NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c-part1/6959

Internet Message Format | 1996-08-05 | 1.9 KB

Path: news.ov.com!news
From: [e-mail] ([e-mail])
Newsgroups: comp.lang.c
Subject: Re: Big and little endians
Date: 16 Feb 1996 18:44:00 GMT
Organization: OpenVision
Message-ID: <4g2j9g$[e-mail]>
References: <4fuuqq$[e-mail]>
Reply-To: [e-mail]
NNTP-Posting-Host: foghorn.pls.ov.com

In article [e-mail], Vidar Moe <[e-mail]> writes:
>What exactly is meant by the terms little-endians and
>big-endians?
>
>Vidar Moe.
>
>

First of all, the terms are derived from Johnathon Swift's "Gulliver's
Travels" referring to the reason behind the war between Lilliput and
Blefesceu.  These two cities were warring over which end of a soft
boiled egg should be opened to eat it.  One city said the "big end" and
the other said the "little end".  The terms now are used to refer to
those who defend a dogma without good reason.

As it applies to computers, the terms refer to the ordering of bytes
in a computer word.  Dec and Intel use the "Big-Endian" order, and
Motorola and SPARC use the "Little-Endian" order.

The "Little-Endian" order places the lowest addressed byte in the high
end of the word, and then next higher addressed byte to the next lower
position.  Thus if you examine a string in memory, you will see the
bytes in the same order as they would appear in a word.

In ancient times, word size and byte ordering would be detected by sending
a 4 character word.  The UNIX world used the word UNIX.  The receiving
end would examine the word, and if it got UNIX, then the machines had
the same byte ordering.  If the machine got XINU, then
the word size was 4 bytes, but the byte order was opposite (the other
endian).  Similarly, if the word was NUXI then the word size was two
bytes and opposite byte ordering.  Clearly, you could not detect the
difference between 2 byte and 4 byte words if the machines had the
same byte ordering.

            [e-mail]