Definition - What does Floating-Point Unit (FPU) mean?
A floating point unit is an integrated circuit which handles all mathematical operations that have anything to do with floating point numbers or fractions. It is a dedicated logic unit specifically designed to work on floating point numbers and nothing else, hence the name. It can be defined as a specialized coprocessor that can manipulate numbers quicker than the basic microprocessor circuitry itself.
The FPU performs simple mathematical tasks which include addition, subtraction, division, multiplication and square root. Older FPUs process transcendental functions like exponential and trigonometric calculations but these can be expensive and complicated to implement, so in modern FPUs, these are done via software library routines.
Not all computer systems have hardware FPU. Those that do not have FPU can emulate its functions in multiple ways:
In an operating system as inherent functions.
It can be emulated in the CPU as a microcode or microprogram.
Or in user code; typically this is what is called as software emulation.
The FPU is a coprocessor that is specifically designed to process mathematical operations on floating point numbers, meaning it is not just a small part of a computer but rather an integral part of the system. It is usually implemented via hardware but not all computer systems can have an FPU due to several reasons including space, power, or price.
An FPU is more complicated to implement in hardware, but luckily it can be emulated by the CPU. In a sense it can be part of the CPU since it can be implemented as a microprogram within the CPU. In such case, the CPU now does all the functions of an FPU but this is less efficient than a dedicated FPU since this would take up valuable CPU time that could have been used to process other things. Emulation can also be done by the OS or via software.