, size_t* offsetPtr);
- searchMax_f const searchMax = searchMethod ? ZSTD_BtFindBestMatch_selectMLS : ZSTD_HcFindBestMatch_selectMLS;
+ searchMax_f const searchMax = dictMode == ZSTD_dictMatchState ?
+ (searchMethod ? ZSTD_BtFindBestMatch_dictMatchState_selectMLS : ZSTD_HcFindBestMatch_dictMatchState_selectMLS) :
+ (searchMethod ? ZSTD_BtFindBestMatch_selectMLS : ZSTD_HcFindBestMatch_selectMLS);
U32 offset_1 = rep[0], offset_2 = rep[1], savedOffset=0;
+ const ZSTD_matchState_t* const dms = ms->dictMatchState;
+ const U32 dictLowestIndex = dictMode == ZSTD_dictMatchState ?
+ dms->window.dictLimit : 0;
+ const BYTE* const dictBase = dictMode == ZSTD_dictMatchState ?
+ dms->window.base : NULL;
+ const BYTE* const dictLowest = dictMode == ZSTD_dictMatchState ?
+ dictBase + dictLowestIndex : NULL;
+ const BYTE* const dictEnd = dictMode == ZSTD_dictMatchState ?
+ dms->window.nextSrc : NULL;
+ const U32 dictIndexDelta = dictMode == ZSTD_dictMatchState ?
+ prefixLowestIndex - (U32)(dictEnd - dictBase) :
+ 0;
+ const U32 dictAndPrefixLength = (U32)(ip - prefixLowest + dictEnd - dictLowest);
+
/* init */
- ip += (ip==base);
- ms->nextToUpdate3 = ms->nextToUpdate;
- { U32 const maxRep = (U32)(ip-base);
+ ip += (dictAndPrefixLength == 0);
+ if (dictMode == ZSTD_noDict) {
+ U32 const maxRep = (U32)(ip - prefixLowest);
if (offset_2 > maxRep) savedOffset = offset_2, offset_2 = 0;
if (offset_1 > maxRep) savedOffset = offset_1, offset_1 = 0;
}
+ if (dictMode == ZSTD_dictMatchState) {
+ /* dictMatchState repCode checks don't currently handle repCode == 0
+ * disabling. */
+ assert(offset_1 <= dictAndPrefixLength);
+ assert(offset_2 <= dictAndPrefixLength);
+ }
/* Match Loop */
while (ip < ilimit) {
@@ -494,15 +679,28 @@
const BYTE* start=ip+1;
/* check repCode */
- if ((offset_1>0) & (MEM_read32(ip+1) == MEM_read32(ip+1 - offset_1))) {
- /* repcode : we take it */
+ if (dictMode == ZSTD_dictMatchState) {
+ const U32 repIndex = (U32)(ip - base) + 1 - offset_1;
+ const BYTE* repMatch = (dictMode == ZSTD_dictMatchState
+ && repIndex < prefixLowestIndex) ?
+ dictBase + (repIndex - dictIndexDelta) :
+ base + repIndex;
+ if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */)
+ && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
+ const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
+ matchLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
+ if (depth==0) goto _storeSequence;
+ }
+ }
+ if ( dictMode == ZSTD_noDict
+ && ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1)))) {
matchLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
if (depth==0) goto _storeSequence;
}
/* first search (depth 0) */
- { size_t offsetFound = 99999999;
- size_t const ml2 = searchMax(ms, cParams, ip, iend, &offsetFound);
+ { size_t offsetFound = 999999999;
+ size_t const ml2 = searchMax(ms, ip, iend, &offsetFound);
if (ml2 > matchLength)
matchLength = ml2, start = ip, offset=offsetFound;
}
@@ -516,15 +714,31 @@
if (depth>=1)
while (ip<ilimit) {
ip ++;
- if ((offset) && ((offset_1>0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) {
+ if ( (dictMode == ZSTD_noDict)
+ && (offset) && ((offset_1>0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) {
size_t const mlRep = ZSTD_count(ip+4, ip+4-offset_1, iend) + 4;
int const gain2 = (int)(mlRep * 3);
int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)offset+1) + 1);
if ((mlRep >= 4) && (gain2 > gain1))
matchLength = mlRep, offset = 0, start = ip;
}
- { size_t offset2=99999999;
- size_t const ml2 = searchMax(ms, cParams, ip, iend, &offset2);
+ if (dictMode == ZSTD_dictMatchState) {
+ const U32 repIndex = (U32)(ip - base) - offset_1;
+ const BYTE* repMatch = repIndex < prefixLowestIndex ?
+ dictBase + (repIndex - dictIndexDelta) :
+ base + repIndex;
+ if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */)
+ && (MEM_read32(repMatch) == MEM_read32(ip)) ) {
+ const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
+ size_t const mlRep = ZSTD_count_2segments(ip+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
+ int const gain2 = (int)(mlRep * 3);
+ int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)offset+1) + 1);
+ if ((mlRep >= 4) && (gain2 > gain1))
+ matchLength = mlRep, offset = 0, start = ip;
+ }
+ }
+ { size_t offset2=999999999;
+ size_t const ml2 = searchMax(ms, ip, iend, &offset2);
int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 4);
if ((ml2 >= 4) && (gain2 > gain1)) {
@@ -535,15 +749,31 @@
/* let's find an even better one */
if ((depth==2) && (ip<ilimit)) {
ip ++;
- if ((offset) && ((offset_1>0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) {
- size_t const ml2 = ZSTD_count(ip+4, ip+4-offset_1, iend) + 4;
- int const gain2 = (int)(ml2 * 4);
+ if ( (dictMode == ZSTD_noDict)
+ && (offset) && ((offset_1>0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) {
+ size_t const mlRep = ZSTD_count(ip+4, ip+4-offset_1, iend) + 4;
+ int const gain2 = (int)(mlRep * 4);
int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 1);
- if ((ml2 >= 4) && (gain2 > gain1))
- matchLength = ml2, offset = 0, start = ip;
+ if ((mlRep >= 4) && (gain2 > gain1))
+ matchLength = mlRep, offset = 0, start = ip;
}
- { size_t offset2=99999999;
- size_t const ml2 = searchMax(ms, cParams, ip, iend, &offset2);
+ if (dictMode == ZSTD_dictMatchState) {
+ const U32 repIndex = (U32)(ip - base) - offset_1;
+ const BYTE* repMatch = repIndex < prefixLowestIndex ?
+ dictBase + (repIndex - dictIndexDelta) :
+ base + repIndex;
+ if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */)
+ && (MEM_read32(repMatch) == MEM_read32(ip)) ) {
+ const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
+ size_t const mlRep = ZSTD_count_2segments(ip+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
+ int const gain2 = (int)(mlRep * 4);
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 1);
+ if ((mlRep >= 4) && (gain2 > gain1))
+ matchLength = mlRep, offset = 0, start = ip;
+ }
+ }
+ { size_t offset2=999999999;
+ size_t const ml2 = searchMax(ms, ip, iend, &offset2);
int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 7);
if ((ml2 >= 4) && (gain2 > gain1)) {
@@ -560,9 +790,17 @@
*/
/* catch up */
if (offset) {
- while ( ((start > anchor) & (start - (offset-ZSTD_REP_MOVE) > base))
- && (start[-1] == (start-(offset-ZSTD_REP_MOVE))[-1]) ) /* only search for offset within prefix */
- { start--; matchLength++; }
+ if (dictMode == ZSTD_noDict) {
+ while ( ((start > anchor) & (start - (offset-ZSTD_REP_MOVE) > prefixLowest))
+ && (start[-1] == (start-(offset-ZSTD_REP_MOVE))[-1]) ) /* only search for offset within prefix */
+ { start--; matchLength++; }
+ }
+ if (dictMode == ZSTD_dictMatchState) {
+ U32 const matchIndex = (U32)((start-base) - (offset - ZSTD_REP_MOVE));
+ const BYTE* match = (matchIndex < prefixLowestIndex) ? dictBase + matchIndex - dictIndexDelta : base + matchIndex;
+ const BYTE* const mStart = (matchIndex < prefixLowestIndex) ? dictLowest : prefixLowest;
+ while ((start>anchor) && (match>mStart) && (start[-1] == match[-1])) { start--; match--; matchLength++; } /* catch up */
+ }
offset_2 = offset_1; offset_1 = (U32)(offset - ZSTD_REP_MOVE);
}
/* store sequence */
@@ -573,16 +811,39 @@
}
/* check immediate repcode */
- while ( ((ip <= ilimit) & (offset_2>0))
- && (MEM_read32(ip) == MEM_read32(ip - offset_2)) ) {
- /* store sequence */
- matchLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
- offset = offset_2; offset_2 = offset_1; offset_1 = (U32)offset; /* swap repcodes */
- ZSTD_storeSeq(seqStore, 0, anchor, 0, matchLength-MINMATCH);
- ip += matchLength;
- anchor = ip;
- continue; /* faster when present ... (?) */
- } }
+ if (dictMode == ZSTD_dictMatchState) {
+ while (ip <= ilimit) {
+ U32 const current2 = (U32)(ip-base);
+ U32 const repIndex = current2 - offset_2;
+ const BYTE* repMatch = dictMode == ZSTD_dictMatchState
+ && repIndex < prefixLowestIndex ?
+ dictBase - dictIndexDelta + repIndex :
+ base + repIndex;
+ if ( ((U32)((prefixLowestIndex-1) - (U32)repIndex) >= 3 /* intentional overflow */)
+ && (MEM_read32(repMatch) == MEM_read32(ip)) ) {
+ const BYTE* const repEnd2 = repIndex < prefixLowestIndex ? dictEnd : iend;
+ matchLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd2, prefixLowest) + 4;
+ offset = offset_2; offset_2 = offset_1; offset_1 = (U32)offset; /* swap offset_2 <=> offset_1 */
+ ZSTD_storeSeq(seqStore, 0, anchor, 0, matchLength-MINMATCH);
+ ip += matchLength;
+ anchor = ip;
+ continue;
+ }
+ break;
+ }
+ }
+
+ if (dictMode == ZSTD_noDict) {
+ while ( ((ip <= ilimit) & (offset_2>0))
+ && (MEM_read32(ip) == MEM_read32(ip - offset_2)) ) {
+ /* store sequence */
+ matchLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
+ offset = offset_2; offset_2 = offset_1; offset_1 = (U32)offset; /* swap repcodes */
+ ZSTD_storeSeq(seqStore, 0, anchor, 0, matchLength-MINMATCH);
+ ip += matchLength;
+ anchor = ip;
+ continue; /* faster when present ... (?) */
+ } } }
/* Save reps for next block */
rep[0] = offset_1 ? offset_1 : savedOffset;
@@ -595,30 +856,58 @@
size_t ZSTD_compressBlock_btlazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, cParams, src, srcSize, 1, 2);
+ return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, 1, 2, ZSTD_noDict);
}
size_t ZSTD_compressBlock_lazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, cParams, src, srcSize, 0, 2);
+ return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, 0, 2, ZSTD_noDict);
}
size_t ZSTD_compressBlock_lazy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, cParams, src, srcSize, 0, 1);
+ return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, 0, 1, ZSTD_noDict);
}
size_t ZSTD_compressBlock_greedy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, cParams, src, srcSize, 0, 0);
+ return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, 0, 0, ZSTD_noDict);
+}
+
+size_t ZSTD_compressBlock_btlazy2_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, 1, 2, ZSTD_dictMatchState);
+}
+
+size_t ZSTD_compressBlock_lazy2_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, 0, 2, ZSTD_dictMatchState);
+}
+
+size_t ZSTD_compressBlock_lazy_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, 0, 1, ZSTD_dictMatchState);
+}
+
+size_t ZSTD_compressBlock_greedy_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, 0, 0, ZSTD_dictMatchState);
}
@@ -626,7 +915,6 @@
size_t ZSTD_compressBlock_lazy_extDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore,
U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams,
const void* src, size_t srcSize,
const U32 searchMethod, const U32 depth)
{
@@ -644,14 +932,13 @@
const BYTE* const dictStart = dictBase + lowestIndex;
typedef size_t (*searchMax_f)(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+ ZSTD_matchState_t* ms,
const BYTE* ip, const BYTE* iLimit, size_t* offsetPtr);
- searchMax_f searchMax = searchMethod ? ZSTD_BtFindBestMatch_selectMLS_extDict : ZSTD_HcFindBestMatch_extDict_selectMLS;
+ searchMax_f searchMax = searchMethod ? ZSTD_BtFindBestMatch_extDict_selectMLS : ZSTD_HcFindBestMatch_extDict_selectMLS;
U32 offset_1 = rep[0], offset_2 = rep[1];
/* init */
- ms->nextToUpdate3 = ms->nextToUpdate;
ip += (ip == prefixStart);
/* Match Loop */
@@ -674,8 +961,8 @@
} }
/* first search (depth 0) */
- { size_t offsetFound = 99999999;
- size_t const ml2 = searchMax(ms, cParams, ip, iend, &offsetFound);
+ { size_t offsetFound = 999999999;
+ size_t const ml2 = searchMax(ms, ip, iend, &offsetFound);
if (ml2 > matchLength)
matchLength = ml2, start = ip, offset=offsetFound;
}
@@ -707,8 +994,8 @@
} }
/* search match, depth 1 */
- { size_t offset2=99999999;
- size_t const ml2 = searchMax(ms, cParams, ip, iend, &offset2);
+ { size_t offset2=999999999;
+ size_t const ml2 = searchMax(ms, ip, iend, &offset2);
int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 4);
if ((ml2 >= 4) && (gain2 > gain1)) {
@@ -737,8 +1024,8 @@
} }
/* search match, depth 2 */
- { size_t offset2=99999999;
- size_t const ml2 = searchMax(ms, cParams, ip, iend, &offset2);
+ { size_t offset2=999999999;
+ size_t const ml2 = searchMax(ms, ip, iend, &offset2);
int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 7);
if ((ml2 >= 4) && (gain2 > gain1)) {
@@ -794,31 +1081,31 @@
size_t ZSTD_compressBlock_greedy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, cParams, src, srcSize, 0, 0);
+ return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, 0, 0);
}
size_t ZSTD_compressBlock_lazy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, cParams, src, srcSize, 0, 1);
+ return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, 0, 1);
}
size_t ZSTD_compressBlock_lazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, cParams, src, srcSize, 0, 2);
+ return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, 0, 2);
}
size_t ZSTD_compressBlock_btlazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ void const* src, size_t srcSize)
{
- return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, cParams, src, srcSize, 1, 2);
+ return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, 1, 2);
}
diff --git a/vendor/github.com/DataDog/zstd/zstd_lazy.h b/vendor/github.com/DataDog/zstd/zstd_lazy.h
index bda064f..bb17630 100644
--- a/vendor/github.com/DataDog/zstd/zstd_lazy.h
+++ b/vendor/github.com/DataDog/zstd/zstd_lazy.h
@@ -17,37 +17,48 @@
#include "zstd_compress_internal.h"
-U32 ZSTD_insertAndFindFirstIndex(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
- const BYTE* ip);
+U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip);
-void ZSTD_preserveUnsortedMark (U32* const table, U32 const size, U32 const reducerValue); /*! used in ZSTD_reduceIndex(). pre-emptively increase value of ZSTD_DUBT_UNSORTED_MARK */
+void ZSTD_preserveUnsortedMark (U32* const table, U32 const size, U32 const reducerValue); /*! used in ZSTD_reduceIndex(). preemptively increase value of ZSTD_DUBT_UNSORTED_MARK */
size_t ZSTD_compressBlock_btlazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
+
+size_t ZSTD_compressBlock_btlazy2_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize);
+size_t ZSTD_compressBlock_lazy2_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize);
+size_t ZSTD_compressBlock_lazy_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize);
+size_t ZSTD_compressBlock_greedy_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btlazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
#if defined (__cplusplus)
}
diff --git a/vendor/github.com/DataDog/zstd/zstd_ldm.c b/vendor/github.com/DataDog/zstd/zstd_ldm.c
index bffd8a3..3dcf86e 100644
--- a/vendor/github.com/DataDog/zstd/zstd_ldm.c
+++ b/vendor/github.com/DataDog/zstd/zstd_ldm.c
@@ -9,6 +9,7 @@
#include "zstd_ldm.h"
+#include "debug.h"
#include "zstd_fast.h" /* ZSTD_fillHashTable() */
#include "zstd_double_fast.h" /* ZSTD_fillDoubleHashTable() */
@@ -20,7 +21,7 @@
void ZSTD_ldm_adjustParameters(ldmParams_t* params,
ZSTD_compressionParameters const* cParams)
{
- U32 const windowLog = cParams->windowLog;
+ params->windowLog = cParams->windowLog;
ZSTD_STATIC_ASSERT(LDM_BUCKET_SIZE_LOG <= ZSTD_LDM_BUCKETSIZELOG_MAX);
DEBUGLOG(4, "ZSTD_ldm_adjustParameters");
if (!params->bucketSizeLog) params->bucketSizeLog = LDM_BUCKET_SIZE_LOG;
@@ -33,12 +34,13 @@
params->minMatchLength = minMatch;
}
if (params->hashLog == 0) {
- params->hashLog = MAX(ZSTD_HASHLOG_MIN, windowLog - LDM_HASH_RLOG);
+ params->hashLog = MAX(ZSTD_HASHLOG_MIN, params->windowLog - LDM_HASH_RLOG);
assert(params->hashLog <= ZSTD_HASHLOG_MAX);
}
- if (params->hashEveryLog == 0) {
- params->hashEveryLog =
- windowLog < params->hashLog ? 0 : windowLog - params->hashLog;
+ if (params->hashRateLog == 0) {
+ params->hashRateLog = params->windowLog < params->hashLog
+ ? 0
+ : params->windowLog - params->hashLog;
}
params->bucketSizeLog = MIN(params->bucketSizeLog, params->hashLog);
}
@@ -117,20 +119,20 @@
*
* Gets the small hash, checksum, and tag from the rollingHash.
*
- * If the tag matches (1 << ldmParams.hashEveryLog)-1, then
+ * If the tag matches (1 << ldmParams.hashRateLog)-1, then
* creates an ldmEntry from the offset, and inserts it into the hash table.
*
* hBits is the length of the small hash, which is the most significant hBits
* of rollingHash. The checksum is the next 32 most significant bits, followed
- * by ldmParams.hashEveryLog bits that make up the tag. */
+ * by ldmParams.hashRateLog bits that make up the tag. */
static void ZSTD_ldm_makeEntryAndInsertByTag(ldmState_t* ldmState,
U64 const rollingHash,
U32 const hBits,
U32 const offset,
ldmParams_t const ldmParams)
{
- U32 const tag = ZSTD_ldm_getTag(rollingHash, hBits, ldmParams.hashEveryLog);
- U32 const tagMask = ((U32)1 << ldmParams.hashEveryLog) - 1;
+ U32 const tag = ZSTD_ldm_getTag(rollingHash, hBits, ldmParams.hashRateLog);
+ U32 const tagMask = ((U32)1 << ldmParams.hashRateLog) - 1;
if (tag == tagMask) {
U32 const hash = ZSTD_ldm_getSmallHash(rollingHash, hBits);
U32 const checksum = ZSTD_ldm_getChecksum(rollingHash, hBits);
@@ -141,56 +143,6 @@
}
}
-/** ZSTD_ldm_getRollingHash() :
- * Get a 64-bit hash using the first len bytes from buf.
- *
- * Giving bytes s = s_1, s_2, ... s_k, the hash is defined to be
- * H(s) = s_1*(a^(k-1)) + s_2*(a^(k-2)) + ... + s_k*(a^0)
- *
- * where the constant a is defined to be prime8bytes.
- *
- * The implementation adds an offset to each byte, so
- * H(s) = (s_1 + HASH_CHAR_OFFSET)*(a^(k-1)) + ... */
-static U64 ZSTD_ldm_getRollingHash(const BYTE* buf, U32 len)
-{
- U64 ret = 0;
- U32 i;
- for (i = 0; i < len; i++) {
- ret *= prime8bytes;
- ret += buf[i] + LDM_HASH_CHAR_OFFSET;
- }
- return ret;
-}
-
-/** ZSTD_ldm_ipow() :
- * Return base^exp. */
-static U64 ZSTD_ldm_ipow(U64 base, U64 exp)
-{
- U64 ret = 1;
- while (exp) {
- if (exp & 1) { ret *= base; }
- exp >>= 1;
- base *= base;
- }
- return ret;
-}
-
-U64 ZSTD_ldm_getHashPower(U32 minMatchLength) {
- DEBUGLOG(4, "ZSTD_ldm_getHashPower: mml=%u", minMatchLength);
- assert(minMatchLength >= ZSTD_LDM_MINMATCH_MIN);
- return ZSTD_ldm_ipow(prime8bytes, minMatchLength - 1);
-}
-
-/** ZSTD_ldm_updateHash() :
- * Updates hash by removing toRemove and adding toAdd. */
-static U64 ZSTD_ldm_updateHash(U64 hash, BYTE toRemove, BYTE toAdd, U64 hashPower)
-{
- hash -= ((toRemove + LDM_HASH_CHAR_OFFSET) * hashPower);
- hash *= prime8bytes;
- hash += toAdd + LDM_HASH_CHAR_OFFSET;
- return hash;
-}
-
/** ZSTD_ldm_countBackwardsMatch() :
* Returns the number of bytes that match backwards before pIn and pMatch.
*
@@ -216,21 +168,18 @@
* The tables for the other strategies are filled within their
* block compressors. */
static size_t ZSTD_ldm_fillFastTables(ZSTD_matchState_t* ms,
- ZSTD_compressionParameters const* cParams,
void const* end)
{
const BYTE* const iend = (const BYTE*)end;
- switch(cParams->strategy)
+ switch(ms->cParams.strategy)
{
case ZSTD_fast:
- ZSTD_fillHashTable(ms, cParams, iend);
- ms->nextToUpdate = (U32)(iend - ms->window.base);
+ ZSTD_fillHashTable(ms, iend, ZSTD_dtlm_fast);
break;
case ZSTD_dfast:
- ZSTD_fillDoubleHashTable(ms, cParams, iend);
- ms->nextToUpdate = (U32)(iend - ms->window.base);
+ ZSTD_fillDoubleHashTable(ms, iend, ZSTD_dtlm_fast);
break;
case ZSTD_greedy:
@@ -239,6 +188,7 @@
case ZSTD_btlazy2:
case ZSTD_btopt:
case ZSTD_btultra:
+ case ZSTD_btultra2:
break;
default:
assert(0); /* not possible : not a valid strategy id */
@@ -262,9 +212,9 @@
const BYTE* cur = lastHashed + 1;
while (cur < iend) {
- rollingHash = ZSTD_ldm_updateHash(rollingHash, cur[-1],
- cur[ldmParams.minMatchLength-1],
- state->hashPower);
+ rollingHash = ZSTD_rollingHash_rotate(rollingHash, cur[-1],
+ cur[ldmParams.minMatchLength-1],
+ state->hashPower);
ZSTD_ldm_makeEntryAndInsertByTag(state,
rollingHash, hBits,
(U32)(cur - base), ldmParams);
@@ -298,8 +248,8 @@
U64 const hashPower = ldmState->hashPower;
U32 const hBits = params->hashLog - params->bucketSizeLog;
U32 const ldmBucketSize = 1U << params->bucketSizeLog;
- U32 const hashEveryLog = params->hashEveryLog;
- U32 const ldmTagMask = (1U << params->hashEveryLog) - 1;
+ U32 const hashRateLog = params->hashRateLog;
+ U32 const ldmTagMask = (1U << params->hashRateLog) - 1;
/* Prefix and extDict parameters */
U32 const dictLimit = ldmState->window.dictLimit;
U32 const lowestIndex = extDict ? ldmState->window.lowLimit : dictLimit;
@@ -325,16 +275,16 @@
size_t forwardMatchLength = 0, backwardMatchLength = 0;
ldmEntry_t* bestEntry = NULL;
if (ip != istart) {
- rollingHash = ZSTD_ldm_updateHash(rollingHash, lastHashed[0],
- lastHashed[minMatchLength],
- hashPower);
+ rollingHash = ZSTD_rollingHash_rotate(rollingHash, lastHashed[0],
+ lastHashed[minMatchLength],
+ hashPower);
} else {
- rollingHash = ZSTD_ldm_getRollingHash(ip, minMatchLength);
+ rollingHash = ZSTD_rollingHash_compute(ip, minMatchLength);
}
lastHashed = ip;
/* Do not insert and do not look for a match */
- if (ZSTD_ldm_getTag(rollingHash, hBits, hashEveryLog) != ldmTagMask) {
+ if (ZSTD_ldm_getTag(rollingHash, hBits, hashRateLog) != ldmTagMask) {
ip++;
continue;
}
@@ -479,7 +429,7 @@
*/
assert(ldmState->window.nextSrc >= (BYTE const*)src + srcSize);
/* The input could be very large (in zstdmt), so it must be broken up into
- * chunks to enforce the maximmum distance and handle overflow correction.
+ * chunks to enforce the maximum distance and handle overflow correction.
*/
assert(sequences->pos <= sequences->size);
assert(sequences->size <= sequences->capacity);
@@ -497,7 +447,7 @@
if (ZSTD_window_needOverflowCorrection(ldmState->window, chunkEnd)) {
U32 const ldmHSize = 1U << params->hashLog;
U32 const correction = ZSTD_window_correctOverflow(
- &ldmState->window, /* cycleLog */ 0, maxDist, src);
+ &ldmState->window, /* cycleLog */ 0, maxDist, chunkStart);
ZSTD_ldm_reduceTable(ldmState->hashTable, ldmHSize, correction);
}
/* 2. We enforce the maximum offset allowed.
@@ -508,7 +458,7 @@
* * Try invalidation after the sequence generation and test the
* the offset against maxDist directly.
*/
- ZSTD_window_enforceMaxDist(&ldmState->window, chunkEnd, maxDist, NULL);
+ ZSTD_window_enforceMaxDist(&ldmState->window, chunkEnd, maxDist, NULL, NULL);
/* 3. Generate the sequences for the chunk, and get newLeftoverSize. */
newLeftoverSize = ZSTD_ldm_generateSequences_internal(
ldmState, sequences, params, chunkStart, chunkSize);
@@ -591,19 +541,19 @@
size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize,
- int const extDict)
+ void const* src, size_t srcSize)
{
- unsigned const minMatch = cParams->searchLength;
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
+ unsigned const minMatch = cParams->minMatch;
ZSTD_blockCompressor const blockCompressor =
- ZSTD_selectBlockCompressor(cParams->strategy, extDict);
- BYTE const* const base = ms->window.base;
+ ZSTD_selectBlockCompressor(cParams->strategy, ZSTD_matchState_dictMode(ms));
/* Input bounds */
BYTE const* const istart = (BYTE const*)src;
BYTE const* const iend = istart + srcSize;
/* Input positions */
BYTE const* ip = istart;
+ DEBUGLOG(5, "ZSTD_ldm_blockCompress: srcSize=%zu", srcSize);
assert(rawSeqStore->pos <= rawSeqStore->size);
assert(rawSeqStore->size <= rawSeqStore->capacity);
/* Loop through each sequence and apply the block compressor to the lits */
@@ -621,14 +571,13 @@
/* Fill tables for block compressor */
ZSTD_ldm_limitTableUpdate(ms, ip);
- ZSTD_ldm_fillFastTables(ms, cParams, ip);
+ ZSTD_ldm_fillFastTables(ms, ip);
/* Run the block compressor */
+ DEBUGLOG(5, "calling block compressor on segment of size %u", sequence.litLength);
{
size_t const newLitLength =
- blockCompressor(ms, seqStore, rep, cParams, ip,
- sequence.litLength);
+ blockCompressor(ms, seqStore, rep, ip, sequence.litLength);
ip += sequence.litLength;
- ms->nextToUpdate = (U32)(ip - base);
/* Update the repcodes */
for (i = ZSTD_REP_NUM - 1; i > 0; i--)
rep[i] = rep[i-1];
@@ -642,12 +591,7 @@
}
/* Fill the tables for the block compressor */
ZSTD_ldm_limitTableUpdate(ms, ip);
- ZSTD_ldm_fillFastTables(ms, cParams, ip);
+ ZSTD_ldm_fillFastTables(ms, ip);
/* Compress the last literals */
- {
- size_t const lastLiterals = blockCompressor(ms, seqStore, rep, cParams,
- ip, iend - ip);
- ms->nextToUpdate = (U32)(iend - base);
- return lastLiterals;
- }
+ return blockCompressor(ms, seqStore, rep, ip, iend - ip);
}
diff --git a/vendor/github.com/DataDog/zstd/zstd_ldm.h b/vendor/github.com/DataDog/zstd/zstd_ldm.h
index 0c3789f..a478461 100644
--- a/vendor/github.com/DataDog/zstd/zstd_ldm.h
+++ b/vendor/github.com/DataDog/zstd/zstd_ldm.h
@@ -21,7 +21,7 @@
* Long distance matching
***************************************/
-#define ZSTD_LDM_DEFAULT_WINDOW_LOG ZSTD_WINDOWLOG_DEFAULTMAX
+#define ZSTD_LDM_DEFAULT_WINDOW_LOG ZSTD_WINDOWLOG_LIMIT_DEFAULT
/**
* ZSTD_ldm_generateSequences():
@@ -61,9 +61,7 @@
*/
size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams,
- void const* src, size_t srcSize,
- int const extDict);
+ void const* src, size_t srcSize);
/**
* ZSTD_ldm_skipSequences():
@@ -88,12 +86,8 @@
*/
size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize);
-/** ZSTD_ldm_getTableSize() :
- * Return prime8bytes^(minMatchLength-1) */
-U64 ZSTD_ldm_getHashPower(U32 minMatchLength);
-
/** ZSTD_ldm_adjustParameters() :
- * If the params->hashEveryLog is not set, set it to its default value based on
+ * If the params->hashRateLog is not set, set it to its default value based on
* windowLog and params->hashLog.
*
* Ensures that params->bucketSizeLog is <= params->hashLog (setting it to
diff --git a/vendor/github.com/DataDog/zstd/zstd_legacy.h b/vendor/github.com/DataDog/zstd/zstd_legacy.h
index 5893cb9..0dbd3c7 100644
--- a/vendor/github.com/DataDog/zstd/zstd_legacy.h
+++ b/vendor/github.com/DataDog/zstd/zstd_legacy.h
@@ -20,7 +20,7 @@
***************************************/
#include "mem.h" /* MEM_STATIC */
#include "error_private.h" /* ERROR */
-#include "zstd.h" /* ZSTD_inBuffer, ZSTD_outBuffer */
+#include "zstd_internal.h" /* ZSTD_inBuffer, ZSTD_outBuffer, ZSTD_frameSizeInfo */
#if !defined (ZSTD_LEGACY_SUPPORT) || (ZSTD_LEGACY_SUPPORT == 0)
# undef ZSTD_LEGACY_SUPPORT
@@ -178,43 +178,77 @@
}
}
-MEM_STATIC size_t ZSTD_findFrameCompressedSizeLegacy(const void *src,
- size_t compressedSize)
+MEM_STATIC ZSTD_frameSizeInfo ZSTD_findFrameSizeInfoLegacy(const void *src, size_t srcSize)
{
- U32 const version = ZSTD_isLegacy(src, compressedSize);
+ ZSTD_frameSizeInfo frameSizeInfo;
+ U32 const version = ZSTD_isLegacy(src, srcSize);
switch(version)
{
#if (ZSTD_LEGACY_SUPPORT <= 1)
case 1 :
- return ZSTDv01_findFrameCompressedSize(src, compressedSize);
+ ZSTDv01_findFrameSizeInfoLegacy(src, srcSize,
+ &frameSizeInfo.compressedSize,
+ &frameSizeInfo.decompressedBound);
+ break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 2)
case 2 :
- return ZSTDv02_findFrameCompressedSize(src, compressedSize);
+ ZSTDv02_findFrameSizeInfoLegacy(src, srcSize,
+ &frameSizeInfo.compressedSize,
+ &frameSizeInfo.decompressedBound);
+ break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 3)
case 3 :
- return ZSTDv03_findFrameCompressedSize(src, compressedSize);
+ ZSTDv03_findFrameSizeInfoLegacy(src, srcSize,
+ &frameSizeInfo.compressedSize,
+ &frameSizeInfo.decompressedBound);
+ break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 4)
case 4 :
- return ZSTDv04_findFrameCompressedSize(src, compressedSize);
+ ZSTDv04_findFrameSizeInfoLegacy(src, srcSize,
+ &frameSizeInfo.compressedSize,
+ &frameSizeInfo.decompressedBound);
+ break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 5)
case 5 :
- return ZSTDv05_findFrameCompressedSize(src, compressedSize);
+ ZSTDv05_findFrameSizeInfoLegacy(src, srcSize,
+ &frameSizeInfo.compressedSize,
+ &frameSizeInfo.decompressedBound);
+ break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 6)
case 6 :
- return ZSTDv06_findFrameCompressedSize(src, compressedSize);
+ ZSTDv06_findFrameSizeInfoLegacy(src, srcSize,
+ &frameSizeInfo.compressedSize,
+ &frameSizeInfo.decompressedBound);
+ break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 7)
case 7 :
- return ZSTDv07_findFrameCompressedSize(src, compressedSize);
+ ZSTDv07_findFrameSizeInfoLegacy(src, srcSize,
+ &frameSizeInfo.compressedSize,
+ &frameSizeInfo.decompressedBound);
+ break;
#endif
default :
- return ERROR(prefix_unknown);
+ frameSizeInfo.compressedSize = ERROR(prefix_unknown);
+ frameSizeInfo.decompressedBound = ZSTD_CONTENTSIZE_ERROR;
+ break;
}
+ if (!ZSTD_isError(frameSizeInfo.compressedSize) && frameSizeInfo.compressedSize > srcSize) {
+ frameSizeInfo.compressedSize = ERROR(srcSize_wrong);
+ frameSizeInfo.decompressedBound = ZSTD_CONTENTSIZE_ERROR;
+ }
+ return frameSizeInfo;
+}
+
+MEM_STATIC size_t ZSTD_findFrameCompressedSizeLegacy(const void *src, size_t srcSize)
+{
+ ZSTD_frameSizeInfo frameSizeInfo = ZSTD_findFrameSizeInfoLegacy(src, srcSize);
+ return frameSizeInfo.compressedSize;
}
MEM_STATIC size_t ZSTD_freeLegacyStreamContext(void* legacyContext, U32 version)
diff --git a/vendor/github.com/DataDog/zstd/zstd_opt.c b/vendor/github.com/DataDog/zstd/zstd_opt.c
index f63f0c5..e32e542 100644
--- a/vendor/github.com/DataDog/zstd/zstd_opt.c
+++ b/vendor/github.com/DataDog/zstd/zstd_opt.c
@@ -9,139 +9,259 @@
*/
#include "zstd_compress_internal.h"
+#include "hist.h"
#include "zstd_opt.h"
-#define ZSTD_LITFREQ_ADD 2 /* scaling factor for litFreq, so that frequencies adapt faster to new stats. Also used for matchSum (?) */
+#define ZSTD_LITFREQ_ADD 2 /* scaling factor for litFreq, so that frequencies adapt faster to new stats */
#define ZSTD_FREQ_DIV 4 /* log factor when using previous stats to init next stats */
#define ZSTD_MAX_PRICE (1<<30)
+#define ZSTD_PREDEF_THRESHOLD 1024 /* if srcSize < ZSTD_PREDEF_THRESHOLD, symbols' cost is assumed static, directly determined by pre-defined distributions */
+
/*-*************************************
* Price functions for optimal parser
***************************************/
-static void ZSTD_setLog2Prices(optState_t* optPtr)
+
+#if 0 /* approximation at bit level */
+# define BITCOST_ACCURACY 0
+# define BITCOST_MULTIPLIER (1 << BITCOST_ACCURACY)
+# define WEIGHT(stat) ((void)opt, ZSTD_bitWeight(stat))
+#elif 0 /* fractional bit accuracy */
+# define BITCOST_ACCURACY 8
+# define BITCOST_MULTIPLIER (1 << BITCOST_ACCURACY)
+# define WEIGHT(stat,opt) ((void)opt, ZSTD_fracWeight(stat))
+#else /* opt==approx, ultra==accurate */
+# define BITCOST_ACCURACY 8
+# define BITCOST_MULTIPLIER (1 << BITCOST_ACCURACY)
+# define WEIGHT(stat,opt) (opt ? ZSTD_fracWeight(stat) : ZSTD_bitWeight(stat))
+#endif
+
+MEM_STATIC U32 ZSTD_bitWeight(U32 stat)
{
- optPtr->log2litSum = ZSTD_highbit32(optPtr->litSum+1);
- optPtr->log2litLengthSum = ZSTD_highbit32(optPtr->litLengthSum+1);
- optPtr->log2matchLengthSum = ZSTD_highbit32(optPtr->matchLengthSum+1);
- optPtr->log2offCodeSum = ZSTD_highbit32(optPtr->offCodeSum+1);
+ return (ZSTD_highbit32(stat+1) * BITCOST_MULTIPLIER);
+}
+
+MEM_STATIC U32 ZSTD_fracWeight(U32 rawStat)
+{
+ U32 const stat = rawStat + 1;
+ U32 const hb = ZSTD_highbit32(stat);
+ U32 const BWeight = hb * BITCOST_MULTIPLIER;
+ U32 const FWeight = (stat << BITCOST_ACCURACY) >> hb;
+ U32 const weight = BWeight + FWeight;
+ assert(hb + BITCOST_ACCURACY < 31);
+ return weight;
+}
+
+#if (DEBUGLEVEL>=2)
+/* debugging function,
+ * @return price in bytes as fractional value
+ * for debug messages only */
+MEM_STATIC double ZSTD_fCost(U32 price)
+{
+ return (double)price / (BITCOST_MULTIPLIER*8);
+}
+#endif
+
+static int ZSTD_compressedLiterals(optState_t const* const optPtr)
+{
+ return optPtr->literalCompressionMode != ZSTD_lcm_uncompressed;
+}
+
+static void ZSTD_setBasePrices(optState_t* optPtr, int optLevel)
+{
+ if (ZSTD_compressedLiterals(optPtr))
+ optPtr->litSumBasePrice = WEIGHT(optPtr->litSum, optLevel);
+ optPtr->litLengthSumBasePrice = WEIGHT(optPtr->litLengthSum, optLevel);
+ optPtr->matchLengthSumBasePrice = WEIGHT(optPtr->matchLengthSum, optLevel);
+ optPtr->offCodeSumBasePrice = WEIGHT(optPtr->offCodeSum, optLevel);
}
-static void ZSTD_rescaleFreqs(optState_t* const optPtr,
- const BYTE* const src, size_t const srcSize)
+/* ZSTD_downscaleStat() :
+ * reduce all elements in table by a factor 2^(ZSTD_FREQ_DIV+malus)
+ * return the resulting sum of elements */
+static U32 ZSTD_downscaleStat(unsigned* table, U32 lastEltIndex, int malus)
{
- optPtr->staticPrices = 0;
+ U32 s, sum=0;
+ DEBUGLOG(5, "ZSTD_downscaleStat (nbElts=%u)", (unsigned)lastEltIndex+1);
+ assert(ZSTD_FREQ_DIV+malus > 0 && ZSTD_FREQ_DIV+malus < 31);
+ for (s=0; s<lastEltIndex+1; s++) {
+ table[s] = 1 + (table[s] >> (ZSTD_FREQ_DIV+malus));
+ sum += table[s];
+ }
+ return sum;
+}
- if (optPtr->litLengthSum == 0) { /* first init */
- unsigned u;
- if (srcSize <= 1024) optPtr->staticPrices = 1;
+/* ZSTD_rescaleFreqs() :
+ * if first block (detected by optPtr->litLengthSum == 0) : init statistics
+ * take hints from dictionary if there is one
+ * or init from zero, using src for literals stats, or flat 1 for match symbols
+ * otherwise downscale existing stats, to be used as seed for next block.
+ */
+static void
+ZSTD_rescaleFreqs(optState_t* const optPtr,
+ const BYTE* const src, size_t const srcSize,
+ int const optLevel)
+{
+ int const compressedLiterals = ZSTD_compressedLiterals(optPtr);
+ DEBUGLOG(5, "ZSTD_rescaleFreqs (srcSize=%u)", (unsigned)srcSize);
+ optPtr->priceType = zop_dynamic;
- assert(optPtr->litFreq!=NULL);
- for (u=0; u<=MaxLit; u++)
- optPtr->litFreq[u] = 0;
- for (u=0; u<srcSize; u++)
- optPtr->litFreq[src[u]]++;
- optPtr->litSum = 0;
- for (u=0; u<=MaxLit; u++) {
- optPtr->litFreq[u] = 1 + (optPtr->litFreq[u] >> ZSTD_FREQ_DIV);
- optPtr->litSum += optPtr->litFreq[u];
+ if (optPtr->litLengthSum == 0) { /* first block : init */
+ if (srcSize <= ZSTD_PREDEF_THRESHOLD) { /* heuristic */
+ DEBUGLOG(5, "(srcSize <= ZSTD_PREDEF_THRESHOLD) => zop_predef");
+ optPtr->priceType = zop_predef;
}
- for (u=0; u<=MaxLL; u++)
- optPtr->litLengthFreq[u] = 1;
- optPtr->litLengthSum = MaxLL+1;
- for (u=0; u<=MaxML; u++)
- optPtr->matchLengthFreq[u] = 1;
- optPtr->matchLengthSum = MaxML+1;
- for (u=0; u<=MaxOff; u++)
- optPtr->offCodeFreq[u] = 1;
- optPtr->offCodeSum = (MaxOff+1);
+ assert(optPtr->symbolCosts != NULL);
+ if (optPtr->symbolCosts->huf.repeatMode == HUF_repeat_valid) {
+ /* huffman table presumed generated by dictionary */
+ optPtr->priceType = zop_dynamic;
- } else {
- unsigned u;
+ if (compressedLiterals) {
+ unsigned lit;
+ assert(optPtr->litFreq != NULL);
+ optPtr->litSum = 0;
+ for (lit=0; lit<=MaxLit; lit++) {
+ U32 const scaleLog = 11; /* scale to 2K */
+ U32 const bitCost = HUF_getNbBits(optPtr->symbolCosts->huf.CTable, lit);
+ assert(bitCost <= scaleLog);
+ optPtr->litFreq[lit] = bitCost ? 1 << (scaleLog-bitCost) : 1 /*minimum to calculate cost*/;
+ optPtr->litSum += optPtr->litFreq[lit];
+ } }
- optPtr->litSum = 0;
- for (u=0; u<=MaxLit; u++) {
- optPtr->litFreq[u] = 1 + (optPtr->litFreq[u] >> (ZSTD_FREQ_DIV+1));
- optPtr->litSum += optPtr->litFreq[u];
+ { unsigned ll;
+ FSE_CState_t llstate;
+ FSE_initCState(&llstate, optPtr->symbolCosts->fse.litlengthCTable);
+ optPtr->litLengthSum = 0;
+ for (ll=0; ll<=MaxLL; ll++) {
+ U32 const scaleLog = 10; /* scale to 1K */
+ U32 const bitCost = FSE_getMaxNbBits(llstate.symbolTT, ll);
+ assert(bitCost < scaleLog);
+ optPtr->litLengthFreq[ll] = bitCost ? 1 << (scaleLog-bitCost) : 1 /*minimum to calculate cost*/;
+ optPtr->litLengthSum += optPtr->litLengthFreq[ll];
+ } }
+
+ { unsigned ml;
+ FSE_CState_t mlstate;
+ FSE_initCState(&mlstate, optPtr->symbolCosts->fse.matchlengthCTable);
+ optPtr->matchLengthSum = 0;
+ for (ml=0; ml<=MaxML; ml++) {
+ U32 const scaleLog = 10;
+ U32 const bitCost = FSE_getMaxNbBits(mlstate.symbolTT, ml);
+ assert(bitCost < scaleLog);
+ optPtr->matchLengthFreq[ml] = bitCost ? 1 << (scaleLog-bitCost) : 1 /*minimum to calculate cost*/;
+ optPtr->matchLengthSum += optPtr->matchLengthFreq[ml];
+ } }
+
+ { unsigned of;
+ FSE_CState_t ofstate;
+ FSE_initCState(&ofstate, optPtr->symbolCosts->fse.offcodeCTable);
+ optPtr->offCodeSum = 0;
+ for (of=0; of<=MaxOff; of++) {
+ U32 const scaleLog = 10;
+ U32 const bitCost = FSE_getMaxNbBits(ofstate.symbolTT, of);
+ assert(bitCost < scaleLog);
+ optPtr->offCodeFreq[of] = bitCost ? 1 << (scaleLog-bitCost) : 1 /*minimum to calculate cost*/;
+ optPtr->offCodeSum += optPtr->offCodeFreq[of];
+ } }
+
+ } else { /* not a dictionary */
+
+ assert(optPtr->litFreq != NULL);
+ if (compressedLiterals) {
+ unsigned lit = MaxLit;
+ HIST_count_simple(optPtr->litFreq, &lit, src, srcSize); /* use raw first block to init statistics */
+ optPtr->litSum = ZSTD_downscaleStat(optPtr->litFreq, MaxLit, 1);
+ }
+
+ { unsigned ll;
+ for (ll=0; ll<=MaxLL; ll++)
+ optPtr->litLengthFreq[ll] = 1;
+ }
+ optPtr->litLengthSum = MaxLL+1;
+
+ { unsigned ml;
+ for (ml=0; ml<=MaxML; ml++)
+ optPtr->matchLengthFreq[ml] = 1;
+ }
+ optPtr->matchLengthSum = MaxML+1;
+
+ { unsigned of;
+ for (of=0; of<=MaxOff; of++)
+ optPtr->offCodeFreq[of] = 1;
+ }
+ optPtr->offCodeSum = MaxOff+1;
+
}
- optPtr->litLengthSum = 0;
- for (u=0; u<=MaxLL; u++) {
- optPtr->litLengthFreq[u] = 1 + (optPtr->litLengthFreq[u]>>(ZSTD_FREQ_DIV+1));
- optPtr->litLengthSum += optPtr->litLengthFreq[u];
- }
- optPtr->matchLengthSum = 0;
- for (u=0; u<=MaxML; u++) {
- optPtr->matchLengthFreq[u] = 1 + (optPtr->matchLengthFreq[u]>>ZSTD_FREQ_DIV);
- optPtr->matchLengthSum += optPtr->matchLengthFreq[u];
- }
- optPtr->offCodeSum = 0;
- for (u=0; u<=MaxOff; u++) {
- optPtr->offCodeFreq[u] = 1 + (optPtr->offCodeFreq[u]>>ZSTD_FREQ_DIV);
- optPtr->offCodeSum += optPtr->offCodeFreq[u];
- }
+
+ } else { /* new block : re-use previous statistics, scaled down */
+
+ if (compressedLiterals)
+ optPtr->litSum = ZSTD_downscaleStat(optPtr->litFreq, MaxLit, 1);
+ optPtr->litLengthSum = ZSTD_downscaleStat(optPtr->litLengthFreq, MaxLL, 0);
+ optPtr->matchLengthSum = ZSTD_downscaleStat(optPtr->matchLengthFreq, MaxML, 0);
+ optPtr->offCodeSum = ZSTD_downscaleStat(optPtr->offCodeFreq, MaxOff, 0);
}
- ZSTD_setLog2Prices(optPtr);
+ ZSTD_setBasePrices(optPtr, optLevel);
}
-
/* ZSTD_rawLiteralsCost() :
- * cost of literals (only) in given segment (which length can be null)
- * does not include cost of literalLength symbol */
+ * price of literals (only) in specified segment (which length can be 0).
+ * does not include price of literalLength symbol */
static U32 ZSTD_rawLiteralsCost(const BYTE* const literals, U32 const litLength,
- const optState_t* const optPtr)
+ const optState_t* const optPtr,
+ int optLevel)
{
- if (optPtr->staticPrices) return (litLength*6); /* 6 bit per literal - no statistic used */
if (litLength == 0) return 0;
- /* literals */
- { U32 u;
- U32 cost = litLength * optPtr->log2litSum;
- for (u=0; u < litLength; u++)
- cost -= ZSTD_highbit32(optPtr->litFreq[literals[u]]+1);
- return cost;
- }
-}
+ if (!ZSTD_compressedLiterals(optPtr))
+ return (litLength << 3) * BITCOST_MULTIPLIER; /* Uncompressed - 8 bytes per literal. */
-/* ZSTD_litLengthPrice() :
- * cost of literalLength symbol */
-static U32 ZSTD_litLengthPrice(U32 const litLength, const optState_t* const optPtr)
-{
- if (optPtr->staticPrices) return ZSTD_highbit32((U32)litLength+1);
+ if (optPtr->priceType == zop_predef)
+ return (litLength*6) * BITCOST_MULTIPLIER; /* 6 bit per literal - no statistic used */
- /* literal Length */
- { U32 const llCode = ZSTD_LLcode(litLength);
- U32 const price = LL_bits[llCode] + optPtr->log2litLengthSum - ZSTD_highbit32(optPtr->litLengthFreq[llCode]+1);
+ /* dynamic statistics */
+ { U32 price = litLength * optPtr->litSumBasePrice;
+ U32 u;
+ for (u=0; u < litLength; u++) {
+ assert(WEIGHT(optPtr->litFreq[literals[u]], optLevel) <= optPtr->litSumBasePrice); /* literal cost should never be negative */
+ price -= WEIGHT(optPtr->litFreq[literals[u]], optLevel);
+ }
return price;
}
}
/* ZSTD_litLengthPrice() :
- * cost of the literal part of a sequence,
- * including literals themselves, and literalLength symbol */
-static U32 ZSTD_fullLiteralsCost(const BYTE* const literals, U32 const litLength,
- const optState_t* const optPtr)
+ * cost of literalLength symbol */
+static U32 ZSTD_litLengthPrice(U32 const litLength, const optState_t* const optPtr, int optLevel)
{
- return ZSTD_rawLiteralsCost(literals, litLength, optPtr)
- + ZSTD_litLengthPrice(litLength, optPtr);
+ if (optPtr->priceType == zop_predef) return WEIGHT(litLength, optLevel);
+
+ /* dynamic statistics */
+ { U32 const llCode = ZSTD_LLcode(litLength);
+ return (LL_bits[llCode] * BITCOST_MULTIPLIER)
+ + optPtr->litLengthSumBasePrice
+ - WEIGHT(optPtr->litLengthFreq[llCode], optLevel);
+ }
}
/* ZSTD_litLengthContribution() :
* @return ( cost(litlength) - cost(0) )
* this value can then be added to rawLiteralsCost()
* to provide a cost which is directly comparable to a match ending at same position */
-static int ZSTD_litLengthContribution(U32 const litLength, const optState_t* const optPtr)
+static int ZSTD_litLengthContribution(U32 const litLength, const optState_t* const optPtr, int optLevel)
{
- if (optPtr->staticPrices) return ZSTD_highbit32(litLength+1);
+ if (optPtr->priceType >= zop_predef) return (int)WEIGHT(litLength, optLevel);
- /* literal Length */
+ /* dynamic statistics */
{ U32 const llCode = ZSTD_LLcode(litLength);
- int const contribution = LL_bits[llCode]
- + ZSTD_highbit32(optPtr->litLengthFreq[0]+1)
- - ZSTD_highbit32(optPtr->litLengthFreq[llCode]+1);
+ int const contribution = (int)(LL_bits[llCode] * BITCOST_MULTIPLIER)
+ + (int)WEIGHT(optPtr->litLengthFreq[0], optLevel) /* note: log2litLengthSum cancel out */
+ - (int)WEIGHT(optPtr->litLengthFreq[llCode], optLevel);
#if 1
return contribution;
#else
@@ -155,10 +275,11 @@
* which can be compared to the ending cost of a match
* should a new match start at this position */
static int ZSTD_literalsContribution(const BYTE* const literals, U32 const litLength,
- const optState_t* const optPtr)
+ const optState_t* const optPtr,
+ int optLevel)
{
- int const contribution = ZSTD_rawLiteralsCost(literals, litLength, optPtr)
- + ZSTD_litLengthContribution(litLength, optPtr);
+ int const contribution = (int)ZSTD_rawLiteralsCost(literals, litLength, optPtr, optLevel)
+ + ZSTD_litLengthContribution(litLength, optPtr, optLevel);
return contribution;
}
@@ -166,37 +287,45 @@
* Provides the cost of the match part (offset + matchLength) of a sequence
* Must be combined with ZSTD_fullLiteralsCost() to get the full cost of a sequence.
* optLevel: when <2, favors small offset for decompression speed (improved cache efficiency) */
-FORCE_INLINE_TEMPLATE U32 ZSTD_getMatchPrice(
- U32 const offset, U32 const matchLength,
- const optState_t* const optPtr,
- int const optLevel)
+FORCE_INLINE_TEMPLATE U32
+ZSTD_getMatchPrice(U32 const offset,
+ U32 const matchLength,
+ const optState_t* const optPtr,
+ int const optLevel)
{
U32 price;
U32 const offCode = ZSTD_highbit32(offset+1);
U32 const mlBase = matchLength - MINMATCH;
assert(matchLength >= MINMATCH);
- if (optPtr->staticPrices) /* fixed scheme, do not use statistics */
- return ZSTD_highbit32((U32)mlBase+1) + 16 + offCode;
+ if (optPtr->priceType == zop_predef) /* fixed scheme, do not use statistics */
+ return WEIGHT(mlBase, optLevel) + ((16 + offCode) * BITCOST_MULTIPLIER);
- price = offCode + optPtr->log2offCodeSum - ZSTD_highbit32(optPtr->offCodeFreq[offCode]+1);
- if ((optLevel<2) /*static*/ && offCode >= 20) price += (offCode-19)*2; /* handicap for long distance offsets, favor decompression speed */
+ /* dynamic statistics */
+ price = (offCode * BITCOST_MULTIPLIER) + (optPtr->offCodeSumBasePrice - WEIGHT(optPtr->offCodeFreq[offCode], optLevel));
+ if ((optLevel<2) /*static*/ && offCode >= 20)
+ price += (offCode-19)*2 * BITCOST_MULTIPLIER; /* handicap for long distance offsets, favor decompression speed */
/* match Length */
{ U32 const mlCode = ZSTD_MLcode(mlBase);
- price += ML_bits[mlCode] + optPtr->log2matchLengthSum - ZSTD_highbit32(optPtr->matchLengthFreq[mlCode]+1);
+ price += (ML_bits[mlCode] * BITCOST_MULTIPLIER) + (optPtr->matchLengthSumBasePrice - WEIGHT(optPtr->matchLengthFreq[mlCode], optLevel));
}
+ price += BITCOST_MULTIPLIER / 5; /* heuristic : make matches a bit more costly to favor less sequences -> faster decompression speed */
+
DEBUGLOG(8, "ZSTD_getMatchPrice(ml:%u) = %u", matchLength, price);
return price;
}
+/* ZSTD_updateStats() :
+ * assumption : literals + litLengtn <= iend */
static void ZSTD_updateStats(optState_t* const optPtr,
U32 litLength, const BYTE* literals,
U32 offsetCode, U32 matchLength)
{
/* literals */
- { U32 u;
+ if (ZSTD_compressedLiterals(optPtr)) {
+ U32 u;
for (u=0; u < litLength; u++)
optPtr->litFreq[literals[u]] += ZSTD_LITFREQ_ADD;
optPtr->litSum += litLength*ZSTD_LITFREQ_ADD;
@@ -243,13 +372,15 @@
/* Update hashTable3 up to ip (excluded)
Assumption : always within prefix (i.e. not within extDict) */
-static U32 ZSTD_insertAndFindFirstIndexHash3 (ZSTD_matchState_t* ms, const BYTE* const ip)
+static U32 ZSTD_insertAndFindFirstIndexHash3 (ZSTD_matchState_t* ms,
+ U32* nextToUpdate3,
+ const BYTE* const ip)
{
U32* const hashTable3 = ms->hashTable3;
U32 const hashLog3 = ms->hashLog3;
const BYTE* const base = ms->window.base;
- U32 idx = ms->nextToUpdate3;
- U32 const target = ms->nextToUpdate3 = (U32)(ip - base);
+ U32 idx = *nextToUpdate3;
+ U32 const target = (U32)(ip - base);
size_t const hash3 = ZSTD_hash3Ptr(ip, hashLog3);
assert(hashLog3 > 0);
@@ -258,6 +389,7 @@
idx++;
}
+ *nextToUpdate3 = target;
return hashTable3[hash3];
}
@@ -269,10 +401,11 @@
* ip : assumed <= iend-8 .
* @return : nb of positions added */
static U32 ZSTD_insertBt1(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+ ZSTD_matchState_t* ms,
const BYTE* const ip, const BYTE* const iend,
- U32 const mls, U32 const extDict)
+ U32 const mls, const int extDict)
{
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hashLog = cParams->hashLog;
size_t const h = ZSTD_hashPtr(ip, hashLog, mls);
@@ -308,7 +441,8 @@
assert(ip <= iend-8); /* required for h calculation */
hashTable[h] = current; /* Update Hash Table */
- while (nbCompares-- && (matchIndex > windowLow)) {
+ assert(windowLow > 0);
+ while (nbCompares-- && (matchIndex >= windowLow)) {
U32* const nextPtr = bt + 2*(matchIndex & btMask);
size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
assert(matchIndex < current);
@@ -334,8 +468,8 @@
}
#endif
- if ((!extDict) || (matchIndex+matchLength >= dictLimit)) {
- assert(matchIndex+matchLength >= dictLimit); /* might be wrong if extDict is incorrectly set to 0 */
+ if (!extDict || (matchIndex+matchLength >= dictLimit)) {
+ assert(matchIndex+matchLength >= dictLimit); /* might be wrong if actually extDict */
match = base + matchIndex;
matchLength += ZSTD_count(ip+matchLength, match+matchLength, iend);
} else {
@@ -372,43 +506,53 @@
} }
*smallerPtr = *largerPtr = 0;
- if (bestLength > 384) return MIN(192, (U32)(bestLength - 384)); /* speed optimization */
- assert(matchEndIdx > current + 8);
- return matchEndIdx - (current + 8);
+ { U32 positions = 0;
+ if (bestLength > 384) positions = MIN(192, (U32)(bestLength - 384)); /* speed optimization */
+ assert(matchEndIdx > current + 8);
+ return MAX(positions, matchEndIdx - (current + 8));
+ }
}
FORCE_INLINE_TEMPLATE
void ZSTD_updateTree_internal(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
+ ZSTD_matchState_t* ms,
const BYTE* const ip, const BYTE* const iend,
- const U32 mls, const U32 extDict)
+ const U32 mls, const ZSTD_dictMode_e dictMode)
{
const BYTE* const base = ms->window.base;
U32 const target = (U32)(ip - base);
U32 idx = ms->nextToUpdate;
- DEBUGLOG(7, "ZSTD_updateTree_internal, from %u to %u (extDict:%u)",
- idx, target, extDict);
+ DEBUGLOG(6, "ZSTD_updateTree_internal, from %u to %u (dictMode:%u)",
+ idx, target, dictMode);
- while(idx < target)
- idx += ZSTD_insertBt1(ms, cParams, base+idx, iend, mls, extDict);
+ while(idx < target) {
+ U32 const forward = ZSTD_insertBt1(ms, base+idx, iend, mls, dictMode == ZSTD_extDict);
+ assert(idx < (U32)(idx + forward));
+ idx += forward;
+ }
+ assert((size_t)(ip - base) <= (size_t)(U32)(-1));
+ assert((size_t)(iend - base) <= (size_t)(U32)(-1));
ms->nextToUpdate = target;
}
-void ZSTD_updateTree(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
- const BYTE* ip, const BYTE* iend)
-{
- ZSTD_updateTree_internal(ms, cParams, ip, iend, cParams->searchLength, 0 /*extDict*/);
+void ZSTD_updateTree(ZSTD_matchState_t* ms, const BYTE* ip, const BYTE* iend) {
+ ZSTD_updateTree_internal(ms, ip, iend, ms->cParams.minMatch, ZSTD_noDict);
}
FORCE_INLINE_TEMPLATE
U32 ZSTD_insertBtAndGetAllMatches (
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
- const BYTE* const ip, const BYTE* const iLimit, int const extDict,
- U32 rep[ZSTD_REP_NUM], U32 const ll0,
- ZSTD_match_t* matches, const U32 lengthToBeat, U32 const mls /* template */)
+ ZSTD_match_t* matches, /* store result (found matches) in this table (presumed large enough) */
+ ZSTD_matchState_t* ms,
+ U32* nextToUpdate3,
+ const BYTE* const ip, const BYTE* const iLimit, const ZSTD_dictMode_e dictMode,
+ const U32 rep[ZSTD_REP_NUM],
+ U32 const ll0, /* tells if associated literal length is 0 or not. This value must be 0 or 1 */
+ const U32 lengthToBeat,
+ U32 const mls /* template */)
{
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32 const sufficient_len = MIN(cParams->targetLength, ZSTD_OPT_NUM -1);
+ U32 const maxDistance = 1U << cParams->windowLog;
const BYTE* const base = ms->window.base;
U32 const current = (U32)(ip-base);
U32 const hashLog = cParams->hashLog;
@@ -424,8 +568,10 @@
U32 const dictLimit = ms->window.dictLimit;
const BYTE* const dictEnd = dictBase + dictLimit;
const BYTE* const prefixStart = base + dictLimit;
- U32 const btLow = btMask >= current ? 0 : current - btMask;
- U32 const windowLow = ms->window.lowLimit;
+ U32 const btLow = (btMask >= current) ? 0 : current - btMask;
+ U32 const windowValid = ms->window.lowLimit;
+ U32 const windowLow = ((current - windowValid) > maxDistance) ? current - maxDistance : windowValid;
+ U32 const matchLow = windowLow ? windowLow : 1;
U32* smallerPtr = bt + 2*(current&btMask);
U32* largerPtr = bt + 2*(current&btMask) + 1;
U32 matchEndIdx = current+8+1; /* farthest referenced position of any match => detects repetitive patterns */
@@ -433,10 +579,24 @@
U32 mnum = 0;
U32 nbCompares = 1U << cParams->searchLog;
+ const ZSTD_matchState_t* dms = dictMode == ZSTD_dictMatchState ? ms->dictMatchState : NULL;
+ const ZSTD_compressionParameters* const dmsCParams =
+ dictMode == ZSTD_dictMatchState ? &dms->cParams : NULL;
+ const BYTE* const dmsBase = dictMode == ZSTD_dictMatchState ? dms->window.base : NULL;
+ const BYTE* const dmsEnd = dictMode == ZSTD_dictMatchState ? dms->window.nextSrc : NULL;
+ U32 const dmsHighLimit = dictMode == ZSTD_dictMatchState ? (U32)(dmsEnd - dmsBase) : 0;
+ U32 const dmsLowLimit = dictMode == ZSTD_dictMatchState ? dms->window.lowLimit : 0;
+ U32 const dmsIndexDelta = dictMode == ZSTD_dictMatchState ? windowLow - dmsHighLimit : 0;
+ U32 const dmsHashLog = dictMode == ZSTD_dictMatchState ? dmsCParams->hashLog : hashLog;
+ U32 const dmsBtLog = dictMode == ZSTD_dictMatchState ? dmsCParams->chainLog - 1 : btLog;
+ U32 const dmsBtMask = dictMode == ZSTD_dictMatchState ? (1U << dmsBtLog) - 1 : 0;
+ U32 const dmsBtLow = dictMode == ZSTD_dictMatchState && dmsBtMask < dmsHighLimit - dmsLowLimit ? dmsHighLimit - dmsBtMask : dmsLowLimit;
+
size_t bestLength = lengthToBeat-1;
- DEBUGLOG(7, "ZSTD_insertBtAndGetAllMatches");
+ DEBUGLOG(8, "ZSTD_insertBtAndGetAllMatches: current=%u", current);
/* check repCode */
+ assert(ll0 <= 1); /* necessarily 1 or 0 */
{ U32 const lastR = ZSTD_REP_NUM + ll0;
U32 repCode;
for (repCode = ll0; repCode < lastR; repCode++) {
@@ -449,18 +609,26 @@
repLen = (U32)ZSTD_count(ip+minMatch, ip+minMatch-repOffset, iLimit) + minMatch;
}
} else { /* repIndex < dictLimit || repIndex >= current */
- const BYTE* const repMatch = dictBase + repIndex;
+ const BYTE* const repMatch = dictMode == ZSTD_dictMatchState ?
+ dmsBase + repIndex - dmsIndexDelta :
+ dictBase + repIndex;
assert(current >= windowLow);
- if ( extDict /* this case only valid in extDict mode */
+ if ( dictMode == ZSTD_extDict
&& ( ((repOffset-1) /*intentional overflow*/ < current - windowLow) /* equivalent to `current > repIndex >= windowLow` */
& (((U32)((dictLimit-1) - repIndex) >= 3) ) /* intentional overflow : do not test positions overlapping 2 memory segments */)
&& (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(repMatch, minMatch)) ) {
repLen = (U32)ZSTD_count_2segments(ip+minMatch, repMatch+minMatch, iLimit, dictEnd, prefixStart) + minMatch;
+ }
+ if (dictMode == ZSTD_dictMatchState
+ && ( ((repOffset-1) /*intentional overflow*/ < current - (dmsLowLimit + dmsIndexDelta)) /* equivalent to `current > repIndex >= dmsLowLimit` */
+ & ((U32)((dictLimit-1) - repIndex) >= 3) ) /* intentional overflow : do not test positions overlapping 2 memory segments */
+ && (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(repMatch, minMatch)) ) {
+ repLen = (U32)ZSTD_count_2segments(ip+minMatch, repMatch+minMatch, iLimit, dmsEnd, prefixStart) + minMatch;
} }
/* save longer solution */
if (repLen > bestLength) {
- DEBUGLOG(8, "found rep-match %u of length %u",
- repCode - ll0, (U32)repLen);
+ DEBUGLOG(8, "found repCode %u (ll0:%u, offset:%u) of length %u",
+ repCode, ll0, repOffset, repLen);
bestLength = repLen;
matches[mnum].off = repCode - ll0;
matches[mnum].len = (U32)repLen;
@@ -472,11 +640,11 @@
/* HC3 match finder */
if ((mls == 3) /*static*/ && (bestLength < mls)) {
- U32 const matchIndex3 = ZSTD_insertAndFindFirstIndexHash3(ms, ip);
- if ((matchIndex3 > windowLow)
+ U32 const matchIndex3 = ZSTD_insertAndFindFirstIndexHash3(ms, nextToUpdate3, ip);
+ if ((matchIndex3 >= matchLow)
& (current - matchIndex3 < (1<<18)) /*heuristic : longer distance likely too expensive*/ ) {
size_t mlen;
- if ((!extDict) /*static*/ || (matchIndex3 >= dictLimit)) {
+ if ((dictMode == ZSTD_noDict) /*static*/ || (dictMode == ZSTD_dictMatchState) /*static*/ || (matchIndex3 >= dictLimit)) {
const BYTE* const match = base + matchIndex3;
mlen = ZSTD_count(ip, match, iLimit);
} else {
@@ -498,17 +666,19 @@
(ip+mlen == iLimit) ) { /* best possible length */
ms->nextToUpdate = current+1; /* skip insertion */
return 1;
- } } } }
+ } } }
+ /* no dictMatchState lookup: dicts don't have a populated HC3 table */
+ }
hashTable[h] = current; /* Update Hash Table */
- while (nbCompares-- && (matchIndex > windowLow)) {
+ while (nbCompares-- && (matchIndex >= matchLow)) {
U32* const nextPtr = bt + 2*(matchIndex & btMask);
size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
const BYTE* match;
assert(current > matchIndex);
- if ((!extDict) || (matchIndex+matchLength >= dictLimit)) {
+ if ((dictMode == ZSTD_noDict) || (dictMode == ZSTD_dictMatchState) || (matchIndex+matchLength >= dictLimit)) {
assert(matchIndex+matchLength >= dictLimit); /* ensure the condition is correct when !extDict */
match = base + matchIndex;
matchLength += ZSTD_count(ip+matchLength, match+matchLength, iLimit);
@@ -520,8 +690,8 @@
}
if (matchLength > bestLength) {
- DEBUGLOG(8, "found match of length %u at distance %u",
- (U32)matchLength, current - matchIndex);
+ DEBUGLOG(8, "found match of length %u at distance %u (offCode=%u)",
+ (U32)matchLength, current - matchIndex, current - matchIndex + ZSTD_REP_MOVE);
assert(matchEndIdx > matchIndex);
if (matchLength > matchEndIdx - matchIndex)
matchEndIdx = matchIndex + (U32)matchLength;
@@ -529,9 +699,10 @@
matches[mnum].off = (current - matchIndex) + ZSTD_REP_MOVE;
matches[mnum].len = (U32)matchLength;
mnum++;
- if (matchLength > ZSTD_OPT_NUM) break;
- if (ip+matchLength == iLimit) { /* equal : no way to know if inf or sup */
- break; /* drop, to preserve bt consistency (miss a little bit of compression) */
+ if ( (matchLength > ZSTD_OPT_NUM)
+ | (ip+matchLength == iLimit) /* equal : no way to know if inf or sup */) {
+ if (dictMode == ZSTD_dictMatchState) nbCompares = 0; /* break should also skip searching dms */
+ break; /* drop, to preserve bt consistency (miss a little bit of compression) */
}
}
@@ -552,6 +723,47 @@
*smallerPtr = *largerPtr = 0;
+ if (dictMode == ZSTD_dictMatchState && nbCompares) {
+ size_t const dmsH = ZSTD_hashPtr(ip, dmsHashLog, mls);
+ U32 dictMatchIndex = dms->hashTable[dmsH];
+ const U32* const dmsBt = dms->chainTable;
+ commonLengthSmaller = commonLengthLarger = 0;
+ while (nbCompares-- && (dictMatchIndex > dmsLowLimit)) {
+ const U32* const nextPtr = dmsBt + 2*(dictMatchIndex & dmsBtMask);
+ size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
+ const BYTE* match = dmsBase + dictMatchIndex;
+ matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iLimit, dmsEnd, prefixStart);
+ if (dictMatchIndex+matchLength >= dmsHighLimit)
+ match = base + dictMatchIndex + dmsIndexDelta; /* to prepare for next usage of match[matchLength] */
+
+ if (matchLength > bestLength) {
+ matchIndex = dictMatchIndex + dmsIndexDelta;
+ DEBUGLOG(8, "found dms match of length %u at distance %u (offCode=%u)",
+ (U32)matchLength, current - matchIndex, current - matchIndex + ZSTD_REP_MOVE);
+ if (matchLength > matchEndIdx - matchIndex)
+ matchEndIdx = matchIndex + (U32)matchLength;
+ bestLength = matchLength;
+ matches[mnum].off = (current - matchIndex) + ZSTD_REP_MOVE;
+ matches[mnum].len = (U32)matchLength;
+ mnum++;
+ if ( (matchLength > ZSTD_OPT_NUM)
+ | (ip+matchLength == iLimit) /* equal : no way to know if inf or sup */) {
+ break; /* drop, to guarantee consistency (miss a little bit of compression) */
+ }
+ }
+
+ if (dictMatchIndex <= dmsBtLow) { break; } /* beyond tree size, stop the search */
+ if (match[matchLength] < ip[matchLength]) {
+ commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */
+ dictMatchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */
+ } else {
+ /* match is larger than current */
+ commonLengthLarger = matchLength;
+ dictMatchIndex = nextPtr[0];
+ }
+ }
+ }
+
assert(matchEndIdx > current+8);
ms->nextToUpdate = matchEndIdx - 8; /* skip repetitive patterns */
return mnum;
@@ -559,23 +771,27 @@
FORCE_INLINE_TEMPLATE U32 ZSTD_BtGetAllMatches (
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
- const BYTE* ip, const BYTE* const iHighLimit, int const extDict,
- U32 rep[ZSTD_REP_NUM], U32 const ll0,
- ZSTD_match_t* matches, U32 const lengthToBeat)
+ ZSTD_match_t* matches, /* store result (match found, increasing size) in this table */
+ ZSTD_matchState_t* ms,
+ U32* nextToUpdate3,
+ const BYTE* ip, const BYTE* const iHighLimit, const ZSTD_dictMode_e dictMode,
+ const U32 rep[ZSTD_REP_NUM],
+ U32 const ll0,
+ U32 const lengthToBeat)
{
- U32 const matchLengthSearch = cParams->searchLength;
- DEBUGLOG(7, "ZSTD_BtGetAllMatches");
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
+ U32 const matchLengthSearch = cParams->minMatch;
+ DEBUGLOG(8, "ZSTD_BtGetAllMatches");
if (ip < ms->window.base + ms->nextToUpdate) return 0; /* skipped area */
- ZSTD_updateTree_internal(ms, cParams, ip, iHighLimit, matchLengthSearch, extDict);
+ ZSTD_updateTree_internal(ms, ip, iHighLimit, matchLengthSearch, dictMode);
switch(matchLengthSearch)
{
- case 3 : return ZSTD_insertBtAndGetAllMatches(ms, cParams, ip, iHighLimit, extDict, rep, ll0, matches, lengthToBeat, 3);
+ case 3 : return ZSTD_insertBtAndGetAllMatches(matches, ms, nextToUpdate3, ip, iHighLimit, dictMode, rep, ll0, lengthToBeat, 3);
default :
- case 4 : return ZSTD_insertBtAndGetAllMatches(ms, cParams, ip, iHighLimit, extDict, rep, ll0, matches, lengthToBeat, 4);
- case 5 : return ZSTD_insertBtAndGetAllMatches(ms, cParams, ip, iHighLimit, extDict, rep, ll0, matches, lengthToBeat, 5);
+ case 4 : return ZSTD_insertBtAndGetAllMatches(matches, ms, nextToUpdate3, ip, iHighLimit, dictMode, rep, ll0, lengthToBeat, 4);
+ case 5 : return ZSTD_insertBtAndGetAllMatches(matches, ms, nextToUpdate3, ip, iHighLimit, dictMode, rep, ll0, lengthToBeat, 5);
case 7 :
- case 6 : return ZSTD_insertBtAndGetAllMatches(ms, cParams, ip, iHighLimit, extDict, rep, ll0, matches, lengthToBeat, 6);
+ case 6 : return ZSTD_insertBtAndGetAllMatches(matches, ms, nextToUpdate3, ip, iHighLimit, dictMode, rep, ll0, lengthToBeat, 6);
}
}
@@ -587,7 +803,7 @@
U32 rep[3];
} repcodes_t;
-repcodes_t ZSTD_updateRep(U32 const rep[3], U32 const offset, U32 const ll0)
+static repcodes_t ZSTD_updateRep(U32 const rep[3], U32 const offset, U32 const ll0)
{
repcodes_t newReps;
if (offset >= ZSTD_REP_NUM) { /* full offset */
@@ -609,65 +825,35 @@
}
-typedef struct {
- const BYTE* anchor;
- U32 litlen;
- U32 rawLitCost;
-} cachedLiteralPrice_t;
-
-static U32 ZSTD_rawLiteralsCost_cached(
- cachedLiteralPrice_t* const cachedLitPrice,
- const BYTE* const anchor, U32 const litlen,
- const optState_t* const optStatePtr)
+static U32 ZSTD_totalLen(ZSTD_optimal_t sol)
{
- U32 startCost;
- U32 remainingLength;
- const BYTE* startPosition;
+ return sol.litlen + sol.mlen;
+}
- if (anchor == cachedLitPrice->anchor) {
- startCost = cachedLitPrice->rawLitCost;
- startPosition = anchor + cachedLitPrice->litlen;
- assert(litlen >= cachedLitPrice->litlen);
- remainingLength = litlen - cachedLitPrice->litlen;
- } else {
- startCost = 0;
- startPosition = anchor;
- remainingLength = litlen;
+#if 0 /* debug */
+
+static void
+listStats(const U32* table, int lastEltID)
+{
+ int const nbElts = lastEltID + 1;
+ int enb;
+ for (enb=0; enb < nbElts; enb++) {
+ (void)table;
+ //RAWLOG(2, "%3i:%3i, ", enb, table[enb]);
+ RAWLOG(2, "%4i,", table[enb]);
}
-
- { U32 const rawLitCost = startCost + ZSTD_rawLiteralsCost(startPosition, remainingLength, optStatePtr);
- cachedLitPrice->anchor = anchor;
- cachedLitPrice->litlen = litlen;
- cachedLitPrice->rawLitCost = rawLitCost;
- return rawLitCost;
- }
+ RAWLOG(2, " \n");
}
-static U32 ZSTD_fullLiteralsCost_cached(
- cachedLiteralPrice_t* const cachedLitPrice,
- const BYTE* const anchor, U32 const litlen,
- const optState_t* const optStatePtr)
-{
- return ZSTD_rawLiteralsCost_cached(cachedLitPrice, anchor, litlen, optStatePtr)
- + ZSTD_litLengthPrice(litlen, optStatePtr);
-}
+#endif
-static int ZSTD_literalsContribution_cached(
- cachedLiteralPrice_t* const cachedLitPrice,
- const BYTE* const anchor, U32 const litlen,
- const optState_t* const optStatePtr)
-{
- int const contribution = ZSTD_rawLiteralsCost_cached(cachedLitPrice, anchor, litlen, optStatePtr)
- + ZSTD_litLengthContribution(litlen, optStatePtr);
- return contribution;
-}
-
-FORCE_INLINE_TEMPLATE
-size_t ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,seqStore_t* seqStore,
- U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams,
- const void* src, size_t srcSize,
- const int optLevel, const int extDict)
+FORCE_INLINE_TEMPLATE size_t
+ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,
+ seqStore_t* seqStore,
+ U32 rep[ZSTD_REP_NUM],
+ const void* src, size_t srcSize,
+ const int optLevel,
+ const ZSTD_dictMode_e dictMode)
{
optState_t* const optStatePtr = &ms->opt;
const BYTE* const istart = (const BYTE*)src;
@@ -677,72 +863,77 @@
const BYTE* const ilimit = iend - 8;
const BYTE* const base = ms->window.base;
const BYTE* const prefixStart = base + ms->window.dictLimit;
+ const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32 const sufficient_len = MIN(cParams->targetLength, ZSTD_OPT_NUM -1);
- U32 const minMatch = (cParams->searchLength == 3) ? 3 : 4;
+ U32 const minMatch = (cParams->minMatch == 3) ? 3 : 4;
+ U32 nextToUpdate3 = ms->nextToUpdate;
ZSTD_optimal_t* const opt = optStatePtr->priceTable;
ZSTD_match_t* const matches = optStatePtr->matchTable;
- cachedLiteralPrice_t cachedLitPrice;
+ ZSTD_optimal_t lastSequence;
/* init */
- DEBUGLOG(5, "ZSTD_compressBlock_opt_generic");
- ms->nextToUpdate3 = ms->nextToUpdate;
- ZSTD_rescaleFreqs(optStatePtr, (const BYTE*)src, srcSize);
+ DEBUGLOG(5, "ZSTD_compressBlock_opt_generic: current=%u, prefix=%u, nextToUpdate=%u",
+ (U32)(ip - base), ms->window.dictLimit, ms->nextToUpdate);
+ assert(optLevel <= 2);
+ ZSTD_rescaleFreqs(optStatePtr, (const BYTE*)src, srcSize, optLevel);
ip += (ip==prefixStart);
- memset(&cachedLitPrice, 0, sizeof(cachedLitPrice));
/* Match Loop */
while (ip < ilimit) {
U32 cur, last_pos = 0;
- U32 best_mlen, best_off;
/* find first match */
{ U32 const litlen = (U32)(ip - anchor);
U32 const ll0 = !litlen;
- U32 const nbMatches = ZSTD_BtGetAllMatches(ms, cParams, ip, iend, extDict, rep, ll0, matches, minMatch);
+ U32 const nbMatches = ZSTD_BtGetAllMatches(matches, ms, &nextToUpdate3, ip, iend, dictMode, rep, ll0, minMatch);
if (!nbMatches) { ip++; continue; }
/* initialize opt[0] */
{ U32 i ; for (i=0; i<ZSTD_REP_NUM; i++) opt[0].rep[i] = rep[i]; }
- opt[0].mlen = 1;
+ opt[0].mlen = 0; /* means is_a_literal */
opt[0].litlen = litlen;
+ opt[0].price = ZSTD_literalsContribution(anchor, litlen, optStatePtr, optLevel);
/* large match -> immediate encoding */
{ U32 const maxML = matches[nbMatches-1].len;
- DEBUGLOG(7, "found %u matches of maxLength=%u and offset=%u at cPos=%u => start new serie",
- nbMatches, maxML, matches[nbMatches-1].off, (U32)(ip-prefixStart));
+ U32 const maxOffset = matches[nbMatches-1].off;
+ DEBUGLOG(6, "found %u matches of maxLength=%u and maxOffCode=%u at cPos=%u => start new series",
+ nbMatches, maxML, maxOffset, (U32)(ip-prefixStart));
if (maxML > sufficient_len) {
- best_mlen = maxML;
- best_off = matches[nbMatches-1].off;
- DEBUGLOG(7, "large match (%u>%u), immediate encoding",
- best_mlen, sufficient_len);
+ lastSequence.litlen = litlen;
+ lastSequence.mlen = maxML;
+ lastSequence.off = maxOffset;
+ DEBUGLOG(6, "large match (%u>%u), immediate encoding",
+ maxML, sufficient_len);
cur = 0;
- last_pos = 1;
+ last_pos = ZSTD_totalLen(lastSequence);
goto _shortestPath;
} }
/* set prices for first matches starting position == 0 */
- { U32 const literalsPrice = ZSTD_fullLiteralsCost_cached(&cachedLitPrice, anchor, litlen, optStatePtr);
+ { U32 const literalsPrice = opt[0].price + ZSTD_litLengthPrice(0, optStatePtr, optLevel);
U32 pos;
U32 matchNb;
- for (pos = 0; pos < minMatch; pos++) {
- opt[pos].mlen = 1;
- opt[pos].price = ZSTD_MAX_PRICE;
+ for (pos = 1; pos < minMatch; pos++) {
+ opt[pos].price = ZSTD_MAX_PRICE; /* mlen, litlen and price will be fixed during forward scanning */
}
for (matchNb = 0; matchNb < nbMatches; matchNb++) {
U32 const offset = matches[matchNb].off;
U32 const end = matches[matchNb].len;
repcodes_t const repHistory = ZSTD_updateRep(rep, offset, ll0);
for ( ; pos <= end ; pos++ ) {
- U32 const matchPrice = literalsPrice + ZSTD_getMatchPrice(offset, pos, optStatePtr, optLevel);
- DEBUGLOG(7, "rPos:%u => set initial price : %u",
- pos, matchPrice);
+ U32 const matchPrice = ZSTD_getMatchPrice(offset, pos, optStatePtr, optLevel);
+ U32 const sequencePrice = literalsPrice + matchPrice;
+ DEBUGLOG(7, "rPos:%u => set initial price : %.2f",
+ pos, ZSTD_fCost(sequencePrice));
opt[pos].mlen = pos;
opt[pos].off = offset;
opt[pos].litlen = litlen;
- opt[pos].price = matchPrice;
+ opt[pos].price = sequencePrice;
+ ZSTD_STATIC_ASSERT(sizeof(opt[pos].rep) == sizeof(repHistory));
memcpy(opt[pos].rep, &repHistory, sizeof(repHistory));
} }
last_pos = pos-1;
@@ -753,55 +944,67 @@
for (cur = 1; cur <= last_pos; cur++) {
const BYTE* const inr = ip + cur;
assert(cur < ZSTD_OPT_NUM);
+ DEBUGLOG(7, "cPos:%zi==rPos:%u", inr-istart, cur)
/* Fix current position with one literal if cheaper */
- { U32 const litlen = (opt[cur-1].mlen == 1) ? opt[cur-1].litlen + 1 : 1;
- int price; /* note : contribution can be negative */
- if (cur > litlen) {
- price = opt[cur - litlen].price + ZSTD_literalsContribution(inr-litlen, litlen, optStatePtr);
- } else {
- price = ZSTD_literalsContribution_cached(&cachedLitPrice, anchor, litlen, optStatePtr);
- }
+ { U32 const litlen = (opt[cur-1].mlen == 0) ? opt[cur-1].litlen + 1 : 1;
+ int const price = opt[cur-1].price
+ + ZSTD_rawLiteralsCost(ip+cur-1, 1, optStatePtr, optLevel)
+ + ZSTD_litLengthPrice(litlen, optStatePtr, optLevel)
+ - ZSTD_litLengthPrice(litlen-1, optStatePtr, optLevel);
assert(price < 1000000000); /* overflow check */
if (price <= opt[cur].price) {
- DEBUGLOG(7, "rPos:%u : better price (%u<%u) using literal",
- cur, price, opt[cur].price);
- opt[cur].mlen = 1;
+ DEBUGLOG(7, "cPos:%zi==rPos:%u : better price (%.2f<=%.2f) using literal (ll==%u) (hist:%u,%u,%u)",
+ inr-istart, cur, ZSTD_fCost(price), ZSTD_fCost(opt[cur].price), litlen,
+ opt[cur-1].rep[0], opt[cur-1].rep[1], opt[cur-1].rep[2]);
+ opt[cur].mlen = 0;
opt[cur].off = 0;
opt[cur].litlen = litlen;
opt[cur].price = price;
memcpy(opt[cur].rep, opt[cur-1].rep, sizeof(opt[cur].rep));
- } }
+ } else {
+ DEBUGLOG(7, "cPos:%zi==rPos:%u : literal would cost more (%.2f>%.2f) (hist:%u,%u,%u)",
+ inr-istart, cur, ZSTD_fCost(price), ZSTD_fCost(opt[cur].price),
+ opt[cur].rep[0], opt[cur].rep[1], opt[cur].rep[2]);
+ }
+ }
/* last match must start at a minimum distance of 8 from oend */
if (inr > ilimit) continue;
if (cur == last_pos) break;
- if ( (optLevel==0) /*static*/
- && (opt[cur+1].price <= opt[cur].price) )
+ if ( (optLevel==0) /*static_test*/
+ && (opt[cur+1].price <= opt[cur].price + (BITCOST_MULTIPLIER/2)) ) {
+ DEBUGLOG(7, "move to next rPos:%u : price is <=", cur+1);
continue; /* skip unpromising positions; about ~+6% speed, -0.01 ratio */
+ }
- { U32 const ll0 = (opt[cur].mlen != 1);
- U32 const litlen = (opt[cur].mlen == 1) ? opt[cur].litlen : 0;
- U32 const previousPrice = (cur > litlen) ? opt[cur-litlen].price : 0;
- U32 const basePrice = previousPrice + ZSTD_fullLiteralsCost(inr-litlen, litlen, optStatePtr);
- U32 const nbMatches = ZSTD_BtGetAllMatches(ms, cParams, inr, iend, extDict, opt[cur].rep, ll0, matches, minMatch);
+ { U32 const ll0 = (opt[cur].mlen != 0);
+ U32 const litlen = (opt[cur].mlen == 0) ? opt[cur].litlen : 0;
+ U32 const previousPrice = opt[cur].price;
+ U32 const basePrice = previousPrice + ZSTD_litLengthPrice(0, optStatePtr, optLevel);
+ U32 const nbMatches = ZSTD_BtGetAllMatches(matches, ms, &nextToUpdate3, inr, iend, dictMode, opt[cur].rep, ll0, minMatch);
U32 matchNb;
- if (!nbMatches) continue;
+ if (!nbMatches) {
+ DEBUGLOG(7, "rPos:%u : no match found", cur);
+ continue;
+ }
{ U32 const maxML = matches[nbMatches-1].len;
- DEBUGLOG(7, "rPos:%u, found %u matches, of maxLength=%u",
- cur, nbMatches, maxML);
+ DEBUGLOG(7, "cPos:%zi==rPos:%u, found %u matches, of maxLength=%u",
+ inr-istart, cur, nbMatches, maxML);
if ( (maxML > sufficient_len)
- | (cur + maxML >= ZSTD_OPT_NUM) ) {
- best_mlen = maxML;
- best_off = matches[nbMatches-1].off;
- last_pos = cur + 1;
+ || (cur + maxML >= ZSTD_OPT_NUM) ) {
+ lastSequence.mlen = maxML;
+ lastSequence.off = matches[nbMatches-1].off;
+ lastSequence.litlen = litlen;
+ cur -= (opt[cur].mlen==0) ? opt[cur].litlen : 0; /* last sequence is actually only literals, fix cur to last match - note : may underflow, in which case, it's first sequence, and it's okay */
+ last_pos = cur + ZSTD_totalLen(lastSequence);
+ if (cur > ZSTD_OPT_NUM) cur = 0; /* underflow => first match */
goto _shortestPath;
- }
- }
+ } }
/* set prices using matches found at position == cur */
for (matchNb = 0; matchNb < nbMatches; matchNb++) {
@@ -811,113 +1014,233 @@
U32 const startML = (matchNb>0) ? matches[matchNb-1].len+1 : minMatch;
U32 mlen;
- DEBUGLOG(7, "testing match %u => offCode=%u, mlen=%u, llen=%u",
+ DEBUGLOG(7, "testing match %u => offCode=%4u, mlen=%2u, llen=%2u",
matchNb, matches[matchNb].off, lastML, litlen);
- for (mlen = lastML; mlen >= startML; mlen--) {
+ for (mlen = lastML; mlen >= startML; mlen--) { /* scan downward */
U32 const pos = cur + mlen;
int const price = basePrice + ZSTD_getMatchPrice(offset, mlen, optStatePtr, optLevel);
if ((pos > last_pos) || (price < opt[pos].price)) {
- DEBUGLOG(7, "rPos:%u => new better price (%u<%u)",
- pos, price, opt[pos].price);
- while (last_pos < pos) { opt[last_pos+1].price = ZSTD_MAX_PRICE; last_pos++; }
+ DEBUGLOG(7, "rPos:%u (ml=%2u) => new better price (%.2f<%.2f)",
+ pos, mlen, ZSTD_fCost(price), ZSTD_fCost(opt[pos].price));
+ while (last_pos < pos) { opt[last_pos+1].price = ZSTD_MAX_PRICE; last_pos++; } /* fill empty positions */
opt[pos].mlen = mlen;
opt[pos].off = offset;
opt[pos].litlen = litlen;
opt[pos].price = price;
+ ZSTD_STATIC_ASSERT(sizeof(opt[pos].rep) == sizeof(repHistory));
memcpy(opt[pos].rep, &repHistory, sizeof(repHistory));
} else {
- if (optLevel==0) break; /* gets ~+10% speed for about -0.01 ratio loss */
+ DEBUGLOG(7, "rPos:%u (ml=%2u) => new price is worse (%.2f>=%.2f)",
+ pos, mlen, ZSTD_fCost(price), ZSTD_fCost(opt[pos].price));
+ if (optLevel==0) break; /* early update abort; gets ~+10% speed for about -0.01 ratio loss */
}
} } }
} /* for (cur = 1; cur <= last_pos; cur++) */
- best_mlen = opt[last_pos].mlen;
- best_off = opt[last_pos].off;
- cur = last_pos - best_mlen;
+ lastSequence = opt[last_pos];
+ cur = last_pos > ZSTD_totalLen(lastSequence) ? last_pos - ZSTD_totalLen(lastSequence) : 0; /* single sequence, and it starts before `ip` */
+ assert(cur < ZSTD_OPT_NUM); /* control overflow*/
_shortestPath: /* cur, last_pos, best_mlen, best_off have to be set */
- assert(opt[0].mlen == 1);
+ assert(opt[0].mlen == 0);
- /* reverse traversal */
- DEBUGLOG(7, "start reverse traversal (last_pos:%u, cur:%u)",
- last_pos, cur);
- { U32 selectedMatchLength = best_mlen;
- U32 selectedOffset = best_off;
- U32 pos = cur;
- while (1) {
- U32 const mlen = opt[pos].mlen;
- U32 const off = opt[pos].off;
- opt[pos].mlen = selectedMatchLength;
- opt[pos].off = selectedOffset;
- selectedMatchLength = mlen;
- selectedOffset = off;
- if (mlen > pos) break;
- pos -= mlen;
- } }
+ { U32 const storeEnd = cur + 1;
+ U32 storeStart = storeEnd;
+ U32 seqPos = cur;
- /* save sequences */
- { U32 pos;
- for (pos=0; pos < last_pos; ) {
- U32 const llen = (U32)(ip - anchor);
- U32 const mlen = opt[pos].mlen;
- U32 const offset = opt[pos].off;
- if (mlen == 1) { ip++; pos++; continue; } /* literal position => move on */
- pos += mlen; ip += mlen;
+ DEBUGLOG(6, "start reverse traversal (last_pos:%u, cur:%u)",
+ last_pos, cur); (void)last_pos;
+ assert(storeEnd < ZSTD_OPT_NUM);
+ DEBUGLOG(6, "last sequence copied into pos=%u (llen=%u,mlen=%u,ofc=%u)",
+ storeEnd, lastSequence.litlen, lastSequence.mlen, lastSequence.off);
+ opt[storeEnd] = lastSequence;
+ while (seqPos > 0) {
+ U32 const backDist = ZSTD_totalLen(opt[seqPos]);
+ storeStart--;
+ DEBUGLOG(6, "sequence from rPos=%u copied into pos=%u (llen=%u,mlen=%u,ofc=%u)",
+ seqPos, storeStart, opt[seqPos].litlen, opt[seqPos].mlen, opt[seqPos].off);
+ opt[storeStart] = opt[seqPos];
+ seqPos = (seqPos > backDist) ? seqPos - backDist : 0;
+ }
- /* repcodes update : like ZSTD_updateRep(), but update in place */
- if (offset >= ZSTD_REP_NUM) { /* full offset */
- rep[2] = rep[1];
- rep[1] = rep[0];
- rep[0] = offset - ZSTD_REP_MOVE;
- } else { /* repcode */
- U32 const repCode = offset + (llen==0);
- if (repCode) { /* note : if repCode==0, no change */
- U32 const currentOffset = (repCode==ZSTD_REP_NUM) ? (rep[0] - 1) : rep[repCode];
- if (repCode >= 2) rep[2] = rep[1];
- rep[1] = rep[0];
- rep[0] = currentOffset;
+ /* save sequences */
+ DEBUGLOG(6, "sending selected sequences into seqStore")
+ { U32 storePos;
+ for (storePos=storeStart; storePos <= storeEnd; storePos++) {
+ U32 const llen = opt[storePos].litlen;
+ U32 const mlen = opt[storePos].mlen;
+ U32 const offCode = opt[storePos].off;
+ U32 const advance = llen + mlen;
+ DEBUGLOG(6, "considering seq starting at %zi, llen=%u, mlen=%u",
+ anchor - istart, (unsigned)llen, (unsigned)mlen);
+
+ if (mlen==0) { /* only literals => must be last "sequence", actually starting a new stream of sequences */
+ assert(storePos == storeEnd); /* must be last sequence */
+ ip = anchor + llen; /* last "sequence" is a bunch of literals => don't progress anchor */
+ continue; /* will finish */
}
- }
- ZSTD_updateStats(optStatePtr, llen, anchor, offset, mlen);
- ZSTD_storeSeq(seqStore, llen, anchor, offset, mlen-MINMATCH);
- anchor = ip;
- } }
- ZSTD_setLog2Prices(optStatePtr);
+ /* repcodes update : like ZSTD_updateRep(), but update in place */
+ if (offCode >= ZSTD_REP_NUM) { /* full offset */
+ rep[2] = rep[1];
+ rep[1] = rep[0];
+ rep[0] = offCode - ZSTD_REP_MOVE;
+ } else { /* repcode */
+ U32 const repCode = offCode + (llen==0);
+ if (repCode) { /* note : if repCode==0, no change */
+ U32 const currentOffset = (repCode==ZSTD_REP_NUM) ? (rep[0] - 1) : rep[repCode];
+ if (repCode >= 2) rep[2] = rep[1];
+ rep[1] = rep[0];
+ rep[0] = currentOffset;
+ } }
+
+ assert(anchor + llen <= iend);
+ ZSTD_updateStats(optStatePtr, llen, anchor, offCode, mlen);
+ ZSTD_storeSeq(seqStore, llen, anchor, offCode, mlen-MINMATCH);
+ anchor += advance;
+ ip = anchor;
+ } }
+ ZSTD_setBasePrices(optStatePtr, optLevel);
+ }
+
} /* while (ip < ilimit) */
/* Return the last literals size */
- return iend - anchor;
+ return (size_t)(iend - anchor);
}
size_t ZSTD_compressBlock_btopt(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ const void* src, size_t srcSize)
{
DEBUGLOG(5, "ZSTD_compressBlock_btopt");
- return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, cParams, src, srcSize, 0 /*optLevel*/, 0 /*extDict*/);
+ return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 0 /*optLevel*/, ZSTD_noDict);
+}
+
+
+/* used in 2-pass strategy */
+static U32 ZSTD_upscaleStat(unsigned* table, U32 lastEltIndex, int bonus)
+{
+ U32 s, sum=0;
+ assert(ZSTD_FREQ_DIV+bonus >= 0);
+ for (s=0; s<lastEltIndex+1; s++) {
+ table[s] <<= ZSTD_FREQ_DIV+bonus;
+ table[s]--;
+ sum += table[s];
+ }
+ return sum;
+}
+
+/* used in 2-pass strategy */
+MEM_STATIC void ZSTD_upscaleStats(optState_t* optPtr)
+{
+ if (ZSTD_compressedLiterals(optPtr))
+ optPtr->litSum = ZSTD_upscaleStat(optPtr->litFreq, MaxLit, 0);
+ optPtr->litLengthSum = ZSTD_upscaleStat(optPtr->litLengthFreq, MaxLL, 0);
+ optPtr->matchLengthSum = ZSTD_upscaleStat(optPtr->matchLengthFreq, MaxML, 0);
+ optPtr->offCodeSum = ZSTD_upscaleStat(optPtr->offCodeFreq, MaxOff, 0);
+}
+
+/* ZSTD_initStats_ultra():
+ * make a first compression pass, just to seed stats with more accurate starting values.
