1<html lang="en"> 2<head> 3<title>Common Predefined Macros - The C Preprocessor</title> 4<meta http-equiv="Content-Type" content="text/html"> 5<meta name="description" content="The C Preprocessor"> 6<meta name="generator" content="makeinfo 4.13"> 7<link title="Top" rel="start" href="index.html#Top"> 8<link rel="up" href="Predefined-Macros.html#Predefined-Macros" title="Predefined Macros"> 9<link rel="prev" href="Standard-Predefined-Macros.html#Standard-Predefined-Macros" title="Standard Predefined Macros"> 10<link rel="next" href="System_002dspecific-Predefined-Macros.html#System_002dspecific-Predefined-Macros" title="System-specific Predefined Macros"> 11<link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage"> 12<!-- 13Copyright (C) 1987, 1989, 1991, 1992, 1993, 1994, 1995, 1996, 141997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 152008, 2009, 2010, 2011 16Free Software Foundation, Inc. 17 18Permission is granted to copy, distribute and/or modify this document 19under the terms of the GNU Free Documentation License, Version 1.3 or 20any later version published by the Free Software Foundation. A copy of 21the license is included in the 22section entitled ``GNU Free Documentation License''. 23 24This manual contains no Invariant Sections. The Front-Cover Texts are 25(a) (see below), and the Back-Cover Texts are (b) (see below). 26 27(a) The FSF's Front-Cover Text is: 28 29 A GNU Manual 30 31(b) The FSF's Back-Cover Text is: 32 33 You have freedom to copy and modify this GNU Manual, like GNU 34 software. Copies published by the Free Software Foundation raise 35 funds for GNU development. 36--> 37<meta http-equiv="Content-Style-Type" content="text/css"> 38<style type="text/css"><!-- 39 pre.display { font-family:inherit } 40 pre.format { font-family:inherit } 41 pre.smalldisplay { font-family:inherit; font-size:smaller } 42 pre.smallformat { font-family:inherit; font-size:smaller } 43 pre.smallexample { font-size:smaller } 44 pre.smalllisp { font-size:smaller } 45 span.sc { font-variant:small-caps } 46 span.roman { font-family:serif; font-weight:normal; } 47 span.sansserif { font-family:sans-serif; font-weight:normal; } 48--></style> 49<link rel="stylesheet" type="text/css" href="../cs.css"> 50</head> 51<body> 52<div class="node"> 53<a name="Common-Predefined-Macros"></a> 54<p> 55Next: <a rel="next" accesskey="n" href="System_002dspecific-Predefined-Macros.html#System_002dspecific-Predefined-Macros">System-specific Predefined Macros</a>, 56Previous: <a rel="previous" accesskey="p" href="Standard-Predefined-Macros.html#Standard-Predefined-Macros">Standard Predefined Macros</a>, 57Up: <a rel="up" accesskey="u" href="Predefined-Macros.html#Predefined-Macros">Predefined Macros</a> 58<hr> 59</div> 60 61<h4 class="subsection">3.7.2 Common Predefined Macros</h4> 62 63<p><a name="index-common-predefined-macros-61"></a> 64The common predefined macros are GNU C extensions. They are available 65with the same meanings regardless of the machine or operating system on 66which you are using GNU C or GNU Fortran. Their names all start with 67double underscores. 