From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cia.veriwave.com (mail.veriwave.com [198.107.59.66]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTP id 45E5867A67 for ; Thu, 14 Apr 2005 05:29:28 +1000 (EST) Message-ID: <425D6C03.1070805@kylo.net> Date: Wed, 13 Apr 2005 11:59:15 -0700 From: Kylo Ginsberg MIME-Version: 1.0 To: Fahd Abidi References: <81C69D96BDD30640952C7A404004AA25018DB9@ultsol01.tewks.ultsol.local> In-Reply-To: <81C69D96BDD30640952C7A404004AA25018DB9@ultsol01.tewks.ultsol.local> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Cc: linuxppc-embedded@ozlabs.org Subject: Re: bdi2000 + mpc8560ads, very early breakpoints List-Id: Linux on Embedded PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Thanks; that was instructive. I hadn't read up on the 85xx debug exceptions yet. For posterity, my issue was that even though the code is running out of ddr at this stage, u-boot has left the IVPR pointed at flash. So once head_e500.S invalidated the tlb entry for the flash region, the debug exception vector was inaccessible and a breakpoint sent the processor off the road. As a workaround, I modified u-boot to re-point IVPR when it relocates itself into ddr. Cheers, Kylo Fahd Abidi wrote: >Try the suggestions pointed out in here: > >http://www.ultsol.com/faq-P308.htm > >This is based off suggestions from another customer. Let me know if this >works for you or not. > >Fahd > >-----Original Message----- >From: Kylo Ginsberg [mailto:kylo@veriwave.com] >Sent: Tuesday, April 12, 2005 3:21 PM >To: linuxppc-embedded@ozlabs.org >Subject: bdi2000 + mpc8560ads, very early breakpoints > >I'm running a bdi2000 (with mpc85xx-gdb f/w 1.03) with a mpc8560ads >target, and I'm trying to set breakpoints early on in linux startup (in >the head_e500.S code that sets up tlb's). I'm setting the breakpoints >from within the bdi telnet interface. > >I can: >--succesfully hit a breakpoint set up to any instruction <= the address >of the first 'tlbwe'. >--single step (with the bdi "TI" command) past that 'tlbwe' as far as >I've cared to try > >I cannot: >--hit a breakpoint once the first 'tlbwe' has executed. If I attempt to > >do so, linux startup doesn't proceed, I lose further communication with >the target and the BDI claims "COP Freeze" and I have to reset the >target. > >Btw, the 'tlbwe' in question marks invalid the first tlb1 entry setup by > >u-boot, which pointed at flash address space. Code is running out of >ram at this point. > >I'm probably missing something very basic here. Any ideas? > >Thanks, >Kylo > >_______________________________________________ >Linuxppc-embedded mailing list >Linuxppc-embedded@ozlabs.org >https://ozlabs.org/mailman/listinfo/linuxppc-embedded > >_______________________________________________ >Linuxppc-embedded mailing list >Linuxppc-embedded@ozlabs.org >https://ozlabs.org/mailman/listinfo/linuxppc-embedded > > >