Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!rutgers!ll-xn!mit-eddie!genrad!decvax!mcnc!rti-sel!dg_rtp!throopw
From: throopw@dg_rtp.UUCP (Wayne Throop)
Newsgroups: net.lang.c
Subject: Re: Structures as arguments, lint checking
Message-ID: <660@dg_rtp.UUCP>
Date: Thu, 23-Oct-86 14:07:27 EDT
Article-I.D.: dg_rtp.660
Posted: Thu Oct 23 14:07:27 1986
Date-Received: Sat, 25-Oct-86 01:22:31 EDT
References: <4738@brl-smoke.ARPA>
Lines: 22
Summary: SysV.2 lint catches this... most compilers don't

> Schauble@MIT-MULTICS.ARPA (Paul Schauble)

> I was recently suprised to discover that Microsoft C and two different
> lints all pass without complaint the sequence

>     struct sa { int a; } a;
>     struct sb { int b; } b;
>     void sub(x) struct sa *x; {}
>     void main(){sub(&b);}

(I've made minor edits to the example to turn it into a lint-able unit.)
Our SysV.2 lint catches this problem, complaining of "argument used
inconsistently".  Our compiler does not complain, but then most
compilers don't complain about such things.  Which two lints didn't
catch it?  This seems a very odd thing for any lint to miss.

--
Old computers go into computer museums,
but old software goes into production.
                                        --- Eric Bush
-- 
Wayne Throop      <the-known-world>!mcnc!rti-sel!dg_rtp!throopw