extzstd 0.2 → 0.3

data/contrib/zstd/lib/README.md

@@ -2,15 +2,35 @@ Zstandard library files
 ================================
 
 The __lib__ directory is split into several sub-directories,
-in order to make it easier to select or exclude specific features.
+in order to make it easier to select or exclude features.
 
 
 #### Building
 
-`Makefile` script is provided, supporting the standard set of commands,
-directories, and variables (see https://www.gnu.org/prep/standards/html_node/Command-Variables.html).
+`Makefile` script is provided, supporting [Makefile conventions](https://www.gnu.org/prep/standards/html_node/Makefile-Conventions.html#Makefile-Conventions),
+including commands variables, staged install, directory variables and standard targets.
 - `make` : generates both static and dynamic libraries
-- `make install` : install libraries in default system directories
+- `make install` : install libraries and headers in target system directories
+
+`libzstd` default scope is pretty large, including compression, decompression, dictionary builder,
+and support for decoding legacy formats >= v0.5.0.
+The scope can be reduced on demand (see paragraph _modular build_).
+
+
+#### Multithreading support
+
+Multithreading is disabled by default when building with `make`.
+Enabling multithreading requires 2 conditions :
+- set build macro `ZSTD_MULTITHREAD` (`-DZSTD_MULTITHREAD` for `gcc`)
+- for POSIX systems : compile with pthread (`-pthread` compilation flag for `gcc`)
+
+Both conditions are automatically applied when invoking `make lib-mt` target.
+
+When linking a POSIX program with a multithreaded version of `libzstd`,
+note that it's necessary to request the `-pthread` flag during link stage.
+
+Multithreading capabilities are exposed
+via the [advanced API defined in `lib/zstd.h`](https://github.com/facebook/zstd/blob/v1.3.8/lib/zstd.h#L592).
 
 
 #### API
@@ -23,53 +43,74 @@ Zstandard's stable API is exposed within [lib/zstd.h](zstd.h).
 Optional advanced features are exposed via :
 
 - `lib/common/zstd_errors.h` : translates `size_t` function results
-                              into an `ZSTD_ErrorCode`, for accurate error handling.
+                               into a `ZSTD_ErrorCode`, for accurate error handling.
+
 - `ZSTD_STATIC_LINKING_ONLY` : if this macro is defined _before_ including `zstd.h`,
-                          it unlocks access to advanced experimental API,
-                          exposed in second part of `zstd.h`.
-                          These APIs shall ___never be used with dynamic library___ !
-                          They are not "stable", their definition may change in the future.
+                          it unlocks access to the experimental API,
+                          exposed in the second part of `zstd.h`.
+                          All definitions in the experimental APIs are unstable,
+                          they may still change in the future, or even be removed.
+                          As a consequence, experimental definitions shall ___never be used with dynamic library___ !
                           Only static linking is allowed.
 
 
 #### Modular build
 
-- Directory `lib/common` is always required, for all variants.
-- Compression source code lies in `lib/compress`
-- Decompression source code lies in `lib/decompress`
-- It's possible to include only `compress` or only `decompress`, they don't depend on each other.
-- `lib/dictBuilder` : makes it possible to generate dictionaries from a set of samples.
-    The API is exposed in `lib/dictBuilder/zdict.h`.
-    This module depends on both `lib/common` and `lib/compress` .
-- `lib/legacy` : source code to decompress older zstd formats, starting from `v0.1`.
-              This module depends on `lib/common` and `lib/decompress`.
-              To enable this feature, it's necessary to define `ZSTD_LEGACY_SUPPORT = 1` during compilation.
-              Typically, with `gcc`, add argument `-DZSTD_LEGACY_SUPPORT=1`.
-              Using higher number limits the number of version supported.
-              For example, `ZSTD_LEGACY_SUPPORT=2` means : "support legacy formats starting from v0.2+".
-              The API is exposed in `lib/legacy/zstd_legacy.h`.
-              Each version also provides a (dedicated) set of advanced API.
-              For example, advanced API for version `v0.4` is exposed in `lib/legacy/zstd_v04.h` .
+It's possible to compile only a limited set of features within `libzstd`.
+The file structure is designed to make this selection manually achievable for any build system :
 
+- Directory `lib/common` is always required, for all variants.
 
