[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Initialization questions
>>>>> "RG" == Rob Gaddi <firstname.lastname@example.org> writes:
RG> We're designing a board using the ETRAX 100LX. When we first get the
RG> boards back with the chips on them are the network and serial boot
RG> code already present on the chip, and if not, how do we flash the
RG> flash? If so, do we have to run flashitall?
== cut ==
All of the tasks above result in two images: kimage and fimage. The kimage
contains the linux kernel and the read-only root partition (cramfs). The
fimage contains a complete flash image (rescue partition, partition table,
decompressor, compressed kernel, read-only root partition and JFFS
partition). They can be transfered to the board using one of four scripts:
ktest: network boots the kimage from RAM.
kflash: writes the kernel and read-only root partition to flash.
flashit: writes all of the fimage except the rescue partition to flash.
flashitall: writes the entire fimage to flash.
Before you run any of these scripts you should first put your devboard in
boot mode by pressing the boot button, pressing and releasing the reset
button and then releasing the boot button (i.e. the boot button must be
kept down while resetting the developer board). The ktest and flashit
scripts are the most useful. The rescue partition contains the serial
number which is used as the ethernet address, so you probably do not want
to overwrite that, i.e. don't use the flashitall script unless you really
want to overwrite the rescue partition. To test a new kernel type:
etc, etc, etc...