Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!ucbvax!bloom-beacon!snorkelwacker!spdcc!ima!haddock!karl
From: karl@haddock.ima.isc.com (Karl Heuer)
Newsgroups: comp.lang.c
Subject: Re: Passing functions in C
Message-ID: <16161@haddock.ima.isc.com>
Date: 14 Mar 90 00:52:46 GMT
References: <1990Mar9.045151.9601@ucselx.sdsu.edu> <3242@hcx1.SSD.CSD.HARRIS.COM> <3243@hcx1.SSD.CSD.HARRIS.COM>
Reply-To: karl@haddock.ima.isc.com (Karl Heuer)
Distribution: usa
Organization: Interactive Systems, Cambridge, MA 02138-5302
Lines: 20

In article <3243@hcx1.SSD.CSD.HARRIS.COM> brad@SSD.CSD.HARRIS.COM (Brad Appleton) writes:
>[The function   int foo() { int i; i = 10; }   will, on some compilers,
>appear to have returned the value 10.  Unreliable?]

Very.  What you're actually getting is whatever happened to be left lying
around in the function-return-value register.

>Should the compiler catch the "missing return" or is that left strictly
>to lint?

As far as the language definition goes, there is no distinction between cc and
lint.  This check is traditionally done only in lint because it generally
requires cross-module analysis to know whether the value that isn't returned
from foo() is actually being used anywhere.

If it *isn't* being used, then the code is legal but sloppy (it should be
declared "void").  This remains legal for for backward compatibility, and,
unfortunately, a lot of code continues to depend on it.

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