Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/17/84; site twitch.UUCP
Path: utzoo!watmath!clyde!cbosgd!ihnp4!twitch!grt
From: grt@twitch.UUCP ( G.R.Tomasevich)
Newsgroups: net.lang.c
Subject: PDP-11 byte order in longs
Message-ID: <356@twitch.UUCP>
Date: Tue, 8-Apr-86 13:16:42 EST
Article-I.D.: twitch.356
Posted: Tue Apr  8 13:16:42 1986
Date-Received: Fri, 11-Apr-86 01:20:15 EST
References: <> <7046@cca.UUCP> <1223@ulysses.UUCP>
Organization: AT&T Bell Labs, Holmdel
Lines: 21

> > 	Well, no, little-endian came about because the engineers at DEC
> > who designed the PDP-11 made an arbitrary decision that was not well
> > thought out.  I will not essay to defend the sanity of DEC engineers,
> > 		Richard Harter, SMDS Inc.
> 
> As an old PDP-11 hacker, I can't agree with the condemnation of the
> DEC engineering decision.
> Griff Smith	AT&T (Bell Laboratories), Murray Hill

I agree with Harter's complaint but for a different reason.  The machine
instruction ASHC, arithmetic shift combined, when given an even-numbered
register, does a 32-bit shift, in which the lower-numbered register has the
high-order bits.  Who ordered that???  This seems to have led the C compiler
writer to store longs with the shorts in big-endian order, even though the
bytes of a short are stored in little-endian order.  I see no reason for that
convention being more convenient than strictly little-endian order, since
there are no long move instructions.  In other words, the byte order is
2,3,0,1 if you number from LSB.
-- 
	George Tomasevich, ihnp4!twitch!grt
	AT&T Bell Laboratories, Holmdel, NJ