📄 other-builtins.html
字号:
<html lang="en"><head><title>Using the GNU Compiler Collection (GCC)</title><meta http-equiv="Content-Type" content="text/html"><meta name="description" content="Using the GNU Compiler Collection (GCC)"><meta name="generator" content="makeinfo 4.6"><!--Copyright © 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998,1999, 2000, 2001, 2002, 2003, 2004 Free Software Foundation, Inc. <p>Permission is granted to copy, distribute and/or modify this documentunder the terms of the GNU Free Documentation License, Version 1.2 orany later version published by the Free Software Foundation; with theInvariant Sections being "GNU General Public License" and "FundingFree Software", the Front-Cover texts being (a) (see below), and withthe Back-Cover Texts being (b) (see below). A copy of the license isincluded in the section entitled "GNU Free Documentation License". <p>(a) The FSF's Front-Cover Text is: <p>A GNU Manual <p>(b) The FSF's Back-Cover Text is: <p>You have freedom to copy and modify this GNU Manual, like GNU software. Copies published by the Free Software Foundation raise funds for GNU development.--><meta http-equiv="Content-Style-Type" content="text/css"><style type="text/css"><!-- pre.display { font-family:inherit } pre.format { font-family:inherit } pre.smalldisplay { font-family:inherit; font-size:smaller } pre.smallformat { font-family:inherit; font-size:smaller } pre.smallexample { font-size:smaller } pre.smalllisp { font-size:smaller }--></style></head><body><div class="node"><p>Node: <a name="Other%20Builtins">Other Builtins</a>,Next: <a rel="next" accesskey="n" href="Target-Builtins.html#Target%20Builtins">Target Builtins</a>,Previous: <a rel="previous" accesskey="p" href="Vector-Extensions.html#Vector%20Extensions">Vector Extensions</a>,Up: <a rel="up" accesskey="u" href="C-Extensions.html#C%20Extensions">C Extensions</a><hr><br></div><h3 class="section">Other built-in functions provided by GCC</h3><p>GCC provides a large number of built-in functions other than the onesmentioned above. Some of these are for internal use in the processingof exceptions or variable-length argument lists and will not bedocumented here because they may change from time to time; we do notrecommend general use of these functions. <p>The remaining functions are provided for optimization purposes. <p>GCC includes built-in versions of many of the functions in the standardC library. The versions prefixed with <code>__builtin_</code> will always betreated as having the same meaning as the C library function even if youspecify the <code>-fno-builtin</code> option. (see <a href="C-Dialect-Options.html#C%20Dialect%20Options">C Dialect Options</a>)Many of these functions are only optimized in certain cases; if they arenot optimized in a particular case, a call to the library function willbe emitted. <p>Outside strict ISO C mode (<code>-ansi</code>, <code>-std=c89</code> or<code>-std=c99</code>), the functions<code>_exit</code>, <code>alloca</code>, <code>bcmp</code>, <code>bzero</code>,<code>dcgettext</code>, <code>dgettext</code>, <code>dremf</code>, <code>dreml</code>,<code>drem</code>, <code>exp10f</code>, <code>exp10l</code>, <code>exp10</code>, <code>ffsll</code>,<code>ffsl</code>, <code>ffs</code>, <code>fprintf_unlocked</code>, <code>fputs_unlocked</code>,<code>gammaf</code>, <code>gammal</code>, <code>gamma</code>, <code>gettext</code>,<code>index</code>, <code>isascii</code>, <code>j0f</code>, <code>j0l</code>, <code>j0</code>,<code>j1f</code>, <code>j1l</code>, <code>j1</code>, <code>jnf</code>, <code>jnl</code>, <code>jn</code>,<code>mempcpy</code>, <code>pow10f</code>, <code>pow10l</code>, <code>pow10</code>,<code>printf_unlocked</code>, <code>rindex</code>, <code>scalbf</code>, <code>scalbl</code>,<code>scalb</code>, <code>signbit</code>, <code>signbitf</code>, <code>signbitl</code>,<code>significandf</code>, <code>significandl</code>, <code>significand</code>,<code>sincosf</code>, <code>sincosl</code>, <code>sincos</code>, <code>stpcpy</code>,<code>strdup</code>, <code>strfmon</code>, <code>toascii</code>, <code>y0f</code>, <code>y0l</code>,<code>y0</code>, <code>y1f</code>, <code>y1l</code>, <code>y1</code>, <code>ynf</code>, <code>ynl</code> and<code>yn</code>may