-#### Multithreading support
+- Compression source code lies in `lib/compress`
 
-Multithreading is disabled by default when building with `make`.
-Enabling multithreading requires 2 conditions :
-- set macro `ZSTD_MULTITHREAD`
-- on POSIX systems : compile with pthread (`-pthread` compilation flag for `gcc` for example)
+- Decompression source code lies in `lib/decompress`
 
-Both conditions are automatically triggered by invoking `make lib-mt` target.
-Note that, when linking a POSIX program with a multithreaded version of `libzstd`,
-it's necessary to trigger `-pthread` flag during link stage.
+- It's possible to include only `compress` or only `decompress`, they don't depend on each other.
 
-Multithreading capabilities are exposed via :
-- private API `lib/compress/zstdmt_compress.h`.
-  Symbols defined in this header are currently exposed in `libzstd`, hence usable.
-  Note however that this API is planned to be locked and remain strictly internal in the future.
-- advanced API `ZSTD_compress_generic()`, defined in `lib/zstd.h`, experimental section.
-  This API is still considered experimental, but is designed to be labelled "stable" at some point in the future.
-  It's the recommended entry point for multi-threading operations.
+- `lib/dictBuilder` : makes it possible to generate dictionaries from a set of samples.
+        The API is exposed in `lib/dictBuilder/zdict.h`.
+        This module depends on both `lib/common` and `lib/compress` .
+
+- `lib/legacy` : makes it possible to decompress legacy zstd formats, starting from `v0.1.0`.
+        This module depends on `lib/common` and `lib/decompress`.
+        To enable this feature, define `ZSTD_LEGACY_SUPPORT` during compilation.
+        Specifying a number limits versions supported to that version onward.
+        For example, `ZSTD_LEGACY_SUPPORT=2` means : "support legacy formats >= v0.2.0".
+        Conversely, `ZSTD_LEGACY_SUPPORT=0` means "do __not__ support legacy formats".
+        By default, this build macro is set as `ZSTD_LEGACY_SUPPORT=5`.
+        Decoding supported legacy format is a transparent capability triggered within decompression functions.
+        It's also allowed to invoke legacy API directly, exposed in `lib/legacy/zstd_legacy.h`.
+        Each version does also provide its own set of advanced API.
+        For example, advanced API for version `v0.4` is exposed in `lib/legacy/zstd_v04.h` .
+
+- While invoking `make libzstd`, it's possible to define build macros
+        `ZSTD_LIB_COMPRESSION, ZSTD_LIB_DECOMPRESSION`, `ZSTD_LIB_DICTBUILDER`,
+        and `ZSTD_LIB_DEPRECATED` as `0` to forgo compilation of the corresponding features.
+        This will also disable compilation of all dependencies
+        (eg. `ZSTD_LIB_COMPRESSION=0` will also disable dictBuilder).
+
+- There are some additional build macros that can be used to minify the decoder.
+
+  Zstandard often has more than one implementation of a piece of functionality,
+  where each implementation optimizes for different scenarios. For example, the
+  Huffman decoder has complementary implementations that decode the stream one
+  symbol at a time or two symbols at a time. Zstd normally includes both (and
+  dispatches between them at runtime), but by defining `HUF_FORCE_DECOMPRESS_X1`
+  or `HUF_FORCE_DECOMPRESS_X2`, you can force the use of one or the other, avoiding
+  compilation of the other. Similarly, `ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT`
+  and `ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG` force the compilation and use of
+  only one or the other of two decompression implementations. The smallest
+  binary is achieved by using `HUF_FORCE_DECOMPRESS_X1` and
+  `ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT`.
+
+  For squeezing the last ounce of size out, you can also define
+  `ZSTD_NO_INLINE`, which disables inlining, and `ZSTD_STRIP_ERROR_STRINGS`,
+  which removes the error messages that are otherwise returned by
+  `ZSTD_getErrorName`.
+
+- While invoking `make libzstd`, the build macro `ZSTD_LEGACY_MULTITHREADED_API=1`
+  will expose the deprecated `ZSTDMT` API exposed by `zstdmt_compress.h` in
+  the shared library, which is now hidden by default.
 
