From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay01.roc.ny.frontiernet.net (relay01.roc.ny.frontiernet.net [66.133.131.34]) by ozlabs.org (Postfix) with ESMTP id 37F212BDB5 for ; Thu, 9 Dec 2004 00:04:25 +1100 (EST) Message-ID: <003801c4dd26$d0e30090$0301a8c0@chuck2> From: "Mark Chambers" To: References: <41B67FCF.6040100@tait.co.nz> <1102498710.414.98.camel@linpc003.aimsys.nl> Date: Wed, 8 Dec 2004 08:06:56 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Cc: ppc embedded list Subject: Re: Exception handling on mpc8xx cores List-Id: Linux on Embedded PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , > On Wed, 2004-12-08 at 05:15, Robin Gilks wrote: > > Greetings > > > > Any Freescale lurkers out there? > > > > The mpc866 manual implies that if the internal core times out an > > external Transfer Acknowledge bus signal (by exerting the internally > > generated Transfer Error Acknowledge signal) an exception is generated. > > > > I've tried extending the external TA signal well beyond 2040 clock > > cycles the maximum timeout allows and I see the busses start up again > > (indicationg that they are no longer waiting) but I don't see anything > > on TEA and I don't get an exception thrown. > > > > Any clues? Manual error?, silicon error?, my brain error? > > Did you enable the bus monitor in the SYPCR register? > You've also got to have ME set in MSR to get the exception. For whatever it's worth, I have an mpc852 design (part of 866 family) that uses external acknowledge and it works as advertised. I've seen the acknowledger, a DSP, fail, and we get a machine check exception. (I've never looked at TEA, though). Mark Chambers