In Chapter 3, Kernel Facilities and Helper Functions, we introduced peripheral IRQs, using request_irq() and request_threaded_irq(). With request_irq(), one registers a handler (top half) that will be executed in atomic context, from which one can schedule a bottom half using one of a differing mechanism discussed in that same chapter. On the other hand, with request_thread_irq(), one can provide top and bottom halves to the function, so that the former will be run as a hardirq handler, which may decide to raise the second and threaded handler, which will be run in a kernel thread.
The problem with those approaches is that sometimes, drivers requesting an IRQ do not know about the nature of the interrupt that provides this IRQ line, especially when the interrupt controller is a discrete chip (typically a GPIO expander connected over SPI or I2C...