68 69 <dl> 70<dt><code>__COUNTER__</code><dd>This macro expands to sequential integral values starting from 0. In 71conjunction with the <code>##</code> operator, this provides a convenient means to 72generate unique identifiers. Care must be taken to ensure that 73<code>__COUNTER__</code> is not expanded prior to inclusion of precompiled headers 74which use it. Otherwise, the precompiled headers will not be used. 75 76 <br><dt><code>__GFORTRAN__</code><dd>The GNU Fortran compiler defines this. 77 78 <br><dt><code>__GNUC__</code><dt><code>__GNUC_MINOR__</code><dt><code>__GNUC_PATCHLEVEL__</code><dd>These macros are defined by all GNU compilers that use the C 79preprocessor: C, C++, Objective-C and Fortran. Their values are the major 80version, minor version, and patch level of the compiler, as integer 81constants. For example, GCC 3.2.1 will define <code>__GNUC__</code> to 3, 82<code>__GNUC_MINOR__</code> to 2, and <code>__GNUC_PATCHLEVEL__</code> to 1. These 83macros are also defined if you invoke the preprocessor directly. 84 85 <p><code>__GNUC_PATCHLEVEL__</code> is new to GCC 3.0; it is also present in the 86widely-used development snapshots leading up to 3.0 (which identify 87themselves as GCC 2.96 or 2.97, depending on which snapshot you have). 88 89 <p>If all you need to know is whether or not your program is being compiled 90by GCC, or a non-GCC compiler that claims to accept the GNU C dialects, 91you can simply test <code>__GNUC__</code>. If you need to write code 92which depends on a specific version, you must be more careful. Each 93time the minor version is increased, the patch level is reset to zero; 94each time the major version is increased (which happens rarely), the 95minor version and patch level are reset. If you wish to use the 96predefined macros directly in the conditional, you will need to write it 97like this: 98 99 <pre class="smallexample"> /* <span class="roman">Test for GCC > 3.2.0</span> */ 100 #if __GNUC__ > 3 || \ 101 (__GNUC__ == 3 && (__GNUC_MINOR__ > 2 || \ 102 (__GNUC_MINOR__ == 2 && \ 103 __GNUC_PATCHLEVEL__ > 0)) 104</pre> 105 <p class="noindent">Another approach is to use the predefined macros to 106calculate a single number, then compare that against a threshold: 107 108 <pre class="smallexample"> #define GCC_VERSION (__GNUC__ * 10000 \ 109 + __GNUC_MINOR__ * 100 \ 110 + __GNUC_PATCHLEVEL__) 111 ... 112 /* <span class="roman">Test for GCC > 3.2.0</span> */ 113 #if GCC_VERSION > 30200 114</pre> 115 <p class="noindent">Many people find this form easier to understand. 116 117 <br><dt><code>__GNUG__</code><dd>The GNU C++ compiler defines this. Testing it is equivalent to 118testing <code>(__GNUC__ && __cplusplus)<!-- /@w --></code>. 119 120 <br><dt><code>__STRICT_ANSI__</code><dd>GCC defines this macro if and only if the <samp><span class="option">-ansi</span></samp> switch, or a 121<samp><span class="option">-std</span></samp> switch specifying strict conformance to some version of ISO C, 122was specified when GCC was invoked. It is defined to ‘<samp><span class="samp">1</span></samp>’. 123This macro exists primarily to direct GNU libc's header files to 124restrict their definitions to the minimal set found in the 1989 C 125standard. 126 127 <br><dt><code>__BASE_FILE__</code><dd>This macro expands to the name of the main input file, in the form 128of a C string constant. This is the source file that was specified 129on the command line of the preprocessor or C compiler. 130 131 <br><dt><code>__INCLUDE_LEVEL__</code><dd>This macro expands to a decimal integer constant that represents the 132depth of nesting in include files. The value of this macro is 133incremented on every ‘<samp><span class="samp">#include</span></samp>’ directive and decremented at the 134end of every included file. It starts out at 0, its value within the 135base file specified on the command line. 136 137 <br><dt><code>__ELF__</code><dd>This macro is defined if the target uses the ELF object format. 