 
 #### Windows : using MinGW+MSYS to create DLL
@@ -92,7 +133,6 @@ The compiled executable will require ZSTD DLL which is available at `dll\libzstd
 
 Obsolete API on their way out are stored in directory `lib/deprecated`.
 At this stage, it contains older streaming prototypes, in `lib/deprecated/zbuff.h`.
-Presence in this directory is temporary.
 These prototypes will be removed in some future version.
 Consider migrating code towards supported streaming API exposed in `zstd.h`.
 
@@ -101,8 +141,8 @@ Consider migrating code towards supported streaming API exposed in `zstd.h`.
 
 The other files are not source code. There are :
 
- - `LICENSE` : contains the BSD license text
- - `Makefile` : `make` script to build and install zstd library (static and dynamic)
  - `BUCK` : support for `buck` build system (https://buckbuild.com/)
- - `libzstd.pc.in` : for `pkg-config` (used in `make install`)
+ - `Makefile` : `make` script to build and install zstd library (static and dynamic)
  - `README.md` : this file
+ - `dll/` : resources directory for Windows compilation
+ - `libzstd.pc.in` : script for `pkg-config` (used in `make install`)

data/contrib/zstd/lib/common/bitstream.h

@@ -1,8 +1,7 @@
 /* ******************************************************************
    bitstream
    Part of FSE library
-   header file (to include)
-   Copyright (C) 2013-2017, Yann Collet.
+   Copyright (C) 2013-present, Yann Collet.
 
    BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
 
@@ -49,21 +48,10 @@ extern "C" {
 *  Dependencies
 ******************************************/
 #include "mem.h"            /* unaligned access routines */
+#include "debug.h"          /* assert(), DEBUGLOG(), RAWLOG() */
 #include "error_private.h"  /* error codes and messages */
 
 
-/*-*************************************
-*  Debug
-***************************************/
-#if defined(BIT_DEBUG) && (BIT_DEBUG>=1)
-#  include <assert.h>
-#else
-#  ifndef assert
-#    define assert(condition) ((void)0)
-#  endif
-#endif
-
-
 /*=========================================
 *  Target specific
 =========================================*/
@@ -83,8 +71,7 @@ extern "C" {
  * A critical property of these streams is that they encode and decode in **reverse** direction.
  * So the first bit sequence you add will be the last to be read, like a LIFO stack.
  */
-typedef struct
-{
+typedef struct {
     size_t bitContainer;
     unsigned bitPos;
     char*  startPtr;
@@ -118,8 +105,7 @@ MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC);
 /*-********************************************
 *  bitStream decoding API (read backward)
 **********************************************/
-typedef struct
-{
+typedef struct {
     size_t   bitContainer;
     unsigned bitsConsumed;
     const char* ptr;
@@ -236,7 +222,8 @@ MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC,
 }
 
 /*! BIT_addBitsFast() :
- *  works only if `value` is _clean_, meaning all high bits above nbBits are 0 */
+ *  works only if `value` is _clean_,
+ *  meaning all high bits above nbBits are 0 */
 MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC,
                                 size_t value, unsigned nbBits)
 {
@@ -352,17 +339,10 @@ MEM_STATIC size_t BIT_getUpperBits(size_t bitContainer, U32 const start)
 
 MEM_STATIC size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits)
 {
-#if defined(__BMI__) && defined(__GNUC__) && __GNUC__*1000+__GNUC_MINOR__ >= 4008  /* experimental */
-#  if defined(__x86_64__)
-    if (sizeof(bitContainer)==8)
-        return _bextr_u64(bitContainer, start, nbBits);
-    else
-#  endif
-        return _bextr_u32(bitContainer, start, nbBits);
-#else
+    U32 const regMask = sizeof(bitContainer)*8 - 1;
+    /* if start > regMask, bitstream is corrupted, and result is undefined */
     assert(nbBits < BIT_MASK_SIZE);
-    return (bitContainer >> start) & BIT_mask[nbBits];
-#endif
+    return (bitContainer >> (start & regMask)) & BIT_mask[nbBits];
 }
 