+ * only works on first block, with no dictionary and no ldm.
+ * this function cannot error, hence its contract must be respected.
+ */
+static void
+ZSTD_initStats_ultra(ZSTD_matchState_t* ms,
+ seqStore_t* seqStore,
+ U32 rep[ZSTD_REP_NUM],
+ const void* src, size_t srcSize)
+{
+ U32 tmpRep[ZSTD_REP_NUM]; /* updated rep codes will sink here */
+ memcpy(tmpRep, rep, sizeof(tmpRep));
+
+ DEBUGLOG(4, "ZSTD_initStats_ultra (srcSize=%zu)", srcSize);
+ assert(ms->opt.litLengthSum == 0); /* first block */
+ assert(seqStore->sequences == seqStore->sequencesStart); /* no ldm */
+ assert(ms->window.dictLimit == ms->window.lowLimit); /* no dictionary */
+ assert(ms->window.dictLimit - ms->nextToUpdate <= 1); /* no prefix (note: intentional overflow, defined as 2-complement) */
+
+ ZSTD_compressBlock_opt_generic(ms, seqStore, tmpRep, src, srcSize, 2 /*optLevel*/, ZSTD_noDict); /* generate stats into ms->opt*/
+
+ /* invalidate first scan from history */
+ ZSTD_resetSeqStore(seqStore);
+ ms->window.base -= srcSize;
+ ms->window.dictLimit += (U32)srcSize;
+ ms->window.lowLimit = ms->window.dictLimit;
+ ms->nextToUpdate = ms->window.dictLimit;
+
+ /* re-inforce weight of collected statistics */
+ ZSTD_upscaleStats(&ms->opt);
}
size_t ZSTD_compressBlock_btultra(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ const void* src, size_t srcSize)
{
- return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, cParams, src, srcSize, 2 /*optLevel*/, 0 /*extDict*/);
+ DEBUGLOG(5, "ZSTD_compressBlock_btultra (srcSize=%zu)", srcSize);
+ return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 2 /*optLevel*/, ZSTD_noDict);
+}
+
+size_t ZSTD_compressBlock_btultra2(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ const void* src, size_t srcSize)
+{
+ U32 const current = (U32)((const BYTE*)src - ms->window.base);
+ DEBUGLOG(5, "ZSTD_compressBlock_btultra2 (srcSize=%zu)", srcSize);
+
+ /* 2-pass strategy:
+ * this strategy makes a first pass over first block to collect statistics
+ * and seed next round's statistics with it.
+ * After 1st pass, function forgets everything, and starts a new block.
+ * Consequently, this can only work if no data has been previously loaded in tables,
+ * aka, no dictionary, no prefix, no ldm preprocessing.
+ * The compression ratio gain is generally small (~0.5% on first block),
+ * the cost is 2x cpu time on first block. */
+ assert(srcSize <= ZSTD_BLOCKSIZE_MAX);
+ if ( (ms->opt.litLengthSum==0) /* first block */
+ && (seqStore->sequences == seqStore->sequencesStart) /* no ldm */
+ && (ms->window.dictLimit == ms->window.lowLimit) /* no dictionary */
+ && (current == ms->window.dictLimit) /* start of frame, nothing already loaded nor skipped */
+ && (srcSize > ZSTD_PREDEF_THRESHOLD)
+ ) {
+ ZSTD_initStats_ultra(ms, seqStore, rep, src, srcSize);
+ }
+
+ return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 2 /*optLevel*/, ZSTD_noDict);
+}
+
+size_t ZSTD_compressBlock_btopt_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 0 /*optLevel*/, ZSTD_dictMatchState);
+}
+
+size_t ZSTD_compressBlock_btultra_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 2 /*optLevel*/, ZSTD_dictMatchState);
}
size_t ZSTD_compressBlock_btopt_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ const void* src, size_t srcSize)
{
- return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, cParams, src, srcSize, 0 /*optLevel*/, 1 /*extDict*/);
+ return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 0 /*optLevel*/, ZSTD_extDict);
}
size_t ZSTD_compressBlock_btultra_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize)
+ const void* src, size_t srcSize)
{
- return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, cParams, src, srcSize, 2 /*optLevel*/, 1 /*extDict*/);
+ return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 2 /*optLevel*/, ZSTD_extDict);
}
+
+/* note : no btultra2 variant for extDict nor dictMatchState,
+ * because btultra2 is not meant to work with dictionaries
+ * and is only specific for the first block (no prefix) */
diff --git a/vendor/github.com/DataDog/zstd/zstd_opt.h b/vendor/github.com/DataDog/zstd/zstd_opt.h
index b8dc389..094f747 100644
--- a/vendor/github.com/DataDog/zstd/zstd_opt.h
+++ b/vendor/github.com/DataDog/zstd/zstd_opt.h
@@ -17,23 +17,37 @@
#include "zstd_compress_internal.h"
-void ZSTD_updateTree(
- ZSTD_matchState_t* ms, ZSTD_compressionParameters const* cParams,
- const BYTE* ip, const BYTE* iend); /* used in ZSTD_loadDictionaryContent() */
+/* used in ZSTD_loadDictionaryContent() */
+void ZSTD_updateTree(ZSTD_matchState_t* ms, const BYTE* ip, const BYTE* iend);
size_t ZSTD_compressBlock_btopt(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btultra(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
+size_t ZSTD_compressBlock_btultra2(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize);
+
+
+size_t ZSTD_compressBlock_btopt_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize);
+size_t ZSTD_compressBlock_btultra_dictMatchState(
+ ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btopt_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btultra_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
- ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+ void const* src, size_t srcSize);
+
+ /* note : no btultra2 variant for extDict nor dictMatchState,
+ * because btultra2 is not meant to work with dictionaries
+ * and is only specific for the first block (no prefix) */
#if defined (__cplusplus)
}
diff --git a/vendor/github.com/DataDog/zstd/zstd_stream.go b/vendor/github.com/DataDog/zstd/zstd_stream.go
index d5d1336..2330353 100644
--- a/vendor/github.com/DataDog/zstd/zstd_stream.go
+++ b/vendor/github.com/DataDog/zstd/zstd_stream.go
@@ -11,6 +11,7 @@
"errors"
"fmt"
"io"
+ "runtime"
"unsafe"
)
@@ -243,6 +244,8 @@
unsafe.Pointer(&src[0]),
&cSrcSize))
+ // Keep src here eventhough, we reuse later, the code might be deleted at some point
+ runtime.KeepAlive(src)
if err = getError(retCode); err != nil {
return 0, fmt.Errorf("failed to decompress: %s", err)
}
diff --git a/vendor/github.com/DataDog/zstd/zstd_v01.c b/vendor/github.com/DataDog/zstd/zstd_v01.c
index ae1cb2c..ae8cba2 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v01.c
+++ b/vendor/github.com/DataDog/zstd/zstd_v01.c
@@ -668,11 +668,17 @@
switch(srcSize)
{
case 7: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[6]) << (sizeof(size_t)*8 - 16);
+ /* fallthrough */
case 6: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[5]) << (sizeof(size_t)*8 - 24);
+ /* fallthrough */
case 5: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[4]) << (sizeof(size_t)*8 - 32);
+ /* fallthrough */
case 4: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[3]) << 24;
+ /* fallthrough */
case 3: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[2]) << 16;
+ /* fallthrough */
case 2: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[1]) << 8;
+ /* fallthrough */
default:;
}
contain32 = ((const BYTE*)srcBuffer)[srcSize-1];
@@ -1067,99 +1073,102 @@
const void* cSrc, size_t cSrcSize,
const U16* DTable)
{
- BYTE* const ostart = (BYTE*) dst;
- BYTE* op = ostart;
- BYTE* const omax = op + maxDstSize;
- BYTE* const olimit = omax-15;
-
- const void* ptr = DTable;
- const HUF_DElt* const dt = (const HUF_DElt*)(ptr)+1;
- const U32 dtLog = DTable[0];
- size_t errorCode;
- U32 reloadStatus;
-
- /* Init */
-
- const U16* jumpTable = (const U16*)cSrc;
- const size_t length1 = FSE_readLE16(jumpTable);
- const size_t length2 = FSE_readLE16(jumpTable+1);
- const size_t length3 = FSE_readLE16(jumpTable+2);
- const size_t length4 = cSrcSize - 6 - length1 - length2 - length3; // check coherency !!
- const char* const start1 = (const char*)(cSrc) + 6;
- const char* const start2 = start1 + length1;
- const char* const start3 = start2 + length2;
- const char* const start4 = start3 + length3;
- FSE_DStream_t bitD1, bitD2, bitD3, bitD4;
-
- if (length1+length2+length3+6 >= cSrcSize) return (size_t)-FSE_ERROR_srcSize_wrong;
-
- errorCode = FSE_initDStream(&bitD1, start1, length1);
- if (FSE_isError(errorCode)) return errorCode;
- errorCode = FSE_initDStream(&bitD2, start2, length2);
- if (FSE_isError(errorCode)) return errorCode;
- errorCode = FSE_initDStream(&bitD3, start3, length3);
- if (FSE_isError(errorCode)) return errorCode;
- errorCode = FSE_initDStream(&bitD4, start4, length4);
- if (FSE_isError(errorCode)) return errorCode;
-
- reloadStatus=FSE_reloadDStream(&bitD2);
-
- /* 16 symbols per loop */
- for ( ; (reloadStatus<FSE_DStream_completed) && (op<olimit); /* D2-3-4 are supposed to be synchronized and finish together */
- op+=16, reloadStatus = FSE_reloadDStream(&bitD2) | FSE_reloadDStream(&bitD3) | FSE_reloadDStream(&bitD4), FSE_reloadDStream(&bitD1))
+ if (cSrcSize < 6) return (size_t)-FSE_ERROR_srcSize_wrong;
{
-#define HUF_DECODE_SYMBOL_0(n, Dstream) \
- op[n] = HUF_decodeSymbol(&Dstream, dt, dtLog);
+ BYTE* const ostart = (BYTE*) dst;
+ BYTE* op = ostart;
+ BYTE* const omax = op + maxDstSize;
+ BYTE* const olimit = omax-15;
-#define HUF_DECODE_SYMBOL_1(n, Dstream) \
- op[n] = HUF_decodeSymbol(&Dstream, dt, dtLog); \
- if (FSE_32bits() && (HUF_MAX_TABLELOG>12)) FSE_reloadDStream(&Dstream)
+ const void* ptr = DTable;
+ const HUF_DElt* const dt = (const HUF_DElt*)(ptr)+1;
+ const U32 dtLog = DTable[0];
+ size_t errorCode;
+ U32 reloadStatus;
-#define HUF_DECODE_SYMBOL_2(n, Dstream) \
- op[n] = HUF_decodeSymbol(&Dstream, dt, dtLog); \
- if (FSE_32bits()) FSE_reloadDStream(&Dstream)
+ /* Init */
- HUF_DECODE_SYMBOL_1( 0, bitD1);
- HUF_DECODE_SYMBOL_1( 1, bitD2);
- HUF_DECODE_SYMBOL_1( 2, bitD3);
- HUF_DECODE_SYMBOL_1( 3, bitD4);
- HUF_DECODE_SYMBOL_2( 4, bitD1);
- HUF_DECODE_SYMBOL_2( 5, bitD2);
- HUF_DECODE_SYMBOL_2( 6, bitD3);
- HUF_DECODE_SYMBOL_2( 7, bitD4);
- HUF_DECODE_SYMBOL_1( 8, bitD1);
- HUF_DECODE_SYMBOL_1( 9, bitD2);
- HUF_DECODE_SYMBOL_1(10, bitD3);
- HUF_DECODE_SYMBOL_1(11, bitD4);
- HUF_DECODE_SYMBOL_0(12, bitD1);
- HUF_DECODE_SYMBOL_0(13, bitD2);
- HUF_DECODE_SYMBOL_0(14, bitD3);
- HUF_DECODE_SYMBOL_0(15, bitD4);
- }
+ const U16* jumpTable = (const U16*)cSrc;
+ const size_t length1 = FSE_readLE16(jumpTable);
+ const size_t length2 = FSE_readLE16(jumpTable+1);
+ const size_t length3 = FSE_readLE16(jumpTable+2);
+ const size_t length4 = cSrcSize - 6 - length1 - length2 - length3; // check coherency !!
+ const char* const start1 = (const char*)(cSrc) + 6;
+ const char* const start2 = start1 + length1;
+ const char* const start3 = start2 + length2;
+ const char* const start4 = start3 + length3;
+ FSE_DStream_t bitD1, bitD2, bitD3, bitD4;
- if (reloadStatus!=FSE_DStream_completed) /* not complete : some bitStream might be FSE_DStream_unfinished */
- return (size_t)-FSE_ERROR_corruptionDetected;
+ if (length1+length2+length3+6 >= cSrcSize) return (size_t)-FSE_ERROR_srcSize_wrong;
- /* tail */
- {
- // bitTail = bitD1; // *much* slower : -20% !??!
- FSE_DStream_t bitTail;
- bitTail.ptr = bitD1.ptr;
- bitTail.bitsConsumed = bitD1.bitsConsumed;
- bitTail.bitContainer = bitD1.bitContainer; // required in case of FSE_DStream_endOfBuffer
- bitTail.start = start1;
- for ( ; (FSE_reloadDStream(&bitTail) < FSE_DStream_completed) && (op<omax) ; op++)
+ errorCode = FSE_initDStream(&bitD1, start1, length1);
+ if (FSE_isError(errorCode)) return errorCode;
+ errorCode = FSE_initDStream(&bitD2, start2, length2);
+ if (FSE_isError(errorCode)) return errorCode;
+ errorCode = FSE_initDStream(&bitD3, start3, length3);
+ if (FSE_isError(errorCode)) return errorCode;
+ errorCode = FSE_initDStream(&bitD4, start4, length4);
+ if (FSE_isError(errorCode)) return errorCode;
+
+ reloadStatus=FSE_reloadDStream(&bitD2);
+
+ /* 16 symbols per loop */
+ for ( ; (reloadStatus<FSE_DStream_completed) && (op<olimit); /* D2-3-4 are supposed to be synchronized and finish together */
+ op+=16, reloadStatus = FSE_reloadDStream(&bitD2) | FSE_reloadDStream(&bitD3) | FSE_reloadDStream(&bitD4), FSE_reloadDStream(&bitD1))
{
- HUF_DECODE_SYMBOL_0(0, bitTail);
+ #define HUF_DECODE_SYMBOL_0(n, Dstream) \
+ op[n] = HUF_decodeSymbol(&Dstream, dt, dtLog);
+
+ #define HUF_DECODE_SYMBOL_1(n, Dstream) \
+ op[n] = HUF_decodeSymbol(&Dstream, dt, dtLog); \
+ if (FSE_32bits() && (HUF_MAX_TABLELOG>12)) FSE_reloadDStream(&Dstream)
+
+ #define HUF_DECODE_SYMBOL_2(n, Dstream) \
+ op[n] = HUF_decodeSymbol(&Dstream, dt, dtLog); \
+ if (FSE_32bits()) FSE_reloadDStream(&Dstream)
+
+ HUF_DECODE_SYMBOL_1( 0, bitD1);
+ HUF_DECODE_SYMBOL_1( 1, bitD2);
+ HUF_DECODE_SYMBOL_1( 2, bitD3);
+ HUF_DECODE_SYMBOL_1( 3, bitD4);
+ HUF_DECODE_SYMBOL_2( 4, bitD1);
+ HUF_DECODE_SYMBOL_2( 5, bitD2);
+ HUF_DECODE_SYMBOL_2( 6, bitD3);
+ HUF_DECODE_SYMBOL_2( 7, bitD4);
+ HUF_DECODE_SYMBOL_1( 8, bitD1);
+ HUF_DECODE_SYMBOL_1( 9, bitD2);
+ HUF_DECODE_SYMBOL_1(10, bitD3);
+ HUF_DECODE_SYMBOL_1(11, bitD4);
+ HUF_DECODE_SYMBOL_0(12, bitD1);
+ HUF_DECODE_SYMBOL_0(13, bitD2);
+ HUF_DECODE_SYMBOL_0(14, bitD3);
+ HUF_DECODE_SYMBOL_0(15, bitD4);
}
- if (FSE_endOfDStream(&bitTail))
- return op-ostart;
+ if (reloadStatus!=FSE_DStream_completed) /* not complete : some bitStream might be FSE_DStream_unfinished */
+ return (size_t)-FSE_ERROR_corruptionDetected;
+
+ /* tail */
+ {
+ // bitTail = bitD1; // *much* slower : -20% !??!
+ FSE_DStream_t bitTail;
+ bitTail.ptr = bitD1.ptr;
+ bitTail.bitsConsumed = bitD1.bitsConsumed;
+ bitTail.bitContainer = bitD1.bitContainer; // required in case of FSE_DStream_endOfBuffer
+ bitTail.start = start1;
+ for ( ; (FSE_reloadDStream(&bitTail) < FSE_DStream_completed) && (op<omax) ; op++)
+ {
+ HUF_DECODE_SYMBOL_0(0, bitTail);
+ }
+
+ if (FSE_endOfDStream(&bitTail))
+ return op-ostart;
+ }
+
+ if (op==omax) return (size_t)-FSE_ERROR_dstSize_tooSmall; /* dst buffer is full, but cSrc unfinished */
+
+ return (size_t)-FSE_ERROR_corruptionDetected;
}
-
- if (op==omax) return (size_t)-FSE_ERROR_dstSize_tooSmall; /* dst buffer is full, but cSrc unfinished */
-
- return (size_t)-FSE_ERROR_corruptionDetected;
}
@@ -1330,6 +1339,8 @@
#define LITERAL_NOENTROPY 63
#define COMMAND_NOENTROPY 7 /* to remove */
+#define ZSTD_CONTENTSIZE_ERROR (0ULL - 2)
+
static const size_t ZSTD_blockHeaderSize = 3;
static const size_t ZSTD_frameHeaderSize = 4;
@@ -1347,8 +1358,6 @@
static U16 ZSTD_read16(const void* p) { U16 r; memcpy(&r, p, sizeof(r)); return r; }
-static U32 ZSTD_read32(const void* p) { U32 r; memcpy(&r, p, sizeof(r)); return r; }
-
static void ZSTD_copy4(void* dst, const void* src) { memcpy(dst, src, 4); }
static void ZSTD_copy8(void* dst, const void* src) { memcpy(dst, src, 8); }
@@ -1373,16 +1382,9 @@
}
}
-
-static U32 ZSTD_readLE32(const void* memPtr)
+static U32 ZSTD_readLE24(const void* memPtr)
{
- if (ZSTD_isLittleEndian())
- return ZSTD_read32(memPtr);
- else
- {
- const BYTE* p = (const BYTE*)memPtr;
- return (U32)((U32)p[0] + ((U32)p[1]<<8) + ((U32)p[2]<<16) + ((U32)p[3]<<24));
- }
+ return ZSTD_readLE16(memPtr) + (((const BYTE*)memPtr)[2] << 16);
}
static U32 ZSTD_readBE32(const void* memPtr)
@@ -1458,7 +1460,7 @@
* Decompression code
**************************************************************/
-size_t ZSTDv01_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr)
+static size_t ZSTDv01_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr)
{
const BYTE* const in = (const BYTE* const)src;
BYTE headerFlags;
@@ -1511,7 +1513,7 @@
}
-size_t ZSTDv01_decodeLiteralsBlock(void* ctx,
+static size_t ZSTDv01_decodeLiteralsBlock(void* ctx,
void* dst, size_t maxDstSize,
const BYTE** litStart, size_t* litSize,
const void* src, size_t srcSize)
@@ -1563,7 +1565,7 @@
}
-size_t ZSTDv01_decodeSeqHeaders(int* nbSeq, const BYTE** dumpsPtr, size_t* dumpsLengthPtr,
+static size_t ZSTDv01_decodeSeqHeaders(int* nbSeq, const BYTE** dumpsPtr, size_t* dumpsLengthPtr,
FSE_DTable* DTableLL, FSE_DTable* DTableML, FSE_DTable* DTableOffb,
const void* src, size_t srcSize)
{
@@ -1696,13 +1698,13 @@
seqState->prevOffset = seq->offset;
if (litLength == MaxLL)
{
- U32 add = dumps<de ? *dumps++ : 0;
+ const U32 add = dumps<de ? *dumps++ : 0;
if (add < 255) litLength += add;
else
{
if (dumps<=(de-3))
{
- litLength = ZSTD_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
+ litLength = ZSTD_readLE24(dumps);
dumps += 3;
}
}
@@ -1724,13 +1726,13 @@
matchLength = FSE_decodeSymbol(&(seqState->stateML), &(seqState->DStream));
if (matchLength == MaxML)
{
- U32 add = dumps<de ? *dumps++ : 0;
+ const U32 add = dumps<de ? *dumps++ : 0;
if (add < 255) matchLength += add;
else
{
if (dumps<=(de-3))
{
- matchLength = ZSTD_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
+ matchLength = ZSTD_readLE24(dumps);
dumps += 3;
}
}
@@ -1751,7 +1753,7 @@
BYTE* const base, BYTE* const oend)
{
static const int dec32table[] = {0, 1, 2, 1, 4, 4, 4, 4}; /* added */
- static const int dec64table[] = {8, 8, 8, 7, 8, 9,10,11}; /* substracted */
+ static const int dec64table[] = {8, 8, 8, 7, 8, 9,10,11}; /* subtracted */
const BYTE* const ostart = op;
const size_t litLength = sequence.litLength;
BYTE* const endMatch = op + litLength + sequence.matchLength; /* risk : address space overflow (32-bits) */
@@ -1993,36 +1995,59 @@
return ZSTDv01_decompressDCtx(&ctx, dst, maxDstSize, src, srcSize);
}
-size_t ZSTDv01_findFrameCompressedSize(const void* src, size_t srcSize)
+/* ZSTD_errorFrameSizeInfoLegacy() :
+ assumes `cSize` and `dBound` are _not_ NULL */
+static void ZSTD_errorFrameSizeInfoLegacy(size_t* cSize, unsigned long long* dBound, size_t ret)
+{
+ *cSize = ret;
+ *dBound = ZSTD_CONTENTSIZE_ERROR;
+}
+
+void ZSTDv01_findFrameSizeInfoLegacy(const void *src, size_t srcSize, size_t* cSize, unsigned long long* dBound)
{
const BYTE* ip = (const BYTE*)src;
size_t remainingSize = srcSize;
+ size_t nbBlocks = 0;
U32 magicNumber;
blockProperties_t blockProperties;
/* Frame Header */
- if (srcSize < ZSTD_frameHeaderSize+ZSTD_blockHeaderSize) return ERROR(srcSize_wrong);
+ if (srcSize < ZSTD_frameHeaderSize+ZSTD_blockHeaderSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
magicNumber = ZSTD_readBE32(src);
- if (magicNumber != ZSTD_magicNumber) return ERROR(prefix_unknown);
+ if (magicNumber != ZSTD_magicNumber) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(prefix_unknown));
+ return;
+ }
ip += ZSTD_frameHeaderSize; remainingSize -= ZSTD_frameHeaderSize;
/* Loop on each block */
while (1)
{
size_t blockSize = ZSTDv01_getcBlockSize(ip, remainingSize, &blockProperties);
- if (ZSTDv01_isError(blockSize)) return blockSize;
+ if (ZSTDv01_isError(blockSize)) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, blockSize);
+ return;
+ }
ip += ZSTD_blockHeaderSize;
remainingSize -= ZSTD_blockHeaderSize;
- if (blockSize > remainingSize) return ERROR(srcSize_wrong);
+ if (blockSize > remainingSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
if (blockSize == 0) break; /* bt_end */
ip += blockSize;
remainingSize -= blockSize;
+ nbBlocks++;
}
- return ip - (const BYTE*)src;
+ *cSize = ip - (const BYTE*)src;
+ *dBound = nbBlocks * BLOCKSIZE;
}
/*******************************
diff --git a/vendor/github.com/DataDog/zstd/zstd_v01.h b/vendor/github.com/DataDog/zstd/zstd_v01.h
index 42f0897..245f9dd 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v01.h
+++ b/vendor/github.com/DataDog/zstd/zstd_v01.h
@@ -35,13 +35,18 @@
size_t ZSTDv01_decompress( void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
-/**
-ZSTDv01_getFrameSrcSize() : get the source length of a ZSTD frame compliant with v0.1.x format
- compressedSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
- return : the number of bytes that would be read to decompress this frame
- or an errorCode if it fails (which can be tested using ZSTDv01_isError())
-*/
-size_t ZSTDv01_findFrameCompressedSize(const void* src, size_t compressedSize);
+ /**
+ ZSTDv01_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.1.x format
+ srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
+ cSize (output parameter) : the number of bytes that would be read to decompress this frame
+ or an error code if it fails (which can be tested using ZSTDv01_isError())
+ dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
+ or ZSTD_CONTENTSIZE_ERROR if an error occurs
+
+ note : assumes `cSize` and `dBound` are _not_ NULL.
+ */
+void ZSTDv01_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
+ size_t* cSize, unsigned long long* dBound);
/**
ZSTDv01_isError() : tells if the result of ZSTDv01_decompress() is an error
diff --git a/vendor/github.com/DataDog/zstd/zstd_v02.c b/vendor/github.com/DataDog/zstd/zstd_v02.c
index 8bc0ece..793df60 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v02.c
+++ b/vendor/github.com/DataDog/zstd/zstd_v02.c
@@ -217,6 +217,11 @@
}
}
+MEM_STATIC U32 MEM_readLE24(const void* memPtr)
+{
+ return MEM_readLE16(memPtr) + (((const BYTE*)memPtr)[2] << 16);
+}
+
MEM_STATIC U32 MEM_readLE32(const void* memPtr)
{
if (MEM_isLittleEndian())
@@ -399,11 +404,17 @@
switch(srcSize)
{
case 7: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[6]) << (sizeof(size_t)*8 - 16);
+ /* fallthrough */
case 6: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[5]) << (sizeof(size_t)*8 - 24);
+ /* fallthrough */
case 5: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[4]) << (sizeof(size_t)*8 - 32);
+ /* fallthrough */
case 4: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[3]) << 24;
+ /* fallthrough */
case 3: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[2]) << 16;
+ /* fallthrough */
case 2: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[1]) << 8;
+ /* fallthrough */
default:;
}
contain32 = ((const BYTE*)srcBuffer)[srcSize-1];
@@ -2722,6 +2733,8 @@
#define LITERAL_NOENTROPY 63
#define COMMAND_NOENTROPY 7 /* to remove */
+#define ZSTD_CONTENTSIZE_ERROR (0ULL - 2)
+
static const size_t ZSTD_blockHeaderSize = 3;
static const size_t ZSTD_frameHeaderSize = 4;
@@ -3035,11 +3048,11 @@
seqState->prevOffset = seq->offset;
if (litLength == MaxLL)
{
- U32 add = *dumps++;
+ const U32 add = dumps<de ? *dumps++ : 0;
if (add < 255) litLength += add;
- else
+ else if (dumps + 3 <= de)
{
- litLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
+ litLength = MEM_readLE24(dumps);
dumps += 3;
}
if (dumps >= de) dumps = de-1; /* late correction, to avoid read overflow (data is now corrupted anyway) */
@@ -3065,11 +3078,11 @@
matchLength = FSE_decodeSymbol(&(seqState->stateML), &(seqState->DStream));
if (matchLength == MaxML)
{
- U32 add = *dumps++;
+ const U32 add = dumps<de ? *dumps++ : 0;
if (add < 255) matchLength += add;
- else
+ else if (dumps + 3 <= de)
{
- matchLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
+ matchLength = MEM_readLE24(dumps);
dumps += 3;
}
if (dumps >= de) dumps = de-1; /* late correction, to avoid read overflow (data is now corrupted anyway) */
@@ -3090,7 +3103,7 @@
BYTE* const base, BYTE* const oend)
{
static const int dec32table[] = {0, 1, 2, 1, 4, 4, 4, 4}; /* added */
- static const int dec64table[] = {8, 8, 8, 7, 8, 9,10,11}; /* substracted */
+ static const int dec64table[] = {8, 8, 8, 7, 8, 9,10,11}; /* subtracted */
const BYTE* const ostart = op;
BYTE* const oLitEnd = op + sequence.litLength;
BYTE* const oMatchEnd = op + sequence.litLength + sequence.matchLength; /* risk : address space overflow (32-bits) */
@@ -3306,37 +3319,59 @@
return ZSTD_decompressDCtx(&ctx, dst, maxDstSize, src, srcSize);
}
-static size_t ZSTD_findFrameCompressedSize(const void *src, size_t srcSize)
+/* ZSTD_errorFrameSizeInfoLegacy() :
+ assumes `cSize` and `dBound` are _not_ NULL */
+static void ZSTD_errorFrameSizeInfoLegacy(size_t* cSize, unsigned long long* dBound, size_t ret)
{
+ *cSize = ret;
+ *dBound = ZSTD_CONTENTSIZE_ERROR;
+}
+void ZSTDv02_findFrameSizeInfoLegacy(const void *src, size_t srcSize, size_t* cSize, unsigned long long* dBound)
+{
const BYTE* ip = (const BYTE*)src;
size_t remainingSize = srcSize;
+ size_t nbBlocks = 0;
U32 magicNumber;
blockProperties_t blockProperties;
/* Frame Header */
- if (srcSize < ZSTD_frameHeaderSize+ZSTD_blockHeaderSize) return ERROR(srcSize_wrong);
+ if (srcSize < ZSTD_frameHeaderSize+ZSTD_blockHeaderSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
magicNumber = MEM_readLE32(src);
- if (magicNumber != ZSTD_magicNumber) return ERROR(prefix_unknown);
+ if (magicNumber != ZSTD_magicNumber) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(prefix_unknown));
+ return;
+ }
ip += ZSTD_frameHeaderSize; remainingSize -= ZSTD_frameHeaderSize;
/* Loop on each block */
while (1)
{
size_t cBlockSize = ZSTD_getcBlockSize(ip, remainingSize, &blockProperties);
- if (ZSTD_isError(cBlockSize)) return cBlockSize;
+ if (ZSTD_isError(cBlockSize)) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, cBlockSize);
+ return;
+ }
ip += ZSTD_blockHeaderSize;
remainingSize -= ZSTD_blockHeaderSize;
- if (cBlockSize > remainingSize) return ERROR(srcSize_wrong);
+ if (cBlockSize > remainingSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
if (cBlockSize == 0) break; /* bt_end */
ip += cBlockSize;
remainingSize -= cBlockSize;
+ nbBlocks++;
}
- return ip - (const BYTE*)src;
+ *cSize = ip - (const BYTE*)src;
+ *dBound = nbBlocks * BLOCKSIZE;
}
/*******************************
@@ -3452,11 +3487,6 @@
return ZSTD_decompress(dst, maxOriginalSize, src, compressedSize);
}
-size_t ZSTDv02_findFrameCompressedSize(const void *src, size_t compressedSize)
-{
- return ZSTD_findFrameCompressedSize(src, compressedSize);
-}
-
ZSTDv02_Dctx* ZSTDv02_createDCtx(void)
{
return (ZSTDv02_Dctx*)ZSTD_createDCtx();
diff --git a/vendor/github.com/DataDog/zstd/zstd_v02.h b/vendor/github.com/DataDog/zstd/zstd_v02.h
index 0dde7a6..9d7d8d9 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v02.h
+++ b/vendor/github.com/DataDog/zstd/zstd_v02.h
@@ -35,13 +35,18 @@
size_t ZSTDv02_decompress( void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
-/**
-ZSTDv02_getFrameSrcSize() : get the source length of a ZSTD frame compliant with v0.2.x format
- compressedSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
- return : the number of bytes that would be read to decompress this frame
- or an errorCode if it fails (which can be tested using ZSTDv02_isError())
-*/
-size_t ZSTDv02_findFrameCompressedSize(const void* src, size_t compressedSize);
+ /**
+ ZSTDv02_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.2.x format
+ srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
+ cSize (output parameter) : the number of bytes that would be read to decompress this frame
+ or an error code if it fails (which can be tested using ZSTDv01_isError())
+ dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
+ or ZSTD_CONTENTSIZE_ERROR if an error occurs
+
+ note : assumes `cSize` and `dBound` are _not_ NULL.
+ */
+void ZSTDv02_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
+ size_t* cSize, unsigned long long* dBound);
/**
ZSTDv02_isError() : tells if the result of ZSTDv02_decompress() is an error
diff --git a/vendor/github.com/DataDog/zstd/zstd_v03.c b/vendor/github.com/DataDog/zstd/zstd_v03.c
index 54445af..7a0e7c9 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v03.c
+++ b/vendor/github.com/DataDog/zstd/zstd_v03.c
@@ -219,6 +219,11 @@
}
}
+MEM_STATIC U32 MEM_readLE24(const void* memPtr)
+{
+ return MEM_readLE16(memPtr) + (((const BYTE*)memPtr)[2] << 16);
+}
+
MEM_STATIC U32 MEM_readLE32(const void* memPtr)
{
if (MEM_isLittleEndian())
@@ -402,11 +407,17 @@
switch(srcSize)
{
case 7: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[6]) << (sizeof(size_t)*8 - 16);
+ /* fallthrough */
case 6: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[5]) << (sizeof(size_t)*8 - 24);
+ /* fallthrough */
case 5: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[4]) << (sizeof(size_t)*8 - 32);
+ /* fallthrough */
case 4: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[3]) << 24;
+ /* fallthrough */
case 3: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[2]) << 16;
+ /* fallthrough */
case 2: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[1]) << 8;
+ /* fallthrough */
default:;
}
contain32 = ((const BYTE*)srcBuffer)[srcSize-1];
@@ -2363,6 +2374,8 @@
#define LITERAL_NOENTROPY 63
#define COMMAND_NOENTROPY 7 /* to remove */
+#define ZSTD_CONTENTSIZE_ERROR (0ULL - 2)
+
static const size_t ZSTD_blockHeaderSize = 3;
static const size_t ZSTD_frameHeaderSize = 4;
@@ -2676,11 +2689,11 @@
seqState->prevOffset = seq->offset;
if (litLength == MaxLL)
{
- U32 add = *dumps++;
+ const U32 add = dumps<de ? *dumps++ : 0;
if (add < 255) litLength += add;
- else
+ else if (dumps + 3 <= de)
{
- litLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
+ litLength = MEM_readLE24(dumps);
dumps += 3;
}
if (dumps >= de) dumps = de-1; /* late correction, to avoid read overflow (data is now corrupted anyway) */
@@ -2706,11 +2719,11 @@
matchLength = FSE_decodeSymbol(&(seqState->stateML), &(seqState->DStream));
if (matchLength == MaxML)
{
- U32 add = *dumps++;
+ const U32 add = dumps<de ? *dumps++ : 0;
if (add < 255) matchLength += add;
- else
+ else if (dumps + 3 <= de)
{
- matchLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
+ matchLength = MEM_readLE24(dumps);
dumps += 3;
}
if (dumps >= de) dumps = de-1; /* late correction, to avoid read overflow (data is now corrupted anyway) */
@@ -2731,7 +2744,7 @@
BYTE* const base, BYTE* const oend)
{
static const int dec32table[] = {0, 1, 2, 1, 4, 4, 4, 4}; /* added */
- static const int dec64table[] = {8, 8, 8, 7, 8, 9,10,11}; /* substracted */
+ static const int dec64table[] = {8, 8, 8, 7, 8, 9,10,11}; /* subtracted */
const BYTE* const ostart = op;
BYTE* const oLitEnd = op + sequence.litLength;
BYTE* const oMatchEnd = op + sequence.litLength + sequence.matchLength; /* risk : address space overflow (32-bits) */
@@ -2947,36 +2960,59 @@
return ZSTD_decompressDCtx(&ctx, dst, maxDstSize, src, srcSize);
}
-static size_t ZSTD_findFrameCompressedSize(const void* src, size_t srcSize)
+/* ZSTD_errorFrameSizeInfoLegacy() :
+ assumes `cSize` and `dBound` are _not_ NULL */
+MEM_STATIC void ZSTD_errorFrameSizeInfoLegacy(size_t* cSize, unsigned long long* dBound, size_t ret)
+{
+ *cSize = ret;
+ *dBound = ZSTD_CONTENTSIZE_ERROR;
+}
+
+void ZSTDv03_findFrameSizeInfoLegacy(const void *src, size_t srcSize, size_t* cSize, unsigned long long* dBound)
{
const BYTE* ip = (const BYTE*)src;
size_t remainingSize = srcSize;
+ size_t nbBlocks = 0;
U32 magicNumber;
blockProperties_t blockProperties;
/* Frame Header */
- if (srcSize < ZSTD_frameHeaderSize+ZSTD_blockHeaderSize) return ERROR(srcSize_wrong);
+ if (srcSize < ZSTD_frameHeaderSize+ZSTD_blockHeaderSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
magicNumber = MEM_readLE32(src);
- if (magicNumber != ZSTD_magicNumber) return ERROR(prefix_unknown);
+ if (magicNumber != ZSTD_magicNumber) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(prefix_unknown));
+ return;
+ }
ip += ZSTD_frameHeaderSize; remainingSize -= ZSTD_frameHeaderSize;
/* Loop on each block */
while (1)
{
size_t cBlockSize = ZSTD_getcBlockSize(ip, remainingSize, &blockProperties);
- if (ZSTD_isError(cBlockSize)) return cBlockSize;
+ if (ZSTD_isError(cBlockSize)) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, cBlockSize);
+ return;
+ }
ip += ZSTD_blockHeaderSize;
remainingSize -= ZSTD_blockHeaderSize;
- if (cBlockSize > remainingSize) return ERROR(srcSize_wrong);
+ if (cBlockSize > remainingSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
if (cBlockSize == 0) break; /* bt_end */
ip += cBlockSize;
remainingSize -= cBlockSize;
+ nbBlocks++;
}
- return ip - (const BYTE*)src;
+ *cSize = ip - (const BYTE*)src;
+ *dBound = nbBlocks * BLOCKSIZE;
}
@@ -3093,11 +3129,6 @@
return ZSTD_decompress(dst, maxOriginalSize, src, compressedSize);
}
-size_t ZSTDv03_findFrameCompressedSize(const void* src, size_t srcSize)
-{
- return ZSTD_findFrameCompressedSize(src, srcSize);
-}
-
ZSTDv03_Dctx* ZSTDv03_createDCtx(void)
{
return (ZSTDv03_Dctx*)ZSTD_createDCtx();
diff --git a/vendor/github.com/DataDog/zstd/zstd_v03.h b/vendor/github.com/DataDog/zstd/zstd_v03.h
index b4449e2..efd8c2b 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v03.h
+++ b/vendor/github.com/DataDog/zstd/zstd_v03.h
@@ -35,13 +35,18 @@
size_t ZSTDv03_decompress( void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
-/**
-ZSTDv03_getFrameSrcSize() : get the source length of a ZSTD frame compliant with v0.3.x format
- compressedSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
- return : the number of bytes that would be read to decompress this frame
- or an errorCode if it fails (which can be tested using ZSTDv03_isError())
-*/
-size_t ZSTDv03_findFrameCompressedSize(const void* src, size_t compressedSize);
+ /**
+ ZSTDv03_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.3.x format
+ srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
+ cSize (output parameter) : the number of bytes that would be read to decompress this frame
+ or an error code if it fails (which can be tested using ZSTDv01_isError())
+ dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
+ or ZSTD_CONTENTSIZE_ERROR if an error occurs
+
+ note : assumes `cSize` and `dBound` are _not_ NULL.
+ */
+ void ZSTDv03_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
+ size_t* cSize, unsigned long long* dBound);
/**
ZSTDv03_isError() : tells if the result of ZSTDv03_decompress() is an error
diff --git a/vendor/github.com/DataDog/zstd/zstd_v04.c b/vendor/github.com/DataDog/zstd/zstd_v04.c
index fb6d1d4..645a6e3 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v04.c
+++ b/vendor/github.com/DataDog/zstd/zstd_v04.c
@@ -9,14 +9,19 @@
*/
-/*- Dependencies -*/
+ /******************************************
+ * Includes
+ ******************************************/
+#include <stddef.h> /* size_t, ptrdiff_t */
+#include <string.h> /* memcpy */
+
#include "zstd_v04.h"
#include "error_private.h"
/* ******************************************************************
- mem.h
-****************************************************************** */
+ * mem.h
+ *******************************************************************/
#ifndef MEM_H_MODULE
#define MEM_H_MODULE
@@ -24,12 +29,6 @@
extern "C" {
#endif
-/******************************************
-* Includes
-******************************************/
-#include <stddef.h> /* size_t, ptrdiff_t */
-#include <string.h> /* memcpy */
-
/******************************************
* Compiler-specific
@@ -75,38 +74,9 @@
/*-*************************************
* Debug
***************************************/
-#if defined(ZSTD_DEBUG) && (ZSTD_DEBUG>=1)
-# include <assert.h>
-#else
-# ifndef assert
-# define assert(condition) ((void)0)
-# endif
-#endif
-
-#define ZSTD_STATIC_ASSERT(c) { enum { ZSTD_static_assert = 1/(int)(!!(c)) }; }
-
-#if defined(ZSTD_DEBUG) && (ZSTD_DEBUG>=2)
-# include <stdio.h>
-extern int g_debuglog_enable;
-/* recommended values for ZSTD_DEBUG display levels :
- * 1 : no display, enables assert() only
- * 2 : reserved for currently active debug path
- * 3 : events once per object lifetime (CCtx, CDict, etc.)
- * 4 : events once per frame
- * 5 : events once per block
- * 6 : events once per sequence (*very* verbose) */
-# define RAWLOG(l, ...) { \
- if ((g_debuglog_enable) & (l<=ZSTD_DEBUG)) { \
- fprintf(stderr, __VA_ARGS__); \
- } }
-# define DEBUGLOG(l, ...) { \
- if ((g_debuglog_enable) & (l<=ZSTD_DEBUG)) { \
- fprintf(stderr, __FILE__ ": " __VA_ARGS__); \
- fprintf(stderr, " \n"); \
- } }
-#else
-# define RAWLOG(l, ...) {} /* disabled */
-# define DEBUGLOG(l, ...) {} /* disabled */
+#include "debug.h"
+#ifndef assert
+# define assert(condition) ((void)0)
#endif
@@ -219,6 +189,11 @@
}
}
+MEM_STATIC U32 MEM_readLE24(const void* memPtr)
+{
+ return MEM_readLE16(memPtr) + (((const BYTE*)memPtr)[2] << 16);
+}
+
MEM_STATIC U32 MEM_readLE32(const void* memPtr)
{
if (MEM_isLittleEndian())
@@ -266,29 +241,11 @@
#ifndef ZSTD_STATIC_H
#define ZSTD_STATIC_H
-/* The objects defined into this file shall be considered experimental.
- * They are not considered stable, as their prototype may change in the future.
- * You can use them for tests, provide feedback, or if you can endure risks of future changes.
- */
-
-#if defined (__cplusplus)
-extern "C" {
-#endif
/* *************************************
* Types
***************************************/
-#define ZSTD_WINDOWLOG_MAX 26
-#define ZSTD_WINDOWLOG_MIN 18
#define ZSTD_WINDOWLOG_ABSOLUTEMIN 11
-#define ZSTD_CONTENTLOG_MAX (ZSTD_WINDOWLOG_MAX+1)
-#define ZSTD_CONTENTLOG_MIN 4
-#define ZSTD_HASHLOG_MAX 28
-#define ZSTD_HASHLOG_MIN 4
-#define ZSTD_SEARCHLOG_MAX (ZSTD_CONTENTLOG_MAX-1)
-#define ZSTD_SEARCHLOG_MIN 1
-#define ZSTD_SEARCHLENGTH_MAX 7
-#define ZSTD_SEARCHLENGTH_MIN 4
/** from faster to stronger */
typedef enum { ZSTD_fast, ZSTD_greedy, ZSTD_lazy, ZSTD_lazy2, ZSTD_btlazy2 } ZSTD_strategy;
@@ -360,9 +317,6 @@
*/
-#if defined (__cplusplus)
-}
-#endif
#endif /* ZSTD_STATIC_H */
@@ -375,10 +329,6 @@
#ifndef ZSTD_CCOMMON_H_MODULE
#define ZSTD_CCOMMON_H_MODULE
-#if defined (__cplusplus)
-extern "C" {
-#endif
-
/* *************************************
* Common macros
***************************************/
@@ -428,6 +378,8 @@
#define MIN_SEQUENCES_SIZE (2 /*seqNb*/ + 2 /*dumps*/ + 3 /*seqTables*/ + 1 /*bitStream*/)
#define MIN_CBLOCK_SIZE (3 /*litCSize*/ + MIN_SEQUENCES_SIZE)
+#define ZSTD_CONTENTSIZE_ERROR (0ULL - 2)
+
typedef enum { bt_compressed, bt_raw, bt_rle, bt_end } blockType_t;
@@ -450,10 +402,6 @@
}
-#if defined (__cplusplus)
-}
-#endif
-
/* ******************************************************************
FSE : Finite State Entropy coder
@@ -1142,6 +1090,7 @@
if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge);
/* Init, lay down lowprob symbols */
+ memset(tableDecode, 0, sizeof(FSE_DECODE_TYPE) * (maxSymbolValue+1) ); /* useless init, but keep static analyzer happy, and we don't need to performance optimize legacy decoders */
DTableH.tableLog = (U16)tableLog;
for (s=0; s<=maxSymbolValue; s++)
{
@@ -2864,13 +2813,12 @@
litLength = FSE_decodeSymbol(&(seqState->stateLL), &(seqState->DStream));
prevOffset = litLength ? seq->offset : seqState->prevOffset;
if (litLength == MaxLL) {
- U32 add = *dumps++;
+ const U32 add = dumps<de ? *dumps++ : 0;
if (add < 255) litLength += add;
- else {
- litLength = dumps[0] + (dumps[1]<<8) + (dumps[2]<<16);
+ else if (dumps + 3 <= de) {
+ litLength = MEM_readLE24(dumps);
dumps += 3;
}
- if (dumps > de) { litLength = MaxLL+255; } /* late correction, to avoid using uninitialized memory */
if (dumps >= de) { dumps = de-1; } /* late correction, to avoid read overflow (data is now corrupted anyway) */
}
@@ -2893,13 +2841,12 @@
/* MatchLength */
matchLength = FSE_decodeSymbol(&(seqState->stateML), &(seqState->DStream));
if (matchLength == MaxML) {
- U32 add = *dumps++;
+ const U32 add = dumps<de ? *dumps++ : 0;
if (add < 255) matchLength += add;
- else {
- matchLength = dumps[0] + (dumps[1]<<8) + (dumps[2]<<16);
+ else if (dumps + 3 <= de){
+ matchLength = MEM_readLE24(dumps);
dumps += 3;
}
- if (dumps > de) { matchLength = MaxML+255; } /* late correction, to avoid using uninitialized memory */
if (dumps >= de) { dumps = de-1; } /* late correction, to avoid read overflow (data is now corrupted anyway) */
}
matchLength += MINMATCH;
@@ -2918,7 +2865,7 @@
const BYTE* const base, const BYTE* const vBase, const BYTE* const dictEnd)
{
static const int dec32table[] = { 0, 1, 2, 1, 4, 4, 4, 4 }; /* added */
- static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* substracted */
+ static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* subtracted */
BYTE* const oLitEnd = op + sequence.litLength;
const size_t sequenceLength = sequence.litLength + sequence.matchLength;
BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */
@@ -2991,7 +2938,7 @@
}
else
{
- ZSTD_wildcopy(op, match, (ptrdiff_t)sequence.matchLength-8); /* works even if matchLength < 8 */
+ ZSTD_wildcopy(op, match, (ptrdiff_t)sequence.matchLength-8); /* works even if matchLength < 8, but must be signed */
}
return sequenceLength;
}
@@ -3177,34 +3124,57 @@
return op-ostart;
}
-static size_t ZSTD_findFrameCompressedSize(const void* src, size_t srcSize)
+/* ZSTD_errorFrameSizeInfoLegacy() :
+ assumes `cSize` and `dBound` are _not_ NULL */
+static void ZSTD_errorFrameSizeInfoLegacy(size_t* cSize, unsigned long long* dBound, size_t ret)
+{
+ *cSize = ret;
+ *dBound = ZSTD_CONTENTSIZE_ERROR;
+}
+
+void ZSTDv04_findFrameSizeInfoLegacy(const void *src, size_t srcSize, size_t* cSize, unsigned long long* dBound)
{
const BYTE* ip = (const BYTE*)src;
size_t remainingSize = srcSize;
+ size_t nbBlocks = 0;
blockProperties_t blockProperties;
/* Frame Header */
- if (srcSize < ZSTD_frameHeaderSize_min) return ERROR(srcSize_wrong);
- if (MEM_readLE32(src) != ZSTD_MAGICNUMBER) return ERROR(prefix_unknown);
+ if (srcSize < ZSTD_frameHeaderSize_min) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
+ if (MEM_readLE32(src) != ZSTD_MAGICNUMBER) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(prefix_unknown));
+ return;
+ }
ip += ZSTD_frameHeaderSize_min; remainingSize -= ZSTD_frameHeaderSize_min;
/* Loop on each block */
while (1)
{
size_t cBlockSize = ZSTD_getcBlockSize(ip, remainingSize, &blockProperties);
- if (ZSTD_isError(cBlockSize)) return cBlockSize;
+ if (ZSTD_isError(cBlockSize)) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, cBlockSize);
+ return;
+ }
ip += ZSTD_blockHeaderSize;
remainingSize -= ZSTD_blockHeaderSize;
- if (cBlockSize > remainingSize) return ERROR(srcSize_wrong);
+ if (cBlockSize > remainingSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
if (cBlockSize == 0) break; /* bt_end */
ip += cBlockSize;
remainingSize -= cBlockSize;
+ nbBlocks++;
}
- return ip - (const BYTE*)src;
+ *cSize = ip - (const BYTE*)src;
+ *dBound = nbBlocks * BLOCKSIZE;
}
/* ******************************
@@ -3636,11 +3606,6 @@
#endif
}
-size_t ZSTDv04_findFrameCompressedSize(const void* src, size_t srcSize)
-{
- return ZSTD_findFrameCompressedSize(src, srcSize);
-}
-
size_t ZSTDv04_resetDCtx(ZSTDv04_Dctx* dctx) { return ZSTD_resetDCtx(dctx); }
size_t ZSTDv04_nextSrcSizeToDecompress(ZSTDv04_Dctx* dctx)
@@ -3670,8 +3635,3 @@
ZSTD_DCtx* ZSTDv04_createDCtx(void) { return ZSTD_createDCtx(); }
size_t ZSTDv04_freeDCtx(ZSTD_DCtx* dctx) { return ZSTD_freeDCtx(dctx); }
-
-size_t ZSTDv04_getFrameParams(ZSTD_parameters* params, const void* src, size_t srcSize)
-{
- return ZSTD_getFrameParams(params, src, srcSize);
-}
diff --git a/vendor/github.com/DataDog/zstd/zstd_v04.h b/vendor/github.com/DataDog/zstd/zstd_v04.h
index 6391631..bb5f3b7 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v04.h
+++ b/vendor/github.com/DataDog/zstd/zstd_v04.h
@@ -35,13 +35,18 @@
size_t ZSTDv04_decompress( void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
-/**
-ZSTDv04_getFrameSrcSize() : get the source length of a ZSTD frame compliant with v0.4.x format
- compressedSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
- return : the number of bytes that would be read to decompress this frame
- or an errorCode if it fails (which can be tested using ZSTDv04_isError())
-*/
-size_t ZSTDv04_findFrameCompressedSize(const void* src, size_t compressedSize);
+ /**
+ ZSTDv04_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.4.x format
+ srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
+ cSize (output parameter) : the number of bytes that would be read to decompress this frame
+ or an error code if it fails (which can be tested using ZSTDv01_isError())
+ dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
+ or ZSTD_CONTENTSIZE_ERROR if an error occurs
+
+ note : assumes `cSize` and `dBound` are _not_ NULL.
+ */
+ void ZSTDv04_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
+ size_t* cSize, unsigned long long* dBound);
/**
ZSTDv04_isError() : tells if the result of ZSTDv04_decompress() is an error
diff --git a/vendor/github.com/DataDog/zstd/zstd_v05.c b/vendor/github.com/DataDog/zstd/zstd_v05.c
index a5e1b1f..a7ea606 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v05.c
+++ b/vendor/github.com/DataDog/zstd/zstd_v05.c
@@ -218,6 +218,11 @@
}
}
+MEM_STATIC U32 MEM_readLE24(const void* memPtr)
+{
+ return MEM_readLE16(memPtr) + (((const BYTE*)memPtr)[2] << 16);
+}
+
MEM_STATIC U32 MEM_readLE32(const void* memPtr)
{
if (MEM_isLittleEndian())
@@ -491,6 +496,8 @@
#define WILDCOPY_OVERLENGTH 8
+#define ZSTD_CONTENTSIZE_ERROR (0ULL - 2)
+
typedef enum { bt_compressed, bt_raw, bt_rle, bt_end } blockType_t;
@@ -836,7 +843,7 @@
bitD->bitsConsumed += nbBits;
}
-MEM_STATIC size_t BITv05_readBits(BITv05_DStream_t* bitD, U32 nbBits)
+MEM_STATIC size_t BITv05_readBits(BITv05_DStream_t* bitD, unsigned nbBits)
{
size_t value = BITv05_lookBits(bitD, nbBits);
BITv05_skipBits(bitD, nbBits);
@@ -845,7 +852,7 @@
/*!BITv05_readBitsFast :
* unsafe version; only works only if nbBits >= 1 */
-MEM_STATIC size_t BITv05_readBitsFast(BITv05_DStream_t* bitD, U32 nbBits)
+MEM_STATIC size_t BITv05_readBitsFast(BITv05_DStream_t* bitD, unsigned nbBits)
{
size_t value = BITv05_lookBitsFast(bitD, nbBits);
BITv05_skipBits(bitD, nbBits);
@@ -1162,7 +1169,7 @@
/* **************************************************************
* Complex types
****************************************************************/
-typedef U32 DTable_max_t[FSEv05_DTABLE_SIZE_U32(FSEv05_MAX_TABLELOG)];
+typedef unsigned DTable_max_t[FSEv05_DTABLE_SIZE_U32(FSEv05_MAX_TABLELOG)];
/* **************************************************************
@@ -1224,6 +1231,7 @@
if (tableLog > FSEv05_MAX_TABLELOG) return ERROR(tableLog_tooLarge);
/* Init, lay down lowprob symbols */
+ memset(tableDecode, 0, sizeof(FSEv05_FUNCTION_TYPE) * (maxSymbolValue+1) ); /* useless init, but keep static analyzer happy, and we don't need to performance optimize legacy decoders */
DTableH.tableLog = (U16)tableLog;
for (s=0; s<=maxSymbolValue; s++) {
if (normalizedCounter[s]==-1) {
@@ -1995,91 +2003,92 @@
const void* cSrc, size_t cSrcSize,
const U16* DTable)
{
- const BYTE* const istart = (const BYTE*) cSrc;
- BYTE* const ostart = (BYTE*) dst;
- BYTE* const oend = ostart + dstSize;
- const void* const dtPtr = DTable;
- const HUFv05_DEltX2* const dt = ((const HUFv05_DEltX2*)dtPtr) +1;
- const U32 dtLog = DTable[0];
- size_t errorCode;
-
- /* Init */
- BITv05_DStream_t bitD1;
- BITv05_DStream_t bitD2;
- BITv05_DStream_t bitD3;
- BITv05_DStream_t bitD4;
- const size_t length1 = MEM_readLE16(istart);
- const size_t length2 = MEM_readLE16(istart+2);
- const size_t length3 = MEM_readLE16(istart+4);
- size_t length4;
- const BYTE* const istart1 = istart + 6; /* jumpTable */
- const BYTE* const istart2 = istart1 + length1;
- const BYTE* const istart3 = istart2 + length2;
- const BYTE* const istart4 = istart3 + length3;
- const size_t segmentSize = (dstSize+3) / 4;
- BYTE* const opStart2 = ostart + segmentSize;
- BYTE* const opStart3 = opStart2 + segmentSize;
- BYTE* const opStart4 = opStart3 + segmentSize;
- BYTE* op1 = ostart;
- BYTE* op2 = opStart2;
- BYTE* op3 = opStart3;
- BYTE* op4 = opStart4;
- U32 endSignal;
-
/* Check */
if (cSrcSize < 10) return ERROR(corruption_detected); /* strict minimum : jump table + 1 byte per stream */
+ {
+ const BYTE* const istart = (const BYTE*) cSrc;
+ BYTE* const ostart = (BYTE*) dst;
+ BYTE* const oend = ostart + dstSize;
+ const void* const dtPtr = DTable;
+ const HUFv05_DEltX2* const dt = ((const HUFv05_DEltX2*)dtPtr) +1;
+ const U32 dtLog = DTable[0];
+ size_t errorCode;
- length4 = cSrcSize - (length1 + length2 + length3 + 6);
- if (length4 > cSrcSize) return ERROR(corruption_detected); /* overflow */
- errorCode = BITv05_initDStream(&bitD1, istart1, length1);
- if (HUFv05_isError(errorCode)) return errorCode;
- errorCode = BITv05_initDStream(&bitD2, istart2, length2);
- if (HUFv05_isError(errorCode)) return errorCode;
- errorCode = BITv05_initDStream(&bitD3, istart3, length3);
- if (HUFv05_isError(errorCode)) return errorCode;
- errorCode = BITv05_initDStream(&bitD4, istart4, length4);
- if (HUFv05_isError(errorCode)) return errorCode;
+ /* Init */
+ BITv05_DStream_t bitD1;
+ BITv05_DStream_t bitD2;
+ BITv05_DStream_t bitD3;
+ BITv05_DStream_t bitD4;
+ const size_t length1 = MEM_readLE16(istart);
+ const size_t length2 = MEM_readLE16(istart+2);
+ const size_t length3 = MEM_readLE16(istart+4);
+ size_t length4;
+ const BYTE* const istart1 = istart + 6; /* jumpTable */
+ const BYTE* const istart2 = istart1 + length1;
+ const BYTE* const istart3 = istart2 + length2;
+ const BYTE* const istart4 = istart3 + length3;
+ const size_t segmentSize = (dstSize+3) / 4;
+ BYTE* const opStart2 = ostart + segmentSize;
+ BYTE* const opStart3 = opStart2 + segmentSize;
+ BYTE* const opStart4 = opStart3 + segmentSize;
+ BYTE* op1 = ostart;
+ BYTE* op2 = opStart2;
+ BYTE* op3 = opStart3;
+ BYTE* op4 = opStart4;
+ U32 endSignal;
- /* 16-32 symbols per loop (4-8 symbols per stream) */
- endSignal = BITv05_reloadDStream(&bitD1) | BITv05_reloadDStream(&bitD2) | BITv05_reloadDStream(&bitD3) | BITv05_reloadDStream(&bitD4);
- for ( ; (endSignal==BITv05_DStream_unfinished) && (op4<(oend-7)) ; ) {
- HUFv05_DECODE_SYMBOLX2_2(op1, &bitD1);
- HUFv05_DECODE_SYMBOLX2_2(op2, &bitD2);
- HUFv05_DECODE_SYMBOLX2_2(op3, &bitD3);
- HUFv05_DECODE_SYMBOLX2_2(op4, &bitD4);
- HUFv05_DECODE_SYMBOLX2_1(op1, &bitD1);
- HUFv05_DECODE_SYMBOLX2_1(op2, &bitD2);
- HUFv05_DECODE_SYMBOLX2_1(op3, &bitD3);
- HUFv05_DECODE_SYMBOLX2_1(op4, &bitD4);
- HUFv05_DECODE_SYMBOLX2_2(op1, &bitD1);
- HUFv05_DECODE_SYMBOLX2_2(op2, &bitD2);
- HUFv05_DECODE_SYMBOLX2_2(op3, &bitD3);
- HUFv05_DECODE_SYMBOLX2_2(op4, &bitD4);
- HUFv05_DECODE_SYMBOLX2_0(op1, &bitD1);
- HUFv05_DECODE_SYMBOLX2_0(op2, &bitD2);
- HUFv05_DECODE_SYMBOLX2_0(op3, &bitD3);
- HUFv05_DECODE_SYMBOLX2_0(op4, &bitD4);
+ length4 = cSrcSize - (length1 + length2 + length3 + 6);
+ if (length4 > cSrcSize) return ERROR(corruption_detected); /* overflow */
+ errorCode = BITv05_initDStream(&bitD1, istart1, length1);
+ if (HUFv05_isError(errorCode)) return errorCode;
+ errorCode = BITv05_initDStream(&bitD2, istart2, length2);
+ if (HUFv05_isError(errorCode)) return errorCode;
+ errorCode = BITv05_initDStream(&bitD3, istart3, length3);
+ if (HUFv05_isError(errorCode)) return errorCode;
+ errorCode = BITv05_initDStream(&bitD4, istart4, length4);
+ if (HUFv05_isError(errorCode)) return errorCode;
+
+ /* 16-32 symbols per loop (4-8 symbols per stream) */
endSignal = BITv05_reloadDStream(&bitD1) | BITv05_reloadDStream(&bitD2) | BITv05_reloadDStream(&bitD3) | BITv05_reloadDStream(&bitD4);
+ for ( ; (endSignal==BITv05_DStream_unfinished) && (op4<(oend-7)) ; ) {
+ HUFv05_DECODE_SYMBOLX2_2(op1, &bitD1);
+ HUFv05_DECODE_SYMBOLX2_2(op2, &bitD2);
+ HUFv05_DECODE_SYMBOLX2_2(op3, &bitD3);
+ HUFv05_DECODE_SYMBOLX2_2(op4, &bitD4);
+ HUFv05_DECODE_SYMBOLX2_1(op1, &bitD1);
+ HUFv05_DECODE_SYMBOLX2_1(op2, &bitD2);
+ HUFv05_DECODE_SYMBOLX2_1(op3, &bitD3);
+ HUFv05_DECODE_SYMBOLX2_1(op4, &bitD4);
+ HUFv05_DECODE_SYMBOLX2_2(op1, &bitD1);
+ HUFv05_DECODE_SYMBOLX2_2(op2, &bitD2);
+ HUFv05_DECODE_SYMBOLX2_2(op3, &bitD3);
+ HUFv05_DECODE_SYMBOLX2_2(op4, &bitD4);
+ HUFv05_DECODE_SYMBOLX2_0(op1, &bitD1);
+ HUFv05_DECODE_SYMBOLX2_0(op2, &bitD2);
+ HUFv05_DECODE_SYMBOLX2_0(op3, &bitD3);
+ HUFv05_DECODE_SYMBOLX2_0(op4, &bitD4);
+ endSignal = BITv05_reloadDStream(&bitD1) | BITv05_reloadDStream(&bitD2) | BITv05_reloadDStream(&bitD3) | BITv05_reloadDStream(&bitD4);
+ }
+
+ /* check corruption */
+ if (op1 > opStart2) return ERROR(corruption_detected);
+ if (op2 > opStart3) return ERROR(corruption_detected);
+ if (op3 > opStart4) return ERROR(corruption_detected);
+ /* note : op4 supposed already verified within main loop */
+
+ /* finish bitStreams one by one */
+ HUFv05_decodeStreamX2(op1, &bitD1, opStart2, dt, dtLog);
+ HUFv05_decodeStreamX2(op2, &bitD2, opStart3, dt, dtLog);
+ HUFv05_decodeStreamX2(op3, &bitD3, opStart4, dt, dtLog);
+ HUFv05_decodeStreamX2(op4, &bitD4, oend, dt, dtLog);
+
+ /* check */
+ endSignal = BITv05_endOfDStream(&bitD1) & BITv05_endOfDStream(&bitD2) & BITv05_endOfDStream(&bitD3) & BITv05_endOfDStream(&bitD4);
+ if (!endSignal) return ERROR(corruption_detected);
+
+ /* decoded size */
+ return dstSize;
}
-
- /* check corruption */
- if (op1 > opStart2) return ERROR(corruption_detected);
- if (op2 > opStart3) return ERROR(corruption_detected);
- if (op3 > opStart4) return ERROR(corruption_detected);
- /* note : op4 supposed already verified within main loop */
-
- /* finish bitStreams one by one */
- HUFv05_decodeStreamX2(op1, &bitD1, opStart2, dt, dtLog);
- HUFv05_decodeStreamX2(op2, &bitD2, opStart3, dt, dtLog);
- HUFv05_decodeStreamX2(op3, &bitD3, opStart4, dt, dtLog);
- HUFv05_decodeStreamX2(op4, &bitD4, oend, dt, dtLog);
-
- /* check */
- endSignal = BITv05_endOfDStream(&bitD1) & BITv05_endOfDStream(&bitD2) & BITv05_endOfDStream(&bitD3) & BITv05_endOfDStream(&bitD4);
- if (!endSignal) return ERROR(corruption_detected);
-
- /* decoded size */
- return dstSize;
}
@@ -2190,7 +2199,7 @@
}
}
-size_t HUFv05_readDTableX4 (U32* DTable, const void* src, size_t srcSize)
+size_t HUFv05_readDTableX4 (unsigned* DTable, const void* src, size_t srcSize)
{
BYTE weightList[HUFv05_MAX_SYMBOL_VALUE + 1];
sortedSymbol_t sortedSymbol[HUFv05_MAX_SYMBOL_VALUE + 1];
@@ -2204,7 +2213,7 @@
void* dtPtr = DTable;
HUFv05_DEltX4* const dt = ((HUFv05_DEltX4*)dtPtr) + 1;
- HUFv05_STATIC_ASSERT(sizeof(HUFv05_DEltX4) == sizeof(U32)); /* if compilation fails here, assertion is false */
+ HUFv05_STATIC_ASSERT(sizeof(HUFv05_DEltX4) == sizeof(unsigned)); /* if compilation fails here, assertion is false */
if (memLog > HUFv05_ABSOLUTEMAX_TABLELOG) return ERROR(tableLog_tooLarge);
//memset(weightList, 0, sizeof(weightList)); /* is not necessary, even though some analyzer complain ... */
@@ -2331,7 +2340,7 @@
size_t HUFv05_decompress1X4_usingDTable(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
- const U32* DTable)
+ const unsigned* DTable)
{
const BYTE* const istart = (const BYTE*) cSrc;
BYTE* const ostart = (BYTE*) dst;
@@ -2374,7 +2383,7 @@
size_t HUFv05_decompress4X4_usingDTable(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
- const U32* DTable)
+ const unsigned* DTable)
{
if (cSrcSize < 10) return ERROR(corruption_detected); /* strict minimum : jump table + 1 byte per stream */
@@ -2658,6 +2667,7 @@
BYTE headerBuffer[ZSTDv05_frameHeaderSize_max];
}; /* typedef'd to ZSTDv05_DCtx within "zstd_static.h" */
+size_t ZSTDv05_sizeofDCtx (void); /* Hidden declaration */
size_t ZSTDv05_sizeofDCtx (void) { return sizeof(ZSTDv05_DCtx); }
size_t ZSTDv05_decompressBegin(ZSTDv05_DCtx* dctx)
@@ -2822,7 +2832,7 @@
}
-size_t ZSTDv05_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr)
+static size_t ZSTDv05_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr)
{
const BYTE* const in = (const BYTE* const)src;
BYTE headerFlags;
@@ -2845,6 +2855,7 @@
static size_t ZSTDv05_copyRawBlock(void* dst, size_t maxDstSize, const void* src, size_t srcSize)
{
+ if (dst==NULL) return ERROR(dstSize_tooSmall);
if (srcSize > maxDstSize) return ERROR(dstSize_tooSmall);
memcpy(dst, src, srcSize);
return srcSize;
@@ -2853,8 +2864,8 @@
/*! ZSTDv05_decodeLiteralsBlock() :
@return : nb of bytes read from src (< srcSize ) */
-size_t ZSTDv05_decodeLiteralsBlock(ZSTDv05_DCtx* dctx,
- const void* src, size_t srcSize) /* note : srcSize < BLOCKSIZE */
+static size_t ZSTDv05_decodeLiteralsBlock(ZSTDv05_DCtx* dctx,
+ const void* src, size_t srcSize) /* note : srcSize < BLOCKSIZE */
{
const BYTE* const istart = (const BYTE*) src;
@@ -2988,7 +2999,7 @@
}
-size_t ZSTDv05_decodeSeqHeaders(int* nbSeq, const BYTE** dumpsPtr, size_t* dumpsLengthPtr,
+static size_t ZSTDv05_decodeSeqHeaders(int* nbSeq, const BYTE** dumpsPtr, size_t* dumpsLengthPtr,
FSEv05_DTable* DTableLL, FSEv05_DTable* DTableML, FSEv05_DTable* DTableOffb,
const void* src, size_t srcSize, U32 flagStaticTable)
{
@@ -2996,7 +3007,7 @@
const BYTE* ip = istart;
const BYTE* const iend = istart + srcSize;
U32 LLtype, Offtype, MLtype;
- U32 LLlog, Offlog, MLlog;
+ unsigned LLlog, Offlog, MLlog;
size_t dumpsLength;
/* check */
@@ -3054,7 +3065,7 @@
break;
case FSEv05_ENCODING_DYNAMIC :
default : /* impossible */
- { U32 max = MaxLL;
+ { unsigned max = MaxLL;
headerSize = FSEv05_readNCount(norm, &max, &LLlog, ip, iend-ip);
if (FSEv05_isError(headerSize)) return ERROR(GENERIC);
if (LLlog > LLFSEv05Log) return ERROR(corruption_detected);
@@ -3078,7 +3089,7 @@
break;
case FSEv05_ENCODING_DYNAMIC :
default : /* impossible */
- { U32 max = MaxOff;
+ { unsigned max = MaxOff;
headerSize = FSEv05_readNCount(norm, &max, &Offlog, ip, iend-ip);
if (FSEv05_isError(headerSize)) return ERROR(GENERIC);
if (Offlog > OffFSEv05Log) return ERROR(corruption_detected);
@@ -3102,7 +3113,7 @@
break;
case FSEv05_ENCODING_DYNAMIC :
default : /* impossible */
- { U32 max = MaxML;
+ { unsigned max = MaxML;
headerSize = FSEv05_readNCount(norm, &max, &MLlog, ip, iend-ip);
if (FSEv05_isError(headerSize)) return ERROR(GENERIC);
if (MLlog > MLFSEv05Log) return ERROR(corruption_detected);
@@ -3145,14 +3156,13 @@
litLength = FSEv05_peakSymbol(&(seqState->stateLL));
prevOffset = litLength ? seq->offset : seqState->prevOffset;
if (litLength == MaxLL) {
- U32 add = *dumps++;
+ const U32 add = *dumps++;
if (add < 255) litLength += add;
- else {
- litLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no risk : dumps is always followed by seq tables > 1 byte */
+ else if (dumps + 3 <= de) {
+ litLength = MEM_readLE24(dumps);
if (litLength&1) litLength>>=1, dumps += 3;
else litLength = (U16)(litLength)>>1, dumps += 2;
}
- if (dumps > de) { litLength = MaxLL+255; } /* late correction, to avoid using uninitialized memory */
if (dumps >= de) { dumps = de-1; } /* late correction, to avoid read overflow (data is now corrupted anyway) */
}
@@ -3179,14 +3189,13 @@
/* MatchLength */
matchLength = FSEv05_decodeSymbol(&(seqState->stateML), &(seqState->DStream));
if (matchLength == MaxML) {
- U32 add = *dumps++;
+ const U32 add = dumps<de ? *dumps++ : 0;
if (add < 255) matchLength += add;
- else {
- matchLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
+ else if (dumps + 3 <= de) {
+ matchLength = MEM_readLE24(dumps);
if (matchLength&1) matchLength>>=1, dumps += 3;
else matchLength = (U16)(matchLength)>>1, dumps += 2;
}
- if (dumps > de) { matchLength = MaxML+255; } /* late correction, to avoid using uninitialized memory */
if (dumps >= de) { dumps = de-1; } /* late correction, to avoid read overflow (data is now corrupted anyway) */
}
matchLength += MINMATCH;
@@ -3214,7 +3223,7 @@
const BYTE* const base, const BYTE* const vBase, const BYTE* const dictEnd)
{
static const int dec32table[] = { 0, 1, 2, 1, 4, 4, 4, 4 }; /* added */
- static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* substracted */
+ static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* subtracted */
BYTE* const oLitEnd = op + sequence.litLength;
const size_t sequenceLength = sequence.litLength + sequence.matchLength;
BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */
@@ -3297,14 +3306,14 @@
BYTE* const ostart = (BYTE* const)dst;
BYTE* op = ostart;
BYTE* const oend = ostart + maxDstSize;
- size_t errorCode, dumpsLength;
+ size_t errorCode, dumpsLength=0;
const BYTE* litPtr = dctx->litPtr;
const BYTE* const litEnd = litPtr + dctx->litSize;
- int nbSeq;
- const BYTE* dumps;
- U32* DTableLL = dctx->LLTable;
- U32* DTableML = dctx->MLTable;
- U32* DTableOffb = dctx->OffTable;
+ int nbSeq=0;
+ const BYTE* dumps = NULL;
+ unsigned* DTableLL = dctx->LLTable;
+ unsigned* DTableML = dctx->MLTable;
+ unsigned* DTableOffb = dctx->OffTable;
const BYTE* const base = (const BYTE*) (dctx->base);
const BYTE* const vBase = (const BYTE*) (dctx->vBase);
const BYTE* const dictEnd = (const BYTE*) (dctx->dictEnd);
@@ -3410,10 +3419,10 @@
BYTE* const oend = ostart + maxDstSize;
size_t remainingSize = srcSize;
blockProperties_t blockProperties;
+ memset(&blockProperties, 0, sizeof(blockProperties));
/* Frame Header */
- {
- size_t frameHeaderSize;
+ { size_t frameHeaderSize;
if (srcSize < ZSTDv05_frameHeaderSize_min+ZSTDv05_blockHeaderSize) return ERROR(srcSize_wrong);
frameHeaderSize = ZSTDv05_decodeFrameHeader_Part1(dctx, src, ZSTDv05_frameHeaderSize_min);
if (ZSTDv05_isError(frameHeaderSize)) return frameHeaderSize;
@@ -3505,34 +3514,57 @@
#endif
}
-size_t ZSTDv05_findFrameCompressedSize(const void *src, size_t srcSize)
+/* ZSTD_errorFrameSizeInfoLegacy() :
+ assumes `cSize` and `dBound` are _not_ NULL */
+static void ZSTD_errorFrameSizeInfoLegacy(size_t* cSize, unsigned long long* dBound, size_t ret)
+{
+ *cSize = ret;
+ *dBound = ZSTD_CONTENTSIZE_ERROR;
+}
+
+void ZSTDv05_findFrameSizeInfoLegacy(const void *src, size_t srcSize, size_t* cSize, unsigned long long* dBound)
{
const BYTE* ip = (const BYTE*)src;
size_t remainingSize = srcSize;
+ size_t nbBlocks = 0;
blockProperties_t blockProperties;
/* Frame Header */
- if (srcSize < ZSTDv05_frameHeaderSize_min) return ERROR(srcSize_wrong);
- if (MEM_readLE32(src) != ZSTDv05_MAGICNUMBER) return ERROR(prefix_unknown);
+ if (srcSize < ZSTDv05_frameHeaderSize_min) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
+ if (MEM_readLE32(src) != ZSTDv05_MAGICNUMBER) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(prefix_unknown));
+ return;
+ }
ip += ZSTDv05_frameHeaderSize_min; remainingSize -= ZSTDv05_frameHeaderSize_min;
/* Loop on each block */
while (1)
{
size_t cBlockSize = ZSTDv05_getcBlockSize(ip, remainingSize, &blockProperties);
- if (ZSTDv05_isError(cBlockSize)) return cBlockSize;
+ if (ZSTDv05_isError(cBlockSize)) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, cBlockSize);
+ return;
+ }
ip += ZSTDv05_blockHeaderSize;
remainingSize -= ZSTDv05_blockHeaderSize;
- if (cBlockSize > remainingSize) return ERROR(srcSize_wrong);
+ if (cBlockSize > remainingSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
if (cBlockSize == 0) break; /* bt_end */
ip += cBlockSize;
remainingSize -= cBlockSize;
+ nbBlocks++;
}
- return ip - (const BYTE*)src;
+ *cSize = ip - (const BYTE*)src;
+ *dBound = nbBlocks * BLOCKSIZE;
}
/* ******************************
@@ -3630,7 +3662,7 @@
{
size_t hSize, offcodeHeaderSize, matchlengthHeaderSize, errorCode, litlengthHeaderSize;
short offcodeNCount[MaxOff+1];
- U32 offcodeMaxValue=MaxOff, offcodeLog;
+ unsigned offcodeMaxValue=MaxOff, offcodeLog;
short matchlengthNCount[MaxML+1];
unsigned matchlengthMaxValue = MaxML, matchlengthLog;
short litlengthNCount[MaxLL+1];
diff --git a/vendor/github.com/DataDog/zstd/zstd_v05.h b/vendor/github.com/DataDog/zstd/zstd_v05.h
index b68fd57..4a97985 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v05.h
+++ b/vendor/github.com/DataDog/zstd/zstd_v05.h
@@ -33,13 +33,18 @@
size_t ZSTDv05_decompress( void* dst, size_t dstCapacity,
const void* src, size_t compressedSize);
-/**
-ZSTDv05_getFrameSrcSize() : get the source length of a ZSTD frame
- compressedSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
- return : the number of bytes that would be read to decompress this frame
- or an errorCode if it fails (which can be tested using ZSTDv05_isError())
-*/
-size_t ZSTDv05_findFrameCompressedSize(const void* src, size_t compressedSize);
+ /**
+ ZSTDv05_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.5.x format
+ srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
+ cSize (output parameter) : the number of bytes that would be read to decompress this frame
+ or an error code if it fails (which can be tested using ZSTDv01_isError())
+ dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
+ or ZSTD_CONTENTSIZE_ERROR if an error occurs
+
+ note : assumes `cSize` and `dBound` are _not_ NULL.
+ */
+void ZSTDv05_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
+ size_t* cSize, unsigned long long* dBound);
/* *************************************
* Helper functions
diff --git a/vendor/github.com/DataDog/zstd/zstd_v06.c b/vendor/github.com/DataDog/zstd/zstd_v06.c
index 8b068b3..f907a3a 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v06.c
+++ b/vendor/github.com/DataDog/zstd/zstd_v06.c
@@ -506,6 +506,8 @@
#define FSEv06_ENCODING_STATIC 2
#define FSEv06_ENCODING_DYNAMIC 3
+#define ZSTD_CONTENTSIZE_ERROR (0ULL - 2)
+
static const U32 LL_bits[MaxLL+1] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
1, 1, 1, 1, 2, 2, 3, 3, 4, 6, 7, 8, 9,10,11,12,
13,14,15,16 };
@@ -1250,9 +1252,7 @@
/* **************************************************************
* HUF Error Management
****************************************************************/
-unsigned HUFv06_isError(size_t code) { return ERR_isError(code); }
-
-const char* HUFv06_getErrorName(size_t code) { return ERR_getErrorName(code); }
+static unsigned HUFv06_isError(size_t code) { return ERR_isError(code); }
/*-**************************************************************
@@ -2823,7 +2823,8 @@
BYTE headerBuffer[ZSTDv06_FRAMEHEADERSIZE_MAX];
}; /* typedef'd to ZSTDv06_DCtx within "zstd_static.h" */
-size_t ZSTDv06_sizeofDCtx (void) { return sizeof(ZSTDv06_DCtx); } /* non published interface */
+size_t ZSTDv06_sizeofDCtx (void); /* Hidden declaration */
+size_t ZSTDv06_sizeofDCtx (void) { return sizeof(ZSTDv06_DCtx); }
size_t ZSTDv06_decompressBegin(ZSTDv06_DCtx* dctx)
{
@@ -3022,7 +3023,7 @@
/*! ZSTDv06_getcBlockSize() :
* Provides the size of compressed block from block header `src` */
-size_t ZSTDv06_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr)
+static size_t ZSTDv06_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr)
{
const BYTE* const in = (const BYTE* const)src;
U32 cSize;
@@ -3041,6 +3042,7 @@
static size_t ZSTDv06_copyRawBlock(void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
+ if (dst==NULL) return ERROR(dstSize_tooSmall);
if (srcSize > dstCapacity) return ERROR(dstSize_tooSmall);
memcpy(dst, src, srcSize);
return srcSize;
@@ -3049,7 +3051,7 @@
/*! ZSTDv06_decodeLiteralsBlock() :
@return : nb of bytes read from src (< srcSize ) */
-size_t ZSTDv06_decodeLiteralsBlock(ZSTDv06_DCtx* dctx,
+static size_t ZSTDv06_decodeLiteralsBlock(ZSTDv06_DCtx* dctx,
const void* src, size_t srcSize) /* note : srcSize < BLOCKSIZE */
{
const BYTE* const istart = (const BYTE*) src;
@@ -3183,7 +3185,7 @@
@return : nb bytes read from src,
or an error code if it fails, testable with ZSTDv06_isError()
*/
-size_t ZSTDv06_buildSeqTable(FSEv06_DTable* DTable, U32 type, U32 max, U32 maxLog,
+static size_t ZSTDv06_buildSeqTable(FSEv06_DTable* DTable, U32 type, U32 max, U32 maxLog,
const void* src, size_t srcSize,
const S16* defaultNorm, U32 defaultLog, U32 flagRepeatTable)
{
@@ -3213,7 +3215,7 @@
}
-size_t ZSTDv06_decodeSeqHeaders(int* nbSeqPtr,
+static size_t ZSTDv06_decodeSeqHeaders(int* nbSeqPtr,
FSEv06_DTable* DTableLL, FSEv06_DTable* DTableML, FSEv06_DTable* DTableOffb, U32 flagRepeatTable,
const void* src, size_t srcSize)
{
@@ -3240,14 +3242,12 @@
}
/* FSE table descriptors */
+ if (ip + 4 > iend) return ERROR(srcSize_wrong); /* min : header byte + all 3 are "raw", hence no header, but at least xxLog bits per type */
{ U32 const LLtype = *ip >> 6;
U32 const Offtype = (*ip >> 4) & 3;
U32 const MLtype = (*ip >> 2) & 3;
ip++;
- /* check */
- if (ip > iend-3) return ERROR(srcSize_wrong); /* min : all 3 are "raw", hence no header, but at least xxLog bits per type */
-
/* Build DTables */
{ size_t const bhSize = ZSTDv06_buildSeqTable(DTableLL, LLtype, MaxLL, LLFSELog, ip, iend-ip, LL_defaultNorm, LL_defaultNormLog, flagRepeatTable);
if (ZSTDv06_isError(bhSize)) return ERROR(corruption_detected);
@@ -3358,7 +3358,7 @@
}
-size_t ZSTDv06_execSequence(BYTE* op,
+static size_t ZSTDv06_execSequence(BYTE* op,
BYTE* const oend, seq_t sequence,
const BYTE** litPtr, const BYTE* const litLimit,
const BYTE* const base, const BYTE* const vBase, const BYTE* const dictEnd)
@@ -3406,7 +3406,7 @@
if (sequence.offset < 8) {
/* close range match, overlap */
static const U32 dec32table[] = { 0, 1, 2, 1, 4, 4, 4, 4 }; /* added */
- static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* substracted */
+ static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* subtracted */
int const sub2 = dec64table[sequence.offset];
op[0] = match[0];
op[1] = match[1];
@@ -3654,36 +3654,62 @@
#endif
}
-size_t ZSTDv06_findFrameCompressedSize(const void* src, size_t srcSize)
+/* ZSTD_errorFrameSizeInfoLegacy() :
+ assumes `cSize` and `dBound` are _not_ NULL */
+static void ZSTD_errorFrameSizeInfoLegacy(size_t* cSize, unsigned long long* dBound, size_t ret)
+{
+ *cSize = ret;
+ *dBound = ZSTD_CONTENTSIZE_ERROR;
+}
+
+void ZSTDv06_findFrameSizeInfoLegacy(const void *src, size_t srcSize, size_t* cSize, unsigned long long* dBound)
{
const BYTE* ip = (const BYTE*)src;
size_t remainingSize = srcSize;
+ size_t nbBlocks = 0;
blockProperties_t blockProperties = { bt_compressed, 0 };
/* Frame Header */
- { size_t const frameHeaderSize = ZSTDv06_frameHeaderSize(src, ZSTDv06_frameHeaderSize_min);
- if (ZSTDv06_isError(frameHeaderSize)) return frameHeaderSize;
- if (MEM_readLE32(src) != ZSTDv06_MAGICNUMBER) return ERROR(prefix_unknown);
- if (srcSize < frameHeaderSize+ZSTDv06_blockHeaderSize) return ERROR(srcSize_wrong);
+ { size_t const frameHeaderSize = ZSTDv06_frameHeaderSize(src, srcSize);
+ if (ZSTDv06_isError(frameHeaderSize)) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, frameHeaderSize);
+ return;
+ }
+ if (MEM_readLE32(src) != ZSTDv06_MAGICNUMBER) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(prefix_unknown));
+ return;
+ }
+ if (srcSize < frameHeaderSize+ZSTDv06_blockHeaderSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
ip += frameHeaderSize; remainingSize -= frameHeaderSize;
}
/* Loop on each block */
while (1) {
size_t const cBlockSize = ZSTDv06_getcBlockSize(ip, remainingSize, &blockProperties);
- if (ZSTDv06_isError(cBlockSize)) return cBlockSize;
+ if (ZSTDv06_isError(cBlockSize)) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, cBlockSize);
+ return;
+ }
ip += ZSTDv06_blockHeaderSize;
remainingSize -= ZSTDv06_blockHeaderSize;
- if (cBlockSize > remainingSize) return ERROR(srcSize_wrong);
+ if (cBlockSize > remainingSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
if (cBlockSize == 0) break; /* bt_end */
ip += cBlockSize;
remainingSize -= cBlockSize;
+ nbBlocks++;
}
- return ip - (const BYTE*)src;
+ *cSize = ip - (const BYTE*)src;
+ *dBound = nbBlocks * ZSTDv06_BLOCKSIZE_MAX;
}
/*_******************************
@@ -4006,7 +4032,7 @@
if (ZSTDv06_isError(hSize)) return hSize;
if (toLoad > (size_t)(iend-ip)) { /* not enough input to load full header */
memcpy(zbd->headerBuffer + zbd->lhSize, ip, iend-ip);
- zbd->lhSize += iend-ip; ip = iend; notDone = 0;
+ zbd->lhSize += iend-ip;
*dstCapacityPtr = 0;
return (hSize - zbd->lhSize) + ZSTDv06_blockHeaderSize; /* remaining header bytes + next block header */
}
diff --git a/vendor/github.com/DataDog/zstd/zstd_v06.h b/vendor/github.com/DataDog/zstd/zstd_v06.h
index fb4eb37..0781857 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v06.h
+++ b/vendor/github.com/DataDog/zstd/zstd_v06.h
@@ -43,12 +43,17 @@
const void* src, size_t compressedSize);
/**
-ZSTDv06_getFrameSrcSize() : get the source length of a ZSTD frame
- compressedSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
- return : the number of bytes that would be read to decompress this frame
- or an errorCode if it fails (which can be tested using ZSTDv06_isError())
+ZSTDv06_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.6.x format
+ srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
+ cSize (output parameter) : the number of bytes that would be read to decompress this frame
+ or an error code if it fails (which can be tested using ZSTDv01_isError())
+ dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
+ or ZSTD_CONTENTSIZE_ERROR if an error occurs
+
+ note : assumes `cSize` and `dBound` are _not_ NULL.
