Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!think!ima!haddock!karl
From: karl@haddock
Newsgroups: net.lang.c
Subject: Re: structure alignment question
Message-ID: <86900062@haddock>
Date: Mon, 22-Sep-86 09:15:00 EDT
Article-I.D.: haddock.86900062
Posted: Mon Sep 22 09:15:00 1986
Date-Received: Mon, 22-Sep-86 21:37:32 EDT
References: <5500003@hcx1>
Lines: 18
Nf-ID: #R:hcx1:5500003:haddock:86900062:000:881
Nf-From: haddock!karl    Sep 22 09:15:00 1986


>In article <1705@mcc-pp.UUCP> tiemann@mcc-pp.UUCP (Michael Tiemann) writes:
>>The last 68000 compiler I used aligned strings on WORD boundaries. ...
>>There was a big speed payoff: I could do word operations in [strncpy, etc.]

Wrong!  Even if strings are word-aligned, substrings (e.g. "p+1") will not
be.  (Unless the chars themselves are word-aligned, in which case your space
penalty is a factor of two instead of a half byte.)

umcp-cs!chris (Chris Torek) writes:
>It is not difficult to do copies in word mode iff the strings are aligned:
>[example of a strcpy() that considers four alignment cases]

Even if the strings are word-aligned, you have to test for a null *byte*
(not a zero word); my guess is that the extra manipulation inside the loop
will make it slower than a byte-by-byte copy.

Karl W. Z. Heuer (ima!haddock!karl or karl@haddock.isc.com), The Walking Lint