zstd-ruby 1.4.5.0 → 1.5.1.1

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

@@ -1,7 +1,7 @@
 /* ******************************************************************
  * bitstream
  * Part of FSE library
- * Copyright (c) 2013-2020, Yann Collet, Facebook, Inc.
+ * Copyright (c) Yann Collet, Facebook, Inc.
  *
  * You can contact the author at :
  * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
@@ -17,7 +17,6 @@
 #if defined (__cplusplus)
 extern "C" {
 #endif
-
 /*
 *  This API consists of small unitary functions, which must be inlined for best performance.
 *  Since link-time-optimization is not available for all compilers,
@@ -36,10 +35,12 @@ extern "C" {
 /*=========================================
 *  Target specific
 =========================================*/
-#if defined(__BMI__) && defined(__GNUC__)
-#  include <immintrin.h>   /* support for bextr (experimental) */
-#elif defined(__ICCARM__)
-#  include <intrinsics.h>
+#ifndef ZSTD_NO_INTRINSICS
+#  if defined(__BMI__) && defined(__GNUC__)
+#    include <immintrin.h>   /* support for bextr (experimental) */
+#  elif defined(__ICCARM__)
+#    include <intrinsics.h>
+#  endif
 #endif
 
 #define STREAM_ACCUMULATOR_MIN_32  25
@@ -141,8 +142,18 @@ MEM_STATIC unsigned BIT_highbit32 (U32 val)
     assert(val != 0);
     {
 #   if defined(_MSC_VER)   /* Visual */
-        unsigned long r=0;
-        return _BitScanReverse ( &r, val ) ? (unsigned)r : 0;
+#       if STATIC_BMI2 == 1
+            return _lzcnt_u32(val) ^ 31;
+#       else
+            if (val != 0) {
+                unsigned long r;
+                _BitScanReverse(&r, val);
+                return (unsigned)r;
+            } else {
+                /* Should not reach this code path */
+                __assume(0);
+            }
+#       endif
 #   elif defined(__GNUC__) && (__GNUC__ >= 3)   /* Use GCC Intrinsic */
         return __builtin_clz (val) ^ 31;
 #   elif defined(__ICCARM__)    /* IAR Intrinsic */
@@ -198,7 +209,7 @@ MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC,
 MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC,
                             size_t value, unsigned nbBits)
 {
-    MEM_STATIC_ASSERT(BIT_MASK_SIZE == 32);
+    DEBUG_STATIC_ASSERT(BIT_MASK_SIZE == 32);
     assert(nbBits < BIT_MASK_SIZE);
     assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8);
     bitC->bitContainer |= (value & BIT_mask[nbBits]) << bitC->bitPos;
@@ -271,7 +282,7 @@ MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC)
  */
 MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize)
 {
-    if (srcSize < 1) { memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); }
+    if (srcSize < 1) { ZSTD_memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); }
 
     bitD->start = (const char*)srcBuffer;
     bitD->limitPtr = bitD->start + sizeof(bitD->bitContainer);
@@ -288,22 +299,22 @@ MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, si
         switch(srcSize)
         {
         case 7: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[6]) << (sizeof(bitD->bitContainer)*8 - 16);
-                /* fall-through */
+                ZSTD_FALLTHROUGH;
 
         case 6: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[5]) << (sizeof(bitD->bitContainer)*8 - 24);
-                /* fall-through */
+                ZSTD_FALLTHROUGH;
 
         case 5: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[4]) << (sizeof(bitD->bitContainer)*8 - 32);
-                /* fall-through */
+                ZSTD_FALLTHROUGH;
 
         case 4: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[3]) << 24;
-                /* fall-through */
+                ZSTD_FALLTHROUGH;
 
         case 3: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[2]) << 16;
-                /* fall-through */
+                ZSTD_FALLTHROUGH;
 
         case 2: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[1]) <<  8;
-                /* fall-through */
+                ZSTD_FALLTHROUGH;
 
         default: break;
         }
@@ -317,23 +328,36 @@ MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, si
     return srcSize;
 }
 
-MEM_STATIC size_t BIT_getUpperBits(size_t bitContainer, U32 const start)
+MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getUpperBits(size_t bitContainer, U32 const start)
 {
     return bitContainer >> start;
 }
 
