A register is a small, fast storage location within a processor that temporarily holds data or instructions during processing. In computing, registers enable rapid data manipulation and reduce the need for main memory access thereby enhancing computer performance. In legal contexts, a register documents important legal information such as property ownership or case details. This information is often stored at the local government level where it may take some time and government assistance to view specific records.
The term register is also used in a number of other applications including accounting and computer hardware. The most common use of the word is in the context of a formal record that documents a transaction, such as a sales invoice or purchase receipt. The record can then be accessed later for reference. The word is also used to describe a set of values or items that are listed together in a particular manner, such as a list of customers or an inventory item listing.
In computing, the word register is also used to refer to a small, fast internal storage space in a central processing unit (CPU) that holds data or instructions during processing. Registers are an element of the CPU memory hierarchy that includes cache main memory and secondary storage with varying capacities and access speeds.
Registers are used extensively in arithmetic and logic operations to temporarily hold operands and intermediate results. They are also used to enable CPU pipelining, where a single instruction can be executed on multiple bits of data thereby increasing computer speed and efficiency.
In addition to the basic storage capabilities of a register, most CPUs include a variety of additional features to enhance its functionality and versatility. For example, many registers contain flags that can be read or cleared and special function registers corresponding to specialized hardware elements.
Typically, the register in a CPU is divided into integer, floating point, stack, and general purpose registers. Each register can be manipulated by various types of instructions and most have different formats or interpretations for the data they store. For instance, an integer register can be manipulated as a half-word or 18-bit address, while a floating point register is usually interpreted as a single precision floating-point value.
The number of registers in a CPU can vary greatly, depending on the architecture and other factors. In the original PDP-10 processor, for example, 16 general purpose registers (GPR) mapped to main memory locations 0-15. Modern CPUs may have fewer or more registers depending on the architecture, with some systems combining GPRs and other memory locations into special purpose registers.