138 139 <br><dt><code>__VERSION__</code><dd>This macro expands to a string constant which describes the version of 140the compiler in use. You should not rely on its contents having any 141particular form, but it can be counted on to contain at least the 142release number. 143 144 <br><dt><code>__OPTIMIZE__</code><dt><code>__OPTIMIZE_SIZE__</code><dt><code>__NO_INLINE__</code><dd>These macros describe the compilation mode. <code>__OPTIMIZE__</code> is 145defined in all optimizing compilations. <code>__OPTIMIZE_SIZE__</code> is 146defined if the compiler is optimizing for size, not speed. 147<code>__NO_INLINE__</code> is defined if no functions will be inlined into 148their callers (when not optimizing, or when inlining has been 149specifically disabled by <samp><span class="option">-fno-inline</span></samp>). 150 151 <p>These macros cause certain GNU header files to provide optimized 152definitions, using macros or inline functions, of system library 153functions. You should not use these macros in any way unless you make 154sure that programs will execute with the same effect whether or not they 155are defined. If they are defined, their value is 1. 156 157 <br><dt><code>__GNUC_GNU_INLINE__</code><dd>GCC defines this macro if functions declared <code>inline</code> will be 158handled in GCC's traditional gnu90 mode. Object files will contain 159externally visible definitions of all functions declared <code>inline</code> 160without <code>extern</code> or <code>static</code>. They will not contain any 161definitions of any functions declared <code>extern inline</code>. 162 163 <br><dt><code>__GNUC_STDC_INLINE__</code><dd>GCC defines this macro if functions declared <code>inline</code> will be 164handled according to the ISO C99 standard. Object files will contain 165externally visible definitions of all functions declared <code>extern 166inline</code>. They will not contain definitions of any functions declared 167<code>inline</code> without <code>extern</code>. 168 169 <p>If this macro is defined, GCC supports the <code>gnu_inline</code> function 170attribute as a way to always get the gnu90 behavior. Support for 171this and <code>__GNUC_GNU_INLINE__</code> was added in GCC 4.1.3. If 172neither macro is defined, an older version of GCC is being used: 173<code>inline</code> functions will be compiled in gnu90 mode, and the 174<code>gnu_inline</code> function attribute will not be recognized. 175 176 <br><dt><code>__CHAR_UNSIGNED__</code><dd>GCC defines this macro if and only if the data type <code>char</code> is 177unsigned on the target machine. It exists to cause the standard header 178file <samp><span class="file">limits.h</span></samp> to work correctly. You should not use this macro 179yourself; instead, refer to the standard macros defined in <samp><span class="file">limits.h</span></samp>. 180 181 <br><dt><code>__WCHAR_UNSIGNED__</code><dd>Like <code>__CHAR_UNSIGNED__</code>, this macro is defined if and only if the 182data type <code>wchar_t</code> is unsigned and the front-end is in C++ mode. 183 184 <br><dt><code>__REGISTER_PREFIX__</code><dd>This macro expands to a single token (not a string constant) which is 185the prefix applied to CPU register names in assembly language for this 186target. You can use it to write assembly that is usable in multiple 187environments. For example, in the <code>m68k-aout</code> environment it 188expands to nothing, but in the <code>m68k-coff</code> environment it expands 189to a single ‘<samp><span class="samp">%</span></samp>’. 