-MEM_STATIC size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits)
+MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits)
 {
     U32 const regMask = sizeof(bitContainer)*8 - 1;
     /* if start > regMask, bitstream is corrupted, and result is undefined */
     assert(nbBits < BIT_MASK_SIZE);
+    /* x86 transform & ((1 << nbBits) - 1) to bzhi instruction, it is better
+     * than accessing memory. When bmi2 instruction is not present, we consider
+     * such cpus old (pre-Haswell, 2013) and their performance is not of that
+     * importance.
+     */
+#if defined(__x86_64__) || defined(_M_X86)
+    return (bitContainer >> (start & regMask)) & ((((U64)1) << nbBits) - 1);
+#else
     return (bitContainer >> (start & regMask)) & BIT_mask[nbBits];
+#endif
 }
 
-MEM_STATIC size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
+MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
 {
+#if defined(STATIC_BMI2) && STATIC_BMI2 == 1
+	return  _bzhi_u64(bitContainer, nbBits);
+#else
     assert(nbBits < BIT_MASK_SIZE);
     return bitContainer & BIT_mask[nbBits];
+#endif
 }
 
 /*! BIT_lookBits() :
@@ -342,7 +366,7 @@ MEM_STATIC size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
  *  On 32-bits, maxNbBits==24.
  *  On 64-bits, maxNbBits==56.
  * @return : value extracted */
-MEM_STATIC size_t BIT_lookBits(const BIT_DStream_t* bitD, U32 nbBits)
+MEM_STATIC  FORCE_INLINE_ATTR size_t BIT_lookBits(const BIT_DStream_t*  bitD, U32 nbBits)
 {
     /* arbitrate between double-shift and shift+mask */
 #if 1
@@ -365,7 +389,7 @@ MEM_STATIC size_t BIT_lookBitsFast(const BIT_DStream_t* bitD, U32 nbBits)
     return (bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> (((regMask+1)-nbBits) & regMask);
 }
 
-MEM_STATIC void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits)
+MEM_STATIC FORCE_INLINE_ATTR void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits)
 {
     bitD->bitsConsumed += nbBits;
 }
@@ -374,7 +398,7 @@ MEM_STATIC void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits)
  *  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, unsigned nbBits)
+MEM_STATIC FORCE_INLINE_ATTR size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits)
 {
     size_t const value = BIT_lookBits(bitD, nbBits);
     BIT_skipBits(bitD, nbBits);

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

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2020, Yann Collet, Facebook, Inc.
+ * Copyright (c) Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -11,6 +11,8 @@
 #ifndef ZSTD_COMPILER_H
 #define ZSTD_COMPILER_H
 
+#include "portability_macros.h"
+
 /*-*******************************************************
 *  Compiler specifics
 *********************************************************/
@@ -38,6 +40,17 @@
 
 #endif
 
+/**
+  On MSVC qsort requires that functions passed into it use the __cdecl calling conversion(CC).
+  This explicitly marks such functions as __cdecl so that the code will still compile
+  if a CC other than __cdecl has been made the default.
+*/
+#if  defined(_MSC_VER)
+#  define WIN_CDECL __cdecl
+#else
+#  define WIN_CDECL
+#endif
+
 /**
  * FORCE_INLINE_TEMPLATE is used to define C "templates", which take constant
  * parameters. They must be inlined for the compiler to eliminate the constant
@@ -79,30 +92,19 @@
 #  endif
 #endif
 
+
 /* target attribute */
-#ifndef __has_attribute
-  #define __has_attribute(x) 0  /* Compatibility with non-clang compilers. */
-#endif
 #if defined(__GNUC__) || defined(__ICCARM__)
 #  define TARGET_ATTRIBUTE(target) __attribute__((__target__(target)))
 #else
 #  define TARGET_ATTRIBUTE(target)
 #endif
 
-/* Enable runtime BMI2 dispatch based on the CPU.
- * Enabled for clang & gcc >=4.8 on x86 when BMI2 isn't enabled by default.
+/* Target attribute for BMI2 dynamic dispatch.
+ * Enable lzcnt, bmi, and bmi2.
+ * We test for bmi1 & bmi2. lzcnt is included in bmi1.
  */
-#ifndef DYNAMIC_BMI2
-  #if ((defined(__clang__) && __has_attribute(__target__)) \
-      || (defined(__GNUC__) \
-          && (__GNUC__ >= 5 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 8)))) \
-      && (defined(__x86_64__) || defined(_M_X86)) \
-      && !defined(__BMI2__)
-  #  define DYNAMIC_BMI2 1
-  #else
-  #  define DYNAMIC_BMI2 0
-  #endif
-#endif
+#define BMI2_TARGET_ATTRIBUTE TARGET_ATTRIBUTE("lzcnt,bmi,bmi2")
 
