Why Does a Computer Freeze? Understanding CPU, Interrupts, and Kernel Deadlocks

Computer freezes are a common experience, but what actually happens inside the CPU when a system becomes unresponsive?

The piece explains that a true hardware CPU stall is rare; most freezes are caused by software where the CPU is trapped in a loop or waiting for resources.

It demonstrates a simple infinite loop in C and shows that modern operating systems use pre‑emptive multitasking and interrupts to regain control, so the loop alone does not freeze the system.

void dead_loop()
{
  while (1) {
    ...
  }
}

The key concept introduced is the interrupt , which allows the OS to temporarily suspend the current execution, run a higher‑priority handler, and later resume the original task.

Interrupt

Interrupts are one of the most important inventions in computer history; they let the CPU stop its current work to handle external events such as the timer interrupt that enables the scheduler.

Because the OS installs interrupt‑handling routines with high priority, a thread stuck in an infinite loop will be pre‑empted once its time slice expires, allowing other processes to run.

However, if an interrupt handler itself enters a deadlock or an infinite spin‑lock, the CPU core can become unresponsive, effectively causing a freeze.

Two scenarios are described:

1) An interrupt that cannot be pre‑empted due to higher priority, meaning low‑priority interrupts cannot break the high‑priority handler.

2) An interrupt that acquires a lock and deadlocks because no other thread can be scheduled, similar to a kernel‑level deadlock.

In multi‑core systems, other cores may continue to run, but a deadlock in the kernel can still halt the whole system.

The article concludes that a simple infinite loop will not crash a modern OS, but bugs in kernel interrupt handling or priority inversion can lead to the dreaded “frozen” computer.