.TH MP 3X
.UC
.SH NAME
itom, mcmp, m_in, m_out, madd, msub, mult, mdiv, min, mout,
pow, gcd, rpow \- multiple precision integer arithmetic
.SH SYNOPSIS
.nf
.B #include <mp.h>
.sp
.B "typedef struct mint { int len; short *val; } MINT;"
.PP
.PP
.B madd(a, b, c)
.B msub(a, b, c)
.B mult(a, b, c)
.B mdiv(a, b, q, r)
.B min(a)
.B mout(a)
.B pow(a, b, m, c)
.B gcd(a, b, c)
.B rpow(a, n, c)
.B msqrt(a, b, r)
.B mcmp(a,b)
.B MINT *a, *b, *c, *m, "*q, *r;"
.B int n;
.PP
.B
.B sdiv(a, n, q, r)
.B MINT *a, *q;
.B short *r;
.PP
.B MINT *itom(n)
.PP
.B cc ... \-lmp
.SH DESCRIPTION
These routines perform arithmetic on integers of
arbitrary length.
The integers are stored using the defined type
.IR MINT .
Pointers to
a
.I MINT
can be initialized using the function
.I itom
which sets the initial value to
.IR n .
After that space is managed automatically by the routines.
.PP
.IR madd , " msub" , " mult" ,
assign to their third arguments the sum, difference, and
product, respectively, of their first two arguments.
.I mdiv
assigns the quotient and remainder, respectively,
to its third and fourth arguments.
.I sdiv
is like
.I mdiv
except that the divisor is an ordinary integer.
.I msqrt
produces the square root and remainder of its first argument.
.I rpow
calculates
.I a
raised to the (``regular'' integral) power
.IR n ,
while
.I pow
calculates this (for
.IR *n " a " MINT )
reduced modulo
.IR m .
.IR min " and " mout
do decimal input and output.
More generally,
.IR fmin " and " fmout
do decimal input and output from file
.IR f ,
and
.IR m_in " and " m_out
do I/O base 
.IR n .
.PP
The functions are obtained with the
loader option
.IR \-lmp .
.SH FILES
.ta 2i
/usr/include/mp.h	include file
.br
/usr/lib/libmp.a	object code library
.SH SEE ALSO
ape(3X)
.SH DIAGNOSTICS
Illegal operations and running out of memory
produce messages and core images.
.SH BUGS
Bases for input and output should be <= 10.
.PP
The
.I mp
package has some bugs--users are advised to use
.I ape
instead.