zstd-ruby 1.5.1.1 → 1.5.5.0

data/ext/zstdruby/libzstd/common/huf.h

@@ -1,7 +1,7 @@
 /* ******************************************************************
  * huff0 huffman codec,
  * part of Finite State Entropy library
- * Copyright (c) Yann Collet, Facebook, Inc.
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
  *
  * You can contact the author at :
  * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
@@ -21,99 +21,22 @@ extern "C" {
 
 /* *** Dependencies *** */
 #include "zstd_deps.h"    /* size_t */
-
-
-/* *** library symbols visibility *** */
-/* Note : when linking with -fvisibility=hidden on gcc, or by default on Visual,
- *        HUF symbols remain "private" (internal symbols for library only).
- *        Set macro FSE_DLL_EXPORT to 1 if you want HUF symbols visible on DLL interface */
-#if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4)
-#  define HUF_PUBLIC_API __attribute__ ((visibility ("default")))
-#elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1)   /* Visual expected */
-#  define HUF_PUBLIC_API __declspec(dllexport)
-#elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1)
-#  define HUF_PUBLIC_API __declspec(dllimport)  /* not required, just to generate faster code (saves a function pointer load from IAT and an indirect jump) */
-#else
-#  define HUF_PUBLIC_API
-#endif
-
-
-/* ========================== */
-/* ***  simple functions  *** */
-/* ========================== */
-
-/** HUF_compress() :
- *  Compress content from buffer 'src', of size 'srcSize', into buffer 'dst'.
- * 'dst' buffer must be already allocated.
- *  Compression runs faster if `dstCapacity` >= HUF_compressBound(srcSize).
- * `srcSize` must be <= `HUF_BLOCKSIZE_MAX` == 128 KB.
- * @return : size of compressed data (<= `dstCapacity`).
- *  Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!!
- *                   if HUF_isError(return), compression failed (more details using HUF_getErrorName())
- */
-HUF_PUBLIC_API size_t HUF_compress(void* dst, size_t dstCapacity,
-                             const void* src, size_t srcSize);
-
-/** HUF_decompress() :
- *  Decompress HUF data from buffer 'cSrc', of size 'cSrcSize',
- *  into already allocated buffer 'dst', of minimum size 'dstSize'.
- * `originalSize` : **must** be the ***exact*** size of original (uncompressed) data.
- *  Note : in contrast with FSE, HUF_decompress can regenerate
- *         RLE (cSrcSize==1) and uncompressed (cSrcSize==dstSize) data,
- *         because it knows size to regenerate (originalSize).
- * @return : size of regenerated data (== originalSize),
- *           or an error code, which can be tested using HUF_isError()
- */
-HUF_PUBLIC_API size_t HUF_decompress(void* dst,  size_t originalSize,
-                               const void* cSrc, size_t cSrcSize);
+#include "mem.h"          /* U32 */
+#define FSE_STATIC_LINKING_ONLY
+#include "fse.h"
 
 
 /* ***   Tool functions *** */
-#define HUF_BLOCKSIZE_MAX (128 * 1024)                  /**< maximum input size for a single block compressed with HUF_compress */
-HUF_PUBLIC_API size_t HUF_compressBound(size_t size);   /**< maximum compressed size (worst case) */
+#define HUF_BLOCKSIZE_MAX (128 * 1024)   /**< maximum input size for a single block compressed with HUF_compress */
+size_t HUF_compressBound(size_t size);   /**< maximum compressed size (worst case) */
 
 /* Error Management */
-HUF_PUBLIC_API unsigned    HUF_isError(size_t code);       /**< tells if a return value is an error code */
-HUF_PUBLIC_API const char* HUF_getErrorName(size_t code);  /**< provides error code string (useful for debugging) */
+unsigned    HUF_isError(size_t code);       /**< tells if a return value is an error code */
+const char* HUF_getErrorName(size_t code);  /**< provides error code string (useful for debugging) */
 
