Update zstd to 1.4.4

thirdparty/README.md

@@ -534,7 +534,7 @@ Files extracted from upstream source:
 ## zstd
 
 - Upstream: https://github.com/facebook/zstd
-- Version: 1.4.3
+- Version: 1.4.4
 - License: BSD-3-Clause
 
 Files extracted from upstream source:

thirdparty/zstd/common/bitstream.h

@@ -164,7 +164,7 @@ MEM_STATIC unsigned BIT_highbit32 (U32 val)
         _BitScanReverse ( &r, val );
         return (unsigned) r;
 #   elif defined(__GNUC__) && (__GNUC__ >= 3)   /* Use GCC Intrinsic */
-        return 31 - __builtin_clz (val);
+        return __builtin_clz (val) ^ 31;
 #   elif defined(__ICCARM__)    /* IAR Intrinsic */
         return 31 - __CLZ(val);
 #   else   /* Software version */
@@ -244,9 +244,9 @@ MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC)
 {
     size_t const nbBytes = bitC->bitPos >> 3;
     assert(bitC->bitPos < sizeof(bitC->bitContainer) * 8);
+    assert(bitC->ptr <= bitC->endPtr);
     MEM_writeLEST(bitC->ptr, bitC->bitContainer);
     bitC->ptr += nbBytes;
-    assert(bitC->ptr <= bitC->endPtr);
     bitC->bitPos &= 7;
     bitC->bitContainer >>= nbBytes*8;
 }
@@ -260,6 +260,7 @@ MEM_STATIC void BIT_flushBits(BIT_CStream_t* bitC)
 {
     size_t const nbBytes = bitC->bitPos >> 3;
     assert(bitC->bitPos < sizeof(bitC->bitContainer) * 8);
+    assert(bitC->ptr <= bitC->endPtr);
     MEM_writeLEST(bitC->ptr, bitC->bitContainer);
     bitC->ptr += nbBytes;
     if (bitC->ptr > bitC->endPtr) bitC->ptr = bitC->endPtr;

thirdparty/zstd/common/compiler.h

@@ -61,6 +61,13 @@
 #  define HINT_INLINE static INLINE_KEYWORD FORCE_INLINE_ATTR
 #endif
 
+/* UNUSED_ATTR tells the compiler it is okay if the function is unused. */
+#if defined(__GNUC__)
+#  define UNUSED_ATTR __attribute__((unused))
+#else
+#  define UNUSED_ATTR
+#endif
+
 /* force no inlining */
 #ifdef _MSC_VER
 #  define FORCE_NOINLINE static __declspec(noinline)
@@ -127,9 +134,14 @@
     }                                     \
 }
 
-/* vectorization */
+/* vectorization
+ * older GCC (pre gcc-4.3 picked as the cutoff) uses a different syntax */
 #if !defined(__clang__) && defined(__GNUC__)
-#  define DONT_VECTORIZE __attribute__((optimize("no-tree-vectorize")))
+#  if (__GNUC__ == 4 && __GNUC_MINOR__ > 3) || (__GNUC__ >= 5)
+#    define DONT_VECTORIZE __attribute__((optimize("no-tree-vectorize")))
+#  else
+#    define DONT_VECTORIZE _Pragma("GCC optimize(\"no-tree-vectorize\")")
+#  endif
 #else
 #  define DONT_VECTORIZE
 #endif

thirdparty/zstd/common/fse.h

@@ -308,7 +308,7 @@ If there is an error, the function will return an error code, which can be teste
 *******************************************/
 /* FSE buffer bounds */
 #define FSE_NCOUNTBOUND 512
-#define FSE_BLOCKBOUND(size) (size + (size>>7))
+#define FSE_BLOCKBOUND(size) (size + (size>>7) + 4 /* fse states */ + sizeof(size_t) /* bitContainer */)
 #define FSE_COMPRESSBOUND(size) (FSE_NCOUNTBOUND + FSE_BLOCKBOUND(size))   /* Macro version, useful for static allocation */
 
 /* It is possible to statically allocate FSE CTable/DTable as a table of FSE_CTable/FSE_DTable using below macros */

thirdparty/zstd/common/fse_decompress.c

@@ -52,7 +52,9 @@
 #define FSE_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c)   /* use only *after* variable declarations */
 
 /* check and forward error code */
+#ifndef CHECK_F
 #define CHECK_F(f) { size_t const e = f; if (FSE_isError(e)) return e; }
+#endif
 
 
 /* **************************************************************

thirdparty/zstd/common/mem.h

@@ -47,6 +47,79 @@ extern "C" {
 #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)); }
 
