Xref: utzoo comp.unix.wizards:22824 comp.unix.i386:6685
Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!asuvax!mcdphx!riscokid.UUCP!fnf
From: fnf@riscokid.UUCP (Fred Fish)
Newsgroups: comp.unix.wizards,comp.unix.i386
Subject: Re: Implementing NULL trapping on AT&T SVR3.2(.2)
Message-ID: <13181@mcdphx.phx.mcd.mot.com>
Date: 10 Jul 90 14:05:11 GMT
References: <412@minya.UUCP> <13291@smoke.BRL.MIL> <1990Jul5.174608.17336@eci386.uucp> <1990Jul6.115941.11096@cbnews.att.com> <PCG.90Jul7172536@odin.cs.aber.ac.uk> <PCG.90Jul9124816@odin.cs.aber.ac.uk>
Sender: listen@mcdphx.phx.mcd.mot.com
Reply-To: fnf@riscokid.UUCP (Fred Fish)
Followup-To: comp.unix.wizards
Organization: Motorola Microcomputer Division, Tempe, Az.
Lines: 11

In article <PCG.90Jul9124816@odin.cs.aber.ac.uk> pcg@cs.aber.ac.uk (Piercarlo Grandi) writes:
>	possible to demand load it. We could have instead aligned it to the
>	address immediately after the end of the COFF headers, but ld does not
>	give us a primitive with the size of the COFF header. We therefore
>	align code to a page boundary, ...

The implementations I am familiar with support the construct SIZEOF(.headers)
to obtain the aggregate size of all the COFF headers.  I don't know if this
is just undocumented in some implementations or actually nonexistant in them.

-Fred