.TH double "" "" "C Keyword"
.PC "Data type"
.PP
A
.B double
is the data type that encodes a double-precision floating-point
number.
On most machines,
.B "sizeof(double)"
is defined as four machine words, or eight \fBchar\fRs.
If you wish your code to be portable, do
\fInot\fR use routines that depend on a
.B double
being 64 bits long.
The ranges of values that can be held by a \*(CO
.BR double
are set in header file
.BR float.h .
.PP
Different formats are used to encode
.BR double s
on various machines.
These formats include
IEEE, DECVAX, and BCD (binary coded decimal), as
described in the entry for
.BR float .
\*(PN 286 uses DECVAX format;
\*(PN 386 uses IEEE format.
.SH "See Also"
.Xr "C keywords," c_keyword
.Xr "data formats," data_form
.Xr "float," float
.Xr "float.h," float.h
.Xr "portability" portabili
.br
\*(AS, \(sc6.1.2.5
