Unlock all 38 lessons of Linux Driver DevelopmentFull course ₹40,000 · paid
Linux Kernel Introduction: History, Modularity, Subsystems and Versions
A free lesson from Skill-Lync's Linux Driver Development course, on Grow by Skill-Lync · Last updated 16 Nov 2026 · Session: “Week 1 Session 1”
What will you learn in this free Linux Driver Development lesson?
The instructor covers the history of Linux from Linus Torvalds in 1991, its modularity, supported architectures and the GPL version 2 licence. It then explains user space versus kernel space, five kernel subsystems, kernel version numbering with LTS releases, checking the version with uname, and the kernel community.
- Linux features: modular, scalable, multi-architecture
- GPL version 2 licence obligations
- User space, kernel space and five subsystems
- Kernel versions, release candidates and LTS
What comes after lesson 1?
The next free lesson is lesson 2, “Getting, Configuring and Building the Linux Kernel Source” (24 min). Later in Skill-Lync's Linux Driver Development course (paid): lesson 4, Week 2 Session 1; lesson 5, Week 2 Session 2; lesson 6, Week 2 Session 3. The full course has 38 lessons (18 h 31 min of video); see the course overview or its price and syllabus.
Lesson transcript
Machine transcript of the lesson audio (lightly cleaned). Technical terms may contain errors; the video is the reference.
Hello everyone. This is a course on Linux device driver development. This is the first week. So at the end of this week, it's like first we will go through what is introduction to Linux. So what is Linux and how it was developed and who developed it. And the next thing is we will be getting the kernel source. So how to. So first of all, we need to get the source. Since it is an open source, we can get it from the internet. So how to get it and what are the procedures for it.
We will be seeing that. Next once we have a kernel, we need to configure it. So how to configure it for a particular PC or a phone or embedded system. So how to configure it. And Linux building. So how to build Linux and kernel. And how to install it and update it in kernel. So this whole course will be like in the two halves. So in the first half, we will be working only on the PC based system. So maybe open to wires or something else. And the second part will be on using embedded hardware.
So in this course, we will be using a bagel bone black. So in the first half, we will be taking only a PC based system for our development or training. So we will just go into the first section. So introduction to Linux. So what is a Linux? Linux kernel. So we will be calling interchangeable Linux kernel. So this is our main objective of this course. So this Linux kernel is free and open source. So this is developed by volunteers. And it is in the public environment. And it is in free. And we don't need to pay anything for it. And it is an open source. And we have a source available in it. So we can go through the source and how it is implemented.
So it is opposed to a proprietary. So in case of a proprietary OS is like Windows. So you will get only a binary or a image or EXE. So we will not have any access to the source code. We don't know what how it is implemented. And opposed to that Linux kernel is free and open. And it is a modular. So the main advantage of Linux kernel is a modular. So which means each software or each task features can be added or removed depends upon the requirement.
So if you are designing a board. So if you are designing a board. You need a UART. We can add UART for it. If you need a SPY. You can add it. If you don't want. We can remove it. That is a kind of a modularity. And multitasking. Multitasking is. It is not a single task. So it is not run from while one. But it is a multitasking. We can have multiple applications run in parallel. So Linux kernel supports that. So this is developed. So this is developed by Linus Torvalds. In 1991. As a hobby project. So this is a hobby project developed by him. And first he developed only for a single platform.
which is Intel 83-86 processors. And he took Unix as a base. But not a Unix platform. So it is Unix like OS. But not as a Unix. The main advantage what we see in this Linux is. It is a platform independent. Initially maybe Linux targeted only for 83-86 processor. But over the course of a period. It is adapted to multiple platforms. So main thing. So it supports all the major architectures available. So the PC what we are using. So it is based upon X86. So it is a X86 processor. And it supports a thing called ARM. So it is a widely used architecture. And Power PC, Spark and so on so forth.
So it also supports 32-bit as well as a 64-bit processor. So it can run on 32-bit architecture. As well as it can run on 64-bit architecture. So this is a I have downloaded the Linux source. So you will see how to download it. But charge architecture we support. For that I have gone into this folder. So this is a base folder. So Linux 5.8.9. So I took this source. And inside that we have a folder called ARM. So I changed the directory to it. And I listed ls.
So you can see each architecture will have each folder for it. So first you can see alpha. Alpha is one of architecture. And the famous architecture is ARM. So ARM itself has two folders. One for 32-bit implementation. And another for 64-bit implementation. We have like MIPS. MIPS is one of the core which is supported by microchip. And the RISP5 will be called it as PowerPC from Freescale, Spark and X86. So X86 is our PC based systems. And in that itself we will be having two folders. One for 32 and 64.
And these are the major architecture supported by this Linux kernel. So what are the major advantages that if you write an application then the source will be same. But if you just give during compilation which architecture it is, it is compiled to that particular architecture. So we do not need to rewrite application for each architecture. Just during compilation process Linux will take care of that. And these are another features of Linux kernel. Scalability. Scalability is one of the biggest advantages. So it can be able to run on very very small systems like watches. For example we have smart watches. In that itself we can have a Linux kernel running into it.