 /* prefetch
  * can be disabled, by declaring NO_PREFETCH build macro */
@@ -114,12 +116,12 @@
 #    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(__aarch64__)
-#     define PREFETCH_L1(ptr)  __asm__ __volatile__("prfm pldl1keep, %0" ::"Q"(*(ptr)))
-#     define PREFETCH_L2(ptr)  __asm__ __volatile__("prfm pldl2keep, %0" ::"Q"(*(ptr)))
 #  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 */)
+#  elif defined(__aarch64__)
+#    define PREFETCH_L1(ptr)  __asm__ __volatile__("prfm pldl1keep, %0" ::"Q"(*(ptr)))
+#    define PREFETCH_L2(ptr)  __asm__ __volatile__("prfm pldl2keep, %0" ::"Q"(*(ptr)))
 #  else
 #    define PREFETCH_L1(ptr) (void)(ptr)  /* disabled */
 #    define PREFETCH_L2(ptr) (void)(ptr)  /* disabled */
@@ -138,8 +140,9 @@
 }
 
 /* vectorization
- * older GCC (pre gcc-4.3 picked as the cutoff) uses a different syntax */
-#if !defined(__INTEL_COMPILER) && !defined(__clang__) && defined(__GNUC__)
+ * older GCC (pre gcc-4.3 picked as the cutoff) uses a different syntax,
+ * and some compilers, like Intel ICC and MCST LCC, do not support it at all. */
+#if !defined(__INTEL_COMPILER) && !defined(__clang__) && defined(__GNUC__) && !defined(__LCC__)
 #  if (__GNUC__ == 4 && __GNUC_MINOR__ > 3) || (__GNUC__ >= 5)
 #    define DONT_VECTORIZE __attribute__((optimize("no-tree-vectorize")))
 #  else
@@ -172,4 +175,161 @@
 #  pragma warning(disable : 4324)        /* disable: C4324: padded structure */
 #endif
 
+/*Like DYNAMIC_BMI2 but for compile time determination of BMI2 support*/
+#ifndef STATIC_BMI2
+#  if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86))
+#    ifdef __AVX2__  //MSVC does not have a BMI2 specific flag, but every CPU that supports AVX2 also supports BMI2
+#       define STATIC_BMI2 1
+#    endif
+#  endif
+#endif
+
+#ifndef STATIC_BMI2
+    #define STATIC_BMI2 0
+#endif
+
+/* compile time determination of SIMD support */
+#if !defined(ZSTD_NO_INTRINSICS)
+#  if defined(__SSE2__) || defined(_M_AMD64) || (defined (_M_IX86) && defined(_M_IX86_FP) && (_M_IX86_FP >= 2))
+#    define ZSTD_ARCH_X86_SSE2
+#  endif
+#  if defined(__ARM_NEON) || defined(_M_ARM64)
+#    define ZSTD_ARCH_ARM_NEON
+#  endif
+#
+#  if defined(ZSTD_ARCH_X86_SSE2)
+#    include <emmintrin.h>
+#  elif defined(ZSTD_ARCH_ARM_NEON)
+#    include <arm_neon.h>
+#  endif
+#endif
+
+/* C-language Attributes are added in C23. */
+#if defined(__STDC_VERSION__) && (__STDC_VERSION__ > 201710L) && defined(__has_c_attribute)
+# define ZSTD_HAS_C_ATTRIBUTE(x) __has_c_attribute(x)
+#else
+# define ZSTD_HAS_C_ATTRIBUTE(x) 0
+#endif
+
+/* Only use C++ attributes in C++. Some compilers report support for C++
+ * attributes when compiling with C.
+ */
+#if defined(__cplusplus) && defined(__has_cpp_attribute)
+# define ZSTD_HAS_CPP_ATTRIBUTE(x) __has_cpp_attribute(x)
+#else
+# define ZSTD_HAS_CPP_ATTRIBUTE(x) 0
+#endif
+
+/* Define ZSTD_FALLTHROUGH macro for annotating switch case with the 'fallthrough' attribute.
+ * - C23: https://en.cppreference.com/w/c/language/attributes/fallthrough
+ * - CPP17: https://en.cppreference.com/w/cpp/language/attributes/fallthrough
+ * - Else: __attribute__((__fallthrough__))
+ */
+#ifndef ZSTD_FALLTHROUGH
+# if ZSTD_HAS_C_ATTRIBUTE(fallthrough)
+#  define ZSTD_FALLTHROUGH [[fallthrough]]
+# elif ZSTD_HAS_CPP_ATTRIBUTE(fallthrough)
+#  define ZSTD_FALLTHROUGH [[fallthrough]]
+# elif __has_attribute(__fallthrough__)
+/* Leading semicolon is to satisfy gcc-11 with -pedantic. Without the semicolon
+ * gcc complains about: a label can only be part of a statement and a declaration is not a statement.
+ */
+#  define ZSTD_FALLTHROUGH ; __attribute__((__fallthrough__))
+# else
+#  define ZSTD_FALLTHROUGH
+# endif
+#endif
+
+/*-**************************************************************
+*  Alignment check
+*****************************************************************/
+
+/* this test was initially positioned in mem.h,
+ * but this file is removed (or replaced) for linux kernel
+ * so it's now hosted in compiler.h,
+ * which remains valid for both user & kernel spaces.
+ */
+
+#ifndef ZSTD_ALIGNOF
+# if defined(__GNUC__) || defined(_MSC_VER)
+/* covers gcc, clang & MSVC */
+/* note : this section must come first, before C11,
+ * due to a limitation in the kernel source generator */
+#  define ZSTD_ALIGNOF(T) __alignof(T)
+
+# elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L)
+/* C11 support */
+#  include <stdalign.h>
+#  define ZSTD_ALIGNOF(T) alignof(T)
+
+# else
+/* No known support for alignof() - imperfect backup */
+#  define ZSTD_ALIGNOF(T) (sizeof(void*) < sizeof(T) ? sizeof(void*) : sizeof(T))
+
+# endif
+#endif /* ZSTD_ALIGNOF */
+
+/*-**************************************************************
+*  Sanitizer
+*****************************************************************/
+
+#if ZSTD_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 <stddef.h>  /* size_t */
+#define ZSTD_DEPS_NEED_STDINT
+#include "zstd_deps.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
+
+#if ZSTD_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... */
+#include <stddef.h>  /* size_t */
+
+/**
+ * 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
+
 #endif /* ZSTD_COMPILER_H */

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

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2020, Facebook, Inc.
+ * Copyright (c) Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -16,8 +16,6 @@
  * https://github.com/facebook/folly/blob/master/folly/CpuId.h
  */
 
-#include <string.h>
-
 #include "mem.h"
 
 #ifdef _MSC_VER

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

@@ -1,7 +1,7 @@
 /* ******************************************************************
  * debug
  * Part of FSE library
- * Copyright (c) 2013-2020, Yann Collet, Facebook, Inc.
+ * Copyright (c) Yann Collet, Facebook, Inc.
  *
  * You can contact the author at :
  * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy

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

@@ -1,7 +1,7 @@
 /* ******************************************************************
  * debug
  * Part of FSE library
- * Copyright (c) 2013-2020, Yann Collet, Facebook, Inc.
+ * Copyright (c) Yann Collet, Facebook, Inc.
  *
  * You can contact the author at :
  * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
@@ -51,15 +51,6 @@ extern "C" {
 #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
@@ -76,7 +67,8 @@ extern "C" {
  */
 
 #if (DEBUGLEVEL>=1)
-#  include <assert.h>
+#  define ZSTD_DEPS_NEED_ASSERT
+#  include "zstd_deps.h"
 #else
 #  ifndef assert   /* assert may be already defined, due to prior #include <assert.h> */
 #    define assert(condition) ((void)0)   /* disable assert (default) */
@@ -84,7 +76,8 @@ extern "C" {
 #endif
 
 #if (DEBUGLEVEL>=2)
-#  include <stdio.h>
+#  define ZSTD_DEPS_NEED_IO
+#  include "zstd_deps.h"
 extern int g_debuglevel; /* the variable is only declared,
                             it actually lives in debug.c,
                             and is shared by the whole process.
@@ -92,14 +85,14 @@ extern int g_debuglevel; /* the variable is only declared,
                             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 RAWLOG(l, ...) {                                       \
+                if (l<=g_debuglevel) {                           \
+                    ZSTD_DEBUG_PRINT(__VA_ARGS__);               \
             }   }
-#  define DEBUGLOG(l, ...) {                                    \
-                if (l<=g_debuglevel) {                          \
-                    fprintf(stderr, __FILE__ ": " __VA_ARGS__); \
-                    fprintf(stderr, " \n");                     \
+#  define DEBUGLOG(l, ...) {                                     \
+                if (l<=g_debuglevel) {                           \
+                    ZSTD_DEBUG_PRINT(__FILE__ ": " __VA_ARGS__); \
+                    ZSTD_DEBUG_PRINT(" \n");                     \
             }   }
 #else
 #  define RAWLOG(l, ...)      {}    /* disabled */