+/* detects whether we are being compiled under msan */
+#if defined (__has_feature)
+#  if __has_feature(memory_sanitizer)
+#    define MEMORY_SANITIZER 1
+#  endif
+#endif
+
+#if defined (MEMORY_SANITIZER)
+/* Not all platforms that support msan provide sanitizers/msan_interface.h.
+ * We therefore declare the functions we need ourselves, rather than trying to
+ * include the header file... */
+
+#include <stdint.h> /* intptr_t */
+
+/* Make memory region fully initialized (without changing its contents). */
+void __msan_unpoison(const volatile void *a, size_t size);
+
+/* Make memory region fully uninitialized (without changing its contents).
+   This is a legacy interface that does not update origin information. Use
+   __msan_allocated_memory() instead. */
+void __msan_poison(const volatile void *a, size_t size);
+
+/* Returns the offset of the first (at least partially) poisoned byte in the
+   memory range, or -1 if the whole range is good. */
+intptr_t __msan_test_shadow(const volatile void *x, size_t size);
+#endif
+
+/* detects whether we are being compiled under asan */
+#if defined (__has_feature)
+#  if __has_feature(address_sanitizer)
+#    define ADDRESS_SANITIZER 1
+#  endif
+#elif defined(__SANITIZE_ADDRESS__)
+#  define ADDRESS_SANITIZER 1
+#endif
+
+#if defined (ADDRESS_SANITIZER)
+/* Not all platforms that support asan provide sanitizers/asan_interface.h.
+ * We therefore declare the functions we need ourselves, rather than trying to
+ * include the header file... */
+
+/**
+ * Marks a memory region (<c>[addr, addr+size)</c>) as unaddressable.
+ *
+ * This memory must be previously allocated by your program. Instrumented
+ * code is forbidden from accessing addresses in this region until it is
+ * unpoisoned. This function is not guaranteed to poison the entire region -
+ * it could poison only a subregion of <c>[addr, addr+size)</c> due to ASan
+ * alignment restrictions.
+ *
+ * \note This function is not thread-safe because no two threads can poison or
+ * unpoison memory in the same memory region simultaneously.
+ *
+ * \param addr Start of memory region.
+ * \param size Size of memory region. */
+void __asan_poison_memory_region(void const volatile *addr, size_t size);
+
+/**
+ * Marks a memory region (<c>[addr, addr+size)</c>) as addressable.
+ *
+ * This memory must be previously allocated by your program. Accessing
+ * addresses in this region is allowed until this region is poisoned again.
+ * This function could unpoison a super-region of <c>[addr, addr+size)</c> due
+ * to ASan alignment restrictions.
+ *
+ * \note This function is not thread-safe because no two threads can
+ * poison or unpoison memory in the same memory region simultaneously.
+ *
+ * \param addr Start of memory region.
+ * \param size Size of memory region. */
+void __asan_unpoison_memory_region(void const volatile *addr, size_t size);
+#endif
+
 
 /*-**************************************************************
 *  Basic Types

thirdparty/zstd/common/pool.c

@@ -127,9 +127,13 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
     ctx->queueTail = 0;
     ctx->numThreadsBusy = 0;
     ctx->queueEmpty = 1;
-    (void)ZSTD_pthread_mutex_init(&ctx->queueMutex, NULL);
-    (void)ZSTD_pthread_cond_init(&ctx->queuePushCond, NULL);
-    (void)ZSTD_pthread_cond_init(&ctx->queuePopCond, NULL);
+    {
+        int error = 0;
+        error |= ZSTD_pthread_mutex_init(&ctx->queueMutex, NULL);
+        error |= ZSTD_pthread_cond_init(&ctx->queuePushCond, NULL);
+        error |= ZSTD_pthread_cond_init(&ctx->queuePopCond, NULL);
+        if (error) { POOL_free(ctx); return NULL; }
+    }
     ctx->shutdown = 0;
     /* Allocate space for the thread handles */
     ctx->threads = (ZSTD_pthread_t*)ZSTD_malloc(numThreads * sizeof(ZSTD_pthread_t), customMem);

thirdparty/zstd/common/threading.c

@@ -14,6 +14,8 @@
  * This file will hold wrapper for systems, which do not support pthreads
  */
 
+#include "threading.h"
+
 /* create fake symbol to avoid empty translation unit warning */
 int g_ZSTD_threading_useless_symbol;
 
@@ -28,7 +30,6 @@ int g_ZSTD_threading_useless_symbol;
 /* ===  Dependencies  === */
 #include <process.h>
 #include <errno.h>
-#include "threading.h"
 
 
 /* ===  Implementation  === */
@@ -73,3 +74,47 @@ int ZSTD_pthread_join(ZSTD_pthread_t thread, void **value_ptr)
 }
 
 #endif   /* ZSTD_MULTITHREAD */
+
+#if defined(ZSTD_MULTITHREAD) && DEBUGLEVEL >= 1 && !defined(_WIN32)
+
+#include <stdlib.h>
+
+int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr)
+{
+    *mutex = (pthread_mutex_t*)malloc(sizeof(pthread_mutex_t));
+    if (!*mutex)
+        return 1;
+    return pthread_mutex_init(*mutex, attr);
+}
+
+int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex)
+{
+    if (!*mutex)
+        return 0;
+    {
+        int const ret = pthread_mutex_destroy(*mutex);
+        free(*mutex);
+        return ret;
+    }
+}
+
+int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr)
+{
+    *cond = (pthread_cond_t*)malloc(sizeof(pthread_cond_t));
+    if (!*cond)
+        return 1;
+    return pthread_cond_init(*cond, attr);
+}
+
+int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond)
+{
+    if (!*cond)
+        return 0;
+    {
+        int const ret = pthread_cond_destroy(*cond);
+        free(*cond);
+        return ret;
+    }
+}
+
+#endif

thirdparty/zstd/common/threading.h

@@ -13,6 +13,8 @@
 #ifndef THREADING_H_938743
 #define THREADING_H_938743
 
+#include "debug.h"
+
 #if defined (__cplusplus)
 extern "C" {
 #endif
@@ -75,10 +77,12 @@ int ZSTD_pthread_join(ZSTD_pthread_t thread, void** value_ptr);
  */
 
 
-#elif defined(ZSTD_MULTITHREAD)   /* posix assumed ; need a better detection method */
+#elif defined(ZSTD_MULTITHREAD)    /* posix assumed ; need a better detection method */
 /* ===   POSIX Systems   === */
 #  include <pthread.h>
 
+#if DEBUGLEVEL < 1
+
 #define ZSTD_pthread_mutex_t            pthread_mutex_t
 #define ZSTD_pthread_mutex_init(a, b)   pthread_mutex_init((a), (b))
 #define ZSTD_pthread_mutex_destroy(a)   pthread_mutex_destroy((a))
@@ -96,6 +100,33 @@ int ZSTD_pthread_join(ZSTD_pthread_t thread, void** value_ptr);
 #define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
 #define ZSTD_pthread_join(a, b)         pthread_join((a),(b))
 