*/
-size_t ZSTDv06_findFrameCompressedSize(const void* src, size_t compressedSize);
+void ZSTDv06_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
+ size_t* cSize, unsigned long long* dBound);
/* *************************************
* Helper functions
diff --git a/vendor/github.com/DataDog/zstd/zstd_v07.c b/vendor/github.com/DataDog/zstd/zstd_v07.c
index 70b170f..a83ddc9 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v07.c
+++ b/vendor/github.com/DataDog/zstd/zstd_v07.c
@@ -2628,7 +2628,7 @@
-void* ZSTDv07_defaultAllocFunction(void* opaque, size_t size)
+static void* ZSTDv07_defaultAllocFunction(void* opaque, size_t size)
{
void* address = malloc(size);
(void)opaque;
@@ -2636,7 +2636,7 @@
return address;
}
-void ZSTDv07_defaultFreeFunction(void* opaque, void* address)
+static void ZSTDv07_defaultFreeFunction(void* opaque, void* address)
{
(void)opaque;
/* if (address) printf("free %p opaque=%p \n", address, opaque); */
@@ -2740,6 +2740,8 @@
#define FSEv07_ENCODING_STATIC 2
#define FSEv07_ENCODING_DYNAMIC 3
+#define ZSTD_CONTENTSIZE_ERROR (0ULL - 2)
+
static const U32 LL_bits[MaxLL+1] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
1, 1, 1, 1, 2, 2, 3, 3, 4, 6, 7, 8, 9,10,11,12,
13,14,15,16 };
@@ -3150,10 +3152,10 @@
const BYTE* ip = (const BYTE*)src;
if (srcSize < ZSTDv07_frameHeaderSize_min) return ZSTDv07_frameHeaderSize_min;
+ memset(fparamsPtr, 0, sizeof(*fparamsPtr));
if (MEM_readLE32(src) != ZSTDv07_MAGICNUMBER) {
if ((MEM_readLE32(src) & 0xFFFFFFF0U) == ZSTDv07_MAGIC_SKIPPABLE_START) {
if (srcSize < ZSTDv07_skippableHeaderSize) return ZSTDv07_skippableHeaderSize; /* magic number + skippable frame length */
- memset(fparamsPtr, 0, sizeof(*fparamsPtr));
fparamsPtr->frameContentSize = MEM_readLE32((const char *)src + 4);
fparamsPtr->windowSize = 0; /* windowSize==0 means a frame is skippable */
return 0;
@@ -3175,11 +3177,13 @@
U32 windowSize = 0;
U32 dictID = 0;
U64 frameContentSize = 0;
- if ((fhdByte & 0x08) != 0) return ERROR(frameParameter_unsupported); /* reserved bits, which must be zero */
+ if ((fhdByte & 0x08) != 0) /* reserved bits, which must be zero */
+ return ERROR(frameParameter_unsupported);
if (!directMode) {
BYTE const wlByte = ip[pos++];
U32 const windowLog = (wlByte >> 3) + ZSTDv07_WINDOWLOG_ABSOLUTEMIN;
- if (windowLog > ZSTDv07_WINDOWLOG_MAX) return ERROR(frameParameter_unsupported);
+ if (windowLog > ZSTDv07_WINDOWLOG_MAX)
+ return ERROR(frameParameter_unsupported);
windowSize = (1U << windowLog);
windowSize += (windowSize >> 3) * (wlByte&7);
}
@@ -3201,7 +3205,8 @@
case 3 : frameContentSize = MEM_readLE64(ip+pos); break;
}
if (!windowSize) windowSize = (U32)frameContentSize;
- if (windowSize > windowSizeMax) return ERROR(frameParameter_unsupported);
+ if (windowSize > windowSizeMax)
+ return ERROR(frameParameter_unsupported);
fparamsPtr->frameContentSize = frameContentSize;
fparamsPtr->windowSize = windowSize;
fparamsPtr->dictID = dictID;
@@ -3220,11 +3225,10 @@
- frame header not completely provided (`srcSize` too small) */
unsigned long long ZSTDv07_getDecompressedSize(const void* src, size_t srcSize)
{
- { ZSTDv07_frameParams fparams;
- size_t const frResult = ZSTDv07_getFrameParams(&fparams, src, srcSize);
- if (frResult!=0) return 0;
- return fparams.frameContentSize;
- }
+ ZSTDv07_frameParams fparams;
+ size_t const frResult = ZSTDv07_getFrameParams(&fparams, src, srcSize);
+ if (frResult!=0) return 0;
+ return fparams.frameContentSize;
}
@@ -3248,7 +3252,7 @@
/*! ZSTDv07_getcBlockSize() :
* Provides the size of compressed block from block header `src` */
-size_t ZSTDv07_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr)
+static size_t ZSTDv07_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr)
{
const BYTE* const in = (const BYTE* const)src;
U32 cSize;
@@ -3275,7 +3279,7 @@
/*! ZSTDv07_decodeLiteralsBlock() :
@return : nb of bytes read from src (< srcSize ) */
-size_t ZSTDv07_decodeLiteralsBlock(ZSTDv07_DCtx* dctx,
+static size_t ZSTDv07_decodeLiteralsBlock(ZSTDv07_DCtx* dctx,
const void* src, size_t srcSize) /* note : srcSize < BLOCKSIZE */
{
const BYTE* const istart = (const BYTE*) src;
@@ -3409,7 +3413,7 @@
@return : nb bytes read from src,
or an error code if it fails, testable with ZSTDv07_isError()
*/
-size_t ZSTDv07_buildSeqTable(FSEv07_DTable* DTable, U32 type, U32 max, U32 maxLog,
+static size_t ZSTDv07_buildSeqTable(FSEv07_DTable* DTable, U32 type, U32 max, U32 maxLog,
const void* src, size_t srcSize,
const S16* defaultNorm, U32 defaultLog, U32 flagRepeatTable)
{
@@ -3439,7 +3443,7 @@
}
-size_t ZSTDv07_decodeSeqHeaders(int* nbSeqPtr,
+static size_t ZSTDv07_decodeSeqHeaders(int* nbSeqPtr,
FSEv07_DTable* DTableLL, FSEv07_DTable* DTableML, FSEv07_DTable* DTableOffb, U32 flagRepeatTable,
const void* src, size_t srcSize)
{
@@ -3466,14 +3470,12 @@
}
/* FSE table descriptors */
+ if (ip + 4 > iend) return ERROR(srcSize_wrong); /* min : header byte + all 3 are "raw", hence no header, but at least xxLog bits per type */
{ U32 const LLtype = *ip >> 6;
U32 const OFtype = (*ip >> 4) & 3;
U32 const MLtype = (*ip >> 2) & 3;
ip++;
- /* check */
- if (ip > iend-3) return ERROR(srcSize_wrong); /* min : all 3 are "raw", hence no header, but at least xxLog bits per type */
-
/* Build DTables */
{ size_t const llhSize = ZSTDv07_buildSeqTable(DTableLL, LLtype, MaxLL, LLFSELog, ip, iend-ip, LL_defaultNorm, LL_defaultNormLog, flagRepeatTable);
if (ZSTDv07_isError(llhSize)) return ERROR(corruption_detected);
@@ -3629,7 +3631,7 @@
if (sequence.offset < 8) {
/* close range match, overlap */
static const U32 dec32table[] = { 0, 1, 2, 1, 4, 4, 4, 4 }; /* added */
- static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* substracted */
+ static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* subtracted */
int const sub2 = dec64table[sequence.offset];
op[0] = match[0];
op[1] = match[1];
@@ -3771,7 +3773,7 @@
}
-size_t ZSTDv07_generateNxBytes(void* dst, size_t dstCapacity, BYTE byte, size_t length)
+static size_t ZSTDv07_generateNxBytes(void* dst, size_t dstCapacity, BYTE byte, size_t length)
{
if (length > dstCapacity) return ERROR(dstSize_tooSmall);
memset(dst, byte, length);
@@ -3851,7 +3853,7 @@
* It avoids reloading the dictionary each time.
* `preparedDCtx` must have been properly initialized using ZSTDv07_decompressBegin_usingDict().
* Requires 2 contexts : 1 for reference (preparedDCtx), which will not be modified, and 1 to run the decompression operation (dctx) */
-size_t ZSTDv07_decompress_usingPreparedDCtx(ZSTDv07_DCtx* dctx, const ZSTDv07_DCtx* refDCtx,
+static size_t ZSTDv07_decompress_usingPreparedDCtx(ZSTDv07_DCtx* dctx, const ZSTDv07_DCtx* refDCtx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize)
{
@@ -3893,19 +3895,40 @@
#endif
}
-size_t ZSTDv07_findFrameCompressedSize(const void* src, size_t srcSize)
+/* ZSTD_errorFrameSizeInfoLegacy() :
+ assumes `cSize` and `dBound` are _not_ NULL */
+static void ZSTD_errorFrameSizeInfoLegacy(size_t* cSize, unsigned long long* dBound, size_t ret)
+{
+ *cSize = ret;
+ *dBound = ZSTD_CONTENTSIZE_ERROR;
+}
+
+void ZSTDv07_findFrameSizeInfoLegacy(const void *src, size_t srcSize, size_t* cSize, unsigned long long* dBound)
{
const BYTE* ip = (const BYTE*)src;
size_t remainingSize = srcSize;
+ size_t nbBlocks = 0;
/* check */
- if (srcSize < ZSTDv07_frameHeaderSize_min+ZSTDv07_blockHeaderSize) return ERROR(srcSize_wrong);
+ if (srcSize < ZSTDv07_frameHeaderSize_min+ZSTDv07_blockHeaderSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
/* Frame Header */
- { size_t const frameHeaderSize = ZSTDv07_frameHeaderSize(src, ZSTDv07_frameHeaderSize_min);
- if (ZSTDv07_isError(frameHeaderSize)) return frameHeaderSize;
- if (MEM_readLE32(src) != ZSTDv07_MAGICNUMBER) return ERROR(prefix_unknown);
- if (srcSize < frameHeaderSize+ZSTDv07_blockHeaderSize) return ERROR(srcSize_wrong);
+ { size_t const frameHeaderSize = ZSTDv07_frameHeaderSize(src, srcSize);
+ if (ZSTDv07_isError(frameHeaderSize)) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, frameHeaderSize);
+ return;
+ }
+ if (MEM_readLE32(src) != ZSTDv07_MAGICNUMBER) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(prefix_unknown));
+ return;
+ }
+ if (srcSize < frameHeaderSize+ZSTDv07_blockHeaderSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
ip += frameHeaderSize; remainingSize -= frameHeaderSize;
}
@@ -3913,20 +3936,28 @@
while (1) {
blockProperties_t blockProperties;
size_t const cBlockSize = ZSTDv07_getcBlockSize(ip, remainingSize, &blockProperties);
- if (ZSTDv07_isError(cBlockSize)) return cBlockSize;
+ if (ZSTDv07_isError(cBlockSize)) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, cBlockSize);
+ return;
+ }
ip += ZSTDv07_blockHeaderSize;
remainingSize -= ZSTDv07_blockHeaderSize;
if (blockProperties.blockType == bt_end) break;
- if (cBlockSize > remainingSize) return ERROR(srcSize_wrong);
+ if (cBlockSize > remainingSize) {
+ ZSTD_errorFrameSizeInfoLegacy(cSize, dBound, ERROR(srcSize_wrong));
+ return;
+ }
ip += cBlockSize;
remainingSize -= cBlockSize;
+ nbBlocks++;
}
- return ip - (const BYTE*)src;
+ *cSize = ip - (const BYTE*)src;
+ *dBound = nbBlocks * ZSTDv07_BLOCKSIZE_ABSOLUTEMAX;
}
/*_******************************
@@ -4146,7 +4177,7 @@
ZSTDv07_DCtx* refContext;
}; /* typedef'd tp ZSTDv07_CDict within zstd.h */
-ZSTDv07_DDict* ZSTDv07_createDDict_advanced(const void* dict, size_t dictSize, ZSTDv07_customMem customMem)
+static ZSTDv07_DDict* ZSTDv07_createDDict_advanced(const void* dict, size_t dictSize, ZSTDv07_customMem customMem)
{
if (!customMem.customAlloc && !customMem.customFree)
customMem = defaultCustomMem;
diff --git a/vendor/github.com/DataDog/zstd/zstd_v07.h b/vendor/github.com/DataDog/zstd/zstd_v07.h
index 6591cd3..a566c1d 100644
--- a/vendor/github.com/DataDog/zstd/zstd_v07.h
+++ b/vendor/github.com/DataDog/zstd/zstd_v07.h
@@ -50,12 +50,17 @@
const void* src, size_t compressedSize);
/**
-ZSTDv07_getFrameSrcSize() : get the source length of a ZSTD frame
- compressedSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
- return : the number of bytes that would be read to decompress this frame
- or an errorCode if it fails (which can be tested using ZSTDv07_isError())
+ZSTDv07_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.7.x format
+ srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
+ cSize (output parameter) : the number of bytes that would be read to decompress this frame
+ or an error code if it fails (which can be tested using ZSTDv01_isError())
+ dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
+ or ZSTD_CONTENTSIZE_ERROR if an error occurs
+
+ note : assumes `cSize` and `dBound` are _not_ NULL.
*/
-size_t ZSTDv07_findFrameCompressedSize(const void* src, size_t compressedSize);
+void ZSTDv07_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
+ size_t* cSize, unsigned long long* dBound);
/*====== Helper functions ======*/
ZSTDLIBv07_API unsigned ZSTDv07_isError(size_t code); /*!< tells if a `size_t` function result is an error code */
diff --git a/vendor/github.com/DataDog/zstd/zstdmt_compress.c b/vendor/github.com/DataDog/zstd/zstdmt_compress.c
index c7a205d..9e537b8 100644
--- a/vendor/github.com/DataDog/zstd/zstdmt_compress.c
+++ b/vendor/github.com/DataDog/zstd/zstdmt_compress.c
@@ -9,21 +9,20 @@
*/
-/* ====== Tuning parameters ====== */
-#define ZSTDMT_NBWORKERS_MAX 200
-#define ZSTDMT_JOBSIZE_MAX (MEM_32bits() ? (512 MB) : (2 GB)) /* note : limited by `jobSize` type, which is `unsigned` */
-#define ZSTDMT_OVERLAPLOG_DEFAULT 6
-
-
/* ====== Compiler specifics ====== */
#if defined(_MSC_VER)
# pragma warning(disable : 4204) /* disable: C4204: non-constant aggregate initializer */
#endif
+/* ====== Constants ====== */
+#define ZSTDMT_OVERLAPLOG_DEFAULT 0
+
+
/* ====== Dependencies ====== */
#include <string.h> /* memcpy, memset */
-#include <limits.h> /* INT_MAX */
+#include <limits.h> /* INT_MAX, UINT_MAX */
+#include "mem.h" /* MEM_STATIC */
#include "pool.h" /* threadpool */
#include "threading.h" /* mutex */
#include "zstd_compress_internal.h" /* MIN, ERROR, ZSTD_*, ZSTD_highbit32 */
@@ -37,18 +36,19 @@
#define ZSTD_RESIZE_SEQPOOL 0
/* ====== Debug ====== */
-#if defined(ZSTD_DEBUG) && (ZSTD_DEBUG>=2)
+#if defined(DEBUGLEVEL) && (DEBUGLEVEL>=2) \
+ && !defined(_MSC_VER) \
+ && !defined(__MINGW32__)
# include <stdio.h>
# include <unistd.h>
# include <sys/times.h>
-# define DEBUGLOGRAW(l, ...) if (l<=ZSTD_DEBUG) { fprintf(stderr, __VA_ARGS__); }
# define DEBUG_PRINTHEX(l,p,n) { \
unsigned debug_u; \
for (debug_u=0; debug_u<(n); debug_u++) \
- DEBUGLOGRAW(l, "%02X ", ((const unsigned char*)(p))[debug_u]); \
- DEBUGLOGRAW(l, " \n"); \
+ RAWLOG(l, "%02X ", ((const unsigned char*)(p))[debug_u]); \
+ RAWLOG(l, " \n"); \
}
static unsigned long long GetCurrentClockTimeMicroseconds(void)
@@ -56,13 +56,13 @@
static clock_t _ticksPerSecond = 0;
if (_ticksPerSecond <= 0) _ticksPerSecond = sysconf(_SC_CLK_TCK);
- { struct tms junk; clock_t newTicks = (clock_t) times(&junk);
- return ((((unsigned long long)newTicks)*(1000000))/_ticksPerSecond); }
-}
+ { struct tms junk; clock_t newTicks = (clock_t) times(&junk);
+ return ((((unsigned long long)newTicks)*(1000000))/_ticksPerSecond);
+} }
#define MUTEX_WAIT_TIME_DLEVEL 6
#define ZSTD_PTHREAD_MUTEX_LOCK(mutex) { \
- if (ZSTD_DEBUG >= MUTEX_WAIT_TIME_DLEVEL) { \
+ if (DEBUGLEVEL >= MUTEX_WAIT_TIME_DLEVEL) { \
unsigned long long const beforeTime = GetCurrentClockTimeMicroseconds(); \
ZSTD_pthread_mutex_lock(mutex); \
{ unsigned long long const afterTime = GetCurrentClockTimeMicroseconds(); \
@@ -160,6 +160,25 @@
ZSTD_pthread_mutex_unlock(&bufPool->poolMutex);
}
+
+static ZSTDMT_bufferPool* ZSTDMT_expandBufferPool(ZSTDMT_bufferPool* srcBufPool, U32 nbWorkers)
+{
+ unsigned const maxNbBuffers = 2*nbWorkers + 3;
+ if (srcBufPool==NULL) return NULL;
+ if (srcBufPool->totalBuffers >= maxNbBuffers) /* good enough */
+ return srcBufPool;
+ /* need a larger buffer pool */
+ { ZSTD_customMem const cMem = srcBufPool->cMem;
+ size_t const bSize = srcBufPool->bufferSize; /* forward parameters */
+ ZSTDMT_bufferPool* newBufPool;
+ ZSTDMT_freeBufferPool(srcBufPool);
+ newBufPool = ZSTDMT_createBufferPool(nbWorkers, cMem);
+ if (newBufPool==NULL) return newBufPool;
+ ZSTDMT_setBufferSize(newBufPool, bSize);
+ return newBufPool;
+ }
+}
+
/** ZSTDMT_getBuffer() :
* assumption : bufPool must be valid
* @return : a buffer, with start pointer and size
@@ -229,8 +248,8 @@
/* store buffer for later re-use, up to pool capacity */
static void ZSTDMT_releaseBuffer(ZSTDMT_bufferPool* bufPool, buffer_t buf)
{
- if (buf.start == NULL) return; /* compatible with release on NULL */
DEBUGLOG(5, "ZSTDMT_releaseBuffer");
+ if (buf.start == NULL) return; /* compatible with release on NULL */
ZSTD_pthread_mutex_lock(&bufPool->poolMutex);
if (bufPool->nbBuffers < bufPool->totalBuffers) {
bufPool->bTable[bufPool->nbBuffers++] = buf; /* stored for later use */
@@ -300,7 +319,8 @@
static ZSTDMT_seqPool* ZSTDMT_createSeqPool(unsigned nbWorkers, ZSTD_customMem cMem)
{
- ZSTDMT_seqPool* seqPool = ZSTDMT_createBufferPool(nbWorkers, cMem);
+ ZSTDMT_seqPool* const seqPool = ZSTDMT_createBufferPool(nbWorkers, cMem);
+ if (seqPool == NULL) return NULL;
ZSTDMT_setNbSeq(seqPool, 0);
return seqPool;
}
@@ -310,6 +330,10 @@
ZSTDMT_freeBufferPool(seqPool);
}
+static ZSTDMT_seqPool* ZSTDMT_expandSeqPool(ZSTDMT_seqPool* pool, U32 nbWorkers)
+{
+ return ZSTDMT_expandBufferPool(pool, nbWorkers);
+}
/* ===== CCtx Pool ===== */
@@ -317,8 +341,8 @@
typedef struct {
ZSTD_pthread_mutex_t poolMutex;
- unsigned totalCCtx;
- unsigned availCCtx;
+ int totalCCtx;
+ int availCCtx;
ZSTD_customMem cMem;
ZSTD_CCtx* cctx[1]; /* variable size */
} ZSTDMT_CCtxPool;
@@ -326,16 +350,16 @@
/* note : all CCtx borrowed from the pool should be released back to the pool _before_ freeing the pool */
static void ZSTDMT_freeCCtxPool(ZSTDMT_CCtxPool* pool)
{
- unsigned u;
- for (u=0; u<pool->totalCCtx; u++)
- ZSTD_freeCCtx(pool->cctx[u]); /* note : compatible with free on NULL */
+ int cid;
+ for (cid=0; cid<pool->totalCCtx; cid++)
+ ZSTD_freeCCtx(pool->cctx[cid]); /* note : compatible with free on NULL */
ZSTD_pthread_mutex_destroy(&pool->poolMutex);
ZSTD_free(pool, pool->cMem);
}
/* ZSTDMT_createCCtxPool() :
* implies nbWorkers >= 1 , checked by caller ZSTDMT_createCCtx() */
-static ZSTDMT_CCtxPool* ZSTDMT_createCCtxPool(unsigned nbWorkers,
+static ZSTDMT_CCtxPool* ZSTDMT_createCCtxPool(int nbWorkers,
ZSTD_customMem cMem)
{
ZSTDMT_CCtxPool* const cctxPool = (ZSTDMT_CCtxPool*) ZSTD_calloc(
@@ -355,6 +379,18 @@
return cctxPool;
}
+static ZSTDMT_CCtxPool* ZSTDMT_expandCCtxPool(ZSTDMT_CCtxPool* srcPool,
+ int nbWorkers)
+{
+ if (srcPool==NULL) return NULL;
+ if (nbWorkers <= srcPool->totalCCtx) return srcPool; /* good enough */
+ /* need a larger cctx pool */
+ { ZSTD_customMem const cMem = srcPool->cMem;
+ ZSTDMT_freeCCtxPool(srcPool);
+ return ZSTDMT_createCCtxPool(nbWorkers, cMem);
+ }
+}
+
/* only works during initialization phase, not during compression */
static size_t ZSTDMT_sizeof_CCtxPool(ZSTDMT_CCtxPool* cctxPool)
{
@@ -421,21 +457,20 @@
* Must be acquired after the main mutex when acquiring both.
*/
ZSTD_pthread_mutex_t ldmWindowMutex;
- ZSTD_pthread_cond_t ldmWindowCond; /* Signaled when ldmWindow is udpated */
+ ZSTD_pthread_cond_t ldmWindowCond; /* Signaled when ldmWindow is updated */
ZSTD_window_t ldmWindow; /* A thread-safe copy of ldmState.window */
} serialState_t;
-static int ZSTDMT_serialState_reset(serialState_t* serialState, ZSTDMT_seqPool* seqPool, ZSTD_CCtx_params params)
+static int ZSTDMT_serialState_reset(serialState_t* serialState, ZSTDMT_seqPool* seqPool, ZSTD_CCtx_params params, size_t jobSize)
{
/* Adjust parameters */
if (params.ldmParams.enableLdm) {
DEBUGLOG(4, "LDM window size = %u KB", (1U << params.cParams.windowLog) >> 10);
- params.ldmParams.windowLog = params.cParams.windowLog;
ZSTD_ldm_adjustParameters(¶ms.ldmParams, ¶ms.cParams);
assert(params.ldmParams.hashLog >= params.ldmParams.bucketSizeLog);
- assert(params.ldmParams.hashEveryLog < 32);
+ assert(params.ldmParams.hashRateLog < 32);
serialState->ldmState.hashPower =
- ZSTD_ldm_getHashPower(params.ldmParams.minMatchLength);
+ ZSTD_rollingHash_primePower(params.ldmParams.minMatchLength);
} else {
memset(¶ms.ldmParams, 0, sizeof(params.ldmParams));
}
@@ -453,7 +488,7 @@
serialState->params.ldmParams.hashLog -
serialState->params.ldmParams.bucketSizeLog;
/* Size the seq pool tables */
- ZSTDMT_setNbSeq(seqPool, ZSTD_ldm_getMaxNbSeq(params.ldmParams, params.jobSize));
+ ZSTDMT_setNbSeq(seqPool, ZSTD_ldm_getMaxNbSeq(params.ldmParams, jobSize));
/* Reset the window */
ZSTD_window_clear(&serialState->ldmState.window);
serialState->ldmWindow = serialState->ldmState.window;
@@ -473,6 +508,7 @@
memset(serialState->ldmState.bucketOffsets, 0, bucketSize);
}
serialState->params = params;
+ serialState->params.jobSize = (U32)jobSize;
return 0;
}
@@ -505,6 +541,7 @@
/* Wait for our turn */
ZSTD_PTHREAD_MUTEX_LOCK(&serialState->mutex);
while (serialState->nextJobID < jobID) {
+ DEBUGLOG(5, "wait for serialState->cond");
ZSTD_pthread_cond_wait(&serialState->cond, &serialState->mutex);
}
/* A future job may error and skip our job */
@@ -514,6 +551,7 @@
size_t error;
assert(seqStore.seq != NULL && seqStore.pos == 0 &&
seqStore.size == 0 && seqStore.capacity > 0);
+ assert(src.size <= serialState->params.jobSize);
ZSTD_window_update(&serialState->ldmState.window, src.start, src.size);
error = ZSTD_ldm_generateSequences(
&serialState->ldmState, &seqStore,
@@ -593,14 +631,32 @@
unsigned frameChecksumNeeded; /* used only by mtctx */
} ZSTDMT_jobDescription;
+#define JOB_ERROR(e) { \
+ ZSTD_PTHREAD_MUTEX_LOCK(&job->job_mutex); \
+ job->cSize = e; \
+ ZSTD_pthread_mutex_unlock(&job->job_mutex); \
+ goto _endJob; \
+}
+
/* ZSTDMT_compressionJob() is a POOL_function type */
-void ZSTDMT_compressionJob(void* jobDescription)
+static void ZSTDMT_compressionJob(void* jobDescription)
{
ZSTDMT_jobDescription* const job = (ZSTDMT_jobDescription*)jobDescription;
ZSTD_CCtx_params jobParams = job->params; /* do not modify job->params ! copy it, modify the copy */
ZSTD_CCtx* const cctx = ZSTDMT_getCCtx(job->cctxPool);
rawSeqStore_t rawSeqStore = ZSTDMT_getSeq(job->seqPool);
buffer_t dstBuff = job->dstBuff;
+ size_t lastCBlockSize = 0;
+
+ /* resources */
+ if (cctx==NULL) JOB_ERROR(ERROR(memory_allocation));
+ if (dstBuff.start == NULL) { /* streaming job : doesn't provide a dstBuffer */
+ dstBuff = ZSTDMT_getBuffer(job->bufPool);
+ if (dstBuff.start==NULL) JOB_ERROR(ERROR(memory_allocation));
+ job->dstBuff = dstBuff; /* this value can be read in ZSTDMT_flush, when it copies the whole job */
+ }
+ if (jobParams.ldmParams.enableLdm && rawSeqStore.seq == NULL)
+ JOB_ERROR(ERROR(memory_allocation));
/* Don't compute the checksum for chunks, since we compute it externally,
* but write it in the header.
@@ -609,47 +665,31 @@
/* Don't run LDM for the chunks, since we handle it externally */
jobParams.ldmParams.enableLdm = 0;
- /* ressources */
- if (cctx==NULL) {
- job->cSize = ERROR(memory_allocation);
- goto _endJob;
- }
- if (dstBuff.start == NULL) { /* streaming job : doesn't provide a dstBuffer */
- dstBuff = ZSTDMT_getBuffer(job->bufPool);
- if (dstBuff.start==NULL) {
- job->cSize = ERROR(memory_allocation);
- goto _endJob;
- }
- job->dstBuff = dstBuff; /* this value can be read in ZSTDMT_flush, when it copies the whole job */
- }
/* init */
if (job->cdict) {
- size_t const initError = ZSTD_compressBegin_advanced_internal(cctx, NULL, 0, ZSTD_dct_auto, job->cdict, jobParams, job->fullFrameSize);
+ size_t const initError = ZSTD_compressBegin_advanced_internal(cctx, NULL, 0, ZSTD_dct_auto, ZSTD_dtlm_fast, job->cdict, jobParams, job->fullFrameSize);
assert(job->firstJob); /* only allowed for first job */
- if (ZSTD_isError(initError)) { job->cSize = initError; goto _endJob; }
+ if (ZSTD_isError(initError)) JOB_ERROR(initError);
} else { /* srcStart points at reloaded section */
U64 const pledgedSrcSize = job->firstJob ? job->fullFrameSize : job->src.size;
- { size_t const forceWindowError = ZSTD_CCtxParam_setParameter(&jobParams, ZSTD_p_forceMaxWindow, !job->firstJob);
- if (ZSTD_isError(forceWindowError)) {
- job->cSize = forceWindowError;
- goto _endJob;
- } }
+ { size_t const forceWindowError = ZSTD_CCtxParams_setParameter(&jobParams, ZSTD_c_forceMaxWindow, !job->firstJob);
+ if (ZSTD_isError(forceWindowError)) JOB_ERROR(forceWindowError);
+ }
{ size_t const initError = ZSTD_compressBegin_advanced_internal(cctx,
job->prefix.start, job->prefix.size, ZSTD_dct_rawContent, /* load dictionary in "content-only" mode (no header analysis) */
+ ZSTD_dtlm_fast,
NULL, /*cdict*/
jobParams, pledgedSrcSize);
- if (ZSTD_isError(initError)) {
- job->cSize = initError;
- goto _endJob;
- } } }
+ if (ZSTD_isError(initError)) JOB_ERROR(initError);
+ } }
/* Perform serial step as early as possible, but after CCtx initialization */
ZSTDMT_serialState_update(job->serial, cctx, rawSeqStore, job->src, job->jobID);
if (!job->firstJob) { /* flush and overwrite frame header when it's not first job */
size_t const hSize = ZSTD_compressContinue(cctx, dstBuff.start, dstBuff.capacity, job->src.start, 0);
- if (ZSTD_isError(hSize)) { job->cSize = hSize; /* save error code */ goto _endJob; }
+ if (ZSTD_isError(hSize)) JOB_ERROR(hSize);
DEBUGLOG(5, "ZSTDMT_compressionJob: flush and overwrite %u bytes of frame header (not first job)", (U32)hSize);
ZSTD_invalidateRepCodes(cctx);
}
@@ -667,7 +707,7 @@
assert(job->cSize == 0);
for (chunkNb = 1; chunkNb < nbChunks; chunkNb++) {
size_t const cSize = ZSTD_compressContinue(cctx, op, oend-op, ip, chunkSize);
- if (ZSTD_isError(cSize)) { job->cSize = cSize; goto _endJob; }
+ if (ZSTD_isError(cSize)) JOB_ERROR(cSize);
ip += chunkSize;
op += cSize; assert(op < oend);
/* stats */
@@ -680,18 +720,16 @@
ZSTD_pthread_mutex_unlock(&job->job_mutex);
}
/* last block */
- assert(chunkSize > 0); assert((chunkSize & (chunkSize - 1)) == 0); /* chunkSize must be power of 2 for mask==(chunkSize-1) to work */
+ assert(chunkSize > 0);
+ assert((chunkSize & (chunkSize - 1)) == 0); /* chunkSize must be power of 2 for mask==(chunkSize-1) to work */
if ((nbChunks > 0) | job->lastJob /*must output a "last block" flag*/ ) {
size_t const lastBlockSize1 = job->src.size & (chunkSize-1);
size_t const lastBlockSize = ((lastBlockSize1==0) & (job->src.size>=chunkSize)) ? chunkSize : lastBlockSize1;
size_t const cSize = (job->lastJob) ?
ZSTD_compressEnd (cctx, op, oend-op, ip, lastBlockSize) :
ZSTD_compressContinue(cctx, op, oend-op, ip, lastBlockSize);
- if (ZSTD_isError(cSize)) { job->cSize = cSize; goto _endJob; }
- /* stats */
- ZSTD_PTHREAD_MUTEX_LOCK(&job->job_mutex);
- job->cSize += cSize;
- ZSTD_pthread_mutex_unlock(&job->job_mutex);
+ if (ZSTD_isError(cSize)) JOB_ERROR(cSize);
+ lastCBlockSize = cSize;
} }
_endJob:
@@ -704,7 +742,9 @@
ZSTDMT_releaseCCtx(job->cctxPool, cctx);
/* report */
ZSTD_PTHREAD_MUTEX_LOCK(&job->job_mutex);
- job->consumed = job->src.size;
+ if (ZSTD_isError(job->cSize)) assert(lastCBlockSize == 0);
+ job->cSize += lastCBlockSize;
+ job->consumed = job->src.size; /* when job->consumed == job->src.size , compression job is presumed completed */
ZSTD_pthread_cond_signal(&job->job_cond);
ZSTD_pthread_mutex_unlock(&job->job_mutex);
}
@@ -736,6 +776,14 @@
static const roundBuff_t kNullRoundBuff = {NULL, 0, 0};
+#define RSYNC_LENGTH 32
+
+typedef struct {
+ U64 hash;
+ U64 hitMask;
+ U64 primePower;
+} rsyncState_t;
+
struct ZSTDMT_CCtx_s {
POOL_ctx* factory;
ZSTDMT_jobDescription* jobs;
@@ -745,10 +793,11 @@
ZSTD_CCtx_params params;
size_t targetSectionSize;
size_t targetPrefixSize;
- roundBuff_t roundBuff;
+ int jobReady; /* 1 => one job is already prepared, but pool has shortage of workers. Don't create a new job. */
inBuff_t inBuff;
- int jobReady; /* 1 => one job is already prepared, but pool has shortage of workers. Don't create another one. */
+ roundBuff_t roundBuff;
serialState_t serial;
+ rsyncState_t rsync;
unsigned singleBlockingThread;
unsigned jobIDMask;
unsigned doneJobID;
@@ -798,18 +847,28 @@
return jobTable;
}
+static size_t ZSTDMT_expandJobsTable (ZSTDMT_CCtx* mtctx, U32 nbWorkers) {
+ U32 nbJobs = nbWorkers + 2;
+ if (nbJobs > mtctx->jobIDMask+1) { /* need more job capacity */
+ ZSTDMT_freeJobsTable(mtctx->jobs, mtctx->jobIDMask+1, mtctx->cMem);
+ mtctx->jobIDMask = 0;
+ mtctx->jobs = ZSTDMT_createJobsTable(&nbJobs, mtctx->cMem);
+ if (mtctx->jobs==NULL) return ERROR(memory_allocation);
+ assert((nbJobs != 0) && ((nbJobs & (nbJobs - 1)) == 0)); /* ensure nbJobs is a power of 2 */
+ mtctx->jobIDMask = nbJobs - 1;
+ }
+ return 0;
+}
+
+
/* ZSTDMT_CCtxParam_setNbWorkers():
* Internal use only */
size_t ZSTDMT_CCtxParam_setNbWorkers(ZSTD_CCtx_params* params, unsigned nbWorkers)
{
- if (nbWorkers > ZSTDMT_NBWORKERS_MAX) nbWorkers = ZSTDMT_NBWORKERS_MAX;
- params->nbWorkers = nbWorkers;
- params->overlapSizeLog = ZSTDMT_OVERLAPLOG_DEFAULT;
- params->jobSize = 0;
- return nbWorkers;
+ return ZSTD_CCtxParams_setParameter(params, ZSTD_c_nbWorkers, (int)nbWorkers);
}
-ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers, ZSTD_customMem cMem)
+MEM_STATIC ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced_internal(unsigned nbWorkers, ZSTD_customMem cMem)
{
ZSTDMT_CCtx* mtctx;
U32 nbJobs = nbWorkers + 2;
@@ -844,6 +903,17 @@
return mtctx;
}
+ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers, ZSTD_customMem cMem)
+{
+#ifdef ZSTD_MULTITHREAD
+ return ZSTDMT_createCCtx_advanced_internal(nbWorkers, cMem);
+#else
+ (void)nbWorkers;
+ (void)cMem;
+ return NULL;
+#endif
+}
+
ZSTDMT_CCtx* ZSTDMT_createCCtx(unsigned nbWorkers)
{
return ZSTDMT_createCCtx_advanced(nbWorkers, ZSTD_defaultCMem);
@@ -875,7 +945,7 @@
unsigned const jobID = mtctx->doneJobID & mtctx->jobIDMask;
ZSTD_PTHREAD_MUTEX_LOCK(&mtctx->jobs[jobID].job_mutex);
while (mtctx->jobs[jobID].consumed < mtctx->jobs[jobID].src.size) {
- DEBUGLOG(5, "waiting for jobCompleted signal from job %u", mtctx->doneJobID); /* we want to block when waiting for data to flush */
+ DEBUGLOG(4, "waiting for jobCompleted signal from job %u", mtctx->doneJobID); /* we want to block when waiting for data to flush */
ZSTD_pthread_cond_wait(&mtctx->jobs[jobID].job_cond, &mtctx->jobs[jobID].job_mutex);
}
ZSTD_pthread_mutex_unlock(&mtctx->jobs[jobID].job_mutex);
@@ -914,38 +984,44 @@
}
/* Internal only */
-size_t ZSTDMT_CCtxParam_setMTCtxParameter(ZSTD_CCtx_params* params,
- ZSTDMT_parameter parameter, unsigned value) {
+size_t
+ZSTDMT_CCtxParam_setMTCtxParameter(ZSTD_CCtx_params* params,
+ ZSTDMT_parameter parameter,
+ int value)
+{
DEBUGLOG(4, "ZSTDMT_CCtxParam_setMTCtxParameter");
switch(parameter)
{
case ZSTDMT_p_jobSize :
- DEBUGLOG(4, "ZSTDMT_CCtxParam_setMTCtxParameter : set jobSize to %u", value);
- if ( (value > 0) /* value==0 => automatic job size */
- & (value < ZSTDMT_JOBSIZE_MIN) )
- value = ZSTDMT_JOBSIZE_MIN;
- params->jobSize = value;
- return value;
- case ZSTDMT_p_overlapSectionLog :
- if (value > 9) value = 9;
- DEBUGLOG(4, "ZSTDMT_p_overlapSectionLog : %u", value);
- params->overlapSizeLog = (value >= 9) ? 9 : value;
- return value;
+ DEBUGLOG(4, "ZSTDMT_CCtxParam_setMTCtxParameter : set jobSize to %i", value);
+ return ZSTD_CCtxParams_setParameter(params, ZSTD_c_jobSize, value);
+ case ZSTDMT_p_overlapLog :
+ DEBUGLOG(4, "ZSTDMT_p_overlapLog : %i", value);
+ return ZSTD_CCtxParams_setParameter(params, ZSTD_c_overlapLog, value);
+ case ZSTDMT_p_rsyncable :
+ DEBUGLOG(4, "ZSTD_p_rsyncable : %i", value);
+ return ZSTD_CCtxParams_setParameter(params, ZSTD_c_rsyncable, value);
default :
return ERROR(parameter_unsupported);
}
}
-size_t ZSTDMT_setMTCtxParameter(ZSTDMT_CCtx* mtctx, ZSTDMT_parameter parameter, unsigned value)
+size_t ZSTDMT_setMTCtxParameter(ZSTDMT_CCtx* mtctx, ZSTDMT_parameter parameter, int value)
{
DEBUGLOG(4, "ZSTDMT_setMTCtxParameter");
- switch(parameter)
- {
- case ZSTDMT_p_jobSize :
- return ZSTDMT_CCtxParam_setMTCtxParameter(&mtctx->params, parameter, value);
- case ZSTDMT_p_overlapSectionLog :
- return ZSTDMT_CCtxParam_setMTCtxParameter(&mtctx->params, parameter, value);
- default :
+ return ZSTDMT_CCtxParam_setMTCtxParameter(&mtctx->params, parameter, value);
+}
+
+size_t ZSTDMT_getMTCtxParameter(ZSTDMT_CCtx* mtctx, ZSTDMT_parameter parameter, int* value)
+{
+ switch (parameter) {
+ case ZSTDMT_p_jobSize:
+ return ZSTD_CCtxParams_getParameter(&mtctx->params, ZSTD_c_jobSize, value);
+ case ZSTDMT_p_overlapLog:
+ return ZSTD_CCtxParams_getParameter(&mtctx->params, ZSTD_c_overlapLog, value);
+ case ZSTDMT_p_rsyncable:
+ return ZSTD_CCtxParams_getParameter(&mtctx->params, ZSTD_c_rsyncable, value);
+ default:
return ERROR(parameter_unsupported);
}
}
@@ -954,19 +1030,38 @@
* initializing others to default values. */
static ZSTD_CCtx_params ZSTDMT_initJobCCtxParams(ZSTD_CCtx_params const params)
{
- ZSTD_CCtx_params jobParams;
- memset(&jobParams, 0, sizeof(jobParams));
-
- jobParams.cParams = params.cParams;
- jobParams.fParams = params.fParams;
- jobParams.compressionLevel = params.compressionLevel;
- jobParams.disableLiteralCompression = params.disableLiteralCompression;
-
+ ZSTD_CCtx_params jobParams = params;
+ /* Clear parameters related to multithreading */
+ jobParams.forceWindow = 0;
+ jobParams.nbWorkers = 0;
+ jobParams.jobSize = 0;
+ jobParams.overlapLog = 0;
+ jobParams.rsyncable = 0;
+ memset(&jobParams.ldmParams, 0, sizeof(ldmParams_t));
+ memset(&jobParams.customMem, 0, sizeof(ZSTD_customMem));
return jobParams;
}
+
+/* ZSTDMT_resize() :
+ * @return : error code if fails, 0 on success */
+static size_t ZSTDMT_resize(ZSTDMT_CCtx* mtctx, unsigned nbWorkers)
+{
+ if (POOL_resize(mtctx->factory, nbWorkers)) return ERROR(memory_allocation);
+ FORWARD_IF_ERROR( ZSTDMT_expandJobsTable(mtctx, nbWorkers) );
+ mtctx->bufPool = ZSTDMT_expandBufferPool(mtctx->bufPool, nbWorkers);
+ if (mtctx->bufPool == NULL) return ERROR(memory_allocation);
+ mtctx->cctxPool = ZSTDMT_expandCCtxPool(mtctx->cctxPool, nbWorkers);
+ if (mtctx->cctxPool == NULL) return ERROR(memory_allocation);
+ mtctx->seqPool = ZSTDMT_expandSeqPool(mtctx->seqPool, nbWorkers);
+ if (mtctx->seqPool == NULL) return ERROR(memory_allocation);
+ ZSTDMT_CCtxParam_setNbWorkers(&mtctx->params, nbWorkers);
+ return 0;
+}
+
+
/*! ZSTDMT_updateCParams_whileCompressing() :
- * Updates only a selected set of compression parameters, to remain compatible with current frame.
+ * Updates a selected set of compression parameters, remaining compatible with currently active frame.
* New parameters will be applied to next compression job. */
void ZSTDMT_updateCParams_whileCompressing(ZSTDMT_CCtx* mtctx, const ZSTD_CCtx_params* cctxParams)
{
@@ -981,38 +1076,36 @@
}
}
-/* ZSTDMT_getNbWorkers():
- * @return nb threads currently active in mtctx.
- * mtctx must be valid */
-unsigned ZSTDMT_getNbWorkers(const ZSTDMT_CCtx* mtctx)
-{
- assert(mtctx != NULL);
- return mtctx->params.nbWorkers;
-}
-
/* ZSTDMT_getFrameProgression():
* tells how much data has been consumed (input) and produced (output) for current frame.
* able to count progression inside worker threads.
- * Note : mutex will be acquired during statistics collection. */
+ * Note : mutex will be acquired during statistics collection inside workers. */
ZSTD_frameProgression ZSTDMT_getFrameProgression(ZSTDMT_CCtx* mtctx)
{
ZSTD_frameProgression fps;
- DEBUGLOG(6, "ZSTDMT_getFrameProgression");
- fps.consumed = mtctx->consumed;
- fps.produced = mtctx->produced;
+ DEBUGLOG(5, "ZSTDMT_getFrameProgression");
fps.ingested = mtctx->consumed + mtctx->inBuff.filled;
+ fps.consumed = mtctx->consumed;
+ fps.produced = fps.flushed = mtctx->produced;
+ fps.currentJobID = mtctx->nextJobID;
+ fps.nbActiveWorkers = 0;
{ unsigned jobNb;
unsigned lastJobNb = mtctx->nextJobID + mtctx->jobReady; assert(mtctx->jobReady <= 1);
DEBUGLOG(6, "ZSTDMT_getFrameProgression: jobs: from %u to <%u (jobReady:%u)",
mtctx->doneJobID, lastJobNb, mtctx->jobReady)
for (jobNb = mtctx->doneJobID ; jobNb < lastJobNb ; jobNb++) {
unsigned const wJobID = jobNb & mtctx->jobIDMask;
- ZSTD_pthread_mutex_lock(&mtctx->jobs[wJobID].job_mutex);
- { size_t const cResult = mtctx->jobs[wJobID].cSize;
+ ZSTDMT_jobDescription* jobPtr = &mtctx->jobs[wJobID];
+ ZSTD_pthread_mutex_lock(&jobPtr->job_mutex);
+ { size_t const cResult = jobPtr->cSize;
size_t const produced = ZSTD_isError(cResult) ? 0 : cResult;
- fps.consumed += mtctx->jobs[wJobID].consumed;
- fps.ingested += mtctx->jobs[wJobID].src.size;
+ size_t const flushed = ZSTD_isError(cResult) ? 0 : jobPtr->dstFlushed;
+ assert(flushed <= produced);
+ fps.ingested += jobPtr->src.size;
+ fps.consumed += jobPtr->consumed;
fps.produced += produced;
+ fps.flushed += flushed;
+ fps.nbActiveWorkers += (jobPtr->consumed < jobPtr->src.size);
}
ZSTD_pthread_mutex_unlock(&mtctx->jobs[wJobID].job_mutex);
}
@@ -1021,26 +1114,107 @@
}
+size_t ZSTDMT_toFlushNow(ZSTDMT_CCtx* mtctx)
+{
+ size_t toFlush;
+ unsigned const jobID = mtctx->doneJobID;
+ assert(jobID <= mtctx->nextJobID);
+ if (jobID == mtctx->nextJobID) return 0; /* no active job => nothing to flush */
+
+ /* look into oldest non-fully-flushed job */
+ { unsigned const wJobID = jobID & mtctx->jobIDMask;
+ ZSTDMT_jobDescription* const jobPtr = &mtctx->jobs[wJobID];
+ ZSTD_pthread_mutex_lock(&jobPtr->job_mutex);
+ { size_t const cResult = jobPtr->cSize;
+ size_t const produced = ZSTD_isError(cResult) ? 0 : cResult;
+ size_t const flushed = ZSTD_isError(cResult) ? 0 : jobPtr->dstFlushed;
+ assert(flushed <= produced);
+ assert(jobPtr->consumed <= jobPtr->src.size);
+ toFlush = produced - flushed;
+ /* if toFlush==0, nothing is available to flush.
+ * However, jobID is expected to still be active:
+ * if jobID was already completed and fully flushed,
+ * ZSTDMT_flushProduced() should have already moved onto next job.
+ * Therefore, some input has not yet been consumed. */
+ if (toFlush==0) {
+ assert(jobPtr->consumed < jobPtr->src.size);
+ }
+ }
+ ZSTD_pthread_mutex_unlock(&mtctx->jobs[wJobID].job_mutex);
+ }
+
+ return toFlush;
+}
+
+
/* ------------------------------------------ */
/* ===== Multi-threaded compression ===== */
/* ------------------------------------------ */
-static size_t ZSTDMT_computeTargetJobLog(ZSTD_CCtx_params const params)
+static unsigned ZSTDMT_computeTargetJobLog(ZSTD_CCtx_params const params)
{
- if (params.ldmParams.enableLdm)
- return MAX(21, params.cParams.chainLog + 4);
- return MAX(20, params.cParams.windowLog + 2);
+ unsigned jobLog;
+ if (params.ldmParams.enableLdm) {
+ /* In Long Range Mode, the windowLog is typically oversized.
+ * In which case, it's preferable to determine the jobSize
+ * based on chainLog instead. */
+ jobLog = MAX(21, params.cParams.chainLog + 4);
+ } else {
+ jobLog = MAX(20, params.cParams.windowLog + 2);
+ }
+ return MIN(jobLog, (unsigned)ZSTDMT_JOBLOG_MAX);
}
-static size_t ZSTDMT_computeOverlapLog(ZSTD_CCtx_params const params)
+static int ZSTDMT_overlapLog_default(ZSTD_strategy strat)
{
- unsigned const overlapRLog = (params.overlapSizeLog>9) ? 0 : 9-params.overlapSizeLog;
- if (params.ldmParams.enableLdm)
- return (MIN(params.cParams.windowLog, ZSTDMT_computeTargetJobLog(params) - 2) - overlapRLog);
- return overlapRLog >= 9 ? 0 : (params.cParams.windowLog - overlapRLog);
+ switch(strat)
+ {
+ case ZSTD_btultra2:
+ return 9;
+ case ZSTD_btultra:
+ case ZSTD_btopt:
+ return 8;
+ case ZSTD_btlazy2:
+ case ZSTD_lazy2:
+ return 7;
+ case ZSTD_lazy:
+ case ZSTD_greedy:
+ case ZSTD_dfast:
+ case ZSTD_fast:
+ default:;
+ }
+ return 6;
}
-static unsigned ZSTDMT_computeNbJobs(ZSTD_CCtx_params params, size_t srcSize, unsigned nbWorkers) {
+static int ZSTDMT_overlapLog(int ovlog, ZSTD_strategy strat)
+{
+ assert(0 <= ovlog && ovlog <= 9);
+ if (ovlog == 0) return ZSTDMT_overlapLog_default(strat);
+ return ovlog;
+}
+
+static size_t ZSTDMT_computeOverlapSize(ZSTD_CCtx_params const params)
+{
+ int const overlapRLog = 9 - ZSTDMT_overlapLog(params.overlapLog, params.cParams.strategy);
+ int ovLog = (overlapRLog >= 8) ? 0 : (params.cParams.windowLog - overlapRLog);
+ assert(0 <= overlapRLog && overlapRLog <= 8);
+ if (params.ldmParams.enableLdm) {
+ /* In Long Range Mode, the windowLog is typically oversized.
+ * In which case, it's preferable to determine the jobSize
+ * based on chainLog instead.
+ * Then, ovLog becomes a fraction of the jobSize, rather than windowSize */
+ ovLog = MIN(params.cParams.windowLog, ZSTDMT_computeTargetJobLog(params) - 2)
+ - overlapRLog;
+ }
+ assert(0 <= ovLog && ovLog <= ZSTD_WINDOWLOG_MAX);
+ DEBUGLOG(4, "overlapLog : %i", params.overlapLog);
+ DEBUGLOG(4, "overlap size : %i", 1 << ovLog);
+ return (ovLog==0) ? 0 : (size_t)1 << ovLog;
+}
+
+static unsigned
+ZSTDMT_computeNbJobs(ZSTD_CCtx_params params, size_t srcSize, unsigned nbWorkers)
+{
assert(nbWorkers>0);
{ size_t const jobSizeTarget = (size_t)1 << ZSTDMT_computeTargetJobLog(params);
size_t const jobMaxSize = jobSizeTarget << 2;
@@ -1063,7 +1237,7 @@
ZSTD_CCtx_params params)
{
ZSTD_CCtx_params const jobParams = ZSTDMT_initJobCCtxParams(params);
- size_t const overlapSize = (size_t)1 << ZSTDMT_computeOverlapLog(params);
+ size_t const overlapSize = ZSTDMT_computeOverlapSize(params);
unsigned const nbJobs = ZSTDMT_computeNbJobs(params, srcSize, params.nbWorkers);
size_t const proposedJobSize = (srcSize + (nbJobs-1)) / nbJobs;
size_t const avgJobSize = (((proposedJobSize-1) & 0x1FFFF) < 0x7FFF) ? proposedJobSize + 0xFFFF : proposedJobSize; /* avoid too small last block */
@@ -1087,18 +1261,10 @@
assert(avgJobSize >= 256 KB); /* condition for ZSTD_compressBound(A) + ZSTD_compressBound(B) <= ZSTD_compressBound(A+B), required to compress directly into Dst (no additional buffer) */
ZSTDMT_setBufferSize(mtctx->bufPool, ZSTD_compressBound(avgJobSize) );
- if (ZSTDMT_serialState_reset(&mtctx->serial, mtctx->seqPool, params))
+ if (ZSTDMT_serialState_reset(&mtctx->serial, mtctx->seqPool, params, avgJobSize))
return ERROR(memory_allocation);
- if (nbJobs > mtctx->jobIDMask+1) { /* enlarge job table */
- U32 jobsTableSize = nbJobs;
- ZSTDMT_freeJobsTable(mtctx->jobs, mtctx->jobIDMask+1, mtctx->cMem);
- mtctx->jobIDMask = 0;
- mtctx->jobs = ZSTDMT_createJobsTable(&jobsTableSize, mtctx->cMem);
- if (mtctx->jobs==NULL) return ERROR(memory_allocation);
- assert((jobsTableSize != 0) && ((jobsTableSize & (jobsTableSize - 1)) == 0)); /* ensure jobsTableSize is a power of 2 */
- mtctx->jobIDMask = jobsTableSize - 1;
- }
+ FORWARD_IF_ERROR( ZSTDMT_expandJobsTable(mtctx, nbJobs) ); /* only expands if necessary */
{ unsigned u;
for (u=0; u<nbJobs; u++) {
@@ -1182,16 +1348,17 @@
}
size_t ZSTDMT_compress_advanced(ZSTDMT_CCtx* mtctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- const ZSTD_CDict* cdict,
- ZSTD_parameters params,
- unsigned overlapLog)
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ const ZSTD_CDict* cdict,
+ ZSTD_parameters params,
+ int overlapLog)
{
ZSTD_CCtx_params cctxParams = mtctx->params;
cctxParams.cParams = params.cParams;
cctxParams.fParams = params.fParams;
- cctxParams.overlapSizeLog = overlapLog;
+ assert(ZSTD_OVERLAPLOG_MIN <= overlapLog && overlapLog <= ZSTD_OVERLAPLOG_MAX);
+ cctxParams.overlapLog = overlapLog;
return ZSTDMT_compress_advanced_internal(mtctx,
dst, dstCapacity,
src, srcSize,
@@ -1204,8 +1371,8 @@
const void* src, size_t srcSize,
int compressionLevel)
{
- U32 const overlapLog = (compressionLevel >= ZSTD_maxCLevel()) ? 9 : ZSTDMT_OVERLAPLOG_DEFAULT;
ZSTD_parameters params = ZSTD_getParams(compressionLevel, srcSize, 0);
+ int const overlapLog = ZSTDMT_overlapLog_default(params.cParams.strategy);
params.fParams.contentSizeFlag = 1;
return ZSTDMT_compress_advanced(mtctx, dst, dstCapacity, src, srcSize, NULL, params, overlapLog);
}
@@ -1221,18 +1388,19 @@
const ZSTD_CDict* cdict, ZSTD_CCtx_params params,
unsigned long long pledgedSrcSize)
{
- DEBUGLOG(4, "ZSTDMT_initCStream_internal (pledgedSrcSize=%u, nbWorkers=%u, cctxPool=%u, disableLiteralCompression=%i)",
- (U32)pledgedSrcSize, params.nbWorkers, mtctx->cctxPool->totalCCtx, params.disableLiteralCompression);
- /* params are supposed to be fully validated at this point */
+ DEBUGLOG(4, "ZSTDMT_initCStream_internal (pledgedSrcSize=%u, nbWorkers=%u, cctxPool=%u)",
+ (U32)pledgedSrcSize, params.nbWorkers, mtctx->cctxPool->totalCCtx);
+
+ /* params supposed partially fully validated at this point */
assert(!ZSTD_isError(ZSTD_checkCParams(params.cParams)));
assert(!((dict) && (cdict))); /* either dict or cdict, not both */
- assert(mtctx->cctxPool->totalCCtx == params.nbWorkers);
/* init */
- if (params.jobSize == 0) {
- params.jobSize = 1U << ZSTDMT_computeTargetJobLog(params);
- }
- if (params.jobSize > ZSTDMT_JOBSIZE_MAX) params.jobSize = ZSTDMT_JOBSIZE_MAX;
+ if (params.nbWorkers != mtctx->params.nbWorkers)
+ FORWARD_IF_ERROR( ZSTDMT_resize(mtctx, params.nbWorkers) );
+
+ if (params.jobSize != 0 && params.jobSize < ZSTDMT_JOBSIZE_MIN) params.jobSize = ZSTDMT_JOBSIZE_MIN;
+ if (params.jobSize > (size_t)ZSTDMT_JOBSIZE_MAX) params.jobSize = (size_t)ZSTDMT_JOBSIZE_MAX;
mtctx->singleBlockingThread = (pledgedSrcSize <= ZSTDMT_JOBSIZE_MIN); /* do not trigger multi-threading when srcSize is too small */
if (mtctx->singleBlockingThread) {
@@ -1267,12 +1435,26 @@
mtctx->cdict = cdict;
}
- mtctx->targetPrefixSize = (size_t)1 << ZSTDMT_computeOverlapLog(params);
- DEBUGLOG(4, "overlapLog=%u => %u KB", params.overlapSizeLog, (U32)(mtctx->targetPrefixSize>>10));
+ mtctx->targetPrefixSize = ZSTDMT_computeOverlapSize(params);
+ DEBUGLOG(4, "overlapLog=%i => %u KB", params.overlapLog, (U32)(mtctx->targetPrefixSize>>10));
mtctx->targetSectionSize = params.jobSize;
- if (mtctx->targetSectionSize < ZSTDMT_JOBSIZE_MIN) mtctx->targetSectionSize = ZSTDMT_JOBSIZE_MIN;
+ if (mtctx->targetSectionSize == 0) {
+ mtctx->targetSectionSize = 1ULL << ZSTDMT_computeTargetJobLog(params);
+ }
+ assert(mtctx->targetSectionSize <= (size_t)ZSTDMT_JOBSIZE_MAX);
+
+ if (params.rsyncable) {
+ /* Aim for the targetsectionSize as the average job size. */
+ U32 const jobSizeMB = (U32)(mtctx->targetSectionSize >> 20);
+ U32 const rsyncBits = ZSTD_highbit32(jobSizeMB) + 20;
+ assert(jobSizeMB >= 1);
+ DEBUGLOG(4, "rsyncLog = %u", rsyncBits);
+ mtctx->rsync.hash = 0;
+ mtctx->rsync.hitMask = (1ULL << rsyncBits) - 1;
+ mtctx->rsync.primePower = ZSTD_rollingHash_primePower(RSYNC_LENGTH);
+ }
if (mtctx->targetSectionSize < mtctx->targetPrefixSize) mtctx->targetSectionSize = mtctx->targetPrefixSize; /* job size must be >= overlap size */
- DEBUGLOG(4, "Job Size : %u KB (note : set to %u)", (U32)(mtctx->targetSectionSize>>10), params.jobSize);
+ DEBUGLOG(4, "Job Size : %u KB (note : set to %u)", (U32)(mtctx->targetSectionSize>>10), (U32)params.jobSize);
DEBUGLOG(4, "inBuff Size : %u KB", (U32)(mtctx->targetSectionSize>>10));
ZSTDMT_setBufferSize(mtctx->bufPool, ZSTD_compressBound(mtctx->targetSectionSize));
{
@@ -1312,7 +1494,7 @@
mtctx->allJobsCompleted = 0;
mtctx->consumed = 0;
mtctx->produced = 0;
- if (ZSTDMT_serialState_reset(&mtctx->serial, mtctx->seqPool, params))
+ if (ZSTDMT_serialState_reset(&mtctx->serial, mtctx->seqPool, params, mtctx->targetSectionSize))
return ERROR(memory_allocation);
return 0;
}
@@ -1368,7 +1550,7 @@
/* ZSTDMT_writeLastEmptyBlock()
* Write a single empty block with an end-of-frame to finish a frame.
* Job must be created from streaming variant.
- * This function is always successfull if expected conditions are fulfilled.
+ * This function is always successful if expected conditions are fulfilled.
*/
static void ZSTDMT_writeLastEmptyBlock(ZSTDMT_jobDescription* job)
{
@@ -1420,7 +1602,7 @@
mtctx->jobs[jobID].jobID = mtctx->nextJobID;
mtctx->jobs[jobID].firstJob = (mtctx->nextJobID==0);
mtctx->jobs[jobID].lastJob = endFrame;
- mtctx->jobs[jobID].frameChecksumNeeded = endFrame && (mtctx->nextJobID>0) && mtctx->params.fParams.checksumFlag;
+ mtctx->jobs[jobID].frameChecksumNeeded = mtctx->params.fParams.checksumFlag && endFrame && (mtctx->nextJobID>0);
mtctx->jobs[jobID].dstFlushed = 0;
/* Update the round buffer pos and clear the input buffer to be reset */
@@ -1468,6 +1650,8 @@
/*! ZSTDMT_flushProduced() :
+ * flush whatever data has been produced but not yet flushed in current job.
+ * move to next job if current one is fully flushed.
* `output` : `pos` will be updated with amount of data flushed .
* `blockToFlush` : if >0, the function will block and wait if there is no data available to flush .
* @return : amount of data remaining within internal buffer, 0 if no more, 1 if unknown but > 0, or an error code */
@@ -1496,7 +1680,7 @@
/* try to flush something */
{ size_t cSize = mtctx->jobs[wJobID].cSize; /* shared */
size_t const srcConsumed = mtctx->jobs[wJobID].consumed; /* shared */
- size_t const srcSize = mtctx->jobs[wJobID].src.size; /* read-only, could be done after mutex lock, but no-declaration-after-statement */
+ size_t const srcSize = mtctx->jobs[wJobID].src.size; /* read-only, could be done after mutex lock, but no-declaration-after-statement */
ZSTD_pthread_mutex_unlock(&mtctx->jobs[wJobID].job_mutex);
if (ZSTD_isError(cSize)) {
DEBUGLOG(5, "ZSTDMT_flushProduced: job %u : compression error detected : %s",
@@ -1516,6 +1700,7 @@
mtctx->jobs[wJobID].cSize += 4; /* can write this shared value, as worker is no longer active */
mtctx->jobs[wJobID].frameChecksumNeeded = 0;
}
+
if (cSize > 0) { /* compression is ongoing or completed */
size_t const toFlush = MIN(cSize - mtctx->jobs[wJobID].dstFlushed, output->size - output->pos);
DEBUGLOG(5, "ZSTDMT_flushProduced: Flushing %u bytes from job %u (completion:%u/%u, generated:%u)",
@@ -1529,11 +1714,12 @@
output->pos += toFlush;
mtctx->jobs[wJobID].dstFlushed += toFlush; /* can write : this value is only used by mtctx */
- if ( (srcConsumed == srcSize) /* job completed */
+ if ( (srcConsumed == srcSize) /* job is completed */
&& (mtctx->jobs[wJobID].dstFlushed == cSize) ) { /* output buffer fully flushed => free this job position */
DEBUGLOG(5, "Job %u completed (%u bytes), moving to next one",
mtctx->doneJobID, (U32)mtctx->jobs[wJobID].dstFlushed);
ZSTDMT_releaseBuffer(mtctx->bufPool, mtctx->jobs[wJobID].dstBuff);
+ DEBUGLOG(5, "dstBuffer released");
mtctx->jobs[wJobID].dstBuff = g_nullBuffer;
mtctx->jobs[wJobID].cSize = 0; /* ensure this job slot is considered "not started" in future check */
mtctx->consumed += srcSize;
@@ -1610,6 +1796,7 @@
range_t extDict;
range_t prefix;
+ DEBUGLOG(5, "ZSTDMT_doesOverlapWindow");
extDict.start = window.dictBase + window.lowLimit;
extDict.size = window.dictLimit - window.lowLimit;
@@ -1630,12 +1817,13 @@
{
if (mtctx->params.ldmParams.enableLdm) {
ZSTD_pthread_mutex_t* mutex = &mtctx->serial.ldmWindowMutex;
+ DEBUGLOG(5, "ZSTDMT_waitForLdmComplete");
DEBUGLOG(5, "source [0x%zx, 0x%zx)",
(size_t)buffer.start,
(size_t)buffer.start + buffer.capacity);
ZSTD_PTHREAD_MUTEX_LOCK(mutex);
while (ZSTDMT_doesOverlapWindow(buffer, mtctx->serial.ldmWindow)) {
- DEBUGLOG(6, "Waiting for LDM to finish...");
+ DEBUGLOG(5, "Waiting for LDM to finish...");
ZSTD_pthread_cond_wait(&mtctx->serial.ldmWindowCond, mutex);
}
DEBUGLOG(6, "Done waiting for LDM to finish");
@@ -1655,6 +1843,7 @@
size_t const target = mtctx->targetSectionSize;
buffer_t buffer;
+ DEBUGLOG(5, "ZSTDMT_tryGetInputRange");
assert(mtctx->inBuff.buffer.start == NULL);
assert(mtctx->roundBuff.capacity >= target);
@@ -1668,7 +1857,7 @@
buffer.start = start;
buffer.capacity = prefixSize;
if (ZSTDMT_isOverlapped(buffer, inUse)) {
- DEBUGLOG(6, "Waiting for buffer...");
+ DEBUGLOG(5, "Waiting for buffer...");
return 0;
}
ZSTDMT_waitForLdmComplete(mtctx, buffer);
@@ -1680,7 +1869,7 @@
buffer.capacity = target;
if (ZSTDMT_isOverlapped(buffer, inUse)) {
- DEBUGLOG(6, "Waiting for buffer...");
+ DEBUGLOG(5, "Waiting for buffer...");
return 0;
}
assert(!ZSTDMT_isOverlapped(buffer, mtctx->inBuff.prefix));
@@ -1701,6 +1890,89 @@
return 1;
}
+typedef struct {
+ size_t toLoad; /* The number of bytes to load from the input. */
+ int flush; /* Boolean declaring if we must flush because we found a synchronization point. */
+} syncPoint_t;
+
+/**
+ * Searches through the input for a synchronization point. If one is found, we
+ * will instruct the caller to flush, and return the number of bytes to load.
+ * Otherwise, we will load as many bytes as possible and instruct the caller
+ * to continue as normal.
+ */
+static syncPoint_t
+findSynchronizationPoint(ZSTDMT_CCtx const* mtctx, ZSTD_inBuffer const input)
+{
+ BYTE const* const istart = (BYTE const*)input.src + input.pos;
+ U64 const primePower = mtctx->rsync.primePower;
+ U64 const hitMask = mtctx->rsync.hitMask;
+
+ syncPoint_t syncPoint;
+ U64 hash;
+ BYTE const* prev;
+ size_t pos;
+
+ syncPoint.toLoad = MIN(input.size - input.pos, mtctx->targetSectionSize - mtctx->inBuff.filled);
+ syncPoint.flush = 0;
+ if (!mtctx->params.rsyncable)
+ /* Rsync is disabled. */
+ return syncPoint;
+ if (mtctx->inBuff.filled + syncPoint.toLoad < RSYNC_LENGTH)
+ /* Not enough to compute the hash.
+ * We will miss any synchronization points in this RSYNC_LENGTH byte
+ * window. However, since it depends only in the internal buffers, if the
+ * state is already synchronized, we will remain synchronized.
+ * Additionally, the probability that we miss a synchronization point is
+ * low: RSYNC_LENGTH / targetSectionSize.
+ */
+ return syncPoint;
+ /* Initialize the loop variables. */
+ if (mtctx->inBuff.filled >= RSYNC_LENGTH) {
+ /* We have enough bytes buffered to initialize the hash.
+ * Start scanning at the beginning of the input.
+ */
+ pos = 0;
+ prev = (BYTE const*)mtctx->inBuff.buffer.start + mtctx->inBuff.filled - RSYNC_LENGTH;
+ hash = ZSTD_rollingHash_compute(prev, RSYNC_LENGTH);
+ } else {
+ /* We don't have enough bytes buffered to initialize the hash, but
+ * we know we have at least RSYNC_LENGTH bytes total.
+ * Start scanning after the first RSYNC_LENGTH bytes less the bytes
+ * already buffered.
+ */
+ pos = RSYNC_LENGTH - mtctx->inBuff.filled;
+ prev = (BYTE const*)mtctx->inBuff.buffer.start - pos;
+ hash = ZSTD_rollingHash_compute(mtctx->inBuff.buffer.start, mtctx->inBuff.filled);
+ hash = ZSTD_rollingHash_append(hash, istart, pos);
+ }
+ /* Starting with the hash of the previous RSYNC_LENGTH bytes, roll
+ * through the input. If we hit a synchronization point, then cut the
+ * job off, and tell the compressor to flush the job. Otherwise, load
+ * all the bytes and continue as normal.
+ * If we go too long without a synchronization point (targetSectionSize)
+ * then a block will be emitted anyways, but this is okay, since if we
+ * are already synchronized we will remain synchronized.
+ */
+ for (; pos < syncPoint.toLoad; ++pos) {
+ BYTE const toRemove = pos < RSYNC_LENGTH ? prev[pos] : istart[pos - RSYNC_LENGTH];
+ /* if (pos >= RSYNC_LENGTH) assert(ZSTD_rollingHash_compute(istart + pos - RSYNC_LENGTH, RSYNC_LENGTH) == hash); */
+ hash = ZSTD_rollingHash_rotate(hash, toRemove, istart[pos], primePower);
+ if ((hash & hitMask) == hitMask) {
+ syncPoint.toLoad = pos + 1;
+ syncPoint.flush = 1;
+ break;
+ }
+ }
+ return syncPoint;
+}
+
+size_t ZSTDMT_nextInputSizeHint(const ZSTDMT_CCtx* mtctx)
+{
+ size_t hintInSize = mtctx->targetSectionSize - mtctx->inBuff.filled;
+ if (hintInSize==0) hintInSize = mtctx->targetSectionSize;
+ return hintInSize;
+}
/** ZSTDMT_compressStream_generic() :
* internal use only - exposed to be invoked from zstd_compress.c
@@ -1718,7 +1990,7 @@
assert(input->pos <= input->size);
if (mtctx->singleBlockingThread) { /* delegate to single-thread (synchronous) */
- return ZSTD_compressStream_generic(mtctx->cctxPool->cctx[0], output, input, endOp);
+ return ZSTD_compressStream2(mtctx->cctxPool->cctx[0], output, input, endOp);
}
if ((mtctx->frameEnded) && (endOp==ZSTD_e_continue)) {
@@ -1727,7 +1999,8 @@
}
/* single-pass shortcut (note : synchronous-mode) */
- if ( (mtctx->nextJobID == 0) /* just started */
+ if ( (!mtctx->params.rsyncable) /* rsyncable mode is disabled */
+ && (mtctx->nextJobID == 0) /* just started */
&& (mtctx->inBuff.filled == 0) /* nothing buffered */
&& (!mtctx->jobReady) /* no job already created */
&& (endOp == ZSTD_e_end) /* end order */
@@ -1753,18 +2026,23 @@
/* It is only possible for this operation to fail if there are
* still compression jobs ongoing.
*/
+ DEBUGLOG(5, "ZSTDMT_tryGetInputRange failed");
assert(mtctx->doneJobID != mtctx->nextJobID);
- }
+ } else
+ DEBUGLOG(5, "ZSTDMT_tryGetInputRange completed successfully : mtctx->inBuff.buffer.start = %p", mtctx->inBuff.buffer.start);
}
if (mtctx->inBuff.buffer.start != NULL) {
- size_t const toLoad = MIN(input->size - input->pos, mtctx->targetSectionSize - mtctx->inBuff.filled);
+ syncPoint_t const syncPoint = findSynchronizationPoint(mtctx, *input);
+ if (syncPoint.flush && endOp == ZSTD_e_continue) {
+ endOp = ZSTD_e_flush;
+ }
assert(mtctx->inBuff.buffer.capacity >= mtctx->targetSectionSize);
DEBUGLOG(5, "ZSTDMT_compressStream_generic: adding %u bytes on top of %u to buffer of size %u",
- (U32)toLoad, (U32)mtctx->inBuff.filled, (U32)mtctx->targetSectionSize);
- memcpy((char*)mtctx->inBuff.buffer.start + mtctx->inBuff.filled, (const char*)input->src + input->pos, toLoad);
- input->pos += toLoad;
- mtctx->inBuff.filled += toLoad;
- forwardInputProgress = toLoad>0;
+ (U32)syncPoint.toLoad, (U32)mtctx->inBuff.filled, (U32)mtctx->targetSectionSize);
+ memcpy((char*)mtctx->inBuff.buffer.start + mtctx->inBuff.filled, (const char*)input->src + input->pos, syncPoint.toLoad);
+ input->pos += syncPoint.toLoad;
+ mtctx->inBuff.filled += syncPoint.toLoad;
+ forwardInputProgress = syncPoint.toLoad>0;
}
if ((input->pos < input->size) && (endOp == ZSTD_e_end))
endOp = ZSTD_e_flush; /* can't end now : not all input consumed */
@@ -1776,12 +2054,13 @@
|| ((endOp == ZSTD_e_end) && (!mtctx->frameEnded)) ) { /* must finish the frame with a zero-size block */
size_t const jobSize = mtctx->inBuff.filled;
assert(mtctx->inBuff.filled <= mtctx->targetSectionSize);
- CHECK_F( ZSTDMT_createCompressionJob(mtctx, jobSize, endOp) );
+ FORWARD_IF_ERROR( ZSTDMT_createCompressionJob(mtctx, jobSize, endOp) );
}
/* check for potential compressed data ready to be flushed */
{ size_t const remainingToFlush = ZSTDMT_flushProduced(mtctx, output, !forwardInputProgress, endOp); /* block if there was no forward input progress */
if (input->pos < input->size) return MAX(remainingToFlush, 1); /* input not consumed : do not end flush yet */
+ DEBUGLOG(5, "end of ZSTDMT_compressStream_generic: remainingToFlush = %u", (U32)remainingToFlush);
return remainingToFlush;
}
}
@@ -1789,7 +2068,7 @@
size_t ZSTDMT_compressStream(ZSTDMT_CCtx* mtctx, ZSTD_outBuffer* output, ZSTD_inBuffer* input)
{
- CHECK_F( ZSTDMT_compressStream_generic(mtctx, output, input, ZSTD_e_continue) );
+ FORWARD_IF_ERROR( ZSTDMT_compressStream_generic(mtctx, output, input, ZSTD_e_continue) );
/* recommended next input size : fill current input buffer */
return mtctx->targetSectionSize - mtctx->inBuff.filled; /* note : could be zero when input buffer is fully filled and no more availability to create new job */
@@ -1806,7 +2085,7 @@
|| ((endFrame==ZSTD_e_end) && !mtctx->frameEnded)) { /* need a last 0-size block to end frame */
DEBUGLOG(5, "ZSTDMT_flushStream_internal : create a new job (%u bytes, end:%u)",
(U32)srcSize, (U32)endFrame);
- CHECK_F( ZSTDMT_createCompressionJob(mtctx, srcSize, endFrame) );
+ FORWARD_IF_ERROR( ZSTDMT_createCompressionJob(mtctx, srcSize, endFrame) );
}
/* check if there is any data available to flush */
diff --git a/vendor/github.com/DataDog/zstd/zstdmt_compress.h b/vendor/github.com/DataDog/zstd/zstdmt_compress.h
index f79e3b4..12a5260 100644
--- a/vendor/github.com/DataDog/zstd/zstdmt_compress.h
+++ b/vendor/github.com/DataDog/zstd/zstdmt_compress.h
@@ -17,10 +17,25 @@
/* Note : This is an internal API.
- * Some methods are still exposed (ZSTDLIB_API),
+ * These APIs used to be exposed with ZSTDLIB_API,
* because it used to be the only way to invoke MT compression.
- * Now, it's recommended to use ZSTD_compress_generic() instead.
- * These methods will stop being exposed in a future version */
+ * Now, it's recommended to use ZSTD_compress2 and ZSTD_compressStream2()
+ * instead.
+ *
+ * If you depend on these APIs and can't switch, then define
+ * ZSTD_LEGACY_MULTITHREADED_API when making the dynamic library.
+ * However, we may completely remove these functions in a future
+ * release, so please switch soon.
+ *
+ * This API requires ZSTD_MULTITHREAD to be defined during compilation,
+ * otherwise ZSTDMT_createCCtx*() will fail.
+ */
+
+#ifdef ZSTD_LEGACY_MULTITHREADED_API
+# define ZSTDMT_API ZSTDLIB_API
+#else
+# define ZSTDMT_API
+#endif
/* === Dependencies === */
#include <stddef.h> /* size_t */
@@ -28,19 +43,32 @@
#include "zstd.h" /* ZSTD_inBuffer, ZSTD_outBuffer, ZSTDLIB_API */
+/* === Constants === */
+#ifndef ZSTDMT_NBWORKERS_MAX
+# define ZSTDMT_NBWORKERS_MAX 200
+#endif
+#ifndef ZSTDMT_JOBSIZE_MIN
+# define ZSTDMT_JOBSIZE_MIN (1 MB)
+#endif
+#define ZSTDMT_JOBLOG_MAX (MEM_32bits() ? 29 : 30)
+#define ZSTDMT_JOBSIZE_MAX (MEM_32bits() ? (512 MB) : (1024 MB))
+
+
/* === Memory management === */
typedef struct ZSTDMT_CCtx_s ZSTDMT_CCtx;
-ZSTDLIB_API ZSTDMT_CCtx* ZSTDMT_createCCtx(unsigned nbWorkers);
-ZSTDLIB_API ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers,
+/* Requires ZSTD_MULTITHREAD to be defined during compilation, otherwise it will return NULL. */
+ZSTDMT_API ZSTDMT_CCtx* ZSTDMT_createCCtx(unsigned nbWorkers);
+/* Requires ZSTD_MULTITHREAD to be defined during compilation, otherwise it will return NULL. */
+ZSTDMT_API ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers,
ZSTD_customMem cMem);
-ZSTDLIB_API size_t ZSTDMT_freeCCtx(ZSTDMT_CCtx* mtctx);
+ZSTDMT_API size_t ZSTDMT_freeCCtx(ZSTDMT_CCtx* mtctx);
-ZSTDLIB_API size_t ZSTDMT_sizeof_CCtx(ZSTDMT_CCtx* mtctx);
+ZSTDMT_API size_t ZSTDMT_sizeof_CCtx(ZSTDMT_CCtx* mtctx);
/* === Simple one-pass compression function === */
-ZSTDLIB_API size_t ZSTDMT_compressCCtx(ZSTDMT_CCtx* mtctx,
+ZSTDMT_API size_t ZSTDMT_compressCCtx(ZSTDMT_CCtx* mtctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
int compressionLevel);
@@ -49,34 +77,31 @@
/* === Streaming functions === */
-ZSTDLIB_API size_t ZSTDMT_initCStream(ZSTDMT_CCtx* mtctx, int compressionLevel);
-ZSTDLIB_API size_t ZSTDMT_resetCStream(ZSTDMT_CCtx* mtctx, unsigned long long pledgedSrcSize); /**< if srcSize is not known at reset time, use ZSTD_CONTENTSIZE_UNKNOWN. Note: for compatibility with older programs, 0 means the same as ZSTD_CONTENTSIZE_UNKNOWN, but it will change in the future to mean "empty" */
+ZSTDMT_API size_t ZSTDMT_initCStream(ZSTDMT_CCtx* mtctx, int compressionLevel);
+ZSTDMT_API size_t ZSTDMT_resetCStream(ZSTDMT_CCtx* mtctx, unsigned long long pledgedSrcSize); /**< if srcSize is not known at reset time, use ZSTD_CONTENTSIZE_UNKNOWN. Note: for compatibility with older programs, 0 means the same as ZSTD_CONTENTSIZE_UNKNOWN, but it will change in the future to mean "empty" */
-ZSTDLIB_API size_t ZSTDMT_compressStream(ZSTDMT_CCtx* mtctx, ZSTD_outBuffer* output, ZSTD_inBuffer* input);
+ZSTDMT_API size_t ZSTDMT_nextInputSizeHint(const ZSTDMT_CCtx* mtctx);
+ZSTDMT_API size_t ZSTDMT_compressStream(ZSTDMT_CCtx* mtctx, ZSTD_outBuffer* output, ZSTD_inBuffer* input);
-ZSTDLIB_API size_t ZSTDMT_flushStream(ZSTDMT_CCtx* mtctx, ZSTD_outBuffer* output); /**< @return : 0 == all flushed; >0 : still some data to be flushed; or an error code (ZSTD_isError()) */
-ZSTDLIB_API size_t ZSTDMT_endStream(ZSTDMT_CCtx* mtctx, ZSTD_outBuffer* output); /**< @return : 0 == all flushed; >0 : still some data to be flushed; or an error code (ZSTD_isError()) */
+ZSTDMT_API size_t ZSTDMT_flushStream(ZSTDMT_CCtx* mtctx, ZSTD_outBuffer* output); /**< @return : 0 == all flushed; >0 : still some data to be flushed; or an error code (ZSTD_isError()) */
+ZSTDMT_API size_t ZSTDMT_endStream(ZSTDMT_CCtx* mtctx, ZSTD_outBuffer* output); /**< @return : 0 == all flushed; >0 : still some data to be flushed; or an error code (ZSTD_isError()) */
/* === Advanced functions and parameters === */
-#ifndef ZSTDMT_JOBSIZE_MIN
-# define ZSTDMT_JOBSIZE_MIN (1U << 20) /* 1 MB - Minimum size of each compression job */
-#endif
+ZSTDMT_API size_t ZSTDMT_compress_advanced(ZSTDMT_CCtx* mtctx,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ const ZSTD_CDict* cdict,
+ ZSTD_parameters params,
+ int overlapLog);
-ZSTDLIB_API size_t ZSTDMT_compress_advanced(ZSTDMT_CCtx* mtctx,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- const ZSTD_CDict* cdict,
- ZSTD_parameters params,
- unsigned overlapLog);
-
-ZSTDLIB_API size_t ZSTDMT_initCStream_advanced(ZSTDMT_CCtx* mtctx,
+ZSTDMT_API size_t ZSTDMT_initCStream_advanced(ZSTDMT_CCtx* mtctx,
const void* dict, size_t dictSize, /* dict can be released after init, a local copy is preserved within zcs */
ZSTD_parameters params,
unsigned long long pledgedSrcSize); /* pledgedSrcSize is optional and can be zero == unknown */
-ZSTDLIB_API size_t ZSTDMT_initCStream_usingCDict(ZSTDMT_CCtx* mtctx,
+ZSTDMT_API size_t ZSTDMT_initCStream_usingCDict(ZSTDMT_CCtx* mtctx,
const ZSTD_CDict* cdict,
ZSTD_frameParameters fparams,
unsigned long long pledgedSrcSize); /* note : zero means empty */
@@ -84,8 +109,9 @@
/* ZSTDMT_parameter :
* List of parameters that can be set using ZSTDMT_setMTCtxParameter() */
typedef enum {
- ZSTDMT_p_jobSize, /* Each job is compressed in parallel. By default, this value is dynamically determined depending on compression parameters. Can be set explicitly here. */
- ZSTDMT_p_overlapSectionLog /* Each job may reload a part of previous job to enhance compressionr ratio; 0 == no overlap, 6(default) == use 1/8th of window, >=9 == use full window. This is a "sticky" parameter : its value will be re-used on next compression job */
+ ZSTDMT_p_jobSize, /* Each job is compressed in parallel. By default, this value is dynamically determined depending on compression parameters. Can be set explicitly here. */
+ ZSTDMT_p_overlapLog, /* Each job may reload a part of previous job to enhance compression ratio; 0 == no overlap, 6(default) == use 1/8th of window, >=9 == use full window. This is a "sticky" parameter : its value will be re-used on next compression job */
+ ZSTDMT_p_rsyncable /* Enables rsyncable mode. */
} ZSTDMT_parameter;
/* ZSTDMT_setMTCtxParameter() :
@@ -93,7 +119,12 @@
* The function must be called typically after ZSTD_createCCtx() but __before ZSTDMT_init*() !__
* Parameters not explicitly reset by ZSTDMT_init*() remain the same in consecutive compression sessions.
* @return : 0, or an error code (which can be tested using ZSTD_isError()) */
-ZSTDLIB_API size_t ZSTDMT_setMTCtxParameter(ZSTDMT_CCtx* mtctx, ZSTDMT_parameter parameter, unsigned value);
+ZSTDMT_API size_t ZSTDMT_setMTCtxParameter(ZSTDMT_CCtx* mtctx, ZSTDMT_parameter parameter, int value);
+
+/* ZSTDMT_getMTCtxParameter() :
+ * Query the ZSTDMT_CCtx for a parameter value.
+ * @return : 0, or an error code (which can be tested using ZSTD_isError()) */
+ZSTDMT_API size_t ZSTDMT_getMTCtxParameter(ZSTDMT_CCtx* mtctx, ZSTDMT_parameter parameter, int* value);
/*! ZSTDMT_compressStream_generic() :
@@ -103,7 +134,7 @@
* 0 if fully flushed
* or an error code
* note : needs to be init using any ZSTD_initCStream*() variant */
-ZSTDLIB_API size_t ZSTDMT_compressStream_generic(ZSTDMT_CCtx* mtctx,
+ZSTDMT_API size_t ZSTDMT_compressStream_generic(ZSTDMT_CCtx* mtctx,
ZSTD_outBuffer* output,
ZSTD_inBuffer* input,
ZSTD_EndDirective endOp);
@@ -114,11 +145,21 @@
* === Not exposed in libzstd. Never invoke directly ===
* ======================================================== */
-size_t ZSTDMT_CCtxParam_setMTCtxParameter(ZSTD_CCtx_params* params, ZSTDMT_parameter parameter, unsigned value);
+ /*! ZSTDMT_toFlushNow()
+ * Tell how many bytes are ready to be flushed immediately.
+ * Probe the oldest active job (not yet entirely flushed) and check its output buffer.
+ * If return 0, it means there is no active job,
+ * or, it means oldest job is still active, but everything produced has been flushed so far,
+ * therefore flushing is limited by speed of oldest job. */
+size_t ZSTDMT_toFlushNow(ZSTDMT_CCtx* mtctx);
-/* ZSTDMT_CCtxParam_setNbWorkers()
- * Set nbWorkers, and clamp it.
- * Also reset jobSize and overlapLog */
+/*! ZSTDMT_CCtxParam_setMTCtxParameter()
+ * like ZSTDMT_setMTCtxParameter(), but into a ZSTD_CCtx_Params */
+size_t ZSTDMT_CCtxParam_setMTCtxParameter(ZSTD_CCtx_params* params, ZSTDMT_parameter parameter, int value);
+
+/*! ZSTDMT_CCtxParam_setNbWorkers()
+ * Set nbWorkers, and clamp it.
+ * Also reset jobSize and overlapLog */
size_t ZSTDMT_CCtxParam_setNbWorkers(ZSTD_CCtx_params* params, unsigned nbWorkers);
/*! ZSTDMT_updateCParams_whileCompressing() :
@@ -126,14 +167,9 @@
* New parameters will be applied to next compression job. */
void ZSTDMT_updateCParams_whileCompressing(ZSTDMT_CCtx* mtctx, const ZSTD_CCtx_params* cctxParams);
-/* ZSTDMT_getNbWorkers():
- * @return nb threads currently active in mtctx.
- * mtctx must be valid */
-unsigned ZSTDMT_getNbWorkers(const ZSTDMT_CCtx* mtctx);
-
-/* ZSTDMT_getFrameProgression():
- * tells how much data has been consumed (input) and produced (output) for current frame.
- * able to count progression inside worker threads.
+/*! ZSTDMT_getFrameProgression():
+ * tells how much data has been consumed (input) and produced (output) for current frame.
+ * able to count progression inside worker threads.
*/
ZSTD_frameProgression ZSTDMT_getFrameProgression(ZSTDMT_CCtx* mtctx);
diff --git a/vendor/github.com/Shopify/sarama/.travis.yml b/vendor/github.com/Shopify/sarama/.travis.yml
index b7f7874..4331fa1 100644
--- a/vendor/github.com/Shopify/sarama/.travis.yml
+++ b/vendor/github.com/Shopify/sarama/.travis.yml
@@ -1,7 +1,8 @@
+dist: xenial
language: go
go:
-- 1.10.x
- 1.11.x
+- 1.12.x
env:
global:
@@ -11,15 +12,16 @@
- KAFKA_HOSTNAME=localhost
- DEBUG=true
matrix:
- - KAFKA_VERSION=1.1.1 KAFKA_SCALA_VERSION=2.11
- - KAFKA_VERSION=2.0.1 KAFKA_SCALA_VERSION=2.12
- - KAFKA_VERSION=2.1.0 KAFKA_SCALA_VERSION=2.12
+ - KAFKA_VERSION=2.1.1 KAFKA_SCALA_VERSION=2.12
+ - KAFKA_VERSION=2.2.1 KAFKA_SCALA_VERSION=2.12
+ - KAFKA_VERSION=2.3.0 KAFKA_SCALA_VERSION=2.12
before_install:
- export REPOSITORY_ROOT=${TRAVIS_BUILD_DIR}
- vagrant/install_cluster.sh
- vagrant/boot_cluster.sh
- vagrant/create_topics.sh
+- vagrant/run_java_producer.sh
install: make install_dependencies
@@ -27,7 +29,7 @@
- make test
- make vet
- make errcheck
-- if [[ "$TRAVIS_GO_VERSION" == 1.11* ]]; then make fmt; fi
+- if [[ "$TRAVIS_GO_VERSION" == 1.12* ]]; then make fmt; fi
after_success:
- bash <(curl -s https://codecov.io/bash)
diff --git a/vendor/github.com/Shopify/sarama/CHANGELOG.md b/vendor/github.com/Shopify/sarama/CHANGELOG.md
index 0e69c2d..02bd0ff 100644
--- a/vendor/github.com/Shopify/sarama/CHANGELOG.md
+++ b/vendor/github.com/Shopify/sarama/CHANGELOG.md
@@ -1,5 +1,109 @@
# Changelog
+#### Version 1.23.1 (2019-07-22)
+
+Bug Fixes:
+- Fix fetch delete bug record
+ ([1425](https://github.com/Shopify/sarama/pull/1425)).
+- Handle SASL/OAUTHBEARER token rejection
+ ([1428](https://github.com/Shopify/sarama/pull/1428)).
+
+#### Version 1.23.0 (2019-07-02)
+
+New Features:
+- Add support for Kafka 2.3.0
+ ([1418](https://github.com/Shopify/sarama/pull/1418)).
+- Add support for ListConsumerGroupOffsets v2
+ ([1374](https://github.com/Shopify/sarama/pull/1374)).
+- Add support for DeleteConsumerGroup
+ ([1417](https://github.com/Shopify/sarama/pull/1417)).
+- Add support for SASLVersion configuration
+ ([1410](https://github.com/Shopify/sarama/pull/1410)).
+- Add kerberos support
+ ([1366](https://github.com/Shopify/sarama/pull/1366)).
+
+Improvements:
+- Improve sasl_scram_client example
+ ([1406](https://github.com/Shopify/sarama/pull/1406)).
+- Fix shutdown and race-condition in consumer-group example
+ ([1404](https://github.com/Shopify/sarama/pull/1404)).
+- Add support for error codes 77—81
+ ([1397](https://github.com/Shopify/sarama/pull/1397)).
+- Pool internal objects allocated per message
+ ([1385](https://github.com/Shopify/sarama/pull/1385)).
+- Reduce packet decoder allocations
+ ([1373](https://github.com/Shopify/sarama/pull/1373)).
+- Support timeout when fetching metadata
+ ([1359](https://github.com/Shopify/sarama/pull/1359)).
+
+Bug Fixes:
+- Fix fetch size integer overflow
+ ([1376](https://github.com/Shopify/sarama/pull/1376)).
+- Handle and log throttled FetchResponses
+ ([1383](https://github.com/Shopify/sarama/pull/1383)).
+- Refactor misspelled word Resouce to Resource
+ ([1368](https://github.com/Shopify/sarama/pull/1368)).
+
+#### Version 1.22.1 (2019-04-29)
+
+Improvements:
+- Use zstd 1.3.8
+ ([1350](https://github.com/Shopify/sarama/pull/1350)).
+- Add support for SaslHandshakeRequest v1
+ ([1354](https://github.com/Shopify/sarama/pull/1354)).
+
+Bug Fixes:
+- Fix V5 MetadataRequest nullable topics array
+ ([1353](https://github.com/Shopify/sarama/pull/1353)).
+- Use a different SCRAM client for each broker connection
+ ([1349](https://github.com/Shopify/sarama/pull/1349)).
+- Fix AllowAutoTopicCreation for MetadataRequest greater than v3
+ ([1344](https://github.com/Shopify/sarama/pull/1344)).
+
+#### Version 1.22.0 (2019-04-09)
+
+New Features:
+- Add Offline Replicas Operation to Client
+ ([1318](https://github.com/Shopify/sarama/pull/1318)).
+- Allow using proxy when connecting to broker
+ ([1326](https://github.com/Shopify/sarama/pull/1326)).
+- Implement ReadCommitted
+ ([1307](https://github.com/Shopify/sarama/pull/1307)).
+- Add support for Kafka 2.2.0
+ ([1331](https://github.com/Shopify/sarama/pull/1331)).
+- Add SASL SCRAM-SHA-512 and SCRAM-SHA-256 mechanismes
+ ([1331](https://github.com/Shopify/sarama/pull/1295)).
+
+Improvements:
+- Unregister all broker metrics on broker stop
+ ([1232](https://github.com/Shopify/sarama/pull/1232)).
+- Add SCRAM authentication example
+ ([1303](https://github.com/Shopify/sarama/pull/1303)).
+- Add consumergroup examples
+ ([1304](https://github.com/Shopify/sarama/pull/1304)).
+- Expose consumer batch size metric
+ ([1296](https://github.com/Shopify/sarama/pull/1296)).
+- Add TLS options to console producer and consumer
+ ([1300](https://github.com/Shopify/sarama/pull/1300)).
+- Reduce client close bookkeeping
+ ([1297](https://github.com/Shopify/sarama/pull/1297)).
