Linux booting halts

[Gian]

hi to all!
So, i have a 547x/548x lite evaluation board with MCF5484 processor.
i build a linux image with LTIB and succesfully downloaded and executed colilo.
i have also downloaded my linux kernel image called "vmlinux.bin". Following the freescale's linux target image user's guide (specific for MCF54x5 and MCF54x4) after the download, it says to type "config save" and "g 2000" commands...
the commands output is shown below:

colilo>set cl mac0=00:04:9f:b5:69:3a noinitrd root=/dev/nfs rw nfsroot=192.168.1.39:/tftpboot/ltib ip= 192.168.1.40:192.168.1.39:192.168.1.1:255.255.255.0:C
oldFire:eth0ff mtdparts=phys_mapped_flash:256k(Colilo),2816k(User)
Command line:'mac0=00:04:9f:b5:69:3a noinitrd root=/dev/nfs rw nfsroot=192.168.1.39:/tftpboot/ltib ip= 192.168.1.40:192.168.1.39:192.168.1.1:255.255.255.0:C
oldFire:eth0ff mtdparts=phys_mapped_flash:256k(Colilo),2816k(User)'
colilo>set kfl 0
colilo>set mac 00:04:9f:b5:69:3a
MAC address:'00:04:9f:b5:69:3a'
colilo>set ip 192.168.1.40
IP address:'192.168.1.40'
colilo>set server 192.168.1.39
TFTP Server: '192.168.1.39'
colilo>set netmask 255.255.255.0
Netmask: '255.255.255.0'
colilo>set gw 192.168.1.1
Gateway IP address:'192.168.1.1'
colilo>set image vmlinux.bin
Image: 'vmlinux.bin'
colilo>tftp 0x1000
TFTP download:
Download address:1000
Using FEC:0
Board MAC: '00:04:9f:b5:69:3a'
Board IP: '192.168.1.40'
Board netmask:'255.255.255.0'
Board gateway:'192.168.1.1'
TFTP server:'192.168.1.39'
Image:'vmlinux.bin'
Initializing TCP/IP stack...
Receiving file 'vmlinux.bin' from '192.168.1.39'
-
Image size = 2699264 bytes

# so the download succeed succesfully!
# but now:

colilo>config save
Code Manufactor detected = F000 (S/B 0x89)
Code Device detected = F000 (S/B 0x8803)
Boot Device detected = 88C5 (S/B 0x88C3)
Burning 848 bytes to flash 0xf0020000 from 0x1febc60
Starting flash sector: 1
Erase 1 128K sectors, starting 1, address f0020000
Retry = 600000
Problems with sector erasing!...
Programing...
Retry = 1024
Problems with data writing!...
data:436F in flash: F002 addr:20000
Problems with data writing!...
Retry = 1024
Problems with data writing!...
data:4C69 in flash: F002 addr:20002
Problems with data writing!...
Retry = 1024
Problems with data writing!...
data:4C6F in flash: F002 addr:20004
.....
Problems with data writing!...
data:BFC4 in flash: F002 addr:2034E
Problems with data writing!...
848 bytes written

# i tried to run the image instead

colilo>g 2000
Linux version 2.6.10 (gianluca@gianluca) (gcc version 3.4.0) #1 Fri Jun 30 15:30:47 CEST 2006
starting up linux rev 0.2: startmem 0xc0296000, size 61MB
Built 1 zonelists
Kernel command line: mac0=00:04:9f:b5:69:3a noinitrd root=/dev/nfs rw nfsroot=192.168.1.39:/tftpboot/ltib ip=192.168.1.40:192.168.1.39:192.168.1.1:255.255.
255.0:ColdFire:eth0ff mtdparts=phys_mapped_flash:256k(Colilo),2816k(User)
PID hash table entries: 512 (order: 9, 8192 bytes)
RTC driver is not initialized yet, returning 1970.
Console: colour dummy device 80x25
Dentry cache hash table entries: 16384 (order: 3, 65536 bytes)
Inode-cache hash table entries: 8192 (order: 2, 32768 bytes)
Memory: 62376k/65536k available (1936k kernel code, 1032k data, 96k init)
Mount-cache hash table entries: 1024 (order: 0, 8192 bytes)
NET: Registered protocol family 16
Linux/m68k PCI BIOS32 revision 0.05
ColdFire PCI Host Bridge (Rev. 0) detected
PCI: Probing PCI hardware
ColdFire internal SEC driver version 0.11