190 191 <br><dt><code>__USER_LABEL_PREFIX__</code><dd>This macro expands to a single token which is the prefix applied to 192user labels (symbols visible to C code) in assembly. For example, in 193the <code>m68k-aout</code> environment it expands to an ‘<samp><span class="samp">_</span></samp>’, but in the 194<code>m68k-coff</code> environment it expands to nothing. 195 196 <p>This macro will have the correct definition even if 197<samp><span class="option">-f(no-)underscores</span></samp> is in use, but it will not be correct if 198target-specific options that adjust this prefix are used (e.g. the 199OSF/rose <samp><span class="option">-mno-underscores</span></samp> option). 200 201 <br><dt><code>__SIZE_TYPE__</code><dt><code>__PTRDIFF_TYPE__</code><dt><code>__WCHAR_TYPE__</code><dt><code>__WINT_TYPE__</code><dt><code>__INTMAX_TYPE__</code><dt><code>__UINTMAX_TYPE__</code><dt><code>__SIG_ATOMIC_TYPE__</code><dt><code>__INT8_TYPE__</code><dt><code>__INT16_TYPE__</code><dt><code>__INT32_TYPE__</code><dt><code>__INT64_TYPE__</code><dt><code>__UINT8_TYPE__</code><dt><code>__UINT16_TYPE__</code><dt><code>__UINT32_TYPE__</code><dt><code>__UINT64_TYPE__</code><dt><code>__INT_LEAST8_TYPE__</code><dt><code>__INT_LEAST16_TYPE__</code><dt><code>__INT_LEAST32_TYPE__</code><dt><code>__INT_LEAST64_TYPE__</code><dt><code>__UINT_LEAST8_TYPE__</code><dt><code>__UINT_LEAST16_TYPE__</code><dt><code>__UINT_LEAST32_TYPE__</code><dt><code>__UINT_LEAST64_TYPE__</code><dt><code>__INT_FAST8_TYPE__</code><dt><code>__INT_FAST16_TYPE__</code><dt><code>__INT_FAST32_TYPE__</code><dt><code>__INT_FAST64_TYPE__</code><dt><code>__UINT_FAST8_TYPE__</code><dt><code>__UINT_FAST16_TYPE__</code><dt><code>__UINT_FAST32_TYPE__</code><dt><code>__UINT_FAST64_TYPE__</code><dt><code>__INTPTR_TYPE__</code><dt><code>__UINTPTR_TYPE__</code><dd>These macros are defined to the correct underlying types for the 202<code>size_t</code>, <code>ptrdiff_t</code>, <code>wchar_t</code>, <code>wint_t</code>, 203<code>intmax_t</code>, <code>uintmax_t</code>, <code>sig_atomic_t</code>, <code>int8_t</code>, 204<code>int16_t</code>, <code>int32_t</code>, <code>int64_t</code>, <code>uint8_t</code>, 205<code>uint16_t</code>, <code>uint32_t</code>, <code>uint64_t</code>, 206<code>int_least8_t</code>, <code>int_least16_t</code>, <code>int_least32_t</code>, 207<code>int_least64_t</code>, <code>uint_least8_t</code>, <code>uint_least16_t</code>, 208<code>uint_least32_t</code>, <code>uint_least64_t</code>, <code>int_fast8_t</code>, 209<code>int_fast16_t</code>, <code>int_fast32_t</code>, <code>int_fast64_t</code>, 210<code>uint_fast8_t</code>, <code>uint_fast16_t</code>, <code>uint_fast32_t</code>, 211<code>uint_fast64_t</code>, <code>intptr_t</code>, and <code>uintptr_t</code> typedefs, 212respectively. They exist to make the standard header files 213<samp><span class="file">stddef.h</span></samp>, <samp><span class="file">stdint.h</span></samp>, and <samp><span class="file">wchar.h</span></samp> work correctly. 214You should not use these macros directly; instead, include the 215appropriate headers and use the typedefs. Some of these macros may 216not be defined on particular systems if GCC does not provide a 217<samp><span class="file">stdint.h</span></samp> header on those systems. 218 219 <br><dt><code>__CHAR_BIT__</code><dd>Defined to the number of bits used in the representation of the 220<code>char</code> data type. It exists to make the standard header given 221numerical limits work correctly. You should not use 222this macro directly; instead, include the appropriate headers. 