veyepar: What is in a tiny Linux installation

Author(s):	Malcolm Tredinnick
Location	Caro
Date	jan Thu 19	Days Raw Files
Start	16:40	First Raw Start	error-in-template
Duration	0:50:00	Offset	None
End	17:30	Last Raw End
Chapters
Total cuts_time	None min.

http://lca2012.linux.org.au/schedule/111/view_talk
raw-playlist raw-mp4-playlist encoded-files-playlist mp4 svg png

assets release.pdf What_is_in_a_tiny_Linux_installation.json
logs
Admin: episode episode list cut list raw files day marks day marks day image_files

State:

Locked:

clear this to unlock

Locked by:

user/process that locked.

Start:

initially scheduled time from master, adjusted to match reality

Duration:

length in hh:mm:ss

Name:

Video Title (shows in video search results)

Emails:

email(s) of the presenter(s)

Released:

has someone authorised pubication

Normalise:

Channelcopy:

m=mono, 01=copy left to right, 10=right to left, 00=ignore.

Thumbnail:

filename.png

Description:

markdown

Let's take a tour of the lower-level pieces of a functional Linux and see what is involved in making a smaller, possibly simpler system. Perhaps for installation on resource-constrained hardware; perhaps for restricted use systems; perhaps just for enjoyment and self-education.

It is "well known" that Linux runs on everything from super computers to toasters (well... maybe). For most people, however, running Linux on their desktop or laptop or even in the server room, they are typically running their distribution supplied kernel with distribution supplied libraries for basic interfaces and graphics.

This talk is aimed at intermediate userspace developers with an interest in expanding their knowledge of how Linux and associated support libraries work together.

The idea is to approach the questions:

- How much can you throw out of the Linux kernel build (as in not configuring them in) and still have a useful system? What functionality is available after, for example, compiling a kernel with "make allnoconfig" and where you might want to add things.

- What is the difference between a full-blown C library like GNU glibc, versus smaller libraries such as dietlibc and uClibc? What do you give up when moving from the larger to smaller versions? What does your favourite application really rely on and where are the built in assumptions.

- How small (space-wise) can you go and still have a system that runs useful software, for whatever your own particular value of "useful" might be.

I will not be covering low-level hardware interaction beyond pointing out that the kernel handles that. I will assume we can get the kernel to boot on whatever hardware is required and the talk explores what can be done from that point.

Brief topic list:

- Viewing the Linux kernel as input-output devices, memory management and process execution and working out what is strictly required in each area.

- What happens after booting: why do you end up with a shell prompt or init scripts running?

- "Everything" relies on a C library and graphical applications rely on something like an X server. What do these things really provide and how much can you give up?

- Comparing and contrasting Linux and software stack on a desktop machine with Linux on an in-flight entertainment system or a phone.

- Building and running tiny Linux systems for fun and education. No special hardware needed (kvm is your friend)!

Comment:

production notes