# and here the booting halts

Could anyone help me?
Thanks, Gianluca Salvador

[OL]

Hello all

Someone looks have send to Gian good answer for this trouble, is it possible to have here this answer? I have exactly the same problem
Linux boot and stop after display:

Coldfire internal SEC driver version 0.11

Please help

Thanks a lot

Olivier

[Gian]

Hi Olivier!

This is a kernel configuration issue. You need to disable:

Device Drivers --->
Misc drivers --->
[*] Security

Device Drivers --->
Memory Technology Drivers --->
[*} MTD support

After that it will work!
Regards,
Gian

Message Edited by Gian on 2006-07-12 01:53 AM

[OL]

Really thanks, now it work until kernel panic because it can't connect NFS server as you notice!

I'm going to try solve this, perhaps have the solution?!!!

Regards

Olivier

[Gian]

Unfortunately yesterday i tried to put colilo in boot-flash (and so erasing DBUG) to avoid the kernel image's download and the net-parameters settings every time. The colilo image work from RAM, but NOT from FLASH!!!!!
So now i must restore the DBUG monitor through BDM interface...
At now i didn't resolve the nfs problem yet, so the first do it, could help the other, ok?
Bye,
Gianluca

[OL]

Hello

Some news, NFS not work but messages change!

Perhap do you not have "portmap" service running?

I have notice a strange problem configuration I have put something like this for IP address that can access to my NFS server: 192.168.1.*
and it looks not work I have replace it by: 192.168.1.0/255.255.255.0
This was need for me for test NFS server from my computer

as root I have done this sort of test:
mount 192.168.1.39:/tftpboot/ltib /mnt/ltib

You should be able to mount it, for me it work

Now I have probably to look in NFS right, I not understand how the coldfire card is identify.

Olivier

[Gian]

Hi OL!
Some news: now my nfs seems work a little bit...
I have updated my nfs-utils, changed the /tftpboot permissions in 775 and now during the kernel execution, it said

[...]
IP-Config: Complete:
device=eth0, addr=192.168.1.40, mask=255.255.255.0, gw=192.168.1.1,
host=ColdFire, domain=, nis-domain=(none),
bootserver=192.168.1.39, rootserver=192.168.1.39, rootpath=
Looking up port of RPC 100003/2 on 192.168.1.39
Looking up port of RPC 100005/1 on 192.168.1.39
VFS: Mounted root (nfs filesystem).
Setting the hostname to freescale
Mounting filesystems
mount: Mounting usbfs on /proc/bus/usb failed: No such file or directory
Starting syslogd and klogd
Setting up networking on loopback device:
Setting up networking on eth0:
Adding static route for default gateway to 192.168.0.1:
RPC: sendmsg returned error 101
nfs: RPC call returned error 101
RPC: sendmsg returned error 101

and here stops.

So i don't know if there is an nfs problem still, but i didn't understand why it has tried to setup eth0 an other time. Perhaps, the gw is wrong... and maybe this could be the origin of the 101 error problem...
If you have any suggestion...
Regards,
Gianluca Salvador

[Gian]

Hi to all!
I have linux prompt on ColdFire!!!!!!!
I solved the nfs problem and also the last problem. That was a eth0 setup issue: In fact i rebuild the linux kernel image checking the dhcp flag for the ethernet configuration. Now i have -sh-2.05b# and theworld is beautifull :smileywink: almost untill the next problem ;-P
Regards,
Gian

[OL]

Yes happy to you

I haven't still totaly solve my problems, but I have a problem issue in my Linux Box I have probably to reinstall it!