223 224 <br><dt><code>__SCHAR_MAX__</code><dt><code>__WCHAR_MAX__</code><dt><code>__SHRT_MAX__</code><dt><code>__INT_MAX__</code><dt><code>__LONG_MAX__</code><dt><code>__LONG_LONG_MAX__</code><dt><code>__WINT_MAX__</code><dt><code>__SIZE_MAX__</code><dt><code>__PTRDIFF_MAX__</code><dt><code>__INTMAX_MAX__</code><dt><code>__UINTMAX_MAX__</code><dt><code>__SIG_ATOMIC_MAX__</code><dt><code>__INT8_MAX__</code><dt><code>__INT16_MAX__</code><dt><code>__INT32_MAX__</code><dt><code>__INT64_MAX__</code><dt><code>__UINT8_MAX__</code><dt><code>__UINT16_MAX__</code><dt><code>__UINT32_MAX__</code><dt><code>__UINT64_MAX__</code><dt><code>__INT_LEAST8_MAX__</code><dt><code>__INT_LEAST16_MAX__</code><dt><code>__INT_LEAST32_MAX__</code><dt><code>__INT_LEAST64_MAX__</code><dt><code>__UINT_LEAST8_MAX__</code><dt><code>__UINT_LEAST16_MAX__</code><dt><code>__UINT_LEAST32_MAX__</code><dt><code>__UINT_LEAST64_MAX__</code><dt><code>__INT_FAST8_MAX__</code><dt><code>__INT_FAST16_MAX__</code><dt><code>__INT_FAST32_MAX__</code><dt><code>__INT_FAST64_MAX__</code><dt><code>__UINT_FAST8_MAX__</code><dt><code>__UINT_FAST16_MAX__</code><dt><code>__UINT_FAST32_MAX__</code><dt><code>__UINT_FAST64_MAX__</code><dt><code>__INTPTR_MAX__</code><dt><code>__UINTPTR_MAX__</code><dt><code>__WCHAR_MIN__</code><dt><code>__WINT_MIN__</code><dt><code>__SIG_ATOMIC_MIN__</code><dd>Defined to the maximum value of the <code>signed char</code>, <code>wchar_t</code>, 225<code>signed short</code>, 226<code>signed int</code>, <code>signed long</code>, <code>signed long long</code>, 227<code>wint_t</code>, <code>size_t</code>, <code>ptrdiff_t</code>, 228<code>intmax_t</code>, <code>uintmax_t</code>, <code>sig_atomic_t</code>, <code>int8_t</code>, 229<code>int16_t</code>, <code>int32_t</code>, <code>int64_t</code>, <code>uint8_t</code>, 230<code>uint16_t</code>, <code>uint32_t</code>, <code>uint64_t</code>, 231<code>int_least8_t</code>, <code>int_least16_t</code>, <code>int_least32_t</code>, 232<code>int_least64_t</code>, <code>uint_least8_t</code>, <code>uint_least16_t</code>, 233<code>uint_least32_t</code>, <code>uint_least64_t</code>, <code>int_fast8_t</code>, 234<code>int_fast16_t</code>, <code>int_fast32_t</code>, <code>int_fast64_t</code>, 235<code>uint_fast8_t</code>, <code>uint_fast16_t</code>, <code>uint_fast32_t</code>, 236<code>uint_fast64_t</code>, <code>intptr_t</code>, and <code>uintptr_t</code> types and 237to the minimum value of the <code>wchar_t</code>, <code>wint_t</code>, and 238<code>sig_atomic_t</code> types respectively. They exist to make the 239standard header given numerical limits work correctly. You should not 240use these macros directly; instead, include the appropriate headers. 241Some of these macros may not be defined on particular systems if GCC 242does not provide a <samp><span class="file">stdint.h</span></samp> header on those systems. 243 244 <br><dt><code>__INT8_C</code><dt><code>__INT16_C</code><dt><code>__INT32_C</code><dt><code>__INT64_C</code><dt><code>__UINT8_C</code><dt><code>__UINT16_C</code><dt><code>__UINT32_C</code><dt><code>__UINT64_C</code><dt><code>__INTMAX_C</code><dt><code>__UINTMAX_C</code><dd>Defined to implementations of the standard <samp><span class="file">stdint.h</span></samp> macros with 245the same names without the leading <code>__</code>. They exist the make the 246implementation of that header work correctly. You should not use 247these macros directly; instead, include the appropriate headers. Some 248of these macros may not be defined on particular systems if GCC does 249not provide a <samp><span class="file">stdint.h</span></samp> header on those systems. 