From owner-freebsd-current  Tue Feb  6 22:10:47 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id WAA27390
          for current-outgoing; Tue, 6 Feb 1996 22:10:47 -0800 (PST)
Received: from bunyip.cc.uq.oz.au (pp@bunyip.cc.uq.oz.au [130.102.2.1])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id WAA27383
          for <freebsd-current@freebsd.org>; Tue, 6 Feb 1996 22:10:44 -0800 (PST)
Received: from cc.uq.oz.au by bunyip.cc.uq.oz.au 
          id <23264-0@bunyip.cc.uq.oz.au>; Wed, 7 Feb 1996 16:09:35 +1000
Received: from orion.devetir.qld.gov.au 
          by pandora.devetir.qld.gov.au (8.6.10/DEVETIR-E0.3a) with ESMTP 
          id QAA29142 for <freebsd-current@freebsd.org>;
          Wed, 7 Feb 1996 16:03:05 +1000
Received: by orion.devetir.qld.gov.au (8.6.10/DEVETIR-0.3) id PAA22140;
          Wed, 7 Feb 1996 15:56:51 +1000
Date: Wed, 7 Feb 1996 15:56:51 +1000
From: Stephen McKay <syssgm@devetir.qld.gov.au>
Message-Id: <199602070556.PAA22140@orion.devetir.qld.gov.au>
To: freebsd-current@freebsd.org
cc: syssgm@devetir.qld.gov.au
Subject: Re: compiles hanging
X-Newsreader: NN version 6.5.0 #1 (NOV)
Sender: owner-current@freebsd.org
Precedence: bulk

Stephen McKay <syssgm@devetir.qld.gov.au> wrote:

>"Marc G. Fournier" <scrappy@ki.net> wrote:
>
>> [Stuck compile description elided]
>
>I have this problem with kernels built from sources dated 1996-01-31
>and 1996-02-05.  A kernel dated 1996-01-08 is ok, but that's kinda old.

>The machine is a 386SX16 with 4MB ram 10MB swap + 3.5MB vnconfig swap using
>NFS heavily.  Memory is short and paging is rampant.

Replying to myself here (proving that everyone else is asleep or there's
some huge party in the USA that I've not been invited to):

I've noticed that, in the output from vmstat -s, both 'traps' and 'total
VM faults taken' are increasing at the rate of about 2150 per second.
If this was for real, this would imply about 8MB/s is being faulted in.
I don't think my little 386SX could do anything of the sort.

Thus, my theory is that it is faulting and the fault handler is not producing
the goods, whether it is the wrong page, or the right page but not in the
page tables, or some such.

Whichever it might be, it seems unlikely that I will be able to find the
answer, seeing as the VM system is so simple and straight-forward. :-)
Over to you John and David!

Stephen.