#define INT_ALIGN_MASK (sizeof(int) - 1)
#include <stdlib.h>

/*
 * MemSet
 *	Exactly the same as standard library function memset(), but considerably
 *	faster for zeroing small word-aligned structures (such as parsetree nodes).
 *	This has to be a macro because the main point is to avoid function-call
 *	overhead.   However, we have also found that the loop is faster than
 *	native libc memset() on some platforms, even those with assembler
 *	memset() functions.  More research needs to be done, perhaps with
 *	platform-specific MEMSET_LOOP_LIMIT values or tests in configure.
 *
 *	bjm 2002-10-08
 */
#define MemSet(start, val, len) \
	do \
	{ \
		int * _start = (int *) (start); \
		int		_val = (val); \
		size_t	_len = (len); \
\
		if ((((long) _start) & INT_ALIGN_MASK) == 0 && \
			(_len & INT_ALIGN_MASK) == 0 && \
			_val == 0 && \
			_len <= MEMSET_LOOP_LIMIT) \
		{ \
			int * _stop = (int *) ((char *) _start + _len); \
			while (_start < _stop) \
				*_start++ = 0; \
		} \
		else \
			memset((char *) _start, _val, _len); \
	} while (0)

#define MEMSET_LOOP_LIMIT  1024

int
main(int argc, char **argv)
{
	int i,j;
	char *buf;

	buf = malloc(256);

	j = 32;
	for (i = 0; i < 1000000; i++)
		MemSet(buf, 0, j /* choose 256 or j here */);
	return 0;
}