be handled as built-in functions. All these functions have corresponding versionsprefixed with <code>__builtin_</code>, which may be used even in strict C89mode. <p>The ISO C99 functions<code>_Exit</code>, <code>acoshf</code>, <code>acoshl</code>, <code>acosh</code>, <code>asinhf</code>,<code>asinhl</code>, <code>asinh</code>, <code>atanhf</code>, <code>atanhl</code>, <code>atanh</code>,<code>cabsf</code>, <code>cabsl</code>, <code>cabs</code>, <code>cacosf</code>, <code>cacoshf</code>,<code>cacoshl</code>, <code>cacosh</code>, <code>cacosl</code>, <code>cacos</code>,<code>cargf</code>, <code>cargl</code>, <code>carg</code>, <code>casinf</code>, <code>casinhf</code>,<code>casinhl</code>, <code>casinh</code>, <code>casinl</code>, <code>casin</code>,<code>catanf</code>, <code>catanhf</code>, <code>catanhl</code>, <code>catanh</code>,<code>catanl</code>, <code>catan</code>, <code>cbrtf</code>, <code>cbrtl</code>, <code>cbrt</code>,<code>ccosf</code>, <code>ccoshf</code>, <code>ccoshl</code>, <code>ccosh</code>, <code>ccosl</code>,<code>ccos</code>, <code>cexpf</code>, <code>cexpl</code>, <code>cexp</code>, <code>cimagf</code>,<code>cimagl</code>, <code>cimag</code>, <code>conjf</code>, <code>conjl</code>, <code>conj</code>,<code>copysignf</code>, <code>copysignl</code>, <code>copysign</code>, <code>cpowf</code>,<code>cpowl</code>, <code>cpow</code>, <code>cprojf</code>, <code>cprojl</code>, <code>cproj</code>,<code>crealf</code>, <code>creall</code>, <code>creal</code>, <code>csinf</code>, <code>csinhf</code>,<code>csinhl</code>, <code>csinh</code>, <code>csinl</code>, <code>csin</code>, <code>csqrtf</code>,<code>csqrtl</code>, <code>csqrt</code>, <code>ctanf</code>, <code>ctanhf</code>, <code>ctanhl</code>,<code>ctanh</code>, <code>ctanl</code>, <code>ctan</code>, <code>erfcf</code>, <code>erfcl</code>,<code>erfc</code>, <code>erff</code>, <code>erfl</code>, <code>erf</code>, <code>exp2f</code>,<code>exp2l</code>, <code>exp2</code>, <code>expm1f</code>, <code>expm1l</code>, <code>expm1</code>,<code>fdimf</code>, <code>fdiml</code>, <code>fdim</code>, <code>fmaf</code>, <code>fmal</code>,<code>fmaxf</code>, <code>fmaxl</code>, <code>fmax</code>, <code>fma</code>, <code>fminf</code>,<code>fminl</code>, <code>fmin</code>, <code>hypotf</code>, <code>hypotl</code>, <code>hypot</code>,<code>ilogbf</code>, <code>ilogbl</code>, <code>ilogb</code>, <code>imaxabs</code>,<code>isblank</code>, <code>iswblank</code>, <code>lgammaf</code>, <code>lgammal</code>,<code>lgamma</code>, <code>llabs</code>, <code>llrintf</code>, <code>llrintl</code>,<code>llrint</code>, <code>llroundf</code>, <code>llroundl</code>, <code>llround</code>,<code>log1pf</code>, <code>log1pl</code>, <code>log1p</code>, <code>log2f</code>, <code>log2l</code>,<code>log2</code>, <code>logbf</code>, <code>logbl</code>, <code>logb</code>, <code>lrintf</code>,<code>lrintl</code>, <code>lrint</code>, <code>lroundf</code>, <code>lroundl</code>,<code>lround</code>, <code>nearbyintf</code>, <code>nearbyintl</code>, <code>nearbyint</code>,<code>nextafterf</code>, <code>nextafterl</code>, <code>nextafter</code>,<code>nexttowardf</code>, <code>nexttowardl</code>, <code>nexttoward</code>,<code>remainderf</code>, <code>remainderl</code>, <code>remainder</code>, <code>remquof</code>,<code>remquol</code>, <code>remquo</code>, <code>rintf</code>, <code>rintl</code>, <code>rint</code>,<code>roundf</code>, <code>roundl</code>, <code>round</code>, <code>scalblnf</code>,<code>scalblnl</code>, <code>scalbln</code>, <code>scalbnf</code>, <code>scalbnl</code>,<code>scalbn</code>, <code>snprintf</code>, <code>tgammaf</code>, <code>tgammal</code>,<code>tgamma</code>, <code>truncf</code>, <code>truncl</code>, <code>trunc</code>,<code>vfscanf</code>, <code>vscanf</code>, <code>vsnprintf</code> and <code>vsscanf</code>are handled as built-in functionsexcept in strict ISO C90 mode (<code>-ansi</code> or <code>-std=c89</code>). <p>There are also built-in versions of the ISO C99 functions<code>acosf</code>, <code>acosl</code>, <code>asinf</code>, <code>asinl</code>, <code>atan2f</code>,<code>atan2l</code>, <code>atanf</code>, <code>atanl</code>, <code>ceilf</code>, <code>ceill</code>,<code>cosf</code>, <code>coshf</code>, <code>coshl</code>, <code>cosl</code>, <code>expf</code>,<code>expl</code>, <code>fabsf</code>, <code>fabsl</code>, <code>floorf</code>, <code>floorl</code>,<code>fmodf</code>, <code>fmodl</code>, <code>frexpf</code>, <code>frexpl</code>, <code>ldexpf</code>,<code>ldexpl</code>, <code>log10f</code>, <code>log10l</code>, <code>logf</code>, <code>logl</code>,<code>modfl</code>, <code>modf</code>, <code>powf</code>, <code>powl</code>, <code>sinf</code>,<code>sinhf</code>, <code>sinhl</code>, <code>sinl</code>, <code>sqrtf</code>, <code>sqrtl</code>,<code>tanf</code>, <code>tanhf</code>, <code>tanhl</code> and <code>tanl</code>that are recognized in any mode since ISO C90 reserves these names forthe purpose to which ISO C99 puts them. All these functions havecorresponding versions prefixed with <code>__builtin_</code>. <p>The ISO C94 functions<code>iswalnum</code>, <code>iswalpha</code>, <code>iswcntrl</code>, <code>iswdigit</code>,<code>iswgraph</code>, <code>iswlower</code>, <code>iswprint</code>, <code>iswpunct</code>,<code>iswspace</code>, <code>iswupper</code>, <code>iswxdigit</code>, <code>towlower</code> and<code>towupper</code>are handled as built-in functionsexcept in strict ISO C90 mode (<code>-ansi</code> or <code>-std=c89</code>). <p>The ISO C90 functions<code>abort</code>, <code>abs</code>, <code>acos</code>, <code>asin</code>, <code>atan2</code>,<code>atan</code>, <code>calloc</code>, <code>ceil</code>, <code>cosh</code>, <code>cos</code>,<code>exit</code>, <code>exp</code>, <code>fabs</code>, <code>floor</code>, <code>fmod</code>,<code>fprintf</code>, <code>fputs</code>, <code>frexp</code>, <code>fscanf</code>,<code>isalnum</code>, <code>isalpha</code>, <code>iscntrl</code>, <code>isdigit</code>,<code>isgraph</code>, <code>islower</code>, <code>isprint</code>, <code>ispunct</code>,<code>isspace</code>, <code>isupper</code>, <code>isxdigit</code>, <code>tolower</code>,<code>toupper</code>, <code>labs</code>, <code>ldexp</code>, <code>log10</code>, <code>log</code>,<code>malloc</code>, <code>memcmp</code>, <code>memcpy</code>, <code>memset</code>, <code>modf</code>,<code>pow</code>, <code>printf</code>, <code>putchar</code>, <code>puts</code>, <code>scanf</code>,<code>sinh</code>, <code>sin</code>, <code>snprintf</code>, <code>sprintf</code>, <code>sqrt</code>,<code>sscanf</code>, <code>strcat</code>, <code>strchr</code>, <code>strcmp</code>,<code>strcpy</code>, <code>strcspn</code>, <code>strlen</code>, <code>strncat</code>,<code>strncmp</code>, <code>strncpy</code>, <code>strpbrk</code>, <code>strrchr</code>,<code>strspn</code>, <code>strstr</code>, <code>tanh</code>, <code>tan</code>, <code>vfprintf</code>,<code>vprintf</code> and <code>vsprintf</code>are all recognized as built-in functions unless<code>-fno-builtin</code> is specified (or <code>-fno-builtin-</code><var>function</var><code></code>is specified for an individual function). All of these functions havecorresponding versions prefixed with <code>__builtin_</code>. <p>GCC provides built-in versions of the ISO C99 floating point comparisonmacros that avoid raising exceptions for unordered operands. They havethe same names as the standard macros ( <code>isgreater</code>,<code>isgreaterequal</code>, <code>isless</code>, <code>islessequal</code>,<code>islessgreater</code>, and <code>isunordered</code>) , with <code>__builtin_</code>prefixed. We intend for a library implementor to be able to simply<code>#define</code> each standard macro to its built-in equivalent.