SUMMARY: sun byte order

From: Darryl Pace (
Date: Mon Apr 20 1998 - 09:45:48 CDT

Sun Managers,
   Apparently, I've committed a faux pas of the highest order, i.e., I
sent spurious information to the list in my summary. After receiving
excoriations via email, I felt it necessary to re-send the summary, this
with the correct information. Thanks go to the following folks for
Arthur Darren Dunham,
Chris Pillips,
Dave Mitchell,
Ian TCollins, and
Jochen Bern

Some of the the replies follow.

Sorry, but this is complete nonsense. Endian is a processor attribute,
not an operating system one. For example,

Solaris running on Sparc is big-endian,
Solaris running on Intel is little-endian.

Just my $0.02

Untrue! Solaris on intel is little-endian, byte order is platform
not OS specific.

Jochen Bern sent the following interesting reply, which I've not yet

I suppose the utmost Detail would be to point out how to determine
it yourself on the Fly ...

penthesilea:/home/TI/bern% uname -s -r ; ./TMP
SunOS 4.1.3_U1
Byte Order: 1 2 3 4
(a.k.a. bigendian)
penthesilea:/home/TI/bern% cat TMP.c
#include <stdio.h>


{ char c[4];
   int *i;

   if (sizeof(int)!=4)
   { printf("Please rewrite me for %d Byte Integers

    i = (int *) c ;
   *i = (( 1 *256+ 2 )*256+ 3 )*256+ 4 ;
   printf("Byte Order: %d %d %d %d\n",c[0],c[1],c[2],c[3]);
   if (c[0]==1 && c[1]==2 && c[2]==3 && c[3]==4)
      printf("(a.k.a. bigendian)\n");
   if (c[0]==4 && c[1]==3 && c[2]==2 && c[3]==1)
      printf("(a.k.a. little-endian)\n");


This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:12:38 CDT