Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 23 Jan 2001 22:49:49 -0800
From:      "Ted Mittelstaedt" <tedm@toybox.placo.com>
To:        <jpaetzel@hutchtel.net>, "'John'" <papalia@udel.edu>
Cc:        <freebsd-questions@FreeBSD.ORG>
Subject:   RE: Install hangs on older system
Message-ID:  <002701c085d1$d9328640$1401a8c0@tedm.placo.com>
In-Reply-To: <3A6DF39D.29657.58F82E@localhost>

next in thread | previous in thread | raw e-mail | index | archive | help
I remember running into this sort of thing during the
transition from the 2.1 series to the 2.2 series, but not
on a Packard Bell, on other hardware.  Other people during
that time also had similar problems.

I don't think there was any consensus on this, but I have a
guesstimate theory on it.  During that time there was a 
change in the compiler version, and I suspect that the newer
version of GCC generated code that does some operation which
has some problem with either certain hardware or certain CPU
versions.  The most likely candidates in my mind are the
caching hardware, because bugs in the motherboard CPU cache
were a known issue with systems of that era, even with other
operating systems like OS/2 and NT.  CPU bugs aren't 
entirely unknown with Intel CPU's either.

I do remember vividly having a 386 system which would crash on
a kernel recompilation with 2.2.x and run fine with a kernel
compilation with 2.1.x  I exchanged the 386 CPU in that system
with another system I happened to be working on one day and
the problem vanished.  Of course the old CPU worked fine in the
other hardware.  The only differences between the CPU's is the
location they were manufactured at, otherwise they were the same
stepping level.  This surprised me quite a bit because there were
no known CPU bugs that even the Intel Secrets site knew about
with those CPU stepping levels.  We probably will never know
the truth but I suspect that over the years there have been a
number of CPU bugs in Intel products that were quietly fixed
with mask changes and such and that were never publicized.

Ted Mittelstaedt                      tedm@toybox.placo.com
Author of:          The FreeBSD Corporate Networker's Guide
Book website:         http://www.freebsd-corp-net-guide.com

>> 
>
>Hmmm...It just so happens that I have one of those, and am able 
>to duplicate the problem.  I can install 2.1.5 just fine, but 4.2 
>chokes.  I even went to far as to put a buslogic SCSI controller and 
>a 400 meg drive in it, and it still hung.  No error message.  I think 
>that you are SOL.
>
>Josh
>



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?002701c085d1$d9328640$1401a8c0>