 
-/* ***   Advanced function   *** */
-
-/** HUF_compress2() :
- *  Same as HUF_compress(), but offers control over `maxSymbolValue` and `tableLog`.
- * `maxSymbolValue` must be <= HUF_SYMBOLVALUE_MAX .
- * `tableLog` must be `<= HUF_TABLELOG_MAX` . */
-HUF_PUBLIC_API size_t HUF_compress2 (void* dst, size_t dstCapacity,
-                               const void* src, size_t srcSize,
-                               unsigned maxSymbolValue, unsigned tableLog);
-
-/** HUF_compress4X_wksp() :
- *  Same as HUF_compress2(), but uses externally allocated `workSpace`.
- * `workspace` must be at least as large as HUF_WORKSPACE_SIZE */
 #define HUF_WORKSPACE_SIZE ((8 << 10) + 512 /* sorting scratch space */)
 #define HUF_WORKSPACE_SIZE_U64 (HUF_WORKSPACE_SIZE / sizeof(U64))
-HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity,
-                                     const void* src, size_t srcSize,
-                                     unsigned maxSymbolValue, unsigned tableLog,
-                                     void* workSpace, size_t wkspSize);
-
-#endif   /* HUF_H_298734234 */
-
-/* ******************************************************************
- *  WARNING !!
- *  The following section contains advanced and experimental definitions
- *  which shall never be used in the context of a dynamic library,
- *  because they are not guaranteed to remain stable in the future.
- *  Only consider them in association with static linking.
- * *****************************************************************/
-#if defined(HUF_STATIC_LINKING_ONLY) && !defined(HUF_H_HUF_STATIC_LINKING_ONLY)
-#define HUF_H_HUF_STATIC_LINKING_ONLY
-
-/* *** Dependencies *** */
-#include "mem.h"   /* U32 */
-#define FSE_STATIC_LINKING_ONLY
-#include "fse.h"
-
 
 /* *** Constants *** */
 #define HUF_TABLELOG_MAX      12      /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */
@@ -154,25 +77,49 @@ typedef U32 HUF_DTable;
 /* ****************************************
 *  Advanced decompression functions
 ******************************************/
-size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
-#ifndef HUF_FORCE_DECOMPRESS_X1
-size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
-#endif
 
-size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< decodes RLE and uncompressed */
-size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< considers RLE and uncompressed as errors */
-size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< considers RLE and uncompressed as errors */
-size_t HUF_decompress4X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
-size_t HUF_decompress4X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);   /**< single-symbol decoder */
-#ifndef HUF_FORCE_DECOMPRESS_X1
-size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
-size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);   /**< double-symbols decoder */
-#endif
+/**
+ * Huffman flags bitset.
+ * For all flags, 0 is the default value.
+ */
+typedef enum {
+    /**
+     * If compiled with DYNAMIC_BMI2: Set flag only if the CPU supports BMI2 at runtime.
+     * Otherwise: Ignored.
+     */
+    HUF_flags_bmi2 = (1 << 0),
+    /**
+     * If set: Test possible table depths to find the one that produces the smallest header + encoded size.
+     * If unset: Use heuristic to find the table depth.
+     */
+    HUF_flags_optimalDepth = (1 << 1),
+    /**
+     * If set: If the previous table can encode the input, always reuse the previous table.
+     * If unset: If the previous table can encode the input, reuse the previous table if it results in a smaller output.
+     */
+    HUF_flags_preferRepeat = (1 << 2),
+    /**
+     * If set: Sample the input and check if the sample is uncompressible, if it is then don't attempt to compress.
+     * If unset: Always histogram the entire input.
+     */
+    HUF_flags_suspectUncompressible = (1 << 3),
+    /**
+     * If set: Don't use assembly implementations
+     * If unset: Allow using assembly implementations
+     */
+    HUF_flags_disableAsm = (1 << 4),
+    /**
+     * If set: Don't use the fast decoding loop, always use the fallback decoding loop.
+     * If unset: Use the fast decoding loop when possible.
+     */
+    HUF_flags_disableFast = (1 << 5)
+} HUF_flags_e;
 
