I also started writing Linux kernel code a while ago and the one thing that I found really hard to follow was the locking hierarchies. It's not completely clear what protects what, and especially when you have two mutually exclusive locks (like a global lock and a lock for a specific variable) and you have to make sure you only grab one. RCU was much simpler to use. But my main worry with series like these is that they'll get out of date very quickly, but I mainly read them to understand the context of "how does this code differ in execution context to userspace code?". This series is very helpful in this regard.