microcodes

Microcode is a layer of low-level instructions that translates the machine code instructions of a computer's instruction set architecture (ISA) into a sequence of micro-operations. These micro-operations are the fundamental operations that a processor's control unit can perform, such as fetching data from a register, performing an arithmetic operation, or writing data to memory. Essentially, microcode acts as an intermediary between the high-level instructions that software understands and the very basic signals that the processor's hardware needs to execute them.

Microcode can be stored in a special type of memory within the processor, often called control store. When the processor encounters a machine code instruction, it looks up the corresponding microcode sequence in the control store. This sequence then dictates the actions of the processor's components, step by step, to complete the instruction. This approach offers flexibility, allowing complex or infrequently used instructions to be implemented efficiently without requiring dedicated hardware for every possible instruction. It also enables the ISA to be updated or modified without necessarily redesigning the entire processor hardware, as the microcode can be updated.

The concept of microcode was pioneered by Maurice Wilkes in 1951. Modern processors still utilize microcode, although its implementation and complexity have evolved significantly. While some instructions might be hardwired directly into the processor's logic for maximum speed, others are still handled by microcode, especially for complex operations or for maintaining backward compatibility with older architectures. Firmware is a related concept, often referring to software embedded in hardware devices, and microcode can be considered a form of firmware.