And we can also see this Linux can be capable of running in the as large as a super computer. Depends upon our needs this Linux is able to scale. So it can scale from a watch to a super computer. So how this is possible is because of a feature called a modularity. So modularity is one of the features that is available in the Linux. So each feature is like a module. So for example UART is one module. And you can see like SPY is one module. I2C is one module. So you don't need all at the time. So you can just if you want I2C, I can just add I2C alone. If I want SPY and I can add SPY alone.
And there is also one more advantage is like we don't need to have it at compile time. So you can have a module can be added at the compile time. So if it is in a part of compile time, then the boot time will increase. If someone wants boot time to be very short and if they want only during the run time. So whenever I need some application to test, I need to load the I2C module. So that is also possible in the Linux. So it gives various features to us. So if you want as a part of a kernel, we can have it. If I want during runtime, I need to load it. That is also possible. And if I don't want, I can unload it.
These are the various advantages Linux is giving to us. And as I said in the first slide, Linux is free and open source. But it is released under distributed under GPL version 2 license. So GPL in the sense it is a general public license. So there are various licenses available. What this GPL version 2 says is right. So we can free to download. So we can download it from the net. And we can modify as per our needs. So we can modify the source code and we can generate the binary and we can test it. Everything we can do.
So if I want to distribute my changes to a customer or if I want to sell the product, then it is a mandatory that we need to give the same right to our recipient, which means we need to give them the source code, whatever we have developed. So that is the main thing. So whatever we modified, it also needs to be free and open source. So others needs to be able to access it. So that is the main thing of this GPL version 3. We need to adhere to it. Otherwise, we will be in the legal. Legal issues will come.
Next thing is kernel. So we are talking about Linux, Linux kernel. And this is a very high level picture of it. So we can see in the top, this application what we work. So this is a user space. So we have applications like editor, browser and many other utilities on that. So all comes under these applications. We can have multiple applications running together at the same time, which is a multitasking. So you can have one editor running. So we have a Vim editor and we can have one more browser running in this application too, which is a Google Chrome application 3.
You can have any editing tools like a GIM, any open source tools, which can be running uncurrently. And there is another layer, which is a kernel space. So this is an important space. It is divided into two major things. One is a kernel subsystem and other is a device. So as I said, there is two subsystems. So one is a kernel subsystem and the device driver. So what we are concerned more about in this course is how writing a device driver. And the major part we will not touch and we will be reusing is a kernel subsystem. So we have this user space application running and the kernel space what we have developed like the kernel subsystem device driver.
So the important thing what bridges between them. So application and this device driver is your system call interface. So this is the thing which will be like what we call signals or interrupts, IOCTL. So application will communicate to this kernel subsystem and if at all there is any data to be transferred back to the application. All will happen when we tell of this system call interface. And device driver. So as the name we communicate with the devices. So this is hardware. So this hardware can be anything. So it can be ARM based or it can be X86 processor. So this is one which is an it can be anything.
So this is an overall picture of it. So we have a user space, we have a kernel space and we have a hardware. So as explained in the previous slide an important thing is a kernel subsystem. So major thing it is the device driver developer we will not be touching this kernel subsystem but we will be using this. So. Our different subsystems are inside this kernel. So, first thing is in process management. So, the kernel is Linux supports multiple processor or multiple task running. So, this process management is very important. So, what is this process management responsible for? It is like creating and destroying a process. So, if you want to create a process and you want to destroy it, so it will keep track of all the resources allocated to it and things like that.
Since the multiple processes are running, we need to have a through-process communication mechanism, like thermophores, minpex, queues, pipes, everything. So, how to communicate with each other. So, that part is also taken care of by this process management. And the important thing is scheduler. So, we need to run the process. So, it should be like, depends upon the scheduling scheme we are selecting. It can be a run-off in scheduling. Depends upon the scheduling mechanism we use, the processes which are ready to be executed will be executed, which are blocked, that will be in the wait queue. There are different states that are available. And the next major thing is memory management. So, computer memory is my major resources. So, we will not be using directly the file physical plan. So, there will be something called as a virtual file system.
So, this is a major topic. And only we will be using a virtual memory, not a physical memory. So, there will be a memory map unit that will take care of mapping this physical memory to the virtual memory. So, that is an important thing. So, it's a memory management. The next thing is a file system. So, we cannot directly access the memory. So, we will be using, from the file system only, we can access the file. So, almost everything is a file. So, it can be a device. It is created as a file. So, for everything, there is a file created in Linux.
And we have multiple file system formats available now. So, from the older legacy, EXPRO, FAT, JFS, many UBIFS, many things are there. So, file system is one of the major components in the kernel. And devices. So, almost every system maps to it physically. So, device drivers are required to access. So, we need devices will be there. We need to write a driver for it. So, then only we can able to access. Next part is networking. So, networking plays an important role as of now. So, we need to communicate with each other. So, that is a very big part.
So, what this networking is capable of. So, what it has to do is, it has to, what are the packets are coming. So, we need to collect it. We need to identify it. So, which process it has to be dispatched. So, everything is taken care of by this network. So, the main part is the networking system. The main part is if there is a routing table. And the ARP is address resolution protocol. So, these are all implemented within the kernel. So, these are the five things. Process management.
