Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!rutgers!ucsd!ames!haven!adm!smoke!gwyn
From: gwyn@smoke.BRL.MIL (Doug Gwyn )
Newsgroups: comp.lang.c
Subject: Re: Question about linking files
Message-ID: <9924@smoke.BRL.MIL>
Date: 26 Mar 89 21:45:18 GMT
References: <18925@iuvax.cs.indiana.edu> <16541@mimsy.UUCP>
Reply-To: gwyn@brl.arpa (Doug Gwyn (VLD/VMB) <gwyn>)
Organization: Ballistic Research Lab (BRL), APG, MD.
Lines: 21

In article <16541@mimsy.UUCP> chris@mimsy.UUCP (Chris Torek) writes:
>Given that the pANS does not have the concept of a `library', or
>even of `separate compilation', ...

The pANS does recognize the notion of library and separate compilation;
see Section 2.1.1.1.

According to the pANS, a program consists of a set of translation units
linked together and communicating by well-defined "external" interfaces.
Nowhere in the pANS (that I could find) is there any idea that only a
portion of a translation unit might be linked into a program.  The
means by which available translation units are selected for linking
together into programs is not within the scope of the standard,
although the usual link-editing of multiple object modules with others
selected from libraries to satisfy external references is clearly among
the methods envisioned.

I would say that any link process which dropped a portion of an object
module (presumed to be produced from a single translation unit) would
be non-standard conforming (unless the dropped portion had no detectable
effect on the final program).