A floating point unit (FPU) is a processor block that performs floating point arithmetic, often accessed through the same addressing arrangements as the rest of the processor. Many processors do not include an FPU, but most processors and software systems implement the IEEE standard for floating point operations. FPUs can perform many operations, including trigonometric functions, and are used in graphics processors, embedded controllers, and encryption processors.
A floating point unit is a functional block of a computer processor that performs floating point arithmetic. Also known by the acronym FPU, a floating point unit can also be a physically separate coprocessor. An FPU usually accesses data through the same addressing arrangements as the rest of the processor and usually includes its own registers. A processor uses an FPU to calculate trigonometric, logarithmic, and basic floating point arithmetic. It also uses an FPU to load, store and compare numbers in different integer and floating point data types.
Many processors do not include a floating point unit. Most microprocessors manufactured by Intel Corporation and Motorola Incorporated before the late 1980s don’t include one, for example. Many Reduced Instruction Set Computing (RISC) processors and integrated controllers also don’t have one. These chips include an arithmetic logic unit (ALU) to perform basic integer arithmetic and bitwise logic functions. However, floating point math is handled by an external coprocessor or software library.
Most processors and software systems implement the Institute of Electrical and Electronics Engineers (IEEE) standard for floating point operations. The IEEE-754 standard was first released in 1985 and expanded in 2008 to include additional data types. Defines several types of integer and floating point data, each with its own precision, representation, and numeric range. It also defines many arithmetic operations, rounding methods, and exception handling such as division by zero. Whether calculations are performed from a floating point unit or not, the IEEE floating point standard guarantees identical results.
In addition to addition, subtraction, multiplication and division, a floating point unit can perform many other operations. These often include scaling, square root and logarithmic calculations. Usually sine, cosine, and partial tangent trigonometric functions are also provided. Integer and floating point numbers can be compared with the FPU and converted from one precision to another. They can also be rounded in both directions and stored directly in the processor’s main memory.
The FPU of a graphics processor may include oft-repeating functions common to rotating or animating three-dimensional objects. Many embedded controllers are designed with processors optimized for digital signal processing (DSP). These can be used in audio and communications applications and can contain FPUs, which are useful for fast Fourier transforms (FFTs), for example. Encryption processors often include an FPU for high-speed computation of heavy encryption algorithms. Generic multi-core processors may also include a floating point unit for each regular processor core.
Protect your devices with Threat Protection by NordVPN