+#else /* DEBUGLEVEL >= 1 */
+
+/* Debug implementation of threading.
+ * In this implementation we use pointers for mutexes and condition variables.
+ * This way, if we forget to init/destroy them the program will crash or ASAN
+ * will report leaks.
+ */
+
+#define ZSTD_pthread_mutex_t            pthread_mutex_t*
+int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr);
+int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex);
+#define ZSTD_pthread_mutex_lock(a)      pthread_mutex_lock(*(a))
+#define ZSTD_pthread_mutex_unlock(a)    pthread_mutex_unlock(*(a))
+
+#define ZSTD_pthread_cond_t             pthread_cond_t*
+int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr);
+int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond);
+#define ZSTD_pthread_cond_wait(a, b)    pthread_cond_wait(*(a), *(b))
+#define ZSTD_pthread_cond_signal(a)     pthread_cond_signal(*(a))
+#define ZSTD_pthread_cond_broadcast(a)  pthread_cond_broadcast(*(a))
+
+#define ZSTD_pthread_t                  pthread_t
+#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
+#define ZSTD_pthread_join(a, b)         pthread_join((a),(b))
+
+#endif
+
 #else  /* ZSTD_MULTITHREAD not defined */
 /* No multithreading support */
 

thirdparty/zstd/common/zstd_internal.h

@@ -197,79 +197,56 @@ static void ZSTD_copy8(void* dst, const void* src) { memcpy(dst, src, 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
+#define WILDCOPY_OVERLENGTH 32
+#define WILDCOPY_VECLEN 16
 
 typedef enum {
     ZSTD_no_overlap,
-    ZSTD_overlap_src_before_dst,
+    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) */
+ *  Custom version of memcpy(), can over read/write up to WILDCOPY_OVERLENGTH bytes (if length==0)
+ *  @param ovtype controls the overlap detection
+ *         - ZSTD_no_overlap: The source and destination are guaranteed to be at least WILDCOPY_VECLEN bytes apart.
+ *         - ZSTD_overlap_src_before_dst: The src and dst may overlap, but they MUST be at least 8 bytes apart.
+ *           The src buffer must be before the dst buffer.
+ */
 MEM_STATIC FORCE_INLINE_ATTR DONT_VECTORIZE
-void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e ovtype)
+void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e const 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(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 {
+    assert(diff >= 8 || (ovtype == ZSTD_no_overlap && diff <= -WILDCOPY_VECLEN));
+
+    if (ovtype == ZSTD_overlap_src_before_dst && diff < WILDCOPY_VECLEN) {
+        /* Handle short offset copies. */
+        do {
+            COPY8(op, ip)
+        } while (op < oend);
+    } else {
+        assert(diff >= WILDCOPY_VECLEN || diff <= -WILDCOPY_VECLEN);
+        /* Separate out the first two COPY16() calls because the copy length is
+         * almost certain to be short, so the branches have different
+         * probabilities.
+         * On gcc-9 unrolling once is +1.6%, twice is +2%, thrice is +1.8%.
+         * On clang-8 unrolling once is +1.4%, twice is +3.3%, thrice is +3%.
+         */
         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);
+        if (op >= oend) return;
+        do {
+            COPY16(op, ip);
+            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 */
-{
-    const BYTE* ip = (const BYTE*)src;
-    BYTE* op = (BYTE*)dst;
-    BYTE* const oend = (BYTE*)dstEnd;
-    do
-        COPY8(op, ip)
-    while (op < oend);
-}
-
 
 /*-*******************************************
 *  Private declarations
@@ -323,7 +300,7 @@ MEM_STATIC U32 ZSTD_highbit32(U32 val)   /* compress, dictBuilder, decodeCorpus
         _BitScanReverse(&r, val);
         return (unsigned)r;
 #   elif defined(__GNUC__) && (__GNUC__ >= 3)   /* GCC Intrinsic */
-        return 31 - __builtin_clz(val);
+        return __builtin_clz (val) ^ 31;
 #   elif defined(__ICCARM__)    /* IAR Intrinsic */
         return 31 - __CLZ(val);
 #   else   /* Software version */

thirdparty/zstd/compress/zstd_compress_internal.h

@@ -19,6 +19,7 @@
 *  Dependencies
 ***************************************/
 #include "zstd_internal.h"
+#include "zstd_cwksp.h"
 #ifdef ZSTD_MULTITHREAD
 #  include "zstdmt_compress.h"
 #endif
@@ -192,6 +193,13 @@ typedef struct {
   size_t capacity; /* The capacity starting from `seq` pointer */
 } rawSeqStore_t;
 
+typedef struct {
+    int collectSequences;
+    ZSTD_Sequence* seqStart;
+    size_t seqIndex;
+    size_t maxSequences;
+} SeqCollector;
+
 struct ZSTD_CCtx_params_s {
     ZSTD_format_e format;
     ZSTD_compressionParameters cParams;
@@ -203,6 +211,9 @@ struct ZSTD_CCtx_params_s {
     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 */
+    int srcSizeHint;           /* User's best guess of source size.
+                                * Hint is not valid when srcSizeHint == 0.
+                                * There is no guarantee that hint is close to actual source size */
 
     ZSTD_dictAttachPref_e attachDictPref;
     ZSTD_literalCompressionMode_e literalCompressionMode;
@@ -228,9 +239,7 @@ struct ZSTD_CCtx_s {
     ZSTD_CCtx_params appliedParams;
     U32   dictID;
 
-    int workSpaceOversizedDuration;
-    void* workSpace;
-    size_t workSpaceSize;
+    ZSTD_cwksp workspace; /* manages buffer for dynamic allocations */
     size_t blockSize;
     unsigned long long pledgedSrcSizePlusOne;  /* this way, 0 (default) == unknown */
     unsigned long long consumedSrcSize;
@@ -238,6 +247,8 @@ struct ZSTD_CCtx_s {
     XXH64_state_t xxhState;
     ZSTD_customMem customMem;
     size_t staticSize;
+    SeqCollector seqCollector;
+    int isFirstBlock;
 
     seqStore_t seqStore;      /* sequences storage ptrs */
     ldmState_t ldmState;      /* long distance matching state */
@@ -337,26 +348,57 @@ MEM_STATIC size_t ZSTD_minGain(size_t srcSize, ZSTD_strategy strat)
     return (srcSize >> minlog) + 2;
 }
 
