| William M. Klein 2004-07-24, 3:55 am |
| "Robert Wagner" <robert.deletethis@wagner.net> wrote in message
news:41016572.9332848@news.optonline.net...
> riplin@Azonic.co.nz (Richard) wrote:
<snip>
>
> Which compiler treats COMP-4 as different from COMP-5? Excepting Unisys, the
> only one I know of was Realia, where COMP-4 guaranteed small-endian. I believe
> all compilers treat BINARY the same as COMP-5 i.e. the machine's native
integer
>
IBM mainframe:
COMP-4 = Binary (sensitive to TRUNC compiler option)
COMP-5 = Binary with TRUNC(OPT)
IBM PC
COMP-4 = Binary = big-endian
COMP-5 = small-endian
IBM OS/400
no COMP-5 support (last time I looked, may have changed recently)
Micro Focus
(default on PC)
COMP-4 = Binary = big-endian (with default of IBM-COMP, if I recall
correctly)
COMP-5 = small-endian (I can't remember but it may also be NOIBM-COMP)
Fujitsu,
Current LRM does not show ANY support for COMP-4
--
Bill Klein
wmklein <at> ix.netcom.com
|