ts-7000
[Top] [All Lists]

Re: [ts-7000] Creation of a 7800 bootable SD card

To:
Subject: Re: [ts-7000] Creation of a 7800 bootable SD card
From: "John Shaffer, PhD" <>
Date: Tue, 08 Jan 2008 13:56:23 -0600
Thanks everyone. That clearly was it.

I happened to have created the SD card using a SuSE 10.3 box (remember, I hosed the TS-7800 and thus couldn't use it to create a bootable SD card there).  But I don't think its a software thing but more of a disk geometry thing of sorts.  Running fdisk on the working SD card on both the suse box and the 7800 report identical info. Furthermore the size of the kernel partition is reported as 8192 on both systems.  However I definitely am NOT a geometry guru.

When I use fdisk on the 7800 to examine the on-board flash device, indeed it shows the kernel partition size as 4096 suggesting a 1024 block size.

This device specific theory is also supported by the fact that when I dd an sdimage.dd image file to the raw device, the card fails to boot, presumably since the partition table specified in the MBR is different from how the actual partitions on the device.

Thanks again!

John Shaffer
Lakota Innovations, LLC
m("lakotainnovations.com","jshaffer");">


Michael Schmidt wrote:


On Mon, January 7, 2008 10:14 am, jhslakota wrote:
> Hello all, I'm a newbie on this group and have a question. Actually I
> solved a problem I had and some of the posts have helped but ultimately
> I did something NOT suggested and solved the problem. Posting here in
> the hope that my description and any discussion will help others. [:)]
>
> Short version:
> The root partition used in the 7800.mbr is only 4096 blocks long (2MB)
> but the kernel image is 3.5MB. This caused me many problems before
> realizing this mismatch.

I suspect what is happening here is that there are some software
disagreements about how big a "block" is. That partition is supposed to
be 4MB in size, which would be 4096 * 1k blocks. In our production
process this all works fine, but there may be some differences between
your environment and ours. For instance, when we blast the flash on a
7800 it is running from an NFS-root environment, and all the commands are
run directly on the 7800.

For example, from the 7800 on my desk I see the following:
/mnt/host # fdisk /dev/tssdcardb

Command (m for help): p

Disk /dev/tssdcardb: 507 MB, 507379712 bytes
255 heads, 63 sectors/track, 61 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System
/dev/tssdcardb1 1 1 4096 c Win95 FAT32 (LBA)
Partition 1 has different physical/logical beginnings (non-Linux?):
phys=(0, 4, 13) logical=(0, 4, 5)
Partition 1 has different physical/logical endings:
phys=(17, 2, 30) logical=(0, 134, 6)
Partition 1 does not end on cylinder boundary
/dev/tssdcardb2 1 2 4096 da Unknown
Partition 2 has different physical/logical beginnings (non-Linux?):
phys=(17, 2, 31) logical=(0, 134, 7)
Partition 2 has different physical/logical endings:
phys=(34, 0, 48) logical=(1, 9, 8)
Partition 2 does not end on cylinder boundary
/dev/tssdcardb3 2 2 4096 da Unknown
Partition 3 has different physical/logical beginnings (non-Linux?):
phys=(34, 0, 49) logical=(1, 9, 9)
Partition 3 has different physical/logical endings:
phys=(50, 7, 5) logical=(1, 139, 10)
Partition 3 does not end on cylinder boundary
/dev/tssdcardb4 2 62 482944 83 Linux
Partition 4 has different physical/logical beginnings (non-Linux?):
phys=(50, 7, 6) logical=(1, 139, 11)
Partition 4 has different physical/logical endings:
phys=(1023, 7, 61) logical=(61, 170, 45)

Partition 4 does not end on cylinder boundary

Command (m for help): q

/mnt/host # dd if=/dev/tssdcardb2 of=tmp
8192+0 records in
8192+0 records out
/mnt/host # ls -l tmp
-rw-r--r-- 1 root dialout 4194304 Jan 7 2008 tmp

Manually resizing the partitions as needed is a valid solution. Both
kernel and initrd partitions are supposed to be 4MB, with the last
(rootfs) parition taking whatever remaining space is left.

______ Best Regards,
|__ __/ Michael Schmidt
|| Software Engineer
||echnologic Systems (EmbeddedARM.com)
|| (480) 16610 East Laser Drive #10
|/ 837-5200 Fountain Hills, AZ 85268
http://oz.embeddedarm.com/~michael


--

John H Shaffer PhD
Managing Partner

507-252-8927
m("lakotainnovations.com","jshaffer");">
logo
Lakota Innovations LLC
2015 Valkyrie Dr NW
Rochester, MN 55901

<Prev in Thread] Current Thread [Next in Thread>
Admin

Disclaimer: Neither Andrew Taylor nor the University of NSW School of Computer and Engineering take any responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU