[U-Boot-Users] Problems with u-boot, PPChameleonEVB and MontaVista kernel

[Andy Hawkins <a.hawkins@cabletime.com>]

Hi all,

I'm trying to get up and running with u-boot on a PPChameleonEVB,
running the MontaVista kernel.

I can successfully compile u-boot 1.0.0-pre and the kernel that came
with the board (2.4.20 from the eldk), and boot the board
successfully.

However, when I try to compile the MontaVista kernel, the board will
not boot. Their kernel doesn't directly support the PPChameleonEVB, so
I'm using the one for the IBM 405EP eval (which the PPChameleon
appears to be very similar to). I'm 'manually' running make pImage to
build the kernel after the MontaVista targetconf tool has run the
build, and booting from vmlinux.PPCBoot using tftp.

I've been in touch with MontaVista, and they suggest re-building
u-boot, but changing the definition of the bd_info struction (in
include/asm-ppc/u-boot.h) to match that in the kernel. I've done this,
but the boot still stops in the same place.

I've compiled u-boot in debug mode, and this is what we see:

reset

U-Boot 1.0.0-pre (Feb 24 2004 - 10:11:24)

CPU:   IBM PowerPC 405EP Rev. B at 133.333 MHz (PLB=133, OPB=66,
EBC=33 MHz)
       IIC Boot EEPROM disabled
       PCI async ext clock used, internal PCI arbiter enabled
       16 kB I-Cache 16 kB D-Cache
Board: ### No HW ID - assuming PPChameleonEVB
I2C:   ready
DRAM:  32 MB
Top of RAM usable for U-Boot at: 02000000
Reserving 182k for U-Boot at: 01fd2000
Reserving 257k for malloc() at: 01f91900
Reserving 132 Bytes for Board Info at: 01f9187c
Reserving 48 Bytes for Global Data at: 01f9184c
Stack Pointer at: 01f91828
New Stack Pointer is: 01f91828
Now running in RAM - U-Boot at: 01fd2000
FLASH:  4 MB
U-Boot relocated to 01fd2000
NAND:Probing at 0xff000000
  32 MB
### main_loop entered: bootdelay=10

### main_loop: bootcmd="run nfsargs addip addcons;bootm
$(kernel_addr)"
Hit any key to stop autoboot: 10 \b\b\b 0
=> run net_selc\b \bf

ENET Speed is 10 Mbps - HALF duplex connection
TFTP from server 191.53.51.21; our IP address is 191.53.51.200
Filename 'vmlinux.PPCBoot'.
Load address: 0x200000
Loading:

[snip tftp progress]

done
Bytes transferred = 629509 (99b05 hex)
## Booting image at 00200000 ...
   Image Name:   Linux-2.4.18_mvl30-405ep_eval
   Created:      2004-02-24  10:18:07 UTC
   Image Type:   PowerPC Linux Kernel Image (gzip compressed)
   Data Size:    629445 Bytes = 614.7 kB
   Load Address: 00000000
   Entry Point:  00000000
   Verifying Checksum ... OK
   Uncompressing Kernel Image ... OK
## Current stack ends at 0x01F911D8 => set upper limit to 0x00800000
## cmdline at 0x007FFF00 ... 0x007FFF70
bd address  = 0x01F9187C
memstart    = 0x00000000
memsize     = 0x02000000
flashstart  = 0xFFFC0000
flashsize   = 0x00400000
flashoffset = 0x00029B00
sramstart   = 0x00000000
sramsize    = 0x00000000
bootflags   = 0x0000A000
procfreq    = 133.333 MHz
plb_busfreq = 133.333 MHz
pci_busfreq = 33.333 MHz
ethaddr     = 00:50:C2:1E:AF:FE
eth1addr    = 00:50:C2:1E:AF:FD
IP addr     = 191.53.51.200
baudrate    = 115200 bps
## Loading RAMDisk Image at ffd00000 ...
   Image Name:   Simple Embedded Linux Framework
   Created:      2002-10-24   9:30:38 UTC
   Image Type:   PowerPC Linux RAMDisk Image (gzip compressed)
   Data Size:    1476478 Bytes =  1.4 MB
   Load Address: 00000000
   Entry Point:  00000000
   Verifying Checksum ... OK
## initrd at 0xFFD00040 ... 0xFFE687BD (len=1476478=0x16877E)
   Loading Ramdisk to 01e28000, end 01f9077e ... OK
## Transferring control to Linux (at address 00000000) ...

I do have access to a BDI2000, that I've used to break into a running
kernel (at start_here), but can't get it to break when I use the
MontaVista kernel. From this, I assume that it's failing somewhere
very early in the linux boot process.

Can anyone offer any advice?

Many thanks.

Andy