 
 /* ****************************************
  *  HUF detailed API
  * ****************************************/
+#define HUF_OPTIMAL_DEPTH_THRESHOLD ZSTD_btultra
 
 /*! HUF_compress() does the following:
  *  1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h")
@@ -185,12 +132,12 @@ size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize,
  *  For example, it's possible to compress several blocks using the same 'CTable',
  *  or to save and regenerate 'CTable' using external methods.
  */
-unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue);
-size_t HUF_buildCTable (HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits);   /* @return : maxNbBits; CTable and count can overlap. In which case, CTable will overwrite count content */
-size_t HUF_writeCTable (void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog);
+unsigned HUF_minTableLog(unsigned symbolCardinality);
+unsigned HUF_cardinality(const unsigned* count, unsigned maxSymbolValue);
+unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, void* workSpace,
+ size_t wkspSize, HUF_CElt* table, const unsigned* count, int flags); /* table is used as scratch space for building and testing tables, not a return value */
 size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize);
-size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
-size_t HUF_compress4X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2);
+size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int flags);
 size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
 int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
 
@@ -199,6 +146,7 @@ typedef enum {
    HUF_repeat_check, /**< Can use the previous table but it must be checked. Note : The previous table must have been constructed by HUF_compress{1, 4}X_repeat */
    HUF_repeat_valid  /**< Can use the previous table and it is assumed to be valid */
  } HUF_repeat;
+
 /** HUF_compress4X_repeat() :
  *  Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
  *  If it uses hufTable it does not modify hufTable or repeat.
@@ -209,13 +157,13 @@ size_t HUF_compress4X_repeat(void* dst, size_t dstSize,
                        const void* src, size_t srcSize,
                        unsigned maxSymbolValue, unsigned tableLog,
                        void* workSpace, size_t wkspSize,    /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
-                       HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible);
+                       HUF_CElt* hufTable, HUF_repeat* repeat, int flags);
 
 /** HUF_buildCTable_wksp() :
  *  Same as HUF_buildCTable(), but using externally allocated scratch buffer.
  * `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE.
  */
-#define HUF_CTABLE_WORKSPACE_SIZE_U32 (2*HUF_SYMBOLVALUE_MAX +1 +1)
+#define HUF_CTABLE_WORKSPACE_SIZE_U32 ((4 * (HUF_SYMBOLVALUE_MAX + 1)) + 192)
 #define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))
 size_t HUF_buildCTable_wksp (HUF_CElt* tree,
                        const unsigned* count, U32 maxSymbolValue, U32 maxNbBits,
@@ -241,7 +189,7 @@ size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize,
                           U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
                           const void* src, size_t srcSize,
                           void* workspace, size_t wkspSize,
-                          int bmi2);
+                          int flags);
 
 /** HUF_readCTable() :
  *  Loading a CTable saved with HUF_writeCTable() */
@@ -279,32 +227,12 @@ U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize);
 #define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9))
 #define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))
 
-#ifndef HUF_FORCE_DECOMPRESS_X2
-size_t HUF_readDTableX1 (HUF_DTable* DTable, const void* src, size_t srcSize);
-size_t HUF_readDTableX1_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
-#endif
-#ifndef HUF_FORCE_DECOMPRESS_X1
-size_t HUF_readDTableX2 (HUF_DTable* DTable, const void* src, size_t srcSize);
-size_t HUF_readDTableX2_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
-#endif
-
-size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
-#ifndef HUF_FORCE_DECOMPRESS_X2
-size_t HUF_decompress4X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
-#endif
-#ifndef HUF_FORCE_DECOMPRESS_X1
-size_t HUF_decompress4X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
-#endif
-
 
 /* ====================== */
 /* single stream variants */
 /* ====================== */
 
