Intermec ck1 Guide De Référence

µClinux System 

With storage cases of Flash and RAM, the processor is able to address 
directly to bits stored in them. In the simplest case, you can run 

µClinux 

kernel by placing the startup code to the Flash into the processors startup 
address. With this kind of a setting, the 

µClinux kernel is in charge of 

doing the hardware setup and placing the necessary segments into the 
RAM. 

A more safe and flexible option is to place a small stand alone piece of code 
called a bootloader to the start offset of the Flash. The bootloader can 
handle the initial settings of the board like the basic hardware setup and 
allows downloading the image to the board. Many times it can also handle 
some environment settings, which enables you to write some simple 
configurations without needing a writable file system for the Flash. You 
can use the 

µClinux with a variety of different bootloaders for specialized 

tasks in different stages of development. Some bootloaders for 

µClinux are 

CoLilo, My Right Boot, Motorola’s dBug, and Power PC boot 
(PPCboot). The PPCboot enables loading the image or images through a 
serial or Ethernet connection with the Trivial File Transfer Protocol 
(TFTP). It also enables the usage of environment variables, booting of a 
compressed and decompressed images and booting kernel from a JFFS2 
partition. 

In 

µClinux, the root file system and kernel are compiled as separate 

entities. In regular Linux distribution, the kernel is embedded as a part of 
the file system. This setting requires a sophisticated bootloader in order to 
place the required parts at the boot to the RAM. The 

µClinux allows the 

file system and kernel partitions to be placed into the devices memory in 
several different ways. The setting depends on whether the image is 
compressed and where the actual execution takes place. When doing the 
development, the most convenient solutions is to download the image 
directly to the fixed offset of RAM and execute it from there, but when the 
image is intended for solid state storage, it should be placed in the systems 
Flash as the Flash memory is usually more expensive and the access to 
32-bit wide RAM is faster than to the 16-bit Flash. In the Flash, you can 
place the image either compressed or as it is. Usually the case of placing 
the kernel plus root file system as it is to the Flash is used with deeply 
embedded devices with this being the only available memory by making 
the root file system read-only. It is also possible to make part of your file 
system read-only and only place the read/write partitions to RAM. 

The amount of used Flash can also be saved by first decompressing the 
kernel and root file system and at power up, uncompressing it to Flash. 
This action requires that the bootloader used is capable of handling the 
uncompressing. With this solution, the footprint of the Flash is 
considerably smaller compared to the image by itself. 

-rw-r--r-- 1 root root 1.6M Oct 16 19:06 image.bin 
-rw-r--r-- 1 root root 712k Oct 16 19:06 image.bin.gz