Recently, I’ve passed a course on Coursera, which is called nand2tetris. The course itself is about building your own computer from the ground up. Though it will be better to quote the instructor himself, what the course is about:

The course guides students and self-learners through the construction of a modern, full-scale computer system – hardware and software – from the ground up. In the process, the students practice many major computer science (CS) abstractions studied in typical CS courses and make them concrete through 12 guided implementation projects.

And so, when I implemented the last module/project from the course, I got a Pong game, written in Jack language (your own language, implemented during the course).

Afterwards, the game is compiled through your own toolchain back into binary code and running on your own CPU. Both toolchain and CPU developed during the course.

Pong Game

If you are interested in the crash course for “How does computers work”, read the article.

Read Full Article

Have you ever been interested in how your code, written in high-level language like TypeScript, transforms into a different representation that a computer can understand, a CPU can understand?

I hope, when you will read this article, the picture of code execution pipeline will be more explicit. We will trace the journey of the code, starting from high-level language to low-level machine instructions. We are going to a deep rabbit hole…

DISCLAIMER: I will not dive into technical details of implementation, which differs from one vendor to another. We will go through a conceptual overview only. Otherwise, the article would take hours to read and months to write.

Read Full Article

Last time, we created a working boot sector, the BIOS can find with the help of magic numbers. You can read more about it here (if you didn’t, I highly recommend doing that now, as you may miss some important details).

The question here is “Why do we need a second stage boot loader?”. We can implement all of it in the boot sector, using Assembly, so… why?

The problem is… size limits. You can’t store over 512 bytes of code in the boot sector, so if you want to make a super-duper boot loader (like GRUB or similar) you need to store all of it somewhere else, but not in the boot sector itself.

And that is one of reasons, we need to have a second stage boot loader.

Read Full Article