-size_t HUF_compress1X (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog);
-size_t HUF_compress1X_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);  /**< `workSpace` must be a table of at least HUF_WORKSPACE_SIZE_U64 U64 */
-size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
-size_t HUF_compress1X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2);
+size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int flags);
 /** HUF_compress1X_repeat() :
  *  Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
  *  If it uses hufTable it does not modify hufTable or repeat.
@@ -315,49 +243,30 @@ size_t HUF_compress1X_repeat(void* dst, size_t dstSize,
                        const void* src, size_t srcSize,
                        unsigned maxSymbolValue, unsigned tableLog,
                        void* workSpace, size_t wkspSize,   /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
-                       HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible);
-
-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);
-#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
+                       HUF_CElt* hufTable, HUF_repeat* repeat, int flags);
 
-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
+size_t HUF_decompress1X_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
 #ifndef HUF_FORCE_DECOMPRESS_X1
-size_t HUF_decompress1X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);   /**< double-symbols decoder */
 #endif
 
 /* BMI2 variants.
  * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
  */
-size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
+size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int flags);
 #ifndef HUF_FORCE_DECOMPRESS_X2
-size_t HUF_decompress1X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
+size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
 #endif
-size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
-size_t HUF_decompress4X_hufOnly_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
+size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int flags);
+size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
 #ifndef HUF_FORCE_DECOMPRESS_X2
-size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2);
+size_t HUF_readDTableX1_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int flags);
 #endif
 #ifndef HUF_FORCE_DECOMPRESS_X1
-size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2);
+size_t HUF_readDTableX2_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int flags);
 #endif
 
-#endif /* HUF_STATIC_LINKING_ONLY */
+#endif   /* HUF_H_298734234 */
 
 #if defined (__cplusplus)
 }

data/ext/zstdruby/libzstd/common/mem.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) Yann Collet, Facebook, Inc.
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -133,21 +133,15 @@ MEM_STATIC size_t MEM_swapST(size_t in);
 /*-**************************************************************
 *  Memory I/O Implementation
 *****************************************************************/
-/* MEM_FORCE_MEMORY_ACCESS :
- * By default, access to unaligned memory is controlled by `memcpy()`, which is safe and portable.
- * Unfortunately, on some target/compiler combinations, the generated assembly is sub-optimal.
- * The below switch allow to select different access method for improved performance.
- * Method 0 (default) : use `memcpy()`. Safe and portable.
- * Method 1 : `__packed` statement. It depends on compiler extension (i.e., not portable).
- *            This method is safe if your compiler supports it, and *generally* as fast or faster than `memcpy`.
+/* MEM_FORCE_MEMORY_ACCESS : For accessing unaligned memory:
+ * Method 0 : always use `memcpy()`. Safe and portable.
+ * Method 1 : Use compiler extension to set unaligned access.
  * Method 2 : direct access. This method is portable but violate C standard.
  *            It can generate buggy code on targets depending on alignment.
- *            In some circumstances, it's the only known way to get the most performance (i.e. GCC + ARMv6)
- * See http://fastcompression.blogspot.fr/2015/08/accessing-unaligned-memory.html for details.
- * Prefer these methods in priority order (0 > 1 > 2)
+ * Default  : method 1 if supported, else method 0
  */
 #ifndef MEM_FORCE_MEMORY_ACCESS   /* can be defined externally, on command line for example */
-#  if defined(__INTEL_COMPILER) || defined(__GNUC__) || defined(__ICCARM__)
+#  ifdef __GNUC__
 #    define MEM_FORCE_MEMORY_ACCESS 1
 #  endif
 #endif
@@ -190,30 +184,19 @@ MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(U64*)memPtr = value; }
 
 #elif defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==1)
 
-/* __pack instructions are safer, but compiler specific, hence potentially problematic for some compilers */
-/* currently only defined for gcc and icc */
-#if defined(_MSC_VER) || (defined(__INTEL_COMPILER) && defined(WIN32))
-    __pragma( pack(push, 1) )
-    typedef struct { U16 v; } unalign16;
-    typedef struct { U32 v; } unalign32;
-    typedef struct { U64 v; } unalign64;
-    typedef struct { size_t v; } unalignArch;
-    __pragma( pack(pop) )
-#else
-    typedef struct { U16 v; } __attribute__((packed)) unalign16;
-    typedef struct { U32 v; } __attribute__((packed)) unalign32;
-    typedef struct { U64 v; } __attribute__((packed)) unalign64;
-    typedef struct { size_t v; } __attribute__((packed)) unalignArch;
-#endif
+typedef __attribute__((aligned(1))) U16 unalign16;
+typedef __attribute__((aligned(1))) U32 unalign32;
+typedef __attribute__((aligned(1))) U64 unalign64;
+typedef __attribute__((aligned(1))) size_t unalignArch;
 
-MEM_STATIC U16 MEM_read16(const void* ptr) { return ((const unalign16*)ptr)->v; }
-MEM_STATIC U32 MEM_read32(const void* ptr) { return ((const unalign32*)ptr)->v; }
-MEM_STATIC U64 MEM_read64(const void* ptr) { return ((const unalign64*)ptr)->v; }
-MEM_STATIC size_t MEM_readST(const void* ptr) { return ((const unalignArch*)ptr)->v; }
+MEM_STATIC U16 MEM_read16(const void* ptr) { return *(const unalign16*)ptr; }
+MEM_STATIC U32 MEM_read32(const void* ptr) { return *(const unalign32*)ptr; }
+MEM_STATIC U64 MEM_read64(const void* ptr) { return *(const unalign64*)ptr; }
+MEM_STATIC size_t MEM_readST(const void* ptr) { return *(const unalignArch*)ptr; }
 
-MEM_STATIC void MEM_write16(void* memPtr, U16 value) { ((unalign16*)memPtr)->v = value; }
-MEM_STATIC void MEM_write32(void* memPtr, U32 value) { ((unalign32*)memPtr)->v = value; }
-MEM_STATIC void MEM_write64(void* memPtr, U64 value) { ((unalign64*)memPtr)->v = value; }
+MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(unalign16*)memPtr = value; }
+MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(unalign32*)memPtr = value; }
+MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(unalign64*)memPtr = value; }
 
 #else
 
@@ -257,6 +240,14 @@ MEM_STATIC void MEM_write64(void* memPtr, U64 value)
 
 #endif /* MEM_FORCE_MEMORY_ACCESS */
 
+MEM_STATIC U32 MEM_swap32_fallback(U32 in)
+{
+    return  ((in << 24) & 0xff000000 ) |
+            ((in <<  8) & 0x00ff0000 ) |
+            ((in >>  8) & 0x0000ff00 ) |
+            ((in >> 24) & 0x000000ff );
+}
+
 MEM_STATIC U32 MEM_swap32(U32 in)
 {
 #if defined(_MSC_VER)     /* Visual Studio */
@@ -265,22 +256,13 @@ MEM_STATIC U32 MEM_swap32(U32 in)
   || (defined(__clang__) && __has_builtin(__builtin_bswap32))
     return __builtin_bswap32(in);
 #else
-    return  ((in << 24) & 0xff000000 ) |
-            ((in <<  8) & 0x00ff0000 ) |
-            ((in >>  8) & 0x0000ff00 ) |
-            ((in >> 24) & 0x000000ff );
+    return MEM_swap32_fallback(in);
 #endif
 }
 
-MEM_STATIC U64 MEM_swap64(U64 in)
+MEM_STATIC U64 MEM_swap64_fallback(U64 in)
 {
-#if defined(_MSC_VER)     /* Visual Studio */
-    return _byteswap_uint64(in);
-#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \
-  || (defined(__clang__) && __has_builtin(__builtin_bswap64))
-    return __builtin_bswap64(in);
-#else
-    return  ((in << 56) & 0xff00000000000000ULL) |
+     return  ((in << 56) & 0xff00000000000000ULL) |
             ((in << 40) & 0x00ff000000000000ULL) |
             ((in << 24) & 0x0000ff0000000000ULL) |
             ((in << 8)  & 0x000000ff00000000ULL) |
@@ -288,6 +270,17 @@ MEM_STATIC U64 MEM_swap64(U64 in)
             ((in >> 24) & 0x0000000000ff0000ULL) |
             ((in >> 40) & 0x000000000000ff00ULL) |
             ((in >> 56) & 0x00000000000000ffULL);
+}
+
+MEM_STATIC U64 MEM_swap64(U64 in)
+{
+#if defined(_MSC_VER)     /* Visual Studio */
+    return _byteswap_uint64(in);
+#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \
+  || (defined(__clang__) && __has_builtin(__builtin_bswap64))
+    return __builtin_bswap64(in);
+#else
+    return MEM_swap64_fallback(in);
 #endif
 }
 