+/*! ZSTD_safecopyLiterals() :
+ *  memcpy() function that won't read beyond more than WILDCOPY_OVERLENGTH bytes past ilimit_w.
+ *  Only called when the sequence ends past ilimit_w, so it only needs to be optimized for single
+ *  large copies.
+ */
+static void ZSTD_safecopyLiterals(BYTE* op, BYTE const* ip, BYTE const* const iend, BYTE const* ilimit_w) {
+    assert(iend > ilimit_w);
+    if (ip <= ilimit_w) {
+        ZSTD_wildcopy(op, ip, ilimit_w - ip, ZSTD_no_overlap);
+        op += ilimit_w - ip;
+        ip = ilimit_w;
+    }
+    while (ip < iend) *op++ = *ip++;
+}
+
 /*! ZSTD_storeSeq() :
- *  Store a sequence (literal length, literals, offset code and match length code) into seqStore_t.
- *  `offsetCode` : distance to match + 3 (values 1-3 are repCodes).
+ *  Store a sequence (litlen, litPtr, offCode and mlBase) into seqStore_t.
+ *  `offCode` : distance to match + ZSTD_REP_MOVE (values <= ZSTD_REP_MOVE are repCodes).
  *  `mlBase` : matchLength - MINMATCH
+ *  Allowed to overread literals up to litLimit.
 */
-MEM_STATIC void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const void* literals, U32 offsetCode, size_t mlBase)
+HINT_INLINE UNUSED_ATTR
+void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const BYTE* literals, const BYTE* litLimit, U32 offCode, size_t mlBase)
 {
+    BYTE const* const litLimit_w = litLimit - WILDCOPY_OVERLENGTH;
+    BYTE const* const litEnd = literals + litLength;
 #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%4u bytes at offCode%7u",
-               pos, (U32)litLength, (U32)mlBase+MINMATCH, (U32)offsetCode);
+               pos, (U32)litLength, (U32)mlBase+MINMATCH, (U32)offCode);
     }
 #endif
     assert((size_t)(seqStorePtr->sequences - seqStorePtr->sequencesStart) < seqStorePtr->maxNbSeq);
     /* copy Literals */
     assert(seqStorePtr->maxNbLit <= 128 KB);
     assert(seqStorePtr->lit + litLength <= seqStorePtr->litStart + seqStorePtr->maxNbLit);
-    ZSTD_wildcopy(seqStorePtr->lit, literals, (ptrdiff_t)litLength, ZSTD_no_overlap);
+    assert(literals + litLength <= litLimit);
+    if (litEnd <= litLimit_w) {
+        /* Common case we can use wildcopy.
+	 * First copy 16 bytes, because literals are likely short.
+	 */
+        assert(WILDCOPY_OVERLENGTH >= 16);
+        ZSTD_copy16(seqStorePtr->lit, literals);
+        if (litLength > 16) {
+            ZSTD_wildcopy(seqStorePtr->lit+16, literals+16, (ptrdiff_t)litLength-16, ZSTD_no_overlap);
+        }
+    } else {
+        ZSTD_safecopyLiterals(seqStorePtr->lit, literals, litEnd, litLimit_w);
+    }
     seqStorePtr->lit += litLength;
 
     /* literal Length */
@@ -368,7 +410,7 @@ MEM_STATIC void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const v
     seqStorePtr->sequences[0].litLength = (U16)litLength;
 
     /* match offset */
-    seqStorePtr->sequences[0].offset = offsetCode + 1;
+    seqStorePtr->sequences[0].offset = offCode + 1;
 
     /* match Length */
     if (mlBase>0xFFFF) {
@@ -910,7 +952,7 @@ ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams(
 size_t ZSTD_initCStream_internal(ZSTD_CStream* zcs,
                      const void* dict, size_t dictSize,
                      const ZSTD_CDict* cdict,
-                     ZSTD_CCtx_params  params, unsigned long long pledgedSrcSize);
+                     const ZSTD_CCtx_params* params, unsigned long long pledgedSrcSize);
 
 void ZSTD_resetSeqStore(seqStore_t* ssPtr);
 
@@ -925,7 +967,7 @@ size_t ZSTD_compressBegin_advanced_internal(ZSTD_CCtx* cctx,
                                     ZSTD_dictContentType_e dictContentType,
                                     ZSTD_dictTableLoadMethod_e dtlm,
                                     const ZSTD_CDict* cdict,
-                                    ZSTD_CCtx_params params,
+                                    const ZSTD_CCtx_params* params,
                                     unsigned long long pledgedSrcSize);
 
 /* ZSTD_compress_advanced_internal() :
@@ -934,7 +976,7 @@ size_t ZSTD_compress_advanced_internal(ZSTD_CCtx* cctx,
                                        void* dst, size_t dstCapacity,
                                  const void* src, size_t srcSize,
                                  const void* dict,size_t dictSize,
-                                 ZSTD_CCtx_params params);
+                                 const ZSTD_CCtx_params* params);
 