+- Satisfy error interface in create responses
+ ([1154](https://github.com/Shopify/sarama/pull/1154)).
+- Please lint gods
+ ([1346](https://github.com/Shopify/sarama/pull/1346)).
+
+Bug Fixes:
+- Fix multi consumer group instance crash
+ ([1338](https://github.com/Shopify/sarama/pull/1338)).
+- Update lz4 to latest version
+ ([1347](https://github.com/Shopify/sarama/pull/1347)).
+- Retry ErrNotCoordinatorForConsumer in new consumergroup session
+ ([1231](https://github.com/Shopify/sarama/pull/1231)).
+- Fix cleanup error handler
+ ([1332](https://github.com/Shopify/sarama/pull/1332)).
+- Fix rate condition in PartitionConsumer
+ ([1156](https://github.com/Shopify/sarama/pull/1156)).
+
#### Version 1.21.0 (2019-02-24)
New Features:
diff --git a/vendor/github.com/Shopify/sarama/Makefile b/vendor/github.com/Shopify/sarama/Makefile
index 09f743c..360b220 100644
--- a/vendor/github.com/Shopify/sarama/Makefile
+++ b/vendor/github.com/Shopify/sarama/Makefile
@@ -1,11 +1,12 @@
export GO111MODULE=on
-default: fmt vet errcheck test
+default: fmt vet errcheck test lint
# Taken from https://github.com/codecov/example-go#caveat-multiple-files
+.PHONY: test
test:
echo "" > coverage.txt
- for d in `go list ./... | grep -v vendor`; do \
+ for d in `go list ./...`; do \
go test -p 1 -v -timeout 240s -race -coverprofile=profile.out -covermode=atomic $$d || exit 1; \
if [ -f profile.out ]; then \
cat profile.out >> coverage.txt; \
@@ -13,20 +14,39 @@
fi \
done
+GOLINT := $(shell command -v golint)
+
+.PHONY: lint
+lint:
+ifndef GOLINT
+ go get golang.org/x/lint/golint
+endif
+ go list ./... | xargs golint
+
+.PHONY: vet
vet:
go vet ./...
+ERRCHECK := $(shell command -v errcheck)
# See https://github.com/kisielk/errcheck/pull/141 for details on ignorepkg
+.PHONY: errcheck
errcheck:
+ifndef ERRCHECK
+ go get github.com/kisielk/errcheck
+endif
errcheck -ignorepkg fmt github.com/Shopify/sarama/...
+.PHONY: fmt
fmt:
@if [ -n "$$(go fmt ./...)" ]; then echo 'Please run go fmt on your code.' && exit 1; fi
-install_dependencies: install_errcheck get
+.PHONY : install_dependencies
+install_dependencies: get
-install_errcheck:
- go get github.com/kisielk/errcheck
-
+.PHONY: get
get:
- go get -t
+ go get -t -v ./...
+
+.PHONY: clean
+clean:
+ go clean ./...
diff --git a/vendor/github.com/Shopify/sarama/README.md b/vendor/github.com/Shopify/sarama/README.md
index f241b89..4cd736b 100644
--- a/vendor/github.com/Shopify/sarama/README.md
+++ b/vendor/github.com/Shopify/sarama/README.md
@@ -21,7 +21,7 @@
Sarama provides a "2 releases + 2 months" compatibility guarantee: we support
the two latest stable releases of Kafka and Go, and we provide a two month
grace period for older releases. This means we currently officially support
-Go 1.8 through 1.11, and Kafka 1.0 through 2.0, although older releases are
+Go 1.11 through 1.12, and Kafka 2.0 through 2.3, although older releases are
still likely to work.
Sarama follows semantic versioning and provides API stability via the gopkg.in service.
diff --git a/vendor/github.com/Shopify/sarama/acl_bindings.go b/vendor/github.com/Shopify/sarama/acl_bindings.go
index b91c282..50b689d 100644
--- a/vendor/github.com/Shopify/sarama/acl_bindings.go
+++ b/vendor/github.com/Shopify/sarama/acl_bindings.go
@@ -1,9 +1,10 @@
package sarama
+//Resource holds information about acl resource type
type Resource struct {
- ResourceType AclResourceType
- ResourceName string
- ResoucePatternType AclResourcePatternType
+ ResourceType AclResourceType
+ ResourceName string
+ ResourcePatternType AclResourcePatternType
}
func (r *Resource) encode(pe packetEncoder, version int16) error {
@@ -14,11 +15,11 @@
}
if version == 1 {
- if r.ResoucePatternType == AclPatternUnknown {
+ if r.ResourcePatternType == AclPatternUnknown {
Logger.Print("Cannot encode an unknown resource pattern type, using Literal instead")
- r.ResoucePatternType = AclPatternLiteral
+ r.ResourcePatternType = AclPatternLiteral
}
- pe.putInt8(int8(r.ResoucePatternType))
+ pe.putInt8(int8(r.ResourcePatternType))
}
return nil
@@ -39,12 +40,13 @@
if err != nil {
return err
}
- r.ResoucePatternType = AclResourcePatternType(pattern)
+ r.ResourcePatternType = AclResourcePatternType(pattern)
}
return nil
}
+//Acl holds information about acl type
type Acl struct {
Principal string
Host string
@@ -91,6 +93,7 @@
return nil
}
+//ResourceAcls is an acl resource type
type ResourceAcls struct {
Resource
Acls []*Acl
diff --git a/vendor/github.com/Shopify/sarama/acl_create_request.go b/vendor/github.com/Shopify/sarama/acl_create_request.go
index d3d5ad8..da1cdef 100644
--- a/vendor/github.com/Shopify/sarama/acl_create_request.go
+++ b/vendor/github.com/Shopify/sarama/acl_create_request.go
@@ -1,5 +1,6 @@
package sarama
+//CreateAclsRequest is an acl creation request
type CreateAclsRequest struct {
Version int16
AclCreations []*AclCreation
@@ -38,16 +39,16 @@
return nil
}
-func (d *CreateAclsRequest) key() int16 {
+func (c *CreateAclsRequest) key() int16 {
return 30
}
-func (d *CreateAclsRequest) version() int16 {
- return d.Version
+func (c *CreateAclsRequest) version() int16 {
+ return c.Version
}
-func (d *CreateAclsRequest) requiredVersion() KafkaVersion {
- switch d.Version {
+func (c *CreateAclsRequest) requiredVersion() KafkaVersion {
+ switch c.Version {
case 1:
return V2_0_0_0
default:
@@ -55,6 +56,7 @@
}
}
+//AclCreation is a wrapper around Resource and Acl type
type AclCreation struct {
Resource
Acl
diff --git a/vendor/github.com/Shopify/sarama/acl_create_response.go b/vendor/github.com/Shopify/sarama/acl_create_response.go
index 8a56f35..f5a5e9a 100644
--- a/vendor/github.com/Shopify/sarama/acl_create_response.go
+++ b/vendor/github.com/Shopify/sarama/acl_create_response.go
@@ -2,6 +2,7 @@
import "time"
+//CreateAclsResponse is a an acl reponse creation type
type CreateAclsResponse struct {
ThrottleTime time.Duration
AclCreationResponses []*AclCreationResponse
@@ -46,18 +47,19 @@
return nil
}
-func (d *CreateAclsResponse) key() int16 {
+func (c *CreateAclsResponse) key() int16 {
return 30
}
-func (d *CreateAclsResponse) version() int16 {
+func (c *CreateAclsResponse) version() int16 {
return 0
}
-func (d *CreateAclsResponse) requiredVersion() KafkaVersion {
+func (c *CreateAclsResponse) requiredVersion() KafkaVersion {
return V0_11_0_0
}
+//AclCreationResponse is an acl creation response type
type AclCreationResponse struct {
Err KError
ErrMsg *string
diff --git a/vendor/github.com/Shopify/sarama/acl_delete_request.go b/vendor/github.com/Shopify/sarama/acl_delete_request.go
index 5e94ad7..15908ea 100644
--- a/vendor/github.com/Shopify/sarama/acl_delete_request.go
+++ b/vendor/github.com/Shopify/sarama/acl_delete_request.go
@@ -1,5 +1,6 @@
package sarama
+//DeleteAclsRequest is a delete acl request
type DeleteAclsRequest struct {
Version int
Filters []*AclFilter
diff --git a/vendor/github.com/Shopify/sarama/acl_delete_response.go b/vendor/github.com/Shopify/sarama/acl_delete_response.go
index a885fe5..6529565 100644
--- a/vendor/github.com/Shopify/sarama/acl_delete_response.go
+++ b/vendor/github.com/Shopify/sarama/acl_delete_response.go
@@ -2,21 +2,22 @@
import "time"
+//DeleteAclsResponse is a delete acl response
type DeleteAclsResponse struct {
Version int16
ThrottleTime time.Duration
FilterResponses []*FilterResponse
}
-func (a *DeleteAclsResponse) encode(pe packetEncoder) error {
- pe.putInt32(int32(a.ThrottleTime / time.Millisecond))
+func (d *DeleteAclsResponse) encode(pe packetEncoder) error {
+ pe.putInt32(int32(d.ThrottleTime / time.Millisecond))
- if err := pe.putArrayLength(len(a.FilterResponses)); err != nil {
+ if err := pe.putArrayLength(len(d.FilterResponses)); err != nil {
return err
}
- for _, filterResponse := range a.FilterResponses {
- if err := filterResponse.encode(pe, a.Version); err != nil {
+ for _, filterResponse := range d.FilterResponses {
+ if err := filterResponse.encode(pe, d.Version); err != nil {
return err
}
}
@@ -24,22 +25,22 @@
return nil
}
-func (a *DeleteAclsResponse) decode(pd packetDecoder, version int16) (err error) {
+func (d *DeleteAclsResponse) decode(pd packetDecoder, version int16) (err error) {
throttleTime, err := pd.getInt32()
if err != nil {
return err
}
- a.ThrottleTime = time.Duration(throttleTime) * time.Millisecond
+ d.ThrottleTime = time.Duration(throttleTime) * time.Millisecond
n, err := pd.getArrayLength()
if err != nil {
return err
}
- a.FilterResponses = make([]*FilterResponse, n)
+ d.FilterResponses = make([]*FilterResponse, n)
for i := 0; i < n; i++ {
- a.FilterResponses[i] = new(FilterResponse)
- if err := a.FilterResponses[i].decode(pd, version); err != nil {
+ d.FilterResponses[i] = new(FilterResponse)
+ if err := d.FilterResponses[i].decode(pd, version); err != nil {
return err
}
}
@@ -59,6 +60,7 @@
return V0_11_0_0
}
+//FilterResponse is a filter response type
type FilterResponse struct {
Err KError
ErrMsg *string
@@ -109,6 +111,7 @@
return nil
}
+//MatchingAcl is a matching acl type
type MatchingAcl struct {
Err KError
ErrMsg *string
diff --git a/vendor/github.com/Shopify/sarama/acl_describe_request.go b/vendor/github.com/Shopify/sarama/acl_describe_request.go
index 3c95320..5222d46 100644
--- a/vendor/github.com/Shopify/sarama/acl_describe_request.go
+++ b/vendor/github.com/Shopify/sarama/acl_describe_request.go
@@ -1,5 +1,6 @@
package sarama
+//DescribeAclsRequest is a secribe acl request type
type DescribeAclsRequest struct {
Version int
AclFilter
diff --git a/vendor/github.com/Shopify/sarama/acl_describe_response.go b/vendor/github.com/Shopify/sarama/acl_describe_response.go
index db3a88c..12126e5 100644
--- a/vendor/github.com/Shopify/sarama/acl_describe_response.go
+++ b/vendor/github.com/Shopify/sarama/acl_describe_response.go
@@ -2,6 +2,7 @@
import "time"
+//DescribeAclsResponse is a describe acl response type
type DescribeAclsResponse struct {
Version int16
ThrottleTime time.Duration
diff --git a/vendor/github.com/Shopify/sarama/acl_types.go b/vendor/github.com/Shopify/sarama/acl_types.go
index 72b7985..c10ad7b 100644
--- a/vendor/github.com/Shopify/sarama/acl_types.go
+++ b/vendor/github.com/Shopify/sarama/acl_types.go
@@ -1,50 +1,51 @@
package sarama
-type AclOperation int
+type (
+ AclOperation int
+
+ AclPermissionType int
+
+ AclResourceType int
+
+ AclResourcePatternType int
+)
// ref: https://github.com/apache/kafka/blob/trunk/clients/src/main/java/org/apache/kafka/common/acl/AclOperation.java
const (
- AclOperationUnknown AclOperation = 0
- AclOperationAny AclOperation = 1
- AclOperationAll AclOperation = 2
- AclOperationRead AclOperation = 3
- AclOperationWrite AclOperation = 4
- AclOperationCreate AclOperation = 5
- AclOperationDelete AclOperation = 6
- AclOperationAlter AclOperation = 7
- AclOperationDescribe AclOperation = 8
- AclOperationClusterAction AclOperation = 9
- AclOperationDescribeConfigs AclOperation = 10
- AclOperationAlterConfigs AclOperation = 11
- AclOperationIdempotentWrite AclOperation = 12
+ AclOperationUnknown AclOperation = iota
+ AclOperationAny
+ AclOperationAll
+ AclOperationRead
+ AclOperationWrite
+ AclOperationCreate
+ AclOperationDelete
+ AclOperationAlter
+ AclOperationDescribe
+ AclOperationClusterAction
+ AclOperationDescribeConfigs
+ AclOperationAlterConfigs
+ AclOperationIdempotentWrite
)
-type AclPermissionType int
-
// ref: https://github.com/apache/kafka/blob/trunk/clients/src/main/java/org/apache/kafka/common/acl/AclPermissionType.java
const (
- AclPermissionUnknown AclPermissionType = 0
- AclPermissionAny AclPermissionType = 1
- AclPermissionDeny AclPermissionType = 2
- AclPermissionAllow AclPermissionType = 3
+ AclPermissionUnknown AclPermissionType = iota
+ AclPermissionAny
+ AclPermissionDeny
+ AclPermissionAllow
)
-type AclResourceType int
-
// ref: https://github.com/apache/kafka/blob/trunk/clients/src/main/java/org/apache/kafka/common/resource/ResourceType.java
const (
- AclResourceUnknown AclResourceType = 0
- AclResourceAny AclResourceType = 1
- AclResourceTopic AclResourceType = 2
- AclResourceGroup AclResourceType = 3
- AclResourceCluster AclResourceType = 4
- AclResourceTransactionalID AclResourceType = 5
+ AclResourceUnknown AclResourceType = iota
+ AclResourceAny
+ AclResourceTopic
+ AclResourceGroup
+ AclResourceCluster
+ AclResourceTransactionalID
)
-type AclResourcePatternType int
-
// ref: https://github.com/apache/kafka/blob/trunk/clients/src/main/java/org/apache/kafka/common/resource/PatternType.java
-
const (
AclPatternUnknown AclResourcePatternType = iota
AclPatternAny
diff --git a/vendor/github.com/Shopify/sarama/add_offsets_to_txn_request.go b/vendor/github.com/Shopify/sarama/add_offsets_to_txn_request.go
index 6da166c..fc227ab 100644
--- a/vendor/github.com/Shopify/sarama/add_offsets_to_txn_request.go
+++ b/vendor/github.com/Shopify/sarama/add_offsets_to_txn_request.go
@@ -1,5 +1,6 @@
package sarama
+//AddOffsetsToTxnRequest adds offsets to a transaction request
type AddOffsetsToTxnRequest struct {
TransactionalID string
ProducerID int64
diff --git a/vendor/github.com/Shopify/sarama/add_offsets_to_txn_response.go b/vendor/github.com/Shopify/sarama/add_offsets_to_txn_response.go
index 3a46151..c88c1f8 100644
--- a/vendor/github.com/Shopify/sarama/add_offsets_to_txn_response.go
+++ b/vendor/github.com/Shopify/sarama/add_offsets_to_txn_response.go
@@ -4,6 +4,7 @@
"time"
)
+//AddOffsetsToTxnResponse is a response type for adding offsets to txns
type AddOffsetsToTxnResponse struct {
ThrottleTime time.Duration
Err KError
diff --git a/vendor/github.com/Shopify/sarama/add_partitions_to_txn_request.go b/vendor/github.com/Shopify/sarama/add_partitions_to_txn_request.go
index a8a5922..8d4b42e 100644
--- a/vendor/github.com/Shopify/sarama/add_partitions_to_txn_request.go
+++ b/vendor/github.com/Shopify/sarama/add_partitions_to_txn_request.go
@@ -1,5 +1,6 @@
package sarama
+//AddPartitionsToTxnRequest is a add paartition request
type AddPartitionsToTxnRequest struct {
TransactionalID string
ProducerID int64
diff --git a/vendor/github.com/Shopify/sarama/add_partitions_to_txn_response.go b/vendor/github.com/Shopify/sarama/add_partitions_to_txn_response.go
index 581c556..eb4f23e 100644
--- a/vendor/github.com/Shopify/sarama/add_partitions_to_txn_response.go
+++ b/vendor/github.com/Shopify/sarama/add_partitions_to_txn_response.go
@@ -4,6 +4,7 @@
"time"
)
+//AddPartitionsToTxnResponse is a partition errors to transaction type
type AddPartitionsToTxnResponse struct {
ThrottleTime time.Duration
Errors map[string][]*PartitionError
@@ -82,6 +83,7 @@
return V0_11_0_0
}
+//PartitionError is a partition error type
type PartitionError struct {
Partition int32
Err KError
diff --git a/vendor/github.com/Shopify/sarama/admin.go b/vendor/github.com/Shopify/sarama/admin.go
index 18b055a..1db6a0e 100644
--- a/vendor/github.com/Shopify/sarama/admin.go
+++ b/vendor/github.com/Shopify/sarama/admin.go
@@ -20,7 +20,7 @@
// List the topics available in the cluster with the default options.
ListTopics() (map[string]TopicDetail, error)
- // Describe some topics in the cluster
+ // Describe some topics in the cluster.
DescribeTopics(topics []string) (metadata []*TopicMetadata, err error)
// Delete a topic. It may take several seconds after the DeleteTopic to returns success
@@ -78,12 +78,15 @@
// List the consumer groups available in the cluster.
ListConsumerGroups() (map[string]string, error)
- // Describe the given consumer group
+ // Describe the given consumer groups.
DescribeConsumerGroups(groups []string) ([]*GroupDescription, error)
// List the consumer group offsets available in the cluster.
ListConsumerGroupOffsets(group string, topicPartitions map[string][]int32) (*OffsetFetchResponse, error)
+ // Delete a consumer group.
+ DeleteConsumerGroup(group string) error
+
// Get information about the nodes in the cluster
DescribeCluster() (brokers []*Broker, controllerID int32, err error)
@@ -131,7 +134,7 @@
}
if detail == nil {
- return errors.New("You must specify topic details")
+ return errors.New("you must specify topic details")
}
topicDetails := make(map[string]*TopicDetail)
@@ -166,7 +169,7 @@
}
if topicErr.Err != ErrNoError {
- return topicErr.Err
+ return topicErr
}
return nil
@@ -183,7 +186,9 @@
AllowAutoTopicCreation: false,
}
- if ca.conf.Version.IsAtLeast(V0_11_0_0) {
+ if ca.conf.Version.IsAtLeast(V1_0_0_0) {
+ request.Version = 5
+ } else if ca.conf.Version.IsAtLeast(V0_11_0_0) {
request.Version = 4
}
@@ -358,7 +363,7 @@
}
if topicErr.Err != ErrNoError {
- return topicErr.Err
+ return topicErr
}
return nil
@@ -369,29 +374,50 @@
if topic == "" {
return ErrInvalidTopic
}
-
- topics := make(map[string]*DeleteRecordsRequestTopic)
- topics[topic] = &DeleteRecordsRequestTopic{PartitionOffsets: partitionOffsets}
- request := &DeleteRecordsRequest{
- Topics: topics,
- Timeout: ca.conf.Admin.Timeout,
+ partitionPerBroker := make(map[*Broker][]int32)
+ for partition := range partitionOffsets {
+ broker, err := ca.client.Leader(topic, partition)
+ if err != nil {
+ return err
+ }
+ if _, ok := partitionPerBroker[broker]; ok {
+ partitionPerBroker[broker] = append(partitionPerBroker[broker], partition)
+ } else {
+ partitionPerBroker[broker] = []int32{partition}
+ }
}
+ errs := make([]error, 0)
+ for broker, partitions := range partitionPerBroker {
+ topics := make(map[string]*DeleteRecordsRequestTopic)
+ recordsToDelete := make(map[int32]int64)
+ for _, p := range partitions {
+ recordsToDelete[p] = partitionOffsets[p]
+ }
+ topics[topic] = &DeleteRecordsRequestTopic{PartitionOffsets: recordsToDelete}
+ request := &DeleteRecordsRequest{
+ Topics: topics,
+ Timeout: ca.conf.Admin.Timeout,
+ }
- b, err := ca.Controller()
- if err != nil {
- return err
+ rsp, err := broker.DeleteRecords(request)
+ if err != nil {
+ errs = append(errs, err)
+ } else {
+ deleteRecordsResponseTopic, ok := rsp.Topics[topic]
+ if !ok {
+ errs = append(errs, ErrIncompleteResponse)
+ } else {
+ for _, deleteRecordsResponsePartition := range deleteRecordsResponseTopic.Partitions {
+ if deleteRecordsResponsePartition.Err != ErrNoError {
+ errs = append(errs, errors.New(deleteRecordsResponsePartition.Err.Error()))
+ }
+ }
+ }
+ }
}
-
- rsp, err := b.DeleteRecords(request)
- if err != nil {
- return err
+ if len(errs) > 0 {
+ return ErrDeleteRecords{MultiError{&errs}}
}
-
- _, ok := rsp.Topics[topic]
- if !ok {
- return ErrIncompleteResponse
- }
-
//todo since we are dealing with couple of partitions it would be good if we return slice of errors
//for each partition instead of one error
return nil
@@ -606,9 +632,38 @@
partitions: topicPartitions,
}
- if ca.conf.Version.IsAtLeast(V0_8_2_2) {
+ if ca.conf.Version.IsAtLeast(V0_10_2_0) {
+ request.Version = 2
+ } else if ca.conf.Version.IsAtLeast(V0_8_2_2) {
request.Version = 1
}
return coordinator.FetchOffset(request)
}
+
+func (ca *clusterAdmin) DeleteConsumerGroup(group string) error {
+ coordinator, err := ca.client.Coordinator(group)
+ if err != nil {
+ return err
+ }
+
+ request := &DeleteGroupsRequest{
+ Groups: []string{group},
+ }
+
+ resp, err := coordinator.DeleteGroups(request)
+ if err != nil {
+ return err
+ }
+
+ groupErr, ok := resp.GroupErrorCodes[group]
+ if !ok {
+ return ErrIncompleteResponse
+ }
+
+ if groupErr != ErrNoError {
+ return groupErr
+ }
+
+ return nil
+}
diff --git a/vendor/github.com/Shopify/sarama/alter_configs_request.go b/vendor/github.com/Shopify/sarama/alter_configs_request.go
index 48c44ea..26c275b 100644
--- a/vendor/github.com/Shopify/sarama/alter_configs_request.go
+++ b/vendor/github.com/Shopify/sarama/alter_configs_request.go
@@ -1,45 +1,47 @@
package sarama
+//AlterConfigsRequest is an alter config request type
type AlterConfigsRequest struct {
Resources []*AlterConfigsResource
ValidateOnly bool
}
+//AlterConfigsResource is an alter config resource type
type AlterConfigsResource struct {
Type ConfigResourceType
Name string
ConfigEntries map[string]*string
}
-func (acr *AlterConfigsRequest) encode(pe packetEncoder) error {
- if err := pe.putArrayLength(len(acr.Resources)); err != nil {
+func (a *AlterConfigsRequest) encode(pe packetEncoder) error {
+ if err := pe.putArrayLength(len(a.Resources)); err != nil {
return err
}
- for _, r := range acr.Resources {
+ for _, r := range a.Resources {
if err := r.encode(pe); err != nil {
return err
}
}
- pe.putBool(acr.ValidateOnly)
+ pe.putBool(a.ValidateOnly)
return nil
}
-func (acr *AlterConfigsRequest) decode(pd packetDecoder, version int16) error {
+func (a *AlterConfigsRequest) decode(pd packetDecoder, version int16) error {
resourceCount, err := pd.getArrayLength()
if err != nil {
return err
}
- acr.Resources = make([]*AlterConfigsResource, resourceCount)
- for i := range acr.Resources {
+ a.Resources = make([]*AlterConfigsResource, resourceCount)
+ for i := range a.Resources {
r := &AlterConfigsResource{}
err = r.decode(pd, version)
if err != nil {
return err
}
- acr.Resources[i] = r
+ a.Resources[i] = r
}
validateOnly, err := pd.getBool()
@@ -47,22 +49,22 @@
return err
}
- acr.ValidateOnly = validateOnly
+ a.ValidateOnly = validateOnly
return nil
}
-func (ac *AlterConfigsResource) encode(pe packetEncoder) error {
- pe.putInt8(int8(ac.Type))
+func (a *AlterConfigsResource) encode(pe packetEncoder) error {
+ pe.putInt8(int8(a.Type))
- if err := pe.putString(ac.Name); err != nil {
+ if err := pe.putString(a.Name); err != nil {
return err
}
- if err := pe.putArrayLength(len(ac.ConfigEntries)); err != nil {
+ if err := pe.putArrayLength(len(a.ConfigEntries)); err != nil {
return err
}
- for configKey, configValue := range ac.ConfigEntries {
+ for configKey, configValue := range a.ConfigEntries {
if err := pe.putString(configKey); err != nil {
return err
}
@@ -74,18 +76,18 @@
return nil
}
-func (ac *AlterConfigsResource) decode(pd packetDecoder, version int16) error {
+func (a *AlterConfigsResource) decode(pd packetDecoder, version int16) error {
t, err := pd.getInt8()
if err != nil {
return err
}
- ac.Type = ConfigResourceType(t)
+ a.Type = ConfigResourceType(t)
name, err := pd.getString()
if err != nil {
return err
}
- ac.Name = name
+ a.Name = name
n, err := pd.getArrayLength()
if err != nil {
@@ -93,13 +95,13 @@
}
if n > 0 {
- ac.ConfigEntries = make(map[string]*string, n)
+ a.ConfigEntries = make(map[string]*string, n)
for i := 0; i < n; i++ {
configKey, err := pd.getString()
if err != nil {
return err
}
- if ac.ConfigEntries[configKey], err = pd.getNullableString(); err != nil {
+ if a.ConfigEntries[configKey], err = pd.getNullableString(); err != nil {
return err
}
}
@@ -107,14 +109,14 @@
return err
}
-func (acr *AlterConfigsRequest) key() int16 {
+func (a *AlterConfigsRequest) key() int16 {
return 33
}
-func (acr *AlterConfigsRequest) version() int16 {
+func (a *AlterConfigsRequest) version() int16 {
return 0
}
-func (acr *AlterConfigsRequest) requiredVersion() KafkaVersion {
+func (a *AlterConfigsRequest) requiredVersion() KafkaVersion {
return V0_11_0_0
}
diff --git a/vendor/github.com/Shopify/sarama/alter_configs_response.go b/vendor/github.com/Shopify/sarama/alter_configs_response.go
index 29b09e1..3893663 100644
--- a/vendor/github.com/Shopify/sarama/alter_configs_response.go
+++ b/vendor/github.com/Shopify/sarama/alter_configs_response.go
@@ -2,11 +2,13 @@
import "time"
+//AlterConfigsResponse is a reponse type for alter config
type AlterConfigsResponse struct {
ThrottleTime time.Duration
Resources []*AlterConfigsResourceResponse
}
+//AlterConfigsResourceResponse is a reponse type for alter config resource
type AlterConfigsResourceResponse struct {
ErrorCode int16
ErrorMsg string
@@ -14,21 +16,21 @@
Name string
}
-func (ct *AlterConfigsResponse) encode(pe packetEncoder) error {
- pe.putInt32(int32(ct.ThrottleTime / time.Millisecond))
+func (a *AlterConfigsResponse) encode(pe packetEncoder) error {
+ pe.putInt32(int32(a.ThrottleTime / time.Millisecond))
- if err := pe.putArrayLength(len(ct.Resources)); err != nil {
+ if err := pe.putArrayLength(len(a.Resources)); err != nil {
return err
}
- for i := range ct.Resources {
- pe.putInt16(ct.Resources[i].ErrorCode)
- err := pe.putString(ct.Resources[i].ErrorMsg)
+ for i := range a.Resources {
+ pe.putInt16(a.Resources[i].ErrorCode)
+ err := pe.putString(a.Resources[i].ErrorMsg)
if err != nil {
return nil
}
- pe.putInt8(int8(ct.Resources[i].Type))
- err = pe.putString(ct.Resources[i].Name)
+ pe.putInt8(int8(a.Resources[i].Type))
+ err = pe.putString(a.Resources[i].Name)
if err != nil {
return nil
}
@@ -37,59 +39,59 @@
return nil
}
-func (acr *AlterConfigsResponse) decode(pd packetDecoder, version int16) error {
+func (a *AlterConfigsResponse) decode(pd packetDecoder, version int16) error {
throttleTime, err := pd.getInt32()
if err != nil {
return err
}
- acr.ThrottleTime = time.Duration(throttleTime) * time.Millisecond
+ a.ThrottleTime = time.Duration(throttleTime) * time.Millisecond
responseCount, err := pd.getArrayLength()
if err != nil {
return err
}
- acr.Resources = make([]*AlterConfigsResourceResponse, responseCount)
+ a.Resources = make([]*AlterConfigsResourceResponse, responseCount)
- for i := range acr.Resources {
- acr.Resources[i] = new(AlterConfigsResourceResponse)
+ for i := range a.Resources {
+ a.Resources[i] = new(AlterConfigsResourceResponse)
errCode, err := pd.getInt16()
if err != nil {
return err
}
- acr.Resources[i].ErrorCode = errCode
+ a.Resources[i].ErrorCode = errCode
e, err := pd.getString()
if err != nil {
return err
}
- acr.Resources[i].ErrorMsg = e
+ a.Resources[i].ErrorMsg = e
t, err := pd.getInt8()
if err != nil {
return err
}
- acr.Resources[i].Type = ConfigResourceType(t)
+ a.Resources[i].Type = ConfigResourceType(t)
name, err := pd.getString()
if err != nil {
return err
}
- acr.Resources[i].Name = name
+ a.Resources[i].Name = name
}
return nil
}
-func (r *AlterConfigsResponse) key() int16 {
+func (a *AlterConfigsResponse) key() int16 {
return 32
}
-func (r *AlterConfigsResponse) version() int16 {
+func (a *AlterConfigsResponse) version() int16 {
return 0
}
-func (r *AlterConfigsResponse) requiredVersion() KafkaVersion {
+func (a *AlterConfigsResponse) requiredVersion() KafkaVersion {
return V0_11_0_0
}
diff --git a/vendor/github.com/Shopify/sarama/api_versions_request.go b/vendor/github.com/Shopify/sarama/api_versions_request.go
index ab65f01..b33167c 100644
--- a/vendor/github.com/Shopify/sarama/api_versions_request.go
+++ b/vendor/github.com/Shopify/sarama/api_versions_request.go
@@ -1,24 +1,25 @@
package sarama
+//ApiVersionsRequest ...
type ApiVersionsRequest struct {
}
-func (r *ApiVersionsRequest) encode(pe packetEncoder) error {
+func (a *ApiVersionsRequest) encode(pe packetEncoder) error {
return nil
}
-func (r *ApiVersionsRequest) decode(pd packetDecoder, version int16) (err error) {
+func (a *ApiVersionsRequest) decode(pd packetDecoder, version int16) (err error) {
return nil
}
-func (r *ApiVersionsRequest) key() int16 {
+func (a *ApiVersionsRequest) key() int16 {
return 18
}
-func (r *ApiVersionsRequest) version() int16 {
+func (a *ApiVersionsRequest) version() int16 {
return 0
}
-func (r *ApiVersionsRequest) requiredVersion() KafkaVersion {
+func (a *ApiVersionsRequest) requiredVersion() KafkaVersion {
return V0_10_0_0
}
diff --git a/vendor/github.com/Shopify/sarama/api_versions_response.go b/vendor/github.com/Shopify/sarama/api_versions_response.go
index 23bc326..bb1f0b3 100644
--- a/vendor/github.com/Shopify/sarama/api_versions_response.go
+++ b/vendor/github.com/Shopify/sarama/api_versions_response.go
@@ -1,5 +1,6 @@
package sarama
+//ApiVersionsResponseBlock is an api version reponse block type
type ApiVersionsResponseBlock struct {
ApiKey int16
MinVersion int16
@@ -31,6 +32,7 @@
return nil
}
+//ApiVersionsResponse is an api version response type
type ApiVersionsResponse struct {
Err KError
ApiVersions []*ApiVersionsResponseBlock
diff --git a/vendor/github.com/Shopify/sarama/async_producer.go b/vendor/github.com/Shopify/sarama/async_producer.go
index 5174a35..11e0849 100644
--- a/vendor/github.com/Shopify/sarama/async_producer.go
+++ b/vendor/github.com/Shopify/sarama/async_producer.go
@@ -92,9 +92,8 @@
}
type asyncProducer struct {
- client Client
- conf *Config
- ownClient bool
+ client Client
+ conf *Config
errors chan *ProducerError
input, successes, retries chan *ProducerMessage
@@ -113,18 +112,19 @@
if err != nil {
return nil, err
}
-
- p, err := NewAsyncProducerFromClient(client)
- if err != nil {
- return nil, err
- }
- p.(*asyncProducer).ownClient = true
- return p, nil
+ return newAsyncProducer(client)
}
// NewAsyncProducerFromClient creates a new Producer using the given client. It is still
// necessary to call Close() on the underlying client when shutting down this producer.
func NewAsyncProducerFromClient(client Client) (AsyncProducer, error) {
+ // For clients passed in by the client, ensure we don't
+ // call Close() on it.
+ cli := &nopCloserClient{client}
+ return newAsyncProducer(cli)
+}
+
+func newAsyncProducer(client Client) (AsyncProducer, error) {
// Check that we are not dealing with a closed Client before processing any other arguments
if client.Closed() {
return nil, ErrClosedClient
@@ -191,10 +191,17 @@
// Partition is the partition that the message was sent to. This is only
// guaranteed to be defined if the message was successfully delivered.
Partition int32
- // Timestamp is the timestamp assigned to the message by the broker. This
- // is only guaranteed to be defined if the message was successfully
- // delivered, RequiredAcks is not NoResponse, and the Kafka broker is at
- // least version 0.10.0.
+ // Timestamp can vary in behaviour depending on broker configuration, being
+ // in either one of the CreateTime or LogAppendTime modes (default CreateTime),
+ // and requiring version at least 0.10.0.
+ //
+ // When configured to CreateTime, the timestamp is specified by the producer
+ // either by explicitly setting this field, or when the message is added
+ // to a produce set.
+ //
+ // When configured to LogAppendTime, the timestamp assigned to the message
+ // by the broker. This is only guaranteed to be defined if the message was
+ // successfully delivered and RequiredAcks is not NoResponse.
Timestamp time.Time
retries int
@@ -999,11 +1006,9 @@
p.inFlight.Wait()
- if p.ownClient {
- err := p.client.Close()
- if err != nil {
- Logger.Println("producer/shutdown failed to close the embedded client:", err)
- }
+ err := p.client.Close()
+ if err != nil {
+ Logger.Println("producer/shutdown failed to close the embedded client:", err)
}
close(p.input)
diff --git a/vendor/github.com/Shopify/sarama/balance_strategy.go b/vendor/github.com/Shopify/sarama/balance_strategy.go
index e78988d..2fce17f 100644
--- a/vendor/github.com/Shopify/sarama/balance_strategy.go
+++ b/vendor/github.com/Shopify/sarama/balance_strategy.go
@@ -24,7 +24,7 @@
// --------------------------------------------------------------------
// BalanceStrategy is used to balance topics and partitions
-// across memebers of a consumer group
+// across members of a consumer group
type BalanceStrategy interface {
// Name uniquely identifies the strategy.
Name() string
@@ -78,7 +78,7 @@
// Name implements BalanceStrategy.
func (s *balanceStrategy) Name() string { return s.name }
-// Balance implements BalanceStrategy.
+// Plan implements BalanceStrategy.
func (s *balanceStrategy) Plan(members map[string]ConsumerGroupMemberMetadata, topics map[string][]int32) (BalanceStrategyPlan, error) {
// Build members by topic map
mbt := make(map[string][]string)
diff --git a/vendor/github.com/Shopify/sarama/broker.go b/vendor/github.com/Shopify/sarama/broker.go
index 9129089..9c3e5a0 100644
--- a/vendor/github.com/Shopify/sarama/broker.go
+++ b/vendor/github.com/Shopify/sarama/broker.go
@@ -13,24 +13,25 @@
"sync/atomic"
"time"
- "github.com/rcrowley/go-metrics"
+ metrics "github.com/rcrowley/go-metrics"
)
// Broker represents a single Kafka broker connection. All operations on this object are entirely concurrency-safe.
type Broker struct {
- id int32
- addr string
+ conf *Config
rack *string
- conf *Config
+ id int32
+ addr string
correlationID int32
conn net.Conn
connErr error
lock sync.Mutex
opened int32
+ responses chan responsePromise
+ done chan bool
- responses chan responsePromise
- done chan bool
+ registeredMetrics []string
incomingByteRate metrics.Meter
requestRate metrics.Meter
@@ -46,6 +47,8 @@
brokerOutgoingByteRate metrics.Meter
brokerResponseRate metrics.Meter
brokerResponseSize metrics.Histogram
+
+ kerberosAuthenticator GSSAPIKerberosAuth
}
// SASLMechanism specifies the SASL mechanism the client uses to authenticate with the broker
@@ -56,6 +59,11 @@
SASLTypeOAuth = "OAUTHBEARER"
// SASLTypePlaintext represents the SASL/PLAIN mechanism
SASLTypePlaintext = "PLAIN"
+ // SASLTypeSCRAMSHA256 represents the SCRAM-SHA-256 mechanism.
+ SASLTypeSCRAMSHA256 = "SCRAM-SHA-256"
+ // SASLTypeSCRAMSHA512 represents the SCRAM-SHA-512 mechanism.
+ SASLTypeSCRAMSHA512 = "SCRAM-SHA-512"
+ SASLTypeGSSAPI = "GSSAPI"
// SASLHandshakeV0 is v0 of the Kafka SASL handshake protocol. Client and
// server negotiate SASL auth using opaque packets.
SASLHandshakeV0 = int16(0)
@@ -92,6 +100,20 @@
Token() (*AccessToken, error)
}
+// SCRAMClient is a an interface to a SCRAM
+// client implementation.
+type SCRAMClient interface {
+ // Begin prepares the client for the SCRAM exchange
+ // with the server with a user name and a password
+ Begin(userName, password, authzID string) error
+ // Step steps client through the SCRAM exchange. It is
+ // called repeatedly until it errors or `Done` returns true.
+ Step(challenge string) (response string, err error)
+ // Done should return true when the SCRAM conversation
+ // is over.
+ Done() bool
+}
+
type responsePromise struct {
requestTime time.Time
correlationID int32
@@ -137,6 +159,8 @@
if conf.Net.TLS.Enable {
b.conn, b.connErr = tls.DialWithDialer(&dialer, "tcp", b.addr, conf.Net.TLS.Config)
+ } else if conf.Net.Proxy.Enable {
+ b.conn, b.connErr = conf.Net.Proxy.Dialer.Dial("tcp", b.addr)
} else {
b.conn, b.connErr = dialer.Dial("tcp", b.addr)
}
@@ -161,13 +185,7 @@
// Do not gather metrics for seeded broker (only used during bootstrap) because they share
// the same id (-1) and are already exposed through the global metrics above
if b.id >= 0 {
- b.brokerIncomingByteRate = getOrRegisterBrokerMeter("incoming-byte-rate", b, conf.MetricRegistry)
- b.brokerRequestRate = getOrRegisterBrokerMeter("request-rate", b, conf.MetricRegistry)
- b.brokerRequestSize = getOrRegisterBrokerHistogram("request-size", b, conf.MetricRegistry)
- b.brokerRequestLatency = getOrRegisterBrokerHistogram("request-latency-in-ms", b, conf.MetricRegistry)
- b.brokerOutgoingByteRate = getOrRegisterBrokerMeter("outgoing-byte-rate", b, conf.MetricRegistry)
- b.brokerResponseRate = getOrRegisterBrokerMeter("response-rate", b, conf.MetricRegistry)
- b.brokerResponseSize = getOrRegisterBrokerHistogram("response-size", b, conf.MetricRegistry)
+ b.registerMetrics()
}
if conf.Net.SASL.Enable {
@@ -210,6 +228,7 @@
return b.conn != nil, b.connErr
}
+//Close closes the broker resources
func (b *Broker) Close() error {
b.lock.Lock()
defer b.lock.Unlock()
@@ -228,12 +247,7 @@
b.done = nil
b.responses = nil
- if b.id >= 0 {
- b.conf.MetricRegistry.Unregister(getMetricNameForBroker("incoming-byte-rate", b))
- b.conf.MetricRegistry.Unregister(getMetricNameForBroker("request-rate", b))
- b.conf.MetricRegistry.Unregister(getMetricNameForBroker("outgoing-byte-rate", b))
- b.conf.MetricRegistry.Unregister(getMetricNameForBroker("response-rate", b))
- }
+ b.unregisterMetrics()
if err == nil {
Logger.Printf("Closed connection to broker %s\n", b.addr)
@@ -267,6 +281,7 @@
return *b.rack
}
+//GetMetadata send a metadata request and returns a metadata response or error
func (b *Broker) GetMetadata(request *MetadataRequest) (*MetadataResponse, error) {
response := new(MetadataResponse)
@@ -279,6 +294,7 @@
return response, nil
}
+//GetConsumerMetadata send a consumer metadata request and returns a consumer metadata response or error
func (b *Broker) GetConsumerMetadata(request *ConsumerMetadataRequest) (*ConsumerMetadataResponse, error) {
response := new(ConsumerMetadataResponse)
@@ -291,6 +307,7 @@
return response, nil
}
+//FindCoordinator sends a find coordinate request and returns a response or error
func (b *Broker) FindCoordinator(request *FindCoordinatorRequest) (*FindCoordinatorResponse, error) {
response := new(FindCoordinatorResponse)
@@ -303,6 +320,7 @@
return response, nil
}
+//GetAvailableOffsets return an offset response or error
func (b *Broker) GetAvailableOffsets(request *OffsetRequest) (*OffsetResponse, error) {
response := new(OffsetResponse)
@@ -315,9 +333,12 @@
return response, nil
}
+//Produce returns a produce response or error
func (b *Broker) Produce(request *ProduceRequest) (*ProduceResponse, error) {
- var response *ProduceResponse
- var err error
+ var (
+ response *ProduceResponse
+ err error
+ )
if request.RequiredAcks == NoResponse {
err = b.sendAndReceive(request, nil)
@@ -333,11 +354,11 @@
return response, nil
}
+//Fetch returns a FetchResponse or error
func (b *Broker) Fetch(request *FetchRequest) (*FetchResponse, error) {
response := new(FetchResponse)
err := b.sendAndReceive(request, response)
-
if err != nil {
return nil, err
}
@@ -345,11 +366,11 @@
return response, nil
}
+//CommitOffset return an Offset commit reponse or error
func (b *Broker) CommitOffset(request *OffsetCommitRequest) (*OffsetCommitResponse, error) {
response := new(OffsetCommitResponse)
err := b.sendAndReceive(request, response)
-
if err != nil {
return nil, err
}
@@ -357,11 +378,11 @@
return response, nil
}
+//FetchOffset returns an offset fetch response or error
func (b *Broker) FetchOffset(request *OffsetFetchRequest) (*OffsetFetchResponse, error) {
response := new(OffsetFetchResponse)
err := b.sendAndReceive(request, response)
-
if err != nil {
return nil, err
}
@@ -369,6 +390,7 @@
return response, nil
}
+//JoinGroup returns a join group response or error
func (b *Broker) JoinGroup(request *JoinGroupRequest) (*JoinGroupResponse, error) {
response := new(JoinGroupResponse)
@@ -380,6 +402,7 @@
return response, nil
}
+//SyncGroup returns a sync group response or error
func (b *Broker) SyncGroup(request *SyncGroupRequest) (*SyncGroupResponse, error) {
response := new(SyncGroupResponse)
@@ -391,6 +414,7 @@
return response, nil
}
+//LeaveGroup return a leave group response or error
func (b *Broker) LeaveGroup(request *LeaveGroupRequest) (*LeaveGroupResponse, error) {
response := new(LeaveGroupResponse)
@@ -402,6 +426,7 @@
return response, nil
}
+//Heartbeat returns a heartbeat response or error
func (b *Broker) Heartbeat(request *HeartbeatRequest) (*HeartbeatResponse, error) {
response := new(HeartbeatResponse)
@@ -413,6 +438,7 @@
return response, nil
}
+//ListGroups return a list group response or error
func (b *Broker) ListGroups(request *ListGroupsRequest) (*ListGroupsResponse, error) {
response := new(ListGroupsResponse)
@@ -424,6 +450,7 @@
return response, nil
}
+//DescribeGroups return describe group response or error
func (b *Broker) DescribeGroups(request *DescribeGroupsRequest) (*DescribeGroupsResponse, error) {
response := new(DescribeGroupsResponse)
@@ -435,6 +462,7 @@
return response, nil
}
+//ApiVersions return api version response or error
func (b *Broker) ApiVersions(request *ApiVersionsRequest) (*ApiVersionsResponse, error) {
response := new(ApiVersionsResponse)
@@ -446,6 +474,7 @@
return response, nil
}
+//CreateTopics send a create topic request and returns create topic response
func (b *Broker) CreateTopics(request *CreateTopicsRequest) (*CreateTopicsResponse, error) {
response := new(CreateTopicsResponse)
@@ -457,6 +486,7 @@
return response, nil
}
+//DeleteTopics sends a delete topic request and returns delete topic response
func (b *Broker) DeleteTopics(request *DeleteTopicsRequest) (*DeleteTopicsResponse, error) {
response := new(DeleteTopicsResponse)
@@ -468,6 +498,8 @@
return response, nil
}
+//CreatePartitions sends a create partition request and returns create
+//partitions response or error
func (b *Broker) CreatePartitions(request *CreatePartitionsRequest) (*CreatePartitionsResponse, error) {
response := new(CreatePartitionsResponse)
@@ -479,6 +511,8 @@
return response, nil
}
+//DeleteRecords send a request to delete records and return delete record
+//response or error
func (b *Broker) DeleteRecords(request *DeleteRecordsRequest) (*DeleteRecordsResponse, error) {
response := new(DeleteRecordsResponse)
@@ -490,6 +524,7 @@
return response, nil
}
+//DescribeAcls sends a describe acl request and returns a response or error
func (b *Broker) DescribeAcls(request *DescribeAclsRequest) (*DescribeAclsResponse, error) {
response := new(DescribeAclsResponse)
@@ -501,6 +536,7 @@
return response, nil
}
+//CreateAcls sends a create acl request and returns a response or error
func (b *Broker) CreateAcls(request *CreateAclsRequest) (*CreateAclsResponse, error) {
response := new(CreateAclsResponse)
@@ -512,6 +548,7 @@
return response, nil
}
+//DeleteAcls sends a delete acl request and returns a response or error
func (b *Broker) DeleteAcls(request *DeleteAclsRequest) (*DeleteAclsResponse, error) {
response := new(DeleteAclsResponse)
@@ -523,6 +560,7 @@
return response, nil
}
+//InitProducerID sends an init producer request and returns a response or error
func (b *Broker) InitProducerID(request *InitProducerIDRequest) (*InitProducerIDResponse, error) {
response := new(InitProducerIDResponse)
@@ -534,6 +572,8 @@
return response, nil
}
+//AddPartitionsToTxn send a request to add partition to txn and returns
+//a response or error
func (b *Broker) AddPartitionsToTxn(request *AddPartitionsToTxnRequest) (*AddPartitionsToTxnResponse, error) {
response := new(AddPartitionsToTxnResponse)
@@ -545,6 +585,8 @@
return response, nil
}
+//AddOffsetsToTxn sends a request to add offsets to txn and returns a response
+//or error
func (b *Broker) AddOffsetsToTxn(request *AddOffsetsToTxnRequest) (*AddOffsetsToTxnResponse, error) {
response := new(AddOffsetsToTxnResponse)
@@ -556,6 +598,7 @@
return response, nil
}
+//EndTxn sends a request to end txn and returns a response or error
func (b *Broker) EndTxn(request *EndTxnRequest) (*EndTxnResponse, error) {
response := new(EndTxnResponse)
@@ -567,6 +610,8 @@
return response, nil
}
+//TxnOffsetCommit sends a request to commit transaction offsets and returns
+//a response or error
func (b *Broker) TxnOffsetCommit(request *TxnOffsetCommitRequest) (*TxnOffsetCommitResponse, error) {
response := new(TxnOffsetCommitResponse)
@@ -578,6 +623,8 @@
return response, nil
}
+//DescribeConfigs sends a request to describe config and returns a response or
+//error
func (b *Broker) DescribeConfigs(request *DescribeConfigsRequest) (*DescribeConfigsResponse, error) {
response := new(DescribeConfigsResponse)
@@ -589,6 +636,7 @@
return response, nil
}
+//AlterConfigs sends a request to alter config and return a response or error
func (b *Broker) AlterConfigs(request *AlterConfigsRequest) (*AlterConfigsResponse, error) {
response := new(AlterConfigsResponse)
@@ -600,6 +648,7 @@
return response, nil
}
+//DeleteGroups sends a request to delete groups and returns a response or error
func (b *Broker) DeleteGroups(request *DeleteGroupsRequest) (*DeleteGroupsResponse, error) {
response := new(DeleteGroupsResponse)
@@ -638,7 +687,7 @@
requestTime := time.Now()
bytes, err := b.conn.Write(buf)
- b.updateOutgoingCommunicationMetrics(bytes)
+ b.updateOutgoingCommunicationMetrics(bytes) //TODO: should it be after error check
if err != nil {
return nil, err
}
@@ -658,7 +707,6 @@
func (b *Broker) sendAndReceive(req protocolBody, res versionedDecoder) error {
promise, err := b.send(req, res != nil)
-
if err != nil {
return err
}
@@ -707,11 +755,11 @@
}
func (b *Broker) encode(pe packetEncoder, version int16) (err error) {
-
host, portstr, err := net.SplitHostPort(b.addr)
if err != nil {
return err
}
+
port, err := strconv.Atoi(portstr)
if err != nil {
return err
@@ -739,6 +787,7 @@
func (b *Broker) responseReceiver() {
var dead error
header := make([]byte, 8)
+
for response := range b.responses {
if dead != nil {
response.errors <- dead
@@ -793,14 +842,28 @@
}
func (b *Broker) authenticateViaSASL() error {
- if b.conf.Net.SASL.Mechanism == SASLTypeOAuth {
+ switch b.conf.Net.SASL.Mechanism {
+ case SASLTypeOAuth:
return b.sendAndReceiveSASLOAuth(b.conf.Net.SASL.TokenProvider)
+ case SASLTypeSCRAMSHA256, SASLTypeSCRAMSHA512:
+ return b.sendAndReceiveSASLSCRAMv1()
+ case SASLTypeGSSAPI:
+ return b.sendAndReceiveKerberos()
+ default:
+ return b.sendAndReceiveSASLPlainAuth()
}
- return b.sendAndReceiveSASLPlainAuth()
}
-func (b *Broker) sendAndReceiveSASLHandshake(saslType string, version int16) error {
- rb := &SaslHandshakeRequest{Mechanism: saslType, Version: version}
+func (b *Broker) sendAndReceiveKerberos() error {
+ b.kerberosAuthenticator.Config = &b.conf.Net.SASL.GSSAPI
+ if b.kerberosAuthenticator.NewKerberosClientFunc == nil {
+ b.kerberosAuthenticator.NewKerberosClientFunc = NewKerberosClient
+ }
+ return b.kerberosAuthenticator.Authorize(b)
+}
+
+func (b *Broker) sendAndReceiveSASLHandshake(saslType SASLMechanism, version int16) error {
+ rb := &SaslHandshakeRequest{Mechanism: string(saslType), Version: version}
req := &request{correlationID: b.correlationID, clientID: b.conf.ClientID, body: rb}
buf, err := encode(req, b.conf.MetricRegistry)
@@ -828,6 +891,7 @@
Logger.Printf("Failed to read SASL handshake header : %s\n", err.Error())
return err
}
+
length := binary.BigEndian.Uint32(header[:4])
payload := make([]byte, length-4)
n, err := io.ReadFull(b.conn, payload)
@@ -835,23 +899,29 @@
Logger.Printf("Failed to read SASL handshake payload : %s\n", err.Error())
return err
}
+
b.updateIncomingCommunicationMetrics(n+8, time.Since(requestTime))
res := &SaslHandshakeResponse{}
+
err = versionedDecode(payload, res, 0)
if err != nil {
Logger.Printf("Failed to parse SASL handshake : %s\n", err.Error())
return err
}
+
if res.Err != ErrNoError {
Logger.Printf("Invalid SASL Mechanism : %s\n", res.Err.Error())
return res.Err
}
- Logger.Print("Successful SASL handshake")
+
+ Logger.Print("Successful SASL handshake. Available mechanisms: ", res.EnabledMechanisms)
return nil
}
-// Kafka 0.10.0 plans to support SASL Plain and Kerberos as per PR #812 (KIP-43)/(JIRA KAFKA-3149)
-// Some hosted kafka services such as IBM Message Hub already offer SASL/PLAIN auth with Kafka 0.9
+// Kafka 0.10.x supported SASL PLAIN/Kerberos via KAFKA-3149 (KIP-43).
+// Kafka 1.x.x onward added a SaslAuthenticate request/response message which
+// wraps the SASL flow in the Kafka protocol, which allows for returning
+// meaningful errors on authentication failure.
//
// In SASL Plain, Kafka expects the auth header to be in the following format
// Message format (from https://tools.ietf.org/html/rfc4616):
@@ -865,17 +935,34 @@
// SAFE = UTF1 / UTF2 / UTF3 / UTF4
// ;; any UTF-8 encoded Unicode character except NUL
//
+// With SASL v0 handshake and auth then:
// When credentials are valid, Kafka returns a 4 byte array of null characters.
-// When credentials are invalid, Kafka closes the connection. This does not seem to be the ideal way
-// of responding to bad credentials but thats how its being done today.
+// When credentials are invalid, Kafka closes the connection.
+//
+// With SASL v1 handshake and auth then:
+// When credentials are invalid, Kafka replies with a SaslAuthenticate response
+// containing an error code and message detailing the authentication failure.
func (b *Broker) sendAndReceiveSASLPlainAuth() error {
+ // default to V0 to allow for backward compatability when SASL is enabled
+ // but not the handshake
if b.conf.Net.SASL.Handshake {
- handshakeErr := b.sendAndReceiveSASLHandshake(SASLTypePlaintext, SASLHandshakeV0)
+
+ handshakeErr := b.sendAndReceiveSASLHandshake(SASLTypePlaintext, b.conf.Net.SASL.Version)
if handshakeErr != nil {
Logger.Printf("Error while performing SASL handshake %s\n", b.addr)
return handshakeErr
}
}
+
+ if b.conf.Net.SASL.Version == SASLHandshakeV1 {
+ return b.sendAndReceiveV1SASLPlainAuth()
+ }
+ return b.sendAndReceiveV0SASLPlainAuth()
+}
+
+// sendAndReceiveV0SASLPlainAuth flows the v0 sasl auth NOT wrapped in the kafka protocol
+func (b *Broker) sendAndReceiveV0SASLPlainAuth() error {
+
length := 1 + len(b.conf.Net.SASL.User) + 1 + len(b.conf.Net.SASL.Password)
authBytes := make([]byte, length+4) //4 byte length header + auth data
binary.BigEndian.PutUint32(authBytes, uint32(length))
@@ -909,55 +996,197 @@
return nil
}
+// sendAndReceiveV1SASLPlainAuth flows the v1 sasl authentication using the kafka protocol
+func (b *Broker) sendAndReceiveV1SASLPlainAuth() error {
+ correlationID := b.correlationID
+
+ requestTime := time.Now()
+
+ bytesWritten, err := b.sendSASLPlainAuthClientResponse(correlationID)
+
+ b.updateOutgoingCommunicationMetrics(bytesWritten)
+
+ if err != nil {
+ Logger.Printf("Failed to write SASL auth header to broker %s: %s\n", b.addr, err.Error())
+ return err
+ }
+
+ b.correlationID++
+
+ bytesRead, err := b.receiveSASLServerResponse(&SaslAuthenticateResponse{}, correlationID)
+ b.updateIncomingCommunicationMetrics(bytesRead, time.Since(requestTime))
+
+ // With v1 sasl we get an error message set in the response we can return
+ if err != nil {
+ Logger.Printf("Error returned from broker during SASL flow %s: %s\n", b.addr, err.Error())
+ return err
+ }
+
+ return nil
+}
+
// sendAndReceiveSASLOAuth performs the authentication flow as described by KIP-255
// https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=75968876
func (b *Broker) sendAndReceiveSASLOAuth(provider AccessTokenProvider) error {
-
if err := b.sendAndReceiveSASLHandshake(SASLTypeOAuth, SASLHandshakeV1); err != nil {
return err
}
token, err := provider.Token()
-
if err != nil {
return err
}
- requestTime := time.Now()
-
- correlationID := b.correlationID
-
- bytesWritten, err := b.sendSASLOAuthBearerClientResponse(token, correlationID)
-
+ message, err := buildClientFirstMessage(token)
if err != nil {
return err
}
+ challenged, err := b.sendClientMessage(message)
+ if err != nil {
+ return err
+ }
+
+ if challenged {
+ // Abort the token exchange. The broker returns the failure code.
+ _, err = b.sendClientMessage([]byte(`\x01`))
+ }
+
+ return err
+}
+
+// sendClientMessage sends a SASL/OAUTHBEARER client message and returns true
+// if the broker responds with a challenge, in which case the token is
+// rejected.
+func (b *Broker) sendClientMessage(message []byte) (bool, error) {
+
+ requestTime := time.Now()
+ correlationID := b.correlationID
+
+ bytesWritten, err := b.sendSASLOAuthBearerClientMessage(message, correlationID)
+ if err != nil {
+ return false, err
+ }
+
b.updateOutgoingCommunicationMetrics(bytesWritten)
-
b.correlationID++
- bytesRead, err := b.receiveSASLOAuthBearerServerResponse(correlationID)
-
- if err != nil {
- return err
- }
+ res := &SaslAuthenticateResponse{}
+ bytesRead, err := b.receiveSASLServerResponse(res, correlationID)
requestLatency := time.Since(requestTime)
b.updateIncomingCommunicationMetrics(bytesRead, requestLatency)
+ isChallenge := len(res.SaslAuthBytes) > 0
+
+ if isChallenge && err != nil {
+ Logger.Printf("Broker rejected authentication token: %s", res.SaslAuthBytes)
+ }
+
+ return isChallenge, err
+}
+
+func (b *Broker) sendAndReceiveSASLSCRAMv1() error {
+ if err := b.sendAndReceiveSASLHandshake(b.conf.Net.SASL.Mechanism, SASLHandshakeV1); err != nil {
+ return err
+ }
+
+ scramClient := b.conf.Net.SASL.SCRAMClientGeneratorFunc()
+ if err := scramClient.Begin(b.conf.Net.SASL.User, b.conf.Net.SASL.Password, b.conf.Net.SASL.SCRAMAuthzID); err != nil {
+ return fmt.Errorf("failed to start SCRAM exchange with the server: %s", err.Error())
+ }
+
+ msg, err := scramClient.Step("")
+ if err != nil {
+ return fmt.Errorf("failed to advance the SCRAM exchange: %s", err.Error())
+
+ }
+
+ for !scramClient.Done() {
+ requestTime := time.Now()
+ correlationID := b.correlationID
+ bytesWritten, err := b.sendSaslAuthenticateRequest(correlationID, []byte(msg))
+ if err != nil {
+ Logger.Printf("Failed to write SASL auth header to broker %s: %s\n", b.addr, err.Error())
+ return err
+ }
+
+ b.updateOutgoingCommunicationMetrics(bytesWritten)
+ b.correlationID++
+ challenge, err := b.receiveSaslAuthenticateResponse(correlationID)
+ if err != nil {
+ Logger.Printf("Failed to read response while authenticating with SASL to broker %s: %s\n", b.addr, err.Error())
+ return err
+ }
+
+ b.updateIncomingCommunicationMetrics(len(challenge), time.Since(requestTime))
+ msg, err = scramClient.Step(string(challenge))
+ if err != nil {
+ Logger.Println("SASL authentication failed", err)
+ return err
+ }
+ }
+
+ Logger.Println("SASL authentication succeeded")
return nil
}
+func (b *Broker) sendSaslAuthenticateRequest(correlationID int32, msg []byte) (int, error) {
+ rb := &SaslAuthenticateRequest{msg}
+ req := &request{correlationID: correlationID, clientID: b.conf.ClientID, body: rb}
+ buf, err := encode(req, b.conf.MetricRegistry)
+ if err != nil {
+ return 0, err
+ }
+
+ if err := b.conn.SetWriteDeadline(time.Now().Add(b.conf.Net.WriteTimeout)); err != nil {
+ return 0, err
+ }
+
+ return b.conn.Write(buf)
+}
+
+func (b *Broker) receiveSaslAuthenticateResponse(correlationID int32) ([]byte, error) {
+ buf := make([]byte, responseLengthSize+correlationIDSize)
+ _, err := io.ReadFull(b.conn, buf)
+ if err != nil {
+ return nil, err
+ }
+
+ header := responseHeader{}
+ err = decode(buf, &header)
+ if err != nil {
+ return nil, err
+ }
+
+ if header.correlationID != correlationID {
+ return nil, fmt.Errorf("correlation ID didn't match, wanted %d, got %d", b.correlationID, header.correlationID)
+ }
+
+ buf = make([]byte, header.length-correlationIDSize)
+ _, err = io.ReadFull(b.conn, buf)
+ if err != nil {
+ return nil, err
+ }
+
+ res := &SaslAuthenticateResponse{}
+ if err := versionedDecode(buf, res, 0); err != nil {
+ return nil, err
+ }
+ if res.Err != ErrNoError {
+ return nil, res.Err
+ }
+ return res.SaslAuthBytes, nil
+}
+
// Build SASL/OAUTHBEARER initial client response as described by RFC-7628
// https://tools.ietf.org/html/rfc7628
-func buildClientInitialResponse(token *AccessToken) ([]byte, error) {
-
+func buildClientFirstMessage(token *AccessToken) ([]byte, error) {
var ext string
if token.Extensions != nil && len(token.Extensions) > 0 {
if _, ok := token.Extensions[SASLExtKeyAuth]; ok {
- return []byte{}, fmt.Errorf("The extension `%s` is invalid", SASLExtKeyAuth)
+ return []byte{}, fmt.Errorf("the extension `%s` is invalid", SASLExtKeyAuth)
}
ext = "\x01" + mapToString(token.Extensions, "=", "\x01")
}
@@ -970,7 +1199,6 @@
// mapToString returns a list of key-value pairs ordered by key.
// keyValSep separates the key from the value. elemSep separates each pair.
func mapToString(extensions map[string]string, keyValSep string, elemSep string) string {
-
buf := make([]string, 0, len(extensions))
for k, v := range extensions {
@@ -982,20 +1210,30 @@
return strings.Join(buf, elemSep)
}
-func (b *Broker) sendSASLOAuthBearerClientResponse(token *AccessToken, correlationID int32) (int, error) {
-
- initialResp, err := buildClientInitialResponse(token)
-
+func (b *Broker) sendSASLPlainAuthClientResponse(correlationID int32) (int, error) {
+ authBytes := []byte("\x00" + b.conf.Net.SASL.User + "\x00" + b.conf.Net.SASL.Password)
+ rb := &SaslAuthenticateRequest{authBytes}
+ req := &request{correlationID: correlationID, clientID: b.conf.ClientID, body: rb}
+ buf, err := encode(req, b.conf.MetricRegistry)
if err != nil {
return 0, err
}
+ err = b.conn.SetWriteDeadline(time.Now().Add(b.conf.Net.WriteTimeout))
+ if err != nil {
+ Logger.Printf("Failed to set write deadline when doing SASL auth with broker %s: %s\n", b.addr, err.Error())
+ return 0, err
+ }
+ return b.conn.Write(buf)
+}
+
+func (b *Broker) sendSASLOAuthBearerClientMessage(initialResp []byte, correlationID int32) (int, error) {
+
rb := &SaslAuthenticateRequest{initialResp}
req := &request{correlationID: correlationID, clientID: b.conf.ClientID, body: rb}
buf, err := encode(req, b.conf.MetricRegistry)
-
if err != nil {
return 0, err
}
@@ -1007,12 +1245,11 @@
return b.conn.Write(buf)
}
-func (b *Broker) receiveSASLOAuthBearerServerResponse(correlationID int32) (int, error) {
+func (b *Broker) receiveSASLServerResponse(res *SaslAuthenticateResponse, correlationID int32) (int, error) {
- buf := make([]byte, 8)
+ buf := make([]byte, responseLengthSize+correlationIDSize)
bytesRead, err := io.ReadFull(b.conn, buf)
-
if err != nil {
return bytesRead, err
}
@@ -1020,7 +1257,6 @@
header := responseHeader{}
err = decode(buf, &header)
-
if err != nil {
return bytesRead, err
}
@@ -1029,48 +1265,39 @@
return bytesRead, fmt.Errorf("correlation ID didn't match, wanted %d, got %d", b.correlationID, header.correlationID)
}
- buf = make([]byte, header.length-4)
+ buf = make([]byte, header.length-correlationIDSize)
c, err := io.ReadFull(b.conn, buf)
-
bytesRead += c
-
if err != nil {
return bytesRead, err
}
- res := &SaslAuthenticateResponse{}
-
if err := versionedDecode(buf, res, 0); err != nil {
return bytesRead, err
}
- if err != nil {
- return bytesRead, err
- }
-
if res.Err != ErrNoError {
return bytesRead, res.Err
}
- if len(res.SaslAuthBytes) > 0 {
- Logger.Printf("Received SASL auth response: %s", res.SaslAuthBytes)
- }
-
return bytesRead, nil
}
func (b *Broker) updateIncomingCommunicationMetrics(bytes int, requestLatency time.Duration) {
b.updateRequestLatencyMetrics(requestLatency)
b.responseRate.Mark(1)
+
if b.brokerResponseRate != nil {
b.brokerResponseRate.Mark(1)
}
+
responseSize := int64(bytes)
b.incomingByteRate.Mark(responseSize)
if b.brokerIncomingByteRate != nil {
b.brokerIncomingByteRate.Mark(responseSize)
}
+
b.responseSize.Update(responseSize)
if b.brokerResponseSize != nil {
b.brokerResponseSize.Update(responseSize)
@@ -1080,9 +1307,11 @@
func (b *Broker) updateRequestLatencyMetrics(requestLatency time.Duration) {
requestLatencyInMs := int64(requestLatency / time.Millisecond)
b.requestLatency.Update(requestLatencyInMs)
+
if b.brokerRequestLatency != nil {
b.brokerRequestLatency.Update(requestLatencyInMs)
}
+
}
func (b *Broker) updateOutgoingCommunicationMetrics(bytes int) {
@@ -1090,13 +1319,44 @@
if b.brokerRequestRate != nil {
b.brokerRequestRate.Mark(1)
}
+
requestSize := int64(bytes)
b.outgoingByteRate.Mark(requestSize)
if b.brokerOutgoingByteRate != nil {
b.brokerOutgoingByteRate.Mark(requestSize)
}
+
b.requestSize.Update(requestSize)
if b.brokerRequestSize != nil {
b.brokerRequestSize.Update(requestSize)
}
+
+}
+
+func (b *Broker) registerMetrics() {
+ b.brokerIncomingByteRate = b.registerMeter("incoming-byte-rate")
+ b.brokerRequestRate = b.registerMeter("request-rate")
+ b.brokerRequestSize = b.registerHistogram("request-size")
+ b.brokerRequestLatency = b.registerHistogram("request-latency-in-ms")
+ b.brokerOutgoingByteRate = b.registerMeter("outgoing-byte-rate")
+ b.brokerResponseRate = b.registerMeter("response-rate")
+ b.brokerResponseSize = b.registerHistogram("response-size")
+}
+
+func (b *Broker) unregisterMetrics() {
+ for _, name := range b.registeredMetrics {
+ b.conf.MetricRegistry.Unregister(name)
+ }
+}
+
+func (b *Broker) registerMeter(name string) metrics.Meter {
+ nameForBroker := getMetricNameForBroker(name, b)
+ b.registeredMetrics = append(b.registeredMetrics, nameForBroker)
+ return metrics.GetOrRegisterMeter(nameForBroker, b.conf.MetricRegistry)
+}
+
+func (b *Broker) registerHistogram(name string) metrics.Histogram {
+ nameForBroker := getMetricNameForBroker(name, b)
+ b.registeredMetrics = append(b.registeredMetrics, nameForBroker)
+ return getOrRegisterHistogram(nameForBroker, b.conf.MetricRegistry)
}
diff --git a/vendor/github.com/Shopify/sarama/client.go b/vendor/github.com/Shopify/sarama/client.go
index 0016f8f..c4c54b2 100644
--- a/vendor/github.com/Shopify/sarama/client.go
+++ b/vendor/github.com/Shopify/sarama/client.go
@@ -46,6 +46,10 @@
// the partition leader.
InSyncReplicas(topic string, partitionID int32) ([]int32, error)
+ // OfflineReplicas returns the set of all offline replica IDs for the given
+ // partition. Offline replicas are replicas which are offline
+ OfflineReplicas(topic string, partitionID int32) ([]int32, error)
+
// RefreshMetadata takes a list of topics and queries the cluster to refresh the
// available metadata for those topics. If no topics are provided, it will refresh
// metadata for all topics.
@@ -288,7 +292,8 @@
partitions = client.cachedPartitions(topic, allPartitions)
}
- if partitions == nil {
+ // no partitions found after refresh metadata
+ if len(partitions) == 0 {
return nil, ErrUnknownTopicOrPartition
}
@@ -373,6 +378,31 @@
return dupInt32Slice(metadata.Isr), nil
}
+func (client *client) OfflineReplicas(topic string, partitionID int32) ([]int32, error) {
+ if client.Closed() {
+ return nil, ErrClosedClient
+ }
+
+ metadata := client.cachedMetadata(topic, partitionID)
+
+ if metadata == nil {
+ err := client.RefreshMetadata(topic)
+ if err != nil {
+ return nil, err
+ }
+ metadata = client.cachedMetadata(topic, partitionID)
+ }
+
+ if metadata == nil {
+ return nil, ErrUnknownTopicOrPartition
+ }
+
+ if metadata.Err == ErrReplicaNotAvailable {
+ return dupInt32Slice(metadata.OfflineReplicas), metadata.Err
+ }
+ return dupInt32Slice(metadata.OfflineReplicas), nil
+}
+
func (client *client) Leader(topic string, partitionID int32) (*Broker, error) {
if client.Closed() {
return nil, ErrClosedClient
@@ -405,7 +435,11 @@
}
}
- return client.tryRefreshMetadata(topics, client.conf.Metadata.Retry.Max)
+ deadline := time.Time{}
+ if client.conf.Metadata.Timeout > 0 {
+ deadline = time.Now().Add(client.conf.Metadata.Timeout)
+ }
+ return client.tryRefreshMetadata(topics, client.conf.Metadata.Retry.Max, deadline)
}
func (client *client) GetOffset(topic string, partitionID int32, time int64) (int64, error) {
@@ -707,32 +741,47 @@
return nil
}
-func (client *client) tryRefreshMetadata(topics []string, attemptsRemaining int) error {
+func (client *client) tryRefreshMetadata(topics []string, attemptsRemaining int, deadline time.Time) error {
+ pastDeadline := func(backoff time.Duration) bool {
+ if !deadline.IsZero() && time.Now().Add(backoff).After(deadline) {
+ // we are past the deadline
+ return true
+ }
+ return false
+ }
retry := func(err error) error {
if attemptsRemaining > 0 {
backoff := client.computeBackoff(attemptsRemaining)
+ if pastDeadline(backoff) {
+ Logger.Println("client/metadata skipping last retries as we would go past the metadata timeout")
+ return err
+ }
Logger.Printf("client/metadata retrying after %dms... (%d attempts remaining)\n", client.conf.Metadata.Retry.Backoff/time.Millisecond, attemptsRemaining)
if backoff > 0 {
time.Sleep(backoff)
}
- return client.tryRefreshMetadata(topics, attemptsRemaining-1)
+ return client.tryRefreshMetadata(topics, attemptsRemaining-1, deadline)
}
return err
}
- for broker := client.any(); broker != nil; broker = client.any() {
+ broker := client.any()
+ for ; broker != nil && !pastDeadline(0); broker = client.any() {
+ allowAutoTopicCreation := true
if len(topics) > 0 {
Logger.Printf("client/metadata fetching metadata for %v from broker %s\n", topics, broker.addr)
} else {
+ allowAutoTopicCreation = false
Logger.Printf("client/metadata fetching metadata for all topics from broker %s\n", broker.addr)
}
- req := &MetadataRequest{Topics: topics}
- if client.conf.Version.IsAtLeast(V0_10_0_0) {
+ req := &MetadataRequest{Topics: topics, AllowAutoTopicCreation: allowAutoTopicCreation}
+ if client.conf.Version.IsAtLeast(V1_0_0_0) {
+ req.Version = 5
+ } else if client.conf.Version.IsAtLeast(V0_10_0_0) {
req.Version = 1
}
response, err := broker.GetMetadata(req)
-
switch err.(type) {
case nil:
allKnownMetaData := len(topics) == 0
@@ -747,6 +796,18 @@
case PacketEncodingError:
// didn't even send, return the error
return err
+
+ case KError:
+ // if SASL auth error return as this _should_ be a non retryable err for all brokers
+ if err.(KError) == ErrSASLAuthenticationFailed {
+ Logger.Println("client/metadata failed SASL authentication")
+ return err
+ }
+ // else remove that broker and try again
+ Logger.Printf("client/metadata got error from broker %d while fetching metadata: %v\n", broker.ID(), err)
+ _ = broker.Close()
+ client.deregisterBroker(broker)
+
default:
// some other error, remove that broker and try again
Logger.Printf("client/metadata got error from broker %d while fetching metadata: %v\n", broker.ID(), err)
@@ -755,6 +816,11 @@
}
}
+ if broker != nil {
+ Logger.Println("client/metadata not fetching metadata from broker %s as we would go past the metadata timeout\n", broker.addr)
+ return retry(ErrOutOfBrokers)
+ }
+
Logger.Println("client/metadata no available broker to send metadata request to")
client.resurrectDeadBrokers()
return retry(ErrOutOfBrokers)
@@ -792,7 +858,7 @@
switch topic.Err {
case ErrNoError:
- break
+ // no-op
case ErrInvalidTopic, ErrTopicAuthorizationFailed: // don't retry, don't store partial results
err = topic.Err
continue
@@ -802,7 +868,6 @@
continue
case ErrLeaderNotAvailable: // retry, but store partial partition results
retry = true
- break
default: // don't retry, don't store partial results
Logger.Printf("Unexpected topic-level metadata error: %s", topic.Err)
err = topic.Err
@@ -847,9 +912,8 @@
maxRetries := client.conf.Metadata.Retry.Max
retries := maxRetries - attemptsRemaining
return client.conf.Metadata.Retry.BackoffFunc(retries, maxRetries)
- } else {
- return client.conf.Metadata.Retry.Backoff
}
+ return client.conf.Metadata.Retry.Backoff
}
func (client *client) getConsumerMetadata(consumerGroup string, attemptsRemaining int) (*FindCoordinatorResponse, error) {
@@ -911,3 +975,18 @@
client.resurrectDeadBrokers()
return retry(ErrOutOfBrokers)
}
+
+// nopCloserClient embeds an existing Client, but disables
+// the Close method (yet all other methods pass
+// through unchanged). This is for use in larger structs
+// where it is undesirable to close the client that was
+// passed in by the caller.
+type nopCloserClient struct {
+ Client
+}
+
+// Close intercepts and purposely does not call the underlying
+// client's Close() method.
+func (ncc *nopCloserClient) Close() error {
+ return nil
+}
diff --git a/vendor/github.com/Shopify/sarama/config.go b/vendor/github.com/Shopify/sarama/config.go
index 9495b7f..e2e6513 100644
--- a/vendor/github.com/Shopify/sarama/config.go
+++ b/vendor/github.com/Shopify/sarama/config.go
@@ -10,6 +10,7 @@
"time"
"github.com/rcrowley/go-metrics"
+ "golang.org/x/net/proxy"
)
const defaultClientID = "sarama"
@@ -57,18 +58,28 @@
// SASLMechanism is the name of the enabled SASL mechanism.
// Possible values: OAUTHBEARER, PLAIN (defaults to PLAIN).
Mechanism SASLMechanism
+ // Version is the SASL Protocol Version to use
+ // Kafka > 1.x should use V1, except on Azure EventHub which use V0
+ Version int16
// Whether or not to send the Kafka SASL handshake first if enabled
// (defaults to true). You should only set this to false if you're using
// a non-Kafka SASL proxy.
Handshake bool
- //username and password for SASL/PLAIN authentication
+ //username and password for SASL/PLAIN or SASL/SCRAM authentication
User string
Password string
+ // authz id used for SASL/SCRAM authentication
+ SCRAMAuthzID string
+ // SCRAMClientGeneratorFunc is a generator of a user provided implementation of a SCRAM
+ // client used to perform the SCRAM exchange with the server.
+ SCRAMClientGeneratorFunc func() SCRAMClient
// TokenProvider is a user-defined callback for generating
// access tokens for SASL/OAUTHBEARER auth. See the
// AccessTokenProvider interface docs for proper implementation
// guidelines.
TokenProvider AccessTokenProvider
+
+ GSSAPI GSSAPIConfig
}
// KeepAlive specifies the keep-alive period for an active network connection.
@@ -80,6 +91,14 @@
// network being dialed.
// If nil, a local address is automatically chosen.
LocalAddr net.Addr
+
+ Proxy struct {
+ // Whether or not to use proxy when connecting to the broker
+ // (defaults to false).
+ Enable bool
+ // The proxy dialer to use enabled (defaults to nil).
+ Dialer proxy.Dialer
+ }
}
// Metadata is the namespace for metadata management properties used by the
@@ -107,6 +126,13 @@
// and usually more convenient, but can take up a substantial amount of
// memory if you have many topics and partitions. Defaults to true.
Full bool
+
+ // How long to wait for a successful metadata response.
+ // Disabled by default which means a metadata request against an unreachable
+ // cluster (all brokers are unreachable or unresponsive) can take up to
+ // `Net.[Dial|Read]Timeout * BrokerCount * (Metadata.Retry.Max + 1) + Metadata.Retry.Backoff * Metadata.Retry.Max`
+ // to fail.
+ Timeout time.Duration
}
// Producer is the namespace for configuration related to producing messages,
@@ -333,6 +359,11 @@
Max int
}
}
+
+ // IsolationLevel support 2 mode:
+ // - use `ReadUncommitted` (default) to consume and return all messages in message channel
+ // - use `ReadCommitted` to hide messages that are part of an aborted transaction
+ IsolationLevel IsolationLevel
}
// A user-provided string sent with every request to the brokers for logging,
@@ -370,6 +401,7 @@
c.Net.ReadTimeout = 30 * time.Second
c.Net.WriteTimeout = 30 * time.Second
c.Net.SASL.Handshake = true
+ c.Net.SASL.Version = SASLHandshakeV0
c.Metadata.Retry.Max = 3
c.Metadata.Retry.Backoff = 250 * time.Millisecond
@@ -414,10 +446,10 @@
// ConfigurationError if the specified values don't make sense.
func (c *Config) Validate() error {
// some configuration values should be warned on but not fail completely, do those first
- if c.Net.TLS.Enable == false && c.Net.TLS.Config != nil {
+ if !c.Net.TLS.Enable && c.Net.TLS.Config != nil {
Logger.Println("Net.TLS is disabled but a non-nil configuration was provided.")
}
- if c.Net.SASL.Enable == false {
+ if !c.Net.SASL.Enable {
if c.Net.SASL.User != "" {
Logger.Println("Net.SASL is disabled but a non-empty username was provided.")
}
@@ -475,22 +507,62 @@
case c.Net.KeepAlive < 0:
return ConfigurationError("Net.KeepAlive must be >= 0")
case c.Net.SASL.Enable:
- // For backwards compatibility, empty mechanism value defaults to PLAIN
- isSASLPlain := len(c.Net.SASL.Mechanism) == 0 || c.Net.SASL.Mechanism == SASLTypePlaintext
- if isSASLPlain {
+ if c.Net.SASL.Mechanism == "" {
+ c.Net.SASL.Mechanism = SASLTypePlaintext
+ }
+
+ switch c.Net.SASL.Mechanism {
+ case SASLTypePlaintext:
if c.Net.SASL.User == "" {
return ConfigurationError("Net.SASL.User must not be empty when SASL is enabled")
}
if c.Net.SASL.Password == "" {
return ConfigurationError("Net.SASL.Password must not be empty when SASL is enabled")
}
- } else if c.Net.SASL.Mechanism == SASLTypeOAuth {
+ case SASLTypeOAuth:
if c.Net.SASL.TokenProvider == nil {
- return ConfigurationError("An AccessTokenProvider instance must be provided to Net.SASL.User.TokenProvider")
+ return ConfigurationError("An AccessTokenProvider instance must be provided to Net.SASL.TokenProvider")
}
- } else {
- msg := fmt.Sprintf("The SASL mechanism configuration is invalid. Possible values are `%s` and `%s`",
- SASLTypeOAuth, SASLTypePlaintext)
+ case SASLTypeSCRAMSHA256, SASLTypeSCRAMSHA512:
+ if c.Net.SASL.User == "" {
+ return ConfigurationError("Net.SASL.User must not be empty when SASL is enabled")
+ }
+ if c.Net.SASL.Password == "" {
+ return ConfigurationError("Net.SASL.Password must not be empty when SASL is enabled")
+ }
+ if c.Net.SASL.SCRAMClientGeneratorFunc == nil {
+ return ConfigurationError("A SCRAMClientGeneratorFunc function must be provided to Net.SASL.SCRAMClientGeneratorFunc")
+ }
+ case SASLTypeGSSAPI:
+ if c.Net.SASL.GSSAPI.ServiceName == "" {
+ return ConfigurationError("Net.SASL.GSSAPI.ServiceName must not be empty when GSS-API mechanism is used")
+ }
+
+ if c.Net.SASL.GSSAPI.AuthType == KRB5_USER_AUTH {
+ if c.Net.SASL.GSSAPI.Password == "" {
+ return ConfigurationError("Net.SASL.GSSAPI.Password must not be empty when GSS-API " +
+ "mechanism is used and Net.SASL.GSSAPI.AuthType = KRB5_USER_AUTH")
+ }
+ } else if c.Net.SASL.GSSAPI.AuthType == KRB5_KEYTAB_AUTH {
+ if c.Net.SASL.GSSAPI.KeyTabPath == "" {
+ return ConfigurationError("Net.SASL.GSSAPI.KeyTabPath must not be empty when GSS-API mechanism is used" +
+ " and Net.SASL.GSSAPI.AuthType = KRB5_KEYTAB_AUTH")
+ }
+ } else {
+ return ConfigurationError("Net.SASL.GSSAPI.AuthType is invalid. Possible values are KRB5_USER_AUTH and KRB5_KEYTAB_AUTH")
+ }
+ if c.Net.SASL.GSSAPI.KerberosConfigPath == "" {
+ return ConfigurationError("Net.SASL.GSSAPI.KerberosConfigPath must not be empty when GSS-API mechanism is used")
+ }
+ if c.Net.SASL.GSSAPI.Username == "" {
+ return ConfigurationError("Net.SASL.GSSAPI.Username must not be empty when GSS-API mechanism is used")
+ }
+ if c.Net.SASL.GSSAPI.Realm == "" {
+ return ConfigurationError("Net.SASL.GSSAPI.Realm must not be empty when GSS-API mechanism is used")
+ }
+ default:
+ msg := fmt.Sprintf("The SASL mechanism configuration is invalid. Possible values are `%s`, `%s`, `%s`, `%s` and `%s`",
+ SASLTypeOAuth, SASLTypePlaintext, SASLTypeSCRAMSHA256, SASLTypeSCRAMSHA512, SASLTypeGSSAPI)
return ConfigurationError(msg)
}
}
@@ -584,6 +656,13 @@
return ConfigurationError("Consumer.Offsets.Initial must be OffsetOldest or OffsetNewest")
case c.Consumer.Offsets.Retry.Max < 0:
return ConfigurationError("Consumer.Offsets.Retry.Max must be >= 0")
+ case c.Consumer.IsolationLevel != ReadUncommitted && c.Consumer.IsolationLevel != ReadCommitted:
+ return ConfigurationError("Consumer.IsolationLevel must be ReadUncommitted or ReadCommitted")
+ }
+
+ // validate IsolationLevel
+ if c.Consumer.IsolationLevel == ReadCommitted && !c.Version.IsAtLeast(V0_11_0_0) {
+ return ConfigurationError("ReadCommitted requires Version >= V0_11_0_0")
}
// validate the Consumer Group values
diff --git a/vendor/github.com/Shopify/sarama/config_resource_type.go b/vendor/github.com/Shopify/sarama/config_resource_type.go
index 848cc9c..5399d75 100644
--- a/vendor/github.com/Shopify/sarama/config_resource_type.go
+++ b/vendor/github.com/Shopify/sarama/config_resource_type.go
@@ -1,15 +1,22 @@
package sarama
+//ConfigResourceType is a type for config resource
type ConfigResourceType int8
// Taken from :
// https://cwiki.apache.org/confluence/display/KAFKA/KIP-133%3A+Describe+and+Alter+Configs+Admin+APIs#KIP-133:DescribeandAlterConfigsAdminAPIs-WireFormattypes
const (
- UnknownResource ConfigResourceType = 0
- AnyResource ConfigResourceType = 1
- TopicResource ConfigResourceType = 2
- GroupResource ConfigResourceType = 3
- ClusterResource ConfigResourceType = 4
- BrokerResource ConfigResourceType = 5
+ //UnknownResource constant type
+ UnknownResource ConfigResourceType = iota
+ //AnyResource constant type
+ AnyResource
+ //TopicResource constant type
+ TopicResource
+ //GroupResource constant type
+ GroupResource
+ //ClusterResource constant type
+ ClusterResource
+ //BrokerResource constant type
+ BrokerResource
)
diff --git a/vendor/github.com/Shopify/sarama/consumer.go b/vendor/github.com/Shopify/sarama/consumer.go
index ce72ff1..72c4d7c 100644
--- a/vendor/github.com/Shopify/sarama/consumer.go
+++ b/vendor/github.com/Shopify/sarama/consumer.go
@@ -3,20 +3,24 @@
import (
"errors"
"fmt"
+ "math"
"sync"
"sync/atomic"
"time"
+
+ "github.com/rcrowley/go-metrics"
)
// ConsumerMessage encapsulates a Kafka message returned by the consumer.
type ConsumerMessage struct {
- Key, Value []byte
- Topic string
- Partition int32
- Offset int64
+ Headers []*RecordHeader // only set if kafka is version 0.11+
Timestamp time.Time // only set if kafka is version 0.10+, inner message timestamp
BlockTimestamp time.Time // only set if kafka is version 0.10+, outer (compressed) block timestamp
- Headers []*RecordHeader // only set if kafka is version 0.11+
+
+ Key, Value []byte
+ Topic string
+ Partition int32
+ Offset int64
}
// ConsumerError is what is provided to the user when an error occurs.
@@ -43,13 +47,7 @@
// Consumer manages PartitionConsumers which process Kafka messages from brokers. You MUST call Close()
// on a consumer to avoid leaks, it will not be garbage-collected automatically when it passes out of
// scope.
-//
-// Sarama's Consumer type does not currently support automatic consumer-group rebalancing and offset tracking.
-// For Zookeeper-based tracking (Kafka 0.8.2 and earlier), the https://github.com/wvanbergen/kafka library
-// builds on Sarama to add this support. For Kafka-based tracking (Kafka 0.9 and later), the
-// https://github.com/bsm/sarama-cluster library builds on Sarama to add this support.
type Consumer interface {
-
// Topics returns the set of available topics as retrieved from the cluster
// metadata. This method is the same as Client.Topics(), and is provided for
// convenience.
@@ -75,13 +73,11 @@
}
type consumer struct {
- client Client
- conf *Config
- ownClient bool
-
- lock sync.Mutex
+ conf *Config
children map[string]map[int32]*partitionConsumer
brokerConsumers map[*Broker]*brokerConsumer
+ client Client
+ lock sync.Mutex
}
// NewConsumer creates a new consumer using the given broker addresses and configuration.
@@ -90,18 +86,19 @@
if err != nil {
return nil, err
}
-
- c, err := NewConsumerFromClient(client)
- if err != nil {
- return nil, err
- }
- c.(*consumer).ownClient = true
- return c, nil
+ return newConsumer(client)
}
// NewConsumerFromClient creates a new consumer using the given client. It is still
// necessary to call Close() on the underlying client when shutting down this consumer.
func NewConsumerFromClient(client Client) (Consumer, error) {
+ // For clients passed in by the client, ensure we don't
+ // call Close() on it.
+ cli := &nopCloserClient{client}
+ return newConsumer(cli)
+}
+
+func newConsumer(client Client) (Consumer, error) {
// Check that we are not dealing with a closed Client before processing any other arguments
if client.Closed() {
return nil, ErrClosedClient
@@ -118,10 +115,7 @@
}
func (c *consumer) Close() error {
- if c.ownClient {
- return c.client.Close()
- }
- return nil
+ return c.client.Close()
}
func (c *consumer) Topics() ([]string, error) {
@@ -261,12 +255,11 @@
// or a separate goroutine. Check out the Consumer examples to see implementations of these different approaches.
