Capability-based addressing

  • Authors:
  • Edward F. Gehringer

  • Affiliations:
  • -

  • Venue:
  • Encyclopedia of Computer Science
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

An operating system must manage access rights, which protect the data of one process from interference by other processes, while also permitting appropriate sharing of data. Capability-based addressing achieves this by using protected pointers (q.v.), known as capabilities, to map from the operand field of an instruction to an operand in memory. A capability contains the unique identifier of an object, together with a set of access rights that it possesses to that object (Fig. 1). For example, a capability for an array might contain the array's unique identifier, and a single access right allowing the holder of the capability to read (but not write) the array. Or a capability might identify an employee record within a personnel system, and grant the right to inspect the employee's job title, department, and work location, but not the employee's salary.