Well done! We covered a lot of ground in this chapter. First, you learned how to create delays in kernel space, both the atomic and the blocking types (via the *delay() and *sleep() routines, respectively). Next, you learned how to set up and use kernel timers within your LKM (or driver) – a very common and required task. Directly creating and working with kernel threads can be a heady (and even difficult) experience, which is why you learned the basics of doing so. After that, you looked at the kernel workqueue subsystem, which solves complexity (and concurrency) issues. You learned what it is and how to practically make use of the kernel-global (default) workqueue to make your work task(s) execute when required.
The series of three sed (simple encrypt decrypt) demo drivers we designed and implemented showed you a bit of a more sophisticated use case for these interesting technologies: sed1 with the timeout implementation, sed2 adding to the kernel thread to perform...
