Document modff() and modfl(). Technically, modff() and modfl()

live in libm, while modf() lives in libc due to historical
mistakes. I'm claiming in the manpage that they all live in libm,
since programmers should not rely on the mistake.

lib/libc/gen/Makefile.inc

@@ -120,6 +120,7 @@ MLINKS+=isgreater.3 isgreaterequal.3 isgreater.3 isless.3 \
 	isgreater.3 isunordered.3
 MLINKS+=ldexp.3 ldexpf.3 ldexp.3 ldexpl.3
 MLINKS+=makecontext.3 swapcontext.3
+MLINKS+=modf.3 modff.3 modf.3 modfl.3
 MLINKS+=popen.3 pclose.3
 MLINKS+=psignal.3 strsignal.3 psignal.3 sys_siglist.3 psignal.3 sys_signame.3
 MLINKS+=pwcache.3 group_from_gid.3 pwcache.3 user_from_uid.3

lib/libc/gen/modf.3

@@ -32,33 +32,40 @@
 .\"     @(#)modf.3	8.1 (Berkeley) 6/4/93
 .\" $FreeBSD$
 .\"
-.Dd June 4, 1993
+.Dd March 29, 2008
 .Dt MODF 3
 .Os
 .Sh NAME
-.Nm modf
+.Nm modf ,
+.Nm modff ,
+.Nm modfl
 .Nd extract signed integral and fractional values from floating-point number
 .Sh LIBRARY
-.Lb libc
+.Lb libm
 .Sh SYNOPSIS
 .In math.h
 .Ft double
 .Fn modf "double value" "double *iptr"
+.Ft float
+.Fn modff "float value" "float *iptr"
+.Ft long double
+.Fn modfl "long double value" "long double *iptr"
 .Sh DESCRIPTION
 The
-.Fn modf
-function breaks the argument
+.Fn modf ,
+.Fn modff ,
+and
+.Fn modfl
+functions break the argument
 .Fa value
 into integral and fractional parts, each of which has the
 same sign as the argument.
 It stores the integral part as a
-.Vt double
+floating point number
 in the object pointed to by
 .Fa iptr .
 .Sh RETURN VALUES
-The
-.Fn modf
-function returns the signed fractional part of
+These functions return the signed fractional part of
 .Fa value .
 .Sh SEE ALSO
 .Xr frexp 3 ,
@@ -66,6 +73,9 @@ function returns the signed fractional part of
 .Xr math 3
 .Sh STANDARDS
 The
-.Fn modf
-function conforms to
-.St -isoC .
+.Fn modf ,
+.Fn modff ,
+and
+.Fn modfl
+functions conform to
+.St -isoC-99 .