Hi Chao-Te and Homin
These are two different problems. In the first case, the USB subsystem
was not initialising properly. Please see Alec's response below. In
the second case, it was complaining that it could not mount the first
partition. I must apologise for the confusion regarding this second
error. The wiki still reports the old method of mounting a loopback
filesystem, but the new U-Boot expects a native ext2/3 partition on
the flash drive. I sent an email to the mailing list with details on
May 25th, but hadn't updated the wiki yet. I'm hoping you have a Linux
machine available and are familiar with creating partitions and
formatting disks.
Alec asked me to forward his reply to the list in answer to the
problems you had in the first case...
Jason
On 17 Jun 2009, at 04:24, Alec Rust wrote:
Hi Chao-Te,
We have noticed this problem an are working on it. There is a
workaround though.
When uboot boots interrupt the autoboot sequence and type "usb
start" at the prompt. This initializes the USB and subsequent USB
operations in Linux will work. Please use this method until we have
figured out why it does not initialize properly in Linux.
Regards
Alec
Chao-Te Li wrote:
Hi Jason,
I adjusted the switch. however, I don't have any luck
with the usb drives. I tried 2 1G usb drives with the
file roach in them. the error messages are different,
as attached. please have a look.
Chao-Te
--
Open WebMail Project (http://openwebmail.org)
---------- Original Message -----------
From: Homin Jiang <[email protected]>
To: Jason Manley <[email protected]>
Cc: Chao-Te Li <[email protected]>, [email protected],
[email protected], [email protected]
Sent: Wed, 17 Jun 2009 14:38:46 +0800
Subject: picky USB
Hi Jason:
We have succeeded installation of Debain linux in the ROACH boards
following your instruction with adding few command lines in the loop
back file system to bring up the ethernet.
One question is there specification for the USB stick ? I have tried
couple of them, only one 1 Gb USB barely works.
cheers
homin
Jason Manley wrote:
Hi Chao-Te
The version of U-Boot you are running is correct, but the new bus
speeds
have not taken effect. If you are happy with it like this, then
it is
not detrimental, but you can improve PPC-FPGA communication speed.
I think Mo shipped the first ROACH boards with the config DIP
SW3.1 in
the "on" position, thereby overriding the PPC's bootstrap options
into
configuration C. We want configuration H (all switches off, flag in
Actel set), which instructs the PPC to retrieve its configuration
data
from an onboard I2C EEPROM.
When you updated, one of the steps was to run the "init_eeprom"
macro in
U-Boot. This updated the EEPROM with the new speeds. In order to
use
them, make sure that all DIP switches are set to "off" and hard-
power
cycle the board.
You should see:
CPU: AMCC PowerPC 440EPx Rev. A at 495 MHz (PLB=165, OPB=82,
EBC=82
MHz)
No Security/Kasumi support
Bootstrap Option H - Boot ROM Location I2C (Addr 0x52)
32 kB I-Cache 32 kB D-Cache
Board: Roach
I2C: ready
DTT: 1 is 31 C
DRAM: (spd v1.0) 512 MB
FLASH: 64 MB
USB: Host(int phy) Device(ext phy)
Net: ppc_4xx_eth0
Roach Information
Serial Number: 020139
Monitor Revision: 6.5.1429
CPLD Revision: 2.0.5
If it still comes up in Bootstrap option C, check the flag in the
Actel
(0xFFFF). You can use this program for querying it:
http://casper.berkeley.edu/svn/trunk/roach/sw/roach_monitor/roach_monitor.py
--
Regards,
Homin Jiang(¦¿§»©ú)
------- End of Original Message -------
