Book Image

Mastering Embedded Linux Programming - Third Edition

By : Frank Vasquez, Chris Simmonds
5 (3)
Book Image

Mastering Embedded Linux Programming - Third Edition

5 (3)
By: Frank Vasquez, Chris Simmonds

Overview of this book

If you’re looking for a book that will demystify embedded Linux, then you’ve come to the right place. Mastering Embedded Linux Programming is a fully comprehensive guide that can serve both as means to learn new things or as a handy reference. The first few chapters of this book will break down the fundamental elements that underpin all embedded Linux projects: the toolchain, the bootloader, the kernel, and the root filesystem. After that, you will learn how to create each of these elements from scratch and automate the process using Buildroot and the Yocto Project. As you progress, the book will show you how to implement an effective storage strategy for flash memory chips and install updates to a device remotely once it’s deployed. You’ll also learn about the key aspects of writing code for embedded Linux, such as how to access hardware from apps, the implications of writing multi-threaded code, and techniques to manage memory in an efficient way. The final chapters demonstrate how to debug your code, whether it resides in apps or in the Linux kernel itself. You’ll also cover the different tracers and profilers that are available for Linux so that you can quickly pinpoint any performance bottlenecks in your system. By the end of this Linux book, you’ll be able to create efficient and secure embedded devices using Linux.
Table of Contents (27 chapters)
1
Section 1: Elements of Embedded Linux
10
Section 2: System Architecture and Design Decisions
18
Section 3: Writing Embedded Applications
22
Section 4: Debugging and Optimizing Performance

The observer effect

Before diving into the tools, let's talk about what the tools will show you. As is the case in many fields, measuring a certain property affects the observation itself. Measuring the electric current in a power supply line requires measuring the voltage drop over a small resistor. However, the resistor itself affects the current. The same is true for profiling: every system observation has a cost in CPU cycles, and that resource is no longer spent on the application. Measurement tools also mess up caching behavior, eat memory space, and write to disk, which all make it worse. There is no measurement without overhead.

I've often heard engineers say that the results of a profiling job were totally misleading. That is usually because they were performing the measurements on something not approaching a real situation. Always try to measure on the target, using release builds of the software, with a valid dataset, using as few extra services as possible...