__builtin_types_compatible_p(3) [netbsd man page]

__BUILTIN_TYPES_COMPATIBLE_P(3) 			   BSD Library Functions Manual 			   __BUILTIN_TYPES_COMPATIBLE_P(3)

NAME

     __builtin_types_compatible_p -- GNU extension to check equivalent types

SYNOPSIS

     int
     __builtin_types_compatible_p(type_a, type_b);

DESCRIPTION

     The __builtin_types_compatible_p() is a GNU extension for determining whether two types are equivalent.  If type_a is equivalent to type_b, a
     value 1 is returned.  Otherwise __builtin_types_compatible_p() returns 0.

     The following remarks should be taken into account.

	   1.	The architecture-specific size of the two types does not have an impact on the result.	For example, sizeof(char *) and
		sizeof(int) result the same value on i386, but the types naturally are not equivalent.

	   2.	Type qualifiers are ignored.  The function returns the same value for long and const long.

	   3.	The amount of pointer indirection affects the result.  For example, double * is not equivalent to double **.

	   4.	Two types defined with typedef are equivalent if and only if their underlying types are equivalent.

	   5.	The enum type is a special case in that two enum types are not considered equivalent.

EXAMPLES

     The following example combines __builtin_types_compatible_p() and the typeof(3) construct:

	   #define __COMPARE_TYPES(v, t)      
		    __builtin_types_compatible_p(__typeof__(v), t)

	   ...

	   if (__COMPARE_TYPES(p, double) != 0)
		   err(EX_DATAERR, "invalid type");

SEE ALSO

     gcc(1), __builtin_constant_p(3), typeof(3)

CAVEATS

     This is a non-standard, compiler-specific extension.

BSD
								 December 21, 2010							       BSD

Check Out this Related Man Page

LRINT(3)						   BSD Library Functions Manual 						  LRINT(3)

NAME

     llrint, llrintf, llrintl, lrint, lrintf, lrintl -- convert to integer

LIBRARY

     Math Library (libm, -lm)

SYNOPSIS

     #include <math.h>

     long long
     llrint(double x);

     long long
     llrintf(float x);

     long long
     llrintl(long double x);

     long
     lrint(double x);

     long
     lrintf(float x);

     long
     lrintl(long double x);

DESCRIPTION

     The lrint() function returns the integer nearest to its argument x according to the current rounding mode.  If the rounded result is too
     large to be represented as a long value, an invalid exception is raised and the return value is undefined.  Otherwise, if x is not an inte-
     ger, lrint() raises an inexact exception.	When the rounded result is representable as a long, the expression lrint(x) is equivalent to
     (long)rint(x) (although the former may be more efficient).

     The llrint(), llrintf(), llrintl(), lrintf(), and lrintl() functions differ from lrint() only in their input and output types.

SEE ALSO

     lround(3), math(3), rint(3), round(3)

STANDARDS

     These functions conform to ISO/IEC 9899:1999 (``ISO C99'').

HISTORY

     The llrint(), llrintf(), lrint(), and lrintf() routines first appeared in FreeBSD 5.4.  The long double variants were introduced in
     FreeBSD 8.0.

BSD
								 January 13, 2008							       BSD

Linux and UNIX Man Pages

__builtin_types_compatible_p(3) [netbsd man page]

Check Out this Related Man Page