 MEM_STATIC size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
@@ -379,9 +359,13 @@ MEM_STATIC size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
  * @return : value extracted */
 MEM_STATIC size_t BIT_lookBits(const BIT_DStream_t* bitD, U32 nbBits)
 {
-#if defined(__BMI__) && defined(__GNUC__)   /* experimental; fails if bitD->bitsConsumed + nbBits > sizeof(bitD->bitContainer)*8 */
+    /* arbitrate between double-shift and shift+mask */
+#if 1
+    /* if bitD->bitsConsumed + nbBits > sizeof(bitD->bitContainer)*8,
+     * bitstream is likely corrupted, and result is undefined */
     return BIT_getMiddleBits(bitD->bitContainer, (sizeof(bitD->bitContainer)*8) - bitD->bitsConsumed - nbBits, nbBits);
 #else
+    /* this code path is slower on my os-x laptop */
     U32 const regMask = sizeof(bitD->bitContainer)*8 - 1;
     return ((bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> 1) >> ((regMask-nbBits) & regMask);
 #endif
@@ -405,7 +389,7 @@ 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, U32 nbBits)
+MEM_STATIC size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits)
 {
     size_t const value = BIT_lookBits(bitD, nbBits);
     BIT_skipBits(bitD, nbBits);
@@ -414,7 +398,7 @@ MEM_STATIC size_t BIT_readBits(BIT_DStream_t* bitD, U32 nbBits)
 
 /*! BIT_readBitsFast() :
  *  unsafe version; only works only if nbBits >= 1 */
-MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, U32 nbBits)
+MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits)
 {
     size_t const value = BIT_lookBitsFast(bitD, nbBits);
     assert(nbBits >= 1);
@@ -426,7 +410,7 @@ MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, U32 nbBits)
  *  Refill `bitD` from buffer previously set in BIT_initDStream() .
  *  This function is safe, it guarantees it will not read beyond src buffer.
  * @return : status of `BIT_DStream_t` internal register.
- *           when status == BIT_DStream_unfinished, internal register is filled with at least 25 or 57 bits */
+ *           when status == BIT_DStream_unfinished, internal register is filled with at least 25 or 57 bits */
 MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD)
 {
     if (bitD->bitsConsumed > (sizeof(bitD->bitContainer)*8))  /* overflow detected, like end of stream */

data/contrib/zstd/lib/common/compiler.h

@@ -15,6 +15,8 @@
 *  Compiler specifics
 *********************************************************/
 /* force inlining */
+
+#if !defined(ZSTD_NO_INLINE)
 #if defined (__GNUC__) || defined(__cplusplus) || defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L   /* C99 */
 #  define INLINE_KEYWORD inline
 #else
@@ -29,9 +31,16 @@
 #  define FORCE_INLINE_ATTR
 #endif
 
+#else
+
+#define INLINE_KEYWORD
+#define FORCE_INLINE_ATTR
+
+#endif
+
 /**
  * FORCE_INLINE_TEMPLATE is used to define C "templates", which take constant
- * parameters. They must be inlined for the compiler to elimininate the constant
+ * parameters. They must be inlined for the compiler to eliminate the constant
  * branches.
  */
 #define FORCE_INLINE_TEMPLATE static INLINE_KEYWORD FORCE_INLINE_ATTR
@@ -63,16 +72,61 @@
 #  endif
 #endif
 