<p><table width="100%"><tr><td align="left">int <b>__builtin_types_compatible_p</b><i> </i>(<i></i><var>type1</var><i>, </i><var>type2</var><i></i>)<i> </i></td><td align="right">Built-in Function</td></tr></table><table width="95%" align="center"><tr><td> <p>You can use the built-in function <code>__builtin_types_compatible_p</code> todetermine whether two types are the same. <p>This built-in function returns 1 if the unqualified versions of thetypes <var>type1</var> and <var>type2</var> (which are types, not expressions) arecompatible, 0 otherwise. The result of this built-in function can beused in integer constant expressions. <p>This built-in function ignores top level qualifiers (e.g., <code>const</code>,<code>volatile</code>). For example, <code>int</code> is equivalent to <code>constint</code>. <p>The type <code>int[]</code> and <code>int[5]</code> are compatible. On the otherhand, <code>int</code> and <code>char *</code> are not compatible, even if the sizeof their types, on the particular architecture are the same. Also, theamount of pointer indirection is taken into account when determiningsimilarity. Consequently, <code>short *</code> is not similar to<code>short **</code>. Furthermore, two types that are typedefed areconsidered compatible if their underlying types are compatible. <p>An <code>enum</code> type is not considered to be compatible with another<code>enum</code> type even if both are compatible with the same integertype; this is what the C standard specifies. For example, <code>enum {foo, bar}</code> is not similar to<code>enum {hot, dog}</code>. <p>You would typically use this function in code whose execution variesdepending on the arguments' types. For example: <pre class="smallexample"> #define foo(x) \ ({ \ typeof (x) tmp; \ if (__builtin_types_compatible_p (typeof (x), long double)) \ tmp = foo_long_double (tmp); \ else if (__builtin_types_compatible_p (typeof (x), double)) \ tmp = foo_double (tmp); \ else if (__builtin_types_compatible_p (typeof (x), float)) \ tmp = foo_float (tmp); \ else \ abort (); \ tmp; \ }) </pre> <p><em>Note:</em> This construct is only available for C. </td></tr></table><p><table width="100%"><tr><td align="left"><var>type</var> <b>__builtin_choose_expr</b><i> </i>(<i></i><var>const_exp</var><i>, </i><var>exp1</var><i>, </i><var>exp2</var><i></i>)<i> </i></td><td align="right">Built-in Function</td></tr></table><table width="95%" align="center"><tr><td> <p>You can use the built-in function <code>__builtin_choose_expr</code> toevaluate code depending on the value of a constant expression. Thisbuilt-in function returns <var>exp1</var> if <var>const_exp</var>, which is aconstant expression that must be able to be determined at compile time,is nonzero. Otherwise it returns 0. <p>This built-in function is analogous to the <code>? :</code> operator in C,except that the expression returned has its type unaltered by promotionrules. Also, the built-in function does not evaluate the expressionthat was not chosen. For example, if <var>const_exp</var> evaluates to true,<var>exp2</var> is not evaluated even if it has side-effects. <p>This built-in function can return an lvalue if the chosen argument is anlvalue. <p>If <var>exp1</var> is returned, the return type is the same as <var>exp1</var>'stype. Similarly, if <var>exp2</var> is returned, its return type is the sameas <var>exp2</var>. <p>Example: <pre class="smallexample"> #define foo(x) \ __builtin_choose_expr ( \ __builtin_types_compatible_p (typeof (x), double), \ foo_double (x), \ __builtin_choose_expr ( \ __builtin_types_compatible_p (typeof (x), float), \ foo_float (x), \ /* The void expression results in a compile-time error \ when assigning the result to something. */ \ (void)0)) </pre>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -