>| > I don't see why we need to have more than two union types. One for double
>| > and one for float.
>|
>| Yes, true.  So "__double_u" and "__single_u" sound reasonable?  What
>| about "long double", or that format not well defined?
>
>long double can be either 80 or 128 bits depending on the machine.

or 64, as on netbsd/sparc.

    % uname -srm
    NetBSD 1.5Y sparc
    % cat ./longdouble.c
    main() { printf("%d\n", sizeof(long double)); }
    % make longdouble
    cc -O2   -o longdouble longdouble.c 
    % ./longdouble 
    8
    % gcc -v
    Using builtin specs.
    gcc version egcs-2.91.66 19990314 (egcs-1.1.2 release)
    % 

which i found confusing, since solaris/sparc uses a 16 byte variety.

-- 
|-----< "CODE WARRIOR" >-----|
codewarrior@daemon.org             * "ah!  i see you have the internet
twofsonet@graffiti.com (Andrew Brown)                that goes *ping*!"
andrew@crossbar.com       * "information is power -- share the wealth."