//
// To terminate such a for/range loop while the loop is executing, call AsyncClose. This will kick off the process of
-// consumer tear-down & return imediately. Continue to loop, servicing the Messages channel until the teardown process
+// consumer tear-down & return immediately. Continue to loop, servicing the Messages channel until the teardown process
// AsyncClose initiated closes it (thus terminating the for/range loop). If you've already ceased reading Messages, call
// Close; this will signal the PartitionConsumer's goroutines to begin shutting down (just like AsyncClose), but will
// also drain the Messages channel, harvest all errors & return them once cleanup has completed.
type PartitionConsumer interface {
-
// AsyncClose initiates a shutdown of the PartitionConsumer. This method will return immediately, after which you
// should continue to service the 'Messages' and 'Errors' channels until they are empty. It is required to call this
// function, or Close before a consumer object passes out of scope, as it will otherwise leak memory. You must call
@@ -298,24 +291,22 @@
type partitionConsumer struct {
highWaterMarkOffset int64 // must be at the top of the struct because https://golang.org/pkg/sync/atomic/#pkg-note-BUG
- consumer *consumer
- conf *Config
- topic string
- partition int32
+ consumer *consumer
+ conf *Config
broker *brokerConsumer
messages chan *ConsumerMessage
errors chan *ConsumerError
feeder chan *FetchResponse
trigger, dying chan none
- responseResult error
closeOnce sync.Once
-
- fetchSize int32
- offset int64
-
- retries int32
+ topic string
+ partition int32
+ responseResult error
+ fetchSize int32
+ offset int64
+ retries int32
}
var errTimedOut = errors.New("timed out feeding messages to the user") // not user-facing
@@ -338,9 +329,8 @@
if child.conf.Consumer.Retry.BackoffFunc != nil {
retries := atomic.AddInt32(&child.retries, 1)
return child.conf.Consumer.Retry.BackoffFunc(int(retries))
- } else {
- return child.conf.Consumer.Retry.Backoff
}
+ return child.conf.Consumer.Retry.Backoff
}
func (child *partitionConsumer) dispatcher() {
@@ -432,12 +422,6 @@
func (child *partitionConsumer) Close() error {
child.AsyncClose()
- go withRecover(func() {
- for range child.messages {
- // drain
- }
- })
-
var errors ConsumerErrors
for err := range child.errors {
errors = append(errors, err)
@@ -469,14 +453,22 @@
for i, msg := range msgs {
messageSelect:
select {
+ case <-child.dying:
+ child.broker.acks.Done()
+ continue feederLoop
case child.messages <- msg:
firstAttempt = true
case <-expiryTicker.C:
if !firstAttempt {
child.responseResult = errTimedOut
child.broker.acks.Done()
+ remainingLoop:
for _, msg = range msgs[i:] {
- child.messages <- msg
+ select {
+ case child.messages <- msg:
+ case <-child.dying:
+ break remainingLoop
+ }
}
child.broker.input <- child
continue feederLoop
@@ -532,7 +524,8 @@
}
func (child *partitionConsumer) parseRecords(batch *RecordBatch) ([]*ConsumerMessage, error) {
- var messages []*ConsumerMessage
+ messages := make([]*ConsumerMessage, 0, len(batch.Records))
+
for _, rec := range batch.Records {
offset := batch.FirstOffset + rec.OffsetDelta
if offset < child.offset {
@@ -560,6 +553,23 @@
}
func (child *partitionConsumer) parseResponse(response *FetchResponse) ([]*ConsumerMessage, error) {
+ var (
+ metricRegistry = child.conf.MetricRegistry
+ consumerBatchSizeMetric metrics.Histogram
+ )
+
+ if metricRegistry != nil {
+ consumerBatchSizeMetric = getOrRegisterHistogram("consumer-batch-size", metricRegistry)
+ }
+
+ // If request was throttled and empty we log and return without error
+ if response.ThrottleTime != time.Duration(0) && len(response.Blocks) == 0 {
+ Logger.Printf(
+ "consumer/broker/%d FetchResponse throttled %v\n",
+ child.broker.broker.ID(), response.ThrottleTime)
+ return nil, nil
+ }
+
block := response.GetBlock(child.topic, child.partition)
if block == nil {
return nil, ErrIncompleteResponse
@@ -573,6 +583,9 @@
if err != nil {
return nil, err
}
+
+ consumerBatchSizeMetric.Update(int64(nRecs))
+
if nRecs == 0 {
partialTrailingMessage, err := block.isPartial()
if err != nil {
@@ -587,6 +600,10 @@
child.offset++ // skip this one so we can keep processing future messages
} else {
child.fetchSize *= 2
+ // check int32 overflow
+ if child.fetchSize < 0 {
+ child.fetchSize = math.MaxInt32
+ }
if child.conf.Consumer.Fetch.Max > 0 && child.fetchSize > child.conf.Consumer.Fetch.Max {
child.fetchSize = child.conf.Consumer.Fetch.Max
}
@@ -600,6 +617,12 @@
child.fetchSize = child.conf.Consumer.Fetch.Default
atomic.StoreInt64(&child.highWaterMarkOffset, block.HighWaterMarkOffset)
+ // abortedProducerIDs contains producerID which message should be ignored as uncommitted
+ // - producerID are added when the partitionConsumer iterate over the offset at which an aborted transaction begins (abortedTransaction.FirstOffset)
+ // - producerID are removed when partitionConsumer iterate over an aborted controlRecord, meaning the aborted transaction for this producer is over
+ abortedProducerIDs := make(map[int64]struct{}, len(block.AbortedTransactions))
+ abortedTransactions := block.getAbortedTransactions()
+
messages := []*ConsumerMessage{}
for _, records := range block.RecordsSet {
switch records.recordsType {
@@ -611,13 +634,55 @@
messages = append(messages, messageSetMessages...)
case defaultRecords:
+ // Consume remaining abortedTransaction up to last offset of current batch
+ for _, txn := range abortedTransactions {
+ if txn.FirstOffset > records.RecordBatch.LastOffset() {
+ break
+ }
+ abortedProducerIDs[txn.ProducerID] = struct{}{}
+ // Pop abortedTransactions so that we never add it again
+ abortedTransactions = abortedTransactions[1:]
+ }
+
recordBatchMessages, err := child.parseRecords(records.RecordBatch)
if err != nil {
return nil, err
}
- if control, err := records.isControl(); err != nil || control {
+
+ // Parse and commit offset but do not expose messages that are:
+ // - control records
+ // - part of an aborted transaction when set to `ReadCommitted`
+
+ // control record
+ isControl, err := records.isControl()
+ if err != nil {
+ // I don't know why there is this continue in case of error to begin with
+ // Safe bet is to ignore control messages if ReadUncommitted
+ // and block on them in case of error and ReadCommitted
+ if child.conf.Consumer.IsolationLevel == ReadCommitted {
+ return nil, err
+ }
continue
}
+ if isControl {
+ controlRecord, err := records.getControlRecord()
+ if err != nil {
+ return nil, err
+ }
+
+ if controlRecord.Type == ControlRecordAbort {
+ delete(abortedProducerIDs, records.RecordBatch.ProducerID)
+ }
+ continue
+ }
+
+ // filter aborted transactions
+ if child.conf.Consumer.IsolationLevel == ReadCommitted {
+ _, isAborted := abortedProducerIDs[records.RecordBatch.ProducerID]
+ if records.RecordBatch.IsTransactional && isAborted {
+ continue
+ }
+ }
messages = append(messages, recordBatchMessages...)
default:
@@ -628,15 +693,13 @@
return messages, nil
}
-// brokerConsumer
-
type brokerConsumer struct {
consumer *consumer
broker *Broker
input chan *partitionConsumer
newSubscriptions chan []*partitionConsumer
- wait chan none
subscriptions map[*partitionConsumer]none
+ wait chan none
acks sync.WaitGroup
refs int
}
@@ -658,14 +721,14 @@
return bc
}
+// The subscriptionManager constantly accepts new subscriptions on `input` (even when the main subscriptionConsumer
+// goroutine is in the middle of a network request) and batches it up. The main worker goroutine picks
+// up a batch of new subscriptions between every network request by reading from `newSubscriptions`, so we give
+// it nil if no new subscriptions are available. We also write to `wait` only when new subscriptions is available,
+// so the main goroutine can block waiting for work if it has none.
func (bc *brokerConsumer) subscriptionManager() {
var buffer []*partitionConsumer
- // The subscriptionManager constantly accepts new subscriptions on `input` (even when the main subscriptionConsumer
- // goroutine is in the middle of a network request) and batches it up. The main worker goroutine picks
- // up a batch of new subscriptions between every network request by reading from `newSubscriptions`, so we give
- // it nil if no new subscriptions are available. We also write to `wait` only when new subscriptions is available,
- // so the main goroutine can block waiting for work if it has none.
for {
if len(buffer) > 0 {
select {
@@ -698,10 +761,10 @@
close(bc.newSubscriptions)
}
+//subscriptionConsumer ensures we will get nil right away if no new subscriptions is available
func (bc *brokerConsumer) subscriptionConsumer() {
<-bc.wait // wait for our first piece of work
- // the subscriptionConsumer ensures we will get nil right away if no new subscriptions is available
for newSubscriptions := range bc.newSubscriptions {
bc.updateSubscriptions(newSubscriptions)
@@ -742,20 +805,20 @@
close(child.trigger)
delete(bc.subscriptions, child)
default:
- break
+ // no-op
}
}
}
+//handleResponses handles the response codes left for us by our subscriptions, and abandons ones that have been closed
func (bc *brokerConsumer) handleResponses() {
- // handles the response codes left for us by our subscriptions, and abandons ones that have been closed
for child := range bc.subscriptions {
result := child.responseResult
child.responseResult = nil
switch result {
case nil:
- break
+ // no-op
case errTimedOut:
Logger.Printf("consumer/broker/%d abandoned subscription to %s/%d because consuming was taking too long\n",
bc.broker.ID(), child.topic, child.partition)
@@ -822,7 +885,7 @@
}
if bc.consumer.conf.Version.IsAtLeast(V0_11_0_0) {
request.Version = 4
- request.Isolation = ReadUncommitted // We don't support yet transactions.
+ request.Isolation = bc.consumer.conf.Consumer.IsolationLevel
}
for child := range bc.subscriptions {
diff --git a/vendor/github.com/Shopify/sarama/consumer_group.go b/vendor/github.com/Shopify/sarama/consumer_group.go
index 8c8babc..8de9513 100644
--- a/vendor/github.com/Shopify/sarama/consumer_group.go
+++ b/vendor/github.com/Shopify/sarama/consumer_group.go
@@ -52,8 +52,7 @@
}
type consumerGroup struct {
- client Client
- ownClient bool
+ client Client
config *Config
consumer Consumer
@@ -73,20 +72,24 @@
return nil, err
}
- c, err := NewConsumerGroupFromClient(groupID, client)
+ c, err := newConsumerGroup(groupID, client)
if err != nil {
_ = client.Close()
- return nil, err
}
-
- c.(*consumerGroup).ownClient = true
- return c, nil
+ return c, err
}
// NewConsumerGroupFromClient creates a new consumer group using the given client. It is still
// necessary to call Close() on the underlying client when shutting down this consumer.
// PLEASE NOTE: consumer groups can only re-use but not share clients.
func NewConsumerGroupFromClient(groupID string, client Client) (ConsumerGroup, error) {
+ // For clients passed in by the client, ensure we don't
+ // call Close() on it.
+ cli := &nopCloserClient{client}
+ return newConsumerGroup(groupID, cli)
+}
+
+func newConsumerGroup(groupID string, client Client) (ConsumerGroup, error) {
config := client.Config()
if !config.Version.IsAtLeast(V0_10_2_0) {
return nil, ConfigurationError("consumer groups require Version to be >= V0_10_2_0")
@@ -131,10 +134,8 @@
err = e
}
- if c.ownClient {
- if e := c.client.Close(); e != nil {
- err = e
- }
+ if e := c.client.Close(); e != nil {
+ err = e
}
})
return
@@ -162,14 +163,8 @@
return err
}
- // Get coordinator
- coordinator, err := c.client.Coordinator(c.groupID)
- if err != nil {
- return err
- }
-
// Init session
- sess, err := c.newSession(ctx, coordinator, topics, handler, c.config.Consumer.Group.Rebalance.Retry.Max)
+ sess, err := c.newSession(ctx, topics, handler, c.config.Consumer.Group.Rebalance.Retry.Max)
if err == ErrClosedClient {
return ErrClosedConsumerGroup
} else if err != nil {
@@ -183,7 +178,33 @@
return sess.release(true)
}
-func (c *consumerGroup) newSession(ctx context.Context, coordinator *Broker, topics []string, handler ConsumerGroupHandler, retries int) (*consumerGroupSession, error) {
+func (c *consumerGroup) retryNewSession(ctx context.Context, topics []string, handler ConsumerGroupHandler, retries int, refreshCoordinator bool) (*consumerGroupSession, error) {
+ select {
+ case <-c.closed:
+ return nil, ErrClosedConsumerGroup
+ case <-time.After(c.config.Consumer.Group.Rebalance.Retry.Backoff):
+ }
+
+ if refreshCoordinator {
+ err := c.client.RefreshCoordinator(c.groupID)
+ if err != nil {
+ return c.retryNewSession(ctx, topics, handler, retries, true)
+ }
+ }
+
+ return c.newSession(ctx, topics, handler, retries-1)
+}
+
+func (c *consumerGroup) newSession(ctx context.Context, topics []string, handler ConsumerGroupHandler, retries int) (*consumerGroupSession, error) {
+ coordinator, err := c.client.Coordinator(c.groupID)
+ if err != nil {
+ if retries <= 0 {
+ return nil, err
+ }
+
+ return c.retryNewSession(ctx, topics, handler, retries, true)
+ }
+
// Join consumer group
join, err := c.joinGroupRequest(coordinator, topics)
if err != nil {
@@ -195,19 +216,19 @@
c.memberID = join.MemberId
case ErrUnknownMemberId, ErrIllegalGeneration: // reset member ID and retry immediately
c.memberID = ""
- return c.newSession(ctx, coordinator, topics, handler, retries)
+ return c.newSession(ctx, topics, handler, retries)
+ case ErrNotCoordinatorForConsumer: // retry after backoff with coordinator refresh
+ if retries <= 0 {
+ return nil, join.Err
+ }
+
+ return c.retryNewSession(ctx, topics, handler, retries, true)
case ErrRebalanceInProgress: // retry after backoff
if retries <= 0 {
return nil, join.Err
}
- select {
- case <-c.closed:
- return nil, ErrClosedConsumerGroup
- case <-time.After(c.config.Consumer.Group.Rebalance.Retry.Backoff):
- }
-
- return c.newSession(ctx, coordinator, topics, handler, retries-1)
+ return c.retryNewSession(ctx, topics, handler, retries, false)
default:
return nil, join.Err
}
@@ -236,19 +257,19 @@
case ErrNoError:
case ErrUnknownMemberId, ErrIllegalGeneration: // reset member ID and retry immediately
c.memberID = ""
- return c.newSession(ctx, coordinator, topics, handler, retries)
+ return c.newSession(ctx, topics, handler, retries)
+ case ErrNotCoordinatorForConsumer: // retry after backoff with coordinator refresh
+ if retries <= 0 {
+ return nil, sync.Err
+ }
+
+ return c.retryNewSession(ctx, topics, handler, retries, true)
case ErrRebalanceInProgress: // retry after backoff
if retries <= 0 {
return nil, sync.Err
}
- select {
- case <-c.closed:
- return nil, ErrClosedConsumerGroup
- case <-time.After(c.config.Consumer.Group.Rebalance.Retry.Backoff):
- }
-
- return c.newSession(ctx, coordinator, topics, handler, retries-1)
+ return c.retryNewSession(ctx, topics, handler, retries, false)
default:
return nil, sync.Err
}
@@ -613,7 +634,7 @@
s.releaseOnce.Do(func() {
if withCleanup {
if e := s.handler.Cleanup(s); e != nil {
- s.parent.handleError(err, "", -1)
+ s.parent.handleError(e, "", -1)
err = e
}
}
diff --git a/vendor/github.com/Shopify/sarama/control_record.go b/vendor/github.com/Shopify/sarama/control_record.go
new file mode 100644
index 0000000..9b75ab5
--- /dev/null
+++ b/vendor/github.com/Shopify/sarama/control_record.go
@@ -0,0 +1,72 @@
+package sarama
+
+//ControlRecordType ...
+type ControlRecordType int
+
+const (
+ //ControlRecordAbort is a control record for abort
+ ControlRecordAbort ControlRecordType = iota
+ //ControlRecordCommit is a control record for commit
+ ControlRecordCommit
+ //ControlRecordUnknown is a control record of unknown type
+ ControlRecordUnknown
+)
+
+// Control records are returned as a record by fetchRequest
+// However unlike "normal" records, they mean nothing application wise.
+// They only serve internal logic for supporting transactions.
+type ControlRecord struct {
+ Version int16
+ CoordinatorEpoch int32
+ Type ControlRecordType
+}
+
+func (cr *ControlRecord) decode(key, value packetDecoder) error {
+ var err error
+ cr.Version, err = value.getInt16()
+ if err != nil {
+ return err
+ }
+
+ cr.CoordinatorEpoch, err = value.getInt32()
+ if err != nil {
+ return err
+ }
+
+ // There a version for the value part AND the key part. And I have no idea if they are supposed to match or not
+ // Either way, all these version can only be 0 for now
+ cr.Version, err = key.getInt16()
+ if err != nil {
+ return err
+ }
+
+ recordType, err := key.getInt16()
+ if err != nil {
+ return err
+ }
+
+ switch recordType {
+ case 0:
+ cr.Type = ControlRecordAbort
+ case 1:
+ cr.Type = ControlRecordCommit
+ default:
+ // from JAVA implementation:
+ // UNKNOWN is used to indicate a control type which the client is not aware of and should be ignored
+ cr.Type = ControlRecordUnknown
+ }
+ return nil
+}
+
+func (cr *ControlRecord) encode(key, value packetEncoder) {
+ value.putInt16(cr.Version)
+ value.putInt32(cr.CoordinatorEpoch)
+ key.putInt16(cr.Version)
+
+ switch cr.Type {
+ case ControlRecordAbort:
+ key.putInt16(0)
+ case ControlRecordCommit:
+ key.putInt16(1)
+ }
+}
diff --git a/vendor/github.com/Shopify/sarama/crc32_field.go b/vendor/github.com/Shopify/sarama/crc32_field.go
index 1f14443..38189a3 100644
--- a/vendor/github.com/Shopify/sarama/crc32_field.go
+++ b/vendor/github.com/Shopify/sarama/crc32_field.go
@@ -4,6 +4,7 @@
"encoding/binary"
"fmt"
"hash/crc32"
+ "sync"
)
type crcPolynomial int8
@@ -13,6 +14,22 @@
crcCastagnoli
)
+var crc32FieldPool = sync.Pool{}
+
+func acquireCrc32Field(polynomial crcPolynomial) *crc32Field {
+ val := crc32FieldPool.Get()
+ if val != nil {
+ c := val.(*crc32Field)
+ c.polynomial = polynomial
+ return c
+ }
+ return newCRC32Field(polynomial)
+}
+
+func releaseCrc32Field(c *crc32Field) {
+ crc32FieldPool.Put(c)
+}
+
var castagnoliTable = crc32.MakeTable(crc32.Castagnoli)
// crc32Field implements the pushEncoder and pushDecoder interfaces for calculating CRC32s.
diff --git a/vendor/github.com/Shopify/sarama/create_partitions_response.go b/vendor/github.com/Shopify/sarama/create_partitions_response.go
index abd621c..bb18204 100644
--- a/vendor/github.com/Shopify/sarama/create_partitions_response.go
+++ b/vendor/github.com/Shopify/sarama/create_partitions_response.go
@@ -1,6 +1,9 @@
package sarama
-import "time"
+import (
+ "fmt"
+ "time"
+)
type CreatePartitionsResponse struct {
ThrottleTime time.Duration
@@ -69,6 +72,14 @@
ErrMsg *string
}
+func (t *TopicPartitionError) Error() string {
+ text := t.Err.Error()
+ if t.ErrMsg != nil {
+ text = fmt.Sprintf("%s - %s", text, *t.ErrMsg)
+ }
+ return text
+}
+
func (t *TopicPartitionError) encode(pe packetEncoder) error {
pe.putInt16(int16(t.Err))
diff --git a/vendor/github.com/Shopify/sarama/create_topics_response.go b/vendor/github.com/Shopify/sarama/create_topics_response.go
index 66207e0..a493e02 100644
--- a/vendor/github.com/Shopify/sarama/create_topics_response.go
+++ b/vendor/github.com/Shopify/sarama/create_topics_response.go
@@ -1,6 +1,9 @@
package sarama
-import "time"
+import (
+ "fmt"
+ "time"
+)
type CreateTopicsResponse struct {
Version int16
@@ -83,6 +86,14 @@
ErrMsg *string
}
+func (t *TopicError) Error() string {
+ text := t.Err.Error()
+ if t.ErrMsg != nil {
+ text = fmt.Sprintf("%s - %s", text, *t.ErrMsg)
+ }
+ return text
+}
+
func (t *TopicError) encode(pe packetEncoder, version int16) error {
pe.putInt16(int16(t.Err))
diff --git a/vendor/github.com/Shopify/sarama/describe_configs_response.go b/vendor/github.com/Shopify/sarama/describe_configs_response.go
index 63fb6ea..5737232 100644
--- a/vendor/github.com/Shopify/sarama/describe_configs_response.go
+++ b/vendor/github.com/Shopify/sarama/describe_configs_response.go
@@ -26,12 +26,12 @@
}
const (
- SourceUnknown ConfigSource = 0
- SourceTopic ConfigSource = 1
- SourceDynamicBroker ConfigSource = 2
- SourceDynamicDefaultBroker ConfigSource = 3
- SourceStaticBroker ConfigSource = 4
- SourceDefault ConfigSource = 5
+ SourceUnknown ConfigSource = iota
+ SourceTopic
+ SourceDynamicBroker
+ SourceDynamicDefaultBroker
+ SourceStaticBroker
+ SourceDefault
)
type DescribeConfigsResponse struct {
diff --git a/vendor/github.com/Shopify/sarama/dev.yml b/vendor/github.com/Shopify/sarama/dev.yml
index 97eed3a..3f4d569 100644
--- a/vendor/github.com/Shopify/sarama/dev.yml
+++ b/vendor/github.com/Shopify/sarama/dev.yml
@@ -2,7 +2,7 @@
up:
- go:
- version: '1.11'
+ version: '1.12'
commands:
test:
diff --git a/vendor/github.com/Shopify/sarama/errors.go b/vendor/github.com/Shopify/sarama/errors.go
index 87a4c61..c6a8be7 100644
--- a/vendor/github.com/Shopify/sarama/errors.go
+++ b/vendor/github.com/Shopify/sarama/errors.go
@@ -81,6 +81,28 @@
// See https://cwiki.apache.org/confluence/display/KAFKA/A+Guide+To+The+Kafka+Protocol#AGuideToTheKafkaProtocol-ErrorCodes
type KError int16
+// MultiError is used to contain multi error.
+type MultiError struct {
+ Errors *[]error
+}
+
+func (mErr MultiError) Error() string {
+ var errString = ""
+ for _, err := range *mErr.Errors {
+ errString += err.Error() + ","
+ }
+ return errString
+}
+
+// ErrDeleteRecords is the type of error returned when fail to delete the required records
+type ErrDeleteRecords struct {
+ MultiError
+}
+
+func (err ErrDeleteRecords) Error() string {
+ return "kafka server: failed to delete records " + err.MultiError.Error()
+}
+
// Numeric error codes returned by the Kafka server.
const (
ErrNoError KError = 0
@@ -161,6 +183,11 @@
ErrFencedLeaderEpoch KError = 74
ErrUnknownLeaderEpoch KError = 75
ErrUnsupportedCompressionType KError = 76
+ ErrStaleBrokerEpoch KError = 77
+ ErrOffsetNotAvailable KError = 78
+ ErrMemberIdRequired KError = 79
+ ErrPreferredLeaderNotAvailable KError = 80
+ ErrGroupMaxSizeReached KError = 81
)
func (err KError) Error() string {
@@ -323,6 +350,16 @@
return "kafka server: The leader epoch in the request is newer than the epoch on the broker."
case ErrUnsupportedCompressionType:
return "kafka server: The requesting client does not support the compression type of given partition."
+ case ErrStaleBrokerEpoch:
+ return "kafka server: Broker epoch has changed"
+ case ErrOffsetNotAvailable:
+ return "kafka server: The leader high watermark has not caught up from a recent leader election so the offsets cannot be guaranteed to be monotonically increasing"
+ case ErrMemberIdRequired:
+ return "kafka server: The group member needs to have a valid member id before actually entering a consumer group"
+ case ErrPreferredLeaderNotAvailable:
+ return "kafka server: The preferred leader was not available"
+ case ErrGroupMaxSizeReached:
+ return "kafka server: Consumer group The consumer group has reached its max size. already has the configured maximum number of members."
}
return fmt.Sprintf("Unknown error, how did this happen? Error code = %d", err)
diff --git a/vendor/github.com/Shopify/sarama/fetch_request.go b/vendor/github.com/Shopify/sarama/fetch_request.go
index 462ab8a..4db9ddd 100644
--- a/vendor/github.com/Shopify/sarama/fetch_request.go
+++ b/vendor/github.com/Shopify/sarama/fetch_request.go
@@ -36,8 +36,8 @@
type IsolationLevel int8
const (
- ReadUncommitted IsolationLevel = 0
- ReadCommitted IsolationLevel = 1
+ ReadUncommitted IsolationLevel = iota
+ ReadCommitted
)
func (r *FetchRequest) encode(pe packetEncoder) (err error) {
diff --git a/vendor/github.com/Shopify/sarama/fetch_response.go b/vendor/github.com/Shopify/sarama/fetch_response.go
index 9df99c1..3afc187 100644
--- a/vendor/github.com/Shopify/sarama/fetch_response.go
+++ b/vendor/github.com/Shopify/sarama/fetch_response.go
@@ -1,6 +1,7 @@
package sarama
import (
+ "sort"
"time"
)
@@ -185,6 +186,17 @@
return pe.pop()
}
+func (b *FetchResponseBlock) getAbortedTransactions() []*AbortedTransaction {
+ // I can't find any doc that guarantee the field `fetchResponse.AbortedTransactions` is ordered
+ // plus Java implementation use a PriorityQueue based on `FirstOffset`. I guess we have to order it ourself
+ at := b.AbortedTransactions
+ sort.Slice(
+ at,
+ func(i, j int) bool { return at[i].FirstOffset < at[j].FirstOffset },
+ )
+ return at
+}
+
type FetchResponse struct {
Blocks map[string]map[int32]*FetchResponseBlock
ThrottleTime time.Duration
@@ -385,6 +397,65 @@
batch.addRecord(rec)
}
+// AddRecordBatchWithTimestamp is similar to AddRecordWithTimestamp
+// But instead of appending 1 record to a batch, it append a new batch containing 1 record to the fetchResponse
+// Since transaction are handled on batch level (the whole batch is either committed or aborted), use this to test transactions
+func (r *FetchResponse) AddRecordBatchWithTimestamp(topic string, partition int32, key, value Encoder, offset int64, producerID int64, isTransactional bool, timestamp time.Time) {
+ frb := r.getOrCreateBlock(topic, partition)
+ kb, vb := encodeKV(key, value)
+
+ records := newDefaultRecords(&RecordBatch{Version: 2, LogAppendTime: r.LogAppendTime, FirstTimestamp: timestamp, MaxTimestamp: r.Timestamp})
+ batch := &RecordBatch{
+ Version: 2,
+ LogAppendTime: r.LogAppendTime,
+ FirstTimestamp: timestamp,
+ MaxTimestamp: r.Timestamp,
+ FirstOffset: offset,
+ LastOffsetDelta: 0,
+ ProducerID: producerID,
+ IsTransactional: isTransactional,
+ }
+ rec := &Record{Key: kb, Value: vb, OffsetDelta: 0, TimestampDelta: timestamp.Sub(batch.FirstTimestamp)}
+ batch.addRecord(rec)
+ records.RecordBatch = batch
+
+ frb.RecordsSet = append(frb.RecordsSet, &records)
+}
+
+func (r *FetchResponse) AddControlRecordWithTimestamp(topic string, partition int32, offset int64, producerID int64, recordType ControlRecordType, timestamp time.Time) {
+ frb := r.getOrCreateBlock(topic, partition)
+
+ // batch
+ batch := &RecordBatch{
+ Version: 2,
+ LogAppendTime: r.LogAppendTime,
+ FirstTimestamp: timestamp,
+ MaxTimestamp: r.Timestamp,
+ FirstOffset: offset,
+ LastOffsetDelta: 0,
+ ProducerID: producerID,
+ IsTransactional: true,
+ Control: true,
+ }
+
+ // records
+ records := newDefaultRecords(nil)
+ records.RecordBatch = batch
+
+ // record
+ crAbort := ControlRecord{
+ Version: 0,
+ Type: recordType,
+ }
+ crKey := &realEncoder{raw: make([]byte, 4)}
+ crValue := &realEncoder{raw: make([]byte, 6)}
+ crAbort.encode(crKey, crValue)
+ rec := &Record{Key: ByteEncoder(crKey.raw), Value: ByteEncoder(crValue.raw), OffsetDelta: 0, TimestampDelta: timestamp.Sub(batch.FirstTimestamp)}
+ batch.addRecord(rec)
+
+ frb.RecordsSet = append(frb.RecordsSet, &records)
+}
+
func (r *FetchResponse) AddMessage(topic string, partition int32, key, value Encoder, offset int64) {
r.AddMessageWithTimestamp(topic, partition, key, value, offset, time.Time{}, 0)
}
@@ -393,6 +464,15 @@
r.AddRecordWithTimestamp(topic, partition, key, value, offset, time.Time{})
}
+func (r *FetchResponse) AddRecordBatch(topic string, partition int32, key, value Encoder, offset int64, producerID int64, isTransactional bool) {
+ r.AddRecordBatchWithTimestamp(topic, partition, key, value, offset, producerID, isTransactional, time.Time{})
+}
+
+func (r *FetchResponse) AddControlRecord(topic string, partition int32, offset int64, producerID int64, recordType ControlRecordType) {
+ // define controlRecord key and value
+ r.AddControlRecordWithTimestamp(topic, partition, offset, producerID, recordType, time.Time{})
+}
+
func (r *FetchResponse) SetLastOffsetDelta(topic string, partition int32, offset int32) {
frb := r.getOrCreateBlock(topic, partition)
if len(frb.RecordsSet) == 0 {
diff --git a/vendor/github.com/Shopify/sarama/find_coordinator_request.go b/vendor/github.com/Shopify/sarama/find_coordinator_request.go
index 0ab5cb5..ff2ad20 100644
--- a/vendor/github.com/Shopify/sarama/find_coordinator_request.go
+++ b/vendor/github.com/Shopify/sarama/find_coordinator_request.go
@@ -3,8 +3,8 @@
type CoordinatorType int8
const (
- CoordinatorGroup CoordinatorType = 0
- CoordinatorTransaction CoordinatorType = 1
+ CoordinatorGroup CoordinatorType = iota
+ CoordinatorTransaction
)
type FindCoordinatorRequest struct {
diff --git a/vendor/github.com/Shopify/sarama/go.mod b/vendor/github.com/Shopify/sarama/go.mod
index 3715129..8c45155 100644
--- a/vendor/github.com/Shopify/sarama/go.mod
+++ b/vendor/github.com/Shopify/sarama/go.mod
@@ -1,13 +1,24 @@
module github.com/Shopify/sarama
require (
- github.com/DataDog/zstd v1.3.5
+ github.com/DataDog/zstd v1.3.6-0.20190409195224-796139022798
github.com/Shopify/toxiproxy v2.1.4+incompatible
github.com/davecgh/go-spew v1.1.1
github.com/eapache/go-resiliency v1.1.0
github.com/eapache/go-xerial-snappy v0.0.0-20180814174437-776d5712da21
github.com/eapache/queue v1.1.0
- github.com/golang/snappy v0.0.0-20180518054509-2e65f85255db // indirect
- github.com/pierrec/lz4 v2.0.5+incompatible
+ github.com/golang/snappy v0.0.1 // indirect
+ github.com/hashicorp/go-uuid v1.0.1 // indirect
+ github.com/jcmturner/gofork v0.0.0-20190328161633-dc7c13fece03
+ github.com/pierrec/lz4 v0.0.0-20190327172049-315a67e90e41
github.com/rcrowley/go-metrics v0.0.0-20181016184325-3113b8401b8a
+ github.com/stretchr/testify v1.3.0
+ github.com/xdg/scram v0.0.0-20180814205039-7eeb5667e42c
+ github.com/xdg/stringprep v1.0.0 // indirect
+ golang.org/x/crypto v0.0.0-20190404164418-38d8ce5564a5 // indirect
+ golang.org/x/net v0.0.0-20190404232315-eb5bcb51f2a3
+ gopkg.in/jcmturner/aescts.v1 v1.0.1 // indirect
+ gopkg.in/jcmturner/dnsutils.v1 v1.0.1 // indirect
+ gopkg.in/jcmturner/gokrb5.v7 v7.2.3
+ gopkg.in/jcmturner/rpc.v1 v1.1.0 // indirect
)
diff --git a/vendor/github.com/Shopify/sarama/go.sum b/vendor/github.com/Shopify/sarama/go.sum
index 58e2e91..4dbc6d2 100644
--- a/vendor/github.com/Shopify/sarama/go.sum
+++ b/vendor/github.com/Shopify/sarama/go.sum
@@ -1,7 +1,8 @@
-github.com/DataDog/zstd v1.3.5 h1:DtpNbljikUepEPD16hD4LvIcmhnhdLTiW/5pHgbmp14=
-github.com/DataDog/zstd v1.3.5/go.mod h1:1jcaCB/ufaK+sKp1NBhlGmpz41jOoPQ35bpF36t7BBo=
+github.com/DataDog/zstd v1.3.6-0.20190409195224-796139022798 h1:2T/jmrHeTezcCM58lvEQXs0UpQJCo5SoGAcg+mbSTIg=
+github.com/DataDog/zstd v1.3.6-0.20190409195224-796139022798/go.mod h1:1jcaCB/ufaK+sKp1NBhlGmpz41jOoPQ35bpF36t7BBo=
github.com/Shopify/toxiproxy v2.1.4+incompatible h1:TKdv8HiTLgE5wdJuEML90aBgNWsokNbMijUGhmcoBJc=
github.com/Shopify/toxiproxy v2.1.4+incompatible/go.mod h1:OXgGpZ6Cli1/URJOF1DMxUHB2q5Ap20/P/eIdh4G0pI=
+github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
github.com/eapache/go-resiliency v1.1.0 h1:1NtRmCAqadE2FN4ZcN6g90TP3uk8cg9rn9eNK2197aU=
@@ -10,9 +11,41 @@
github.com/eapache/go-xerial-snappy v0.0.0-20180814174437-776d5712da21/go.mod h1:+020luEh2TKB4/GOp8oxxtq0Daoen/Cii55CzbTV6DU=
github.com/eapache/queue v1.1.0 h1:YOEu7KNc61ntiQlcEeUIoDTJ2o8mQznoNvUhiigpIqc=
github.com/eapache/queue v1.1.0/go.mod h1:6eCeP0CKFpHLu8blIFXhExK/dRa7WDZfr6jVFPTqq+I=
-github.com/golang/snappy v0.0.0-20180518054509-2e65f85255db h1:woRePGFeVFfLKN/pOkfl+p/TAqKOfFu+7KPlMVpok/w=
-github.com/golang/snappy v0.0.0-20180518054509-2e65f85255db/go.mod h1:/XxbfmMg8lxefKM7IXC3fBNl/7bRcc72aCRzEWrmP2Q=
-github.com/pierrec/lz4 v2.0.5+incompatible h1:2xWsjqPFWcplujydGg4WmhC/6fZqK42wMM8aXeqhl0I=
-github.com/pierrec/lz4 v2.0.5+incompatible/go.mod h1:pdkljMzZIN41W+lC3N2tnIh5sFi+IEE17M5jbnwPHcY=
+github.com/golang/snappy v0.0.1 h1:Qgr9rKW7uDUkrbSmQeiDsGa8SjGyCOGtuasMWwvp2P4=
+github.com/golang/snappy v0.0.1/go.mod h1:/XxbfmMg8lxefKM7IXC3fBNl/7bRcc72aCRzEWrmP2Q=
+github.com/hashicorp/go-uuid v1.0.1 h1:fv1ep09latC32wFoVwnqcnKJGnMSdBanPczbHAYm1BE=
+github.com/hashicorp/go-uuid v1.0.1/go.mod h1:6SBZvOh/SIDV7/2o3Jml5SYk/TvGqwFJ/bN7x4byOro=
+github.com/jcmturner/gofork v0.0.0-20190328161633-dc7c13fece03 h1:FUwcHNlEqkqLjLBdCp5PRlCFijNjvcYANOZXzCfXwCM=
+github.com/jcmturner/gofork v0.0.0-20190328161633-dc7c13fece03/go.mod h1:MK8+TM0La+2rjBD4jE12Kj1pCCxK7d2LK/UM3ncEo0o=
+github.com/pierrec/lz4 v0.0.0-20190327172049-315a67e90e41 h1:GeinFsrjWz97fAxVUEd748aV0cYL+I6k44gFJTCVvpU=
+github.com/pierrec/lz4 v0.0.0-20190327172049-315a67e90e41/go.mod h1:3/3N9NVKO0jef7pBehbT1qWhCMrIgbYNnFAZCqQ5LRc=
+github.com/pkg/profile v1.2.1/go.mod h1:hJw3o1OdXxsrSjjVksARp5W95eeEaEfptyVZyv6JUPA=
+github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
+github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
github.com/rcrowley/go-metrics v0.0.0-20181016184325-3113b8401b8a h1:9ZKAASQSHhDYGoxY8uLVpewe1GDZ2vu2Tr/vTdVAkFQ=
github.com/rcrowley/go-metrics v0.0.0-20181016184325-3113b8401b8a/go.mod h1:bCqnVzQkZxMG4s8nGwiZ5l3QUCyqpo9Y+/ZMZ9VjZe4=
+github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
+github.com/stretchr/testify v1.3.0 h1:TivCn/peBQ7UY8ooIcPgZFpTNSz0Q2U6UrFlUfqbe0Q=
+github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
+github.com/xdg/scram v0.0.0-20180814205039-7eeb5667e42c h1:u40Z8hqBAAQyv+vATcGgV0YCnDjqSL7/q/JyPhhJSPk=
+github.com/xdg/scram v0.0.0-20180814205039-7eeb5667e42c/go.mod h1:lB8K/P019DLNhemzwFU4jHLhdvlE6uDZjXFejJXr49I=
+github.com/xdg/stringprep v1.0.0 h1:d9X0esnoa3dFsV0FG35rAT0RIhYFlPq7MiP+DW89La0=
+github.com/xdg/stringprep v1.0.0/go.mod h1:Jhud4/sHMO4oL310DaZAKk9ZaJ08SJfe+sJh0HrGL1Y=
+golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2 h1:VklqNMn3ovrHsnt90PveolxSbWFaJdECFbxSq0Mqo2M=
+golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
+golang.org/x/crypto v0.0.0-20190404164418-38d8ce5564a5 h1:bselrhR0Or1vomJZC8ZIjWtbDmn9OYFLX5Ik9alpJpE=
+golang.org/x/crypto v0.0.0-20190404164418-38d8ce5564a5/go.mod h1:WFFai1msRO1wXaEeE5yQxYXgSfI8pQAWXbQop6sCtWE=
+golang.org/x/net v0.0.0-20190404232315-eb5bcb51f2a3 h1:0GoQqolDA55aaLxZyTzK/Y2ePZzZTUrRacwib7cNsYQ=
+golang.org/x/net v0.0.0-20190404232315-eb5bcb51f2a3/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
+golang.org/x/sys v0.0.0-20190215142949-d0b11bdaac8a/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20190403152447-81d4e9dc473e/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/text v0.3.0 h1:g61tztE5qeGQ89tm6NTjjM9VPIm088od1l6aSorWRWg=
+golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
+gopkg.in/jcmturner/aescts.v1 v1.0.1 h1:cVVZBK2b1zY26haWB4vbBiZrfFQnfbTVrE3xZq6hrEw=
+gopkg.in/jcmturner/aescts.v1 v1.0.1/go.mod h1:nsR8qBOg+OucoIW+WMhB3GspUQXq9XorLnQb9XtvcOo=
+gopkg.in/jcmturner/dnsutils.v1 v1.0.1 h1:cIuC1OLRGZrld+16ZJvvZxVJeKPsvd5eUIvxfoN5hSM=
+gopkg.in/jcmturner/dnsutils.v1 v1.0.1/go.mod h1:m3v+5svpVOhtFAP/wSz+yzh4Mc0Fg7eRhxkJMWSIz9Q=
+gopkg.in/jcmturner/gokrb5.v7 v7.2.3 h1:hHMV/yKPwMnJhPuPx7pH2Uw/3Qyf+thJYlisUc44010=
+gopkg.in/jcmturner/gokrb5.v7 v7.2.3/go.mod h1:l8VISx+WGYp+Fp7KRbsiUuXTTOnxIc3Tuvyavf11/WM=
+gopkg.in/jcmturner/rpc.v1 v1.1.0 h1:QHIUxTX1ISuAv9dD2wJ9HWQVuWDX/Zc0PfeC2tjc4rU=
+gopkg.in/jcmturner/rpc.v1 v1.1.0/go.mod h1:YIdkC4XfD6GXbzje11McwsDuOlZQSb9W4vfLvuNnlv8=
diff --git a/vendor/github.com/Shopify/sarama/gssapi_kerberos.go b/vendor/github.com/Shopify/sarama/gssapi_kerberos.go
new file mode 100644
index 0000000..49b632d
--- /dev/null
+++ b/vendor/github.com/Shopify/sarama/gssapi_kerberos.go
@@ -0,0 +1,257 @@
+package sarama
+
+import (
+ "encoding/binary"
+ "fmt"
+ "github.com/jcmturner/gofork/encoding/asn1"
+ "gopkg.in/jcmturner/gokrb5.v7/asn1tools"
+ "gopkg.in/jcmturner/gokrb5.v7/gssapi"
+ "gopkg.in/jcmturner/gokrb5.v7/iana/chksumtype"
+ "gopkg.in/jcmturner/gokrb5.v7/iana/keyusage"
+ "gopkg.in/jcmturner/gokrb5.v7/messages"
+ "gopkg.in/jcmturner/gokrb5.v7/types"
+ "io"
+ "strings"
+ "time"
+)
+
+const (
+ TOK_ID_KRB_AP_REQ = 256
+ GSS_API_GENERIC_TAG = 0x60
+ KRB5_USER_AUTH = 1
+ KRB5_KEYTAB_AUTH = 2
+ GSS_API_INITIAL = 1
+ GSS_API_VERIFY = 2
+ GSS_API_FINISH = 3
+)
+
+type GSSAPIConfig struct {
+ AuthType int
+ KeyTabPath string
+ KerberosConfigPath string
+ ServiceName string
+ Username string
+ Password string
+ Realm string
+}
+
+type GSSAPIKerberosAuth struct {
+ Config *GSSAPIConfig
+ ticket messages.Ticket
+ encKey types.EncryptionKey
+ NewKerberosClientFunc func(config *GSSAPIConfig) (KerberosClient, error)
+ step int
+}
+
+type KerberosClient interface {
+ Login() error
+ GetServiceTicket(spn string) (messages.Ticket, types.EncryptionKey, error)
+ Domain() string
+ CName() types.PrincipalName
+ Destroy()
+}
+
+/*
+*
+* Appends length in big endian before payload, and send it to kafka
+*
+ */
+
+func (krbAuth *GSSAPIKerberosAuth) writePackage(broker *Broker, payload []byte) (int, error) {
+ length := len(payload)
+ finalPackage := make([]byte, length+4) //4 byte length header + payload
+ copy(finalPackage[4:], payload)
+ binary.BigEndian.PutUint32(finalPackage, uint32(length))
+ bytes, err := broker.conn.Write(finalPackage)
+ if err != nil {
+ return bytes, err
+ }
+ return bytes, nil
+}
+
+/*
+*
+* Read length (4 bytes) and then read the payload
+*
+ */
+
+func (krbAuth *GSSAPIKerberosAuth) readPackage(broker *Broker) ([]byte, int, error) {
+ bytesRead := 0
+ lengthInBytes := make([]byte, 4)
+ bytes, err := io.ReadFull(broker.conn, lengthInBytes)
+ if err != nil {
+ return nil, bytesRead, err
+ }
+ bytesRead += bytes
+ payloadLength := binary.BigEndian.Uint32(lengthInBytes)
+ payloadBytes := make([]byte, payloadLength) // buffer for read..
+ bytes, err = io.ReadFull(broker.conn, payloadBytes) // read bytes
+ if err != nil {
+ return payloadBytes, bytesRead, err
+ }
+ bytesRead += bytes
+ return payloadBytes, bytesRead, nil
+}
+
+func (krbAuth *GSSAPIKerberosAuth) newAuthenticatorChecksum() []byte {
+ a := make([]byte, 24)
+ flags := []int{gssapi.ContextFlagInteg, gssapi.ContextFlagConf}
+ binary.LittleEndian.PutUint32(a[:4], 16)
+ for _, i := range flags {
+ f := binary.LittleEndian.Uint32(a[20:24])
+ f |= uint32(i)
+ binary.LittleEndian.PutUint32(a[20:24], f)
+ }
+ return a
+}
+
+/*
+*
+* Construct Kerberos AP_REQ package, conforming to RFC-4120
+* https://tools.ietf.org/html/rfc4120#page-84
+*
+ */
+func (krbAuth *GSSAPIKerberosAuth) createKrb5Token(
+ domain string, cname types.PrincipalName,
+ ticket messages.Ticket,
+ sessionKey types.EncryptionKey) ([]byte, error) {
+ auth, err := types.NewAuthenticator(domain, cname)
+ if err != nil {
+ return nil, err
+ }
+ auth.Cksum = types.Checksum{
+ CksumType: chksumtype.GSSAPI,
+ Checksum: krbAuth.newAuthenticatorChecksum(),
+ }
+ APReq, err := messages.NewAPReq(
+ ticket,
+ sessionKey,
+ auth,
+ )
+ if err != nil {
+ return nil, err
+ }
+ aprBytes := make([]byte, 2)
+ binary.BigEndian.PutUint16(aprBytes, TOK_ID_KRB_AP_REQ)
+ tb, err := APReq.Marshal()
+ if err != nil {
+ return nil, err
+ }
+ aprBytes = append(aprBytes, tb...)
+ return aprBytes, nil
+}
+
+/*
+*
+* Append the GSS-API header to the payload, conforming to RFC-2743
+* Section 3.1, Mechanism-Independent Token Format
+*
+* https://tools.ietf.org/html/rfc2743#page-81
+*
+* GSSAPIHeader + <specific mechanism payload>
+*
+ */
+func (krbAuth *GSSAPIKerberosAuth) appendGSSAPIHeader(payload []byte) ([]byte, error) {
+ oidBytes, err := asn1.Marshal(gssapi.OID(gssapi.OIDKRB5))
+ if err != nil {
+ return nil, err
+ }
+ tkoLengthBytes := asn1tools.MarshalLengthBytes(len(oidBytes) + len(payload))
+ GSSHeader := append([]byte{GSS_API_GENERIC_TAG}, tkoLengthBytes...)
+ GSSHeader = append(GSSHeader, oidBytes...)
+ GSSPackage := append(GSSHeader, payload...)
+ return GSSPackage, nil
+}
+
+func (krbAuth *GSSAPIKerberosAuth) initSecContext(bytes []byte, kerberosClient KerberosClient) ([]byte, error) {
+ switch krbAuth.step {
+ case GSS_API_INITIAL:
+ aprBytes, err := krbAuth.createKrb5Token(
+ kerberosClient.Domain(),
+ kerberosClient.CName(),
+ krbAuth.ticket,
+ krbAuth.encKey)
+ if err != nil {
+ return nil, err
+ }
+ krbAuth.step = GSS_API_VERIFY
+ return krbAuth.appendGSSAPIHeader(aprBytes)
+ case GSS_API_VERIFY:
+ wrapTokenReq := gssapi.WrapToken{}
+ if err := wrapTokenReq.Unmarshal(bytes, true); err != nil {
+ return nil, err
+ }
+ // Validate response.
+ isValid, err := wrapTokenReq.Verify(krbAuth.encKey, keyusage.GSSAPI_ACCEPTOR_SEAL)
+ if !isValid {
+ return nil, err
+ }
+
+ wrapTokenResponse, err := gssapi.NewInitiatorWrapToken(wrapTokenReq.Payload, krbAuth.encKey)
+ if err != nil {
+ return nil, err
+ }
+ krbAuth.step = GSS_API_FINISH
+ return wrapTokenResponse.Marshal()
+ }
+ return nil, nil
+}
+
+/* This does the handshake for authorization */
+func (krbAuth *GSSAPIKerberosAuth) Authorize(broker *Broker) error {
+
+ kerberosClient, err := krbAuth.NewKerberosClientFunc(krbAuth.Config)
+ if err != nil {
+ Logger.Printf("Kerberos client error: %s", err)
+ return err
+ }
+
+ err = kerberosClient.Login()
+ if err != nil {
+ Logger.Printf("Kerberos client error: %s", err)
+ return err
+ }
+ // Construct SPN using serviceName and host
+ // SPN format: <SERVICE>/<FQDN>
+
+ host := strings.SplitN(broker.addr, ":", 2)[0] // Strip port part
+ spn := fmt.Sprintf("%s/%s", broker.conf.Net.SASL.GSSAPI.ServiceName, host)
+
+ ticket, encKey, err := kerberosClient.GetServiceTicket(spn)
+
+ if err != nil {
+ Logger.Printf("Error getting Kerberos service ticket : %s", err)
+ return err
+ }
+ krbAuth.ticket = ticket
+ krbAuth.encKey = encKey
+ krbAuth.step = GSS_API_INITIAL
+ var receivedBytes []byte = nil
+ defer kerberosClient.Destroy()
+ for {
+ packBytes, err := krbAuth.initSecContext(receivedBytes, kerberosClient)
+ if err != nil {
+ Logger.Printf("Error while performing GSSAPI Kerberos Authentication: %s\n", err)
+ return err
+ }
+ requestTime := time.Now()
+ bytesWritten, err := krbAuth.writePackage(broker, packBytes)
+ if err != nil {
+ Logger.Printf("Error while performing GSSAPI Kerberos Authentication: %s\n", err)
+ return err
+ }
+ broker.updateOutgoingCommunicationMetrics(bytesWritten)
+ if krbAuth.step == GSS_API_VERIFY {
+ var bytesRead = 0
+ receivedBytes, bytesRead, err = krbAuth.readPackage(broker)
+ requestLatency := time.Since(requestTime)
+ broker.updateIncomingCommunicationMetrics(bytesRead, requestLatency)
+ if err != nil {
+ Logger.Printf("Error while performing GSSAPI Kerberos Authentication: %s\n", err)
+ return err
+ }
+ } else if krbAuth.step == GSS_API_FINISH {
+ return nil
+ }
+ }
+}
diff --git a/vendor/github.com/Shopify/sarama/kerberos_client.go b/vendor/github.com/Shopify/sarama/kerberos_client.go
new file mode 100644
index 0000000..91b998f
--- /dev/null
+++ b/vendor/github.com/Shopify/sarama/kerberos_client.go
@@ -0,0 +1,51 @@
+package sarama
+
+import (
+ krb5client "gopkg.in/jcmturner/gokrb5.v7/client"
+ krb5config "gopkg.in/jcmturner/gokrb5.v7/config"
+ "gopkg.in/jcmturner/gokrb5.v7/keytab"
+ "gopkg.in/jcmturner/gokrb5.v7/types"
+)
+
+type KerberosGoKrb5Client struct {
+ krb5client.Client
+}
+
+func (c *KerberosGoKrb5Client) Domain() string {
+ return c.Credentials.Domain()
+}
+
+func (c *KerberosGoKrb5Client) CName() types.PrincipalName {
+ return c.Credentials.CName()
+}
+
+/*
+*
+* Create kerberos client used to obtain TGT and TGS tokens
+* used gokrb5 library, which is a pure go kerberos client with
+* some GSS-API capabilities, and SPNEGO support. Kafka does not use SPNEGO
+* it uses pure Kerberos 5 solution (RFC-4121 and RFC-4120).
+*
+ */
+func NewKerberosClient(config *GSSAPIConfig) (KerberosClient, error) {
+ cfg, err := krb5config.Load(config.KerberosConfigPath)
+ if err != nil {
+ return nil, err
+ }
+ return createClient(config, cfg)
+}
+
+func createClient(config *GSSAPIConfig, cfg *krb5config.Config) (KerberosClient, error) {
+ var client *krb5client.Client
+ if config.AuthType == KRB5_KEYTAB_AUTH {
+ kt, err := keytab.Load(config.KeyTabPath)
+ if err != nil {
+ return nil, err
+ }
+ client = krb5client.NewClientWithKeytab(config.Username, config.Realm, kt, cfg)
+ } else {
+ client = krb5client.NewClientWithPassword(config.Username,
+ config.Realm, config.Password, cfg)
+ }
+ return &KerberosGoKrb5Client{*client}, nil
+}
diff --git a/vendor/github.com/Shopify/sarama/length_field.go b/vendor/github.com/Shopify/sarama/length_field.go
index da199a7..7d864f6 100644
--- a/vendor/github.com/Shopify/sarama/length_field.go
+++ b/vendor/github.com/Shopify/sarama/length_field.go
@@ -1,6 +1,9 @@
package sarama
-import "encoding/binary"
+import (
+ "encoding/binary"
+ "sync"
+)
// LengthField implements the PushEncoder and PushDecoder interfaces for calculating 4-byte lengths.
type lengthField struct {
@@ -8,6 +11,20 @@
length int32
}
+var lengthFieldPool = sync.Pool{}
+
+func acquireLengthField() *lengthField {
+ val := lengthFieldPool.Get()
+ if val != nil {
+ return val.(*lengthField)
+ }
+ return &lengthField{}
+}
+
+func releaseLengthField(m *lengthField) {
+ lengthFieldPool.Put(m)
+}
+
func (l *lengthField) decode(pd packetDecoder) error {
var err error
l.length, err = pd.getInt32()
diff --git a/vendor/github.com/Shopify/sarama/message.go b/vendor/github.com/Shopify/sarama/message.go
index f64c79b..7c54748 100644
--- a/vendor/github.com/Shopify/sarama/message.go
+++ b/vendor/github.com/Shopify/sarama/message.go
@@ -5,37 +5,44 @@
"time"
)
-// The lowest 3 bits contain the compression codec used for the message
-const compressionCodecMask int8 = 0x07
+const (
+ //CompressionNone no compression
+ CompressionNone CompressionCodec = iota
+ //CompressionGZIP compression using GZIP
+ CompressionGZIP
+ //CompressionSnappy compression using snappy
+ CompressionSnappy
+ //CompressionLZ4 compression using LZ4
+ CompressionLZ4
+ //CompressionZSTD compression using ZSTD
+ CompressionZSTD
-// Bit 3 set for "LogAppend" timestamps
-const timestampTypeMask = 0x08
+ // The lowest 3 bits contain the compression codec used for the message
+ compressionCodecMask int8 = 0x07
+
+ // Bit 3 set for "LogAppend" timestamps
+ timestampTypeMask = 0x08
+
+ // CompressionLevelDefault is the constant to use in CompressionLevel
+ // to have the default compression level for any codec. The value is picked
+ // that we don't use any existing compression levels.
+ CompressionLevelDefault = -1000
+)
// CompressionCodec represents the various compression codecs recognized by Kafka in messages.
type CompressionCodec int8
-const (
- CompressionNone CompressionCodec = 0
- CompressionGZIP CompressionCodec = 1
- CompressionSnappy CompressionCodec = 2
- CompressionLZ4 CompressionCodec = 3
- CompressionZSTD CompressionCodec = 4
-)
-
func (cc CompressionCodec) String() string {
return []string{
"none",
"gzip",
"snappy",
"lz4",
+ "zstd",
}[int(cc)]
}
-// CompressionLevelDefault is the constant to use in CompressionLevel
-// to have the default compression level for any codec. The value is picked
-// that we don't use any existing compression levels.
-const CompressionLevelDefault = -1000
-
+//Message is a kafka message type
type Message struct {
Codec CompressionCodec // codec used to compress the message contents
CompressionLevel int // compression level
@@ -96,7 +103,10 @@
}
func (m *Message) decode(pd packetDecoder) (err error) {
- err = pd.push(newCRC32Field(crcIEEE))
+ crc32Decoder := acquireCrc32Field(crcIEEE)
+ defer releaseCrc32Field(crc32Decoder)
+
+ err = pd.push(crc32Decoder)
if err != nil {
return err
}
diff --git a/vendor/github.com/Shopify/sarama/message_set.go b/vendor/github.com/Shopify/sarama/message_set.go
index 600c7c4..6523ec2 100644
--- a/vendor/github.com/Shopify/sarama/message_set.go
+++ b/vendor/github.com/Shopify/sarama/message_set.go
@@ -29,7 +29,10 @@
return err
}
- if err = pd.push(&lengthField{}); err != nil {
+ lengthDecoder := acquireLengthField()
+ defer releaseLengthField(lengthDecoder)
+
+ if err = pd.push(lengthDecoder); err != nil {
return err
}
diff --git a/vendor/github.com/Shopify/sarama/metadata_request.go b/vendor/github.com/Shopify/sarama/metadata_request.go
index 17dc428..1b590d3 100644
--- a/vendor/github.com/Shopify/sarama/metadata_request.go
+++ b/vendor/github.com/Shopify/sarama/metadata_request.go
@@ -37,15 +37,8 @@
if err != nil {
return err
}
- if size < 0 {
- return nil
- } else {
- topicCount := size
- if topicCount == 0 {
- return nil
- }
-
- r.Topics = make([]string, topicCount)
+ if size > 0 {
+ r.Topics = make([]string, size)
for i := range r.Topics {
topic, err := pd.getString()
if err != nil {
diff --git a/vendor/github.com/Shopify/sarama/metadata_response.go b/vendor/github.com/Shopify/sarama/metadata_response.go
index c402d05..b2d532e 100644
--- a/vendor/github.com/Shopify/sarama/metadata_response.go
+++ b/vendor/github.com/Shopify/sarama/metadata_response.go
@@ -296,7 +296,7 @@
return tmatch
}
-func (r *MetadataResponse) AddTopicPartition(topic string, partition, brokerID int32, replicas, isr []int32, err KError) {
+func (r *MetadataResponse) AddTopicPartition(topic string, partition, brokerID int32, replicas, isr []int32, offline []int32, err KError) {
tmatch := r.AddTopic(topic, ErrNoError)
var pmatch *PartitionMetadata
@@ -316,6 +316,7 @@
pmatch.Leader = brokerID
pmatch.Replicas = replicas
pmatch.Isr = isr
+ pmatch.OfflineReplicas = offline
pmatch.Err = err
}
diff --git a/vendor/github.com/Shopify/sarama/metrics.go b/vendor/github.com/Shopify/sarama/metrics.go
index 4869708..90e5a87 100644
--- a/vendor/github.com/Shopify/sarama/metrics.go
+++ b/vendor/github.com/Shopify/sarama/metrics.go
@@ -28,14 +28,6 @@
return fmt.Sprintf(name+"-for-broker-%d", broker.ID())
}
-func getOrRegisterBrokerMeter(name string, broker *Broker, r metrics.Registry) metrics.Meter {
- return metrics.GetOrRegisterMeter(getMetricNameForBroker(name, broker), r)
-}
-
-func getOrRegisterBrokerHistogram(name string, broker *Broker, r metrics.Registry) metrics.Histogram {
- return getOrRegisterHistogram(getMetricNameForBroker(name, broker), r)
-}
-
func getMetricNameForTopic(name string, topic string) string {
// Convert dot to _ since reporters like Graphite typically use dot to represent hierarchy
// cf. KAFKA-1902 and KAFKA-2337
diff --git a/vendor/github.com/Shopify/sarama/mockbroker.go b/vendor/github.com/Shopify/sarama/mockbroker.go
index 55ef1e2..4ed46a6 100644
--- a/vendor/github.com/Shopify/sarama/mockbroker.go
+++ b/vendor/github.com/Shopify/sarama/mockbroker.go
@@ -18,6 +18,8 @@
expectationTimeout = 500 * time.Millisecond
)
+type GSSApiHandlerFunc func([]byte) []byte
+
type requestHandlerFunc func(req *request) (res encoder)
// RequestNotifierFunc is invoked when a mock broker processes a request successfully
@@ -49,18 +51,19 @@
// It is not necessary to prefix message length or correlation ID to your
// response bytes, the server does that automatically as a convenience.
type MockBroker struct {
- brokerID int32
- port int32
- closing chan none
- stopper chan none
- expectations chan encoder
- listener net.Listener
- t TestReporter
- latency time.Duration
- handler requestHandlerFunc
- notifier RequestNotifierFunc
- history []RequestResponse
- lock sync.Mutex
+ brokerID int32
+ port int32
+ closing chan none
+ stopper chan none
+ expectations chan encoder
+ listener net.Listener
+ t TestReporter
+ latency time.Duration
+ handler requestHandlerFunc
+ notifier RequestNotifierFunc
+ history []RequestResponse
+ lock sync.Mutex
+ gssApiHandler GSSApiHandlerFunc
}
// RequestResponse represents a Request/Response pair processed by MockBroker.
@@ -173,6 +176,43 @@
Logger.Printf("*** mockbroker/%d: listener closed, err=%v", b.BrokerID(), err)
}
+func (b *MockBroker) SetGSSAPIHandler(handler GSSApiHandlerFunc) {
+ b.gssApiHandler = handler
+}
+
+func (b *MockBroker) readToBytes(r io.Reader) ([]byte, error) {
+ var (
+ bytesRead int
+ lengthBytes = make([]byte, 4)
+ )
+
+ if _, err := io.ReadFull(r, lengthBytes); err != nil {
+ return nil, err
+ }
+
+ bytesRead += len(lengthBytes)
+ length := int32(binary.BigEndian.Uint32(lengthBytes))
+
+ if length <= 4 || length > MaxRequestSize {
+ return nil, PacketDecodingError{fmt.Sprintf("message of length %d too large or too small", length)}
+ }
+
+ encodedReq := make([]byte, length)
+ if _, err := io.ReadFull(r, encodedReq); err != nil {
+ return nil, err
+ }
+
+ bytesRead += len(encodedReq)
+
+ fullBytes := append(lengthBytes, encodedReq...)
+
+ return fullBytes, nil
+}
+
+func (b *MockBroker) isGSSAPI(buffer []byte) bool {
+ return buffer[4] == 0x60 || bytes.Equal(buffer[4:6], []byte{0x05, 0x04})
+}
+
func (b *MockBroker) handleRequests(conn net.Conn, idx int, wg *sync.WaitGroup) {
defer wg.Done()
defer func() {
@@ -192,59 +232,92 @@
}()
resHeader := make([]byte, 8)
+ var bytesWritten int
+ var bytesRead int
for {
- req, bytesRead, err := decodeRequest(conn)
+
+ buffer, err := b.readToBytes(conn)
if err != nil {
- Logger.Printf("*** mockbroker/%d/%d: invalid request: err=%+v, %+v", b.brokerID, idx, err, spew.Sdump(req))
+ Logger.Printf("*** mockbroker/%d/%d: invalid request: err=%+v, %+v", b.brokerID, idx, err, spew.Sdump(buffer))
b.serverError(err)
break
}
- if b.latency > 0 {
- time.Sleep(b.latency)
- }
+ bytesWritten = 0
+ if !b.isGSSAPI(buffer) {
- b.lock.Lock()
- res := b.handler(req)
- b.history = append(b.history, RequestResponse{req.body, res})
- b.lock.Unlock()
-
- if res == nil {
- Logger.Printf("*** mockbroker/%d/%d: ignored %v", b.brokerID, idx, spew.Sdump(req))
- continue
- }
- Logger.Printf("*** mockbroker/%d/%d: served %v -> %v", b.brokerID, idx, req, res)
-
- encodedRes, err := encode(res, nil)
- if err != nil {
- b.serverError(err)
- break
- }
- if len(encodedRes) == 0 {
- b.lock.Lock()
- if b.notifier != nil {
- b.notifier(bytesRead, 0)
+ req, br, err := decodeRequest(bytes.NewReader(buffer))
+ bytesRead = br
+ if err != nil {
+ Logger.Printf("*** mockbroker/%d/%d: invalid request: err=%+v, %+v", b.brokerID, idx, err, spew.Sdump(req))
+ b.serverError(err)
+ break
}
- b.lock.Unlock()
- continue
- }
- binary.BigEndian.PutUint32(resHeader, uint32(len(encodedRes)+4))
- binary.BigEndian.PutUint32(resHeader[4:], uint32(req.correlationID))
- if _, err = conn.Write(resHeader); err != nil {
- b.serverError(err)
- break
- }
- if _, err = conn.Write(encodedRes); err != nil {
- b.serverError(err)
- break
+ if b.latency > 0 {
+ time.Sleep(b.latency)
+ }
+
+ b.lock.Lock()
+ res := b.handler(req)
+ b.history = append(b.history, RequestResponse{req.body, res})
+ b.lock.Unlock()
+
+ if res == nil {
+ Logger.Printf("*** mockbroker/%d/%d: ignored %v", b.brokerID, idx, spew.Sdump(req))
+ continue
+ }
+ Logger.Printf("*** mockbroker/%d/%d: served %v -> %v", b.brokerID, idx, req, res)
+
+ encodedRes, err := encode(res, nil)
+ if err != nil {
+ b.serverError(err)
+ break
+ }
+ if len(encodedRes) == 0 {
+ b.lock.Lock()
+ if b.notifier != nil {
+ b.notifier(bytesRead, 0)
+ }
+ b.lock.Unlock()
+ continue
+ }
+
+ binary.BigEndian.PutUint32(resHeader, uint32(len(encodedRes)+4))
+ binary.BigEndian.PutUint32(resHeader[4:], uint32(req.correlationID))
+ if _, err = conn.Write(resHeader); err != nil {
+ b.serverError(err)
+ break
+ }
+ if _, err = conn.Write(encodedRes); err != nil {
+ b.serverError(err)
+ break
+ }
+ bytesWritten = len(resHeader) + len(encodedRes)
+
+ } else {
+ // GSSAPI is not part of kafka protocol, but is supported for authentication proposes.
+ // Don't support history for this kind of request as is only used for test GSSAPI authentication mechanism
+ b.lock.Lock()
+ res := b.gssApiHandler(buffer)
+ b.lock.Unlock()
+ if res == nil {
+ Logger.Printf("*** mockbroker/%d/%d: ignored %v", b.brokerID, idx, spew.Sdump(buffer))
+ continue
+ }
+ if _, err = conn.Write(res); err != nil {
+ b.serverError(err)
+ break
+ }
+ bytesWritten = len(res)
}
b.lock.Lock()
if b.notifier != nil {
- b.notifier(bytesRead, len(resHeader)+len(encodedRes))
+ b.notifier(bytesRead, bytesWritten)
}
b.lock.Unlock()
+
}
Logger.Printf("*** mockbroker/%d/%d: connection closed, err=%v", b.BrokerID(), idx, err)
}
diff --git a/vendor/github.com/Shopify/sarama/mockkerberos.go b/vendor/github.com/Shopify/sarama/mockkerberos.go
new file mode 100644
index 0000000..affeb2d
--- /dev/null
+++ b/vendor/github.com/Shopify/sarama/mockkerberos.go
@@ -0,0 +1,123 @@
+package sarama
+
+import (
+ "encoding/binary"
+ "encoding/hex"
+ "gopkg.in/jcmturner/gokrb5.v7/credentials"
+ "gopkg.in/jcmturner/gokrb5.v7/gssapi"
+ "gopkg.in/jcmturner/gokrb5.v7/iana/keyusage"
+ "gopkg.in/jcmturner/gokrb5.v7/messages"
+ "gopkg.in/jcmturner/gokrb5.v7/types"
+)
+
+type KafkaGSSAPIHandler struct {
+ client *MockKerberosClient
+ badResponse bool
+ badKeyChecksum bool
+}
+
+func (h *KafkaGSSAPIHandler) MockKafkaGSSAPI(buffer []byte) []byte {
+ // Default payload used for verify
+ err := h.client.Login() // Mock client construct keys when login
+ if err != nil {
+ return nil
+ }
+ if h.badResponse { // Returns trash
+ return []byte{0x00, 0x00, 0x00, 0x01, 0xAD}
+ }
+
+ var pack = gssapi.WrapToken{
+ Flags: KRB5_USER_AUTH,
+ EC: 12,
+ RRC: 0,
+ SndSeqNum: 3398292281,
+ Payload: []byte{0x11, 0x00}, // 1100
+ }
+ // Compute checksum
+ if h.badKeyChecksum {
+ pack.CheckSum = []byte{0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF}
+ } else {
+ err = pack.SetCheckSum(h.client.ASRep.DecryptedEncPart.Key, keyusage.GSSAPI_ACCEPTOR_SEAL)
+ if err != nil {
+ return nil
+ }
+ }
+
+ packBytes, err := pack.Marshal()
+ if err != nil {
+ return nil
+ }
+ lenBytes := len(packBytes)
+ response := make([]byte, lenBytes+4)
+ copy(response[4:], packBytes)
+ binary.BigEndian.PutUint32(response, uint32(lenBytes))
+ return response
+}
+
+type MockKerberosClient struct {
+ asReqBytes string
+ asRepBytes string
+ ASRep messages.ASRep
+ credentials *credentials.Credentials
+ mockError error
+ errorStage string
+}
+
+func (c *MockKerberosClient) Login() error {
+ if c.errorStage == "login" && c.mockError != nil {
+ return c.mockError
+ }
+ c.asRepBytes = "6b8202e9308202e5a003020105a10302010ba22b30293027a103020113a220041e301c301aa003020112a1131b114" +
+ "558414d504c452e434f4d636c69656e74a30d1b0b4558414d504c452e434f4da4133011a003020101a10a30081b06636c69656e7" +
+ "4a5820156618201523082014ea003020105a10d1b0b4558414d504c452e434f4da220301ea003020102a11730151b066b7262746" +
+ "7741b0b4558414d504c452e434f4da382011430820110a003020112a103020101a28201020481ffdb9891175d106818e61008c51" +
+ "d0b3462bca92f3bf9d4cfa82de4c4d7aff9994ec87c573e3a3d54dcb2bb79618c76f2bf4a3d006f90d5bdbd049bc18f48be39203" +
+ "549ca02acaf63f292b12404f9b74c34b83687119d8f56552ccc0c50ebee2a53bb114c1b4619bb1d5d31f0f49b4d40a08a9b4c046" +
+ "2e1398d0b648be1c0e50c552ad16e1d8d8e74263dd0bf0ec591e4797dfd40a9a1be4ae830d03a306e053fd7586fef84ffc5e4a83" +
+ "7c3122bf3e6a40fe87e84019f6283634461b955712b44a5f7386c278bff94ec2c2dc0403247e29c2450e853471ceababf9b8911f" +
+ "997f2e3010b046d2c49eb438afb0f4c210821e80d4ffa4c9521eb895dcd68610b3feaa682012c30820128a003020112a282011f0" +
+ "482011bce73cbce3f1dd17661c412005f0f2257c756fe8e98ff97e6ec24b7bab66e5fd3a3827aeeae4757af0c6e892948122d8b2" +
+ "03c8df48df0ef5d142d0e416d688f11daa0fcd63d96bdd431d02b8e951c664eeff286a2be62383d274a04016d5f0e141da58cb86" +
+ "331de64063062f4f885e8e9ce5b181ca2fdc67897c5995e0ae1ae0c171a64493ff7bd91bc6d89cd4fce1e2b3ea0a10e34b0d5eda" +
+ "aa38ee727b50c5632ed1d2f2b457908e616178d0d80b72af209fb8ac9dbaa1768fa45931392b36b6d8c12400f8ded2efaa0654d0" +
+ "da1db966e8b5aab4706c800f95d559664646041fdb38b411c62fc0fbe0d25083a28562b0e1c8df16e62e9d5626b0addee489835f" +
+ "eedb0f26c05baa596b69b17f47920aa64b29dc77cfcc97ba47885"
+ apRepBytes, err := hex.DecodeString(c.asRepBytes)
+ if err != nil {
+ return err
+ }
+ err = c.ASRep.Unmarshal(apRepBytes)
+ if err != nil {
+ return err
+ }
+ c.credentials = credentials.New("client", "EXAMPLE.COM").WithPassword("qwerty")
+ _, err = c.ASRep.DecryptEncPart(c.credentials)
+ if err != nil {
+ return err
+ }
+ return nil
+}
+
+func (c *MockKerberosClient) GetServiceTicket(spn string) (messages.Ticket, types.EncryptionKey, error) {
+ if c.errorStage == "service_ticket" && c.mockError != nil {
+ return messages.Ticket{}, types.EncryptionKey{}, c.mockError
+ }
+ return c.ASRep.Ticket, c.ASRep.DecryptedEncPart.Key, nil
+}
+
+func (c *MockKerberosClient) Domain() string {
+ return "EXAMPLE.COM"
+}
+func (c *MockKerberosClient) CName() types.PrincipalName {
+ var p = types.PrincipalName{
+ NameType: KRB5_USER_AUTH,
+ NameString: []string{
+ "kafka",
+ "kafka",
+ },
+ }
+ return p
+}
+func (c *MockKerberosClient) Destroy() {
+ // Do nothing.
+}
diff --git a/vendor/github.com/Shopify/sarama/mockresponses.go b/vendor/github.com/Shopify/sarama/mockresponses.go
index 348c223..c78f0ac 100644
--- a/vendor/github.com/Shopify/sarama/mockresponses.go
+++ b/vendor/github.com/Shopify/sarama/mockresponses.go
@@ -2,6 +2,7 @@
import (
"fmt"
+ "strings"
)
// TestReporter has methods matching go's testing.T to avoid importing
@@ -177,7 +178,7 @@
// Generate set of replicas
replicas := []int32{}
-
+ offlineReplicas := []int32{}
for _, brokerID := range mmr.brokers {
replicas = append(replicas, brokerID)
}
@@ -185,14 +186,14 @@
if len(metadataRequest.Topics) == 0 {
for topic, partitions := range mmr.leaders {
for partition, brokerID := range partitions {
- metadataResponse.AddTopicPartition(topic, partition, brokerID, replicas, replicas, ErrNoError)
+ metadataResponse.AddTopicPartition(topic, partition, brokerID, replicas, replicas, offlineReplicas, ErrNoError)
}
}
return metadataResponse
}
for _, topic := range metadataRequest.Topics {
for partition, brokerID := range mmr.leaders[topic] {
- metadataResponse.AddTopicPartition(topic, partition, brokerID, replicas, replicas, ErrNoError)
+ metadataResponse.AddTopicPartition(topic, partition, brokerID, replicas, replicas, offlineReplicas, ErrNoError)
}
}
return metadataResponse
@@ -573,6 +574,7 @@
// MockOffsetFetchResponse is a `OffsetFetchResponse` builder.
type MockOffsetFetchResponse struct {
offsets map[string]map[string]map[int32]*OffsetFetchResponseBlock
+ error KError
t TestReporter
}
@@ -598,15 +600,25 @@
return mr
}
+func (mr *MockOffsetFetchResponse) SetError(kerror KError) *MockOffsetFetchResponse {
+ mr.error = kerror
+ return mr
+}
+
func (mr *MockOffsetFetchResponse) For(reqBody versionedDecoder) encoder {
req := reqBody.(*OffsetFetchRequest)
group := req.ConsumerGroup
- res := &OffsetFetchResponse{}
+ res := &OffsetFetchResponse{Version: req.Version}
+
for topic, partitions := range mr.offsets[group] {
for partition, block := range partitions {
res.AddBlock(topic, partition, block)
}
}
+
+ if res.Version >= 2 {
+ res.Err = mr.error
+ }
return res
}
@@ -620,10 +632,20 @@
func (mr *MockCreateTopicsResponse) For(reqBody versionedDecoder) encoder {
req := reqBody.(*CreateTopicsRequest)
- res := &CreateTopicsResponse{}
+ res := &CreateTopicsResponse{
+ Version: req.Version,
+ }
res.TopicErrors = make(map[string]*TopicError)
- for topic, _ := range req.TopicDetails {
+ for topic := range req.TopicDetails {
+ if res.Version >= 1 && strings.HasPrefix(topic, "_") {
+ msg := "insufficient permissions to create topic with reserved prefix"
+ res.TopicErrors[topic] = &TopicError{
+ Err: ErrTopicAuthorizationFailed,
+ ErrMsg: &msg,
+ }
+ continue
+ }
res.TopicErrors[topic] = &TopicError{Err: ErrNoError}
}
return res
@@ -661,7 +683,15 @@
res := &CreatePartitionsResponse{}
res.TopicPartitionErrors = make(map[string]*TopicPartitionError)
- for topic, _ := range req.TopicPartitions {
+ for topic := range req.TopicPartitions {
+ if strings.HasPrefix(topic, "_") {
+ msg := "insufficient permissions to create partition on topic with reserved prefix"
+ res.TopicPartitionErrors[topic] = &TopicPartitionError{
+ Err: ErrTopicAuthorizationFailed,
+ ErrMsg: &msg,
+ }
+ continue
+ }
res.TopicPartitionErrors[topic] = &TopicPartitionError{Err: ErrNoError}
}
return res
@@ -682,7 +712,7 @@
for topic, deleteRecordRequestTopic := range req.Topics {
partitions := make(map[int32]*DeleteRecordsResponsePartition)
- for partition, _ := range deleteRecordRequestTopic.PartitionOffsets {
+ for partition := range deleteRecordRequestTopic.PartitionOffsets {
partitions[partition] = &DeleteRecordsResponsePartition{Err: ErrNoError}
}
res.Topics[topic] = &DeleteRecordsResponseTopic{Partitions: partitions}
@@ -866,3 +896,26 @@
}
return res
}
+
+type MockDeleteGroupsResponse struct {
+ deletedGroups []string
+}
+
+func NewMockDeleteGroupsRequest(t TestReporter) *MockDeleteGroupsResponse {
+ return &MockDeleteGroupsResponse{}
+}
+
+func (m *MockDeleteGroupsResponse) SetDeletedGroups(groups []string) *MockDeleteGroupsResponse {
+ m.deletedGroups = groups
+ return m
+}
+
+func (m *MockDeleteGroupsResponse) For(reqBody versionedDecoder) encoder {
+ resp := &DeleteGroupsResponse{
+ GroupErrorCodes: map[string]KError{},
+ }
+ for _, group := range m.deletedGroups {
+ resp.GroupErrorCodes[group] = ErrNoError
+ }
+ return resp
+}
diff --git a/vendor/github.com/Shopify/sarama/offset_commit_request.go b/vendor/github.com/Shopify/sarama/offset_commit_request.go
index 1ec583e..5732ed9 100644
--- a/vendor/github.com/Shopify/sarama/offset_commit_request.go
+++ b/vendor/github.com/Shopify/sarama/offset_commit_request.go
@@ -200,11 +200,11 @@
func (r *OffsetCommitRequest) Offset(topic string, partitionID int32) (int64, string, error) {
partitions := r.blocks[topic]
if partitions == nil {
- return 0, "", errors.New("No such offset")
+ return 0, "", errors.New("no such offset")
}
block := partitions[partitionID]
if block == nil {
- return 0, "", errors.New("No such offset")
+ return 0, "", errors.New("no such offset")
}
return block.offset, block.metadata, nil
}
diff --git a/vendor/github.com/Shopify/sarama/offset_manager.go b/vendor/github.com/Shopify/sarama/offset_manager.go
index 2432f7b..923972f 100644
--- a/vendor/github.com/Shopify/sarama/offset_manager.go
+++ b/vendor/github.com/Shopify/sarama/offset_manager.go
@@ -333,7 +333,6 @@
pom.handleError(err)
case ErrOffsetsLoadInProgress:
// nothing wrong but we didn't commit, we'll get it next time round
- break
case ErrUnknownTopicOrPartition:
// let the user know *and* try redispatching - if topic-auto-create is
// enabled, redispatching should trigger a metadata req and create the
@@ -576,6 +575,6 @@
func (pom *partitionOffsetManager) release() {
pom.releaseOnce.Do(func() {
- go close(pom.errors)
+ close(pom.errors)
})
}
diff --git a/vendor/github.com/Shopify/sarama/packet_decoder.go b/vendor/github.com/Shopify/sarama/packet_decoder.go
index 74805cc..9be854c 100644
--- a/vendor/github.com/Shopify/sarama/packet_decoder.go
+++ b/vendor/github.com/Shopify/sarama/packet_decoder.go
@@ -27,6 +27,7 @@
remaining() int
getSubset(length int) (packetDecoder, error)
peek(offset, length int) (packetDecoder, error) // similar to getSubset, but it doesn't advance the offset
+ peekInt8(offset int) (int8, error) // similar to peek, but just one byte
// Stacks, see PushDecoder
push(in pushDecoder) error
diff --git a/vendor/github.com/Shopify/sarama/produce_set.go b/vendor/github.com/Shopify/sarama/produce_set.go
index 4b42d9c..bba0f7e 100644
--- a/vendor/github.com/Shopify/sarama/produce_set.go
+++ b/vendor/github.com/Shopify/sarama/produce_set.go
@@ -81,7 +81,7 @@
if ps.parent.conf.Version.IsAtLeast(V0_11_0_0) {
if ps.parent.conf.Producer.Idempotent && msg.sequenceNumber < set.recordsToSend.RecordBatch.FirstSequence {
- return errors.New("Assertion failed: Message out of sequence added to a batch")
+ return errors.New("assertion failed: message out of sequence added to a batch")
}
// We are being conservative here to avoid having to prep encode the record
size += maximumRecordOverhead
diff --git a/vendor/github.com/Shopify/sarama/real_decoder.go b/vendor/github.com/Shopify/sarama/real_decoder.go
index 23045e7..085cbb3 100644
--- a/vendor/github.com/Shopify/sarama/real_decoder.go
+++ b/vendor/github.com/Shopify/sarama/real_decoder.go
@@ -290,6 +290,14 @@
return &realDecoder{raw: rd.raw[off : off+length]}, nil
}
+func (rd *realDecoder) peekInt8(offset int) (int8, error) {
+ const byteLen = 1
+ if rd.remaining() < offset+byteLen {
+ return -1, ErrInsufficientData
+ }
+ return int8(rd.raw[rd.off+offset]), nil
+}
+
// stacks
func (rd *realDecoder) push(in pushDecoder) error {
diff --git a/vendor/github.com/Shopify/sarama/record.go b/vendor/github.com/Shopify/sarama/record.go
index cded308..cdccfe3 100644
--- a/vendor/github.com/Shopify/sarama/record.go
+++ b/vendor/github.com/Shopify/sarama/record.go
@@ -6,10 +6,12 @@
)
const (
+ isTransactionalMask = 0x10
controlMask = 0x20
maximumRecordOverhead = 5*binary.MaxVarintLen32 + binary.MaxVarintLen64 + 1
)
+//RecordHeader stores key and value for a record header
type RecordHeader struct {
Key []byte
Value []byte
@@ -33,15 +35,16 @@
return nil
}
+//Record is kafka record type
type Record struct {
+ Headers []*RecordHeader
+
Attributes int8
TimestampDelta time.Duration
OffsetDelta int64
Key []byte
Value []byte
- Headers []*RecordHeader
-
- length varintLengthField
+ length varintLengthField
}
func (r *Record) encode(pe packetEncoder) error {
diff --git a/vendor/github.com/Shopify/sarama/record_batch.go b/vendor/github.com/Shopify/sarama/record_batch.go
index a36f7e6..c653763 100644
--- a/vendor/github.com/Shopify/sarama/record_batch.go
+++ b/vendor/github.com/Shopify/sarama/record_batch.go
@@ -45,11 +45,16 @@
FirstSequence int32
Records []*Record
PartialTrailingRecord bool
+ IsTransactional bool
compressedRecords []byte
recordsLen int // uncompressed records size
}
+func (b *RecordBatch) LastOffset() int64 {
+ return b.FirstOffset + int64(b.LastOffsetDelta)
+}
+
func (b *RecordBatch) encode(pe packetEncoder) error {
if b.Version != 2 {
return PacketEncodingError{fmt.Sprintf("unsupported compression codec (%d)", b.Codec)}
@@ -111,7 +116,10 @@
return err
}
- if err = pd.push(&crc32Field{polynomial: crcCastagnoli}); err != nil {
+ crc32Decoder := acquireCrc32Field(crcCastagnoli)
+ defer releaseCrc32Field(crc32Decoder)
+
+ if err = pd.push(crc32Decoder); err != nil {
return err
}
@@ -122,6 +130,7 @@
b.Codec = CompressionCodec(int8(attributes) & compressionCodecMask)
b.Control = attributes&controlMask == controlMask
b.LogAppendTime = attributes×tampTypeMask == timestampTypeMask
+ b.IsTransactional = attributes&isTransactionalMask == isTransactionalMask
if b.LastOffsetDelta, err = pd.getInt32(); err != nil {
return err
@@ -205,6 +214,9 @@
if b.LogAppendTime {
attr |= timestampTypeMask
}
+ if b.IsTransactional {
+ attr |= isTransactionalMask
+ }
return attr
}
diff --git a/vendor/github.com/Shopify/sarama/records.go b/vendor/github.com/Shopify/sarama/records.go
index 192f592..98160c7 100644
--- a/vendor/github.com/Shopify/sarama/records.go
+++ b/vendor/github.com/Shopify/sarama/records.go
@@ -185,10 +185,20 @@
}
func magicValue(pd packetDecoder) (int8, error) {
- dec, err := pd.peek(magicOffset, magicLength)
- if err != nil {
- return 0, err
+ return pd.peekInt8(magicOffset)
+}
+
+func (r *Records) getControlRecord() (ControlRecord, error) {
+ if r.RecordBatch == nil || len(r.RecordBatch.Records) <= 0 {
+ return ControlRecord{}, fmt.Errorf("cannot get control record, record batch is empty")
}
- return dec.getInt8()
+ firstRecord := r.RecordBatch.Records[0]
+ controlRecord := ControlRecord{}
+ err := controlRecord.decode(&realDecoder{raw: firstRecord.Key}, &realDecoder{raw: firstRecord.Value})
+ if err != nil {
+ return ControlRecord{}, err
+ }
+
+ return controlRecord, nil
}
diff --git a/vendor/github.com/Shopify/sarama/request.go b/vendor/github.com/Shopify/sarama/request.go
index f7eea59..5ed8ca4 100644
--- a/vendor/github.com/Shopify/sarama/request.go
+++ b/vendor/github.com/Shopify/sarama/request.go
@@ -20,51 +20,67 @@
body protocolBody
}
-func (r *request) encode(pe packetEncoder) (err error) {
+func (r *request) encode(pe packetEncoder) error {
pe.push(&lengthField{})
pe.putInt16(r.body.key())
pe.putInt16(r.body.version())
pe.putInt32(r.correlationID)
- err = pe.putString(r.clientID)
+
+ err := pe.putString(r.clientID)
if err != nil {
return err
}
+
err = r.body.encode(pe)
if err != nil {
return err
}
+
return pe.pop()
}
func (r *request) decode(pd packetDecoder) (err error) {
- var key int16
- if key, err = pd.getInt16(); err != nil {
+ key, err := pd.getInt16()
+ if err != nil {
return err
}
- var version int16
- if version, err = pd.getInt16(); err != nil {
+
+ version, err := pd.getInt16()
+ if err != nil {
return err
}
- if r.correlationID, err = pd.getInt32(); err != nil {
+
+ r.correlationID, err = pd.getInt32()
+ if err != nil {
return err
}
+
r.clientID, err = pd.getString()
+ if err != nil {
+ return err
+ }
r.body = allocateBody(key, version)
if r.body == nil {
return PacketDecodingError{fmt.Sprintf("unknown request key (%d)", key)}
}
+
return r.body.decode(pd, version)
}
-func decodeRequest(r io.Reader) (req *request, bytesRead int, err error) {
- lengthBytes := make([]byte, 4)
+func decodeRequest(r io.Reader) (*request, int, error) {
+ var (
+ bytesRead int
+ lengthBytes = make([]byte, 4)
+ )
+
if _, err := io.ReadFull(r, lengthBytes); err != nil {
return nil, bytesRead, err
}
- bytesRead += len(lengthBytes)
+ bytesRead += len(lengthBytes)
length := int32(binary.BigEndian.Uint32(lengthBytes))
+
if length <= 4 || length > MaxRequestSize {
return nil, bytesRead, PacketDecodingError{fmt.Sprintf("message of length %d too large or too small", length)}
}
@@ -73,12 +89,14 @@
if _, err := io.ReadFull(r, encodedReq); err != nil {
return nil, bytesRead, err
}
+
bytesRead += len(encodedReq)
- req = &request{}
+ req := &request{}
if err := decode(encodedReq, req); err != nil {
return nil, bytesRead, err
}
+
return req, bytesRead, nil
}
diff --git a/vendor/github.com/Shopify/sarama/response_header.go b/vendor/github.com/Shopify/sarama/response_header.go
index f3f4d27..7a75918 100644
--- a/vendor/github.com/Shopify/sarama/response_header.go
+++ b/vendor/github.com/Shopify/sarama/response_header.go
@@ -2,6 +2,9 @@
import "fmt"
+const responseLengthSize = 4
+const correlationIDSize = 4
+
type responseHeader struct {
length int32
correlationID int32
diff --git a/vendor/github.com/Shopify/sarama/sarama.go b/vendor/github.com/Shopify/sarama/sarama.go
index 7d5dc60..1e0277a 100644
--- a/vendor/github.com/Shopify/sarama/sarama.go
+++ b/vendor/github.com/Shopify/sarama/sarama.go
@@ -10,10 +10,7 @@
depend on the configured value of `Producer.RequiredAcks`. There are configurations where a message acknowledged by the
SyncProducer can still sometimes be lost.
-To consume messages, use the Consumer. Note that Sarama's Consumer implementation does not currently support automatic
-consumer-group rebalancing and offset tracking. For Zookeeper-based tracking (Kafka 0.8.2 and earlier), the
-https://github.com/wvanbergen/kafka library builds on Sarama to add this support. For Kafka-based tracking (Kafka 0.9
-and later), the https://github.com/bsm/sarama-cluster library builds on Sarama to add this support.
+To consume messages, use Consumer or Consumer-Group API.
For lower-level needs, the Broker and Request/Response objects permit precise control over each connection
and message sent on the wire; the Client provides higher-level metadata management that is shared between
@@ -61,6 +58,14 @@
| compression-ratio-for-topic-<topic> | histogram | Distribution of the compression ratio times 100 of record batches for a given topic |
+-------------------------------------------+------------+--------------------------------------------------------------------------------------+
+Consumer related metrics:
+
+ +-------------------------------------------+------------+--------------------------------------------------------------------------------------+
+ | Name | Type | Description |
+ +-------------------------------------------+------------+--------------------------------------------------------------------------------------+
+ | consumer-batch-size | histogram | Distribution of the number of messages in a batch |
+ +-------------------------------------------+------------+--------------------------------------------------------------------------------------+
+
*/
package sarama
@@ -69,10 +74,29 @@
"log"
)
-// Logger is the instance of a StdLogger interface that Sarama writes connection
-// management events to. By default it is set to discard all log messages via ioutil.Discard,
-// but you can set it to redirect wherever you want.
-var Logger StdLogger = log.New(ioutil.Discard, "[Sarama] ", log.LstdFlags)
+var (
+ // Logger is the instance of a StdLogger interface that Sarama writes connection
+ // management events to. By default it is set to discard all log messages via ioutil.Discard,
+ // but you can set it to redirect wherever you want.
+ Logger StdLogger = log.New(ioutil.Discard, "[Sarama] ", log.LstdFlags)
+
+ // PanicHandler is called for recovering from panics spawned internally to the library (and thus
+ // not recoverable by the caller's goroutine). Defaults to nil, which means panics are not recovered.
+ PanicHandler func(interface{})
+
+ // MaxRequestSize is the maximum size (in bytes) of any request that Sarama will attempt to send. Trying
+ // to send a request larger than this will result in an PacketEncodingError. The default of 100 MiB is aligned
+ // with Kafka's default `socket.request.max.bytes`, which is the largest request the broker will attempt
+ // to process.
+ MaxRequestSize int32 = 100 * 1024 * 1024
+
+ // MaxResponseSize is the maximum size (in bytes) of any response that Sarama will attempt to parse. If
+ // a broker returns a response message larger than this value, Sarama will return a PacketDecodingError to
+ // protect the client from running out of memory. Please note that brokers do not have any natural limit on
+ // the size of responses they send. In particular, they can send arbitrarily large fetch responses to consumers
+ // (see https://issues.apache.org/jira/browse/KAFKA-2063).
+ MaxResponseSize int32 = 100 * 1024 * 1024
+)
// StdLogger is used to log error messages.
type StdLogger interface {
@@ -80,20 +104,3 @@
Printf(format string, v ...interface{})
Println(v ...interface{})
}
-
-// PanicHandler is called for recovering from panics spawned internally to the library (and thus
-// not recoverable by the caller's goroutine). Defaults to nil, which means panics are not recovered.
-var PanicHandler func(interface{})
-
-// MaxRequestSize is the maximum size (in bytes) of any request that Sarama will attempt to send. Trying
-// to send a request larger than this will result in an PacketEncodingError. The default of 100 MiB is aligned
-// with Kafka's default `socket.request.max.bytes`, which is the largest request the broker will attempt
-// to process.
-var MaxRequestSize int32 = 100 * 1024 * 1024
-
-// MaxResponseSize is the maximum size (in bytes) of any response that Sarama will attempt to parse. If
-// a broker returns a response message larger than this value, Sarama will return a PacketDecodingError to
-// protect the client from running out of memory. Please note that brokers do not have any natural limit on
-// the size of responses they send. In particular, they can send arbitrarily large fetch responses to consumers
-// (see https://issues.apache.org/jira/browse/KAFKA-2063).
-var MaxResponseSize int32 = 100 * 1024 * 1024
diff --git a/vendor/github.com/Shopify/sarama/utils.go b/vendor/github.com/Shopify/sarama/utils.go
index 7dcbf03..7c815cd 100644
--- a/vendor/github.com/Shopify/sarama/utils.go
+++ b/vendor/github.com/Shopify/sarama/utils.go
@@ -159,6 +159,8 @@
V2_0_0_0 = newKafkaVersion(2, 0, 0, 0)
V2_0_1_0 = newKafkaVersion(2, 0, 1, 0)
V2_1_0_0 = newKafkaVersion(2, 1, 0, 0)
+ V2_2_0_0 = newKafkaVersion(2, 2, 0, 0)
+ V2_3_0_0 = newKafkaVersion(2, 3, 0, 0)
SupportedVersions = []KafkaVersion{
V0_8_2_0,
@@ -181,11 +183,14 @@
V2_0_0_0,
V2_0_1_0,
V2_1_0_0,
+ V2_2_0_0,
+ V2_3_0_0,
}
MinVersion = V0_8_2_0
- MaxVersion = V2_1_0_0
+ MaxVersion = V2_3_0_0
)
+//ParseKafkaVersion parses and returns kafka version or error from a string
func ParseKafkaVersion(s string) (KafkaVersion, error) {
if len(s) < 5 {
return MinVersion, fmt.Errorf("invalid version `%s`", s)
diff --git a/vendor/github.com/armon/go-metrics/.travis.yml b/vendor/github.com/armon/go-metrics/.travis.yml
new file mode 100644
index 0000000..87d230c
--- /dev/null
+++ b/vendor/github.com/armon/go-metrics/.travis.yml
@@ -0,0 +1,13 @@
+language: go
+
+go:
+ - "1.x"
+
+env:
+ - GO111MODULE=on
+
+install:
+ - go get ./...
+
+script:
+ - go test ./...
diff --git a/vendor/github.com/armon/go-metrics/go.mod b/vendor/github.com/armon/go-metrics/go.mod
new file mode 100644
index 0000000..88e1e98
--- /dev/null
+++ b/vendor/github.com/armon/go-metrics/go.mod
@@ -0,0 +1,16 @@
+module github.com/armon/go-metrics
+
+go 1.12
+
+require (
+ github.com/DataDog/datadog-go v2.2.0+incompatible
+ github.com/circonus-labs/circonus-gometrics v2.3.1+incompatible
+ github.com/circonus-labs/circonusllhist v0.1.3 // indirect
+ github.com/hashicorp/go-immutable-radix v1.0.0
+ github.com/hashicorp/go-retryablehttp v0.5.3 // indirect
+ github.com/pascaldekloe/goe v0.1.0
+ github.com/pkg/errors v0.8.1 // indirect
+ github.com/prometheus/client_golang v0.9.2
+ github.com/stretchr/testify v1.3.0 // indirect
+ github.com/tv42/httpunix v0.0.0-20150427012821-b75d8614f926 // indirect
+)
diff --git a/vendor/github.com/armon/go-metrics/go.sum b/vendor/github.com/armon/go-metrics/go.sum
new file mode 100644
index 0000000..5ffd832
--- /dev/null
+++ b/vendor/github.com/armon/go-metrics/go.sum
@@ -0,0 +1,46 @@
+github.com/DataDog/datadog-go v2.2.0+incompatible h1:V5BKkxACZLjzHjSgBbr2gvLA2Ae49yhc6CSY7MLy5k4=
+github.com/DataDog/datadog-go v2.2.0+incompatible/go.mod h1:LButxg5PwREeZtORoXG3tL4fMGNddJ+vMq1mwgfaqoQ=
+github.com/beorn7/perks v0.0.0-20180321164747-3a771d992973 h1:xJ4a3vCFaGF/jqvzLMYoU8P317H5OQ+Via4RmuPwCS0=
+github.com/beorn7/perks v0.0.0-20180321164747-3a771d992973/go.mod h1:Dwedo/Wpr24TaqPxmxbtue+5NUziq4I4S80YR8gNf3Q=
+github.com/circonus-labs/circonus-gometrics v2.3.1+incompatible h1:C29Ae4G5GtYyYMm1aztcyj/J5ckgJm2zwdDajFbx1NY=
+github.com/circonus-labs/circonus-gometrics v2.3.1+incompatible/go.mod h1:nmEj6Dob7S7YxXgwXpfOuvO54S+tGdZdw9fuRZt25Ag=
+github.com/circonus-labs/circonusllhist v0.1.3 h1:TJH+oke8D16535+jHExHj4nQvzlZrj7ug5D7I/orNUA=
+github.com/circonus-labs/circonusllhist v0.1.3/go.mod h1:kMXHVDlOchFAehlya5ePtbp5jckzBHf4XRpQvBOLI+I=
+github.com/davecgh/go-spew v1.1.0 h1:ZDRjVQ15GmhC3fiQ8ni8+OwkZQO4DARzQgrnXU1Liz8=
+github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/golang/protobuf v1.2.0 h1:P3YflyNX/ehuJFLhxviNdFxQPkGK5cDcApsge1SqnvM=
+github.com/golang/protobuf v1.2.0/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
+github.com/hashicorp/go-cleanhttp v0.5.0 h1:wvCrVc9TjDls6+YGAF2hAifE1E5U1+b4tH6KdvN3Gig=
+github.com/hashicorp/go-cleanhttp v0.5.0/go.mod h1:JpRdi6/HCYpAwUzNwuwqhbovhLtngrth3wmdIIUrZ80=
+github.com/hashicorp/go-immutable-radix v1.0.0 h1:AKDB1HM5PWEA7i4nhcpwOrO2byshxBjXVn/J/3+z5/0=
+github.com/hashicorp/go-immutable-radix v1.0.0/go.mod h1:0y9vanUI8NX6FsYoO3zeMjhV/C5i9g4Q3DwcSNZ4P60=
+github.com/hashicorp/go-retryablehttp v0.5.3 h1:QlWt0KvWT0lq8MFppF9tsJGF+ynG7ztc2KIPhzRGk7s=
+github.com/hashicorp/go-retryablehttp v0.5.3/go.mod h1:9B5zBasrRhHXnJnui7y6sL7es7NDiJgTc6Er0maI1Xs=
+github.com/hashicorp/go-uuid v1.0.0 h1:RS8zrF7PhGwyNPOtxSClXXj9HA8feRnJzgnI1RJCSnM=
+github.com/hashicorp/go-uuid v1.0.0/go.mod h1:6SBZvOh/SIDV7/2o3Jml5SYk/TvGqwFJ/bN7x4byOro=
+github.com/hashicorp/golang-lru v0.5.0 h1:CL2msUPvZTLb5O648aiLNJw3hnBxN2+1Jq8rCOH9wdo=
+github.com/hashicorp/golang-lru v0.5.0/go.mod h1:/m3WP610KZHVQ1SGc6re/UDhFvYD7pJ4Ao+sR/qLZy8=
+github.com/matttproud/golang_protobuf_extensions v1.0.1 h1:4hp9jkHxhMHkqkrB3Ix0jegS5sx/RkqARlsWZ6pIwiU=
+github.com/matttproud/golang_protobuf_extensions v1.0.1/go.mod h1:D8He9yQNgCq6Z5Ld7szi9bcBfOoFv/3dc6xSMkL2PC0=
+github.com/pascaldekloe/goe v0.1.0 h1:cBOtyMzM9HTpWjXfbbunk26uA6nG3a8n06Wieeh0MwY=
+github.com/pascaldekloe/goe v0.1.0/go.mod h1:lzWF7FIEvWOWxwDKqyGYQf6ZUaNfKdP144TG7ZOy1lc=
+github.com/pkg/errors v0.8.1 h1:iURUrRGxPUNPdy5/HRSm+Yj6okJ6UtLINN0Q9M4+h3I=
+github.com/pkg/errors v0.8.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
+github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
+github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/prometheus/client_golang v0.9.2 h1:awm861/B8OKDd2I/6o1dy3ra4BamzKhYOiGItCeZ740=
+github.com/prometheus/client_golang v0.9.2/go.mod h1:OsXs2jCmiKlQ1lTBmv21f2mNfw4xf/QclQDMrYNZzcM=
+github.com/prometheus/client_model v0.0.0-20180712105110-5c3871d89910 h1:idejC8f05m9MGOsuEi1ATq9shN03HrxNkD/luQvxCv8=
+github.com/prometheus/client_model v0.0.0-20180712105110-5c3871d89910/go.mod h1:MbSGuTsp3dbXC40dX6PRTWyKYBIrTGTE9sqQNg2J8bo=
+github.com/prometheus/common v0.0.0-20181126121408-4724e9255275 h1:PnBWHBf+6L0jOqq0gIVUe6Yk0/QMZ640k6NvkxcBf+8=
+github.com/prometheus/common v0.0.0-20181126121408-4724e9255275/go.mod h1:daVV7qP5qjZbuso7PdcryaAu0sAZbrN9i7WWcTMWvro=
+github.com/prometheus/procfs v0.0.0-20181204211112-1dc9a6cbc91a h1:9a8MnZMP0X2nLJdBg+pBmGgkJlSaKC2KaQmTCk1XDtE=
+github.com/prometheus/procfs v0.0.0-20181204211112-1dc9a6cbc91a/go.mod h1:c3At6R/oaqEKCNdg8wHV1ftS6bRYblBhIjjI8uT2IGk=
+github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
+github.com/stretchr/testify v1.3.0 h1:TivCn/peBQ7UY8ooIcPgZFpTNSz0Q2U6UrFlUfqbe0Q=
+github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
+github.com/tv42/httpunix v0.0.0-20150427012821-b75d8614f926 h1:G3dpKMzFDjgEh2q1Z7zUUtKa8ViPtH+ocF0bE0g00O8=
+github.com/tv42/httpunix v0.0.0-20150427012821-b75d8614f926/go.mod h1:9ESjWnEqriFuLhtthL60Sar/7RFoluCcXsuvEwTV5KM=
+golang.org/x/net v0.0.0-20181201002055-351d144fa1fc/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/sync v0.0.0-20181108010431-42b317875d0f h1:Bl/8QSvNqXvPGPGXa2z5xUTmV7VDcZyvRZ+QQXkXTZQ=
+golang.org/x/sync v0.0.0-20181108010431-42b317875d0f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
diff --git a/vendor/github.com/armon/go-metrics/inmem.go b/vendor/github.com/armon/go-metrics/inmem.go
index 4e2d6a7..93b0e0a 100644
--- a/vendor/github.com/armon/go-metrics/inmem.go
+++ b/vendor/github.com/armon/go-metrics/inmem.go
@@ -255,11 +255,11 @@
}
copyCurrent.Counters = make(map[string]SampledValue, len(current.Counters))
for k, v := range current.Counters {
- copyCurrent.Counters[k] = v
+ copyCurrent.Counters[k] = v.deepCopy()
}
copyCurrent.Samples = make(map[string]SampledValue, len(current.Samples))
for k, v := range current.Samples {
- copyCurrent.Samples[k] = v
+ copyCurrent.Samples[k] = v.deepCopy()
}
current.RUnlock()
diff --git a/vendor/github.com/armon/go-metrics/inmem_endpoint.go b/vendor/github.com/armon/go-metrics/inmem_endpoint.go
index 504f1b3..5fac958 100644
--- a/vendor/github.com/armon/go-metrics/inmem_endpoint.go
+++ b/vendor/github.com/armon/go-metrics/inmem_endpoint.go
@@ -41,6 +41,16 @@
DisplayLabels map[string]string `json:"Labels"`
}
+// deepCopy allocates a new instance of AggregateSample
+func (source *SampledValue) deepCopy() SampledValue {
+ dest := *source
+ if source.AggregateSample != nil {
+ dest.AggregateSample = &AggregateSample{}
+ *dest.AggregateSample = *source.AggregateSample
+ }
+ return dest
+}
+
// DisplayMetrics returns a summary of the metrics from the most recent finished interval.
func (i *InmemSink) DisplayMetrics(resp http.ResponseWriter, req *http.Request) (interface{}, error) {
data := i.Data()
@@ -52,12 +62,15 @@
return nil, fmt.Errorf("no metric intervals have been initialized yet")
case n == 1:
// Show the current interval if it's all we have
- interval = i.intervals[0]
+ interval = data[0]
default:
// Show the most recent finished interval if we have one
- interval = i.intervals[n-2]
+ interval = data[n-2]
}
+ interval.RLock()
+ defer interval.RUnlock()
+
summary := MetricsSummary{
Timestamp: interval.Interval.Round(time.Second).UTC().String(),
Gauges: make([]GaugeValue, 0, len(interval.Gauges)),
diff --git a/vendor/github.com/armon/go-metrics/metrics.go b/vendor/github.com/armon/go-metrics/metrics.go
index cf9def7..4920d68 100644
--- a/vendor/github.com/armon/go-metrics/metrics.go
+++ b/vendor/github.com/armon/go-metrics/metrics.go
@@ -197,7 +197,7 @@
if labels == nil {
return nil
}
- toReturn := labels[:0]
+ toReturn := []Label{}
for _, label := range labels {
if m.labelIsAllowed(&label) {
toReturn = append(toReturn, label)
diff --git a/vendor/github.com/gogo/protobuf/gogoproto/gogo.pb.go b/vendor/github.com/gogo/protobuf/gogoproto/gogo.pb.go
index e352808..1e91766 100644
--- a/vendor/github.com/gogo/protobuf/gogoproto/gogo.pb.go
+++ b/vendor/github.com/gogo/protobuf/gogoproto/gogo.pb.go
@@ -19,7 +19,7 @@
// is compatible with the proto package it is being compiled against.
// A compilation error at this line likely means your copy of the
// proto package needs to be updated.
-const _ = proto.GoGoProtoPackageIsVersion2 // please upgrade the proto package
+const _ = proto.GoGoProtoPackageIsVersion3 // please upgrade the proto package
var E_GoprotoEnumPrefix = &proto.ExtensionDesc{
ExtendedType: (*descriptor.EnumOptions)(nil),
diff --git a/vendor/github.com/gogo/protobuf/proto/extensions.go b/vendor/github.com/gogo/protobuf/proto/extensions.go
index 686bd2a..341c6f5 100644
--- a/vendor/github.com/gogo/protobuf/proto/extensions.go
+++ b/vendor/github.com/gogo/protobuf/proto/extensions.go
@@ -527,6 +527,7 @@
// SetExtension sets the specified extension of pb to the specified value.
func SetExtension(pb Message, extension *ExtensionDesc, value interface{}) error {
if epb, ok := pb.(extensionsBytes); ok {
+ ClearExtension(pb, extension)
newb, err := encodeExtension(extension, value)
if err != nil {
return err
diff --git a/vendor/github.com/gogo/protobuf/proto/extensions_gogo.go b/vendor/github.com/gogo/protobuf/proto/extensions_gogo.go
index 53ebd8c..6f1ae12 100644
--- a/vendor/github.com/gogo/protobuf/proto/extensions_gogo.go
+++ b/vendor/github.com/gogo/protobuf/proto/extensions_gogo.go
@@ -154,6 +154,10 @@
return EncodeExtensionMap(m.extensionsWrite(), data)
}
+func EncodeInternalExtensionBackwards(m extendableProto, data []byte) (n int, err error) {
+ return EncodeExtensionMapBackwards(m.extensionsWrite(), data)
+}
+
func EncodeExtensionMap(m map[int32]Extension, data []byte) (n int, err error) {
o := 0
for _, e := range m {
@@ -169,6 +173,23 @@
return o, nil
}
+func EncodeExtensionMapBackwards(m map[int32]Extension, data []byte) (n int, err error) {
+ o := 0
+ end := len(data)
+ for _, e := range m {
+ if err := e.Encode(); err != nil {
+ return 0, err
+ }
+ n := copy(data[end-len(e.enc):], e.enc)
+ if n != len(e.enc) {
+ return 0, io.ErrShortBuffer
+ }
+ end -= n
+ o += n
+ }
+ return o, nil
+}
+
func GetRawExtension(m map[int32]Extension, id int32) ([]byte, error) {
e := m[id]
if err := e.Encode(); err != nil {
diff --git a/vendor/github.com/gogo/protobuf/proto/lib.go b/vendor/github.com/gogo/protobuf/proto/lib.go
index d17f802..80db1c1 100644
--- a/vendor/github.com/gogo/protobuf/proto/lib.go
+++ b/vendor/github.com/gogo/protobuf/proto/lib.go
@@ -948,13 +948,19 @@
return false
}
-// ProtoPackageIsVersion2 is referenced from generated protocol buffer files
-// to assert that that code is compatible with this version of the proto package.
-const GoGoProtoPackageIsVersion2 = true
+const (
+ // ProtoPackageIsVersion3 is referenced from generated protocol buffer files
+ // to assert that that code is compatible with this version of the proto package.
+ GoGoProtoPackageIsVersion3 = true
-// ProtoPackageIsVersion1 is referenced from generated protocol buffer files
-// to assert that that code is compatible with this version of the proto package.
-const GoGoProtoPackageIsVersion1 = true
+ // ProtoPackageIsVersion2 is referenced from generated protocol buffer files
+ // to assert that that code is compatible with this version of the proto package.
+ GoGoProtoPackageIsVersion2 = true
+
+ // ProtoPackageIsVersion1 is referenced from generated protocol buffer files
+ // to assert that that code is compatible with this version of the proto package.
+ GoGoProtoPackageIsVersion1 = true
+)
// InternalMessageInfo is a type used internally by generated .pb.go files.
// This type is not intended to be used by non-generated code.
diff --git a/vendor/github.com/gogo/protobuf/proto/properties.go b/vendor/github.com/gogo/protobuf/proto/properties.go
index c9e5fa0..62c5562 100644
--- a/vendor/github.com/gogo/protobuf/proto/properties.go
+++ b/vendor/github.com/gogo/protobuf/proto/properties.go
@@ -400,6 +400,15 @@
return sprop
}
+type (
+ oneofFuncsIface interface {
+ XXX_OneofFuncs() (func(Message, *Buffer) error, func(Message, int, int, *Buffer) (bool, error), func(Message) int, []interface{})
+ }
+ oneofWrappersIface interface {
+ XXX_OneofWrappers() []interface{}
+ }
+)
+
// getPropertiesLocked requires that propertiesMu is held.
func getPropertiesLocked(t reflect.Type) *StructProperties {
if prop, ok := propertiesMap[t]; ok {
@@ -441,37 +450,40 @@
// Re-order prop.order.
sort.Sort(prop)
- type oneofMessage interface {
- XXX_OneofFuncs() (func(Message, *Buffer) error, func(Message, int, int, *Buffer) (bool, error), func(Message) int, []interface{})
- }
- if om, ok := reflect.Zero(reflect.PtrTo(t)).Interface().(oneofMessage); isOneofMessage && ok {
+ if isOneofMessage {
var oots []interface{}
- _, _, _, oots = om.XXX_OneofFuncs()
-
- // Interpret oneof metadata.
- prop.OneofTypes = make(map[string]*OneofProperties)
- for _, oot := range oots {
- oop := &OneofProperties{
- Type: reflect.ValueOf(oot).Type(), // *T
- Prop: new(Properties),
- }
- sft := oop.Type.Elem().Field(0)
- oop.Prop.Name = sft.Name
- oop.Prop.Parse(sft.Tag.Get("protobuf"))
- // There will be exactly one interface field that
- // this new value is assignable to.
- for i := 0; i < t.NumField(); i++ {
- f := t.Field(i)
- if f.Type.Kind() != reflect.Interface {
- continue
+ switch m := reflect.Zero(reflect.PtrTo(t)).Interface().(type) {
+ case oneofFuncsIface:
+ _, _, _, oots = m.XXX_OneofFuncs()
+ case oneofWrappersIface:
+ oots = m.XXX_OneofWrappers()
+ }
+ if len(oots) > 0 {
+ // Interpret oneof metadata.
+ prop.OneofTypes = make(map[string]*OneofProperties)
+ for _, oot := range oots {
+ oop := &OneofProperties{
+ Type: reflect.ValueOf(oot).Type(), // *T
+ Prop: new(Properties),
}
- if !oop.Type.AssignableTo(f.Type) {
- continue
+ sft := oop.Type.Elem().Field(0)
+ oop.Prop.Name = sft.Name
+ oop.Prop.Parse(sft.Tag.Get("protobuf"))
+ // There will be exactly one interface field that
+ // this new value is assignable to.
+ for i := 0; i < t.NumField(); i++ {
+ f := t.Field(i)
+ if f.Type.Kind() != reflect.Interface {
+ continue
+ }
+ if !oop.Type.AssignableTo(f.Type) {
+ continue
+ }
+ oop.Field = i
+ break
}
- oop.Field = i
- break
+ prop.OneofTypes[oop.Prop.OrigName] = oop
}
- prop.OneofTypes[oop.Prop.OrigName] = oop
}
}
diff --git a/vendor/github.com/gogo/protobuf/proto/table_marshal.go b/vendor/github.com/gogo/protobuf/proto/table_marshal.go
index 9b1538d..db9927a 100644
--- a/vendor/github.com/gogo/protobuf/proto/table_marshal.go
+++ b/vendor/github.com/gogo/protobuf/proto/table_marshal.go
@@ -389,8 +389,13 @@
// get oneof implementers
var oneofImplementers []interface{}
// gogo: isOneofMessage is needed for embedded oneof messages, without a marshaler and unmarshaler
- if m, ok := reflect.Zero(reflect.PtrTo(t)).Interface().(oneofMessage); ok && isOneofMessage {
- _, _, _, oneofImplementers = m.XXX_OneofFuncs()
+ if isOneofMessage {
+ switch m := reflect.Zero(reflect.PtrTo(t)).Interface().(type) {
+ case oneofFuncsIface:
+ _, _, _, oneofImplementers = m.XXX_OneofFuncs()
+ case oneofWrappersIface:
+ oneofImplementers = m.XXX_OneofWrappers()
+ }
}
// normal fields
@@ -519,10 +524,6 @@
}
}
-type oneofMessage interface {
- XXX_OneofFuncs() (func(Message, *Buffer) error, func(Message, int, int, *Buffer) (bool, error), func(Message) int, []interface{})
-}
-
// wiretype returns the wire encoding of the type.
func wiretype(encoding string) uint64 {
switch encoding {
diff --git a/vendor/github.com/gogo/protobuf/proto/table_merge.go b/vendor/github.com/gogo/protobuf/proto/table_merge.go
index f520106..60dcf70 100644
--- a/vendor/github.com/gogo/protobuf/proto/table_merge.go
+++ b/vendor/github.com/gogo/protobuf/proto/table_merge.go
@@ -530,6 +530,25 @@
}
case reflect.Struct:
switch {
+ case isSlice && !isPointer: // E.g. []pb.T
+ mergeInfo := getMergeInfo(tf)
+ zero := reflect.Zero(tf)
+ mfi.merge = func(dst, src pointer) {
+ // TODO: Make this faster?
+ dstsp := dst.asPointerTo(f.Type)
+ dsts := dstsp.Elem()
+ srcs := src.asPointerTo(f.Type).Elem()
+ for i := 0; i < srcs.Len(); i++ {
+ dsts = reflect.Append(dsts, zero)
+ srcElement := srcs.Index(i).Addr()
+ dstElement := dsts.Index(dsts.Len() - 1).Addr()
+ mergeInfo.merge(valToPointer(dstElement), valToPointer(srcElement))
+ }
+ if dsts.IsNil() {
+ dsts = reflect.MakeSlice(f.Type, 0, 0)
+ }
+ dstsp.Elem().Set(dsts)
+ }
case !isPointer:
mergeInfo := getMergeInfo(tf)
mfi.merge = func(dst, src pointer) {
diff --git a/vendor/github.com/gogo/protobuf/proto/table_unmarshal.go b/vendor/github.com/gogo/protobuf/proto/table_unmarshal.go
index bb2622f..9372293 100644
--- a/vendor/github.com/gogo/protobuf/proto/table_unmarshal.go
+++ b/vendor/github.com/gogo/protobuf/proto/table_unmarshal.go
@@ -371,15 +371,18 @@
}
// Find any types associated with oneof fields.
- // TODO: XXX_OneofFuncs returns more info than we need. Get rid of some of it?
- fn := reflect.Zero(reflect.PtrTo(t)).MethodByName("XXX_OneofFuncs")
// gogo: len(oneofFields) > 0 is needed for embedded oneof messages, without a marshaler and unmarshaler
- if fn.IsValid() && len(oneofFields) > 0 {
- res := fn.Call(nil)[3] // last return value from XXX_OneofFuncs: []interface{}
- for i := res.Len() - 1; i >= 0; i-- {
- v := res.Index(i) // interface{}
- tptr := reflect.ValueOf(v.Interface()).Type() // *Msg_X
- typ := tptr.Elem() // Msg_X
+ if len(oneofFields) > 0 {
+ var oneofImplementers []interface{}
+ switch m := reflect.Zero(reflect.PtrTo(t)).Interface().(type) {
+ case oneofFuncsIface:
+ _, _, _, oneofImplementers = m.XXX_OneofFuncs()
+ case oneofWrappersIface:
+ oneofImplementers = m.XXX_OneofWrappers()
+ }
+ for _, v := range oneofImplementers {
+ tptr := reflect.TypeOf(v) // *Msg_X
+ typ := tptr.Elem() // Msg_X
f := typ.Field(0) // oneof implementers have one field
baseUnmarshal := fieldUnmarshaler(&f)
@@ -407,11 +410,12 @@
u.setTag(fieldNum, of.field, unmarshal, 0, name)
}
}
+
}
}
// Get extension ranges, if any.
- fn = reflect.Zero(reflect.PtrTo(t)).MethodByName("ExtensionRangeArray")
+ fn := reflect.Zero(reflect.PtrTo(t)).MethodByName("ExtensionRangeArray")
if fn.IsValid() {
if !u.extensions.IsValid() && !u.oldExtensions.IsValid() && !u.bytesExtensions.IsValid() {
panic("a message with extensions, but no extensions field in " + t.Name())
diff --git a/vendor/github.com/gogo/protobuf/protoc-gen-gogo/descriptor/descriptor.pb.go b/vendor/github.com/gogo/protobuf/protoc-gen-gogo/descriptor/descriptor.pb.go
index cacfa39..d1307d9 100644
--- a/vendor/github.com/gogo/protobuf/protoc-gen-gogo/descriptor/descriptor.pb.go
+++ b/vendor/github.com/gogo/protobuf/protoc-gen-gogo/descriptor/descriptor.pb.go
@@ -18,7 +18,7 @@
// is compatible with the proto package it is being compiled against.
// A compilation error at this line likely means your copy of the
// proto package needs to be updated.
-const _ = proto.GoGoProtoPackageIsVersion2 // please upgrade the proto package
+const _ = proto.GoGoProtoPackageIsVersion3 // please upgrade the proto package
type FieldDescriptorProto_Type int32
diff --git a/vendor/github.com/golang/protobuf/proto/properties.go b/vendor/github.com/golang/protobuf/proto/properties.go
index 79668ff..a4b8c0c 100644
--- a/vendor/github.com/golang/protobuf/proto/properties.go
+++ b/vendor/github.com/golang/protobuf/proto/properties.go
@@ -38,7 +38,6 @@
import (
"fmt"
"log"
- "os"
"reflect"
"sort"
"strconv"
@@ -194,7 +193,7 @@
// "bytes,49,opt,name=foo,def=hello!"
fields := strings.Split(s, ",") // breaks def=, but handled below.
if len(fields) < 2 {
- fmt.Fprintf(os.Stderr, "proto: tag has too few fields: %q\n", s)
+ log.Printf("proto: tag has too few fields: %q", s)
return
}
@@ -214,7 +213,7 @@
p.WireType = WireBytes
// no numeric converter for non-numeric types
default:
- fmt.Fprintf(os.Stderr, "proto: tag has unknown wire type: %q\n", s)
+ log.Printf("proto: tag has unknown wire type: %q", s)
return
}
diff --git a/vendor/github.com/hashicorp/consul/NOTICE.md b/vendor/github.com/hashicorp/consul/NOTICE.md
deleted file mode 100644
index fe34b5e..0000000
--- a/vendor/github.com/hashicorp/consul/NOTICE.md
+++ /dev/null
@@ -1,3 +0,0 @@
-Copyright © 2014-2018 HashiCorp, Inc.
-
-This Source Code Form is subject to the terms of the Mozilla Public License, v. 2.0. If a copy of the MPL was not distributed with this project, you can obtain one at http://mozilla.org/MPL/2.0/.
diff --git a/vendor/github.com/hashicorp/consul/LICENSE b/vendor/github.com/hashicorp/consul/api/LICENSE
similarity index 100%
rename from vendor/github.com/hashicorp/consul/LICENSE
rename to vendor/github.com/hashicorp/consul/api/LICENSE
diff --git a/vendor/github.com/hashicorp/consul/api/acl.go b/vendor/github.com/hashicorp/consul/api/acl.go
index 53a0523..124409f 100644
--- a/vendor/github.com/hashicorp/consul/api/acl.go
+++ b/vendor/github.com/hashicorp/consul/api/acl.go
@@ -4,7 +4,10 @@
"fmt"
"io"
"io/ioutil"
+ "net/url"
"time"
+
+ "github.com/mitchellh/mapstructure"
)
const (
@@ -19,18 +22,26 @@
ID string
Name string
}
+type ACLTokenRoleLink struct {
+ ID string
+ Name string
+}
// ACLToken represents an ACL Token
type ACLToken struct {
- CreateIndex uint64
- ModifyIndex uint64
- AccessorID string
- SecretID string
- Description string
- Policies []*ACLTokenPolicyLink
- Local bool
- CreateTime time.Time `json:",omitempty"`
- Hash []byte `json:",omitempty"`
+ CreateIndex uint64
+ ModifyIndex uint64
+ AccessorID string
+ SecretID string
+ Description string
+ Policies []*ACLTokenPolicyLink `json:",omitempty"`
+ Roles []*ACLTokenRoleLink `json:",omitempty"`
+ ServiceIdentities []*ACLServiceIdentity `json:",omitempty"`
+ Local bool
+ ExpirationTTL time.Duration `json:",omitempty"`
+ ExpirationTime *time.Time `json:",omitempty"`
+ CreateTime time.Time `json:",omitempty"`
+ Hash []byte `json:",omitempty"`
// DEPRECATED (ACL-Legacy-Compat)
// Rules will only be present for legacy tokens returned via the new APIs
@@ -38,15 +49,18 @@
}
type ACLTokenListEntry struct {
- CreateIndex uint64
- ModifyIndex uint64
- AccessorID string
- Description string
- Policies []*ACLTokenPolicyLink
- Local bool
- CreateTime time.Time
- Hash []byte
- Legacy bool
+ CreateIndex uint64
+ ModifyIndex uint64
+ AccessorID string
+ Description string
+ Policies []*ACLTokenPolicyLink `json:",omitempty"`
+ Roles []*ACLTokenRoleLink `json:",omitempty"`
+ ServiceIdentities []*ACLServiceIdentity `json:",omitempty"`
+ Local bool
+ ExpirationTime *time.Time `json:",omitempty"`
+ CreateTime time.Time
+ Hash []byte
+ Legacy bool
}
// ACLEntry is used to represent a legacy ACL token
@@ -67,11 +81,20 @@
SourceDatacenter string
ReplicationType string
ReplicatedIndex uint64
+ ReplicatedRoleIndex uint64
ReplicatedTokenIndex uint64
LastSuccess time.Time
LastError time.Time
}
+// ACLServiceIdentity represents a high-level grant of all necessary privileges
+// to assume the identity of the named Service in the Catalog and within
+// Connect.
+type ACLServiceIdentity struct {
+ ServiceName string
+ Datacenters []string `json:",omitempty"`
+}
+
// ACLPolicy represents an ACL Policy.
type ACLPolicy struct {
ID string
@@ -94,6 +117,113 @@
ModifyIndex uint64
}
+type ACLRolePolicyLink struct {
+ ID string
+ Name string
+}
+
+// ACLRole represents an ACL Role.
+type ACLRole struct {
+ ID string
+ Name string
+ Description string
+ Policies []*ACLRolePolicyLink `json:",omitempty"`
+ ServiceIdentities []*ACLServiceIdentity `json:",omitempty"`
+ Hash []byte
+ CreateIndex uint64
+ ModifyIndex uint64
+}
+
+// BindingRuleBindType is the type of binding rule mechanism used.
+type BindingRuleBindType string
+
+const (
+ // BindingRuleBindTypeService binds to a service identity with the given name.
+ BindingRuleBindTypeService BindingRuleBindType = "service"
+
+ // BindingRuleBindTypeRole binds to pre-existing roles with the given name.
+ BindingRuleBindTypeRole BindingRuleBindType = "role"
+)
+
+type ACLBindingRule struct {
+ ID string
+ Description string
+ AuthMethod string
+ Selector string
+ BindType BindingRuleBindType
+ BindName string
+
+ CreateIndex uint64
+ ModifyIndex uint64
+}
+
+type ACLAuthMethod struct {
+ Name string
+ Type string
+ Description string
+
+ // Configuration is arbitrary configuration for the auth method. This
+ // should only contain primitive values and containers (such as lists and
+ // maps).
+ Config map[string]interface{}
+
+ CreateIndex uint64
+ ModifyIndex uint64
+}
+
+type ACLAuthMethodListEntry struct {
+ Name string
+ Type string
+ Description string
+ CreateIndex uint64
+ ModifyIndex uint64
+}
+
+// ParseKubernetesAuthMethodConfig takes a raw config map and returns a parsed
+// KubernetesAuthMethodConfig.
+func ParseKubernetesAuthMethodConfig(raw map[string]interface{}) (*KubernetesAuthMethodConfig, error) {
+ var config KubernetesAuthMethodConfig
+ decodeConf := &mapstructure.DecoderConfig{
+ Result: &config,
+ WeaklyTypedInput: true,
+ }
+
+ decoder, err := mapstructure.NewDecoder(decodeConf)
+ if err != nil {
+ return nil, err
+ }
+
+ if err := decoder.Decode(raw); err != nil {
+ return nil, fmt.Errorf("error decoding config: %s", err)
+ }
+
+ return &config, nil
+}
+
+// KubernetesAuthMethodConfig is the config for the built-in Consul auth method
+// for Kubernetes.
+type KubernetesAuthMethodConfig struct {
+ Host string `json:",omitempty"`
+ CACert string `json:",omitempty"`
+ ServiceAccountJWT string `json:",omitempty"`
+}
+
+// RenderToConfig converts this into a map[string]interface{} suitable for use
+// in the ACLAuthMethod.Config field.
+func (c *KubernetesAuthMethodConfig) RenderToConfig() map[string]interface{} {
+ return map[string]interface{}{
+ "Host": c.Host,
+ "CACert": c.CACert,
+ "ServiceAccountJWT": c.ServiceAccountJWT,
+ }
+}
+
+type ACLLoginParams struct {
+ AuthMethod string
+ BearerToken string
+ Meta map[string]string `json:",omitempty"`
+}
+
// ACL can be used to query the ACL endpoints
type ACL struct {
c *Client
@@ -266,17 +396,9 @@
return entries, qm, nil
}
-// TokenCreate creates a new ACL token. It requires that the AccessorID and SecretID fields
-// of the ACLToken structure to be empty as these will be filled in by Consul.
+// TokenCreate creates a new ACL token. If either the AccessorID or SecretID fields
+// of the ACLToken structure are empty they will be filled in by Consul.
func (a *ACL) TokenCreate(token *ACLToken, q *WriteOptions) (*ACLToken, *WriteMeta, error) {
- if token.AccessorID != "" {
- return nil, nil, fmt.Errorf("Cannot specify an AccessorID in Token Creation")
- }
-
- if token.SecretID != "" {
- return nil, nil, fmt.Errorf("Cannot specify a SecretID in Token Creation")
- }
-
r := a.c.newRequest("PUT", "/v1/acl/token")
r.setWriteOptions(q)
r.obj = token
@@ -437,7 +559,6 @@
if policy.ID != "" {
return nil, nil, fmt.Errorf("Cannot specify an ID in Policy Creation")
}
-
r := a.c.newRequest("PUT", "/v1/acl/policy")
r.setWriteOptions(q)
r.obj = policy
@@ -460,7 +581,7 @@
// existing policy ID
func (a *ACL) PolicyUpdate(policy *ACLPolicy, q *WriteOptions) (*ACLPolicy, *WriteMeta, error) {
if policy.ID == "" {
- return nil, nil, fmt.Errorf("Must specify an ID in Policy Creation")
+ return nil, nil, fmt.Errorf("Must specify an ID in Policy Update")
}
r := a.c.newRequest("PUT", "/v1/acl/policy/"+policy.ID)
@@ -586,3 +707,410 @@
return string(ruleBytes), nil
}
+
+// RoleCreate will create a new role. It is not allowed for the role parameters
+// ID field to be set as this will be generated by Consul while processing the request.
+func (a *ACL) RoleCreate(role *ACLRole, q *WriteOptions) (*ACLRole, *WriteMeta, error) {
+ if role.ID != "" {
+ return nil, nil, fmt.Errorf("Cannot specify an ID in Role Creation")
+ }
+
+ r := a.c.newRequest("PUT", "/v1/acl/role")
+ r.setWriteOptions(q)
+ r.obj = role
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ var out ACLRole
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, wm, nil
+}
+
+// RoleUpdate updates a role. The ID field of the role parameter must be set to an
+// existing role ID
+func (a *ACL) RoleUpdate(role *ACLRole, q *WriteOptions) (*ACLRole, *WriteMeta, error) {
+ if role.ID == "" {
+ return nil, nil, fmt.Errorf("Must specify an ID in Role Update")
+ }
+
+ r := a.c.newRequest("PUT", "/v1/acl/role/"+role.ID)
+ r.setWriteOptions(q)
+ r.obj = role
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ var out ACLRole
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, wm, nil
+}
+
+// RoleDelete deletes a role given its ID.
+func (a *ACL) RoleDelete(roleID string, q *WriteOptions) (*WriteMeta, error) {
+ r := a.c.newRequest("DELETE", "/v1/acl/role/"+roleID)
+ r.setWriteOptions(q)
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, err
+ }
+ resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ return wm, nil
+}
+
+// RoleRead retrieves the role details (by ID). Returns nil if not found.
+func (a *ACL) RoleRead(roleID string, q *QueryOptions) (*ACLRole, *QueryMeta, error) {
+ r := a.c.newRequest("GET", "/v1/acl/role/"+roleID)
+ r.setQueryOptions(q)
+ found, rtt, resp, err := requireNotFoundOrOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ qm := &QueryMeta{}
+ parseQueryMeta(resp, qm)
+ qm.RequestTime = rtt
+
+ if !found {
+ return nil, qm, nil
+ }
+
+ var out ACLRole
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, qm, nil
+}
+
+// RoleReadByName retrieves the role details (by name). Returns nil if not found.
+func (a *ACL) RoleReadByName(roleName string, q *QueryOptions) (*ACLRole, *QueryMeta, error) {
+ r := a.c.newRequest("GET", "/v1/acl/role/name/"+url.QueryEscape(roleName))
+ r.setQueryOptions(q)
+ found, rtt, resp, err := requireNotFoundOrOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ qm := &QueryMeta{}
+ parseQueryMeta(resp, qm)
+ qm.RequestTime = rtt
+
+ if !found {
+ return nil, qm, nil
+ }
+
+ var out ACLRole
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, qm, nil
+}
+
+// RoleList retrieves a listing of all roles. The listing does not include some
+// metadata for the role as those should be retrieved by subsequent calls to
+// RoleRead.
+func (a *ACL) RoleList(q *QueryOptions) ([]*ACLRole, *QueryMeta, error) {
+ r := a.c.newRequest("GET", "/v1/acl/roles")
+ r.setQueryOptions(q)
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ qm := &QueryMeta{}
+ parseQueryMeta(resp, qm)
+ qm.RequestTime = rtt
+
+ var entries []*ACLRole
+ if err := decodeBody(resp, &entries); err != nil {
+ return nil, nil, err
+ }
+ return entries, qm, nil
+}
+
+// AuthMethodCreate will create a new auth method.
+func (a *ACL) AuthMethodCreate(method *ACLAuthMethod, q *WriteOptions) (*ACLAuthMethod, *WriteMeta, error) {
+ if method.Name == "" {
+ return nil, nil, fmt.Errorf("Must specify a Name in Auth Method Creation")
+ }
+
+ r := a.c.newRequest("PUT", "/v1/acl/auth-method")
+ r.setWriteOptions(q)
+ r.obj = method
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ var out ACLAuthMethod
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, wm, nil
+}
+
+// AuthMethodUpdate updates an auth method.
+func (a *ACL) AuthMethodUpdate(method *ACLAuthMethod, q *WriteOptions) (*ACLAuthMethod, *WriteMeta, error) {
+ if method.Name == "" {
+ return nil, nil, fmt.Errorf("Must specify a Name in Auth Method Update")
+ }
+
+ r := a.c.newRequest("PUT", "/v1/acl/auth-method/"+url.QueryEscape(method.Name))
+ r.setWriteOptions(q)
+ r.obj = method
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ var out ACLAuthMethod
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, wm, nil
+}
+
+// AuthMethodDelete deletes an auth method given its Name.
+func (a *ACL) AuthMethodDelete(methodName string, q *WriteOptions) (*WriteMeta, error) {
+ if methodName == "" {
+ return nil, fmt.Errorf("Must specify a Name in Auth Method Delete")
+ }
+
+ r := a.c.newRequest("DELETE", "/v1/acl/auth-method/"+url.QueryEscape(methodName))
+ r.setWriteOptions(q)
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, err
+ }
+ resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ return wm, nil
+}
+
+// AuthMethodRead retrieves the auth method. Returns nil if not found.
+func (a *ACL) AuthMethodRead(methodName string, q *QueryOptions) (*ACLAuthMethod, *QueryMeta, error) {
+ if methodName == "" {
+ return nil, nil, fmt.Errorf("Must specify a Name in Auth Method Read")
+ }
+
+ r := a.c.newRequest("GET", "/v1/acl/auth-method/"+url.QueryEscape(methodName))
+ r.setQueryOptions(q)
+ found, rtt, resp, err := requireNotFoundOrOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ qm := &QueryMeta{}
+ parseQueryMeta(resp, qm)
+ qm.RequestTime = rtt
+
+ if !found {
+ return nil, qm, nil
+ }
+
+ var out ACLAuthMethod
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, qm, nil
+}
+
+// AuthMethodList retrieves a listing of all auth methods. The listing does not
+// include some metadata for the auth method as those should be retrieved by
+// subsequent calls to AuthMethodRead.
+func (a *ACL) AuthMethodList(q *QueryOptions) ([]*ACLAuthMethodListEntry, *QueryMeta, error) {
+ r := a.c.newRequest("GET", "/v1/acl/auth-methods")
+ r.setQueryOptions(q)
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ qm := &QueryMeta{}
+ parseQueryMeta(resp, qm)
+ qm.RequestTime = rtt
+
+ var entries []*ACLAuthMethodListEntry
+ if err := decodeBody(resp, &entries); err != nil {
+ return nil, nil, err
+ }
+ return entries, qm, nil
+}
+
+// BindingRuleCreate will create a new binding rule. It is not allowed for the
+// binding rule parameter's ID field to be set as this will be generated by
+// Consul while processing the request.
+func (a *ACL) BindingRuleCreate(rule *ACLBindingRule, q *WriteOptions) (*ACLBindingRule, *WriteMeta, error) {
+ if rule.ID != "" {
+ return nil, nil, fmt.Errorf("Cannot specify an ID in Binding Rule Creation")
+ }
+
+ r := a.c.newRequest("PUT", "/v1/acl/binding-rule")
+ r.setWriteOptions(q)
+ r.obj = rule
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ var out ACLBindingRule
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, wm, nil
+}
+
+// BindingRuleUpdate updates a binding rule. The ID field of the role binding
+// rule parameter must be set to an existing binding rule ID.
+func (a *ACL) BindingRuleUpdate(rule *ACLBindingRule, q *WriteOptions) (*ACLBindingRule, *WriteMeta, error) {
+ if rule.ID == "" {
+ return nil, nil, fmt.Errorf("Must specify an ID in Binding Rule Update")
+ }
+
+ r := a.c.newRequest("PUT", "/v1/acl/binding-rule/"+rule.ID)
+ r.setWriteOptions(q)
+ r.obj = rule
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ var out ACLBindingRule
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, wm, nil
+}
+
+// BindingRuleDelete deletes a binding rule given its ID.
+func (a *ACL) BindingRuleDelete(bindingRuleID string, q *WriteOptions) (*WriteMeta, error) {
+ r := a.c.newRequest("DELETE", "/v1/acl/binding-rule/"+bindingRuleID)
+ r.setWriteOptions(q)
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, err
+ }
+ resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ return wm, nil
+}
+
+// BindingRuleRead retrieves the binding rule details. Returns nil if not found.
+func (a *ACL) BindingRuleRead(bindingRuleID string, q *QueryOptions) (*ACLBindingRule, *QueryMeta, error) {
+ r := a.c.newRequest("GET", "/v1/acl/binding-rule/"+bindingRuleID)
+ r.setQueryOptions(q)
+ found, rtt, resp, err := requireNotFoundOrOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ qm := &QueryMeta{}
+ parseQueryMeta(resp, qm)
+ qm.RequestTime = rtt
+
+ if !found {
+ return nil, qm, nil
+ }
+
+ var out ACLBindingRule
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, qm, nil
+}
+
+// BindingRuleList retrieves a listing of all binding rules.
+func (a *ACL) BindingRuleList(methodName string, q *QueryOptions) ([]*ACLBindingRule, *QueryMeta, error) {
+ r := a.c.newRequest("GET", "/v1/acl/binding-rules")
+ if methodName != "" {
+ r.params.Set("authmethod", methodName)
+ }
+ r.setQueryOptions(q)
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ qm := &QueryMeta{}
+ parseQueryMeta(resp, qm)
+ qm.RequestTime = rtt
+
+ var entries []*ACLBindingRule
+ if err := decodeBody(resp, &entries); err != nil {
+ return nil, nil, err
+ }
+ return entries, qm, nil
+}
+
+// Login is used to exchange auth method credentials for a newly-minted Consul Token.
+func (a *ACL) Login(auth *ACLLoginParams, q *WriteOptions) (*ACLToken, *WriteMeta, error) {
+ r := a.c.newRequest("POST", "/v1/acl/login")
+ r.setWriteOptions(q)
+ r.obj = auth
+
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ var out ACLToken
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+ return &out, wm, nil
+}
+
+// Logout is used to destroy a Consul Token created via Login().
+func (a *ACL) Logout(q *WriteOptions) (*WriteMeta, error) {
+ r := a.c.newRequest("POST", "/v1/acl/logout")
+ r.setWriteOptions(q)
+ rtt, resp, err := requireOK(a.c.doRequest(r))
+ if err != nil {
+ return nil, err
+ }
+ resp.Body.Close()
+
+ wm := &WriteMeta{RequestTime: rtt}
+ return wm, nil
+}
diff --git a/vendor/github.com/hashicorp/consul/api/agent.go b/vendor/github.com/hashicorp/consul/api/agent.go
index 6acf8ad..1ef3312 100644
--- a/vendor/github.com/hashicorp/consul/api/agent.go
+++ b/vendor/github.com/hashicorp/consul/api/agent.go
@@ -2,7 +2,9 @@
import (
"bufio"
+ "bytes"
"fmt"
+ "io"
"net/http"
"net/url"
)
@@ -21,23 +23,11 @@
// service proxies another service within Consul and speaks the connect
// protocol.
ServiceKindConnectProxy ServiceKind = "connect-proxy"
-)
-// ProxyExecMode is the execution mode for a managed Connect proxy.
-type ProxyExecMode string
-
-const (
- // ProxyExecModeDaemon indicates that the proxy command should be long-running
- // and should be started and supervised by the agent until it's target service
- // is deregistered.
- ProxyExecModeDaemon ProxyExecMode = "daemon"
-
- // ProxyExecModeScript indicates that the proxy command should be invoke to
- // completion on each change to the configuration of lifecycle event. The
- // script typically fetches the config and certificates from the agent API and
- // then configures an externally managed daemon, perhaps starting and stopping
- // it if necessary.
- ProxyExecModeScript ProxyExecMode = "script"
+ // ServiceKindMeshGateway is a Mesh Gateway for the Connect feature. This
+ // service will proxy connections based off the SNI header set by other
+ // connect proxies
+ ServiceKindMeshGateway ServiceKind = "mesh-gateway"
)
// UpstreamDestType is the type of upstream discovery mechanism.
@@ -80,15 +70,14 @@
Meta map[string]string
Port int
Address string
+ TaggedAddresses map[string]ServiceAddress `json:",omitempty"`
Weights AgentWeights
EnableTagOverride bool
- CreateIndex uint64 `json:",omitempty"`
- ModifyIndex uint64 `json:",omitempty"`
- ContentHash string `json:",omitempty"`
- // DEPRECATED (ProxyDestination) - remove this field
- ProxyDestination string `json:",omitempty"`
- Proxy *AgentServiceConnectProxyConfig `json:",omitempty"`
- Connect *AgentServiceConnect `json:",omitempty"`
+ CreateIndex uint64 `json:",omitempty" bexpr:"-"`
+ ModifyIndex uint64 `json:",omitempty" bexpr:"-"`
+ ContentHash string `json:",omitempty" bexpr:"-"`
+ Proxy *AgentServiceConnectProxyConfig `json:",omitempty"`
+ Connect *AgentServiceConnect `json:",omitempty"`
}
// AgentServiceChecksInfo returns information about a Service and its checks
@@ -101,28 +90,19 @@
// AgentServiceConnect represents the Connect configuration of a service.
type AgentServiceConnect struct {
Native bool `json:",omitempty"`
- Proxy *AgentServiceConnectProxy `json:",omitempty"`
- SidecarService *AgentServiceRegistration `json:",omitempty"`
-}
-
-// AgentServiceConnectProxy represents the Connect Proxy configuration of a
-// service.
-type AgentServiceConnectProxy struct {
- ExecMode ProxyExecMode `json:",omitempty"`
- Command []string `json:",omitempty"`
- Config map[string]interface{} `json:",omitempty"`
- Upstreams []Upstream `json:",omitempty"`
+ SidecarService *AgentServiceRegistration `json:",omitempty" bexpr:"-"`
}
// AgentServiceConnectProxyConfig is the proxy configuration in a connect-proxy
// ServiceDefinition or response.
type AgentServiceConnectProxyConfig struct {
- DestinationServiceName string
+ DestinationServiceName string `json:",omitempty"`
DestinationServiceID string `json:",omitempty"`
LocalServiceAddress string `json:",omitempty"`
LocalServicePort int `json:",omitempty"`
- Config map[string]interface{} `json:",omitempty"`
- Upstreams []Upstream
+ Config map[string]interface{} `json:",omitempty" bexpr:"-"`
+ Upstreams []Upstream `json:",omitempty"`
+ MeshGateway MeshGatewayConfig `json:",omitempty"`
}
// AgentMember represents a cluster member known to the agent
@@ -155,21 +135,20 @@
// AgentServiceRegistration is used to register a new service
type AgentServiceRegistration struct {
- Kind ServiceKind `json:",omitempty"`
- ID string `json:",omitempty"`
- Name string `json:",omitempty"`
- Tags []string `json:",omitempty"`
- Port int `json:",omitempty"`
- Address string `json:",omitempty"`
- EnableTagOverride bool `json:",omitempty"`
- Meta map[string]string `json:",omitempty"`
- Weights *AgentWeights `json:",omitempty"`
+ Kind ServiceKind `json:",omitempty"`
+ ID string `json:",omitempty"`
+ Name string `json:",omitempty"`
+ Tags []string `json:",omitempty"`
+ Port int `json:",omitempty"`
+ Address string `json:",omitempty"`
+ TaggedAddresses map[string]ServiceAddress `json:",omitempty"`
+ EnableTagOverride bool `json:",omitempty"`
+ Meta map[string]string `json:",omitempty"`
+ Weights *AgentWeights `json:",omitempty"`
Check *AgentServiceCheck
Checks AgentServiceChecks
- // DEPRECATED (ProxyDestination) - remove this field
- ProxyDestination string `json:",omitempty"`
- Proxy *AgentServiceConnectProxyConfig `json:",omitempty"`
- Connect *AgentServiceConnect `json:",omitempty"`
+ Proxy *AgentServiceConnectProxyConfig `json:",omitempty"`
+ Connect *AgentServiceConnect `json:",omitempty"`
}
// AgentCheckRegistration is used to register a new check
@@ -274,12 +253,8 @@
TargetServiceID string
TargetServiceName string
ContentHash string
- // DEPRECATED(managed-proxies) - this struct is re-used for sidecar configs
- // but they don't need ExecMode or Command
- ExecMode ProxyExecMode `json:",omitempty"`
- Command []string `json:",omitempty"`
- Config map[string]interface{}
- Upstreams []Upstream
+ Config map[string]interface{} `bexpr:"-"`
+ Upstreams []Upstream
}
// Upstream is the response structure for a proxy upstream configuration.
@@ -290,7 +265,8 @@
Datacenter string `json:",omitempty"`
LocalBindAddress string `json:",omitempty"`
LocalBindPort int `json:",omitempty"`
- Config map[string]interface{} `json:",omitempty"`
+ Config map[string]interface{} `json:",omitempty" bexpr:"-"`
+ MeshGateway MeshGatewayConfig `json:",omitempty"`
}
// Agent can be used to query the Agent endpoints
@@ -385,7 +361,14 @@
// Checks returns the locally registered checks
func (a *Agent) Checks() (map[string]*AgentCheck, error) {
+ return a.ChecksWithFilter("")
+}
+
+// ChecksWithFilter returns a subset of the locally registered checks that match
+// the given filter expression
+func (a *Agent) ChecksWithFilter(filter string) (map[string]*AgentCheck, error) {
r := a.c.newRequest("GET", "/v1/agent/checks")
+ r.filterQuery(filter)
_, resp, err := requireOK(a.c.doRequest(r))
if err != nil {
return nil, err
@@ -401,7 +384,14 @@
// Services returns the locally registered services
func (a *Agent) Services() (map[string]*AgentService, error) {
+ return a.ServicesWithFilter("")
+}
+
+// ServicesWithFilter returns a subset of the locally registered services that match
+// the given filter expression
+func (a *Agent) ServicesWithFilter(filter string) (map[string]*AgentService, error) {
r := a.c.newRequest("GET", "/v1/agent/services")
+ r.filterQuery(filter)
_, resp, err := requireOK(a.c.doRequest(r))
if err != nil {
return nil, err
@@ -799,31 +789,6 @@
return &out, qm, nil
}
-// ConnectProxyConfig gets the configuration for a local managed proxy instance.
-//
-// Note that this uses an unconventional blocking mechanism since it's
-// agent-local state. That means there is no persistent raft index so we block
-// based on object hash instead.
-func (a *Agent) ConnectProxyConfig(proxyServiceID string, q *QueryOptions) (*ConnectProxyConfig, *QueryMeta, error) {
- r := a.c.newRequest("GET", "/v1/agent/connect/proxy/"+proxyServiceID)
- r.setQueryOptions(q)
- rtt, resp, err := requireOK(a.c.doRequest(r))
- if err != nil {
- return nil, nil, err
- }
- defer resp.Body.Close()
-
- qm := &QueryMeta{}
- parseQueryMeta(resp, qm)
- qm.RequestTime = rtt
-
- var out ConnectProxyConfig
- if err := decodeBody(resp, &out); err != nil {
- return nil, nil, err
- }
- return &out, qm, nil
-}
-
// EnableServiceMaintenance toggles service maintenance mode on
// for the given service ID.
func (a *Agent) EnableServiceMaintenance(serviceID, reason string) error {
@@ -1000,12 +965,20 @@
r := a.c.newRequest("PUT", fmt.Sprintf("/v1/agent/token/%s", target))
r.setWriteOptions(q)
r.obj = &AgentToken{Token: token}
- rtt, resp, err := requireOK(a.c.doRequest(r))
+
+ rtt, resp, err := a.c.doRequest(r)
if err != nil {
- return nil, resp.StatusCode, err
+ return nil, 0, err
}
- resp.Body.Close()
+ defer resp.Body.Close()
wm := &WriteMeta{RequestTime: rtt}
+
+ if resp.StatusCode != 200 {
+ var buf bytes.Buffer
+ io.Copy(&buf, resp.Body)
+ return wm, resp.StatusCode, fmt.Errorf("Unexpected response code: %d (%s)", resp.StatusCode, buf.Bytes())
+ }
+
return wm, resp.StatusCode, nil
}
diff --git a/vendor/github.com/hashicorp/consul/api/api.go b/vendor/github.com/hashicorp/consul/api/api.go
index 39a0ad3..b624b3c 100644
--- a/vendor/github.com/hashicorp/consul/api/api.go
+++ b/vendor/github.com/hashicorp/consul/api/api.go
@@ -30,6 +30,10 @@
// the HTTP token.
HTTPTokenEnvName = "CONSUL_HTTP_TOKEN"
+ // HTTPTokenFileEnvName defines an environment variable name which sets
+ // the HTTP token file.
+ HTTPTokenFileEnvName = "CONSUL_HTTP_TOKEN_FILE"
+
// HTTPAuthEnvName defines an environment variable name which sets
// the HTTP authentication header.
HTTPAuthEnvName = "CONSUL_HTTP_AUTH"
@@ -85,7 +89,7 @@
RequireConsistent bool
// UseCache requests that the agent cache results locally. See
- // https://www.consul.io/api/index.html#agent-caching for more details on the
+ // https://www.consul.io/api/features/caching.html for more details on the
// semantics.
UseCache bool
@@ -95,14 +99,14 @@
// returned. Clients that wish to allow for stale results on error can set
// StaleIfError to a longer duration to change this behavior. It is ignored
// if the endpoint supports background refresh caching. See
- // https://www.consul.io/api/index.html#agent-caching for more details.
+ // https://www.consul.io/api/features/caching.html for more details.
MaxAge time.Duration
// StaleIfError specifies how stale the client will accept a cached response
// if the servers are unavailable to fetch a fresh one. Only makes sense when
// UseCache is true and MaxAge is set to a lower, non-zero value. It is
// ignored if the endpoint supports background refresh caching. See
- // https://www.consul.io/api/index.html#agent-caching for more details.
+ // https://www.consul.io/api/features/caching.html for more details.
StaleIfError time.Duration
// WaitIndex is used to enable a blocking query. Waits
@@ -139,6 +143,10 @@
// a value from 0 to 5 (inclusive).
RelayFactor uint8
+ // LocalOnly is used in keyring list operation to force the keyring
+ // query to only hit local servers (no WAN traffic).
+ LocalOnly bool
+
// Connect filters prepared query execution to only include Connect-capable
// services. This currently affects prepared query execution.
Connect bool
@@ -146,6 +154,10 @@
// ctx is an optional context pass through to the underlying HTTP
// request layer. Use Context() and WithContext() to manage this.
ctx context.Context
+
+ // Filter requests filtering data prior to it being returned. The string
+ // is a go-bexpr compatible expression.
+ Filter string
}
func (o *QueryOptions) Context() context.Context {
@@ -276,6 +288,10 @@
// which overrides the agent's default token.
Token string
+ // TokenFile is a file containing the current token to use for this client.
+ // If provided it is read once at startup and never again.
+ TokenFile string
+
TLSConfig TLSConfig
}
@@ -339,6 +355,10 @@
config.Address = addr
}
+ if tokenFile := os.Getenv(HTTPTokenFileEnvName); tokenFile != "" {
+ config.TokenFile = tokenFile
+ }
+
if token := os.Getenv(HTTPTokenEnvName); token != "" {
config.Token = token
}
@@ -445,6 +465,7 @@
env = append(env,
fmt.Sprintf("%s=%s", HTTPAddrEnvName, c.Address),
fmt.Sprintf("%s=%s", HTTPTokenEnvName, c.Token),
+ fmt.Sprintf("%s=%s", HTTPTokenFileEnvName, c.TokenFile),
fmt.Sprintf("%s=%t", HTTPSSLEnvName, c.Scheme == "https"),
fmt.Sprintf("%s=%s", HTTPCAFile, c.TLSConfig.CAFile),
fmt.Sprintf("%s=%s", HTTPCAPath, c.TLSConfig.CAPath),
@@ -537,6 +558,19 @@
config.Address = parts[1]
}
+ // If the TokenFile is set, always use that, even if a Token is configured.
+ // This is because when TokenFile is set it is read into the Token field.
+ // We want any derived clients to have to re-read the token file.
+ if config.TokenFile != "" {
+ data, err := ioutil.ReadFile(config.TokenFile)
+ if err != nil {
+ return nil, fmt.Errorf("Error loading token file: %s", err)
+ }
+
+ if token := strings.TrimSpace(string(data)); token != "" {
+ config.Token = token
+ }
+ }
if config.Token == "" {
config.Token = defConfig.Token
}
@@ -614,6 +648,9 @@
if q.Near != "" {
r.params.Set("near", q.Near)
}
+ if q.Filter != "" {
+ r.params.Set("filter", q.Filter)
+ }
if len(q.NodeMeta) > 0 {
for key, value := range q.NodeMeta {
r.params.Add("node-meta", key+":"+value)
@@ -622,6 +659,9 @@
if q.RelayFactor != 0 {
r.params.Set("relay-factor", strconv.Itoa(int(q.RelayFactor)))
}
+ if q.LocalOnly {
+ r.params.Set("local-only", fmt.Sprintf("%t", q.LocalOnly))
+ }
if q.Connect {
r.params.Set("connect", "true")
}
@@ -813,6 +853,8 @@
}
// parseQueryMeta is used to help parse query meta-data
+//
+// TODO(rb): bug? the error from this function is never handled
func parseQueryMeta(resp *http.Response, q *QueryMeta) error {
header := resp.Header
@@ -890,10 +932,42 @@
return d, nil, e
}
if resp.StatusCode != 200 {
- var buf bytes.Buffer
- io.Copy(&buf, resp.Body)
- resp.Body.Close()
- return d, nil, fmt.Errorf("Unexpected response code: %d (%s)", resp.StatusCode, buf.Bytes())
+ return d, nil, generateUnexpectedResponseCodeError(resp)
}
return d, resp, nil
}
+
+func (req *request) filterQuery(filter string) {
+ if filter == "" {
+ return
+ }
+
+ req.params.Set("filter", filter)
+}
+
+// generateUnexpectedResponseCodeError consumes the rest of the body, closes
+// the body stream and generates an error indicating the status code was
+// unexpected.
+func generateUnexpectedResponseCodeError(resp *http.Response) error {
+ var buf bytes.Buffer
+ io.Copy(&buf, resp.Body)
+ resp.Body.Close()
+ return fmt.Errorf("Unexpected response code: %d (%s)", resp.StatusCode, buf.Bytes())
+}
+
+func requireNotFoundOrOK(d time.Duration, resp *http.Response, e error) (bool, time.Duration, *http.Response, error) {
+ if e != nil {
+ if resp != nil {
+ resp.Body.Close()
+ }
+ return false, d, nil, e
+ }
+ switch resp.StatusCode {
+ case 200:
+ return true, d, resp, nil
+ case 404:
+ return false, d, resp, nil
+ default:
+ return false, d, nil, generateUnexpectedResponseCodeError(resp)
+ }
+}
diff --git a/vendor/github.com/hashicorp/consul/api/catalog.go b/vendor/github.com/hashicorp/consul/api/catalog.go
index c175c3f..3fb0553 100644
--- a/vendor/github.com/hashicorp/consul/api/catalog.go
+++ b/vendor/github.com/hashicorp/consul/api/catalog.go
@@ -1,5 +1,10 @@
package api
+import (
+ "net"
+ "strconv"
+)
+
type Weights struct {
Passing int
Warning int
@@ -16,6 +21,11 @@
ModifyIndex uint64
}
+type ServiceAddress struct {
+ Address string
+ Port int
+}
+
type CatalogService struct {
ID string
Node string
@@ -26,17 +36,16 @@
ServiceID string
ServiceName string
ServiceAddress string
+ ServiceTaggedAddresses map[string]ServiceAddress
ServiceTags []string
ServiceMeta map[string]string
ServicePort int
ServiceWeights Weights
ServiceEnableTagOverride bool
- // DEPRECATED (ProxyDestination) - remove the next comment!
- // We forgot to ever add ServiceProxyDestination here so no need to deprecate!
- ServiceProxy *AgentServiceConnectProxyConfig
- CreateIndex uint64
- Checks HealthChecks
- ModifyIndex uint64
+ ServiceProxy *AgentServiceConnectProxyConfig
+ CreateIndex uint64
+ Checks HealthChecks
+ ModifyIndex uint64
}
type CatalogNode struct {
@@ -242,3 +251,12 @@
}
return out, qm, nil
}
+
+func ParseServiceAddr(addrPort string) (ServiceAddress, error) {
+ port := 0
+ host, portStr, err := net.SplitHostPort(addrPort)
+ if err == nil {
+ port, err = strconv.Atoi(portStr)
+ }
+ return ServiceAddress{Address: host, Port: port}, err
+}
diff --git a/vendor/github.com/hashicorp/consul/api/config_entry.go b/vendor/github.com/hashicorp/consul/api/config_entry.go
new file mode 100644
index 0000000..1588f2e
--- /dev/null
+++ b/vendor/github.com/hashicorp/consul/api/config_entry.go
@@ -0,0 +1,319 @@
+package api
+
+import (
+ "bytes"
+ "encoding/json"
+ "fmt"
+ "io"
+ "strconv"
+ "strings"
+
+ "github.com/mitchellh/mapstructure"
+)
+
+const (
+ ServiceDefaults string = "service-defaults"
+ ProxyDefaults string = "proxy-defaults"
+ ServiceRouter string = "service-router"
+ ServiceSplitter string = "service-splitter"
+ ServiceResolver string = "service-resolver"
+
+ ProxyConfigGlobal string = "global"
+)
+
+type ConfigEntry interface {
+ GetKind() string
+ GetName() string
+ GetCreateIndex() uint64
+ GetModifyIndex() uint64
+}
+
+type MeshGatewayMode string
+
+const (
+ // MeshGatewayModeDefault represents no specific mode and should
+ // be used to indicate that a different layer of the configuration
+ // chain should take precedence
+ MeshGatewayModeDefault MeshGatewayMode = ""
+
+ // MeshGatewayModeNone represents that the Upstream Connect connections
+ // should be direct and not flow through a mesh gateway.
+ MeshGatewayModeNone MeshGatewayMode = "none"
+
+ // MeshGatewayModeLocal represents that the Upstrea Connect connections
+ // should be made to a mesh gateway in the local datacenter. This is
+ MeshGatewayModeLocal MeshGatewayMode = "local"
+
+ // MeshGatewayModeRemote represents that the Upstream Connect connections
+ // should be made to a mesh gateway in a remote datacenter.
+ MeshGatewayModeRemote MeshGatewayMode = "remote"
+)
+
+// MeshGatewayConfig controls how Mesh Gateways are used for upstream Connect
+// services
+type MeshGatewayConfig struct {
+ // Mode is the mode that should be used for the upstream connection.
+ Mode MeshGatewayMode `json:",omitempty"`
+}
+
+type ServiceConfigEntry struct {
+ Kind string
+ Name string
+ Protocol string `json:",omitempty"`
+ MeshGateway MeshGatewayConfig `json:",omitempty"`
+ ExternalSNI string `json:",omitempty"`
+ CreateIndex uint64
+ ModifyIndex uint64
+}
+
+func (s *ServiceConfigEntry) GetKind() string {
+ return s.Kind
+}
+
+func (s *ServiceConfigEntry) GetName() string {
+ return s.Name
+}
+
+func (s *ServiceConfigEntry) GetCreateIndex() uint64 {
+ return s.CreateIndex
+}
+
+func (s *ServiceConfigEntry) GetModifyIndex() uint64 {
+ return s.ModifyIndex
+}
+
+type ProxyConfigEntry struct {
+ Kind string
+ Name string
+ Config map[string]interface{} `json:",omitempty"`
+ MeshGateway MeshGatewayConfig `json:",omitempty"`
+ CreateIndex uint64
+ ModifyIndex uint64
+}
+
+func (p *ProxyConfigEntry) GetKind() string {
+ return p.Kind
+}
+
+func (p *ProxyConfigEntry) GetName() string {
+ return p.Name
+}
+
+func (p *ProxyConfigEntry) GetCreateIndex() uint64 {
+ return p.CreateIndex
+}
+
+func (p *ProxyConfigEntry) GetModifyIndex() uint64 {
+ return p.ModifyIndex
+}
+
+type rawEntryListResponse struct {
+ kind string
+ Entries []map[string]interface{}
+}
+
+func makeConfigEntry(kind, name string) (ConfigEntry, error) {
+ switch kind {
+ case ServiceDefaults:
+ return &ServiceConfigEntry{Kind: kind, Name: name}, nil
+ case ProxyDefaults:
+ return &ProxyConfigEntry{Kind: kind, Name: name}, nil
+ case ServiceRouter:
+ return &ServiceRouterConfigEntry{Kind: kind, Name: name}, nil
+ case ServiceSplitter:
+ return &ServiceSplitterConfigEntry{Kind: kind, Name: name}, nil
+ case ServiceResolver:
+ return &ServiceResolverConfigEntry{Kind: kind, Name: name}, nil
+ default:
+ return nil, fmt.Errorf("invalid config entry kind: %s", kind)
+ }
+}
+
+func MakeConfigEntry(kind, name string) (ConfigEntry, error) {
+ return makeConfigEntry(kind, name)
+}
+
+// DecodeConfigEntry will decode the result of using json.Unmarshal of a config
+// entry into a map[string]interface{}.
+//
+// Important caveats:
+//
+// - This will NOT work if the map[string]interface{} was produced using HCL
+// decoding as that requires more extensive parsing to work around the issues
+// with map[string][]interface{} that arise.
+//
+// - This will only decode fields using their camel case json field
+// representations.
+func DecodeConfigEntry(raw map[string]interface{}) (ConfigEntry, error) {
+ var entry ConfigEntry
+
+ kindVal, ok := raw["Kind"]
+ if !ok {
+ kindVal, ok = raw["kind"]
+ }
+ if !ok {
+ return nil, fmt.Errorf("Payload does not contain a kind/Kind key at the top level")
+ }
+
+ if kindStr, ok := kindVal.(string); ok {
+ newEntry, err := makeConfigEntry(kindStr, "")
+ if err != nil {
+ return nil, err
+ }
+ entry = newEntry
+ } else {
+ return nil, fmt.Errorf("Kind value in payload is not a string")
+ }
+
+ decodeConf := &mapstructure.DecoderConfig{
+ DecodeHook: mapstructure.StringToTimeDurationHookFunc(),
+ Result: &entry,
+ WeaklyTypedInput: true,
+ }
+
+ decoder, err := mapstructure.NewDecoder(decodeConf)
+ if err != nil {
+ return nil, err
+ }
+
+ return entry, decoder.Decode(raw)
+}
+
+func DecodeConfigEntryFromJSON(data []byte) (ConfigEntry, error) {
+ var raw map[string]interface{}
+ if err := json.Unmarshal(data, &raw); err != nil {
+ return nil, err
+ }
+
+ return DecodeConfigEntry(raw)
+}
+
+func decodeConfigEntrySlice(raw []map[string]interface{}) ([]ConfigEntry, error) {
+ var entries []ConfigEntry
+ for _, rawEntry := range raw {
+ entry, err := DecodeConfigEntry(rawEntry)
+ if err != nil {
+ return nil, err
+ }
+ entries = append(entries, entry)
+ }
+ return entries, nil
+}
+
+// ConfigEntries can be used to query the Config endpoints
+type ConfigEntries struct {
+ c *Client
+}
+
+// Config returns a handle to the Config endpoints
+func (c *Client) ConfigEntries() *ConfigEntries {
+ return &ConfigEntries{c}
+}
+
+func (conf *ConfigEntries) Get(kind string, name string, q *QueryOptions) (ConfigEntry, *QueryMeta, error) {
+ if kind == "" || name == "" {
+ return nil, nil, fmt.Errorf("Both kind and name parameters must not be empty")
+ }
+
+ entry, err := makeConfigEntry(kind, name)
+ if err != nil {
+ return nil, nil, err
+ }
+
+ r := conf.c.newRequest("GET", fmt.Sprintf("/v1/config/%s/%s", kind, name))
+ r.setQueryOptions(q)
+ rtt, resp, err := requireOK(conf.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+
+ defer resp.Body.Close()
+
+ qm := &QueryMeta{}
+ parseQueryMeta(resp, qm)
+ qm.RequestTime = rtt
+
+ if err := decodeBody(resp, entry); err != nil {
+ return nil, nil, err
+ }
+
+ return entry, qm, nil
+}
+
+func (conf *ConfigEntries) List(kind string, q *QueryOptions) ([]ConfigEntry, *QueryMeta, error) {
+ if kind == "" {
+ return nil, nil, fmt.Errorf("The kind parameter must not be empty")
+ }
+
+ r := conf.c.newRequest("GET", fmt.Sprintf("/v1/config/%s", kind))
+ r.setQueryOptions(q)
+ rtt, resp, err := requireOK(conf.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+
+ defer resp.Body.Close()
+
+ qm := &QueryMeta{}
+ parseQueryMeta(resp, qm)
+ qm.RequestTime = rtt
+
+ var raw []map[string]interface{}
+ if err := decodeBody(resp, &raw); err != nil {
+ return nil, nil, err
+ }
+
+ entries, err := decodeConfigEntrySlice(raw)
+ if err != nil {
+ return nil, nil, err
+ }
+
+ return entries, qm, nil
+}
+
+func (conf *ConfigEntries) Set(entry ConfigEntry, w *WriteOptions) (bool, *WriteMeta, error) {
+ return conf.set(entry, nil, w)
+}
+
+func (conf *ConfigEntries) CAS(entry ConfigEntry, index uint64, w *WriteOptions) (bool, *WriteMeta, error) {
+ return conf.set(entry, map[string]string{"cas": strconv.FormatUint(index, 10)}, w)
+}
+
+func (conf *ConfigEntries) set(entry ConfigEntry, params map[string]string, w *WriteOptions) (bool, *WriteMeta, error) {
+ r := conf.c.newRequest("PUT", "/v1/config")
+ r.setWriteOptions(w)
+ for param, value := range params {
+ r.params.Set(param, value)
+ }
+ r.obj = entry
+ rtt, resp, err := requireOK(conf.c.doRequest(r))
+ if err != nil {
+ return false, nil, err
+ }
+ defer resp.Body.Close()
+
+ var buf bytes.Buffer
+ if _, err := io.Copy(&buf, resp.Body); err != nil {
+ return false, nil, fmt.Errorf("Failed to read response: %v", err)
+ }
+ res := strings.Contains(buf.String(), "true")
+
+ wm := &WriteMeta{RequestTime: rtt}
+ return res, wm, nil
+}
+
+func (conf *ConfigEntries) Delete(kind string, name string, w *WriteOptions) (*WriteMeta, error) {
+ if kind == "" || name == "" {
+ return nil, fmt.Errorf("Both kind and name parameters must not be empty")
+ }
+
+ r := conf.c.newRequest("DELETE", fmt.Sprintf("/v1/config/%s/%s", kind, name))
+ r.setWriteOptions(w)
+ rtt, resp, err := requireOK(conf.c.doRequest(r))
+ if err != nil {
+ return nil, err
+ }
+ resp.Body.Close()
+ wm := &WriteMeta{RequestTime: rtt}
+ return wm, nil
+}
diff --git a/vendor/github.com/hashicorp/consul/api/config_entry_discoverychain.go b/vendor/github.com/hashicorp/consul/api/config_entry_discoverychain.go
new file mode 100644
index 0000000..77acfbd
--- /dev/null
+++ b/vendor/github.com/hashicorp/consul/api/config_entry_discoverychain.go
@@ -0,0 +1,200 @@
+package api
+
+import (
+ "encoding/json"
+ "time"
+)
+
+type ServiceRouterConfigEntry struct {
+ Kind string
+ Name string
+
+ Routes []ServiceRoute `json:",omitempty"`
+
+ CreateIndex uint64
+ ModifyIndex uint64
+}
+
+func (e *ServiceRouterConfigEntry) GetKind() string { return e.Kind }
+func (e *ServiceRouterConfigEntry) GetName() string { return e.Name }
+func (e *ServiceRouterConfigEntry) GetCreateIndex() uint64 { return e.CreateIndex }
+func (e *ServiceRouterConfigEntry) GetModifyIndex() uint64 { return e.ModifyIndex }
+
+type ServiceRoute struct {
+ Match *ServiceRouteMatch `json:",omitempty"`
+ Destination *ServiceRouteDestination `json:",omitempty"`
+}
+
+type ServiceRouteMatch struct {
+ HTTP *ServiceRouteHTTPMatch `json:",omitempty"`
+}
+
+type ServiceRouteHTTPMatch struct {
+ PathExact string `json:",omitempty"`
+ PathPrefix string `json:",omitempty"`
+ PathRegex string `json:",omitempty"`
+
+ Header []ServiceRouteHTTPMatchHeader `json:",omitempty"`
+ QueryParam []ServiceRouteHTTPMatchQueryParam `json:",omitempty"`
+ Methods []string `json:",omitempty"`
+}
+
+type ServiceRouteHTTPMatchHeader struct {
+ Name string
+ Present bool `json:",omitempty"`
+ Exact string `json:",omitempty"`
+ Prefix string `json:",omitempty"`
+ Suffix string `json:",omitempty"`
+ Regex string `json:",omitempty"`
+ Invert bool `json:",omitempty"`
+}
+
+type ServiceRouteHTTPMatchQueryParam struct {
+ Name string
+ Present bool `json:",omitempty"`
+ Exact string `json:",omitempty"`
+ Regex string `json:",omitempty"`
+}
+
+type ServiceRouteDestination struct {
+ Service string `json:",omitempty"`
+ ServiceSubset string `json:",omitempty"`
+ Namespace string `json:",omitempty"`
+ PrefixRewrite string `json:",omitempty"`
+ RequestTimeout time.Duration `json:",omitempty"`
+ NumRetries uint32 `json:",omitempty"`
+ RetryOnConnectFailure bool `json:",omitempty"`
+ RetryOnStatusCodes []uint32 `json:",omitempty"`
+}
+
+func (e *ServiceRouteDestination) MarshalJSON() ([]byte, error) {
+ type Alias ServiceRouteDestination
+ exported := &struct {
+ RequestTimeout string `json:",omitempty"`
+ *Alias
+ }{
+ RequestTimeout: e.RequestTimeout.String(),
+ Alias: (*Alias)(e),
+ }
+ if e.RequestTimeout == 0 {
+ exported.RequestTimeout = ""
+ }
+
+ return json.Marshal(exported)
+}
+
+func (e *ServiceRouteDestination) UnmarshalJSON(data []byte) error {
+ type Alias ServiceRouteDestination
+ aux := &struct {
+ RequestTimeout string
+ *Alias
+ }{
+ Alias: (*Alias)(e),
+ }
+ if err := json.Unmarshal(data, &aux); err != nil {
+ return err
+ }
+ var err error
+ if aux.RequestTimeout != "" {
+ if e.RequestTimeout, err = time.ParseDuration(aux.RequestTimeout); err != nil {
+ return err
+ }
+ }
+ return nil
+}
+
+type ServiceSplitterConfigEntry struct {
+ Kind string
+ Name string
+
+ Splits []ServiceSplit `json:",omitempty"`
+
+ CreateIndex uint64
+ ModifyIndex uint64
+}
+
+func (e *ServiceSplitterConfigEntry) GetKind() string { return e.Kind }
+func (e *ServiceSplitterConfigEntry) GetName() string { return e.Name }
+func (e *ServiceSplitterConfigEntry) GetCreateIndex() uint64 { return e.CreateIndex }
+func (e *ServiceSplitterConfigEntry) GetModifyIndex() uint64 { return e.ModifyIndex }
+
+type ServiceSplit struct {
+ Weight float32
+ Service string `json:",omitempty"`
+ ServiceSubset string `json:",omitempty"`
+ Namespace string `json:",omitempty"`
+}
+
+type ServiceResolverConfigEntry struct {
+ Kind string
+ Name string
+
+ DefaultSubset string `json:",omitempty"`
+ Subsets map[string]ServiceResolverSubset `json:",omitempty"`
+ Redirect *ServiceResolverRedirect `json:",omitempty"`
+ Failover map[string]ServiceResolverFailover `json:",omitempty"`
+ ConnectTimeout time.Duration `json:",omitempty"`
+
+ CreateIndex uint64
+ ModifyIndex uint64
+}
+
+func (e *ServiceResolverConfigEntry) MarshalJSON() ([]byte, error) {
+ type Alias ServiceResolverConfigEntry
+ exported := &struct {
+ ConnectTimeout string `json:",omitempty"`
+ *Alias
+ }{
+ ConnectTimeout: e.ConnectTimeout.String(),
+ Alias: (*Alias)(e),
+ }
+ if e.ConnectTimeout == 0 {
+ exported.ConnectTimeout = ""
+ }
+
+ return json.Marshal(exported)
+}
+
+func (e *ServiceResolverConfigEntry) UnmarshalJSON(data []byte) error {
+ type Alias ServiceResolverConfigEntry
+ aux := &struct {
+ ConnectTimeout string
+ *Alias
+ }{
+ Alias: (*Alias)(e),
+ }
+ if err := json.Unmarshal(data, &aux); err != nil {
+ return err
+ }
+ var err error
+ if aux.ConnectTimeout != "" {
+ if e.ConnectTimeout, err = time.ParseDuration(aux.ConnectTimeout); err != nil {
+ return err
+ }
+ }
+ return nil
+}
+
+func (e *ServiceResolverConfigEntry) GetKind() string { return e.Kind }
+func (e *ServiceResolverConfigEntry) GetName() string { return e.Name }
+func (e *ServiceResolverConfigEntry) GetCreateIndex() uint64 { return e.CreateIndex }
+func (e *ServiceResolverConfigEntry) GetModifyIndex() uint64 { return e.ModifyIndex }
+
+type ServiceResolverSubset struct {
+ Filter string `json:",omitempty"`
+ OnlyPassing bool `json:",omitempty"`
+}
+
+type ServiceResolverRedirect struct {
+ Service string `json:",omitempty"`
+ ServiceSubset string `json:",omitempty"`
+ Namespace string `json:",omitempty"`
+ Datacenter string `json:",omitempty"`
+}
+
+type ServiceResolverFailover struct {
+ Service string `json:",omitempty"`
+ ServiceSubset string `json:",omitempty"`
+ Namespace string `json:",omitempty"`
+ Datacenters []string `json:",omitempty"`
+}
diff --git a/vendor/github.com/hashicorp/consul/api/connect_intention.go b/vendor/github.com/hashicorp/consul/api/connect_intention.go
index a996c03..d25cb84 100644
--- a/vendor/github.com/hashicorp/consul/api/connect_intention.go
+++ b/vendor/github.com/hashicorp/consul/api/connect_intention.go
@@ -54,6 +54,13 @@
// or modified.
CreatedAt, UpdatedAt time.Time
+ // Hash of the contents of the intention
+ //
+ // This is needed mainly for replication purposes. When replicating from
+ // one DC to another keeping the content Hash will allow us to detect
+ // content changes more efficiently than checking every single field
+ Hash []byte
+
CreateIndex uint64
ModifyIndex uint64
}
diff --git a/vendor/github.com/hashicorp/consul/api/discovery_chain.go b/vendor/github.com/hashicorp/consul/api/discovery_chain.go
new file mode 100644
index 0000000..407a3b0
--- /dev/null
+++ b/vendor/github.com/hashicorp/consul/api/discovery_chain.go
@@ -0,0 +1,230 @@
+package api
+
+import (
+ "encoding/json"
+ "fmt"
+ "time"
+)
+
+// DiscoveryChain can be used to query the discovery-chain endpoints
+type DiscoveryChain struct {
+ c *Client
+}
+
+// DiscoveryChain returns a handle to the discovery-chain endpoints
+func (c *Client) DiscoveryChain() *DiscoveryChain {
+ return &DiscoveryChain{c}
+}
+
+func (d *DiscoveryChain) Get(name string, opts *DiscoveryChainOptions, q *QueryOptions) (*DiscoveryChainResponse, *QueryMeta, error) {
+ if name == "" {
+ return nil, nil, fmt.Errorf("Name parameter must not be empty")
+ }
+
+ method := "GET"
+ if opts != nil && opts.requiresPOST() {
+ method = "POST"
+ }
+
+ r := d.c.newRequest(method, fmt.Sprintf("/v1/discovery-chain/%s", name))
+ r.setQueryOptions(q)
+
+ if opts != nil {
+ if opts.EvaluateInDatacenter != "" {
+ r.params.Set("compile-dc", opts.EvaluateInDatacenter)
+ }
+ // TODO(namespaces): handle possible EvaluateInNamespace here
+ }
+
+ if method == "POST" {
+ r.obj = opts
+ }
+
+ rtt, resp, err := requireOK(d.c.doRequest(r))
+ if err != nil {
+ return nil, nil, err
+ }
+ defer resp.Body.Close()
+
+ qm := &QueryMeta{}
+ parseQueryMeta(resp, qm)
+ qm.RequestTime = rtt
+
+ var out DiscoveryChainResponse
+
+ if err := decodeBody(resp, &out); err != nil {
+ return nil, nil, err
+ }
+
+ return &out, qm, nil
+}
+
+type DiscoveryChainOptions struct {
+ EvaluateInDatacenter string `json:"-"`
+
+ // OverrideMeshGateway allows for the mesh gateway setting to be overridden
+ // for any resolver in the compiled chain.
+ OverrideMeshGateway MeshGatewayConfig `json:",omitempty"`
+
+ // OverrideProtocol allows for the final protocol for the chain to be
+ // altered.
+ //
+ // - If the chain ordinarily would be TCP and an L7 protocol is passed here
+ // the chain will not include Routers or Splitters.
+ //
+ // - If the chain ordinarily would be L7 and TCP is passed here the chain
+ // will not include Routers or Splitters.
+ OverrideProtocol string `json:",omitempty"`
+
+ // OverrideConnectTimeout allows for the ConnectTimeout setting to be
+ // overridden for any resolver in the compiled chain.
+ OverrideConnectTimeout time.Duration `json:",omitempty"`
+}
+
+func (o *DiscoveryChainOptions) requiresPOST() bool {
+ if o == nil {
+ return false
+ }
+ return o.OverrideMeshGateway.Mode != "" ||
+ o.OverrideProtocol != "" ||
+ o.OverrideConnectTimeout != 0
+}
+
+type DiscoveryChainResponse struct {
+ Chain *CompiledDiscoveryChain
+}
+
+type CompiledDiscoveryChain struct {
+ ServiceName string
+ Namespace string
+ Datacenter string
+
+ // CustomizationHash is a unique hash of any data that affects the
+ // compilation of the discovery chain other than config entries or the
+ // name/namespace/datacenter evaluation criteria.
+ //
+ // If set, this value should be used to prefix/suffix any generated load
+ // balancer data plane objects to avoid sharing customized and
+ // non-customized versions.
+ CustomizationHash string
+
+ // Protocol is the overall protocol shared by everything in the chain.
+ Protocol string
+
+ // StartNode is the first key into the Nodes map that should be followed
+ // when walking the discovery chain.
+ StartNode string
+
+ // Nodes contains all nodes available for traversal in the chain keyed by a
+ // unique name. You can walk this by starting with StartNode.
+ //
+ // NOTE: The names should be treated as opaque values and are only
+ // guaranteed to be consistent within a single compilation.
+ Nodes map[string]*DiscoveryGraphNode
+
+ // Targets is a list of all targets used in this chain.
+ //
+ // NOTE: The names should be treated as opaque values and are only
+ // guaranteed to be consistent within a single compilation.
+ Targets map[string]*DiscoveryTarget
+}
+
+const (
+ DiscoveryGraphNodeTypeRouter = "router"
+ DiscoveryGraphNodeTypeSplitter = "splitter"
+ DiscoveryGraphNodeTypeResolver = "resolver"
+)
+
+// DiscoveryGraphNode is a single node in the compiled discovery chain.
+type DiscoveryGraphNode struct {
+ Type string
+ Name string // this is NOT necessarily a service
+
+ // fields for Type==router
+ Routes []*DiscoveryRoute
+
+ // fields for Type==splitter
+ Splits []*DiscoverySplit
+
+ // fields for Type==resolver
+ Resolver *DiscoveryResolver
+}
+
+// compiled form of ServiceRoute
+type DiscoveryRoute struct {
+ Definition *ServiceRoute
+ NextNode string
+}
+
+// compiled form of ServiceSplit
+type DiscoverySplit struct {
+ Weight float32
+ NextNode string
+}
+
+// compiled form of ServiceResolverConfigEntry
+type DiscoveryResolver struct {
+ Default bool
+ ConnectTimeout time.Duration
+ Target string
+ Failover *DiscoveryFailover
+}
+
+func (r *DiscoveryResolver) MarshalJSON() ([]byte, error) {
+ type Alias DiscoveryResolver
+ exported := &struct {
+ ConnectTimeout string `json:",omitempty"`
+ *Alias
+ }{
+ ConnectTimeout: r.ConnectTimeout.String(),
+ Alias: (*Alias)(r),
+ }
+ if r.ConnectTimeout == 0 {
+ exported.ConnectTimeout = ""
+ }
+
+ return json.Marshal(exported)
+}
+
+func (r *DiscoveryResolver) UnmarshalJSON(data []byte) error {
+ type Alias DiscoveryResolver
+ aux := &struct {
+ ConnectTimeout string
+ *Alias
+ }{
+ Alias: (*Alias)(r),
+ }
+ if err := json.Unmarshal(data, &aux); err != nil {
+ return err
+ }
+ var err error
+ if aux.ConnectTimeout != "" {
+ if r.ConnectTimeout, err = time.ParseDuration(aux.ConnectTimeout); err != nil {
+ return err
+ }
+ }
+ return nil
+}
+
+// compiled form of ServiceResolverFailover
+type DiscoveryFailover struct {
+ Targets []string
+}
+
+// DiscoveryTarget represents all of the inputs necessary to use a resolver
+// config entry to execute a catalog query to generate a list of service
+// instances during discovery.
