Operating system for smart home in C++ from scratch. How?

Question

Operating Systems

In short, I and 5 other guys got together and decided to make a COMMERCIAL smart home system.
Taking something as a basis for such a project is not solid, and therefore it was decided to make our own OS. But immediately faced with questions:
- How?
- What to learn?
- In what direction to work? (Design --- Functionality)
Development will be carried out on ARM processors (Raspberry PI or Banana PI or CubieBoard have not yet been decided exactly)
I ask knowledgeable people to suggest what literature can be read on writing an OS.
Please note that the development will be carried out from scratch.
File system, kernel, etc. we will do it ourselves. Ready for difficulties.
If clarifications are needed, I will write directly on demand, I am on a similar resource for the 1st time.

Answer 1

And why arm and raspberries? Well it's undignified. Make your microcontrollers! Mine silicon, develop a scheme. And there, both your ASM and the axle will tighten up.
Here is your question from about the same category, if that.

Answer 2

In my opinion, asking a similar question is not even worth starting ... It's just that a "smart home" is a zoo of technologies, with a bunch of nuances.
And with this approach, you should think about changing your profession. Seriously.
Another nonsense ... Although the legs are from there. You don't need an OS, write software for the finished one. In your case under Linux.
0) Determine the budget for the final product (output price). You will often come back here and here you will find quite a few unpleasant surprises.
1) At least be defined with iron. Raspberries and analogues are certainly cool, but what will be the performance equipment ? The choice is very large and not obvious. And this is just the main question.
2)

- How?

Find the answer to this question. And here someone is unlikely to help you, because it is not clear what you want to get at the output.
3)

- What to learn?

Well, if you are going to write on these, then teach them. Learn a lot. LOT! But I would choose a simpler language like PHP, Node JS, Go. In this case, MUCH will be slightly less.
4) Since there are a lot of you, try to break up in different directions.
5)

- In what direction to work? (Design --- Functionality)

In all, but according to the classics, they first saw the functionality, and then they pull the picture.

Answer 3

COMMERCIAL smart home system.
Taking something as a basis for such a project is not solid, and therefore it was decided to make our own OS

You don’t have to read further ... But if, in the order of delirium, we assume that this is not fat, fat trolling, then you need to start with a business plan . And not at all with the choice of hardware, software, PL, etc. You, hee hee, are going to saw a COMMERCIAL system: ROFL: Therefore, before doing another under-linux, sit down and think:
- who is the consumer
- what are the costs
- what are the expected sales
- what kind of licensing scheme
...
and if after all the questions don't barf from all this boring non-technical tedious nonsense - go for it. Will you also develop interaction protocols yourself? So what? Existing protocols are undignified...
:ROFL:

Answer 4

1. Take a programming language (at least Javascript compiled by V8), write PROFIT.
   I will not attach a picture with a trolley bus yet, but the question is "Why, when there is a Linux kernel that is under GPL2?" I'll still ask.

Answer 5

spoiler

Вы что фейки регаете или какое-то весеннее обострение? Такой же вопрос был N дней назад

Then Vanga came and said that nothing would work out for you. And on a subject, throw this silly idea. Why are you not satisfied with the zoo of ready-made solutions based on linux and unix? Make the next Linux distribution, only stuffed with software for interacting with microcontrollers, sensors, cameras and everything smart in your home. Those. the core of a smart home, and then expand it modularly. You can make a distributed fault-tolerant system, in general, there is a lot of room for imagination and you don’t need to write your own OS at all.

Answer 6

Answer 7

For any idea, first of all, you need a concept, as well as an understanding of what you expect to get at the end. I can say right away that you will need to understand which of you will be engaged in which direction.
I think you should look at how the simplest device (system) is being developed, usually development includes:
- creation of technical specifications
- creation of a draft design
- development of E1 (creation of a block diagram of a device (system))
- development of E2 (creation of a functional diagram of a device (system))
- development of a circuit diagram and programs
- preliminary tests (a prototype device (system) is used)
- final
For development, it will be necessary to study the following main areas:
- Working with embedded solutions (MK, peripherals, equipment, etc.)
a. development of E1, E2, selection of components and creation of E3, selection of interaction protocols
b. creating a prototype on development boards
c. development of firmware for MK
d. testing
- Working with the server and client
a. development of structural and functional schemes
b. server and client development (Using a ready-made solution / using a ready-made solution, but with changes taking into account the specifics of your project / creating your own solution)
c. testing the interaction between the server and the client
- Networking
- Integration of embedded solutions into information systems (here, mainly the choice of protocol)
Another important point is the final interface and the stability of the system. The end client should get a beautiful and simple management interface.

Answer 8

I can see your company of clear boys. I believe in your success, good luck, now there is a shortage of operating systems on a commercial basis all over the world. I advise you to start with the development of your microcontrollers and assembler for them, and from there move smoothly to writing your own OS. For more details, you can consult Denis Popov, a well-known developer of operating systems from scratch.

spoiler

P. S.
Сначала вам нужно добыть руду в шахте для производства своего железа под свою ос.
Могу помочь с выбором оптимальной кирки, посоветовать лопату.