Good work

Olivier

[OL]

Hello

There is a specific colilo for flash, and it is wrote it is possible to have dbug and colilo on the flash.

For NFS, of course if I find something I put some information here, probably this is a mistake on NFS server config.

Olivier

[OL]

Thanks a lot Gian for your help

I try it tonight!

Regards

Olivier

[Gian]

Hi to all!
Thanks to Coray for the suggestion and for his linux.bin image:
I tried it and works for itself, but now i have problems with my server PC, not with linux image. Infact as shown in the output below, target linux cannot mount /tftpboot/ltib because the server returned error (-13 it says).
The nfs server was configurated and restarted as sad in the user's guide.
I looked in my /dev directory but i didn't find any nfs device, so i thought that it could be the problem. Isn't it?
If so, how could i solve it?


As you request me by mail i show you also the colilo "?" output
now it works correctly and my beginning problem was a bit stupid: i didn't know that colilo works by default at 19200 baud rate, while my coldfire and minicom was setted for 115200!
So I seen garbage when i tried to run colilo!!!

Another thing: where can i found documentation that explain the rules for set colilo and colilo kernel command line?

Thanks and regards,
Gianluca Salvador


colilo>?
Coldfire Linux Loader
Platform: -- Colilo M547X - M548X Bootloader --
-- 2005 Freescale/2004 Motorola(c) --
Commands available:
b - boot default image
m [addr] - memory dump at addr
g [xfer_addr] - go to xfer_addr
l [down_addr] [port] - load image from port
f [down_addr] [size] [val] - fill memory with value
pl [addr] [val] - set the 32 bit value at address
ps [addr] [val] - set the 16 bit value at address
pc [addr] [val] - set the 8 bit value at address
rl [addr] - read a 32 bit value at address
rs [addr] - read a 16 bit value at address
rc [addr] - read a 8 bit value at address
c [source_addr] [dest_addr] [size] - copy memory
d [source_addr] [dest_addr] [size] - decompress image
x [down_addr] [dest_addr] [xfer_addr] [port] - cmds l, d, g
z [source_addr] [dest_addr] [xfer_addr] - cmds d, g
TCP/IP commands:
set mac [MAC address] - set board MAC address
set ip [IP address] - set board IP address
set gw [IP address] - set board gateway address
set netmask [IP address] - set board netmask
set server [IP address] - set address of TFTP server for TFTP download
set image [Image name] - set name of image for downloading
set fec [fec number] - select FEC to use
set cl [command_line] - set kernel command line
set kfl [0/1] - set to 1 if kernel image is in flash, 0 - otherwise
tftp [down_addr] - download image from TFTP server
bootp - request bootp information
Where:
flash_addr - address in flash to start burning from
xfer_addr: hex address to start running at
down_addr: hex address of downloaded image
dest_addr: hex address of image destination
source_addr: hex address of compressed image
size: size of compressed image
port: 0 = /dev/ttyS0, 1 = /dev/ttyS1
IP address: IP address in format xxx.xxx.xxx.xxx
MAC address: MAC address in format xx:xx:xx:xx:xx:xx
Image name: string, filename
fec number: 0 or 1
Default values for paramaters:
down_addr: 1000
source_addr: 7fc10000
dest_addr: 1000
xfer_addr: 2000
flash_addr: 0
size: 130000
downloadPort: 1
kernel command line: root=/dev/ram
kernel image in flash: no
start booting automaticaly: no
Current TCP/IP configuration:
Board MAC: 00:04:9f:b5:69:3a
Board IP: 192.168.1.40
Netmask: 255.255.255.0
Board gateway: 192.168.10.1
TFTP server: 192.168.1.39
Image name: linux.bin
colilo>