Memory management. Memory management. File system. Device control. Networking. These are the major components in the Linux kernel. So, not a device favor. It is our support of a. And next slide. So, this is the kernel version number. And another is a distributed version number. So, these are not different. Meaning is, you can have a laptop. So, which runs on a version, say for example, it is in 18.04. So, Ubuntu or any other distribution like OpenSUSE, ABN, Red Hat. So, we will have their own version number.
But we have a kernel version. So, which will be like 4.15, 5.0. So, this both are not same. So, it means if a laptop comes with 18.4 Ubuntu, doesn't mean it is running a Linux kernel of 18.4. But it uses Linux kernel version of 4.15. And similarly, if you have a version called 19.04, it uses 5.0 kernel version. So, you can check it in the Ubuntu website. We can use a command line to find it. So, ultimately, both are different. So, distribution version and the kernel versions are. So, kernel version number. So, how it is numbered. So, it is following a pattern like X dot Y dot Z.
So, X is a kernel version. And Y is a major version. And Z is a minor version. So, initially what they have done is like if there is an even number, then they called it as a stable version. And if it is an odd number, then it is an experimental version. So, this was followed earlier. But recently, what they did is. So, X dot Y dot Z pattern remains the same. But odd and even number for stable experimental versions. So, what they are doing is. So, they will release like if you want to release a 5.10. So, before that there will be a release called 5.9. So, which is a stable version.
After that there will be N number of experimental. So, each experimental they will be releasing with you as RCN. So, which is an RC stands for release candidate. So, RC1, RC2. So, they will be releasing one by one. And they will go to testing. So, if it is found a bug free, then they will be releasing a stable version called 5.10. So, we will take an example with this 2.6 release. So, what they did is. This is an stable release. So, it is in this color. So, 2.6.17. So, what happens is. After that there will be a 2.6.18 RC1. So, which is experimental version. So, new features will be added. And if there is any bug fixes are there. And that will be added into it. So, 1.2. So, it will go to. There is no fixed rule. We need to have 5 RC release.
It can go up to 6 or 7. The max it can go up to 8 also. So, once this finalized. So, this funnel source is ready for release. Then they will tag it as a 2.6.18. And they will release it. So, then once it is 2.16.18 is released. Then they will start RC release for 2.6.19. And they will go on so on. There are few things like few versions they will call LTS which will be seen in the next slide. And for that there is any bug fixes. So, they will use it as 2.6.17.1.2.3.4. So, it will go on like that. So, bug fix only for but if they have found any bug in the 2.6.17 in later case.
In 2.6.19. So, they can backport it. So, this previous version says. As I said in the previous slide. LTS is a LTS long term support. This terminology is applicable not only for Linux but for also applications like Google, Google, Google distribution, let it tell. So, this is a long time supported version. So, which means if at all they find a bug in the later releases. For example, one consider this 5.04 is long term support. If any bug is found in 5.10 or 5.15, what they will do is they will backport the bug fixes to this long time support kernel.
But in between the releases write 5.04 to 5.10506. These are not long term support which means the bug fixes are not backported to it. So, only in the LTS support bug fixes found in the recently found bugs are backported to it. So, how to find the kernel version number. So, if you are running your distributions like Ubuntu, how to find it. So, there is a user space command called union. So, using this union we can be able to find the kernel version number. So, this is I have used SSH for my Linux laptop. This Linux laptop runs on Ubuntu OS. So, I just get this terminal.
We can just type uname. So, uname underscore R. So, uname underscore R shows what version of kernel I am just using. So, which means now I am using version 5.0.15.0. If I use command uname underscore R. So, it prints. So, find out the kernel version number you are running. So, for example, if you are using Ubuntu OS and you want to find out which kernel version number laptop is running. Then there is a utility called uname. Then there is a utility called uname which is an application in Linux using which we can able to find out which kernel version number you are trying to load. So, this is a command prompt. So, this is using SSH.
So, in this we can open your terminal and then you can type uname underscore R. So, it will tell which version of the kernel your Ubuntu is running. So, in my case it is a version 5.15.0. So, similarly you can run the command of US. Next is a kernel development community. So, kernel development for Linux kernel. So, we have a community which has an active volunteers all around the world. As of now, more than 2,000 people are working on it. And we can become a part of a kernel. So, when you develop a device driver, then eventually we become a part of a kernel community.
And there is a main forum which is there. It is an LKML which means Linux kernel mailing list. So, we can subscribe to it. So, once we subscribe to it. So, what are the things currently working? So, Linux kernel development we can be able to see it. So, we can check it and what actually development is going on. So, in this community we can find not only developers, we can also have reviewers. So, only if our code is reviewed, then the source code will be accepted in the mainline kernel. And we can have testers and we can have questions.
Any doubts during the development, we can actively engage. So, this is the end of our first session. So, in this first session, we have seen history of a Linux kernel and some features of it and how it is. Thank you very much. architecture and how what is the Linux version number and distribution version number and what is the kernel development community. This is the end of our question.