250 251 <br><dt><code>__SIZEOF_INT__</code><dt><code>__SIZEOF_LONG__</code><dt><code>__SIZEOF_LONG_LONG__</code><dt><code>__SIZEOF_SHORT__</code><dt><code>__SIZEOF_POINTER__</code><dt><code>__SIZEOF_FLOAT__</code><dt><code>__SIZEOF_DOUBLE__</code><dt><code>__SIZEOF_LONG_DOUBLE__</code><dt><code>__SIZEOF_SIZE_T__</code><dt><code>__SIZEOF_WCHAR_T__</code><dt><code>__SIZEOF_WINT_T__</code><dt><code>__SIZEOF_PTRDIFF_T__</code><dd>Defined to the number of bytes of the C standard data types: <code>int</code>, 252<code>long</code>, <code>long long</code>, <code>short</code>, <code>void *</code>, <code>float</code>, 253<code>double</code>, <code>long double</code>, <code>size_t</code>, <code>wchar_t</code>, <code>wint_t</code> 254and <code>ptrdiff_t</code>. 255 256 <br><dt><code>__BYTE_ORDER__</code><dt><code>__ORDER_LITTLE_ENDIAN__</code><dt><code>__ORDER_BIG_ENDIAN__</code><dt><code>__ORDER_PDP_ENDIAN__</code><dd><code>__BYTE_ORDER__</code> is defined to one of the values 257<code>__ORDER_LITTLE_ENDIAN__</code>, <code>__ORDER_BIG_ENDIAN__</code>, or 258<code>__ORDER_PDP_ENDIAN__</code> to reflect the layout of multi-byte and 259multi-word quantities in memory. If <code>__BYTE_ORDER__</code> is equal to 260<code>__ORDER_LITTLE_ENDIAN__</code> or <code>__ORDER_BIG_ENDIAN__</code>, then 261multi-byte and multi-word quantities are laid out identically: the 262byte (word) at the lowest address is the least significant or most 263significant byte (word) of the quantity, respectively. If 264<code>__BYTE_ORDER__</code> is equal to <code>__ORDER_PDP_ENDIAN__</code>, then 265bytes in 16-bit words are laid out in a little-endian fashion, whereas 266the 16-bit subwords of a 32-bit quantity are laid out in big-endian 267fashion. 268 269 <p>You should use these macros for testing like this: 270 271 <pre class="smallexample"> /* <span class="roman">Test for a little-endian machine</span> */ 272 #if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__ 273</pre> 274 <br><dt><code>__FLOAT_WORD_ORDER__</code><dd><code>__FLOAT_WORD_ORDER__</code> is defined to one of the values 275<code>__ORDER_LITTLE_ENDIAN__</code> or <code>__ORDER_BIG_ENDIAN__</code> to reflect 276the layout of the words of multi-word floating-point quantities. 277 278 <br><dt><code>__DEPRECATED</code><dd>This macro is defined, with value 1, when compiling a C++ source file 279with warnings about deprecated constructs enabled. These warnings are 280enabled by default, but can be disabled with <samp><span class="option">-Wno-deprecated</span></samp>. 281 282 <br><dt><code>__EXCEPTIONS</code><dd>This macro is defined, with value 1, when compiling a C++ source file 283with exceptions enabled. If <samp><span class="option">-fno-exceptions</span></samp> is used when 284compiling the file, then this macro is not defined. 285 286 <br><dt><code>__GXX_RTTI</code><dd>This macro is defined, with value 1, when compiling a C++ source file 287with runtime type identification enabled. If <samp><span class="option">-fno-rtti</span></samp> is 288used when compiling the file, then this macro is not defined. 289 290 <br><dt><code>__USING_SJLJ_EXCEPTIONS__</code><dd>This macro is defined, with value 1, if the compiler uses the old 291mechanism based on <code>setjmp</code> and <code>longjmp</code> for exception 292handling. 