 
 /* ZSTD_writeLastEmptyBlock() :

thirdparty/zstd/compress/zstd_compress_literals.c

@@ -70,7 +70,7 @@ size_t ZSTD_compressLiterals (ZSTD_hufCTables_t const* prevHuf,
                               ZSTD_strategy strategy, int disableLiteralCompression,
                               void* dst, size_t dstCapacity,
                         const void* src, size_t srcSize,
-                              void* workspace, size_t wkspSize,
+                              void* entropyWorkspace, size_t entropyWorkspaceSize,
                         const int bmi2)
 {
     size_t const minGain = ZSTD_minGain(srcSize, strategy);
@@ -99,10 +99,15 @@ size_t ZSTD_compressLiterals (ZSTD_hufCTables_t const* prevHuf,
     {   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, wkspSize, (HUF_CElt*)nextHuf->CTable, &repeat, preferRepeat, bmi2)
-                                : HUF_compress4X_repeat(ostart+lhSize, dstCapacity-lhSize, src, srcSize, 255, 11,
-                                      workspace, wkspSize, (HUF_CElt*)nextHuf->CTable, &repeat, preferRepeat, bmi2);
+        cLitSize = singleStream ?
+            HUF_compress1X_repeat(
+                ostart+lhSize, dstCapacity-lhSize, src, srcSize,
+                255, 11, entropyWorkspace, entropyWorkspaceSize,
+                (HUF_CElt*)nextHuf->CTable, &repeat, preferRepeat, bmi2) :
+            HUF_compress4X_repeat(
+                ostart+lhSize, dstCapacity-lhSize, src, srcSize,
+                255, 11, entropyWorkspace, entropyWorkspaceSize,
+                (HUF_CElt*)nextHuf->CTable, &repeat, preferRepeat, bmi2);
         if (repeat != HUF_repeat_none) {
             /* reused the existing table */
             hType = set_repeat;

thirdparty/zstd/compress/zstd_compress_literals.h

@@ -23,7 +23,7 @@ size_t ZSTD_compressLiterals (ZSTD_hufCTables_t const* prevHuf,
                               ZSTD_strategy strategy, int disableLiteralCompression,
                               void* dst, size_t dstCapacity,
                         const void* src, size_t srcSize,
-                              void* workspace, size_t wkspSize,
+                              void* entropyWorkspace, size_t entropyWorkspaceSize,
                         const int bmi2);
 
 #endif /* ZSTD_COMPRESS_LITERALS_H */

thirdparty/zstd/compress/zstd_compress_sequences.c

@@ -222,7 +222,7 @@ ZSTD_buildCTable(void* dst, size_t dstCapacity,
                 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)
+                void* entropyWorkspace, size_t entropyWorkspaceSize)
 {
     BYTE* op = (BYTE*)dst;
     const BYTE* const oend = op + dstCapacity;
@@ -238,7 +238,7 @@ ZSTD_buildCTable(void* dst, size_t dstCapacity,
         memcpy(nextCTable, prevCTable, prevCTableSize);
         return 0;
     case set_basic:
-        FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, defaultNorm, defaultMax, defaultNormLog, workspace, workspaceSize));  /* note : could be pre-calculated */
+        FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, defaultNorm, defaultMax, defaultNormLog, entropyWorkspace, entropyWorkspaceSize));  /* note : could be pre-calculated */
         return 0;
     case set_compressed: {
         S16 norm[MaxSeq + 1];
@@ -252,7 +252,7 @@ ZSTD_buildCTable(void* dst, size_t dstCapacity,
         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 */
             FORWARD_IF_ERROR(NCountSize);
-            FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, norm, max, tableLog, workspace, workspaceSize));
+            FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, norm, max, tableLog, entropyWorkspace, entropyWorkspaceSize));
             return NCountSize;
         }
     }

thirdparty/zstd/compress/zstd_compress_sequences.h

@@ -35,7 +35,7 @@ ZSTD_buildCTable(void* dst, size_t dstCapacity,
                 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);
+                void* entropyWorkspace, size_t entropyWorkspaceSize);
 
 size_t ZSTD_encodeSequences(
             void* dst, size_t dstCapacity,

thirdparty/zstd/compress/zstd_double_fast.c

@@ -148,7 +148,7 @@ size_t ZSTD_compressBlock_doubleFast_generic(
             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);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
             goto _match_stored;
         }
 
@@ -157,7 +157,7 @@ size_t ZSTD_compressBlock_doubleFast_generic(
           && ((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, (size_t)(ip-anchor), anchor, 0, mLength-MINMATCH);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
             goto _match_stored;
         }
 
@@ -247,7 +247,7 @@ _match_found:
         offset_2 = offset_1;
         offset_1 = offset;
 
-        ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+        ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
 
 _match_stored:
         /* match found */
@@ -278,7 +278,7 @@ _match_stored:
                         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);
+                        ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, repLength2-MINMATCH);
                         hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
                         hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
                         ip += repLength2;
@@ -297,7 +297,7 @@ _match_stored:
                     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);
+                    ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, rLength-MINMATCH);
                     ip += rLength;
                     anchor = ip;
                     continue;   /* faster when present ... (?) */
@@ -411,7 +411,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
             const BYTE* repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
             mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
             ip++;
-            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, 0, mLength-MINMATCH);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
         } else {
             if ((matchLongIndex > dictStartIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) {
                 const BYTE* const matchEnd = matchLongIndex < prefixStartIndex ? dictEnd : iend;
@@ -422,7 +422,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
                 while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; }   /* catch up */
                 offset_2 = offset_1;
                 offset_1 = offset;
-                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
 
             } else if ((matchIndex > dictStartIndex) && (MEM_read32(match) == MEM_read32(ip))) {
                 size_t const h3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
@@ -447,7 +447,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
                 }
                 offset_2 = offset_1;
                 offset_1 = offset;
-                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
 
             } else {
                 ip += ((ip-anchor) >> kSearchStrength) + 1;
@@ -479,7 +479,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
                     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);
+                    ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, repLength2-MINMATCH);
                     hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
                     hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
                     ip += repLength2;

thirdparty/zstd/compress/zstd_lazy.c

@@ -810,7 +810,7 @@ ZSTD_compressBlock_lazy_generic(
         /* store sequence */
 _storeSequence:
         {   size_t const litLength = start - anchor;
-            ZSTD_storeSeq(seqStore, litLength, anchor, (U32)offset, matchLength-MINMATCH);
+            ZSTD_storeSeq(seqStore, litLength, anchor, iend, (U32)offset, matchLength-MINMATCH);
             anchor = ip = start + matchLength;
         }
 
