NAME

libmp —

traditional BSD multiple precision integer arithmetic library

SYNOPSIS

#include <mp.h>

Function prototypes are given in the main body of the text.

Applications using this interface must be linked with -lmp (this library) and -lcrypto (crypto(3)).

DESCRIPTION

This interface is obsolete in favor of the crypto(3) BIGNUM library.

libmp is the traditional BSD multiple precision integer arithmetic library. It has a number of problems, and is unsuitable for use in any programs where reliability is a concern. It is provided here for compatibility only.

These routines perform arithmetic on integers of arbitrary precision stored using the defined type MINT. Pointers to MINT are initialized using mp_itom() or mp_xtom(), and must be recycled with mp_mfree() when they are no longer needed. Routines which store a result in one of their arguments expect that the latter has also been initialized prior to being passed to it. The following routines are defined and implemented:

MINT *mp_itom(short n);

MINT *mp_xtom(const char *s);

char *mp_mtox(const MINT *mp);

void mp_mfree(MINT *mp);

mp_itom() returns an MINT with the value of n. mp_xtom() returns an MINT with the value of s, which is treated to be in hexadecimal. The return values from mp_itom() and mp_xtom() must be released with mp_mfree() when they are no longer needed. mp_mtox() returns a null-terminated hexadecimal string having the value of mp; its return value must be released with free() (free(3)) when it is no longer needed.

void mp_madd(const MINT *mp1, const MINT *mp2, MINT *rmp);

void mp_msub(const MINT *mp1, const MINT *mp2, MINT *rmp);

void mp_mult(const MINT *mp1, const MINT *mp2, MINT *rmp);

mp_madd(), mp_msub(), and mp_mult() store the sum, difference, or product, respectively, of mp1 and mp2 in rmp.

void mp_mdiv(const MINT *nmp, const MINT *dmp, MINT *qmp, MINT *rmp);

void mp_sdiv(const MINT *nmp, short d, MINT *qmp, short *ro);

mp_mdiv() computes the quotient and remainder of nmp and dmp and stores the result in qmp and rmp, respectively. mp_sdiv() is similar to mp_mdiv() except the divisor (dmp or d) and remainder (rmp or ro) are ordinary integers.

void mp_pow(const MINT *bmp, const MINT *emp, const MINT *mmp, MINT *rmp);

void mp_rpow(const MINT *bmp, short e, MINT *rmp);

mp_rpow() computes the result of bmp raised to the empth power and reduced modulo mmp; the result is stored in rmp. mp_pow() computes the result of bmp raised to the eth power and stores the result in rmp.

void mp_min(MINT *mp);

void mp_mout(const MINT *mp);

mp_min() reads a line from standard input, tries to interpret it as a decimal number, and if successful, stores the result in mp. mp_mout() prints the value, in decimal, of mp to standard output (without a trailing newline).

void mp_gcd(const MINT *mp1, const MINT *mp2, MINT *rmp);

mp_gcd() computes the greatest common divisor of mp1 and mp2 and stores the result in rmp.

int mp_mcmp(const MINT *mp1, const MINT *mp2);

mcmp compares the values of mp1 and mp2 and returns 0 if the two values are equal, a value greater than 0 if mp1 is greater than mp2, and a value less than 0 if mp2 is greater than mp1.

void mp_move(const MINT *smp, MINT *tmp);

mp_move() copies the value of smp to tmp (both values must be initialized).

void mp_msqrt(const MINT *nmp, MINT *xmp, MINT *rmp);

mp_msqrt() computes the square root and remainder of nmp and stores them in xmp and rmp, respectively.

IMPLEMENTATION NOTES

This version of libmp is implemented in terms of the crypto(3) BIGNUM library.

DIAGNOSTICS

Running out of memory or illegal operations result in error messages on standard error and a call to abort(3).

HISTORY

A libmp library appeared in 4.3BSD. FreeBSD 2.2 shipped with a libmp implemented in terms of libgmp. This implementation appeared in FreeBSD 5.0.

BUGS

Errors are reported via output to standard error and abnormal program termination instead of via return values. The application cannot control this behavior.

It is not clear whether the string returned by mp_mtox() may be written to by the caller. This implementation allows it, but others may not. Ideally, mp_mtox() would take a pointer to a buffer to fill in.

It is not clear whether using the same variable as both source and destination in a single invocation is permitted. Some of the calls in this implementation allow this, while others do not.