+type DiscoveryTarget struct {
+ ID string
+
+ Service string
+ ServiceSubset string
+ Namespace string
+ Datacenter string
+
+ MeshGateway MeshGatewayConfig
+ Subset ServiceResolverSubset
+ External bool
+ SNI string
+ Name string
+}
diff --git a/vendor/github.com/hashicorp/consul/api/go.mod b/vendor/github.com/hashicorp/consul/api/go.mod
new file mode 100644
index 0000000..78fe8a3
--- /dev/null
+++ b/vendor/github.com/hashicorp/consul/api/go.mod
@@ -0,0 +1,16 @@
+module github.com/hashicorp/consul/api
+
+go 1.12
+
+replace github.com/hashicorp/consul/sdk => ../sdk
+
+require (
+ github.com/hashicorp/consul/sdk v0.2.0
+ github.com/hashicorp/go-cleanhttp v0.5.1
+ github.com/hashicorp/go-rootcerts v1.0.0
+ github.com/hashicorp/go-uuid v1.0.1
+ github.com/hashicorp/serf v0.8.2
+ github.com/mitchellh/mapstructure v1.1.2
+ github.com/pascaldekloe/goe v0.0.0-20180627143212-57f6aae5913c
+ github.com/stretchr/testify v1.3.0
+)
diff --git a/vendor/github.com/hashicorp/consul/api/go.sum b/vendor/github.com/hashicorp/consul/api/go.sum
new file mode 100644
index 0000000..01591f9
--- /dev/null
+++ b/vendor/github.com/hashicorp/consul/api/go.sum
@@ -0,0 +1,78 @@
+github.com/armon/circbuf v0.0.0-20150827004946-bbbad097214e/go.mod h1:3U/XgcO3hCbHZ8TKRvWD2dDTCfh9M9ya+I9JpbB7O8o=
+github.com/armon/go-metrics v0.0.0-20180917152333-f0300d1749da h1:8GUt8eRujhVEGZFFEjBj46YV4rDjvGrNxb0KMWYkL2I=
+github.com/armon/go-metrics v0.0.0-20180917152333-f0300d1749da/go.mod h1:Q73ZrmVTwzkszR9V5SSuryQ31EELlFMUz1kKyl939pY=
+github.com/armon/go-radix v0.0.0-20180808171621-7fddfc383310/go.mod h1:ufUuZ+zHj4x4TnLV4JWEpy2hxWSpsRywHrMgIH9cCH8=
+github.com/bgentry/speakeasy v0.1.0/go.mod h1:+zsyZBPWlz7T6j88CTgSN5bM796AkVf0kBD4zp0CCIs=
+github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
+github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/fatih/color v1.7.0/go.mod h1:Zm6kSWBoL9eyXnKyktHP6abPY2pDugNf5KwzbycvMj4=
+github.com/google/btree v0.0.0-20180813153112-4030bb1f1f0c h1:964Od4U6p2jUkFxvCydnIczKteheJEzHRToSGK3Bnlw=
+github.com/google/btree v0.0.0-20180813153112-4030bb1f1f0c/go.mod h1:lNA+9X1NB3Zf8V7Ke586lFgjr2dZNuvo3lPJSGZ5JPQ=
+github.com/hashicorp/consul/sdk v0.2.0 h1:GWFYFmry/k4b1hEoy7kSkmU8e30GAyI4VZHk0fRxeL4=
+github.com/hashicorp/consul/sdk v0.2.0/go.mod h1:VKf9jXwCTEY1QZP2MOLRhb5i/I/ssyNV1vwHyQBF0x8=
+github.com/hashicorp/errwrap v1.0.0 h1:hLrqtEDnRye3+sgx6z4qVLNuviH3MR5aQ0ykNJa/UYA=
+github.com/hashicorp/errwrap v1.0.0/go.mod h1:YH+1FKiLXxHSkmPseP+kNlulaMuP3n2brvKWEqk/Jc4=
+github.com/hashicorp/go-cleanhttp v0.5.1 h1:dH3aiDG9Jvb5r5+bYHsikaOUIpcM0xvgMXVoDkXMzJM=
+github.com/hashicorp/go-cleanhttp v0.5.1/go.mod h1:JpRdi6/HCYpAwUzNwuwqhbovhLtngrth3wmdIIUrZ80=
+github.com/hashicorp/go-immutable-radix v1.0.0 h1:AKDB1HM5PWEA7i4nhcpwOrO2byshxBjXVn/J/3+z5/0=
+github.com/hashicorp/go-immutable-radix v1.0.0/go.mod h1:0y9vanUI8NX6FsYoO3zeMjhV/C5i9g4Q3DwcSNZ4P60=
+github.com/hashicorp/go-msgpack v0.5.3 h1:zKjpN5BK/P5lMYrLmBHdBULWbJ0XpYR+7NGzqkZzoD4=
+github.com/hashicorp/go-msgpack v0.5.3/go.mod h1:ahLV/dePpqEmjfWmKiqvPkv/twdG7iPBM1vqhUKIvfM=
+github.com/hashicorp/go-multierror v1.0.0 h1:iVjPR7a6H0tWELX5NxNe7bYopibicUzc7uPribsnS6o=
+github.com/hashicorp/go-multierror v1.0.0/go.mod h1:dHtQlpGsu+cZNNAkkCN/P3hoUDHhCYQXV3UM06sGGrk=
+github.com/hashicorp/go-rootcerts v1.0.0 h1:Rqb66Oo1X/eSV1x66xbDccZjhJigjg0+e82kpwzSwCI=
+github.com/hashicorp/go-rootcerts v1.0.0/go.mod h1:K6zTfqpRlCUIjkwsN4Z+hiSfzSTQa6eBIzfwKfwNnHU=
+github.com/hashicorp/go-sockaddr v1.0.0 h1:GeH6tui99pF4NJgfnhp+L6+FfobzVW3Ah46sLo0ICXs=
+github.com/hashicorp/go-sockaddr v1.0.0/go.mod h1:7Xibr9yA9JjQq1JpNB2Vw7kxv8xerXegt+ozgdvDeDU=
+github.com/hashicorp/go-syslog v1.0.0/go.mod h1:qPfqrKkXGihmCqbJM2mZgkZGvKG1dFdvsLplgctolz4=
+github.com/hashicorp/go-uuid v1.0.0/go.mod h1:6SBZvOh/SIDV7/2o3Jml5SYk/TvGqwFJ/bN7x4byOro=
+github.com/hashicorp/go-uuid v1.0.1 h1:fv1ep09latC32wFoVwnqcnKJGnMSdBanPczbHAYm1BE=
+github.com/hashicorp/go-uuid v1.0.1/go.mod h1:6SBZvOh/SIDV7/2o3Jml5SYk/TvGqwFJ/bN7x4byOro=
+github.com/hashicorp/go.net v0.0.1/go.mod h1:hjKkEWcCURg++eb33jQU7oqQcI9XDCnUzHA0oac0k90=
+github.com/hashicorp/golang-lru v0.5.0 h1:CL2msUPvZTLb5O648aiLNJw3hnBxN2+1Jq8rCOH9wdo=
+github.com/hashicorp/golang-lru v0.5.0/go.mod h1:/m3WP610KZHVQ1SGc6re/UDhFvYD7pJ4Ao+sR/qLZy8=
+github.com/hashicorp/logutils v1.0.0/go.mod h1:QIAnNjmIWmVIIkWDTG1z5v++HQmx9WQRO+LraFDTW64=
+github.com/hashicorp/mdns v1.0.0/go.mod h1:tL+uN++7HEJ6SQLQ2/p+z2pH24WQKWjBPkE0mNTz8vQ=
+github.com/hashicorp/memberlist v0.1.3 h1:EmmoJme1matNzb+hMpDuR/0sbJSUisxyqBGG676r31M=
+github.com/hashicorp/memberlist v0.1.3/go.mod h1:ajVTdAv/9Im8oMAAj5G31PhhMCZJV2pPBoIllUwCN7I=
+github.com/hashicorp/serf v0.8.2 h1:YZ7UKsJv+hKjqGVUUbtE3HNj79Eln2oQ75tniF6iPt0=
+github.com/hashicorp/serf v0.8.2/go.mod h1:6hOLApaqBFA1NXqRQAsxw9QxuDEvNxSQRwA/JwenrHc=
+github.com/mattn/go-colorable v0.0.9/go.mod h1:9vuHe8Xs5qXnSaW/c/ABM9alt+Vo+STaOChaDxuIBZU=
+github.com/mattn/go-isatty v0.0.3/go.mod h1:M+lRXTBqGeGNdLjl/ufCoiOlB5xdOkqRJdNxMWT7Zi4=
+github.com/miekg/dns v1.0.14 h1:9jZdLNd/P4+SfEJ0TNyxYpsK8N4GtfylBLqtbYN1sbA=
+github.com/miekg/dns v1.0.14/go.mod h1:W1PPwlIAgtquWBMBEV9nkV9Cazfe8ScdGz/Lj7v3Nrg=
+github.com/mitchellh/cli v1.0.0/go.mod h1:hNIlj7HEI86fIcpObd7a0FcrxTWetlwJDGcceTlRvqc=
+github.com/mitchellh/go-homedir v1.0.0 h1:vKb8ShqSby24Yrqr/yDYkuFz8d0WUjys40rvnGC8aR0=
+github.com/mitchellh/go-homedir v1.0.0/go.mod h1:SfyaCUpYCn1Vlf4IUYiD9fPX4A5wJrkLzIz1N1q0pr0=
+github.com/mitchellh/go-testing-interface v1.0.0 h1:fzU/JVNcaqHQEcVFAKeR41fkiLdIPrefOvVG1VZ96U0=
+github.com/mitchellh/go-testing-interface v1.0.0/go.mod h1:kRemZodwjscx+RGhAo8eIhFbs2+BFgRtFPeD/KE+zxI=
+github.com/mitchellh/gox v0.4.0/go.mod h1:Sd9lOJ0+aimLBi73mGofS1ycjY8lL3uZM3JPS42BGNg=
+github.com/mitchellh/iochan v1.0.0/go.mod h1:JwYml1nuB7xOzsp52dPpHFffvOCDupsG0QubkSMEySY=
+github.com/mitchellh/mapstructure v0.0.0-20160808181253-ca63d7c062ee/go.mod h1:FVVH3fgwuzCH5S8UJGiWEs2h04kUh9fWfEaFds41c1Y=
+github.com/mitchellh/mapstructure v1.1.2 h1:fmNYVwqnSfB9mZU6OS2O6GsXM+wcskZDuKQzvN1EDeE=
+github.com/mitchellh/mapstructure v1.1.2/go.mod h1:FVVH3fgwuzCH5S8UJGiWEs2h04kUh9fWfEaFds41c1Y=
+github.com/pascaldekloe/goe v0.0.0-20180627143212-57f6aae5913c h1:Lgl0gzECD8GnQ5QCWA8o6BtfL6mDH5rQgM4/fX3avOs=
+github.com/pascaldekloe/goe v0.0.0-20180627143212-57f6aae5913c/go.mod h1:lzWF7FIEvWOWxwDKqyGYQf6ZUaNfKdP144TG7ZOy1lc=
+github.com/pkg/errors v0.8.1 h1:iURUrRGxPUNPdy5/HRSm+Yj6okJ6UtLINN0Q9M4+h3I=
+github.com/pkg/errors v0.8.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
+github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
+github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/posener/complete v1.1.1/go.mod h1:em0nMJCgc9GFtwrmVmEMR/ZL6WyhyjMBndrE9hABlRI=
+github.com/ryanuber/columnize v0.0.0-20160712163229-9b3edd62028f/go.mod h1:sm1tb6uqfes/u+d4ooFouqFdy9/2g9QGwK3SQygK0Ts=
+github.com/sean-/seed v0.0.0-20170313163322-e2103e2c3529 h1:nn5Wsu0esKSJiIVhscUtVbo7ada43DJhG55ua/hjS5I=
+github.com/sean-/seed v0.0.0-20170313163322-e2103e2c3529/go.mod h1:DxrIzT+xaE7yg65j358z/aeFdxmN0P9QXhEzd20vsDc=
+github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
+github.com/stretchr/testify v1.2.2/go.mod h1:a8OnRcib4nhh0OaRAV+Yts87kKdq0PP7pXfy6kDkUVs=
+github.com/stretchr/testify v1.3.0 h1:TivCn/peBQ7UY8ooIcPgZFpTNSz0Q2U6UrFlUfqbe0Q=
+github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
+golang.org/x/crypto v0.0.0-20181029021203-45a5f77698d3 h1:KYQXGkl6vs02hK7pK4eIbw0NpNPedieTSTEiJ//bwGs=
+golang.org/x/crypto v0.0.0-20181029021203-45a5f77698d3/go.mod h1:6SG95UA2DQfeDnfUPMdvaQW0Q7yPrPDi9nlGo2tz2b4=
+golang.org/x/net v0.0.0-20181023162649-9b4f9f5ad519/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20181201002055-351d144fa1fc h1:a3CU5tJYVj92DY2LaA1kUkrsqD5/3mLDhx2NcNqyW+0=
+golang.org/x/net v0.0.0-20181201002055-351d144fa1fc/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/sync v0.0.0-20181221193216-37e7f081c4d4 h1:YUO/7uOKsKeq9UokNS62b8FYywz3ker1l1vDZRCRefw=
+golang.org/x/sync v0.0.0-20181221193216-37e7f081c4d4/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+golang.org/x/sys v0.0.0-20180823144017-11551d06cbcc/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20181026203630-95b1ffbd15a5 h1:x6r4Jo0KNzOOzYd8lbcRsqjuqEASK6ob3auvWYM4/8U=
+golang.org/x/sys v0.0.0-20181026203630-95b1ffbd15a5/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
diff --git a/vendor/github.com/hashicorp/consul/api/operator_license.go b/vendor/github.com/hashicorp/consul/api/operator_license.go
new file mode 100644
index 0000000..25aa702
--- /dev/null
+++ b/vendor/github.com/hashicorp/consul/api/operator_license.go
@@ -0,0 +1,111 @@
+package api
+
+import (
+ "io/ioutil"
+ "strings"
+ "time"
+)
+
+type License struct {
+ // The unique identifier of the license
+ LicenseID string `json:"license_id"`
+
+ // The customer ID associated with the license
+ CustomerID string `json:"customer_id"`
+
+ // If set, an identifier that should be used to lock the license to a
+ // particular site, cluster, etc.
+ InstallationID string `json:"installation_id"`
+
+ // The time at which the license was issued
+ IssueTime time.Time `json:"issue_time"`
+
+ // The time at which the license starts being valid
+ StartTime time.Time `json:"start_time"`
+
+ // The time after which the license expires
+ ExpirationTime time.Time `json:"expiration_time"`
+
+ // The time at which the license ceases to function and can
+ // no longer be used in any capacity
+ TerminationTime time.Time `json:"termination_time"`
+
+ // The product the license is valid for
+ Product string `json:"product"`
+
+ // License Specific Flags
+ Flags map[string]interface{} `json:"flags"`
+
+ // List of features enabled by the license
+ Features []string `json:"features"`
+}
+
+type LicenseReply struct {
+ Valid bool
+ License *License
+ Warnings []string
+}
+
+func (op *Operator) LicenseGet(q *QueryOptions) (*LicenseReply, error) {
+ var reply LicenseReply
+ if _, err := op.c.query("/v1/operator/license", &reply, q); err != nil {
+ return nil, err
+ } else {
+ return &reply, nil
+ }
+}
+
+func (op *Operator) LicenseGetSigned(q *QueryOptions) (string, error) {
+ r := op.c.newRequest("GET", "/v1/operator/license")
+ r.params.Set("signed", "1")
+ r.setQueryOptions(q)
+ _, resp, err := requireOK(op.c.doRequest(r))
+ if err != nil {
+ return "", err
+ }
+ defer resp.Body.Close()
+
+ data, err := ioutil.ReadAll(resp.Body)
+ if err != nil {
+ return "", err
+ }
+
+ return string(data), nil
+}
+
+// LicenseReset will reset the license to the builtin one if it is still valid.
+// If the builtin license is invalid, the current license stays active.
+func (op *Operator) LicenseReset(opts *WriteOptions) (*LicenseReply, error) {
+ var reply LicenseReply
+ r := op.c.newRequest("DELETE", "/v1/operator/license")
+ r.setWriteOptions(opts)
+ _, resp, err := requireOK(op.c.doRequest(r))
+ if err != nil {
+ return nil, err
+ }
+ defer resp.Body.Close()
+
+ if err := decodeBody(resp, &reply); err != nil {
+ return nil, err
+ }
+
+ return &reply, nil
+}
+
+func (op *Operator) LicensePut(license string, opts *WriteOptions) (*LicenseReply, error) {
+ var reply LicenseReply
+ r := op.c.newRequest("PUT", "/v1/operator/license")
+ r.setWriteOptions(opts)
+ r.body = strings.NewReader(license)
+ _, resp, err := requireOK(op.c.doRequest(r))
+ if err != nil {
+ return nil, err
+ }
+ defer resp.Body.Close()
+
+ if err := decodeBody(resp, &reply); err != nil {
+ return nil, err
+ }
+
+ return &reply, nil
+}
diff --git a/vendor/github.com/hashicorp/go-immutable-radix/CHANGELOG.md b/vendor/github.com/hashicorp/go-immutable-radix/CHANGELOG.md
new file mode 100644
index 0000000..dd7c0ef
--- /dev/null
+++ b/vendor/github.com/hashicorp/go-immutable-radix/CHANGELOG.md
@@ -0,0 +1,9 @@
+# 1.1.0 (May 22nd, 2019)
+
+FEATURES
+
+* Add `SeekLowerBound` to allow for range scans. [[GH-24](https://github.com/hashicorp/go-immutable-radix/pull/24)]
+
+# 1.0.0 (August 30th, 2018)
+
+* go mod adopted
diff --git a/vendor/github.com/hashicorp/go-immutable-radix/README.md b/vendor/github.com/hashicorp/go-immutable-radix/README.md
index 8910fcc..4b6338b 100644
--- a/vendor/github.com/hashicorp/go-immutable-radix/README.md
+++ b/vendor/github.com/hashicorp/go-immutable-radix/README.md
@@ -39,3 +39,28 @@
}
```
+Here is an example of performing a range scan of the keys.
+
+```go
+// Create a tree
+r := iradix.New()
+r, _, _ = r.Insert([]byte("001"), 1)
+r, _, _ = r.Insert([]byte("002"), 2)
+r, _, _ = r.Insert([]byte("005"), 5)
+r, _, _ = r.Insert([]byte("010"), 10)
+r, _, _ = r.Insert([]byte("100"), 10)
+
+// Range scan over the keys that sort lexicographically between [003, 050)
+it := r.Root().Iterator()
+it.SeekLowerBound([]byte("003"))
+for key, _, ok := it.Next(); ok; key, _, ok = it.Next() {
+ if key >= "050" {
+ break
+ }
+ fmt.Println(key)
+}
+// Output:
+// 005
+// 010
+```
+
diff --git a/vendor/github.com/hashicorp/go-immutable-radix/iter.go b/vendor/github.com/hashicorp/go-immutable-radix/iter.go
index 9815e02..1ecaf83 100644
--- a/vendor/github.com/hashicorp/go-immutable-radix/iter.go
+++ b/vendor/github.com/hashicorp/go-immutable-radix/iter.go
@@ -1,6 +1,8 @@
package iradix
-import "bytes"
+import (
+ "bytes"
+)
// Iterator is used to iterate over a set of nodes
// in pre-order
@@ -53,6 +55,101 @@
i.SeekPrefixWatch(prefix)
}
+func (i *Iterator) recurseMin(n *Node) *Node {
+ // Traverse to the minimum child
+ if n.leaf != nil {
+ return n
+ }
+ if len(n.edges) > 0 {
+ // Add all the other edges to the stack (the min node will be added as
+ // we recurse)
+ i.stack = append(i.stack, n.edges[1:])
+ return i.recurseMin(n.edges[0].node)
+ }
+ // Shouldn't be possible
+ return nil
+}
+
+// SeekLowerBound is used to seek the iterator to the smallest key that is
+// greater or equal to the given key. There is no watch variant as it's hard to
+// predict based on the radix structure which node(s) changes might affect the
+// result.
+func (i *Iterator) SeekLowerBound(key []byte) {
+ // Wipe the stack. Unlike Prefix iteration, we need to build the stack as we
+ // go because we need only a subset of edges of many nodes in the path to the
+ // leaf with the lower bound.
+ i.stack = []edges{}
+ n := i.node
+ search := key
+
+ found := func(n *Node) {
+ i.node = n
+ i.stack = append(i.stack, edges{edge{node: n}})
+ }
+
+ for {
+ // Compare current prefix with the search key's same-length prefix.
+ var prefixCmp int
+ if len(n.prefix) < len(search) {
+ prefixCmp = bytes.Compare(n.prefix, search[0:len(n.prefix)])
+ } else {
+ prefixCmp = bytes.Compare(n.prefix, search)
+ }
+
+ if prefixCmp > 0 {
+ // Prefix is larger, that means the lower bound is greater than the search
+ // and from now on we need to follow the minimum path to the smallest
+ // leaf under this subtree.
+ n = i.recurseMin(n)
+ if n != nil {
+ found(n)
+ }
+ return
+ }
+
+ if prefixCmp < 0 {
+ // Prefix is smaller than search prefix, that means there is no lower
+ // bound
+ i.node = nil
+ return
+ }
+
+ // Prefix is equal, we are still heading for an exact match. If this is a
+ // leaf we're done.
+ if n.leaf != nil {
+ if bytes.Compare(n.leaf.key, key) < 0 {
+ i.node = nil
+ return
+ }
+ found(n)
+ return
+ }
+
+ // Consume the search prefix
+ if len(n.prefix) > len(search) {
+ search = []byte{}
+ } else {
+ search = search[len(n.prefix):]
+ }
+
+ // Otherwise, take the lower bound next edge.
+ idx, lbNode := n.getLowerBoundEdge(search[0])
+ if lbNode == nil {
+ i.node = nil
+ return
+ }
+
+ // Create stack edges for the all strictly higher edges in this node.
+ if idx+1 < len(n.edges) {
+ i.stack = append(i.stack, n.edges[idx+1:])
+ }
+
+ i.node = lbNode
+ // Recurse
+ n = lbNode
+ }
+}
+
// Next returns the next node in order
func (i *Iterator) Next() ([]byte, interface{}, bool) {
// Initialize our stack if needed
diff --git a/vendor/github.com/hashicorp/go-immutable-radix/node.go b/vendor/github.com/hashicorp/go-immutable-radix/node.go
index 7a065e7..3ab904e 100644
--- a/vendor/github.com/hashicorp/go-immutable-radix/node.go
+++ b/vendor/github.com/hashicorp/go-immutable-radix/node.go
@@ -79,6 +79,18 @@
return -1, nil
}
+func (n *Node) getLowerBoundEdge(label byte) (int, *Node) {
+ num := len(n.edges)
+ idx := sort.Search(num, func(i int) bool {
+ return n.edges[i].label >= label
+ })
+ // we want lower bound behavior so return even if it's not an exact match
+ if idx < num {
+ return idx, n.edges[idx].node
+ }
+ return -1, nil
+}
+
func (n *Node) delEdge(label byte) {
num := len(n.edges)
idx := sort.Search(num, func(i int) bool {
diff --git a/vendor/github.com/hashicorp/go-rootcerts/go.mod b/vendor/github.com/hashicorp/go-rootcerts/go.mod
index 3c0e0e6..e2dd024 100644
--- a/vendor/github.com/hashicorp/go-rootcerts/go.mod
+++ b/vendor/github.com/hashicorp/go-rootcerts/go.mod
@@ -1,3 +1,5 @@
module github.com/hashicorp/go-rootcerts
-require github.com/mitchellh/go-homedir v1.0.0
+go 1.12
+
+require github.com/mitchellh/go-homedir v1.1.0
diff --git a/vendor/github.com/hashicorp/go-rootcerts/go.sum b/vendor/github.com/hashicorp/go-rootcerts/go.sum
index d12bb75..ae38d14 100644
--- a/vendor/github.com/hashicorp/go-rootcerts/go.sum
+++ b/vendor/github.com/hashicorp/go-rootcerts/go.sum
@@ -1,2 +1,2 @@
-github.com/mitchellh/go-homedir v1.0.0 h1:vKb8ShqSby24Yrqr/yDYkuFz8d0WUjys40rvnGC8aR0=
-github.com/mitchellh/go-homedir v1.0.0/go.mod h1:SfyaCUpYCn1Vlf4IUYiD9fPX4A5wJrkLzIz1N1q0pr0=
+github.com/mitchellh/go-homedir v1.1.0 h1:lukF9ziXFxDFPkA1vsr5zpc1XuPDn/wFntq5mG+4E0Y=
+github.com/mitchellh/go-homedir v1.1.0/go.mod h1:SfyaCUpYCn1Vlf4IUYiD9fPX4A5wJrkLzIz1N1q0pr0=
diff --git a/vendor/github.com/hashicorp/go-uuid/.travis.yml b/vendor/github.com/hashicorp/go-uuid/.travis.yml
new file mode 100644
index 0000000..7698490
--- /dev/null
+++ b/vendor/github.com/hashicorp/go-uuid/.travis.yml
@@ -0,0 +1,12 @@
+language: go
+
+sudo: false
+
+go:
+ - 1.4
+ - 1.5
+ - 1.6
+ - tip
+
+script:
+ - go test -bench . -benchmem -v ./...
diff --git a/vendor/github.com/hashicorp/go-uuid/LICENSE b/vendor/github.com/hashicorp/go-uuid/LICENSE
new file mode 100644
index 0000000..e87a115
--- /dev/null
+++ b/vendor/github.com/hashicorp/go-uuid/LICENSE
@@ -0,0 +1,363 @@
+Mozilla Public License, version 2.0
+
+1. Definitions
+
+1.1. "Contributor"
+
+ means each individual or legal entity that creates, contributes to the
+ creation of, or owns Covered Software.
+
+1.2. "Contributor Version"
+
+ means the combination of the Contributions of others (if any) used by a
+ Contributor and that particular Contributor's Contribution.
+
+1.3. "Contribution"
+
+ means Covered Software of a particular Contributor.
+
+1.4. "Covered Software"
+
+ means Source Code Form to which the initial Contributor has attached the
+ notice in Exhibit A, the Executable Form of such Source Code Form, and
+ Modifications of such Source Code Form, in each case including portions
+ thereof.
+
+1.5. "Incompatible With Secondary Licenses"
+ means
+
+ a. that the initial Contributor has attached the notice described in
+ Exhibit B to the Covered Software; or
+
+ b. that the Covered Software was made available under the terms of
+ version 1.1 or earlier of the License, but not also under the terms of
+ a Secondary License.
+
+1.6. "Executable Form"
+
+ means any form of the work other than Source Code Form.
+
+1.7. "Larger Work"
+
+ means a work that combines Covered Software with other material, in a
+ separate file or files, that is not Covered Software.
+
+1.8. "License"
+
+ means this document.
+
+1.9. "Licensable"
+
+ means having the right to grant, to the maximum extent possible, whether
+ at the time of the initial grant or subsequently, any and all of the
+ rights conveyed by this License.
+
+1.10. "Modifications"
+
+ means any of the following:
+
+ a. any file in Source Code Form that results from an addition to,
+ deletion from, or modification of the contents of Covered Software; or
+
+ b. any new file in Source Code Form that contains any Covered Software.
+
+1.11. "Patent Claims" of a Contributor
+
+ means any patent claim(s), including without limitation, method,
+ process, and apparatus claims, in any patent Licensable by such
+ Contributor that would be infringed, but for the grant of the License,
+ by the making, using, selling, offering for sale, having made, import,
+ or transfer of either its Contributions or its Contributor Version.
+
+1.12. "Secondary License"
+
+ means either the GNU General Public License, Version 2.0, the GNU Lesser
+ General Public License, Version 2.1, the GNU Affero General Public
+ License, Version 3.0, or any later versions of those licenses.
+
+1.13. "Source Code Form"
+
+ means the form of the work preferred for making modifications.
+
+1.14. "You" (or "Your")
+
+ means an individual or a legal entity exercising rights under this
+ License. For legal entities, "You" includes any entity that controls, is
+ controlled by, or is under common control with You. For purposes of this
+ definition, "control" means (a) the power, direct or indirect, to cause
+ the direction or management of such entity, whether by contract or
+ otherwise, or (b) ownership of more than fifty percent (50%) of the
+ outstanding shares or beneficial ownership of such entity.
+
+
+2. License Grants and Conditions
+
+2.1. Grants
+
+ Each Contributor hereby grants You a world-wide, royalty-free,
+ non-exclusive license:
+
+ a. under intellectual property rights (other than patent or trademark)
+ Licensable by such Contributor to use, reproduce, make available,
+ modify, display, perform, distribute, and otherwise exploit its
+ Contributions, either on an unmodified basis, with Modifications, or
+ as part of a Larger Work; and
+
+ b. under Patent Claims of such Contributor to make, use, sell, offer for
+ sale, have made, import, and otherwise transfer either its
+ Contributions or its Contributor Version.
+
+2.2. Effective Date
+
+ The licenses granted in Section 2.1 with respect to any Contribution
+ become effective for each Contribution on the date the Contributor first
+ distributes such Contribution.
+
+2.3. Limitations on Grant Scope
+
+ The licenses granted in this Section 2 are the only rights granted under
+ this License. No additional rights or licenses will be implied from the
+ distribution or licensing of Covered Software under this License.
+ Notwithstanding Section 2.1(b) above, no patent license is granted by a
+ Contributor:
+
+ a. for any code that a Contributor has removed from Covered Software; or
+
+ b. for infringements caused by: (i) Your and any other third party's
+ modifications of Covered Software, or (ii) the combination of its
+ Contributions with other software (except as part of its Contributor
+ Version); or
+
+ c. under Patent Claims infringed by Covered Software in the absence of
+ its Contributions.
+
+ This License does not grant any rights in the trademarks, service marks,
+ or logos of any Contributor (except as may be necessary to comply with
+ the notice requirements in Section 3.4).
+
+2.4. Subsequent Licenses
+
+ No Contributor makes additional grants as a result of Your choice to
+ distribute the Covered Software under a subsequent version of this
+ License (see Section 10.2) or under the terms of a Secondary License (if
+ permitted under the terms of Section 3.3).
+
+2.5. Representation
+
+ Each Contributor represents that the Contributor believes its
+ Contributions are its original creation(s) or it has sufficient rights to
+ grant the rights to its Contributions conveyed by this License.
+
+2.6. Fair Use
+
+ This License is not intended to limit any rights You have under
+ applicable copyright doctrines of fair use, fair dealing, or other
+ equivalents.
+
+2.7. Conditions
+
+ Sections 3.1, 3.2, 3.3, and 3.4 are conditions of the licenses granted in
+ Section 2.1.
+
+
+3. Responsibilities
+
+3.1. Distribution of Source Form
+
+ All distribution of Covered Software in Source Code Form, including any
+ Modifications that You create or to which You contribute, must be under
+ the terms of this License. You must inform recipients that the Source
+ Code Form of the Covered Software is governed by the terms of this
+ License, and how they can obtain a copy of this License. You may not
+ attempt to alter or restrict the recipients' rights in the Source Code
+ Form.
+
+3.2. Distribution of Executable Form
+
+ If You distribute Covered Software in Executable Form then:
+
+ a. such Covered Software must also be made available in Source Code Form,
+ as described in Section 3.1, and You must inform recipients of the
+ Executable Form how they can obtain a copy of such Source Code Form by
+ reasonable means in a timely manner, at a charge no more than the cost
+ of distribution to the recipient; and
+
+ b. You may distribute such Executable Form under the terms of this
+ License, or sublicense it under different terms, provided that the
+ license for the Executable Form does not attempt to limit or alter the
+ recipients' rights in the Source Code Form under this License.
+
+3.3. Distribution of a Larger Work
+
+ You may create and distribute a Larger Work under terms of Your choice,
+ provided that You also comply with the requirements of this License for
+ the Covered Software. If the Larger Work is a combination of Covered
+ Software with a work governed by one or more Secondary Licenses, and the
+ Covered Software is not Incompatible With Secondary Licenses, this
+ License permits You to additionally distribute such Covered Software
+ under the terms of such Secondary License(s), so that the recipient of
+ the Larger Work may, at their option, further distribute the Covered
+ Software under the terms of either this License or such Secondary
+ License(s).
+
+3.4. Notices
+
+ You may not remove or alter the substance of any license notices
+ (including copyright notices, patent notices, disclaimers of warranty, or
+ limitations of liability) contained within the Source Code Form of the
+ Covered Software, except that You may alter any license notices to the
+ extent required to remedy known factual inaccuracies.
+
+3.5. Application of Additional Terms
+
+ You may choose to offer, and to charge a fee for, warranty, support,
+ indemnity or liability obligations to one or more recipients of Covered
+ Software. However, You may do so only on Your own behalf, and not on
+ behalf of any Contributor. You must make it absolutely clear that any
+ such warranty, support, indemnity, or liability obligation is offered by
+ You alone, and You hereby agree to indemnify every Contributor for any
+ liability incurred by such Contributor as a result of warranty, support,
+ indemnity or liability terms You offer. You may include additional
+ disclaimers of warranty and limitations of liability specific to any
+ jurisdiction.
+
+4. Inability to Comply Due to Statute or Regulation
+
+ If it is impossible for You to comply with any of the terms of this License
+ with respect to some or all of the Covered Software due to statute,
+ judicial order, or regulation then You must: (a) comply with the terms of
+ this License to the maximum extent possible; and (b) describe the
+ limitations and the code they affect. Such description must be placed in a
+ text file included with all distributions of the Covered Software under
+ this License. Except to the extent prohibited by statute or regulation,
+ such description must be sufficiently detailed for a recipient of ordinary
+ skill to be able to understand it.
+
+5. Termination
+
+5.1. The rights granted under this License will terminate automatically if You
+ fail to comply with any of its terms. However, if You become compliant,
+ then the rights granted under this License from a particular Contributor
+ are reinstated (a) provisionally, unless and until such Contributor
+ explicitly and finally terminates Your grants, and (b) on an ongoing
+ basis, if such Contributor fails to notify You of the non-compliance by
+ some reasonable means prior to 60 days after You have come back into
+ compliance. Moreover, Your grants from a particular Contributor are
+ reinstated on an ongoing basis if such Contributor notifies You of the
+ non-compliance by some reasonable means, this is the first time You have
+ received notice of non-compliance with this License from such
+ Contributor, and You become compliant prior to 30 days after Your receipt
+ of the notice.
+
+5.2. If You initiate litigation against any entity by asserting a patent
+ infringement claim (excluding declaratory judgment actions,
+ counter-claims, and cross-claims) alleging that a Contributor Version
+ directly or indirectly infringes any patent, then the rights granted to
+ You by any and all Contributors for the Covered Software under Section
+ 2.1 of this License shall terminate.
+
+5.3. In the event of termination under Sections 5.1 or 5.2 above, all end user
+ license agreements (excluding distributors and resellers) which have been
+ validly granted by You or Your distributors under this License prior to
+ termination shall survive termination.
+
+6. Disclaimer of Warranty
+
+ Covered Software is provided under this License on an "as is" basis,
+ without warranty of any kind, either expressed, implied, or statutory,
+ including, without limitation, warranties that the Covered Software is free
+ of defects, merchantable, fit for a particular purpose or non-infringing.
+ The entire risk as to the quality and performance of the Covered Software
+ is with You. Should any Covered Software prove defective in any respect,
+ You (not any Contributor) assume the cost of any necessary servicing,
+ repair, or correction. This disclaimer of warranty constitutes an essential
+ part of this License. No use of any Covered Software is authorized under
+ this License except under this disclaimer.
+
+7. Limitation of Liability
+
+ Under no circumstances and under no legal theory, whether tort (including
+ negligence), contract, or otherwise, shall any Contributor, or anyone who
+ distributes Covered Software as permitted above, be liable to You for any
+ direct, indirect, special, incidental, or consequential damages of any
+ character including, without limitation, damages for lost profits, loss of
+ goodwill, work stoppage, computer failure or malfunction, or any and all
+ other commercial damages or losses, even if such party shall have been
+ informed of the possibility of such damages. This limitation of liability
+ shall not apply to liability for death or personal injury resulting from
+ such party's negligence to the extent applicable law prohibits such
+ limitation. Some jurisdictions do not allow the exclusion or limitation of
+ incidental or consequential damages, so this exclusion and limitation may
+ not apply to You.
+
+8. Litigation
+
+ Any litigation relating to this License may be brought only in the courts
+ of a jurisdiction where the defendant maintains its principal place of
+ business and such litigation shall be governed by laws of that
+ jurisdiction, without reference to its conflict-of-law provisions. Nothing
+ in this Section shall prevent a party's ability to bring cross-claims or
+ counter-claims.
+
+9. Miscellaneous
+
+ This License represents the complete agreement concerning the subject
+ matter hereof. If any provision of this License is held to be
+ unenforceable, such provision shall be reformed only to the extent
+ necessary to make it enforceable. Any law or regulation which provides that
+ the language of a contract shall be construed against the drafter shall not
+ be used to construe this License against a Contributor.
+
+
+10. Versions of the License
+
+10.1. New Versions
+
+ Mozilla Foundation is the license steward. Except as provided in Section
+ 10.3, no one other than the license steward has the right to modify or
+ publish new versions of this License. Each version will be given a
+ distinguishing version number.
+
+10.2. Effect of New Versions
+
+ You may distribute the Covered Software under the terms of the version
+ of the License under which You originally received the Covered Software,
+ or under the terms of any subsequent version published by the license
+ steward.
+
+10.3. Modified Versions
+
+ If you create software not governed by this License, and you want to
+ create a new license for such software, you may create and use a
+ modified version of this License if you rename the license and remove
+ any references to the name of the license steward (except to note that
+ such modified license differs from this License).
+
+10.4. Distributing Source Code Form that is Incompatible With Secondary
+ Licenses If You choose to distribute Source Code Form that is
+ Incompatible With Secondary Licenses under the terms of this version of
+ the License, the notice described in Exhibit B of this License must be
+ attached.
+
+Exhibit A - Source Code Form License Notice
+
+ This Source Code Form is subject to the
+ terms of the Mozilla Public License, v.
+ 2.0. If a copy of the MPL was not
+ distributed with this file, You can
+ obtain one at
+ http://mozilla.org/MPL/2.0/.
+
+If it is not possible or desirable to put the notice in a particular file,
+then You may include the notice in a location (such as a LICENSE file in a
+relevant directory) where a recipient would be likely to look for such a
+notice.
+
+You may add additional accurate notices of copyright ownership.
+
+Exhibit B - "Incompatible With Secondary Licenses" Notice
+
+ This Source Code Form is "Incompatible
+ With Secondary Licenses", as defined by
+ the Mozilla Public License, v. 2.0.
+
diff --git a/vendor/github.com/hashicorp/go-uuid/README.md b/vendor/github.com/hashicorp/go-uuid/README.md
new file mode 100644
index 0000000..fbde8b9
--- /dev/null
+++ b/vendor/github.com/hashicorp/go-uuid/README.md
@@ -0,0 +1,8 @@
+# uuid [](https://travis-ci.org/hashicorp/go-uuid)
+
+Generates UUID-format strings using high quality, _purely random_ bytes. It is **not** intended to be RFC compliant, merely to use a well-understood string representation of a 128-bit value. It can also parse UUID-format strings into their component bytes.
+
+Documentation
+=============
+
+The full documentation is available on [Godoc](http://godoc.org/github.com/hashicorp/go-uuid).
diff --git a/vendor/github.com/hashicorp/go-uuid/go.mod b/vendor/github.com/hashicorp/go-uuid/go.mod
new file mode 100644
index 0000000..dd57f9d
--- /dev/null
+++ b/vendor/github.com/hashicorp/go-uuid/go.mod
@@ -0,0 +1 @@
+module github.com/hashicorp/go-uuid
diff --git a/vendor/github.com/hashicorp/go-uuid/uuid.go b/vendor/github.com/hashicorp/go-uuid/uuid.go
new file mode 100644
index 0000000..911227f
--- /dev/null
+++ b/vendor/github.com/hashicorp/go-uuid/uuid.go
@@ -0,0 +1,65 @@
+package uuid
+
+import (
+ "crypto/rand"
+ "encoding/hex"
+ "fmt"
+)
+
+// GenerateRandomBytes is used to generate random bytes of given size.
+func GenerateRandomBytes(size int) ([]byte, error) {
+ buf := make([]byte, size)
+ if _, err := rand.Read(buf); err != nil {
+ return nil, fmt.Errorf("failed to read random bytes: %v", err)
+ }
+ return buf, nil
+}
+
+const uuidLen = 16
+
+// GenerateUUID is used to generate a random UUID
+func GenerateUUID() (string, error) {
+ buf, err := GenerateRandomBytes(uuidLen)
+ if err != nil {
+ return "", err
+ }
+ return FormatUUID(buf)
+}
+
+func FormatUUID(buf []byte) (string, error) {
+ if buflen := len(buf); buflen != uuidLen {
+ return "", fmt.Errorf("wrong length byte slice (%d)", buflen)
+ }
+
+ return fmt.Sprintf("%x-%x-%x-%x-%x",
+ buf[0:4],
+ buf[4:6],
+ buf[6:8],
+ buf[8:10],
+ buf[10:16]), nil
+}
+
+func ParseUUID(uuid string) ([]byte, error) {
+ if len(uuid) != 2 * uuidLen + 4 {
+ return nil, fmt.Errorf("uuid string is wrong length")
+ }
+
+ if uuid[8] != '-' ||
+ uuid[13] != '-' ||
+ uuid[18] != '-' ||
+ uuid[23] != '-' {
+ return nil, fmt.Errorf("uuid is improperly formatted")
+ }
+
+ hexStr := uuid[0:8] + uuid[9:13] + uuid[14:18] + uuid[19:23] + uuid[24:36]
+
+ ret, err := hex.DecodeString(hexStr)
+ if err != nil {
+ return nil, err
+ }
+ if len(ret) != uuidLen {
+ return nil, fmt.Errorf("decoded hex is the wrong length")
+ }
+
+ return ret, nil
+}
diff --git a/vendor/github.com/hashicorp/golang-lru/simplelru/lru.go b/vendor/github.com/hashicorp/golang-lru/simplelru/lru.go
index 5673773..a86c853 100644
--- a/vendor/github.com/hashicorp/golang-lru/simplelru/lru.go
+++ b/vendor/github.com/hashicorp/golang-lru/simplelru/lru.go
@@ -73,6 +73,9 @@
func (c *LRU) Get(key interface{}) (value interface{}, ok bool) {
if ent, ok := c.items[key]; ok {
c.evictList.MoveToFront(ent)
+ if ent.Value.(*entry) == nil {
+ return nil, false
+ }
return ent.Value.(*entry).value, true
}
return
@@ -142,6 +145,19 @@
return c.evictList.Len()
}
+// Resize changes the cache size.
+func (c *LRU) Resize(size int) (evicted int) {
+ diff := c.Len() - size
+ if diff < 0 {
+ diff = 0
+ }
+ for i := 0; i < diff; i++ {
+ c.removeOldest()
+ }
+ c.size = size
+ return diff
+}
+
// removeOldest removes the oldest item from the cache.
func (c *LRU) removeOldest() {
ent := c.evictList.Back()
diff --git a/vendor/github.com/hashicorp/golang-lru/simplelru/lru_interface.go b/vendor/github.com/hashicorp/golang-lru/simplelru/lru_interface.go
index 74c7077..92d7093 100644
--- a/vendor/github.com/hashicorp/golang-lru/simplelru/lru_interface.go
+++ b/vendor/github.com/hashicorp/golang-lru/simplelru/lru_interface.go
@@ -10,7 +10,7 @@
// updates the "recently used"-ness of the key. #value, isFound
Get(key interface{}) (value interface{}, ok bool)
- // Check if a key exsists in cache without updating the recent-ness.
+ // Checks if a key exists in cache without updating the recent-ness.
Contains(key interface{}) (ok bool)
// Returns key's value without updating the "recently used"-ness of the key.
@@ -31,6 +31,9 @@
// Returns the number of items in the cache.
Len() int
- // Clear all cache entries
+ // Clears all cache entries.
Purge()
+
+ // Resizes cache, returning number evicted
+ Resize(int) int
}
diff --git a/vendor/github.com/jcmturner/gofork/LICENSE b/vendor/github.com/jcmturner/gofork/LICENSE
new file mode 100644
index 0000000..6a66aea
--- /dev/null
+++ b/vendor/github.com/jcmturner/gofork/LICENSE
@@ -0,0 +1,27 @@
+Copyright (c) 2009 The Go Authors. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+ * Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+ * Redistributions in binary form must reproduce the above
+copyright notice, this list of conditions and the following disclaimer
+in the documentation and/or other materials provided with the
+distribution.
+ * Neither the name of Google Inc. nor the names of its
+contributors may be used to endorse or promote products derived from
+this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
diff --git a/vendor/github.com/jcmturner/gofork/encoding/asn1/README.md b/vendor/github.com/jcmturner/gofork/encoding/asn1/README.md
new file mode 100644
index 0000000..66a2a8c
--- /dev/null
+++ b/vendor/github.com/jcmturner/gofork/encoding/asn1/README.md
@@ -0,0 +1,5 @@
+This is a temporary repository that will be removed when the issues below are fixed in the core golang code.
+
+## Issues
+* [encoding/asn1: cannot marshal into a GeneralString](https://github.com/golang/go/issues/18832)
+* [encoding/asn1: cannot marshal into slice of strings and pass stringtype parameter tags to members](https://github.com/golang/go/issues/18834)
\ No newline at end of file
diff --git a/vendor/github.com/jcmturner/gofork/encoding/asn1/asn1.go b/vendor/github.com/jcmturner/gofork/encoding/asn1/asn1.go
new file mode 100644
index 0000000..f1bb767
--- /dev/null
+++ b/vendor/github.com/jcmturner/gofork/encoding/asn1/asn1.go
@@ -0,0 +1,1003 @@
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// Package asn1 implements parsing of DER-encoded ASN.1 data structures,
+// as defined in ITU-T Rec X.690.
+//
+// See also ``A Layman's Guide to a Subset of ASN.1, BER, and DER,''
+// http://luca.ntop.org/Teaching/Appunti/asn1.html.
+package asn1
+
+// ASN.1 is a syntax for specifying abstract objects and BER, DER, PER, XER etc
+// are different encoding formats for those objects. Here, we'll be dealing
+// with DER, the Distinguished Encoding Rules. DER is used in X.509 because
+// it's fast to parse and, unlike BER, has a unique encoding for every object.
+// When calculating hashes over objects, it's important that the resulting
+// bytes be the same at both ends and DER removes this margin of error.
+//
+// ASN.1 is very complex and this package doesn't attempt to implement
+// everything by any means.
+
+import (
+ "errors"
+ "fmt"
+ "math/big"
+ "reflect"
+ "strconv"
+ "time"
+ "unicode/utf8"
+)
+
+// A StructuralError suggests that the ASN.1 data is valid, but the Go type
+// which is receiving it doesn't match.
+type StructuralError struct {
+ Msg string
+}
+
+func (e StructuralError) Error() string { return "asn1: structure error: " + e.Msg }
+
+// A SyntaxError suggests that the ASN.1 data is invalid.
+type SyntaxError struct {
+ Msg string
+}
+
+func (e SyntaxError) Error() string { return "asn1: syntax error: " + e.Msg }
+
+// We start by dealing with each of the primitive types in turn.
+
+// BOOLEAN
+
+func parseBool(bytes []byte) (ret bool, err error) {
+ if len(bytes) != 1 {
+ err = SyntaxError{"invalid boolean"}
+ return
+ }
+
+ // DER demands that "If the encoding represents the boolean value TRUE,
+ // its single contents octet shall have all eight bits set to one."
+ // Thus only 0 and 255 are valid encoded values.
+ switch bytes[0] {
+ case 0:
+ ret = false
+ case 0xff:
+ ret = true
+ default:
+ err = SyntaxError{"invalid boolean"}
+ }
+
+ return
+}
+
+// INTEGER
+
+// checkInteger returns nil if the given bytes are a valid DER-encoded
+// INTEGER and an error otherwise.
+func checkInteger(bytes []byte) error {
+ if len(bytes) == 0 {
+ return StructuralError{"empty integer"}
+ }
+ if len(bytes) == 1 {
+ return nil
+ }
+ if (bytes[0] == 0 && bytes[1]&0x80 == 0) || (bytes[0] == 0xff && bytes[1]&0x80 == 0x80) {
+ return StructuralError{"integer not minimally-encoded"}
+ }
+ return nil
+}
+
+// parseInt64 treats the given bytes as a big-endian, signed integer and
+// returns the result.
+func parseInt64(bytes []byte) (ret int64, err error) {
+ err = checkInteger(bytes)
+ if err != nil {
+ return
+ }
+ if len(bytes) > 8 {
+ // We'll overflow an int64 in this case.
+ err = StructuralError{"integer too large"}
+ return
+ }
+ for bytesRead := 0; bytesRead < len(bytes); bytesRead++ {
+ ret <<= 8
+ ret |= int64(bytes[bytesRead])
+ }
+
+ // Shift up and down in order to sign extend the result.
+ ret <<= 64 - uint8(len(bytes))*8
+ ret >>= 64 - uint8(len(bytes))*8
+ return
+}
+
+// parseInt treats the given bytes as a big-endian, signed integer and returns
+// the result.
+func parseInt32(bytes []byte) (int32, error) {
+ if err := checkInteger(bytes); err != nil {
+ return 0, err
+ }
+ ret64, err := parseInt64(bytes)
+ if err != nil {
+ return 0, err
+ }
+ if ret64 != int64(int32(ret64)) {
+ return 0, StructuralError{"integer too large"}
+ }
+ return int32(ret64), nil
+}
+
+var bigOne = big.NewInt(1)
+
+// parseBigInt treats the given bytes as a big-endian, signed integer and returns
+// the result.
+func parseBigInt(bytes []byte) (*big.Int, error) {
+ if err := checkInteger(bytes); err != nil {
+ return nil, err
+ }
+ ret := new(big.Int)
+ if len(bytes) > 0 && bytes[0]&0x80 == 0x80 {
+ // This is a negative number.
+ notBytes := make([]byte, len(bytes))
+ for i := range notBytes {
+ notBytes[i] = ^bytes[i]
+ }
+ ret.SetBytes(notBytes)
+ ret.Add(ret, bigOne)
+ ret.Neg(ret)
+ return ret, nil
+ }
+ ret.SetBytes(bytes)
+ return ret, nil
+}
+
+// BIT STRING
+
+// BitString is the structure to use when you want an ASN.1 BIT STRING type. A
+// bit string is padded up to the nearest byte in memory and the number of
+// valid bits is recorded. Padding bits will be zero.
+type BitString struct {
+ Bytes []byte // bits packed into bytes.
+ BitLength int // length in bits.
+}
+
+// At returns the bit at the given index. If the index is out of range it
+// returns false.
+func (b BitString) At(i int) int {
+ if i < 0 || i >= b.BitLength {
+ return 0
+ }
+ x := i / 8
+ y := 7 - uint(i%8)
+ return int(b.Bytes[x]>>y) & 1
+}
+
+// RightAlign returns a slice where the padding bits are at the beginning. The
+// slice may share memory with the BitString.
+func (b BitString) RightAlign() []byte {
+ shift := uint(8 - (b.BitLength % 8))
+ if shift == 8 || len(b.Bytes) == 0 {
+ return b.Bytes
+ }
+
+ a := make([]byte, len(b.Bytes))
+ a[0] = b.Bytes[0] >> shift
+ for i := 1; i < len(b.Bytes); i++ {
+ a[i] = b.Bytes[i-1] << (8 - shift)
+ a[i] |= b.Bytes[i] >> shift
+ }
+
+ return a
+}
+
+// parseBitString parses an ASN.1 bit string from the given byte slice and returns it.
+func parseBitString(bytes []byte) (ret BitString, err error) {
+ if len(bytes) == 0 {
+ err = SyntaxError{"zero length BIT STRING"}
+ return
+ }
+ paddingBits := int(bytes[0])
+ if paddingBits > 7 ||
+ len(bytes) == 1 && paddingBits > 0 ||
+ bytes[len(bytes)-1]&((1<<bytes[0])-1) != 0 {
+ err = SyntaxError{"invalid padding bits in BIT STRING"}
+ return
+ }
+ ret.BitLength = (len(bytes)-1)*8 - paddingBits
+ ret.Bytes = bytes[1:]
+ return
+}
+
+// OBJECT IDENTIFIER
+
+// An ObjectIdentifier represents an ASN.1 OBJECT IDENTIFIER.
+type ObjectIdentifier []int
+
+// Equal reports whether oi and other represent the same identifier.
+func (oi ObjectIdentifier) Equal(other ObjectIdentifier) bool {
+ if len(oi) != len(other) {
+ return false
+ }
+ for i := 0; i < len(oi); i++ {
+ if oi[i] != other[i] {
+ return false
+ }
+ }
+
+ return true
+}
+
+func (oi ObjectIdentifier) String() string {
+ var s string
+
+ for i, v := range oi {
+ if i > 0 {
+ s += "."
+ }
+ s += strconv.Itoa(v)
+ }
+
+ return s
+}
+
+// parseObjectIdentifier parses an OBJECT IDENTIFIER from the given bytes and
+// returns it. An object identifier is a sequence of variable length integers
+// that are assigned in a hierarchy.
+func parseObjectIdentifier(bytes []byte) (s []int, err error) {
+ if len(bytes) == 0 {
+ err = SyntaxError{"zero length OBJECT IDENTIFIER"}
+ return
+ }
+
+ // In the worst case, we get two elements from the first byte (which is
+ // encoded differently) and then every varint is a single byte long.
+ s = make([]int, len(bytes)+1)
+
+ // The first varint is 40*value1 + value2:
+ // According to this packing, value1 can take the values 0, 1 and 2 only.
+ // When value1 = 0 or value1 = 1, then value2 is <= 39. When value1 = 2,
+ // then there are no restrictions on value2.
+ v, offset, err := parseBase128Int(bytes, 0)
+ if err != nil {
+ return
+ }
+ if v < 80 {
+ s[0] = v / 40
+ s[1] = v % 40
+ } else {
+ s[0] = 2
+ s[1] = v - 80
+ }
+
+ i := 2
+ for ; offset < len(bytes); i++ {
+ v, offset, err = parseBase128Int(bytes, offset)
+ if err != nil {
+ return
+ }
+ s[i] = v
+ }
+ s = s[0:i]
+ return
+}
+
+// ENUMERATED
+
+// An Enumerated is represented as a plain int.
+type Enumerated int
+
+// FLAG
+
+// A Flag accepts any data and is set to true if present.
+type Flag bool
+
+// parseBase128Int parses a base-128 encoded int from the given offset in the
+// given byte slice. It returns the value and the new offset.
+func parseBase128Int(bytes []byte, initOffset int) (ret, offset int, err error) {
+ offset = initOffset
+ for shifted := 0; offset < len(bytes); shifted++ {
+ if shifted == 4 {
+ err = StructuralError{"base 128 integer too large"}
+ return
+ }
+ ret <<= 7
+ b := bytes[offset]
+ ret |= int(b & 0x7f)
+ offset++
+ if b&0x80 == 0 {
+ return
+ }
+ }
+ err = SyntaxError{"truncated base 128 integer"}
+ return
+}
+
+// UTCTime
+
+func parseUTCTime(bytes []byte) (ret time.Time, err error) {
+ s := string(bytes)
+
+ formatStr := "0601021504Z0700"
+ ret, err = time.Parse(formatStr, s)
+ if err != nil {
+ formatStr = "060102150405Z0700"
+ ret, err = time.Parse(formatStr, s)
+ }
+ if err != nil {
+ return
+ }
+
+ if serialized := ret.Format(formatStr); serialized != s {
+ err = fmt.Errorf("asn1: time did not serialize back to the original value and may be invalid: given %q, but serialized as %q", s, serialized)
+ return
+ }
+
+ if ret.Year() >= 2050 {
+ // UTCTime only encodes times prior to 2050. See https://tools.ietf.org/html/rfc5280#section-4.1.2.5.1
+ ret = ret.AddDate(-100, 0, 0)
+ }
+
+ return
+}
+
+// parseGeneralizedTime parses the GeneralizedTime from the given byte slice
+// and returns the resulting time.
+func parseGeneralizedTime(bytes []byte) (ret time.Time, err error) {
+ const formatStr = "20060102150405Z0700"
+ s := string(bytes)
+
+ if ret, err = time.Parse(formatStr, s); err != nil {
+ return
+ }
+
+ if serialized := ret.Format(formatStr); serialized != s {
+ err = fmt.Errorf("asn1: time did not serialize back to the original value and may be invalid: given %q, but serialized as %q", s, serialized)
+ }
+
+ return
+}
+
+// PrintableString
+
+// parsePrintableString parses a ASN.1 PrintableString from the given byte
+// array and returns it.
+func parsePrintableString(bytes []byte) (ret string, err error) {
+ for _, b := range bytes {
+ if !isPrintable(b) {
+ err = SyntaxError{"PrintableString contains invalid character"}
+ return
+ }
+ }
+ ret = string(bytes)
+ return
+}
+
+// isPrintable reports whether the given b is in the ASN.1 PrintableString set.
+func isPrintable(b byte) bool {
+ return 'a' <= b && b <= 'z' ||
+ 'A' <= b && b <= 'Z' ||
+ '0' <= b && b <= '9' ||
+ '\'' <= b && b <= ')' ||
+ '+' <= b && b <= '/' ||
+ b == ' ' ||
+ b == ':' ||
+ b == '=' ||
+ b == '?' ||
+ // This is technically not allowed in a PrintableString.
+ // However, x509 certificates with wildcard strings don't
+ // always use the correct string type so we permit it.
+ b == '*'
+}
+
+// IA5String
+
+// parseIA5String parses a ASN.1 IA5String (ASCII string) from the given
+// byte slice and returns it.
+func parseIA5String(bytes []byte) (ret string, err error) {
+ for _, b := range bytes {
+ if b >= utf8.RuneSelf {
+ err = SyntaxError{"IA5String contains invalid character"}
+ return
+ }
+ }
+ ret = string(bytes)
+ return
+}
+
+// T61String
+
+// parseT61String parses a ASN.1 T61String (8-bit clean string) from the given
+// byte slice and returns it.
+func parseT61String(bytes []byte) (ret string, err error) {
+ return string(bytes), nil
+}
+
+// UTF8String
+
+// parseUTF8String parses a ASN.1 UTF8String (raw UTF-8) from the given byte
+// array and returns it.
+func parseUTF8String(bytes []byte) (ret string, err error) {
+ if !utf8.Valid(bytes) {
+ return "", errors.New("asn1: invalid UTF-8 string")
+ }
+ return string(bytes), nil
+}
+
+// A RawValue represents an undecoded ASN.1 object.
+type RawValue struct {
+ Class, Tag int
+ IsCompound bool
+ Bytes []byte
+ FullBytes []byte // includes the tag and length
+}
+
+// RawContent is used to signal that the undecoded, DER data needs to be
+// preserved for a struct. To use it, the first field of the struct must have
+// this type. It's an error for any of the other fields to have this type.
+type RawContent []byte
+
+// Tagging
+
+// parseTagAndLength parses an ASN.1 tag and length pair from the given offset
+// into a byte slice. It returns the parsed data and the new offset. SET and
+// SET OF (tag 17) are mapped to SEQUENCE and SEQUENCE OF (tag 16) since we
+// don't distinguish between ordered and unordered objects in this code.
+func parseTagAndLength(bytes []byte, initOffset int) (ret tagAndLength, offset int, err error) {
+ offset = initOffset
+ // parseTagAndLength should not be called without at least a single
+ // byte to read. Thus this check is for robustness:
+ if offset >= len(bytes) {
+ err = errors.New("asn1: internal error in parseTagAndLength")
+ return
+ }
+ b := bytes[offset]
+ offset++
+ ret.class = int(b >> 6)
+ ret.isCompound = b&0x20 == 0x20
+ ret.tag = int(b & 0x1f)
+
+ // If the bottom five bits are set, then the tag number is actually base 128
+ // encoded afterwards
+ if ret.tag == 0x1f {
+ ret.tag, offset, err = parseBase128Int(bytes, offset)
+ if err != nil {
+ return
+ }
+ // Tags should be encoded in minimal form.
+ if ret.tag < 0x1f {
+ err = SyntaxError{"non-minimal tag"}
+ return
+ }
+ }
+ if offset >= len(bytes) {
+ err = SyntaxError{"truncated tag or length"}
+ return
+ }
+ b = bytes[offset]
+ offset++
+ if b&0x80 == 0 {
+ // The length is encoded in the bottom 7 bits.
+ ret.length = int(b & 0x7f)
+ } else {
+ // Bottom 7 bits give the number of length bytes to follow.
+ numBytes := int(b & 0x7f)
+ if numBytes == 0 {
+ err = SyntaxError{"indefinite length found (not DER)"}
+ return
+ }
+ ret.length = 0
+ for i := 0; i < numBytes; i++ {
+ if offset >= len(bytes) {
+ err = SyntaxError{"truncated tag or length"}
+ return
+ }
+ b = bytes[offset]
+ offset++
+ if ret.length >= 1<<23 {
+ // We can't shift ret.length up without
+ // overflowing.
+ err = StructuralError{"length too large"}
+ return
+ }
+ ret.length <<= 8
+ ret.length |= int(b)
+ if ret.length == 0 {
+ // DER requires that lengths be minimal.
+ err = StructuralError{"superfluous leading zeros in length"}
+ return
+ }
+ }
+ // Short lengths must be encoded in short form.
+ if ret.length < 0x80 {
+ err = StructuralError{"non-minimal length"}
+ return
+ }
+ }
+
+ return
+}
+
+// parseSequenceOf is used for SEQUENCE OF and SET OF values. It tries to parse
+// a number of ASN.1 values from the given byte slice and returns them as a
+// slice of Go values of the given type.
+func parseSequenceOf(bytes []byte, sliceType reflect.Type, elemType reflect.Type) (ret reflect.Value, err error) {
+ expectedTag, compoundType, ok := getUniversalType(elemType)
+ if !ok {
+ err = StructuralError{"unknown Go type for slice"}
+ return
+ }
+
+ // First we iterate over the input and count the number of elements,
+ // checking that the types are correct in each case.
+ numElements := 0
+ for offset := 0; offset < len(bytes); {
+ var t tagAndLength
+ t, offset, err = parseTagAndLength(bytes, offset)
+ if err != nil {
+ return
+ }
+ switch t.tag {
+ case TagIA5String, TagGeneralString, TagT61String, TagUTF8String:
+ // We pretend that various other string types are
+ // PRINTABLE STRINGs so that a sequence of them can be
+ // parsed into a []string.
+ t.tag = TagPrintableString
+ case TagGeneralizedTime, TagUTCTime:
+ // Likewise, both time types are treated the same.
+ t.tag = TagUTCTime
+ }
+
+ if t.class != ClassUniversal || t.isCompound != compoundType || t.tag != expectedTag {
+ err = StructuralError{"sequence tag mismatch"}
+ return
+ }
+ if invalidLength(offset, t.length, len(bytes)) {
+ err = SyntaxError{"truncated sequence"}
+ return
+ }
+ offset += t.length
+ numElements++
+ }
+ ret = reflect.MakeSlice(sliceType, numElements, numElements)
+ params := fieldParameters{}
+ offset := 0
+ for i := 0; i < numElements; i++ {
+ offset, err = parseField(ret.Index(i), bytes, offset, params)
+ if err != nil {
+ return
+ }
+ }
+ return
+}
+
+var (
+ bitStringType = reflect.TypeOf(BitString{})
+ objectIdentifierType = reflect.TypeOf(ObjectIdentifier{})
+ enumeratedType = reflect.TypeOf(Enumerated(0))
+ flagType = reflect.TypeOf(Flag(false))
+ timeType = reflect.TypeOf(time.Time{})
+ rawValueType = reflect.TypeOf(RawValue{})
+ rawContentsType = reflect.TypeOf(RawContent(nil))
+ bigIntType = reflect.TypeOf(new(big.Int))
+)
+
+// invalidLength returns true iff offset + length > sliceLength, or if the
+// addition would overflow.
+func invalidLength(offset, length, sliceLength int) bool {
+ return offset+length < offset || offset+length > sliceLength
+}
+
+// parseField is the main parsing function. Given a byte slice and an offset
+// into the array, it will try to parse a suitable ASN.1 value out and store it
+// in the given Value.
+func parseField(v reflect.Value, bytes []byte, initOffset int, params fieldParameters) (offset int, err error) {
+ offset = initOffset
+ fieldType := v.Type()
+
+ // If we have run out of data, it may be that there are optional elements at the end.
+ if offset == len(bytes) {
+ if !setDefaultValue(v, params) {
+ err = SyntaxError{"sequence truncated"}
+ }
+ return
+ }
+
+ // Deal with raw values.
+ if fieldType == rawValueType {
+ var t tagAndLength
+ t, offset, err = parseTagAndLength(bytes, offset)
+ if err != nil {
+ return
+ }
+ if invalidLength(offset, t.length, len(bytes)) {
+ err = SyntaxError{"data truncated"}
+ return
+ }
+ result := RawValue{t.class, t.tag, t.isCompound, bytes[offset : offset+t.length], bytes[initOffset : offset+t.length]}
+ offset += t.length
+ v.Set(reflect.ValueOf(result))
+ return
+ }
+
+ // Deal with the ANY type.
+ if ifaceType := fieldType; ifaceType.Kind() == reflect.Interface && ifaceType.NumMethod() == 0 {
+ var t tagAndLength
+ t, offset, err = parseTagAndLength(bytes, offset)
+ if err != nil {
+ return
+ }
+ if invalidLength(offset, t.length, len(bytes)) {
+ err = SyntaxError{"data truncated"}
+ return
+ }
+ var result interface{}
+ if !t.isCompound && t.class == ClassUniversal {
+ innerBytes := bytes[offset : offset+t.length]
+ switch t.tag {
+ case TagPrintableString:
+ result, err = parsePrintableString(innerBytes)
+ case TagIA5String:
+ result, err = parseIA5String(innerBytes)
+ // jtasn1 addition of following case
+ case TagGeneralString:
+ result, err = parseIA5String(innerBytes)
+ case TagT61String:
+ result, err = parseT61String(innerBytes)
+ case TagUTF8String:
+ result, err = parseUTF8String(innerBytes)
+ case TagInteger:
+ result, err = parseInt64(innerBytes)
+ case TagBitString:
+ result, err = parseBitString(innerBytes)
+ case TagOID:
+ result, err = parseObjectIdentifier(innerBytes)
+ case TagUTCTime:
+ result, err = parseUTCTime(innerBytes)
+ case TagGeneralizedTime:
+ result, err = parseGeneralizedTime(innerBytes)
+ case TagOctetString:
+ result = innerBytes
+ default:
+ // If we don't know how to handle the type, we just leave Value as nil.
+ }
+ }
+ offset += t.length
+ if err != nil {
+ return
+ }
+ if result != nil {
+ v.Set(reflect.ValueOf(result))
+ }
+ return
+ }
+ universalTag, compoundType, ok1 := getUniversalType(fieldType)
+ if !ok1 {
+ err = StructuralError{fmt.Sprintf("unknown Go type: %v", fieldType)}
+ return
+ }
+
+ t, offset, err := parseTagAndLength(bytes, offset)
+ if err != nil {
+ return
+ }
+ if params.explicit {
+ expectedClass := ClassContextSpecific
+ if params.application {
+ expectedClass = ClassApplication
+ }
+ if offset == len(bytes) {
+ err = StructuralError{"explicit tag has no child"}
+ return
+ }
+ if t.class == expectedClass && t.tag == *params.tag && (t.length == 0 || t.isCompound) {
+ if t.length > 0 {
+ t, offset, err = parseTagAndLength(bytes, offset)
+ if err != nil {
+ return
+ }
+ } else {
+ if fieldType != flagType {
+ err = StructuralError{"zero length explicit tag was not an asn1.Flag"}
+ return
+ }
+ v.SetBool(true)
+ return
+ }
+ } else {
+ // The tags didn't match, it might be an optional element.
+ ok := setDefaultValue(v, params)
+ if ok {
+ offset = initOffset
+ } else {
+ err = StructuralError{"explicitly tagged member didn't match"}
+ }
+ return
+ }
+ }
+
+ // Special case for strings: all the ASN.1 string types map to the Go
+ // type string. getUniversalType returns the tag for PrintableString
+ // when it sees a string, so if we see a different string type on the
+ // wire, we change the universal type to match.
+ if universalTag == TagPrintableString {
+ if t.class == ClassUniversal {
+ switch t.tag {
+ case TagIA5String, TagGeneralString, TagT61String, TagUTF8String:
+ universalTag = t.tag
+ }
+ } else if params.stringType != 0 {
+ universalTag = params.stringType
+ }
+ }
+
+ // Special case for time: UTCTime and GeneralizedTime both map to the
+ // Go type time.Time.
+ if universalTag == TagUTCTime && t.tag == TagGeneralizedTime && t.class == ClassUniversal {
+ universalTag = TagGeneralizedTime
+ }
+
+ if params.set {
+ universalTag = TagSet
+ }
+
+ expectedClass := ClassUniversal
+ expectedTag := universalTag
+
+ if !params.explicit && params.tag != nil {
+ expectedClass = ClassContextSpecific
+ expectedTag = *params.tag
+ }
+
+ if !params.explicit && params.application && params.tag != nil {
+ expectedClass = ClassApplication
+ expectedTag = *params.tag
+ }
+
+ // We have unwrapped any explicit tagging at this point.
+ if t.class != expectedClass || t.tag != expectedTag || t.isCompound != compoundType {
+ // Tags don't match. Again, it could be an optional element.
+ ok := setDefaultValue(v, params)
+ if ok {
+ offset = initOffset
+ } else {
+ err = StructuralError{fmt.Sprintf("tags don't match (%d vs %+v) %+v %s @%d", expectedTag, t, params, fieldType.Name(), offset)}
+ }
+ return
+ }
+ if invalidLength(offset, t.length, len(bytes)) {
+ err = SyntaxError{"data truncated"}
+ return
+ }
+ innerBytes := bytes[offset : offset+t.length]
+ offset += t.length
+
+ // We deal with the structures defined in this package first.
+ switch fieldType {
+ case objectIdentifierType:
+ newSlice, err1 := parseObjectIdentifier(innerBytes)
+ v.Set(reflect.MakeSlice(v.Type(), len(newSlice), len(newSlice)))
+ if err1 == nil {
+ reflect.Copy(v, reflect.ValueOf(newSlice))
+ }
+ err = err1
+ return
+ case bitStringType:
+ bs, err1 := parseBitString(innerBytes)
+ if err1 == nil {
+ v.Set(reflect.ValueOf(bs))
+ }
+ err = err1
+ return
+ case timeType:
+ var time time.Time
+ var err1 error
+ if universalTag == TagUTCTime {
+ time, err1 = parseUTCTime(innerBytes)
+ } else {
+ time, err1 = parseGeneralizedTime(innerBytes)
+ }
+ if err1 == nil {
+ v.Set(reflect.ValueOf(time))
+ }
+ err = err1
+ return
+ case enumeratedType:
+ parsedInt, err1 := parseInt32(innerBytes)
+ if err1 == nil {
+ v.SetInt(int64(parsedInt))
+ }
+ err = err1
+ return
+ case flagType:
+ v.SetBool(true)
+ return
+ case bigIntType:
+ parsedInt, err1 := parseBigInt(innerBytes)
+ if err1 == nil {
+ v.Set(reflect.ValueOf(parsedInt))
+ }
+ err = err1
+ return
+ }
+ switch val := v; val.Kind() {
+ case reflect.Bool:
+ parsedBool, err1 := parseBool(innerBytes)
+ if err1 == nil {
+ val.SetBool(parsedBool)
+ }
+ err = err1
+ return
+ case reflect.Int, reflect.Int32, reflect.Int64:
+ if val.Type().Size() == 4 {
+ parsedInt, err1 := parseInt32(innerBytes)
+ if err1 == nil {
+ val.SetInt(int64(parsedInt))
+ }
+ err = err1
+ } else {
+ parsedInt, err1 := parseInt64(innerBytes)
+ if err1 == nil {
+ val.SetInt(parsedInt)
+ }
+ err = err1
+ }
+ return
+ // TODO(dfc) Add support for the remaining integer types
+ case reflect.Struct:
+ structType := fieldType
+
+ if structType.NumField() > 0 &&
+ structType.Field(0).Type == rawContentsType {
+ bytes := bytes[initOffset:offset]
+ val.Field(0).Set(reflect.ValueOf(RawContent(bytes)))
+ }
+
+ innerOffset := 0
+ for i := 0; i < structType.NumField(); i++ {
+ field := structType.Field(i)
+ if i == 0 && field.Type == rawContentsType {
+ continue
+ }
+ innerOffset, err = parseField(val.Field(i), innerBytes, innerOffset, parseFieldParameters(field.Tag.Get("asn1")))
+ if err != nil {
+ return
+ }
+ }
+ // We allow extra bytes at the end of the SEQUENCE because
+ // adding elements to the end has been used in X.509 as the
+ // version numbers have increased.
+ return
+ case reflect.Slice:
+ sliceType := fieldType
+ if sliceType.Elem().Kind() == reflect.Uint8 {
+ val.Set(reflect.MakeSlice(sliceType, len(innerBytes), len(innerBytes)))
+ reflect.Copy(val, reflect.ValueOf(innerBytes))
+ return
+ }
+ newSlice, err1 := parseSequenceOf(innerBytes, sliceType, sliceType.Elem())
+ if err1 == nil {
+ val.Set(newSlice)
+ }
+ err = err1
+ return
+ case reflect.String:
+ var v string
+ switch universalTag {
+ case TagPrintableString:
+ v, err = parsePrintableString(innerBytes)
+ case TagIA5String:
+ v, err = parseIA5String(innerBytes)
+ case TagT61String:
+ v, err = parseT61String(innerBytes)
+ case TagUTF8String:
+ v, err = parseUTF8String(innerBytes)
+ case TagGeneralString:
+ // GeneralString is specified in ISO-2022/ECMA-35,
+ // A brief review suggests that it includes structures
+ // that allow the encoding to change midstring and
+ // such. We give up and pass it as an 8-bit string.
+ v, err = parseT61String(innerBytes)
+ default:
+ err = SyntaxError{fmt.Sprintf("internal error: unknown string type %d", universalTag)}
+ }
+ if err == nil {
+ val.SetString(v)
+ }
+ return
+ }
+ err = StructuralError{"unsupported: " + v.Type().String()}
+ return
+}
+
+// canHaveDefaultValue reports whether k is a Kind that we will set a default
+// value for. (A signed integer, essentially.)
+func canHaveDefaultValue(k reflect.Kind) bool {
+ switch k {
+ case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64:
+ return true
+ }
+
+ return false
+}
+
+// setDefaultValue is used to install a default value, from a tag string, into
+// a Value. It is successful if the field was optional, even if a default value
+// wasn't provided or it failed to install it into the Value.
+func setDefaultValue(v reflect.Value, params fieldParameters) (ok bool) {
+ if !params.optional {
+ return
+ }
+ ok = true
+ if params.defaultValue == nil {
+ return
+ }
+ if canHaveDefaultValue(v.Kind()) {
+ v.SetInt(*params.defaultValue)
+ }
+ return
+}
+
+// Unmarshal parses the DER-encoded ASN.1 data structure b
+// and uses the reflect package to fill in an arbitrary value pointed at by val.
+// Because Unmarshal uses the reflect package, the structs
+// being written to must use upper case field names.
+//
+// An ASN.1 INTEGER can be written to an int, int32, int64,
+// or *big.Int (from the math/big package).
+// If the encoded value does not fit in the Go type,
+// Unmarshal returns a parse error.
+//
+// An ASN.1 BIT STRING can be written to a BitString.
+//
+// An ASN.1 OCTET STRING can be written to a []byte.
+//
+// An ASN.1 OBJECT IDENTIFIER can be written to an
+// ObjectIdentifier.
+//
+// An ASN.1 ENUMERATED can be written to an Enumerated.
+//
+// An ASN.1 UTCTIME or GENERALIZEDTIME can be written to a time.Time.
+//
+// An ASN.1 PrintableString or IA5String can be written to a string.
+//
+// Any of the above ASN.1 values can be written to an interface{}.
+// The value stored in the interface has the corresponding Go type.
+// For integers, that type is int64.
+//
+// An ASN.1 SEQUENCE OF x or SET OF x can be written
+// to a slice if an x can be written to the slice's element type.
+//
+// An ASN.1 SEQUENCE or SET can be written to a struct
+// if each of the elements in the sequence can be
+// written to the corresponding element in the struct.
+//
+// The following tags on struct fields have special meaning to Unmarshal:
+//
+// application specifies that a APPLICATION tag is used
+// default:x sets the default value for optional integer fields
+// explicit specifies that an additional, explicit tag wraps the implicit one
+// optional marks the field as ASN.1 OPTIONAL
+// set causes a SET, rather than a SEQUENCE type to be expected
+// tag:x specifies the ASN.1 tag number; implies ASN.1 CONTEXT SPECIFIC
+//
+// If the type of the first field of a structure is RawContent then the raw
+// ASN1 contents of the struct will be stored in it.
+//
+// If the type name of a slice element ends with "SET" then it's treated as if
+// the "set" tag was set on it. This can be used with nested slices where a
+// struct tag cannot be given.
+//
+// Other ASN.1 types are not supported; if it encounters them,
+// Unmarshal returns a parse error.
+func Unmarshal(b []byte, val interface{}) (rest []byte, err error) {
+ return UnmarshalWithParams(b, val, "")
+}
+
+// UnmarshalWithParams allows field parameters to be specified for the
+// top-level element. The form of the params is the same as the field tags.
+func UnmarshalWithParams(b []byte, val interface{}, params string) (rest []byte, err error) {
+ v := reflect.ValueOf(val).Elem()
+ offset, err := parseField(v, b, 0, parseFieldParameters(params))
+ if err != nil {
+ return nil, err
+ }
+ return b[offset:], nil
+}
diff --git a/vendor/github.com/jcmturner/gofork/encoding/asn1/common.go b/vendor/github.com/jcmturner/gofork/encoding/asn1/common.go
new file mode 100644
index 0000000..7a9da49
--- /dev/null
+++ b/vendor/github.com/jcmturner/gofork/encoding/asn1/common.go
@@ -0,0 +1,173 @@
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package asn1
+
+import (
+ "reflect"
+ "strconv"
+ "strings"
+)
+
+// ASN.1 objects have metadata preceding them:
+// the tag: the type of the object
+// a flag denoting if this object is compound or not
+// the class type: the namespace of the tag
+// the length of the object, in bytes
+
+// Here are some standard tags and classes
+
+// ASN.1 tags represent the type of the following object.
+const (
+ TagBoolean = 1
+ TagInteger = 2
+ TagBitString = 3
+ TagOctetString = 4
+ TagOID = 6
+ TagEnum = 10
+ TagUTF8String = 12
+ TagSequence = 16
+ TagSet = 17
+ TagPrintableString = 19
+ TagT61String = 20
+ TagIA5String = 22
+ TagUTCTime = 23
+ TagGeneralizedTime = 24
+ TagGeneralString = 27
+)
+
+// ASN.1 class types represent the namespace of the tag.
+const (
+ ClassUniversal = 0
+ ClassApplication = 1
+ ClassContextSpecific = 2
+ ClassPrivate = 3
+)
+
+type tagAndLength struct {
+ class, tag, length int
+ isCompound bool
+}
+
+// ASN.1 has IMPLICIT and EXPLICIT tags, which can be translated as "instead
+// of" and "in addition to". When not specified, every primitive type has a
+// default tag in the UNIVERSAL class.
+//
+// For example: a BIT STRING is tagged [UNIVERSAL 3] by default (although ASN.1
+// doesn't actually have a UNIVERSAL keyword). However, by saying [IMPLICIT
+// CONTEXT-SPECIFIC 42], that means that the tag is replaced by another.
+//
+// On the other hand, if it said [EXPLICIT CONTEXT-SPECIFIC 10], then an
+// /additional/ tag would wrap the default tag. This explicit tag will have the
+// compound flag set.
+//
+// (This is used in order to remove ambiguity with optional elements.)
+//
+// You can layer EXPLICIT and IMPLICIT tags to an arbitrary depth, however we
+// don't support that here. We support a single layer of EXPLICIT or IMPLICIT
+// tagging with tag strings on the fields of a structure.
+
+// fieldParameters is the parsed representation of tag string from a structure field.
+type fieldParameters struct {
+ optional bool // true iff the field is OPTIONAL
+ explicit bool // true iff an EXPLICIT tag is in use.
+ application bool // true iff an APPLICATION tag is in use.
+ defaultValue *int64 // a default value for INTEGER typed fields (maybe nil).
+ tag *int // the EXPLICIT or IMPLICIT tag (maybe nil).
+ stringType int // the string tag to use when marshaling.
+ timeType int // the time tag to use when marshaling.
+ set bool // true iff this should be encoded as a SET
+ omitEmpty bool // true iff this should be omitted if empty when marshaling.
+
+ // Invariants:
+ // if explicit is set, tag is non-nil.
+}
+
+// Given a tag string with the format specified in the package comment,
+// parseFieldParameters will parse it into a fieldParameters structure,
+// ignoring unknown parts of the string.
+func parseFieldParameters(str string) (ret fieldParameters) {
+ for _, part := range strings.Split(str, ",") {
+ switch {
+ case part == "optional":
+ ret.optional = true
+ case part == "explicit":
+ ret.explicit = true
+ if ret.tag == nil {
+ ret.tag = new(int)
+ }
+ case part == "generalized":
+ ret.timeType = TagGeneralizedTime
+ case part == "utc":
+ ret.timeType = TagUTCTime
+ case part == "ia5":
+ ret.stringType = TagIA5String
+ // jtasn1 case below added
+ case part == "generalstring":
+ ret.stringType = TagGeneralString
+ case part == "printable":
+ ret.stringType = TagPrintableString
+ case part == "utf8":
+ ret.stringType = TagUTF8String
+ case strings.HasPrefix(part, "default:"):
+ i, err := strconv.ParseInt(part[8:], 10, 64)
+ if err == nil {
+ ret.defaultValue = new(int64)
+ *ret.defaultValue = i
+ }
+ case strings.HasPrefix(part, "tag:"):
+ i, err := strconv.Atoi(part[4:])
+ if err == nil {
+ ret.tag = new(int)
+ *ret.tag = i
+ }
+ case part == "set":
+ ret.set = true
+ case part == "application":
+ ret.application = true
+ if ret.tag == nil {
+ ret.tag = new(int)
+ }
+ case part == "omitempty":
+ ret.omitEmpty = true
+ }
+ }
+ return
+}
+
+// Given a reflected Go type, getUniversalType returns the default tag number
+// and expected compound flag.
+func getUniversalType(t reflect.Type) (tagNumber int, isCompound, ok bool) {
+ switch t {
+ case objectIdentifierType:
+ return TagOID, false, true
+ case bitStringType:
+ return TagBitString, false, true
+ case timeType:
+ return TagUTCTime, false, true
+ case enumeratedType:
+ return TagEnum, false, true
+ case bigIntType:
+ return TagInteger, false, true
+ }
+ switch t.Kind() {
+ case reflect.Bool:
+ return TagBoolean, false, true
+ case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64:
+ return TagInteger, false, true
+ case reflect.Struct:
+ return TagSequence, true, true
+ case reflect.Slice:
+ if t.Elem().Kind() == reflect.Uint8 {
+ return TagOctetString, false, true
+ }
+ if strings.HasSuffix(t.Name(), "SET") {
+ return TagSet, true, true
+ }
+ return TagSequence, true, true
+ case reflect.String:
+ return TagPrintableString, false, true
+ }
+ return 0, false, false
+}
diff --git a/vendor/github.com/jcmturner/gofork/encoding/asn1/marshal.go b/vendor/github.com/jcmturner/gofork/encoding/asn1/marshal.go
new file mode 100644
index 0000000..f52eee9
--- /dev/null
+++ b/vendor/github.com/jcmturner/gofork/encoding/asn1/marshal.go
@@ -0,0 +1,659 @@
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package asn1
+
+import (
+ "bytes"
+ "errors"
+ "fmt"
+ "io"
+ "math/big"
+ "reflect"
+ "time"
+ "unicode/utf8"
+)
+
+// A forkableWriter is an in-memory buffer that can be
+// 'forked' to create new forkableWriters that bracket the
+// original. After
+// pre, post := w.fork()
+// the overall sequence of bytes represented is logically w+pre+post.
+type forkableWriter struct {
+ *bytes.Buffer
+ pre, post *forkableWriter
+}
+
+func newForkableWriter() *forkableWriter {
+ return &forkableWriter{new(bytes.Buffer), nil, nil}
+}
+
+func (f *forkableWriter) fork() (pre, post *forkableWriter) {
+ if f.pre != nil || f.post != nil {
+ panic("have already forked")
+ }
+ f.pre = newForkableWriter()
+ f.post = newForkableWriter()
+ return f.pre, f.post
+}
+
+func (f *forkableWriter) Len() (l int) {
+ l += f.Buffer.Len()
+ if f.pre != nil {
+ l += f.pre.Len()
+ }
+ if f.post != nil {
+ l += f.post.Len()
+ }
+ return
+}
+
+func (f *forkableWriter) writeTo(out io.Writer) (n int, err error) {
+ n, err = out.Write(f.Bytes())
+ if err != nil {
+ return
+ }
+
+ var nn int
+
+ if f.pre != nil {
+ nn, err = f.pre.writeTo(out)
+ n += nn
+ if err != nil {
+ return
+ }
+ }
+
+ if f.post != nil {
+ nn, err = f.post.writeTo(out)
+ n += nn
+ }
+ return
+}
+
+func marshalBase128Int(out *forkableWriter, n int64) (err error) {
+ if n == 0 {
+ err = out.WriteByte(0)
+ return
+ }
+
+ l := 0
+ for i := n; i > 0; i >>= 7 {
+ l++
+ }
+
+ for i := l - 1; i >= 0; i-- {
+ o := byte(n >> uint(i*7))
+ o &= 0x7f
+ if i != 0 {
+ o |= 0x80
+ }
+ err = out.WriteByte(o)
+ if err != nil {
+ return
+ }
+ }
+
+ return nil
+}
+
+func marshalInt64(out *forkableWriter, i int64) (err error) {
+ n := int64Length(i)
+
+ for ; n > 0; n-- {
+ err = out.WriteByte(byte(i >> uint((n-1)*8)))
+ if err != nil {
+ return
+ }
+ }
+
+ return nil
+}
+
+func int64Length(i int64) (numBytes int) {
+ numBytes = 1
+
+ for i > 127 {
+ numBytes++
+ i >>= 8
+ }
+
+ for i < -128 {
+ numBytes++
+ i >>= 8
+ }
+
+ return
+}
+
+func marshalBigInt(out *forkableWriter, n *big.Int) (err error) {
+ if n.Sign() < 0 {
+ // A negative number has to be converted to two's-complement
+ // form. So we'll subtract 1 and invert. If the
+ // most-significant-bit isn't set then we'll need to pad the
+ // beginning with 0xff in order to keep the number negative.
+ nMinus1 := new(big.Int).Neg(n)
+ nMinus1.Sub(nMinus1, bigOne)
+ bytes := nMinus1.Bytes()
+ for i := range bytes {
+ bytes[i] ^= 0xff
+ }
+ if len(bytes) == 0 || bytes[0]&0x80 == 0 {
+ err = out.WriteByte(0xff)
+ if err != nil {
+ return
+ }
+ }
+ _, err = out.Write(bytes)
+ } else if n.Sign() == 0 {
+ // Zero is written as a single 0 zero rather than no bytes.
+ err = out.WriteByte(0x00)
+ } else {
+ bytes := n.Bytes()
+ if len(bytes) > 0 && bytes[0]&0x80 != 0 {
+ // We'll have to pad this with 0x00 in order to stop it
+ // looking like a negative number.
+ err = out.WriteByte(0)
+ if err != nil {
+ return
+ }
+ }
+ _, err = out.Write(bytes)
+ }
+ return
+}
+
+func marshalLength(out *forkableWriter, i int) (err error) {
+ n := lengthLength(i)
+
+ for ; n > 0; n-- {
+ err = out.WriteByte(byte(i >> uint((n-1)*8)))
+ if err != nil {
+ return
+ }
+ }
+
+ return nil
+}
+
+func lengthLength(i int) (numBytes int) {
+ numBytes = 1
+ for i > 255 {
+ numBytes++
+ i >>= 8
+ }
+ return
+}
+
+func marshalTagAndLength(out *forkableWriter, t tagAndLength) (err error) {
+ b := uint8(t.class) << 6
+ if t.isCompound {
+ b |= 0x20
+ }
+ if t.tag >= 31 {
+ b |= 0x1f
+ err = out.WriteByte(b)
+ if err != nil {
+ return
+ }
+ err = marshalBase128Int(out, int64(t.tag))
+ if err != nil {
+ return
+ }
+ } else {
+ b |= uint8(t.tag)
+ err = out.WriteByte(b)
+ if err != nil {
+ return
+ }
+ }
+
+ if t.length >= 128 {
+ l := lengthLength(t.length)
+ err = out.WriteByte(0x80 | byte(l))
+ if err != nil {
+ return
+ }
+ err = marshalLength(out, t.length)
+ if err != nil {
+ return
+ }
+ } else {
+ err = out.WriteByte(byte(t.length))
+ if err != nil {
+ return
+ }
+ }
+
+ return nil
+}
+
+func marshalBitString(out *forkableWriter, b BitString) (err error) {
+ paddingBits := byte((8 - b.BitLength%8) % 8)
+ err = out.WriteByte(paddingBits)
+ if err != nil {
+ return
+ }
+ _, err = out.Write(b.Bytes)
+ return
+}
+
+func marshalObjectIdentifier(out *forkableWriter, oid []int) (err error) {
+ if len(oid) < 2 || oid[0] > 2 || (oid[0] < 2 && oid[1] >= 40) {
+ return StructuralError{"invalid object identifier"}
+ }
+
+ err = marshalBase128Int(out, int64(oid[0]*40+oid[1]))
+ if err != nil {
+ return
+ }
+ for i := 2; i < len(oid); i++ {
+ err = marshalBase128Int(out, int64(oid[i]))
+ if err != nil {
+ return
+ }
+ }
+
+ return
+}
+
+func marshalPrintableString(out *forkableWriter, s string) (err error) {
+ b := []byte(s)
+ for _, c := range b {
+ if !isPrintable(c) {
+ return StructuralError{"PrintableString contains invalid character"}
+ }
+ }
+
+ _, err = out.Write(b)
+ return
+}
+
+func marshalIA5String(out *forkableWriter, s string) (err error) {
+ b := []byte(s)
+ for _, c := range b {
+ if c > 127 {
+ return StructuralError{"IA5String contains invalid character"}
+ }
+ }
+
+ _, err = out.Write(b)
+ return
+}
+
+func marshalUTF8String(out *forkableWriter, s string) (err error) {
+ _, err = out.Write([]byte(s))
+ return
+}
+
+func marshalTwoDigits(out *forkableWriter, v int) (err error) {
+ err = out.WriteByte(byte('0' + (v/10)%10))
+ if err != nil {
+ return
+ }
+ return out.WriteByte(byte('0' + v%10))
+}
+
+func marshalFourDigits(out *forkableWriter, v int) (err error) {
+ var bytes [4]byte
+ for i := range bytes {
+ bytes[3-i] = '0' + byte(v%10)
+ v /= 10
+ }
+ _, err = out.Write(bytes[:])
+ return
+}
+
+func outsideUTCRange(t time.Time) bool {
+ year := t.Year()
+ return year < 1950 || year >= 2050
+}
+
+func marshalUTCTime(out *forkableWriter, t time.Time) (err error) {
+ year := t.Year()
+
+ switch {
+ case 1950 <= year && year < 2000:
+ err = marshalTwoDigits(out, year-1900)
+ case 2000 <= year && year < 2050:
+ err = marshalTwoDigits(out, year-2000)
+ default:
+ return StructuralError{"cannot represent time as UTCTime"}
+ }
+ if err != nil {
+ return
+ }
+
+ return marshalTimeCommon(out, t)
+}
+
+func marshalGeneralizedTime(out *forkableWriter, t time.Time) (err error) {
+ year := t.Year()
+ if year < 0 || year > 9999 {
+ return StructuralError{"cannot represent time as GeneralizedTime"}
+ }
+ if err = marshalFourDigits(out, year); err != nil {
+ return
+ }
+
+ return marshalTimeCommon(out, t)
+}
+
+func marshalTimeCommon(out *forkableWriter, t time.Time) (err error) {
+ _, month, day := t.Date()
+
+ err = marshalTwoDigits(out, int(month))
+ if err != nil {
+ return
+ }
+
+ err = marshalTwoDigits(out, day)
+ if err != nil {
+ return
+ }
+
+ hour, min, sec := t.Clock()
+
+ err = marshalTwoDigits(out, hour)
+ if err != nil {
+ return
+ }
+
+ err = marshalTwoDigits(out, min)
+ if err != nil {
+ return
+ }
+
+ err = marshalTwoDigits(out, sec)
+ if err != nil {
+ return
+ }
+
+ _, offset := t.Zone()
+
+ switch {
+ case offset/60 == 0:
+ err = out.WriteByte('Z')
+ return
+ case offset > 0:
+ err = out.WriteByte('+')
+ case offset < 0:
+ err = out.WriteByte('-')
+ }
+
+ if err != nil {
+ return
+ }
+
+ offsetMinutes := offset / 60
+ if offsetMinutes < 0 {
+ offsetMinutes = -offsetMinutes
+ }
+
+ err = marshalTwoDigits(out, offsetMinutes/60)
+ if err != nil {
+ return
+ }
+
+ err = marshalTwoDigits(out, offsetMinutes%60)
+ return
+}
+
+func stripTagAndLength(in []byte) []byte {
+ _, offset, err := parseTagAndLength(in, 0)
+ if err != nil {
+ return in
+ }
+ return in[offset:]
+}
+
+func marshalBody(out *forkableWriter, value reflect.Value, params fieldParameters) (err error) {
+ switch value.Type() {
+ case flagType:
+ return nil
+ case timeType:
+ t := value.Interface().(time.Time)
+ if params.timeType == TagGeneralizedTime || outsideUTCRange(t) {
+ return marshalGeneralizedTime(out, t)
+ } else {
+ return marshalUTCTime(out, t)
+ }
+ case bitStringType:
+ return marshalBitString(out, value.Interface().(BitString))
+ case objectIdentifierType:
+ return marshalObjectIdentifier(out, value.Interface().(ObjectIdentifier))
+ case bigIntType:
+ return marshalBigInt(out, value.Interface().(*big.Int))
+ }
+
+ switch v := value; v.Kind() {
+ case reflect.Bool:
+ if v.Bool() {
+ return out.WriteByte(255)
+ } else {
+ return out.WriteByte(0)
+ }
+ case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64:
+ return marshalInt64(out, v.Int())
+ case reflect.Struct:
+ t := v.Type()
+
+ startingField := 0
+
+ // If the first element of the structure is a non-empty
+ // RawContents, then we don't bother serializing the rest.
+ if t.NumField() > 0 && t.Field(0).Type == rawContentsType {
+ s := v.Field(0)
+ if s.Len() > 0 {
+ bytes := make([]byte, s.Len())
+ for i := 0; i < s.Len(); i++ {
+ bytes[i] = uint8(s.Index(i).Uint())
+ }
+ /* The RawContents will contain the tag and
+ * length fields but we'll also be writing
+ * those ourselves, so we strip them out of
+ * bytes */
+ _, err = out.Write(stripTagAndLength(bytes))
+ return
+ } else {
+ startingField = 1
+ }
+ }
+
+ for i := startingField; i < t.NumField(); i++ {
+ var pre *forkableWriter
+ pre, out = out.fork()
+ err = marshalField(pre, v.Field(i), parseFieldParameters(t.Field(i).Tag.Get("asn1")))
+ if err != nil {
+ return
+ }
+ }
+ return
+ case reflect.Slice:
+ sliceType := v.Type()
+ if sliceType.Elem().Kind() == reflect.Uint8 {
+ bytes := make([]byte, v.Len())
+ for i := 0; i < v.Len(); i++ {
+ bytes[i] = uint8(v.Index(i).Uint())
+ }
+ _, err = out.Write(bytes)
+ return
+ }
+
+ // jtasn1 Pass on the tags to the members but need to unset explicit switch and implicit value
+ //var fp fieldParameters
+ params.explicit = false
+ params.tag = nil
+ for i := 0; i < v.Len(); i++ {
+ var pre *forkableWriter
+ pre, out = out.fork()
+ err = marshalField(pre, v.Index(i), params)
+ if err != nil {
+ return
+ }
+ }
+ return
+ case reflect.String:
+ switch params.stringType {
+ case TagIA5String:
+ return marshalIA5String(out, v.String())
+ case TagPrintableString:
+ return marshalPrintableString(out, v.String())
+ default:
+ return marshalUTF8String(out, v.String())
+ }
+ }
+
+ return StructuralError{"unknown Go type"}
+}
+
+func marshalField(out *forkableWriter, v reflect.Value, params fieldParameters) (err error) {
+ if !v.IsValid() {
+ return fmt.Errorf("asn1: cannot marshal nil value")
+ }
+ // If the field is an interface{} then recurse into it.
+ if v.Kind() == reflect.Interface && v.Type().NumMethod() == 0 {
+ return marshalField(out, v.Elem(), params)
+ }
+
+ if v.Kind() == reflect.Slice && v.Len() == 0 && params.omitEmpty {
+ return
+ }
+
+ if params.optional && params.defaultValue != nil && canHaveDefaultValue(v.Kind()) {
+ defaultValue := reflect.New(v.Type()).Elem()
+ defaultValue.SetInt(*params.defaultValue)
+
+ if reflect.DeepEqual(v.Interface(), defaultValue.Interface()) {
+ return
+ }
+ }
+
+ // If no default value is given then the zero value for the type is
+ // assumed to be the default value. This isn't obviously the correct
+ // behaviour, but it's what Go has traditionally done.