@@ -828,7 +828,7 @@ _storeSequence:
                     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);
+                    ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, matchLength-MINMATCH);
                     ip += matchLength;
                     anchor = ip;
                     continue;
@@ -843,7 +843,7 @@ _storeSequence:
                 /* 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);
+                ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, matchLength-MINMATCH);
                 ip += matchLength;
                 anchor = ip;
                 continue;   /* faster when present ... (?) */
@@ -1051,7 +1051,7 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
         /* store sequence */
 _storeSequence:
         {   size_t const litLength = start - anchor;
-            ZSTD_storeSeq(seqStore, litLength, anchor, (U32)offset, matchLength-MINMATCH);
+            ZSTD_storeSeq(seqStore, litLength, anchor, iend, (U32)offset, matchLength-MINMATCH);
             anchor = ip = start + matchLength;
         }
 
@@ -1066,7 +1066,7 @@ _storeSequence:
                 const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
                 matchLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd, prefixStart) + 4;
                 offset = offset_2; offset_2 = offset_1; offset_1 = (U32)offset;   /* swap offset history */
-                ZSTD_storeSeq(seqStore, 0, anchor, 0, matchLength-MINMATCH);
+                ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, matchLength-MINMATCH);
                 ip += matchLength;
                 anchor = ip;
                 continue;   /* faster when present ... (?) */

thirdparty/zstd/compress/zstd_ldm.c

@@ -49,9 +49,9 @@ size_t ZSTD_ldm_getTableSize(ldmParams_t params)
 {
     size_t const ldmHSize = ((size_t)1) << params.hashLog;
     size_t const ldmBucketSizeLog = MIN(params.bucketSizeLog, params.hashLog);
-    size_t const ldmBucketSize =
-        ((size_t)1) << (params.hashLog - ldmBucketSizeLog);
-    size_t const totalSize = ldmBucketSize + ldmHSize * sizeof(ldmEntry_t);
+    size_t const ldmBucketSize = ((size_t)1) << (params.hashLog - ldmBucketSizeLog);
+    size_t const totalSize = ZSTD_cwksp_alloc_size(ldmBucketSize)
+                           + ZSTD_cwksp_alloc_size(ldmHSize * sizeof(ldmEntry_t));
     return params.enableLdm ? totalSize : 0;
 }
 
@@ -583,7 +583,7 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
                 rep[i] = rep[i-1];
             rep[0] = sequence.offset;
             /* Store the sequence */
-            ZSTD_storeSeq(seqStore, newLitLength, ip - newLitLength,
+            ZSTD_storeSeq(seqStore, newLitLength, ip - newLitLength, iend,
                           sequence.offset + ZSTD_REP_MOVE,
                           sequence.matchLength - MINMATCH);
             ip += sequence.matchLength;

thirdparty/zstd/compress/zstd_opt.c

@@ -1098,7 +1098,7 @@ _shortestPath:   /* cur, last_pos, best_mlen, best_off have to be set */
 
                     assert(anchor + llen <= iend);
                     ZSTD_updateStats(optStatePtr, llen, anchor, offCode, mlen);
-                    ZSTD_storeSeq(seqStore, llen, anchor, offCode, mlen-MINMATCH);
+                    ZSTD_storeSeq(seqStore, llen, anchor, iend, offCode, mlen-MINMATCH);
                     anchor += advance;
                     ip = anchor;
             }   }

thirdparty/zstd/compress/zstdmt_compress.c

@@ -668,7 +668,7 @@ static void ZSTDMT_compressionJob(void* jobDescription)
 
     /* init */
     if (job->cdict) {
-        size_t const initError = ZSTD_compressBegin_advanced_internal(cctx, NULL, 0, ZSTD_dct_auto, ZSTD_dtlm_fast, 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_ERROR(initError);
     } else {  /* srcStart points at reloaded section */
@@ -680,7 +680,7 @@ static void ZSTDMT_compressionJob(void* jobDescription)
                                         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);
+                                        &jobParams, pledgedSrcSize);
             if (ZSTD_isError(initError)) JOB_ERROR(initError);
     }   }
 
@@ -927,12 +927,18 @@ static void ZSTDMT_releaseAllJobResources(ZSTDMT_CCtx* mtctx)
     unsigned jobID;
     DEBUGLOG(3, "ZSTDMT_releaseAllJobResources");
     for (jobID=0; jobID <= mtctx->jobIDMask; jobID++) {
+        /* Copy the mutex/cond out */
+        ZSTD_pthread_mutex_t const mutex = mtctx->jobs[jobID].job_mutex;
+        ZSTD_pthread_cond_t const cond = mtctx->jobs[jobID].job_cond;
+
         DEBUGLOG(4, "job%02u: release dst address %08X", jobID, (U32)(size_t)mtctx->jobs[jobID].dstBuff.start);
         ZSTDMT_releaseBuffer(mtctx->bufPool, mtctx->jobs[jobID].dstBuff);
-        mtctx->jobs[jobID].dstBuff = g_nullBuffer;
-        mtctx->jobs[jobID].cSize = 0;
+
+        /* Clear the job description, but keep the mutex/cond */
+        memset(&mtctx->jobs[jobID], 0, sizeof(mtctx->jobs[jobID]));
+        mtctx->jobs[jobID].job_mutex = mutex;
+        mtctx->jobs[jobID].job_cond = cond;
     }
-    memset(mtctx->jobs, 0, (mtctx->jobIDMask+1)*sizeof(ZSTDMT_jobDescription));
     mtctx->inBuff.buffer = g_nullBuffer;
     mtctx->inBuff.filled = 0;
     mtctx->allJobsCompleted = 1;
@@ -1028,9 +1034,9 @@ size_t ZSTDMT_getMTCtxParameter(ZSTDMT_CCtx* mtctx, ZSTDMT_parameter parameter,
 
