Re: [Gumstix-users] BAD MAGIC NUMBER

[<jes...@ma...>]

Hi-

I'm having the same problem.  I suspect it's something more than a CRC
issue.  Perhaps a u-boot problem?

I've been following this and other similar threads.  The latest wisdom
suggested that this was fixed in 1522 by removing some kernel patch.  So I
started from scratch, from svn co ....

Anyway, I'm working on a brand new Connex 400 Mhz board with a NetDuo MMC
(and a serial Console-hw/st card).  I'm flashing from an MMC card, which has
*always* worked in the past.  When I first ventured into the post-Verdex
merge code stream, I was encouraged to upgrade to u-boot 1.2, which seemed
to go just fine.  Now here's what happens:

1.  Power-on to U-Boot 1.2 (GUM>) prompt
2.  mmcinit  (works fine, sees the card)
3.  fatload mmc 1 a2000000 rootfs.arm_nofpu.jffs2 (works fine)
4.  pro on 1:0-1 && jera all && cp.b a2000000 40000 ${filesize} (works fine)
5.  fatload mmc 1 a2000000 uImage  (works fine)
6.  katinstall 100000  (works fine)
7.  katload 100000  (works fine)
8.  bootm

At this point the gumstix reboots.  Although there is a bunch of error
messages screaming by related to the '-e' characters in /etc/modprobe.conf
(which have always been there), the logon prompt eventually appears.  Login
with root/gumstix and everything is fine.

NOW, here's where it gets weird:  At the shell prompt I type 'reboot' to
restart the gumstix.  Here's what comes up:

   *** Welcome to Gumstix ***

   DRAM:  64 MB
   Flash: 16 MB
   SMC91C1111-0
   Can't overwrite "serial#"
   Can't overwrite "ethaddr"
   Net:   SMC91C1111-0
   Hit any key to stop autoboot:  0
   Instruction Cache is ON
   Copying kernel to 0xa2000000 from 0x00f00000 (length 0x00100000)...done
   ## Booting image at a2000000 ...
   Bad Magic Number
   GUM> 

Here's the printenv for u-boot:

   GUM> printenv
   bootcmd=icache on; setenv stderr nulldev; setenv stdout nulldev; if pinit
on &&
     fatload ide 0 a2000000 gumstix-factory.script; then setenv stdout
serial; setenv
     stderr serial; echo Found gumstix-factory.script on CF...; autoscr;
else if mmc
     init && fatload mmc 0 a2000000 gumstix-factory.script; then setenv
stdout serial
     ; setenv stderr serial; echo Found gumstix-factory.script on MMC...;
autoscr; el
     se setenv stdout serial; setenv stderr serial; katload 100000 && bootm;
fi; fi
   bootdelay=2
   baudrate=115200
   bootfile=boot/uImage
   verify=no
   ethact=SMC91C1111-0
   serial#=4D551FD90E42A3AF
   ethaddr=1E:D9:0E:42:A3:90
   bootargs=console=ttys0,115200n8 root=1f01 rootfstype=jffs2
reboot=cold,hard
   stdin=serial
   stdout=serial
   stderr=serial

   Environment size: 726/4092 bytes
   GUM>
 
Anyone have any ideas?  I've tried several builds.  Should I go back to
u-boot 1.1.4?  Do I have an invalid u-boot env?  I'm dead in the water.

Thanks in advance.



Hi Chris,

On 9/17/07, Chris Cheshire <che...@ho...> wrote:
> Hi Dave,
>
> I am using build revision 1522 at the moment or though I have tried this
> on
> 1510 and 1482.

I have a prebuilt version of 1421 (that has worked properly in the
past) you could try:
<http://www.davehylands.com/gumstix-wiki/verdex-1421/>

After loading each of rootfs and uImage, you should run crc32 (in u-boot)

crc32 a2000000 ${filesize}

The CRC's for the 1421 files are:
5a56178e        rootfs.arm_nofpu.jffs2
1af44661        uImage


-- 
View this message in context: http://www.nabble.com/BAD-MAGIC-NUMBER-tf4441051.html#a12770456
Sent from the Gumstix mailing list archive at Nabble.com.