From owner-freebsd-hackers  Tue Oct 28 18:21:34 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id SAA10988
          for hackers-outgoing; Tue, 28 Oct 1997 18:21:34 -0800 (PST)
          (envelope-from owner-freebsd-hackers)
Received: from ocean.campus.luth.se (ocean.campus.luth.se [130.240.194.116])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id SAA10983
          for <hackers@FreeBSD.ORG>; Tue, 28 Oct 1997 18:21:28 -0800 (PST)
          (envelope-from karpen@ocean.campus.luth.se)
Received: (from karpen@localhost)
	by ocean.campus.luth.se (8.8.5/8.8.5) id DAA07708;
	Wed, 29 Oct 1997 03:29:13 +0100 (CET)
From: Mikael Karpberg <karpen@ocean.campus.luth.se>
Message-Id: <199710290229.DAA07708@ocean.campus.luth.se>
Subject: Re: Parity Ram
In-Reply-To: <199710290129.RAA20291@implode.root.com> from David Greenman at "Oct 28, 97 05:29:28 pm"
To: dg@root.com
Date: Wed, 29 Oct 1997 03:29:12 +0100 (CET)
Cc: hackers@FreeBSD.ORG
X-Mailer: ELM [version 2.4ME+ PL31H (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

According to David Greenman:
> >
> >[...Discussion on ECC/parity/no-parity memory...]
> >
> >I seem to recall something about partiy and/or ECC memory having slower
> >access rates, or something, and therefor being a bad thing preformace-wise
> >but a good thing safety-wise?
> >
> >I don't know where I got this, but could anyone with knowledge in the
> >subject maybe enlighten me on the amount of truth behind this?
> 
>   In order to update the memory, the ECC must be recalculated over the
> entire 64bit quadword. This escentially means that you have to read the
> memory first, apply the changes/calculate the new ECC and then write it
> back. Obviously,this makes memory writes quite a bit slower.

Hmm... It's still not quite clear to me. That is, does this slow my
computer down, in case I use ECC?

It seems to me all this could be done on the DIMM/SIMM, or something,
possibly clocked at multiple of the bus clockspeed, and therefor
not effect the rate at which memory could be read/written over the bus
by the CPU.

If that's not the case, and the computer is actually slowed down by ECC,
how much performace do you loose? 0.1%? 5%? 30%?

  /Mikael