 /* Sets parameters relevant to the compression job,
  * initializing others to default values. */
-static ZSTD_CCtx_params ZSTDMT_initJobCCtxParams(ZSTD_CCtx_params const params)
+static ZSTD_CCtx_params ZSTDMT_initJobCCtxParams(const ZSTD_CCtx_params* params)
 {
-    ZSTD_CCtx_params jobParams = params;
+    ZSTD_CCtx_params jobParams = *params;
     /* Clear parameters related to multithreading */
     jobParams.forceWindow = 0;
     jobParams.nbWorkers = 0;
@@ -1151,16 +1157,16 @@ size_t ZSTDMT_toFlushNow(ZSTDMT_CCtx* mtctx)
 /* =====   Multi-threaded compression   ===== */
 /* ------------------------------------------ */
 
-static unsigned ZSTDMT_computeTargetJobLog(ZSTD_CCtx_params const params)
+static unsigned ZSTDMT_computeTargetJobLog(const ZSTD_CCtx_params* params)
 {
     unsigned jobLog;
-    if (params.ldmParams.enableLdm) {
+    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);
+        jobLog = MAX(21, params->cParams.chainLog + 4);
     } else {
-        jobLog = MAX(20, params.cParams.windowLog + 2);
+        jobLog = MAX(20, params->cParams.windowLog + 2);
     }
     return MIN(jobLog, (unsigned)ZSTDMT_JOBLOG_MAX);
 }
@@ -1193,27 +1199,27 @@ static int ZSTDMT_overlapLog(int ovlog, ZSTD_strategy strat)
     return ovlog;
 }
 
-static size_t ZSTDMT_computeOverlapSize(ZSTD_CCtx_params const params)
+static size_t ZSTDMT_computeOverlapSize(const ZSTD_CCtx_params* params)
 {
-    int const overlapRLog = 9 - ZSTDMT_overlapLog(params.overlapLog, params.cParams.strategy);
-    int ovLog = (overlapRLog >= 8) ? 0 : (params.cParams.windowLog - overlapRLog);
+    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) {
+    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)
+        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, "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)
+ZSTDMT_computeNbJobs(const ZSTD_CCtx_params* params, size_t srcSize, unsigned nbWorkers)
 {
     assert(nbWorkers>0);
     {   size_t const jobSizeTarget = (size_t)1 << ZSTDMT_computeTargetJobLog(params);
@@ -1236,9 +1242,9 @@ static size_t ZSTDMT_compress_advanced_internal(
           const ZSTD_CDict* cdict,
                 ZSTD_CCtx_params params)
 {
-    ZSTD_CCtx_params const jobParams = ZSTDMT_initJobCCtxParams(params);
-    size_t const overlapSize = ZSTDMT_computeOverlapSize(params);
-    unsigned const nbJobs = ZSTDMT_computeNbJobs(params, srcSize, params.nbWorkers);
+    ZSTD_CCtx_params const jobParams = ZSTDMT_initJobCCtxParams(&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 */
     const char* const srcStart = (const char*)src;
@@ -1256,7 +1262,7 @@ static size_t ZSTDMT_compress_advanced_internal(
         ZSTD_CCtx* const cctx = mtctx->cctxPool->cctx[0];
         DEBUGLOG(4, "ZSTDMT_compress_advanced_internal: fallback to single-thread mode");
         if (cdict) return ZSTD_compress_usingCDict_advanced(cctx, dst, dstCapacity, src, srcSize, cdict, jobParams.fParams);
-        return ZSTD_compress_advanced_internal(cctx, dst, dstCapacity, src, srcSize, NULL, 0, jobParams);
+        return ZSTD_compress_advanced_internal(cctx, dst, dstCapacity, src, srcSize, NULL, 0, &jobParams);
     }
 
     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) */
@@ -1404,12 +1410,12 @@ size_t ZSTDMT_initCStream_internal(
 
     mtctx->singleBlockingThread = (pledgedSrcSize <= ZSTDMT_JOBSIZE_MIN);  /* do not trigger multi-threading when srcSize is too small */
     if (mtctx->singleBlockingThread) {
-        ZSTD_CCtx_params const singleThreadParams = ZSTDMT_initJobCCtxParams(params);
+        ZSTD_CCtx_params const singleThreadParams = ZSTDMT_initJobCCtxParams(&params);
         DEBUGLOG(5, "ZSTDMT_initCStream_internal: switch to single blocking thread mode");
         assert(singleThreadParams.nbWorkers == 0);
         return ZSTD_initCStream_internal(mtctx->cctxPool->cctx[0],
                                          dict, dictSize, cdict,
-                                         singleThreadParams, pledgedSrcSize);
+                                         &singleThreadParams, pledgedSrcSize);
     }
 
     DEBUGLOG(4, "ZSTDMT_initCStream_internal: %u workers", params.nbWorkers);
@@ -1435,11 +1441,11 @@ size_t ZSTDMT_initCStream_internal(
         mtctx->cdict = cdict;
     }
 
-    mtctx->targetPrefixSize = ZSTDMT_computeOverlapSize(params);
+    mtctx->targetPrefixSize = ZSTDMT_computeOverlapSize(&params);
     DEBUGLOG(4, "overlapLog=%i => %u KB", params.overlapLog, (U32)(mtctx->targetPrefixSize>>10));
     mtctx->targetSectionSize = params.jobSize;
     if (mtctx->targetSectionSize == 0) {
-        mtctx->targetSectionSize = 1ULL << ZSTDMT_computeTargetJobLog(params);
+        mtctx->targetSectionSize = 1ULL << ZSTDMT_computeTargetJobLog(&params);
     }
     assert(mtctx->targetSectionSize <= (size_t)ZSTDMT_JOBSIZE_MAX);
 

thirdparty/zstd/decompress/huf_decompress.c

@@ -61,7 +61,9 @@
 *  Error Management
 ****************************************************************/
 #define HUF_isError ERR_isError
+#ifndef CHECK_F
 #define CHECK_F(f) { size_t const err_ = (f); if (HUF_isError(err_)) return err_; }
+#endif
 
