Learning how not to write SPI device drivers
The usual way to deal with SPI devices is to write kernel code to drive this device. Nowadays the spidev
interface makes it possible to deal with such devices without even writing a line of kernel code. The use of this interface should be limited, however, to simple use cases such as talking to a slave microcontroller or for prototyping. Using this interface, you will not be able to deal with various interrupts (IRQs) the device may support nor leverage other kernel frameworks.
The spidev
interface exposes a character device node in the form /dev/spidevX.Y
where X
represents the bus our device sits on, and Y
represents the CS index (relative to the controller) assigned to the device node in the device tree. For example, /dev/spidev1.0
means device 0
on SPI bus 1
. The same applies to the sysfs directory entry, which would be in the form /sys/class/spidev/spidevX.Y
.
Prior to the character device appearing in the user space, the device...