Date: Fri, 31 Mar 2017 12:56:06 +0200 From: Rasmus Liland <jensrasmus@gmail.com> To: Jan Sieka <jan.sieka@gmail.com>, freebsd-arm <freebsd-arm@freebsd.org> Subject: Re: DB-88F6XXX kernel on 88F6281_A0 (GoFlex Net) Message-ID: <20170331105606.GA31398@jrl.uk.to> In-Reply-To: <CAGeh=dUE1_mCUSeRGUYr5iqJu-scuJ64M1-i0-7Beyskgb_ZPg@mail.gmail.com> References: <20170330232907.GA21389@jrl.uk.to> <CAGeh=dX1g1kHNMr3Mhb=bwEgpx_W=F_EXTeO1_ihP8qE=sysew@mail.gmail.com> <CAGeh=dUE1_mCUSeRGUYr5iqJu-scuJ64M1-i0-7Beyskgb_ZPg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2017-03-31 09:01 +0200, Jan Sieka wrote: > Hi Rasmus! > > I searched a little and based on that file: > https://svnweb.freebsd.org/base/releng/10.3/sys/arm/mv/kirkwood/std.kirkwood?view=markup > and the wiki you mentioned I'd try to load the kernel under > 900000 and go to that address: > > tftpboot 900000 kernel.bin > go 900000 Hi Jan! When I try to load the kernel.bin at address 900000, this happens: | Marvell>> tftpboot 900000 kernel.bin | Using egiga0 device | TFTP from server 10.6.6.1; our IP address is 10.6.6.26 | Filename 'kernel.bin'. | Load address: 0x900000 | Loading: ################################################################# | ################################################################# | ############# U-boot is stalling here, and at the other end of the ethernet cable the FreeBSD tftpd process is discovering it as a timeout ACK in stderr. Thus, what I did in the earlier email was decrementing the address to 500000, or even as high as 600000, then this happens: | Marvell>> tftpboot 600000 kernel.bin | Using egiga0 device | TFTP from server 10.6.6.1; our IP address is 10.6.6.26 | Filename 'kernel.bin'. | Load address: 0x600000 | Loading: ################################################################# | ################################################################# | ################################################################# | ################################################################# | ####################################### | done | Bytes transferred = 4380232 (42d648 hex) | Marvell>> go 600000 | ## Starting application at 0x00600000 ... | KDB: debugger backends: ddb | KDB: current backend: ddb | Copyright (c) 1992-2016 The FreeBSD Project. | Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 | The Regents of the University of California. All rights reserved. | FreeBSD is a registered trademark of The FreeBSD Foundation. | FreeBSD 10.3-STABLE #0 r304297: Thu Mar 30 19:28:22 CEST 2017 | root@node:/usr/obj/arm.arm/usr/src/stable/10/sys/DB-88F6XXX arm | FreeBSD clang version 3.4.1 (tags/RELEASE_34/dot1-final 208032) 20140512 The std.kirkwood file header specifically mentions 900000 to be the right place for the kernel to account for the page table in another place, and the possibility for these to be bounced higher if there is a problem. The next step would be to change the addresses of KERNPHYSADDR, and KERNVIRTADDR to something new. What is the right procedure to calculate new addresses for these? /Rasmus
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20170331105606.GA31398>