Path: utzoo!mnetor!uunet!husc6!bloom-beacon!gatech!purdue!i.cc.purdue.edu!j.cc.purdue.edu!pur-ee!iuvax!inuxc!ihnp4!alberta!sask!zaphod!billj
From: billj@zaphod.UUCP (Bill Jones)
Newsgroups: comp.unix.wizards
Subject: Re: Problems with nmake (from the toolchest) under 4.3BSD
Message-ID: <2215@zaphod.UUCP>
Date: 6 Feb 88 03:47:45 GMT
References: <1773@rayssd.ray.com>
Reply-To: billj@zaphod.UUCP (Bill Jones)
Organization: Develcon Electronics, Saskatoon SK Canada
Lines: 22

In article <1773@rayssd.ray.com> dhb@rayssd.ray.com (David H. Brierley) writes:
>Is anyone successfully using nmake (fourth generation make from the
>ATT toolchest) on a 4.3BSD system.  I dont think it makes any difference
>but specifically, MtXinu 4.3+NFS.  I brought over the source code from
>our 4.2 machine, compiled it and installed it.  No indications of any
>errors so far.  However, when I run it I get one or two strange messages
>and then it core dumps.  The messages are usually complaints about
>"source for (insert module or include file here) has changed".  The only
>changes that I have made to the source are fiddling with "MSGFD" to get
>it to run correctly.

We're not running 4.3 here, but I think I may have found and fixed the
same bug.  The Achilles heel is that .h files which #include each other
with a  #ifndef to break the chain cause isintdep to recurse and dump
core.  This will show up as an enormous repetitive backtrace when
you look at it with dbx or adb.  I have a patch to fix this which
borrows the r->mark field to determine if we've been by already, and
takes care not to recurse unless we know it's 0.   I'll send it via
mail to anyone who has a licence for nmake.
-- 
Bill Jones, Develcon Electronics, 856 51 St E, Saskatoon S7K 5C7 Canada
uucp:  ...ihnp4!sask!zaphod!billj               phone:  +1 306 931 1504