colilo>g 2000
Linux version 2.6.10 (coray@fastball) (gcc version 3.4.0) #3 Thu Jul 6 16:48:26 CDT 2006
starting up linux rev 0.2: startmem 0xc0270000, size 61MB
Built 1 zonelists
Kernel command line: mac0=00:04:9f:b5:69:3a noinitrd root=/dev/nfs rw nfsroot=192.168.1.39:/tftpboot/ltib ip=192.168.1.40:192.168.1.39:192.168.1.1:255.255.255.0:ColdFire:eth0f
f
PID hash table entries: 512 (order: 9, 8192 bytes)
RTC driver is not initialized yet, returning 1970.
Console: colour dummy device 80x25
Dentry cache hash table entries: 16384 (order: 3, 65536 bytes)
Inode-cache hash table entries: 8192 (order: 2, 32768 bytes)
Memory: 62528k/65536k available (1800k kernel code, 1016k data, 96k init)
Mount-cache hash table entries: 1024 (order: 0, 8192 bytes)
NET: Registered protocol family 16
Linux/m68k PCI BIOS32 revision 0.05
ColdFire PCI Host Bridge (Rev. 0) detected
PCI: Probing PCI hardware
SCSI subsystem initialized
Initialize Multi-channel DMA API Alpha v0.3 (2004-04-26)
Initializing Cryptographic API
rtc: RV5C387 RTC driver successfully loaded
MCF547x/8x DSPI device driver installed
Coldfire DSPI driver is loaded
ISA-Philips-Basic-CAN memory mapped CAN Driver 3.1_ColdFire_FlexCAN (c) Jul 6 2006 16:22:57
ColdFire internal UART serial driver version 1.00
ttyS0 at 0xe0008600 (irq = 99) is a builtin ColdFire UART
ttyS1 at 0xe0008700 (irq = 98) is a builtin ColdFire UART
ttyS2 at 0xe0008800 (irq = 97) is a builtin ColdFire UART
ttyS3 at 0xe0008900 (irq = 96) is a builtin ColdFire UART
io scheduler noop registered
io scheduler anticipatory registered
io scheduler deadline registered
io scheduler cfq registered
RAMDISK driver initialized: 16 RAM disks of 8192K size 1024 blocksize
ColdFire internal FEC driver version 0.11
Loaded prism54 driver, version 1.2
i2c /dev entries driver
i2c-algo-mcf.o: I2C ColdFire algorithm module is loaded.
NET: Registered protocol family 2
IP: routing cache hash table of 1024 buckets, 8Kbytes
TCP: Hash tables configured (established 8192 bind 16384)
Initializing IPsec netlink socket
NET: Registered protocol family 1
NET: Registered protocol family 17
NET: Registered protocol family 15
IP-Config: Complete:
device=eth0, addr=192.168.1.40, mask=255.255.255.0, gw=192.168.1.1,
host=ColdFire, domain=, nis-domain=(none),
bootserver=192.168.1.39, rootserver=192.168.1.39, rootpath=
Looking up port of RPC 100003/2 on 192.168.1.39
Looking up port of RPC 100005/1 on 192.168.1.39
Root-NFS: Server returned error -13 while mounting /tftpboot/ltib
VFS: Unable to mount root fs via NFS, trying floppy.
VFS: Cannot open root device "nfs" or unknown-block(2,0)
Please append a correct "root=" boot option
Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(2,0)

[dyon]

Thanks guys,

This was just the info I needed.
I've started working on coldfire-linux almost 2 months now, and
this trick helped a great deal.

I don't have the final answer to the "kernel panic", but I got it running with a workaround: I exported the original directory in which the root-file system is generated by LTIB (/my-ltib-root-dir/rootfs) directly, in stead of using the link in /tftpboot.

I'm new to linux but for your information, this is how it worked:
I used the /etc/exports - file and added:
/path-to-my-ltib-root-dir/rootfs 192.168.1.202(rw,async,no_root_squash)

in which "192.168.1.202" is the IP-address of my target board

Remarks:
* In the setup for the command-line (colilo>set cl mac0= ... etc etc) refer directly to:
nfsroot=my-host-IP-address:/path-to-my-ltib-root-dir/rootfs

* When I do the start-up sequence I have to switch serial communication to 115200 baud-rate, after the linux kernel has started.