+ if params.optional && params.defaultValue == nil {
+ if reflect.DeepEqual(v.Interface(), reflect.Zero(v.Type()).Interface()) {
+ return
+ }
+ }
+
+ if v.Type() == rawValueType {
+ rv := v.Interface().(RawValue)
+ if len(rv.FullBytes) != 0 {
+ _, err = out.Write(rv.FullBytes)
+ } else {
+ err = marshalTagAndLength(out, tagAndLength{rv.Class, rv.Tag, len(rv.Bytes), rv.IsCompound})
+ if err != nil {
+ return
+ }
+ _, err = out.Write(rv.Bytes)
+ }
+ return
+ }
+
+ tag, isCompound, ok := getUniversalType(v.Type())
+ if !ok {
+ err = StructuralError{fmt.Sprintf("unknown Go type: %v", v.Type())}
+ return
+ }
+ class := ClassUniversal
+
+ if params.timeType != 0 && tag != TagUTCTime {
+ return StructuralError{"explicit time type given to non-time member"}
+ }
+
+ // jtasn1 updated to allow slices of strings
+ if params.stringType != 0 && !(tag == TagPrintableString || (v.Kind() == reflect.Slice && tag == 16 && v.Type().Elem().Kind() == reflect.String)) {
+ return StructuralError{"explicit string type given to non-string member"}
+ }
+
+ switch tag {
+ case TagPrintableString:
+ if params.stringType == 0 {
+ // This is a string without an explicit string type. We'll use
+ // a PrintableString if the character set in the string is
+ // sufficiently limited, otherwise we'll use a UTF8String.
+ for _, r := range v.String() {
+ if r >= utf8.RuneSelf || !isPrintable(byte(r)) {
+ if !utf8.ValidString(v.String()) {
+ return errors.New("asn1: string not valid UTF-8")
+ }
+ tag = TagUTF8String
+ break
+ }
+ }
+ } else {
+ tag = params.stringType
+ }
+ case TagUTCTime:
+ if params.timeType == TagGeneralizedTime || outsideUTCRange(v.Interface().(time.Time)) {
+ tag = TagGeneralizedTime
+ }
+ }
+
+ if params.set {
+ if tag != TagSequence {
+ return StructuralError{"non sequence tagged as set"}
+ }
+ tag = TagSet
+ }
+
+ tags, body := out.fork()
+
+ err = marshalBody(body, v, params)
+ if err != nil {
+ return
+ }
+
+ bodyLen := body.Len()
+
+ var explicitTag *forkableWriter
+ if params.explicit {
+ explicitTag, tags = tags.fork()
+ }
+
+ if !params.explicit && params.tag != nil {
+ // implicit tag.
+ tag = *params.tag
+ class = ClassContextSpecific
+ }
+
+ err = marshalTagAndLength(tags, tagAndLength{class, tag, bodyLen, isCompound})
+ if err != nil {
+ return
+ }
+
+ if params.explicit {
+ err = marshalTagAndLength(explicitTag, tagAndLength{
+ class: ClassContextSpecific,
+ tag: *params.tag,
+ length: bodyLen + tags.Len(),
+ isCompound: true,
+ })
+ }
+
+ return err
+}
+
+// Marshal returns the ASN.1 encoding of val.
+//
+// In addition to the struct tags recognised by Unmarshal, the following can be
+// used:
+//
+// ia5: causes strings to be marshaled as ASN.1, IA5 strings
+// omitempty: causes empty slices to be skipped
+// printable: causes strings to be marshaled as ASN.1, PrintableString strings.
+// utf8: causes strings to be marshaled as ASN.1, UTF8 strings
+func Marshal(val interface{}) ([]byte, error) {
+ var out bytes.Buffer
+ v := reflect.ValueOf(val)
+ f := newForkableWriter()
+ err := marshalField(f, v, fieldParameters{})
+ if err != nil {
+ return nil, err
+ }
+ _, err = f.writeTo(&out)
+ return out.Bytes(), err
+}
diff --git a/vendor/github.com/jcmturner/gofork/x/crypto/pbkdf2/pbkdf2.go b/vendor/github.com/jcmturner/gofork/x/crypto/pbkdf2/pbkdf2.go
new file mode 100644
index 0000000..75d4187
--- /dev/null
+++ b/vendor/github.com/jcmturner/gofork/x/crypto/pbkdf2/pbkdf2.go
@@ -0,0 +1,98 @@
+// Copyright 2012 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+/*
+Package pbkdf2 implements the key derivation function PBKDF2 as defined in RFC
+2898 / PKCS #5 v2.0.
+
+A key derivation function is useful when encrypting data based on a password
+or any other not-fully-random data. It uses a pseudorandom function to derive
+a secure encryption key based on the password.
+
+While v2.0 of the standard defines only one pseudorandom function to use,
+HMAC-SHA1, the drafted v2.1 specification allows use of all five FIPS Approved
+Hash Functions SHA-1, SHA-224, SHA-256, SHA-384 and SHA-512 for HMAC. To
+choose, you can pass the `New` functions from the different SHA packages to
+pbkdf2.Key.
+*/
+package pbkdf2
+
+import (
+ "crypto/hmac"
+ "hash"
+)
+
+// Key derives a key from the password, salt and iteration count, returning a
+// []byte of length keylen that can be used as cryptographic key. The key is
+// derived based on the method described as PBKDF2 with the HMAC variant using
+// the supplied hash function.
+//
+// For example, to use a HMAC-SHA-1 based PBKDF2 key derivation function, you
+// can get a derived key for e.g. AES-256 (which needs a 32-byte key) by
+// doing:
+//
+// dk := pbkdf2.Key([]byte("some password"), salt, 4096, 32, sha1.New)
+//
+// Remember to get a good random salt. At least 8 bytes is recommended by the
+// RFC.
+//
+// Using a higher iteration count will increase the cost of an exhaustive
+// search but will also make derivation proportionally slower.
+func Key(password, salt []byte, iter, keyLen int, h func() hash.Hash) []byte {
+ return Key64(password, salt, int64(iter), int64(keyLen), h)
+}
+
+// Key64 derives a key from the password, salt and iteration count, returning a
+// []byte of length keylen that can be used as cryptographic key. Key64 uses
+// int64 for the iteration count and key length to allow larger values.
+// The key is derived based on the method described as PBKDF2 with the HMAC
+// variant using the supplied hash function.
+//
+// For example, to use a HMAC-SHA-1 based PBKDF2 key derivation function, you
+// can get a derived key for e.g. AES-256 (which needs a 32-byte key) by
+// doing:
+//
+// dk := pbkdf2.Key([]byte("some password"), salt, 4096, 32, sha1.New)
+//
+// Remember to get a good random salt. At least 8 bytes is recommended by the
+// RFC.
+//
+// Using a higher iteration count will increase the cost of an exhaustive
+// search but will also make derivation proportionally slower.
+func Key64(password, salt []byte, iter, keyLen int64, h func() hash.Hash) []byte {
+ prf := hmac.New(h, password)
+ hashLen := int64(prf.Size())
+ numBlocks := (keyLen + hashLen - 1) / hashLen
+
+ var buf [4]byte
+ dk := make([]byte, 0, numBlocks*hashLen)
+ U := make([]byte, hashLen)
+ for block := int64(1); block <= numBlocks; block++ {
+ // N.B.: || means concatenation, ^ means XOR
+ // for each block T_i = U_1 ^ U_2 ^ ... ^ U_iter
+ // U_1 = PRF(password, salt || uint(i))
+ prf.Reset()
+ prf.Write(salt)
+ buf[0] = byte(block >> 24)
+ buf[1] = byte(block >> 16)
+ buf[2] = byte(block >> 8)
+ buf[3] = byte(block)
+ prf.Write(buf[:4])
+ dk = prf.Sum(dk)
+ T := dk[int64(len(dk))-hashLen:]
+ copy(U, T)
+
+ // U_n = PRF(password, U_(n-1))
+ for n := int64(2); n <= iter; n++ {
+ prf.Reset()
+ prf.Write(U)
+ U = U[:0]
+ U = prf.Sum(U)
+ for x := range U {
+ T[x] ^= U[x]
+ }
+ }
+ }
+ return dk[:keyLen]
+}
diff --git a/vendor/github.com/opencord/voltha-go/adapters/README.md b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/README.md
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/adapters/README.md
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/README.md
diff --git a/vendor/github.com/opencord/voltha-go/adapters/adapterif/adapter_proxy_if.go b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/adapterif/adapter_proxy_if.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/adapters/adapterif/adapter_proxy_if.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/adapterif/adapter_proxy_if.go
diff --git a/vendor/github.com/opencord/voltha-go/adapters/adapterif/core_proxy_if.go b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/adapterif/core_proxy_if.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/adapters/adapterif/core_proxy_if.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/adapterif/core_proxy_if.go
diff --git a/vendor/github.com/opencord/voltha-go/adapters/adapterif/events_proxy_if.go b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/adapterif/events_proxy_if.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/adapters/adapterif/events_proxy_if.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/adapterif/events_proxy_if.go
diff --git a/vendor/github.com/opencord/voltha-go/adapters/common/adapter_proxy.go b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/adapter_proxy.go
similarity index 96%
rename from vendor/github.com/opencord/voltha-go/adapters/common/adapter_proxy.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/adapter_proxy.go
index 6c32422..fee70c8 100644
--- a/vendor/github.com/opencord/voltha-go/adapters/common/adapter_proxy.go
+++ b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/adapter_proxy.go
@@ -21,8 +21,8 @@
"github.com/golang/protobuf/ptypes"
"github.com/golang/protobuf/ptypes/any"
"github.com/google/uuid"
- "github.com/opencord/voltha-go/common/log"
- "github.com/opencord/voltha-go/kafka"
+ "github.com/opencord/voltha-lib-go/pkg/kafka"
+ "github.com/opencord/voltha-lib-go/pkg/log"
ic "github.com/opencord/voltha-protos/go/inter_container"
"time"
)
diff --git a/vendor/github.com/opencord/voltha-go/adapters/common/core_proxy.go b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/core_proxy.go
similarity index 99%
rename from vendor/github.com/opencord/voltha-go/adapters/common/core_proxy.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/core_proxy.go
index c9f332c..30117ca 100644
--- a/vendor/github.com/opencord/voltha-go/adapters/common/core_proxy.go
+++ b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/core_proxy.go
@@ -19,8 +19,8 @@
"context"
"github.com/golang/protobuf/ptypes"
a "github.com/golang/protobuf/ptypes/any"
- "github.com/opencord/voltha-go/common/log"
- "github.com/opencord/voltha-go/kafka"
+ "github.com/opencord/voltha-lib-go/pkg/kafka"
+ "github.com/opencord/voltha-lib-go/pkg/log"
ic "github.com/opencord/voltha-protos/go/inter_container"
"github.com/opencord/voltha-protos/go/voltha"
"google.golang.org/grpc/codes"
diff --git a/vendor/github.com/opencord/voltha-go/adapters/common/events_proxy.go b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/events_proxy.go
similarity index 95%
rename from vendor/github.com/opencord/voltha-go/adapters/common/events_proxy.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/events_proxy.go
index 34fcde7..25b1be4 100644
--- a/vendor/github.com/opencord/voltha-go/adapters/common/events_proxy.go
+++ b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/events_proxy.go
@@ -23,9 +23,9 @@
"strings"
"time"
- "github.com/opencord/voltha-go/adapters/adapterif"
- "github.com/opencord/voltha-go/common/log"
- "github.com/opencord/voltha-go/kafka"
+ "github.com/opencord/voltha-lib-go/pkg/adapters/adapterif"
+ "github.com/opencord/voltha-lib-go/pkg/kafka"
+ "github.com/opencord/voltha-lib-go/pkg/log"
"github.com/opencord/voltha-protos/go/voltha"
)
diff --git a/vendor/github.com/opencord/voltha-go/adapters/common/performance_metrics.go b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/performance_metrics.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/adapters/common/performance_metrics.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/performance_metrics.go
diff --git a/vendor/github.com/opencord/voltha-go/adapters/common/request_handler.go b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/request_handler.go
similarity index 98%
rename from vendor/github.com/opencord/voltha-go/adapters/common/request_handler.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/request_handler.go
index 55e04d7..27f9846 100644
--- a/vendor/github.com/opencord/voltha-go/adapters/common/request_handler.go
+++ b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/request_handler.go
@@ -19,10 +19,10 @@
"errors"
"github.com/golang/protobuf/ptypes"
"github.com/golang/protobuf/ptypes/empty"
- "github.com/opencord/voltha-go/adapters"
- "github.com/opencord/voltha-go/adapters/adapterif"
- "github.com/opencord/voltha-go/common/log"
- "github.com/opencord/voltha-go/kafka"
+ "github.com/opencord/voltha-lib-go/pkg/adapters"
+ "github.com/opencord/voltha-lib-go/pkg/adapters/adapterif"
+ "github.com/opencord/voltha-lib-go/pkg/kafka"
+ "github.com/opencord/voltha-lib-go/pkg/log"
ic "github.com/opencord/voltha-protos/go/inter_container"
"github.com/opencord/voltha-protos/go/openflow_13"
"github.com/opencord/voltha-protos/go/voltha"
diff --git a/vendor/github.com/opencord/voltha-go/adapters/common/utils.go b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/utils.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/adapters/common/utils.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/common/utils.go
diff --git a/vendor/github.com/opencord/voltha-go/adapters/iAdapter.go b/vendor/github.com/opencord/voltha-lib-go/pkg/adapters/iAdapter.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/adapters/iAdapter.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/adapters/iAdapter.go
diff --git a/vendor/github.com/opencord/voltha-go/db/kvstore/client.go b/vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/client.go
similarity index 98%
rename from vendor/github.com/opencord/voltha-go/db/kvstore/client.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/client.go
index 67c9219..f40d10e 100644
--- a/vendor/github.com/opencord/voltha-go/db/kvstore/client.go
+++ b/vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/client.go
@@ -16,7 +16,7 @@
package kvstore
import (
- "github.com/opencord/voltha-go/common/log"
+ "github.com/opencord/voltha-lib-go/pkg/log"
)
const (
diff --git a/vendor/github.com/opencord/voltha-go/db/kvstore/consulclient.go b/vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/consulclient.go
similarity index 99%
rename from vendor/github.com/opencord/voltha-go/db/kvstore/consulclient.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/consulclient.go
index c4fa0af..c3e3999 100644
--- a/vendor/github.com/opencord/voltha-go/db/kvstore/consulclient.go
+++ b/vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/consulclient.go
@@ -19,7 +19,7 @@
"bytes"
"context"
"errors"
- log "github.com/opencord/voltha-go/common/log"
+ log "github.com/opencord/voltha-lib-go/pkg/log"
"sync"
"time"
//log "ciena.com/coordinator/common"
diff --git a/vendor/github.com/opencord/voltha-go/db/kvstore/etcdclient.go b/vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/etcdclient.go
similarity index 96%
rename from vendor/github.com/opencord/voltha-go/db/kvstore/etcdclient.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/etcdclient.go
index 88c13ae..beac4e0 100644
--- a/vendor/github.com/opencord/voltha-go/db/kvstore/etcdclient.go
+++ b/vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/etcdclient.go
@@ -19,7 +19,7 @@
"context"
"errors"
"fmt"
- "github.com/opencord/voltha-go/common/log"
+ "github.com/opencord/voltha-lib-go/pkg/log"
v3Client "go.etcd.io/etcd/clientv3"
v3Concurrency "go.etcd.io/etcd/clientv3/concurrency"
v3rpcTypes "go.etcd.io/etcd/etcdserver/api/v3rpc/rpctypes"
@@ -127,7 +127,15 @@
c.writeLock.Lock()
defer c.writeLock.Unlock()
- _, err := c.ectdAPI.Put(ctx, key, val)
+
+ var err error
+ // Check if there is already a lease for this key - if there is then use it, otherwise a PUT will make
+ // that KV key permanent instead of automatically removing it after a lease expiration
+ if leaseID, ok := c.keyReservations[key]; ok {
+ _, err = c.ectdAPI.Put(ctx, key, val, v3Client.WithLease(*leaseID))
+ } else {
+ _, err = c.ectdAPI.Put(ctx, key, val)
+ }
cancel()
if err != nil {
switch err {
@@ -158,8 +166,8 @@
c.writeLock.Lock()
defer c.writeLock.Unlock()
- // delete the keys
- if _, err := c.ectdAPI.Delete(ctx, key, v3Client.WithPrefix()); err != nil {
+ // delete the key
+ if _, err := c.ectdAPI.Delete(ctx, key); err != nil {
log.Errorw("failed-to-delete-key", log.Fields{"key": key, "error": err})
return err
}
@@ -309,7 +317,7 @@
func (c *EtcdClient) Watch(key string) chan *Event {
w := v3Client.NewWatcher(c.ectdAPI)
ctx, cancel := context.WithCancel(context.Background())
- channel := w.Watch(ctx, key, v3Client.WithPrefix())
+ channel := w.Watch(ctx, key)
// Create a new channel
ch := make(chan *Event, maxClientChannelBufferSize)
@@ -317,8 +325,6 @@
// Keep track of the created channels so they can be closed when required
channelMap := make(map[chan *Event]v3Client.Watcher)
channelMap[ch] = w
- //c.writeLock.Lock()
- //defer c.writeLock.Unlock()
channelMaps := c.addChannelMap(key, channelMap)
@@ -412,7 +418,6 @@
defer close(ch)
for resp := range channel {
for _, ev := range resp.Events {
- //log.Debugf("%s %q : %q\n", ev.Type, ev.Kv.Key, ev.Kv.Value)
ch <- NewEvent(getEventType(ev), ev.Kv.Key, ev.Kv.Value, ev.Kv.Version)
}
}
diff --git a/vendor/github.com/opencord/voltha-go/db/kvstore/kvutils.go b/vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/kvutils.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/db/kvstore/kvutils.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/db/kvstore/kvutils.go
diff --git a/vendor/github.com/opencord/voltha-go/kafka/client.go b/vendor/github.com/opencord/voltha-lib-go/pkg/kafka/client.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/kafka/client.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/kafka/client.go
diff --git a/vendor/github.com/opencord/voltha-go/kafka/kafka_inter_container_library.go b/vendor/github.com/opencord/voltha-lib-go/pkg/kafka/kafka_inter_container_library.go
similarity index 97%
rename from vendor/github.com/opencord/voltha-go/kafka/kafka_inter_container_library.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/kafka/kafka_inter_container_library.go
index 56b5fa1..d2e0702 100644
--- a/vendor/github.com/opencord/voltha-go/kafka/kafka_inter_container_library.go
+++ b/vendor/github.com/opencord/voltha-lib-go/pkg/kafka/kafka_inter_container_library.go
@@ -23,7 +23,7 @@
"github.com/golang/protobuf/ptypes"
"github.com/golang/protobuf/ptypes/any"
"github.com/google/uuid"
- "github.com/opencord/voltha-go/common/log"
+ "github.com/opencord/voltha-lib-go/pkg/log"
ic "github.com/opencord/voltha-protos/go/inter_container"
"reflect"
"strings"
@@ -46,6 +46,9 @@
FromTopic = "fromTopic"
)
+var ErrorTransactionNotAcquired = errors.New("transaction-not-acquired")
+var ErrorTransactionInvalidId = errors.New("transaction-invalid-id")
+
// requestHandlerChannel represents an interface associated with a channel. Whenever, an event is
// obtained from that channel, this interface is invoked. This is used to handle
// async requests into the Core via the kafka messaging bus
@@ -674,15 +677,20 @@
// Check for errors first
lastIndex := len(out) - 1
if out[lastIndex].Interface() != nil { // Error
- if goError, ok := out[lastIndex].Interface().(error); ok {
- returnError = &ic.Error{Reason: goError.Error()}
+ if retError, ok := out[lastIndex].Interface().(error); ok {
+ if retError.Error() == ErrorTransactionNotAcquired.Error() {
+ log.Debugw("Ignoring request", log.Fields{"error": retError, "txId": msg.Header.Id})
+ return // Ignore - process is in competing mode and ignored transaction
+ }
+ returnError = &ic.Error{Reason: retError.Error()}
returnedValues = append(returnedValues, returnError)
} else { // Should never happen
returnError = &ic.Error{Reason: "incorrect-error-returns"}
returnedValues = append(returnedValues, returnError)
}
} else if len(out) == 2 && reflect.ValueOf(out[0].Interface()).IsValid() && reflect.ValueOf(out[0].Interface()).IsNil() {
- return // Ignore case - when core is in competing mode
+ log.Warnw("Unexpected response of (nil,nil)", log.Fields{"txId": msg.Header.Id})
+ return // Ignore - should not happen
} else { // Non-error case
success = true
for idx, val := range out {
diff --git a/vendor/github.com/opencord/voltha-go/kafka/sarama_client.go b/vendor/github.com/opencord/voltha-lib-go/pkg/kafka/sarama_client.go
similarity index 99%
rename from vendor/github.com/opencord/voltha-go/kafka/sarama_client.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/kafka/sarama_client.go
index 9e3ce0c..cec18ec 100644
--- a/vendor/github.com/opencord/voltha-go/kafka/sarama_client.go
+++ b/vendor/github.com/opencord/voltha-lib-go/pkg/kafka/sarama_client.go
@@ -22,7 +22,7 @@
scc "github.com/bsm/sarama-cluster"
"github.com/golang/protobuf/proto"
"github.com/google/uuid"
- "github.com/opencord/voltha-go/common/log"
+ "github.com/opencord/voltha-lib-go/pkg/log"
ic "github.com/opencord/voltha-protos/go/inter_container"
"strings"
"sync"
diff --git a/vendor/github.com/opencord/voltha-go/kafka/utils.go b/vendor/github.com/opencord/voltha-lib-go/pkg/kafka/utils.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/kafka/utils.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/kafka/utils.go
diff --git a/vendor/github.com/opencord/voltha-go/common/log/log.go b/vendor/github.com/opencord/voltha-lib-go/pkg/log/log.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/common/log/log.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/log/log.go
diff --git a/vendor/github.com/opencord/voltha-go/common/probe/probe.go b/vendor/github.com/opencord/voltha-lib-go/pkg/probe/probe.go
similarity index 99%
rename from vendor/github.com/opencord/voltha-go/common/probe/probe.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/probe/probe.go
index 8a8e485..192c7ca 100644
--- a/vendor/github.com/opencord/voltha-go/common/probe/probe.go
+++ b/vendor/github.com/opencord/voltha-lib-go/pkg/probe/probe.go
@@ -18,7 +18,7 @@
import (
"context"
"fmt"
- "github.com/opencord/voltha-go/common/log"
+ "github.com/opencord/voltha-lib-go/pkg/log"
"net/http"
"sync"
)
diff --git a/vendor/github.com/opencord/voltha-go/common/version/version.go b/vendor/github.com/opencord/voltha-lib-go/pkg/version/version.go
similarity index 100%
rename from vendor/github.com/opencord/voltha-go/common/version/version.go
rename to vendor/github.com/opencord/voltha-lib-go/pkg/version/version.go
diff --git a/vendor/github.com/pierrec/lz4/.gitignore b/vendor/github.com/pierrec/lz4/.gitignore
index e48bab3..5e98735 100644
--- a/vendor/github.com/pierrec/lz4/.gitignore
+++ b/vendor/github.com/pierrec/lz4/.gitignore
@@ -30,4 +30,5 @@
# End of https://www.gitignore.io/api/macos
-lz4c/lz4c
+cmd/*/*exe
+.idea
\ No newline at end of file
diff --git a/vendor/github.com/pierrec/lz4/.travis.yml b/vendor/github.com/pierrec/lz4/.travis.yml
index b2c806d..fd6c6db 100644
--- a/vendor/github.com/pierrec/lz4/.travis.yml
+++ b/vendor/github.com/pierrec/lz4/.travis.yml
@@ -1,9 +1,13 @@
language: go
+env:
+ - GO111MODULE=off
+
go:
- - 1.8.x
- 1.9.x
- 1.10.x
+ - 1.11.x
+ - 1.12.x
- master
matrix:
@@ -16,3 +20,5 @@
script:
- go test -v -cpu=2
- go test -v -cpu=2 -race
+ - go test -v -cpu=2 -tags noasm
+ - go test -v -cpu=2 -race -tags noasm
diff --git a/vendor/github.com/pierrec/lz4/README.md b/vendor/github.com/pierrec/lz4/README.md
index 50a10ee..be1f52a 100644
--- a/vendor/github.com/pierrec/lz4/README.md
+++ b/vendor/github.com/pierrec/lz4/README.md
@@ -1,24 +1,106 @@
-[](https://godoc.org/github.com/pierrec/lz4)
+# lz4 : LZ4 compression in pure Go
-# lz4
-LZ4 compression and decompression in pure Go.
+[](https://godoc.org/github.com/pierrec/lz4)
+[](https://travis-ci.org/pierrec/lz4)
+[](https://goreportcard.com/report/github.com/pierrec/lz4)
+[](https://github.com/pierrec/lz4/tags)
-## Usage
+## Overview
-```go
-import "github.com/pierrec/lz4"
+This package provides a streaming interface to [LZ4 data streams](http://fastcompression.blogspot.fr/2013/04/lz4-streaming-format-final.html) as well as low level compress and uncompress functions for LZ4 data blocks.
+The implementation is based on the reference C [one](https://github.com/lz4/lz4).
+
+## Install
+
+Assuming you have the go toolchain installed:
+
+```
+go get github.com/pierrec/lz4
```
-## Description
-Package lz4 implements reading and writing lz4 compressed data (a frame),
-as specified in http://fastcompression.blogspot.fr/2013/04/lz4-streaming-format-final.html.
+There is a command line interface tool to compress and decompress LZ4 files.
-This package is **compatible with the LZ4 frame format** although the block level compression
-and decompression functions are exposed and are fully compatible with the lz4 block format
-definition, they are low level and should not be used directly.
+```
+go install github.com/pierrec/lz4/cmd/lz4c
+```
-For a complete description of an lz4 compressed block, see:
-http://fastcompression.blogspot.fr/2011/05/lz4-explained.html
+Usage
-See https://github.com/Cyan4973/lz4 for the reference C implementation.
+```
+Usage of lz4c:
+ -version
+ print the program version
+Subcommands:
+Compress the given files or from stdin to stdout.
+compress [arguments] [<file name> ...]
+ -bc
+ enable block checksum
+ -l int
+ compression level (0=fastest)
+ -sc
+ disable stream checksum
+ -size string
+ block max size [64K,256K,1M,4M] (default "4M")
+
+Uncompress the given files or from stdin to stdout.
+uncompress [arguments] [<file name> ...]
+
+```
+
+
+## Example
+
+```
+// Compress and uncompress an input string.
+s := "hello world"
+r := strings.NewReader(s)
+
+// The pipe will uncompress the data from the writer.
+pr, pw := io.Pipe()
+zw := lz4.NewWriter(pw)
+zr := lz4.NewReader(pr)
+
+go func() {
+ // Compress the input string.
+ _, _ = io.Copy(zw, r)
+ _ = zw.Close() // Make sure the writer is closed
+ _ = pw.Close() // Terminate the pipe
+}()
+
+_, _ = io.Copy(os.Stdout, zr)
+
+// Output:
+// hello world
+```
+
+## Contributing
+
+Contributions are very welcome for bug fixing, performance improvements...!
+
+- Open an issue with a proper description
+- Send a pull request with appropriate test case(s)
+
+## Contributors
+
+Thanks to all contributors so far:
+
+- [@klauspost](https://github.com/klauspost)
+- [@heidawei](https://github.com/heidawei)
+- [@x4m](https://github.com/x4m)
+- [@Zariel](https://github.com/Zariel)
+- [@edwingeng](https://github.com/edwingeng)
+- [@danielmoy-google](https://github.com/danielmoy-google)
+- [@honda-tatsuya](https://github.com/honda-tatsuya)
+- [@h8liu](https://github.com/h8liu)
+- [@sbinet](https://github.com/sbinet)
+- [@fingon](https://github.com/fingon)
+- [@emfree](https://github.com/emfree)
+- [@lhemala](https://github.com/lhemala)
+- [@connor4312](https://github.com/connor4312)
+- [@oov](https://github.com/oov)
+- [@arya](https://github.com/arya)
+- [@ikkeps](https://github.com/ikkeps)
+
+Special thanks to [@Zariel](https://github.com/Zariel) for his asm implementation of the decoder
+Special thanks to [@klauspost](https://github.com/klauspost) for his work on optimizing the code
diff --git a/vendor/github.com/pierrec/lz4/block.go b/vendor/github.com/pierrec/lz4/block.go
index ef24f17..5755cda 100644
--- a/vendor/github.com/pierrec/lz4/block.go
+++ b/vendor/github.com/pierrec/lz4/block.go
@@ -2,21 +2,14 @@
import (
"encoding/binary"
- "errors"
+ "fmt"
+ "math/bits"
)
-var (
- // ErrInvalidSourceShortBuffer is returned by UncompressBlock or CompressBLock when a compressed
- // block is corrupted or the destination buffer is not large enough for the uncompressed data.
- ErrInvalidSourceShortBuffer = errors.New("lz4: invalid source or destination buffer too short")
- // ErrInvalid is returned when reading an invalid LZ4 archive.
- ErrInvalid = errors.New("lz4: bad magic number")
-)
-
-// blockHash hashes 4 bytes into a value < winSize.
-func blockHash(x uint32) uint32 {
- const hasher uint32 = 2654435761 // Knuth multiplicative hash.
- return x * hasher >> hashShift
+// blockHash hashes the lower 6 bytes into a value < htSize.
+func blockHash(x uint64) uint32 {
+ const prime6bytes = 227718039650203
+ return uint32(((x << (64 - 48)) * prime6bytes) >> (64 - hashLog))
}
// CompressBlockBound returns the maximum size of a given buffer of size n, when not compressible.
@@ -30,75 +23,14 @@
// The destination buffer must be sized appropriately.
//
// An error is returned if the source data is invalid or the destination buffer is too small.
-func UncompressBlock(src, dst []byte) (si int, err error) {
- defer func() {
- // It is now faster to let the runtime panic and recover on out of bound slice access
- // than checking indices as we go along.
- if recover() != nil {
- err = ErrInvalidSourceShortBuffer
- }
- }()
- sn := len(src)
- if sn == 0 {
+func UncompressBlock(src, dst []byte) (int, error) {
+ if len(src) == 0 {
return 0, nil
}
- var di int
-
- for {
- // Literals and match lengths (token).
- b := int(src[si])
- si++
-
- // Literals.
- if lLen := b >> 4; lLen > 0 {
- if lLen == 0xF {
- for src[si] == 0xFF {
- lLen += 0xFF
- si++
- }
- lLen += int(src[si])
- si++
- }
- i := si
- si += lLen
- di += copy(dst[di:], src[i:si])
-
- if si >= sn {
- return di, nil
- }
- }
-
- si++
- _ = src[si] // Bound check elimination.
- offset := int(src[si-1]) | int(src[si])<<8
- si++
-
- // Match.
- mLen := b & 0xF
- if mLen == 0xF {
- for src[si] == 0xFF {
- mLen += 0xFF
- si++
- }
- mLen += int(src[si])
- si++
- }
- mLen += minMatch
-
- // Copy the match.
- i := di - offset
- if offset > 0 && mLen >= offset {
- // Efficiently copy the match dst[di-offset:di] into the dst slice.
- bytesToCopy := offset * (mLen / offset)
- expanded := dst[i:]
- for n := offset; n <= bytesToCopy+offset; n *= 2 {
- copy(expanded[n:], expanded[:n])
- }
- di += bytesToCopy
- mLen -= bytesToCopy
- }
- di += copy(dst[di:], dst[i:i+mLen])
+ if di := decodeBlock(dst, src); di >= 0 {
+ return di, nil
}
+ return 0, ErrInvalidSourceShortBuffer
}
// CompressBlock compresses the source buffer into the destination one.
@@ -109,58 +41,98 @@
//
// An error is returned if the destination buffer is too small.
func CompressBlock(src, dst []byte, hashTable []int) (di int, err error) {
- defer func() {
- if recover() != nil {
- err = ErrInvalidSourceShortBuffer
- }
- }()
+ defer recoverBlock(&err)
+ // adaptSkipLog sets how quickly the compressor begins skipping blocks when data is incompressible.
+ // This significantly speeds up incompressible data and usually has very small impact on compresssion.
+ // bytes to skip = 1 + (bytes since last match >> adaptSkipLog)
+ const adaptSkipLog = 7
sn, dn := len(src)-mfLimit, len(dst)
if sn <= 0 || dn == 0 {
return 0, nil
}
- var si int
+ if len(hashTable) < htSize {
+ return 0, fmt.Errorf("hash table too small, should be at least %d in size", htSize)
+ }
+ // Prove to the compiler the table has at least htSize elements.
+ // The compiler can see that "uint32() >> hashShift" cannot be out of bounds.
+ hashTable = hashTable[:htSize]
+
+ // si: Current position of the search.
+ // anchor: Position of the current literals.
+ var si, anchor int
// Fast scan strategy: the hash table only stores the last 4 bytes sequences.
- // const accInit = 1 << skipStrength
-
- anchor := si // Position of the current literals.
- // acc := accInit // Variable step: improves performance on non-compressible data.
-
for si < sn {
- // Hash the next 4 bytes (sequence)...
- match := binary.LittleEndian.Uint32(src[si:])
+ // Hash the next 6 bytes (sequence)...
+ match := binary.LittleEndian.Uint64(src[si:])
h := blockHash(match)
+ h2 := blockHash(match >> 8)
+ // We check a match at s, s+1 and s+2 and pick the first one we get.
+ // Checking 3 only requires us to load the source one.
ref := hashTable[h]
+ ref2 := hashTable[h2]
hashTable[h] = si
- if ref >= sn { // Invalid reference (dirty hashtable).
- si++
- continue
- }
+ hashTable[h2] = si + 1
offset := si - ref
+
+ // If offset <= 0 we got an old entry in the hash table.
if offset <= 0 || offset >= winSize || // Out of window.
- match != binary.LittleEndian.Uint32(src[ref:]) { // Hash collision on different matches.
- // si += acc >> skipStrength
- // acc++
- si++
- continue
+ uint32(match) != binary.LittleEndian.Uint32(src[ref:]) { // Hash collision on different matches.
+ // No match. Start calculating another hash.
+ // The processor can usually do this out-of-order.
+ h = blockHash(match >> 16)
+ ref = hashTable[h]
+
+ // Check the second match at si+1
+ si += 1
+ offset = si - ref2
+
+ if offset <= 0 || offset >= winSize ||
+ uint32(match>>8) != binary.LittleEndian.Uint32(src[ref2:]) {
+ // No match. Check the third match at si+2
+ si += 1
+ offset = si - ref
+ hashTable[h] = si
+
+ if offset <= 0 || offset >= winSize ||
+ uint32(match>>16) != binary.LittleEndian.Uint32(src[ref:]) {
+ // Skip one extra byte (at si+3) before we check 3 matches again.
+ si += 2 + (si-anchor)>>adaptSkipLog
+ continue
+ }
+ }
}
// Match found.
- // acc = accInit
lLen := si - anchor // Literal length.
+ // We already matched 4 bytes.
+ mLen := 4
- // Encode match length part 1.
- si += minMatch
- mLen := si // Match length has minMatch already.
- // Find the longest match, first looking by batches of 8 bytes.
- for si < sn && binary.LittleEndian.Uint64(src[si:]) == binary.LittleEndian.Uint64(src[si-offset:]) {
- si += 8
+ // Extend backwards if we can, reducing literals.
+ tOff := si - offset - 1
+ for lLen > 0 && tOff >= 0 && src[si-1] == src[tOff] {
+ si--
+ tOff--
+ lLen--
+ mLen++
}
- // Then byte by byte.
- for si < sn && src[si] == src[si-offset] {
- si++
+
+ // Add the match length, so we continue search at the end.
+ // Use mLen to store the offset base.
+ si, mLen = si+mLen, si+minMatch
+
+ // Find the longest match by looking by batches of 8 bytes.
+ for si < sn {
+ x := binary.LittleEndian.Uint64(src[si:]) ^ binary.LittleEndian.Uint64(src[si-offset:])
+ if x == 0 {
+ si += 8
+ } else {
+ // Stop is first non-zero byte.
+ si += bits.TrailingZeros64(x) >> 3
+ break
+ }
}
mLen = si - mLen
@@ -186,7 +158,7 @@
di++
// Literals.
- copy(dst[di:], src[anchor:anchor+lLen])
+ copy(dst[di:di+lLen], src[anchor:anchor+lLen])
di += lLen + 2
anchor = si
@@ -203,6 +175,13 @@
dst[di] = byte(mLen)
di++
}
+ // Check if we can load next values.
+ if si >= sn {
+ break
+ }
+ // Hash match end-2
+ h = blockHash(binary.LittleEndian.Uint64(src[si-2:]))
+ hashTable[h] = si - 2
}
if anchor == 0 {
@@ -230,10 +209,16 @@
// Incompressible.
return 0, nil
}
- di += copy(dst[di:], src[anchor:])
+ di += copy(dst[di:di+len(src)-anchor], src[anchor:])
return di, nil
}
+// blockHash hashes 4 bytes into a value < winSize.
+func blockHashHC(x uint32) uint32 {
+ const hasher uint32 = 2654435761 // Knuth multiplicative hash.
+ return x * hasher >> (32 - winSizeLog)
+}
+
// CompressBlockHC compresses the source buffer src into the destination dst
// with max search depth (use 0 or negative value for no max).
//
@@ -243,11 +228,12 @@
//
// An error is returned if the destination buffer is too small.
func CompressBlockHC(src, dst []byte, depth int) (di int, err error) {
- defer func() {
- if recover() != nil {
- err = ErrInvalidSourceShortBuffer
- }
- }()
+ defer recoverBlock(&err)
+
+ // adaptSkipLog sets how quickly the compressor begins skipping blocks when data is incompressible.
+ // This significantly speeds up incompressible data and usually has very small impact on compresssion.
+ // bytes to skip = 1 + (bytes since last match >> adaptSkipLog)
+ const adaptSkipLog = 7
sn, dn := len(src)-mfLimit, len(dst)
if sn <= 0 || dn == 0 {
@@ -256,7 +242,7 @@
var si int
// hashTable: stores the last position found for a given hash
- // chaingTable: stores previous positions for a given hash
+ // chainTable: stores previous positions for a given hash
var hashTable, chainTable [winSize]int
if depth <= 0 {
@@ -267,7 +253,7 @@
for si < sn {
// Hash the next 4 bytes (sequence).
match := binary.LittleEndian.Uint32(src[si:])
- h := blockHash(match)
+ h := blockHashHC(match)
// Follow the chain until out of window and give the longest match.
mLen := 0
@@ -280,13 +266,17 @@
}
ml := 0
// Compare the current position with a previous with the same hash.
- for ml < sn-si && binary.LittleEndian.Uint64(src[next+ml:]) == binary.LittleEndian.Uint64(src[si+ml:]) {
- ml += 8
+ for ml < sn-si {
+ x := binary.LittleEndian.Uint64(src[next+ml:]) ^ binary.LittleEndian.Uint64(src[si+ml:])
+ if x == 0 {
+ ml += 8
+ } else {
+ // Stop is first non-zero byte.
+ ml += bits.TrailingZeros64(x) >> 3
+ break
+ }
}
- for ml < sn-si && src[next+ml] == src[si+ml] {
- ml++
- }
- if ml+1 < minMatch || ml <= mLen {
+ if ml < minMatch || ml <= mLen {
// Match too small (<minMath) or smaller than the current match.
continue
}
@@ -301,7 +291,7 @@
// No match found.
if mLen == 0 {
- si++
+ si += 1 + (si-anchor)>>adaptSkipLog
continue
}
@@ -315,7 +305,7 @@
for si, ml := winStart, si+mLen; si < ml; {
match >>= 8
match |= uint32(src[si+3]) << 24
- h := blockHash(match)
+ h := blockHashHC(match)
chainTable[si&winMask] = hashTable[h]
hashTable[h] = si
si++
@@ -347,7 +337,7 @@
di++
// Literals.
- copy(dst[di:], src[anchor:anchor+lLen])
+ copy(dst[di:di+lLen], src[anchor:anchor+lLen])
di += lLen
anchor = si
@@ -392,6 +382,6 @@
// Incompressible.
return 0, nil
}
- di += copy(dst[di:], src[anchor:])
+ di += copy(dst[di:di+len(src)-anchor], src[anchor:])
return di, nil
}
diff --git a/vendor/github.com/pierrec/lz4/decode_amd64.go b/vendor/github.com/pierrec/lz4/decode_amd64.go
new file mode 100644
index 0000000..43cc14f
--- /dev/null
+++ b/vendor/github.com/pierrec/lz4/decode_amd64.go
@@ -0,0 +1,8 @@
+// +build !appengine
+// +build gc
+// +build !noasm
+
+package lz4
+
+//go:noescape
+func decodeBlock(dst, src []byte) int
diff --git a/vendor/github.com/pierrec/lz4/decode_amd64.s b/vendor/github.com/pierrec/lz4/decode_amd64.s
new file mode 100644
index 0000000..20fef39
--- /dev/null
+++ b/vendor/github.com/pierrec/lz4/decode_amd64.s
@@ -0,0 +1,375 @@
+// +build !appengine
+// +build gc
+// +build !noasm
+
+#include "textflag.h"
+
+// AX scratch
+// BX scratch
+// CX scratch
+// DX token
+//
+// DI &dst
+// SI &src
+// R8 &dst + len(dst)
+// R9 &src + len(src)
+// R11 &dst
+// R12 short output end
+// R13 short input end
+// func decodeBlock(dst, src []byte) int
+// using 50 bytes of stack currently
+TEXT ·decodeBlock(SB), NOSPLIT, $64-56
+ MOVQ dst_base+0(FP), DI
+ MOVQ DI, R11
+ MOVQ dst_len+8(FP), R8
+ ADDQ DI, R8
+
+ MOVQ src_base+24(FP), SI
+ MOVQ src_len+32(FP), R9
+ ADDQ SI, R9
+
+ // shortcut ends
+ // short output end
+ MOVQ R8, R12
+ SUBQ $32, R12
+ // short input end
+ MOVQ R9, R13
+ SUBQ $16, R13
+
+loop:
+ // for si < len(src)
+ CMPQ SI, R9
+ JGE end
+
+ // token := uint32(src[si])
+ MOVBQZX (SI), DX
+ INCQ SI
+
+ // lit_len = token >> 4
+ // if lit_len > 0
+ // CX = lit_len
+ MOVQ DX, CX
+ SHRQ $4, CX
+
+ // if lit_len != 0xF
+ CMPQ CX, $0xF
+ JEQ lit_len_loop_pre
+ CMPQ DI, R12
+ JGE lit_len_loop_pre
+ CMPQ SI, R13
+ JGE lit_len_loop_pre
+
+ // copy shortcut
+
+ // A two-stage shortcut for the most common case:
+ // 1) If the literal length is 0..14, and there is enough space,
+ // enter the shortcut and copy 16 bytes on behalf of the literals
+ // (in the fast mode, only 8 bytes can be safely copied this way).
+ // 2) Further if the match length is 4..18, copy 18 bytes in a similar
+ // manner; but we ensure that there's enough space in the output for
+ // those 18 bytes earlier, upon entering the shortcut (in other words,
+ // there is a combined check for both stages).
+
+ // copy literal
+ MOVOU (SI), X0
+ MOVOU X0, (DI)
+ ADDQ CX, DI
+ ADDQ CX, SI
+
+ MOVQ DX, CX
+ ANDQ $0xF, CX
+
+ // The second stage: prepare for match copying, decode full info.
+ // If it doesn't work out, the info won't be wasted.
+ // offset := uint16(data[:2])
+ MOVWQZX (SI), DX
+ ADDQ $2, SI
+
+ MOVQ DI, AX
+ SUBQ DX, AX
+ CMPQ AX, DI
+ JGT err_short_buf
+
+ // if we can't do the second stage then jump straight to read the
+ // match length, we already have the offset.
+ CMPQ CX, $0xF
+ JEQ match_len_loop_pre
+ CMPQ DX, $8
+ JLT match_len_loop_pre
+ CMPQ AX, R11
+ JLT err_short_buf
+
+ // memcpy(op + 0, match + 0, 8);
+ MOVQ (AX), BX
+ MOVQ BX, (DI)
+ // memcpy(op + 8, match + 8, 8);
+ MOVQ 8(AX), BX
+ MOVQ BX, 8(DI)
+ // memcpy(op +16, match +16, 2);
+ MOVW 16(AX), BX
+ MOVW BX, 16(DI)
+
+ ADDQ $4, DI // minmatch
+ ADDQ CX, DI
+
+ // shortcut complete, load next token
+ JMP loop
+
+lit_len_loop_pre:
+ // if lit_len > 0
+ CMPQ CX, $0
+ JEQ offset
+ CMPQ CX, $0xF
+ JNE copy_literal
+
+lit_len_loop:
+ // for src[si] == 0xFF
+ CMPB (SI), $0xFF
+ JNE lit_len_finalise
+
+ // bounds check src[si+1]
+ MOVQ SI, AX
+ ADDQ $1, AX
+ CMPQ AX, R9
+ JGT err_short_buf
+
+ // lit_len += 0xFF
+ ADDQ $0xFF, CX
+ INCQ SI
+ JMP lit_len_loop
+
+lit_len_finalise:
+ // lit_len += int(src[si])
+ // si++
+ MOVBQZX (SI), AX
+ ADDQ AX, CX
+ INCQ SI
+
+copy_literal:
+ // bounds check src and dst
+ MOVQ SI, AX
+ ADDQ CX, AX
+ CMPQ AX, R9
+ JGT err_short_buf
+
+ MOVQ DI, AX
+ ADDQ CX, AX
+ CMPQ AX, R8
+ JGT err_short_buf
+
+ // whats a good cut off to call memmove?
+ CMPQ CX, $16
+ JGT memmove_lit
+
+ // if len(dst[di:]) < 16
+ MOVQ R8, AX
+ SUBQ DI, AX
+ CMPQ AX, $16
+ JLT memmove_lit
+
+ // if len(src[si:]) < 16
+ MOVQ R9, AX
+ SUBQ SI, AX
+ CMPQ AX, $16
+ JLT memmove_lit
+
+ MOVOU (SI), X0
+ MOVOU X0, (DI)
+
+ JMP finish_lit_copy
+
+memmove_lit:
+ // memmove(to, from, len)
+ MOVQ DI, 0(SP)
+ MOVQ SI, 8(SP)
+ MOVQ CX, 16(SP)
+ // spill
+ MOVQ DI, 24(SP)
+ MOVQ SI, 32(SP)
+ MOVQ CX, 40(SP) // need len to inc SI, DI after
+ MOVB DX, 48(SP)
+ CALL runtime·memmove(SB)
+
+ // restore registers
+ MOVQ 24(SP), DI
+ MOVQ 32(SP), SI
+ MOVQ 40(SP), CX
+ MOVB 48(SP), DX
+
+ // recalc initial values
+ MOVQ dst_base+0(FP), R8
+ MOVQ R8, R11
+ ADDQ dst_len+8(FP), R8
+ MOVQ src_base+24(FP), R9
+ ADDQ src_len+32(FP), R9
+ MOVQ R8, R12
+ SUBQ $32, R12
+ MOVQ R9, R13
+ SUBQ $16, R13
+
+finish_lit_copy:
+ ADDQ CX, SI
+ ADDQ CX, DI
+
+ CMPQ SI, R9
+ JGE end
+
+offset:
+ // CX := mLen
+ // free up DX to use for offset
+ MOVQ DX, CX
+
+ MOVQ SI, AX
+ ADDQ $2, AX
+ CMPQ AX, R9
+ JGT err_short_buf
+
+ // offset
+ // DX := int(src[si]) | int(src[si+1])<<8
+ MOVWQZX (SI), DX
+ ADDQ $2, SI
+
+ // 0 offset is invalid
+ CMPQ DX, $0
+ JEQ err_corrupt
+
+ ANDB $0xF, CX
+
+match_len_loop_pre:
+ // if mlen != 0xF
+ CMPB CX, $0xF
+ JNE copy_match
+
+match_len_loop:
+ // for src[si] == 0xFF
+ // lit_len += 0xFF
+ CMPB (SI), $0xFF
+ JNE match_len_finalise
+
+ // bounds check src[si+1]
+ MOVQ SI, AX
+ ADDQ $1, AX
+ CMPQ AX, R9
+ JGT err_short_buf
+
+ ADDQ $0xFF, CX
+ INCQ SI
+ JMP match_len_loop
+
+match_len_finalise:
+ // lit_len += int(src[si])
+ // si++
+ MOVBQZX (SI), AX
+ ADDQ AX, CX
+ INCQ SI
+
+copy_match:
+ // mLen += minMatch
+ ADDQ $4, CX
+
+ // check we have match_len bytes left in dst
+ // di+match_len < len(dst)
+ MOVQ DI, AX
+ ADDQ CX, AX
+ CMPQ AX, R8
+ JGT err_short_buf
+
+ // DX = offset
+ // CX = match_len
+ // BX = &dst + (di - offset)
+ MOVQ DI, BX
+ SUBQ DX, BX
+
+ // check BX is within dst
+ // if BX < &dst
+ CMPQ BX, R11
+ JLT err_short_buf
+
+ // if offset + match_len < di
+ MOVQ BX, AX
+ ADDQ CX, AX
+ CMPQ DI, AX
+ JGT copy_interior_match
+
+ // AX := len(dst[:di])
+ // MOVQ DI, AX
+ // SUBQ R11, AX
+
+ // copy 16 bytes at a time
+ // if di-offset < 16 copy 16-(di-offset) bytes to di
+ // then do the remaining
+
+copy_match_loop:
+ // for match_len >= 0
+ // dst[di] = dst[i]
+ // di++
+ // i++
+ MOVB (BX), AX
+ MOVB AX, (DI)
+ INCQ DI
+ INCQ BX
+ DECQ CX
+
+ CMPQ CX, $0
+ JGT copy_match_loop
+
+ JMP loop
+
+copy_interior_match:
+ CMPQ CX, $16
+ JGT memmove_match
+
+ // if len(dst[di:]) < 16
+ MOVQ R8, AX
+ SUBQ DI, AX
+ CMPQ AX, $16
+ JLT memmove_match
+
+ MOVOU (BX), X0
+ MOVOU X0, (DI)
+
+ ADDQ CX, DI
+ JMP loop
+
+memmove_match:
+ // memmove(to, from, len)
+ MOVQ DI, 0(SP)
+ MOVQ BX, 8(SP)
+ MOVQ CX, 16(SP)
+ // spill
+ MOVQ DI, 24(SP)
+ MOVQ SI, 32(SP)
+ MOVQ CX, 40(SP) // need len to inc SI, DI after
+ CALL runtime·memmove(SB)
+
+ // restore registers
+ MOVQ 24(SP), DI
+ MOVQ 32(SP), SI
+ MOVQ 40(SP), CX
+
+ // recalc initial values
+ MOVQ dst_base+0(FP), R8
+ MOVQ R8, R11 // TODO: make these sensible numbers
+ ADDQ dst_len+8(FP), R8
+ MOVQ src_base+24(FP), R9
+ ADDQ src_len+32(FP), R9
+ MOVQ R8, R12
+ SUBQ $32, R12
+ MOVQ R9, R13
+ SUBQ $16, R13
+
+ ADDQ CX, DI
+ JMP loop
+
+err_corrupt:
+ MOVQ $-1, ret+48(FP)
+ RET
+
+err_short_buf:
+ MOVQ $-2, ret+48(FP)
+ RET
+
+end:
+ SUBQ R11, DI
+ MOVQ DI, ret+48(FP)
+ RET
diff --git a/vendor/github.com/pierrec/lz4/decode_other.go b/vendor/github.com/pierrec/lz4/decode_other.go
new file mode 100644
index 0000000..919888e
--- /dev/null
+++ b/vendor/github.com/pierrec/lz4/decode_other.go
@@ -0,0 +1,98 @@
+// +build !amd64 appengine !gc noasm
+
+package lz4
+
+func decodeBlock(dst, src []byte) (ret int) {
+ const hasError = -2
+ defer func() {
+ if recover() != nil {
+ ret = hasError
+ }
+ }()
+
+ var si, di int
+ for {
+ // Literals and match lengths (token).
+ b := int(src[si])
+ si++
+
+ // Literals.
+ if lLen := b >> 4; lLen > 0 {
+ switch {
+ case lLen < 0xF && si+16 < len(src):
+ // Shortcut 1
+ // if we have enough room in src and dst, and the literals length
+ // is small enough (0..14) then copy all 16 bytes, even if not all
+ // are part of the literals.
+ copy(dst[di:], src[si:si+16])
+ si += lLen
+ di += lLen
+ if mLen := b & 0xF; mLen < 0xF {
+ // Shortcut 2
+ // if the match length (4..18) fits within the literals, then copy
+ // all 18 bytes, even if not all are part of the literals.
+ mLen += 4
+ if offset := int(src[si]) | int(src[si+1])<<8; mLen <= offset {
+ i := di - offset
+ end := i + 18
+ if end > len(dst) {
+ // The remaining buffer may not hold 18 bytes.
+ // See https://github.com/pierrec/lz4/issues/51.
+ end = len(dst)
+ }
+ copy(dst[di:], dst[i:end])
+ si += 2
+ di += mLen
+ continue
+ }
+ }
+ case lLen == 0xF:
+ for src[si] == 0xFF {
+ lLen += 0xFF
+ si++
+ }
+ lLen += int(src[si])
+ si++
+ fallthrough
+ default:
+ copy(dst[di:di+lLen], src[si:si+lLen])
+ si += lLen
+ di += lLen
+ }
+ }
+ if si >= len(src) {
+ return di
+ }
+
+ offset := int(src[si]) | int(src[si+1])<<8
+ if offset == 0 {
+ return hasError
+ }
+ si += 2
+
+ // Match.
+ mLen := b & 0xF
+ if mLen == 0xF {
+ for src[si] == 0xFF {
+ mLen += 0xFF
+ si++
+ }
+ mLen += int(src[si])
+ si++
+ }
+ mLen += minMatch
+
+ // Copy the match.
+ expanded := dst[di-offset:]
+ if mLen > offset {
+ // Efficiently copy the match dst[di-offset:di] into the dst slice.
+ bytesToCopy := offset * (mLen / offset)
+ for n := offset; n <= bytesToCopy+offset; n *= 2 {
+ copy(expanded[n:], expanded[:n])
+ }
+ di += bytesToCopy
+ mLen -= bytesToCopy
+ }
+ di += copy(dst[di:di+mLen], expanded[:mLen])
+ }
+}
diff --git a/vendor/github.com/pierrec/lz4/errors.go b/vendor/github.com/pierrec/lz4/errors.go
new file mode 100644
index 0000000..1c45d18
--- /dev/null
+++ b/vendor/github.com/pierrec/lz4/errors.go
@@ -0,0 +1,30 @@
+package lz4
+
+import (
+ "errors"
+ "fmt"
+ "os"
+ rdebug "runtime/debug"
+)
+
+var (
+ // ErrInvalidSourceShortBuffer is returned by UncompressBlock or CompressBLock when a compressed
+ // block is corrupted or the destination buffer is not large enough for the uncompressed data.
+ ErrInvalidSourceShortBuffer = errors.New("lz4: invalid source or destination buffer too short")
+ // ErrInvalid is returned when reading an invalid LZ4 archive.
+ ErrInvalid = errors.New("lz4: bad magic number")
+ // ErrBlockDependency is returned when attempting to decompress an archive created with block dependency.
+ ErrBlockDependency = errors.New("lz4: block dependency not supported")
+ // ErrUnsupportedSeek is returned when attempting to Seek any way but forward from the current position.
+ ErrUnsupportedSeek = errors.New("lz4: can only seek forward from io.SeekCurrent")
+)
+
+func recoverBlock(e *error) {
+ if r := recover(); r != nil && *e == nil {
+ if debugFlag {
+ fmt.Fprintln(os.Stderr, r)
+ rdebug.PrintStack()
+ }
+ *e = ErrInvalidSourceShortBuffer
+ }
+}
diff --git a/vendor/github.com/pierrec/lz4/internal/xxh32/xxh32zero.go b/vendor/github.com/pierrec/lz4/internal/xxh32/xxh32zero.go
index 850a6fd..7a76a6b 100644
--- a/vendor/github.com/pierrec/lz4/internal/xxh32/xxh32zero.go
+++ b/vendor/github.com/pierrec/lz4/internal/xxh32/xxh32zero.go
@@ -7,14 +7,15 @@
)
const (
- prime32_1 uint32 = 2654435761
- prime32_2 uint32 = 2246822519
- prime32_3 uint32 = 3266489917
- prime32_4 uint32 = 668265263
- prime32_5 uint32 = 374761393
+ prime1 uint32 = 2654435761
+ prime2 uint32 = 2246822519
+ prime3 uint32 = 3266489917
+ prime4 uint32 = 668265263
+ prime5 uint32 = 374761393
- prime32_1plus2 uint32 = 606290984
- prime32_minus1 uint32 = 1640531535
+ primeMask = 0xFFFFFFFF
+ prime1plus2 = uint32((uint64(prime1) + uint64(prime2)) & primeMask) // 606290984
+ prime1minus = uint32((-int64(prime1)) & primeMask) // 1640531535
)
// XXHZero represents an xxhash32 object with seed 0.
@@ -37,10 +38,10 @@
// Reset resets the Hash to its initial state.
func (xxh *XXHZero) Reset() {
- xxh.v1 = prime32_1plus2
- xxh.v2 = prime32_2
+ xxh.v1 = prime1plus2
+ xxh.v2 = prime2
xxh.v3 = 0
- xxh.v4 = prime32_minus1
+ xxh.v4 = prime1minus
xxh.totalLen = 0
xxh.bufused = 0
}
@@ -83,20 +84,20 @@
// fast rotl(13)
buf := xxh.buf[:16] // BCE hint.
- v1 = rol13(v1+binary.LittleEndian.Uint32(buf[:])*prime32_2) * prime32_1
- v2 = rol13(v2+binary.LittleEndian.Uint32(buf[4:])*prime32_2) * prime32_1
- v3 = rol13(v3+binary.LittleEndian.Uint32(buf[8:])*prime32_2) * prime32_1
- v4 = rol13(v4+binary.LittleEndian.Uint32(buf[12:])*prime32_2) * prime32_1
+ v1 = rol13(v1+binary.LittleEndian.Uint32(buf[:])*prime2) * prime1
+ v2 = rol13(v2+binary.LittleEndian.Uint32(buf[4:])*prime2) * prime1
+ v3 = rol13(v3+binary.LittleEndian.Uint32(buf[8:])*prime2) * prime1
+ v4 = rol13(v4+binary.LittleEndian.Uint32(buf[12:])*prime2) * prime1
p = r
xxh.bufused = 0
}
for n := n - 16; p <= n; p += 16 {
sub := input[p:][:16] //BCE hint for compiler
- v1 = rol13(v1+binary.LittleEndian.Uint32(sub[:])*prime32_2) * prime32_1
- v2 = rol13(v2+binary.LittleEndian.Uint32(sub[4:])*prime32_2) * prime32_1
- v3 = rol13(v3+binary.LittleEndian.Uint32(sub[8:])*prime32_2) * prime32_1
- v4 = rol13(v4+binary.LittleEndian.Uint32(sub[12:])*prime32_2) * prime32_1
+ v1 = rol13(v1+binary.LittleEndian.Uint32(sub[:])*prime2) * prime1
+ v2 = rol13(v2+binary.LittleEndian.Uint32(sub[4:])*prime2) * prime1
+ v3 = rol13(v3+binary.LittleEndian.Uint32(sub[8:])*prime2) * prime1
+ v4 = rol13(v4+binary.LittleEndian.Uint32(sub[12:])*prime2) * prime1
}
xxh.v1, xxh.v2, xxh.v3, xxh.v4 = v1, v2, v3, v4
@@ -112,25 +113,25 @@
if h32 >= 16 {
h32 += rol1(xxh.v1) + rol7(xxh.v2) + rol12(xxh.v3) + rol18(xxh.v4)
} else {
- h32 += prime32_5
+ h32 += prime5
}
p := 0
n := xxh.bufused
buf := xxh.buf
for n := n - 4; p <= n; p += 4 {
- h32 += binary.LittleEndian.Uint32(buf[p:p+4]) * prime32_3
- h32 = rol17(h32) * prime32_4
+ h32 += binary.LittleEndian.Uint32(buf[p:p+4]) * prime3
+ h32 = rol17(h32) * prime4
}
for ; p < n; p++ {
- h32 += uint32(buf[p]) * prime32_5
- h32 = rol11(h32) * prime32_1
+ h32 += uint32(buf[p]) * prime5
+ h32 = rol11(h32) * prime1
}
h32 ^= h32 >> 15
- h32 *= prime32_2
+ h32 *= prime2
h32 ^= h32 >> 13
- h32 *= prime32_3
+ h32 *= prime3
h32 ^= h32 >> 16
return h32
@@ -142,19 +143,19 @@
h32 := uint32(n)
if n < 16 {
- h32 += prime32_5
+ h32 += prime5
} else {
- v1 := prime32_1plus2
- v2 := prime32_2
+ v1 := prime1plus2
+ v2 := prime2
v3 := uint32(0)
- v4 := prime32_minus1
+ v4 := prime1minus
p := 0
for n := n - 16; p <= n; p += 16 {
sub := input[p:][:16] //BCE hint for compiler
- v1 = rol13(v1+binary.LittleEndian.Uint32(sub[:])*prime32_2) * prime32_1
- v2 = rol13(v2+binary.LittleEndian.Uint32(sub[4:])*prime32_2) * prime32_1
- v3 = rol13(v3+binary.LittleEndian.Uint32(sub[8:])*prime32_2) * prime32_1
- v4 = rol13(v4+binary.LittleEndian.Uint32(sub[12:])*prime32_2) * prime32_1
+ v1 = rol13(v1+binary.LittleEndian.Uint32(sub[:])*prime2) * prime1
+ v2 = rol13(v2+binary.LittleEndian.Uint32(sub[4:])*prime2) * prime1
+ v3 = rol13(v3+binary.LittleEndian.Uint32(sub[8:])*prime2) * prime1
+ v4 = rol13(v4+binary.LittleEndian.Uint32(sub[12:])*prime2) * prime1
}
input = input[p:]
n -= p
@@ -163,19 +164,19 @@
p := 0
for n := n - 4; p <= n; p += 4 {
- h32 += binary.LittleEndian.Uint32(input[p:p+4]) * prime32_3
- h32 = rol17(h32) * prime32_4
+ h32 += binary.LittleEndian.Uint32(input[p:p+4]) * prime3
+ h32 = rol17(h32) * prime4
}
for p < n {
- h32 += uint32(input[p]) * prime32_5
- h32 = rol11(h32) * prime32_1
+ h32 += uint32(input[p]) * prime5
+ h32 = rol11(h32) * prime1
p++
}
h32 ^= h32 >> 15
- h32 *= prime32_2
+ h32 *= prime2
h32 ^= h32 >> 13
- h32 *= prime32_3
+ h32 *= prime3
h32 ^= h32 >> 16
return h32
@@ -183,12 +184,12 @@
// Uint32Zero hashes x with seed 0.
func Uint32Zero(x uint32) uint32 {
- h := prime32_5 + 4 + x*prime32_3
- h = rol17(h) * prime32_4
+ h := prime5 + 4 + x*prime3
+ h = rol17(h) * prime4
h ^= h >> 15
- h *= prime32_2
+ h *= prime2
h ^= h >> 13
- h *= prime32_3
+ h *= prime3
h ^= h >> 16
return h
}
diff --git a/vendor/github.com/pierrec/lz4/lz4.go b/vendor/github.com/pierrec/lz4/lz4.go
index 3580275..cdbf961 100644
--- a/vendor/github.com/pierrec/lz4/lz4.go
+++ b/vendor/github.com/pierrec/lz4/lz4.go
@@ -30,27 +30,25 @@
// hashLog determines the size of the hash table used to quickly find a previous match position.
// Its value influences the compression speed and memory usage, the lower the faster,
// but at the expense of the compression ratio.
- // 16 seems to be the best compromise.
- hashLog = 16
- hashTableSize = 1 << hashLog
- hashShift = uint((minMatch * 8) - hashLog)
+ // 16 seems to be the best compromise for fast compression.
+ hashLog = 16
+ htSize = 1 << hashLog
- mfLimit = 8 + minMatch // The last match cannot start within the last 12 bytes.
- skipStrength = 6 // variable step for fast scan
+ mfLimit = 8 + minMatch // The last match cannot start within the last 12 bytes.
)
// map the block max size id with its value in bytes: 64Kb, 256Kb, 1Mb and 4Mb.
-var (
- bsMapID = map[byte]int{4: 64 << 10, 5: 256 << 10, 6: 1 << 20, 7: 4 << 20}
- bsMapValue = make(map[int]byte, len(bsMapID))
+const (
+ blockSize64K = 64 << 10
+ blockSize256K = 256 << 10
+ blockSize1M = 1 << 20
+ blockSize4M = 4 << 20
)
-// Reversed.
-func init() {
- for i, v := range bsMapID {
- bsMapValue[v] = i
- }
-}
+var (
+ bsMapID = map[byte]int{4: blockSize64K, 5: blockSize256K, 6: blockSize1M, 7: blockSize4M}
+ bsMapValue = map[int]byte{blockSize64K: 4, blockSize256K: 5, blockSize1M: 6, blockSize4M: 7}
+)
// Header describes the various flags that can be set on a Writer or obtained from a Reader.
// The default values match those of the LZ4 frame format definition
diff --git a/vendor/github.com/pierrec/lz4/reader.go b/vendor/github.com/pierrec/lz4/reader.go
index f08db47..126b792 100644
--- a/vendor/github.com/pierrec/lz4/reader.go
+++ b/vendor/github.com/pierrec/lz4/reader.go
@@ -14,6 +14,9 @@
// The Header may change between Read() calls in case of concatenated frames.
type Reader struct {
Header
+ // Handler called when a block has been successfully read.
+ // It provides the number of bytes read.
+ OnBlockDone func(size int)
buf [8]byte // Scrap buffer.
pos int64 // Current position in src.
@@ -22,6 +25,8 @@
data []byte // Uncompressed data.
idx int // Index of unread bytes into data.
checksum xxh32.XXHZero // Frame hash.
+ skip int64 // Bytes to skip before next read.
+ dpos int64 // Position in dest
}
// NewReader returns a new LZ4 frame decoder.
@@ -76,7 +81,7 @@
return fmt.Errorf("lz4: invalid version: got %d; expected %d", v, Version)
}
if b>>5&1 == 0 {
- return fmt.Errorf("lz4: block dependency not supported")
+ return ErrBlockDependency
}
z.BlockChecksum = b>>4&1 > 0
frameSize := b>>3&1 > 0
@@ -101,7 +106,7 @@
z.data = z.zdata[:cap(z.zdata)][bSize:]
z.idx = len(z.data)
- z.checksum.Write(buf[0:2])
+ _, _ = z.checksum.Write(buf[0:2])
if frameSize {
buf := buf[:8]
@@ -110,7 +115,7 @@
}
z.Size = binary.LittleEndian.Uint64(buf)
z.pos += 8
- z.checksum.Write(buf)
+ _, _ = z.checksum.Write(buf)
}
// Header checksum.
@@ -158,6 +163,9 @@
if debugFlag {
debug("reading block from writer")
}
+ // Reset uncompressed buffer
+ z.data = z.zdata[:cap(z.zdata)][len(z.zdata):]
+
// Block length: 0 = end of frame, highest bit set: uncompressed.
bLen, err := z.readUint32()
if err != nil {
@@ -208,6 +216,9 @@
return 0, err
}
z.pos += int64(bLen)
+ if z.OnBlockDone != nil {
+ z.OnBlockDone(int(bLen))
+ }
if z.BlockChecksum {
checksum, err := z.readUint32()
@@ -252,10 +263,13 @@
return 0, err
}
z.data = z.data[:n]
+ if z.OnBlockDone != nil {
+ z.OnBlockDone(n)
+ }
}
if !z.NoChecksum {
- z.checksum.Write(z.data)
+ _, _ = z.checksum.Write(z.data)
if debugFlag {
debug("current frame checksum %x", z.checksum.Sum32())
}
@@ -263,8 +277,20 @@
z.idx = 0
}
+ if z.skip > int64(len(z.data[z.idx:])) {
+ z.skip -= int64(len(z.data[z.idx:]))
+ z.dpos += int64(len(z.data[z.idx:]))
+ z.idx = len(z.data)
+ return 0, nil
+ }
+
+ z.idx += int(z.skip)
+ z.dpos += z.skip
+ z.skip = 0
+
n := copy(buf, z.data[z.idx:])
z.idx += n
+ z.dpos += int64(n)
if debugFlag {
debug("copied %d bytes to input", n)
}
@@ -272,6 +298,20 @@
return n, nil
}
+// Seek implements io.Seeker, but supports seeking forward from the current
+// position only. Any other seek will return an error. Allows skipping output
+// bytes which aren't needed, which in some scenarios is faster than reading
+// and discarding them.
+// Note this may cause future calls to Read() to read 0 bytes if all of the
+// data they would have returned is skipped.
+func (z *Reader) Seek(offset int64, whence int) (int64, error) {
+ if offset < 0 || whence != io.SeekCurrent {
+ return z.dpos + z.skip, ErrUnsupportedSeek
+ }
+ z.skip += offset
+ return z.dpos + z.skip, nil
+}
+
// Reset discards the Reader's state and makes it equivalent to the
// result of its original state from NewReader, but reading from r instead.
// This permits reusing a Reader rather than allocating a new one.
diff --git a/vendor/github.com/pierrec/lz4/writer.go b/vendor/github.com/pierrec/lz4/writer.go
index 0120438..2cc8d95 100644
--- a/vendor/github.com/pierrec/lz4/writer.go
+++ b/vendor/github.com/pierrec/lz4/writer.go
@@ -11,6 +11,9 @@
// Writer implements the LZ4 frame encoder.
type Writer struct {
Header
+ // Handler called when a block has been successfully written out.
+ // It provides the number of bytes written.
+ OnBlockDone func(size int)
buf [19]byte // magic number(4) + header(flags(2)+[Size(8)+DictID(4)]+checksum(1)) does not exceed 19 bytes
dst io.Writer // Destination.
@@ -43,11 +46,13 @@
}
// Allocate the compressed/uncompressed buffers.
// The compressed buffer cannot exceed the uncompressed one.
- if n := 2 * bSize; cap(z.zdata) < n {
- z.zdata = make([]byte, n, n)
+ if cap(z.zdata) < bSize {
+ // Only allocate if there is not enough capacity.
+ // Allocate both buffers at once.
+ z.zdata = make([]byte, 2*bSize)
}
- z.zdata = z.zdata[:bSize]
- z.data = z.zdata[:cap(z.zdata)][bSize:]
+ z.data = z.zdata[:bSize] // Uncompressed buffer is the first half.
+ z.zdata = z.zdata[:cap(z.zdata)][bSize:] // Compressed buffer is the second half.
z.idx = 0
// Size is optional.
@@ -182,24 +187,26 @@
if err := z.writeUint32(bLen); err != nil {
return err
}
- if _, err := z.dst.Write(zdata); err != nil {
+ written, err := z.dst.Write(zdata)
+ if err != nil {
return err
}
+ if h := z.OnBlockDone; h != nil {
+ h(written)
+ }
- if z.BlockChecksum {
- checksum := xxh32.ChecksumZero(zdata)
+ if !z.BlockChecksum {
if debugFlag {
- debug("block checksum %x", checksum)
+ debug("current frame checksum %x", z.checksum.Sum32())
}
- if err := z.writeUint32(checksum); err != nil {
- return err
- }
+ return nil
}
+ checksum := xxh32.ChecksumZero(zdata)
if debugFlag {
- debug("current frame checksum %x", z.checksum.Sum32())
+ debug("block checksum %x", checksum)
+ defer func() { debug("current frame checksum %x", z.checksum.Sum32()) }()
}
-
- return nil
+ return z.writeUint32(checksum)
}
// Flush flushes any pending compressed data to the underlying writer.
@@ -213,7 +220,11 @@
return nil
}
- return z.compressBlock(z.data[:z.idx])
+ if err := z.compressBlock(z.data[:z.idx]); err != nil {
+ return err
+ }
+ z.idx = 0
+ return nil
}
// Close closes the Writer, flushing any unwritten data to the underlying io.Writer, but does not close the underlying io.Writer.
@@ -223,7 +234,6 @@
return err
}
}
-
if err := z.Flush(); err != nil {
return err
}
@@ -234,16 +244,14 @@
if err := z.writeUint32(0); err != nil {
return err
}
- if !z.NoChecksum {
- checksum := z.checksum.Sum32()
- if debugFlag {
- debug("stream checksum %x", checksum)
- }
- if err := z.writeUint32(checksum); err != nil {
- return err
- }
+ if z.NoChecksum {
+ return nil
}
- return nil
+ checksum := z.checksum.Sum32()
+ if debugFlag {
+ debug("stream checksum %x", checksum)
+ }
+ return z.writeUint32(checksum)
}
// Reset clears the state of the Writer z such that it is equivalent to its
diff --git a/vendor/github.com/rcrowley/go-metrics/.travis.yml b/vendor/github.com/rcrowley/go-metrics/.travis.yml
index 117763e..aead076 100644
--- a/vendor/github.com/rcrowley/go-metrics/.travis.yml
+++ b/vendor/github.com/rcrowley/go-metrics/.travis.yml
@@ -1,13 +1,16 @@
language: go
go:
- - 1.3
- - 1.4
- - 1.5
- - 1.6
- - 1.7
- - 1.8
- - 1.9
+ - "1.3"
+ - "1.4"
+ - "1.5"
+ - "1.6"
+ - "1.7"
+ - "1.8"
+ - "1.9"
+ - "1.10"
+ - "1.11"
+ - "1.12"
script:
- ./validate.sh
diff --git a/vendor/github.com/rcrowley/go-metrics/README.md b/vendor/github.com/rcrowley/go-metrics/README.md
index b7356b5..27ddfee 100644
--- a/vendor/github.com/rcrowley/go-metrics/README.md
+++ b/vendor/github.com/rcrowley/go-metrics/README.md
@@ -157,6 +157,7 @@
Clients are available for the following destinations:
+* AppOptics - https://github.com/ysamlan/go-metrics-appoptics
* Librato - https://github.com/mihasya/go-metrics-librato
* Graphite - https://github.com/cyberdelia/go-metrics-graphite
* InfluxDB - https://github.com/vrischmann/go-metrics-influxdb
@@ -166,3 +167,5 @@
* SignalFX - https://github.com/pascallouisperez/go-metrics-signalfx
* Honeycomb - https://github.com/getspine/go-metrics-honeycomb
* Wavefront - https://github.com/wavefrontHQ/go-metrics-wavefront
+* Open-Falcon - https://github.com/g4zhuj/go-metrics-falcon
+* AWS CloudWatch - [https://github.com/savaki/cloudmetrics](https://github.com/savaki/cloudmetrics)
diff --git a/vendor/github.com/rcrowley/go-metrics/debug.go b/vendor/github.com/rcrowley/go-metrics/debug.go
index 043ccef..179e5aa 100644
--- a/vendor/github.com/rcrowley/go-metrics/debug.go
+++ b/vendor/github.com/rcrowley/go-metrics/debug.go
@@ -2,6 +2,7 @@
import (
"runtime/debug"
+ "sync"
"time"
)
@@ -16,7 +17,8 @@
}
ReadGCStats Timer
}
- gcStats debug.GCStats
+ gcStats debug.GCStats
+ registerDebugMetricsOnce = sync.Once{}
)
// Capture new values for the Go garbage collector statistics exported in
@@ -54,19 +56,21 @@
// debug.GCStats. The metrics are named by their fully-qualified Go symbols,
// i.e. debug.GCStats.PauseTotal.
func RegisterDebugGCStats(r Registry) {
- debugMetrics.GCStats.LastGC = NewGauge()
- debugMetrics.GCStats.NumGC = NewGauge()
- debugMetrics.GCStats.Pause = NewHistogram(NewExpDecaySample(1028, 0.015))
- //debugMetrics.GCStats.PauseQuantiles = NewHistogram(NewExpDecaySample(1028, 0.015))
- debugMetrics.GCStats.PauseTotal = NewGauge()
- debugMetrics.ReadGCStats = NewTimer()
+ registerDebugMetricsOnce.Do(func() {
+ debugMetrics.GCStats.LastGC = NewGauge()
+ debugMetrics.GCStats.NumGC = NewGauge()
+ debugMetrics.GCStats.Pause = NewHistogram(NewExpDecaySample(1028, 0.015))
+ //debugMetrics.GCStats.PauseQuantiles = NewHistogram(NewExpDecaySample(1028, 0.015))
+ debugMetrics.GCStats.PauseTotal = NewGauge()
+ debugMetrics.ReadGCStats = NewTimer()
- r.Register("debug.GCStats.LastGC", debugMetrics.GCStats.LastGC)
- r.Register("debug.GCStats.NumGC", debugMetrics.GCStats.NumGC)
- r.Register("debug.GCStats.Pause", debugMetrics.GCStats.Pause)
- //r.Register("debug.GCStats.PauseQuantiles", debugMetrics.GCStats.PauseQuantiles)
- r.Register("debug.GCStats.PauseTotal", debugMetrics.GCStats.PauseTotal)
- r.Register("debug.ReadGCStats", debugMetrics.ReadGCStats)
+ r.Register("debug.GCStats.LastGC", debugMetrics.GCStats.LastGC)
+ r.Register("debug.GCStats.NumGC", debugMetrics.GCStats.NumGC)
+ r.Register("debug.GCStats.Pause", debugMetrics.GCStats.Pause)
+ //r.Register("debug.GCStats.PauseQuantiles", debugMetrics.GCStats.PauseQuantiles)
+ r.Register("debug.GCStats.PauseTotal", debugMetrics.GCStats.PauseTotal)
+ r.Register("debug.ReadGCStats", debugMetrics.ReadGCStats)
+ })
}
// Allocate an initial slice for gcStats.Pause to avoid allocations during
diff --git a/vendor/github.com/rcrowley/go-metrics/log.go b/vendor/github.com/rcrowley/go-metrics/log.go
index f8074c0..2614a0a 100644
--- a/vendor/github.com/rcrowley/go-metrics/log.go
+++ b/vendor/github.com/rcrowley/go-metrics/log.go
@@ -8,17 +8,37 @@
Printf(format string, v ...interface{})
}
+// Log outputs each metric in the given registry periodically using the given logger.
func Log(r Registry, freq time.Duration, l Logger) {
LogScaled(r, freq, time.Nanosecond, l)
}
-// Output each metric in the given registry periodically using the given
+// LogOnCue outputs each metric in the given registry on demand through the channel
+// using the given logger
+func LogOnCue(r Registry, ch chan interface{}, l Logger) {
+ LogScaledOnCue(r, ch, time.Nanosecond, l)
+}
+
+// LogScaled outputs each metric in the given registry periodically using the given
// logger. Print timings in `scale` units (eg time.Millisecond) rather than nanos.
func LogScaled(r Registry, freq time.Duration, scale time.Duration, l Logger) {
+ ch := make(chan interface{})
+ go func(channel chan interface{}) {
+ for _ = range time.Tick(freq) {
+ channel <- struct{}{}
+ }
+ }(ch)
+ LogScaledOnCue(r, ch, scale, l)
+}
+
+// LogScaledOnCue outputs each metric in the given registry on demand through the channel
+// using the given logger. Print timings in `scale` units (eg time.Millisecond) rather
+// than nanos.
+func LogScaledOnCue(r Registry, ch chan interface{}, scale time.Duration, l Logger) {
du := float64(scale)
duSuffix := scale.String()[1:]
- for _ = range time.Tick(freq) {
+ for _ = range ch {
r.Each(func(name string, i interface{}) {
switch metric := i.(type) {
case Counter:
diff --git a/vendor/github.com/rcrowley/go-metrics/registry.go b/vendor/github.com/rcrowley/go-metrics/registry.go
index b3bab64..a8e6722 100644
--- a/vendor/github.com/rcrowley/go-metrics/registry.go
+++ b/vendor/github.com/rcrowley/go-metrics/registry.go
@@ -64,8 +64,10 @@
// Call the given function for each registered metric.
func (r *StandardRegistry) Each(f func(string, interface{})) {
- for name, i := range r.registered() {
- f(name, i)
+ metrics := r.registered()
+ for i := range metrics {
+ kv := &metrics[i]
+ f(kv.name, kv.value)
}
}
@@ -211,12 +213,20 @@
return nil
}
-func (r *StandardRegistry) registered() map[string]interface{} {
- r.mutex.Lock()
- defer r.mutex.Unlock()
- metrics := make(map[string]interface{}, len(r.metrics))
+type metricKV struct {
+ name string
+ value interface{}
+}
+
+func (r *StandardRegistry) registered() []metricKV {
+ r.mutex.RLock()
+ defer r.mutex.RUnlock()
+ metrics := make([]metricKV, 0, len(r.metrics))
for name, i := range r.metrics {
- metrics[name] = i
+ metrics = append(metrics, metricKV{
+ name: name,
+ value: i,
+ })
}
return metrics
}
diff --git a/vendor/github.com/rcrowley/go-metrics/runtime.go b/vendor/github.com/rcrowley/go-metrics/runtime.go
index 11c6b78..4047ab3 100644
--- a/vendor/github.com/rcrowley/go-metrics/runtime.go
+++ b/vendor/github.com/rcrowley/go-metrics/runtime.go
@@ -3,6 +3,7 @@
import (
"runtime"
"runtime/pprof"
+ "sync"
"time"
)
@@ -49,7 +50,8 @@
numGC uint32
numCgoCalls int64
- threadCreateProfile = pprof.Lookup("threadcreate")
+ threadCreateProfile = pprof.Lookup("threadcreate")
+ registerRuntimeMetricsOnce = sync.Once{}
)
// Capture new values for the Go runtime statistics exported in
@@ -146,67 +148,69 @@
// specifically runtime.MemStats. The runtimeMetrics are named by their
// fully-qualified Go symbols, i.e. runtime.MemStats.Alloc.
func RegisterRuntimeMemStats(r Registry) {
- runtimeMetrics.MemStats.Alloc = NewGauge()
- runtimeMetrics.MemStats.BuckHashSys = NewGauge()
- runtimeMetrics.MemStats.DebugGC = NewGauge()
- runtimeMetrics.MemStats.EnableGC = NewGauge()
- runtimeMetrics.MemStats.Frees = NewGauge()
- runtimeMetrics.MemStats.HeapAlloc = NewGauge()
- runtimeMetrics.MemStats.HeapIdle = NewGauge()
- runtimeMetrics.MemStats.HeapInuse = NewGauge()
- runtimeMetrics.MemStats.HeapObjects = NewGauge()
- runtimeMetrics.MemStats.HeapReleased = NewGauge()
- runtimeMetrics.MemStats.HeapSys = NewGauge()
- runtimeMetrics.MemStats.LastGC = NewGauge()
- runtimeMetrics.MemStats.Lookups = NewGauge()
- runtimeMetrics.MemStats.Mallocs = NewGauge()
- runtimeMetrics.MemStats.MCacheInuse = NewGauge()
- runtimeMetrics.MemStats.MCacheSys = NewGauge()
- runtimeMetrics.MemStats.MSpanInuse = NewGauge()
- runtimeMetrics.MemStats.MSpanSys = NewGauge()
- runtimeMetrics.MemStats.NextGC = NewGauge()
- runtimeMetrics.MemStats.NumGC = NewGauge()
- runtimeMetrics.MemStats.GCCPUFraction = NewGaugeFloat64()
- runtimeMetrics.MemStats.PauseNs = NewHistogram(NewExpDecaySample(1028, 0.015))
- runtimeMetrics.MemStats.PauseTotalNs = NewGauge()
- runtimeMetrics.MemStats.StackInuse = NewGauge()
- runtimeMetrics.MemStats.StackSys = NewGauge()
- runtimeMetrics.MemStats.Sys = NewGauge()
- runtimeMetrics.MemStats.TotalAlloc = NewGauge()
- runtimeMetrics.NumCgoCall = NewGauge()
- runtimeMetrics.NumGoroutine = NewGauge()
- runtimeMetrics.NumThread = NewGauge()
- runtimeMetrics.ReadMemStats = NewTimer()
+ registerRuntimeMetricsOnce.Do(func() {
+ runtimeMetrics.MemStats.Alloc = NewGauge()
+ runtimeMetrics.MemStats.BuckHashSys = NewGauge()
+ runtimeMetrics.MemStats.DebugGC = NewGauge()
+ runtimeMetrics.MemStats.EnableGC = NewGauge()
+ runtimeMetrics.MemStats.Frees = NewGauge()
+ runtimeMetrics.MemStats.HeapAlloc = NewGauge()
+ runtimeMetrics.MemStats.HeapIdle = NewGauge()
+ runtimeMetrics.MemStats.HeapInuse = NewGauge()
+ runtimeMetrics.MemStats.HeapObjects = NewGauge()
+ runtimeMetrics.MemStats.HeapReleased = NewGauge()
+ runtimeMetrics.MemStats.HeapSys = NewGauge()
+ runtimeMetrics.MemStats.LastGC = NewGauge()
+ runtimeMetrics.MemStats.Lookups = NewGauge()
+ runtimeMetrics.MemStats.Mallocs = NewGauge()
+ runtimeMetrics.MemStats.MCacheInuse = NewGauge()
+ runtimeMetrics.MemStats.MCacheSys = NewGauge()
+ runtimeMetrics.MemStats.MSpanInuse = NewGauge()
+ runtimeMetrics.MemStats.MSpanSys = NewGauge()
+ runtimeMetrics.MemStats.NextGC = NewGauge()
+ runtimeMetrics.MemStats.NumGC = NewGauge()
+ runtimeMetrics.MemStats.GCCPUFraction = NewGaugeFloat64()
+ runtimeMetrics.MemStats.PauseNs = NewHistogram(NewExpDecaySample(1028, 0.015))
+ runtimeMetrics.MemStats.PauseTotalNs = NewGauge()
+ runtimeMetrics.MemStats.StackInuse = NewGauge()
+ runtimeMetrics.MemStats.StackSys = NewGauge()
+ runtimeMetrics.MemStats.Sys = NewGauge()
+ runtimeMetrics.MemStats.TotalAlloc = NewGauge()
+ runtimeMetrics.NumCgoCall = NewGauge()
+ runtimeMetrics.NumGoroutine = NewGauge()
+ runtimeMetrics.NumThread = NewGauge()
+ runtimeMetrics.ReadMemStats = NewTimer()
- r.Register("runtime.MemStats.Alloc", runtimeMetrics.MemStats.Alloc)
- r.Register("runtime.MemStats.BuckHashSys", runtimeMetrics.MemStats.BuckHashSys)
- r.Register("runtime.MemStats.DebugGC", runtimeMetrics.MemStats.DebugGC)
- r.Register("runtime.MemStats.EnableGC", runtimeMetrics.MemStats.EnableGC)
- r.Register("runtime.MemStats.Frees", runtimeMetrics.MemStats.Frees)
- r.Register("runtime.MemStats.HeapAlloc", runtimeMetrics.MemStats.HeapAlloc)
- r.Register("runtime.MemStats.HeapIdle", runtimeMetrics.MemStats.HeapIdle)
- r.Register("runtime.MemStats.HeapInuse", runtimeMetrics.MemStats.HeapInuse)
- r.Register("runtime.MemStats.HeapObjects", runtimeMetrics.MemStats.HeapObjects)
- r.Register("runtime.MemStats.HeapReleased", runtimeMetrics.MemStats.HeapReleased)
- r.Register("runtime.MemStats.HeapSys", runtimeMetrics.MemStats.HeapSys)
- r.Register("runtime.MemStats.LastGC", runtimeMetrics.MemStats.LastGC)
- r.Register("runtime.MemStats.Lookups", runtimeMetrics.MemStats.Lookups)
- r.Register("runtime.MemStats.Mallocs", runtimeMetrics.MemStats.Mallocs)
- r.Register("runtime.MemStats.MCacheInuse", runtimeMetrics.MemStats.MCacheInuse)
- r.Register("runtime.MemStats.MCacheSys", runtimeMetrics.MemStats.MCacheSys)
- r.Register("runtime.MemStats.MSpanInuse", runtimeMetrics.MemStats.MSpanInuse)
- r.Register("runtime.MemStats.MSpanSys", runtimeMetrics.MemStats.MSpanSys)
- r.Register("runtime.MemStats.NextGC", runtimeMetrics.MemStats.NextGC)
- r.Register("runtime.MemStats.NumGC", runtimeMetrics.MemStats.NumGC)
- r.Register("runtime.MemStats.GCCPUFraction", runtimeMetrics.MemStats.GCCPUFraction)
- r.Register("runtime.MemStats.PauseNs", runtimeMetrics.MemStats.PauseNs)
- r.Register("runtime.MemStats.PauseTotalNs", runtimeMetrics.MemStats.PauseTotalNs)
- r.Register("runtime.MemStats.StackInuse", runtimeMetrics.MemStats.StackInuse)
- r.Register("runtime.MemStats.StackSys", runtimeMetrics.MemStats.StackSys)
- r.Register("runtime.MemStats.Sys", runtimeMetrics.MemStats.Sys)
- r.Register("runtime.MemStats.TotalAlloc", runtimeMetrics.MemStats.TotalAlloc)
- r.Register("runtime.NumCgoCall", runtimeMetrics.NumCgoCall)
- r.Register("runtime.NumGoroutine", runtimeMetrics.NumGoroutine)
- r.Register("runtime.NumThread", runtimeMetrics.NumThread)
- r.Register("runtime.ReadMemStats", runtimeMetrics.ReadMemStats)
+ r.Register("runtime.MemStats.Alloc", runtimeMetrics.MemStats.Alloc)
+ r.Register("runtime.MemStats.BuckHashSys", runtimeMetrics.MemStats.BuckHashSys)
+ r.Register("runtime.MemStats.DebugGC", runtimeMetrics.MemStats.DebugGC)
+ r.Register("runtime.MemStats.EnableGC", runtimeMetrics.MemStats.EnableGC)
+ r.Register("runtime.MemStats.Frees", runtimeMetrics.MemStats.Frees)
+ r.Register("runtime.MemStats.HeapAlloc", runtimeMetrics.MemStats.HeapAlloc)
+ r.Register("runtime.MemStats.HeapIdle", runtimeMetrics.MemStats.HeapIdle)
+ r.Register("runtime.MemStats.HeapInuse", runtimeMetrics.MemStats.HeapInuse)
+ r.Register("runtime.MemStats.HeapObjects", runtimeMetrics.MemStats.HeapObjects)
+ r.Register("runtime.MemStats.HeapReleased", runtimeMetrics.MemStats.HeapReleased)
+ r.Register("runtime.MemStats.HeapSys", runtimeMetrics.MemStats.HeapSys)
+ r.Register("runtime.MemStats.LastGC", runtimeMetrics.MemStats.LastGC)
+ r.Register("runtime.MemStats.Lookups", runtimeMetrics.MemStats.Lookups)
+ r.Register("runtime.MemStats.Mallocs", runtimeMetrics.MemStats.Mallocs)
+ r.Register("runtime.MemStats.MCacheInuse", runtimeMetrics.MemStats.MCacheInuse)
+ r.Register("runtime.MemStats.MCacheSys", runtimeMetrics.MemStats.MCacheSys)
+ r.Register("runtime.MemStats.MSpanInuse", runtimeMetrics.MemStats.MSpanInuse)
+ r.Register("runtime.MemStats.MSpanSys", runtimeMetrics.MemStats.MSpanSys)
+ r.Register("runtime.MemStats.NextGC", runtimeMetrics.MemStats.NextGC)
+ r.Register("runtime.MemStats.NumGC", runtimeMetrics.MemStats.NumGC)
+ r.Register("runtime.MemStats.GCCPUFraction", runtimeMetrics.MemStats.GCCPUFraction)
+ r.Register("runtime.MemStats.PauseNs", runtimeMetrics.MemStats.PauseNs)
+ r.Register("runtime.MemStats.PauseTotalNs", runtimeMetrics.MemStats.PauseTotalNs)
+ r.Register("runtime.MemStats.StackInuse", runtimeMetrics.MemStats.StackInuse)
+ r.Register("runtime.MemStats.StackSys", runtimeMetrics.MemStats.StackSys)
+ r.Register("runtime.MemStats.Sys", runtimeMetrics.MemStats.Sys)
+ r.Register("runtime.MemStats.TotalAlloc", runtimeMetrics.MemStats.TotalAlloc)
+ r.Register("runtime.NumCgoCall", runtimeMetrics.NumCgoCall)
+ r.Register("runtime.NumGoroutine", runtimeMetrics.NumGoroutine)
+ r.Register("runtime.NumThread", runtimeMetrics.NumThread)
+ r.Register("runtime.ReadMemStats", runtimeMetrics.ReadMemStats)
+ })
}