How Linux Belongs to Everything: Key Takeaways from Greg Kroah-Hartman Interview

Linux is the world's most widely used operating system, powering billions of Android devices, cloud servers, smart TVs, 5G modems, financial markets, and even the International Space Station. In a detailed conversation on How Linux is built with Greg Kroah-Hartman—a Linux kernel maintainer for over 13 years and a Linux Foundation fellow—shared technical and organizational insights into how Linux is built.

---

1. Codebase Scale: Why Your Phone Code is 3x Larger Than a Server

The Linux kernel repository contains nearly 40 million lines of code, though the core kernel accounts for only about 5% of that total. The remaining 95% consists of hardware support, architecture code, and device drivers.

Surprisingly, a standard enterprise server requires only about 1.5 million lines of code because server hardware (CPU, network interface, storage) is relatively straightforward. In contrast, a modern smartphone runs approximately 4 million lines of code—nearly three times as much—due to complex System-on-Chip (SoC) architectures, power management, audio routing, and multiple internal buses.

---

2. The 9-Week Pulse: A Time-Based Release Model

Linux operates on a strict, time-based release cadence of every 9 weeks.

• The 2-Week Merge Window: When a new release drops, a 2-week window opens for subsystem maintainers to submit pre-tested features to Linus Torvalds.
• 7 Weeks of Stabilization: Following Release Candidate 1 (RC1), the next 7 weeks are strictly reserved for bug fixes, regression testing, and stabilization—no new features are allowed.

This predictable schedule removes pressure from maintainers, as unready features simply wait for the next 9-week window rather than being rushed into production.

---

3. Scaling to 4,000 Contributors Without Project Managers

Despite recording around 4,000 active contributors per year, the Linux project operates without product or project managers. Instead, the workflow relies on a pyramid hierarchy of roughly 800 maintainers who oversee specific subsystems. The core organizational mechanism is human trust: maintainers accept patches because they trust that contributors will remain accountable to fix bugs if issues arise later.

---

4. The Golden Rule of Kernel Development

The primary responsibility of the kernel is to manage hardware resources and provide a uniform, hardware-agnostic interface to user-space applications. Because of this responsibility, kernel developers follow one strict cardinal rule: "Don't break user space on purpose". Users must always be able to upgrade their kernel without worrying that existing applications will crash.

---

5. Embracing Rust for Memory Safety

While C remains the foundational language of the kernel, Rust is actively being integrated, with over 25,000 lines of Rust code already in the repository.

Rust brings strong memory safety rules and structured object lifecycle management, making it easier to write safe drivers and eliminate common buffer overflow or lock-management errors.

---

Conclusion

Linux's ubiquity proves that open collaboration, transparent code reviews, and strong trust networks can produce highly reliable software at an unprecedented global scale.

Thanks for his attitudes, Greg Linkedin ID