|
The Cambridge CAP computer was the first successful experimental computer that demonstrated the use of security capabilities, both in hardware and software.〔Levy, p.96〕 It was developed at the University of Cambridge Computer Laboratory in the 1970s. Unlike most research machines of the time, it was also a useful service machine.〔 The sign currently on the front of the machine reads:
==Design== The CAP was designed such that any access to a memory segment or hardware required that the current process held the necessary capabilities. The 32-bit processor featured microprogramming control, two 256-entry caches, a 32-entry write buffer and the capability unit itself, which had 64 registers for holding evaluated capabilities. Floating point operations were available using a single 72-bit accumulator. The instruction set featured over 200 instructions, including basic ALU and memory operations, to capability- and process-control instructions. Instead of the programmer-visible registers used in Chicago and Plessey System 250 designs, the CAP would load internal registers silently when a program defined a capability.〔Levy, p. 79〕 The memory was divided into segments of up to 64K 32-bit words. Each segment could contain data or capabilities, but not both. Hardware was accessed via an associated minicomputer. All procedures constituting the operating system were written in ALGOL 68C, although a number of other closely associated protected procedures - such as a paginator - are written in BCPL.〔Wilkes and Needham, p. 32〕 抄文引用元・出典: フリー百科事典『 ウィキペディア（Wikipedia）』 ■ウィキペディアで「CAP computer」の詳細全文を読む スポンサード リンク
|