data/ext/zstdruby/libzstd/common/pool.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) Yann Collet, Facebook, Inc.
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -10,9 +10,9 @@
 
 
 /* ======   Dependencies   ======= */
+#include "../common/allocations.h"  /* ZSTD_customCalloc, ZSTD_customFree */
 #include "zstd_deps.h" /* size_t */
 #include "debug.h"     /* assert */
-#include "zstd_internal.h"  /* ZSTD_customMalloc, ZSTD_customFree */
 #include "pool.h"
 
 /* ======   Compiler specifics   ====== */
@@ -86,7 +86,7 @@ static void* POOL_thread(void* opaque) {
         {   POOL_job const job = ctx->queue[ctx->queueHead];
             ctx->queueHead = (ctx->queueHead + 1) % ctx->queueSize;
             ctx->numThreadsBusy++;
-            ctx->queueEmpty = ctx->queueHead == ctx->queueTail;
+            ctx->queueEmpty = (ctx->queueHead == ctx->queueTail);
             /* Unlock the mutex, signal a pusher, and run the job */
             ZSTD_pthread_cond_signal(&ctx->queuePushCond);
             ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
@@ -96,15 +96,14 @@ static void* POOL_thread(void* opaque) {
             /* If the intended queue size was 0, signal after finishing job */
             ZSTD_pthread_mutex_lock(&ctx->queueMutex);
             ctx->numThreadsBusy--;
-            if (ctx->queueSize == 1) {
-                ZSTD_pthread_cond_signal(&ctx->queuePushCond);
-            }
+            ZSTD_pthread_cond_signal(&ctx->queuePushCond);
             ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
         }
     }  /* for (;;) */
     assert(0);  /* Unreachable */
 }
 
+/* ZSTD_createThreadPool() : public access point */
 POOL_ctx* ZSTD_createThreadPool(size_t numThreads) {
     return POOL_create (numThreads, 0);
 }
@@ -114,7 +113,8 @@ POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
 }
 
 POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
-                               ZSTD_customMem customMem) {
+                               ZSTD_customMem customMem)
+{
     POOL_ctx* ctx;
     /* Check parameters */
     if (!numThreads) { return NULL; }
@@ -126,7 +126,7 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
      * empty and full queues.
      */
     ctx->queueSize = queueSize + 1;
-    ctx->queue = (POOL_job*)ZSTD_customMalloc(ctx->queueSize * sizeof(POOL_job), customMem);
+    ctx->queue = (POOL_job*)ZSTD_customCalloc(ctx->queueSize * sizeof(POOL_job), customMem);
     ctx->queueHead = 0;
     ctx->queueTail = 0;
     ctx->numThreadsBusy = 0;
@@ -140,7 +140,7 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
     }
     ctx->shutdown = 0;
     /* Allocate space for the thread handles */
-    ctx->threads = (ZSTD_pthread_t*)ZSTD_customMalloc(numThreads * sizeof(ZSTD_pthread_t), customMem);
+    ctx->threads = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), customMem);
     ctx->threadCapacity = 0;
     ctx->customMem = customMem;
     /* Check for errors */
@@ -173,7 +173,7 @@ static void POOL_join(POOL_ctx* ctx) {
     /* Join all of the threads */
     {   size_t i;
         for (i = 0; i < ctx->threadCapacity; ++i) {
-            ZSTD_pthread_join(ctx->threads[i], NULL);  /* note : could fail */
+            ZSTD_pthread_join(ctx->threads[i]);  /* note : could fail */
     }   }
 }
 
