[linux-2.6-block.git] / include / linux / reciprocal_div.h

#ifndef _LINUX_RECIPROCAL_DIV_H
#define _LINUX_RECIPROCAL_DIV_H

#include <linux/types.h>

/*
 * This algorithm is based on the paper "Division by Invariant
 * Integers Using Multiplication" by Torbjörn Granlund and Peter
 * L. Montgomery.
 *
 * The assembler implementation from Agner Fog, which this code is
 * based on, can be found here:
 * http://www.agner.org/optimize/asmlib.zip
 *
 * This optimization for A/B is helpful if the divisor B is mostly
 * runtime invariant. The reciprocal of B is calculated in the
 * slow-path with reciprocal_value(). The fast-path can then just use
 * a much faster multiplication operation with a variable dividend A
 * to calculate the division A/B.
 */

struct reciprocal_value {
	u32 m;
	u8 sh1, sh2;
};

struct reciprocal_value reciprocal_value(u32 d);

static inline u32 reciprocal_divide(u32 a, struct reciprocal_value R)
{
	u32 t = (u32)(((u64)a * R.m) >> 32);
	return (t + ((a - t) >> R.sh1)) >> R.sh2;
}

#endif /* _LINUX_RECIPROCAL_DIV_H */
Commit	Line	Data
6a2d7a95 ED	1	#ifndef _LINUX_RECIPROCAL_DIV_H
	2	#define _LINUX_RECIPROCAL_DIV_H
	3
	4	#include <linux/types.h>
	5
	6	/*
809fa972 HFS	7	* This algorithm is based on the paper "Division by Invariant
	8	* Integers Using Multiplication" by Torbjörn Granlund and Peter
	9	* L. Montgomery.
6a2d7a95	10	*
809fa972 HFS	11	* The assembler implementation from Agner Fog, which this code is
	12	* based on, can be found here:
	13	* http://www.agner.org/optimize/asmlib.zip
6a2d7a95	14	*
809fa972 HFS	15	* This optimization for A/B is helpful if the divisor B is mostly
	16	* runtime invariant. The reciprocal of B is calculated in the
	17	* slow-path with reciprocal_value(). The fast-path can then just use
	18	* a much faster multiplication operation with a variable dividend A
	19	* to calculate the division A/B.
6a2d7a95 ED	20	*/
6a2d7a95 ED	21
809fa972 HFS	22	struct reciprocal_value {
	23	u32 m;
	24	u8 sh1, sh2;
	25	};
6a2d7a95	26
809fa972	27	struct reciprocal_value reciprocal_value(u32 d);
6a2d7a95	28
809fa972	29	static inline u32 reciprocal_divide(u32 a, struct reciprocal_value R)
6a2d7a95	30	{
809fa972 HFS	31	u32 t = (u32)(((u64)a * R.m) >> 32);
809fa972 HFS	32	return (t + ((a - t) >> R.sh1)) >> R.sh2;
6a2d7a95	33	}
809fa972 HFS	34
809fa972 HFS	35	#endif /* _LINUX_RECIPROCAL_DIV_H */