-/* prefetch */
-#if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86))  /* _mm_prefetch() is not defined outside of x86/x64 */
-#  include <mmintrin.h>   /* https://msdn.microsoft.com/fr-fr/library/84szxsww(v=vs.90).aspx */
-#  define PREFETCH(ptr)   _mm_prefetch((const char*)ptr, _MM_HINT_T0)
-#elif defined(__GNUC__)
-#  define PREFETCH(ptr)   __builtin_prefetch(ptr, 0, 0)
+/* target attribute */
+#ifndef __has_attribute
+  #define __has_attribute(x) 0  /* Compatibility with non-clang compilers. */
+#endif
+#if defined(__GNUC__)
+#  define TARGET_ATTRIBUTE(target) __attribute__((__target__(target)))
 #else
-#  define PREFETCH(ptr)   /* disabled */
+#  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.
+ */
+#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
+
+/* prefetch
+ * can be disabled, by declaring NO_PREFETCH build macro */
+#if defined(NO_PREFETCH)
+#  define PREFETCH_L1(ptr)  (void)(ptr)  /* disabled */
+#  define PREFETCH_L2(ptr)  (void)(ptr)  /* disabled */
+#else
+#  if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86))  /* _mm_prefetch() is not defined outside of x86/x64 */
+#    include <mmintrin.h>   /* https://msdn.microsoft.com/fr-fr/library/84szxsww(v=vs.90).aspx */
+#    define PREFETCH_L1(ptr)  _mm_prefetch((const char*)(ptr), _MM_HINT_T0)
+#    define PREFETCH_L2(ptr)  _mm_prefetch((const char*)(ptr), _MM_HINT_T1)
+#  elif defined(__GNUC__) && ( (__GNUC__ >= 4) || ( (__GNUC__ == 3) && (__GNUC_MINOR__ >= 1) ) )
+#    define PREFETCH_L1(ptr)  __builtin_prefetch((ptr), 0 /* rw==read */, 3 /* locality */)
+#    define PREFETCH_L2(ptr)  __builtin_prefetch((ptr), 0 /* rw==read */, 2 /* locality */)
+#  else
+#    define PREFETCH_L1(ptr) (void)(ptr)  /* disabled */
+#    define PREFETCH_L2(ptr) (void)(ptr)  /* disabled */
+#  endif
+#endif  /* NO_PREFETCH */
+
+#define CACHELINE_SIZE 64
+
+#define PREFETCH_AREA(p, s)  {            \
+    const char* const _ptr = (const char*)(p);  \
+    size_t const _size = (size_t)(s);     \
+    size_t _pos;                          \
+    for (_pos=0; _pos<_size; _pos+=CACHELINE_SIZE) {  \
+        PREFETCH_L2(_ptr + _pos);         \
+    }                                     \
+}
+
 /* disable warnings */
 #ifdef _MSC_VER    /* Visual Studio */
 #  include <intrin.h>                    /* For Visual 2005 */

data/contrib/zstd/lib/common/cpu.h

@@ -0,0 +1,215 @@
+/*
+ * Copyright (c) 2018-present, Facebook, Inc.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+#ifndef ZSTD_COMMON_CPU_H
+#define ZSTD_COMMON_CPU_H
+
+/**
+ * Implementation taken from folly/CpuId.h
+ * https://github.com/facebook/folly/blob/master/folly/CpuId.h
+ */
+
+#include <string.h>
+
+#include "mem.h"
+
+#ifdef _MSC_VER
+#include <intrin.h>
+#endif
+
+typedef struct {
+    U32 f1c;
+    U32 f1d;
+    U32 f7b;
+    U32 f7c;
+} ZSTD_cpuid_t;
+
+MEM_STATIC ZSTD_cpuid_t ZSTD_cpuid(void) {
+    U32 f1c = 0;
+    U32 f1d = 0;
+    U32 f7b = 0;
+    U32 f7c = 0;
+#if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_IX86))
+    int reg[4];
+    __cpuid((int*)reg, 0);
+    {
+        int const n = reg[0];
+        if (n >= 1) {
+            __cpuid((int*)reg, 1);
+            f1c = (U32)reg[2];
+            f1d = (U32)reg[3];
+        }
+        if (n >= 7) {
+            __cpuidex((int*)reg, 7, 0);
+            f7b = (U32)reg[1];
+            f7c = (U32)reg[2];
+        }
+    }
+#elif defined(__i386__) && defined(__PIC__) && !defined(__clang__) && defined(__GNUC__)
+    /* The following block like the normal cpuid branch below, but gcc
+     * reserves ebx for use of its pic register so we must specially
+     * handle the save and restore to avoid clobbering the register
+     */
+    U32 n;
+    __asm__(
+        "pushl %%ebx\n\t"
+        "cpuid\n\t"
+        "popl %%ebx\n\t"
+        : "=a"(n)
+        : "a"(0)
+        : "ecx", "edx");
+    if (n >= 1) {
+      U32 f1a;
+      __asm__(
+          "pushl %%ebx\n\t"
+          "cpuid\n\t"
+          "popl %%ebx\n\t"
+          : "=a"(f1a), "=c"(f1c), "=d"(f1d)
+          : "a"(1));
+    }
+    if (n >= 7) {
+      __asm__(
+          "pushl %%ebx\n\t"
+          "cpuid\n\t"
+          "movl %%ebx, %%eax\n\t"
+          "popl %%ebx"
+          : "=a"(f7b), "=c"(f7c)
+          : "a"(7), "c"(0)
+          : "edx");
+    }
+#elif defined(__x86_64__) || defined(_M_X64) || defined(__i386__)
+    U32 n;
+    __asm__("cpuid" : "=a"(n) : "a"(0) : "ebx", "ecx", "edx");
+    if (n >= 1) {
+      U32 f1a;
+      __asm__("cpuid" : "=a"(f1a), "=c"(f1c), "=d"(f1d) : "a"(1) : "ebx");
+    }
+    if (n >= 7) {
+      U32 f7a;
+      __asm__("cpuid"
+              : "=a"(f7a), "=b"(f7b), "=c"(f7c)
+              : "a"(7), "c"(0)
+              : "edx");
+    }
+#endif
+    {
+        ZSTD_cpuid_t cpuid;
+        cpuid.f1c = f1c;
+        cpuid.f1d = f1d;
+        cpuid.f7b = f7b;
+        cpuid.f7c = f7c;
+        return cpuid;
+    }
+}
+
+#define X(name, r, bit)                                                        \
+  MEM_STATIC int ZSTD_cpuid_##name(ZSTD_cpuid_t const cpuid) {                 \
+    return ((cpuid.r) & (1U << bit)) != 0;                                     \
+  }
+
+/* cpuid(1): Processor Info and Feature Bits. */
+#define C(name, bit) X(name, f1c, bit)
+  C(sse3, 0)
+  C(pclmuldq, 1)
+  C(dtes64, 2)
+  C(monitor, 3)
+  C(dscpl, 4)
+  C(vmx, 5)
+  C(smx, 6)
+  C(eist, 7)
+  C(tm2, 8)
+  C(ssse3, 9)
+  C(cnxtid, 10)
+  C(fma, 12)
+  C(cx16, 13)
+  C(xtpr, 14)
+  C(pdcm, 15)
+  C(pcid, 17)
+  C(dca, 18)
+  C(sse41, 19)
+  C(sse42, 20)
+  C(x2apic, 21)
+  C(movbe, 22)
+  C(popcnt, 23)
+  C(tscdeadline, 24)
+  C(aes, 25)
+  C(xsave, 26)
+  C(osxsave, 27)
+  C(avx, 28)
+  C(f16c, 29)
+  C(rdrand, 30)
+#undef C
+#define D(name, bit) X(name, f1d, bit)
+  D(fpu, 0)
+  D(vme, 1)
+  D(de, 2)
+  D(pse, 3)
+  D(tsc, 4)
+  D(msr, 5)
+  D(pae, 6)
+  D(mce, 7)
+  D(cx8, 8)
+  D(apic, 9)
+  D(sep, 11)
+  D(mtrr, 12)
+  D(pge, 13)
+  D(mca, 14)
+  D(cmov, 15)
+  D(pat, 16)
+  D(pse36, 17)
+  D(psn, 18)
+  D(clfsh, 19)
+  D(ds, 21)
+  D(acpi, 22)
+  D(mmx, 23)
+  D(fxsr, 24)
+  D(sse, 25)
+  D(sse2, 26)
+  D(ss, 27)
+  D(htt, 28)
+  D(tm, 29)
+  D(pbe, 31)
+#undef D
+
+/* cpuid(7): Extended Features. */
+#define B(name, bit) X(name, f7b, bit)
+  B(bmi1, 3)
+  B(hle, 4)
+  B(avx2, 5)
+  B(smep, 7)
+  B(bmi2, 8)
+  B(erms, 9)
+  B(invpcid, 10)
+  B(rtm, 11)
+  B(mpx, 14)
+  B(avx512f, 16)
+  B(avx512dq, 17)
+  B(rdseed, 18)
+  B(adx, 19)
+  B(smap, 20)
+  B(avx512ifma, 21)
+  B(pcommit, 22)
+  B(clflushopt, 23)
+  B(clwb, 24)
+  B(avx512pf, 26)
+  B(avx512er, 27)
+  B(avx512cd, 28)
+  B(sha, 29)
+  B(avx512bw, 30)
+  B(avx512vl, 31)
+#undef B
+#define C(name, bit) X(name, f7c, bit)
+  C(prefetchwt1, 0)
+  C(avx512vbmi, 1)
+#undef C
+
+#undef X
+
+#endif /* ZSTD_COMMON_CPU_H */