293 294 <br><dt><code>__GXX_EXPERIMENTAL_CXX0X__</code><dd>This macro is defined when compiling a C++ source file with the option 295<samp><span class="option">-std=c++0x</span></samp> or <samp><span class="option">-std=gnu++0x</span></samp>. It indicates that some 296features likely to be included in C++0x are available. Note that these 297features are experimental, and may change or be removed in future 298versions of GCC. 299 300 <br><dt><code>__GXX_WEAK__</code><dd>This macro is defined when compiling a C++ source file. It has the 301value 1 if the compiler will use weak symbols, COMDAT sections, or 302other similar techniques to collapse symbols with “vague linkage” 303that are defined in multiple translation units. If the compiler will 304not collapse such symbols, this macro is defined with value 0. In 305general, user code should not need to make use of this macro; the 306purpose of this macro is to ease implementation of the C++ runtime 307library provided with G++. 308 309 <br><dt><code>__NEXT_RUNTIME__</code><dd>This macro is defined, with value 1, if (and only if) the NeXT runtime 310(as in <samp><span class="option">-fnext-runtime</span></samp>) is in use for Objective-C. If the GNU 311runtime is used, this macro is not defined, so that you can use this 312macro to determine which runtime (NeXT or GNU) is being used. 313 314 <br><dt><code>__LP64__</code><dt><code>_LP64</code><dd>These macros are defined, with value 1, if (and only if) the compilation 315is for a target where <code>long int</code> and pointer both use 64-bits and 316<code>int</code> uses 32-bit. 317 318 <br><dt><code>__SSP__</code><dd>This macro is defined, with value 1, when <samp><span class="option">-fstack-protector</span></samp> is in 319use. 320 321 <br><dt><code>__SSP_ALL__</code><dd>This macro is defined, with value 2, when <samp><span class="option">-fstack-protector-all</span></samp> is 322in use. 323 324 <br><dt><code>__TIMESTAMP__</code><dd>This macro expands to a string constant that describes the date and time 325of the last modification of the current source file. The string constant 326contains abbreviated day of the week, month, day of the month, time in 327hh:mm:ss form, year and looks like <code>"Sun Sep 16 01:03:52 1973"<!-- /@w --></code>. 328If the day of the month is less than 10, it is padded with a space on the left. 329 330 <p>If GCC cannot determine the current date, it will emit a warning message 331(once per compilation) and <code>__TIMESTAMP__</code> will expand to 332<code>"??? ??? ?? ??:??:?? ????"<!-- /@w --></code>. 333 334 <br><dt><code>__GCC_HAVE_SYNC_COMPARE_AND_SWAP_1</code><dt><code>__GCC_HAVE_SYNC_COMPARE_AND_SWAP_2</code><dt><code>__GCC_HAVE_SYNC_COMPARE_AND_SWAP_4</code><dt><code>__GCC_HAVE_SYNC_COMPARE_AND_SWAP_8</code><dt><code>__GCC_HAVE_SYNC_COMPARE_AND_SWAP_16</code><dd>These macros are defined when the target processor supports atomic compare 335and swap operations on operands 1, 2, 4, 8 or 16 bytes in length, respectively. 336 337 <br><dt><code>__GCC_HAVE_DWARF2_CFI_ASM</code><dd>This macro is defined when the compiler is emitting Dwarf2 CFI directives 338to the assembler. When this is defined, it is possible to emit those same 339directives in inline assembly. 340 341 <br><dt><code>__FP_FAST_FMA</code><dt><code>__FP_FAST_FMAF</code><dt><code>__FP_FAST_FMAL</code><dd>These macros are defined with value 1 if the backend supports the 342<code>fma</code>, <code>fmaf</code>, and <code>fmal</code> builtin functions, so that 343the include file <samp><span class="file">math.h</span></samp> can define the macros 344<code>FP_FAST_FMA</code>, <code>FP_FAST_FMAF</code>, and <code>FP_FAST_FMAL</code> 345for compatibility with the 1999 C standard. 346</dl> 347 348 </body></html> 349 350