@@ -188,11 +188,22 @@ void POOL_free(POOL_ctx *ctx) {
     ZSTD_customFree(ctx, ctx->customMem);
 }
 
+/*! POOL_joinJobs() :
+ *  Waits for all queued jobs to finish executing.
+ */
+void POOL_joinJobs(POOL_ctx* ctx) {
+    ZSTD_pthread_mutex_lock(&ctx->queueMutex);
+    while(!ctx->queueEmpty || ctx->numThreadsBusy > 0) {
+        ZSTD_pthread_cond_wait(&ctx->queuePushCond, &ctx->queueMutex);
+    }
+    ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+}
+
 void ZSTD_freeThreadPool (ZSTD_threadPool* pool) {
   POOL_free (pool);
 }
 
-size_t POOL_sizeof(POOL_ctx *ctx) {
+size_t POOL_sizeof(const POOL_ctx* ctx) {
     if (ctx==NULL) return 0;  /* supports sizeof NULL */
     return sizeof(*ctx)
         + ctx->queueSize * sizeof(POOL_job)
@@ -209,7 +220,7 @@ static int POOL_resize_internal(POOL_ctx* ctx, size_t numThreads)
         return 0;
     }
     /* numThreads > threadCapacity */
-    {   ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customMalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem);
+    {   ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem);
         if (!threadPool) return 1;
         /* replace existing thread pool */
         ZSTD_memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(*threadPool));
@@ -257,9 +268,12 @@ static int isQueueFull(POOL_ctx const* ctx) {
 }
 
 
-static void POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque)
+static void
+POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque)
 {
-    POOL_job const job = {function, opaque};
+    POOL_job job;
+    job.function = function;
+    job.opaque = opaque;
     assert(ctx != NULL);
     if (ctx->shutdown) return;
 
@@ -313,7 +327,9 @@ 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)
+{
     (void)numThreads;
     (void)queueSize;
     (void)customMem;
@@ -325,6 +341,11 @@ void POOL_free(POOL_ctx* ctx) {
     (void)ctx;
 }
 
+void POOL_joinJobs(POOL_ctx* ctx){
+    assert(!ctx || ctx == &g_poolCtx);
+    (void)ctx;
+}
+
 int POOL_resize(POOL_ctx* ctx, size_t numThreads) {
     (void)ctx; (void)numThreads;
     return 0;
@@ -341,7 +362,7 @@ int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque) {
     return 1;
 }
 
-size_t POOL_sizeof(POOL_ctx* ctx) {
+size_t POOL_sizeof(const POOL_ctx* ctx) {
     if (ctx==NULL) return 0;  /* supports sizeof NULL */
     assert(ctx == &g_poolCtx);
     return sizeof(*ctx);

data/ext/zstdruby/libzstd/common/pool.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) Yann Collet, Facebook, Inc.
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -38,6 +38,12 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
  */
 void POOL_free(POOL_ctx* ctx);
 
+
+/*! POOL_joinJobs() :
+ *  Waits for all queued jobs to finish executing.
+ */
+void POOL_joinJobs(POOL_ctx* ctx);
+
 /*! POOL_resize() :
  *  Expands or shrinks pool's number of threads.
  *  This is more efficient than releasing + creating a new context,
@@ -53,7 +59,7 @@ int POOL_resize(POOL_ctx* ctx, size_t numThreads);
  * @return threadpool memory usage
  *  note : compatible with NULL (returns 0 in this case)
  */
-size_t POOL_sizeof(POOL_ctx* ctx);
+size_t POOL_sizeof(const POOL_ctx* ctx);
 
 /*! POOL_function :
  *  The function type that can be added to a thread pool.
@@ -70,7 +76,7 @@ void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque);
 
 
 /*! POOL_tryAdd() :
- *  Add the job `function(opaque)` to thread pool _if_ a worker is available.
+ *  Add the job `function(opaque)` to thread pool _if_ a queue slot is available.
  *  Returns immediately even if not (does not block).
  * @return : 1 if successful, 0 if not.
  */