 
 /* **************************************************************

thirdparty/zstd/decompress/zstd_decompress.c

@@ -88,10 +88,7 @@ size_t ZSTD_estimateDCtxSize(void) { return sizeof(ZSTD_DCtx); }
 
 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_STATIC_ASSERT(ZSTD_FRAMEHEADERSIZE_PREFIX >= ZSTD_FRAMEIDSIZE);
+    size_t const startingInputLength = ZSTD_FRAMEHEADERSIZE_PREFIX(format);
     /* only supports formats ZSTD_f_zstd1 and ZSTD_f_zstd1_magicless */
     assert( (format == ZSTD_f_zstd1) || (format == ZSTD_f_zstd1_magicless) );
     return startingInputLength;
@@ -376,7 +373,7 @@ unsigned long long ZSTD_findDecompressedSize(const void* src, size_t srcSize)
 {
     unsigned long long totalDstSize = 0;
 
-    while (srcSize >= ZSTD_FRAMEHEADERSIZE_PREFIX) {
+    while (srcSize >= ZSTD_startingInputLength(ZSTD_f_zstd1)) {
         U32 const magicNumber = MEM_readLE32(src);
 
         if ((magicNumber & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) {
@@ -629,11 +626,12 @@ static size_t ZSTD_decompressFrame(ZSTD_DCtx* dctx,
 
     /* check */
     RETURN_ERROR_IF(
-        remainingSrcSize < ZSTD_FRAMEHEADERSIZE_MIN+ZSTD_blockHeaderSize,
+        remainingSrcSize < ZSTD_FRAMEHEADERSIZE_MIN(dctx->format)+ZSTD_blockHeaderSize,
         srcSize_wrong);
 
     /* Frame Header */
-    {   size_t const frameHeaderSize = ZSTD_frameHeaderSize(ip, ZSTD_FRAMEHEADERSIZE_PREFIX);
+    {   size_t const frameHeaderSize = ZSTD_frameHeaderSize_internal(
+                ip, ZSTD_FRAMEHEADERSIZE_PREFIX(dctx->format), dctx->format);
         if (ZSTD_isError(frameHeaderSize)) return frameHeaderSize;
         RETURN_ERROR_IF(remainingSrcSize < frameHeaderSize+ZSTD_blockHeaderSize,
                         srcSize_wrong);
@@ -714,7 +712,7 @@ static size_t ZSTD_decompressMultiFrame(ZSTD_DCtx* dctx,
         dictSize = ZSTD_DDict_dictSize(ddict);
     }
 
-    while (srcSize >= ZSTD_FRAMEHEADERSIZE_PREFIX) {
+    while (srcSize >= ZSTD_startingInputLength(dctx->format)) {
 
 #if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1)
         if (ZSTD_isLegacy(src, srcSize)) {
@@ -1098,7 +1096,7 @@ ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy,
         size_t const dictContentSize = (size_t)(dictEnd - (dictPtr+12));
         for (i=0; i<3; i++) {
             U32 const rep = MEM_readLE32(dictPtr); dictPtr += 4;
-            RETURN_ERROR_IF(rep==0 || rep >= dictContentSize,
+            RETURN_ERROR_IF(rep==0 || rep > dictContentSize,
                             dictionary_corrupted);
             entropy->rep[i] = rep;
     }   }
@@ -1267,7 +1265,7 @@ size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx,
 {
     RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong);
     ZSTD_clearDict(dctx);
-    if (dict && dictSize >= 8) {
+    if (dict && dictSize != 0) {
         dctx->ddictLocal = ZSTD_createDDict_advanced(dict, dictSize, dictLoadMethod, dictContentType, dctx->customMem);
         RETURN_ERROR_IF(dctx->ddictLocal == NULL, memory_allocation);
         dctx->ddict = dctx->ddictLocal;
@@ -1300,14 +1298,14 @@ size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSiz
 
 
 /* ZSTD_initDStream_usingDict() :
- * return : expected size, aka ZSTD_FRAMEHEADERSIZE_PREFIX.
+ * return : expected size, aka ZSTD_startingInputLength().
  * this function cannot fail */
 size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize)
 {
     DEBUGLOG(4, "ZSTD_initDStream_usingDict");
     FORWARD_IF_ERROR( ZSTD_DCtx_reset(zds, ZSTD_reset_session_only) );
     FORWARD_IF_ERROR( ZSTD_DCtx_loadDictionary(zds, dict, dictSize) );
-    return ZSTD_FRAMEHEADERSIZE_PREFIX;
+    return ZSTD_startingInputLength(zds->format);
 }
 
 /* note : this variant can't fail */
@@ -1324,16 +1322,16 @@ size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* dctx, const ZSTD_DDict* ddict)
 {
     FORWARD_IF_ERROR( ZSTD_DCtx_reset(dctx, ZSTD_reset_session_only) );
     FORWARD_IF_ERROR( ZSTD_DCtx_refDDict(dctx, ddict) );
-    return ZSTD_FRAMEHEADERSIZE_PREFIX;
+    return ZSTD_startingInputLength(dctx->format);
 }
 
 /* ZSTD_resetDStream() :
- * return : expected size, aka ZSTD_FRAMEHEADERSIZE_PREFIX.
+ * return : expected size, aka ZSTD_startingInputLength().
  * this function cannot fail */
 size_t ZSTD_resetDStream(ZSTD_DStream* dctx)
 {
     FORWARD_IF_ERROR(ZSTD_DCtx_reset(dctx, ZSTD_reset_session_only));
-    return ZSTD_FRAMEHEADERSIZE_PREFIX;
+    return ZSTD_startingInputLength(dctx->format);
 }
 
 
@@ -1564,7 +1562,7 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB
                             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((size_t)ZSTD_FRAMEHEADERSIZE_MIN(zds->format), 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;