Advanced Micro Devices Corporation

Siemens Aktiengesellschaft

Ole H Moller

Rod G Fleck

William M Johnson

Cheng-Gang Kong

Gigy Baror

Brian W Case

Philip M Freidin

Timothy A Olson

Smeeta Gupta

David I Sorensen

A data processor is disclosed which comprises a first pipeline for decoding and executing data instructions, a second pipeline for decoding and executing address instructions, a unit for issuing multiple instructions to the pipelines, a first set of registers being coupled with the first pipeline, and a second set of registers being coupled with the second pipeline, wherein first and second pipeline process data in parallel.

Execution of a loop instructing in a loop pipeline after detection of a

A streamlined instruction processor processes data in response to a program composed of prespecified instructions in pipeline cycles. The processor comprises an instruction fetch unit, including an instruction interface adapted for connection to an instruction memory and for fetching instructions from the instruction memory. The instruction fetch unit includes an instruction prefetch buffer coupled to the instruction interface for buffering a sequence of instructions supplied to the instruction interface. A branch target cache is coupled with the prefetch buffer for storing sets of instructions retrieved from a corresponding set of locations in the instruction memory, having sequential instruction addresses. The first instruction in each such set is a branch target instruction in the program. In addition, an execution unit including a data interface adapted for connection to the data memory, executes the instructions in pipeline cycles. The execution unit includes a storage facility, coupled to the data interface, for storing data in a file of data locations identified by file addresses. The storage facility includes at least two read ports and one write port operable in response to file addresses. An addressing unit coupled to receive the instructions from the instruction register, supplies the file addresses to the read ports and the write port under program control. In addition, the addressing unit is operable in response to a stack pointer providing dynamic allocation of the file of data locations to processes within the program. A memory management unit is coupled to the data interface. The memory management unit includes an address interface adapted for connection to the data memory and the instruction memory for supplying instruction addresses to the instruction memory and data addresses to the data memory, in a simple single access mode, a pipeline mode and a burst mode.

Streamlined instruction processor

A high speed register file for use by an instruction processor suitable for reduced instruction-set computers (RISCs) is disclosed which is preferably used with an efficient register allocation method. The register file facilitates the passing of parameters between procedures by dynamically providing overlapping registers which are accessible to both procedures. Each procedure also has a set of 'local' registers assigned to it which are inaccessible from other procedures. The register file is divided into a number of blocks and a protection register stores a word which proscribes access by a particular procedure or task to certain blocks. In this manner, an instruction processor using the register file can operate on multiple tasks maintaining the integrity of each from undesired changes occuring in the others.

General-purpose register file optimized for intraprocedural register

A simple architecture to implement a mechanism for performing data references to storage in parallel with instruction execution. The architecture is particularly suited to reduced instruction-set computers (RISCs) and employs a channel address register to store the main memory load or store address, a channel data register which temporarily stores the data from a store operation and, a channel control register which contains control information including the number of the register loaded within the file, in the case of a load operation. This number is used to detect instruction dependency of the data to be loaded. Logic circuitry suspends further instruction processing if the data required from a load is not yet available. A data-in register is used to store load data until an instruction execution cycle is available for writing it back to the register file. Logic circuitry detects storage of data prior to its writing back, so as to effectively replace the register file location. During page faults, the contents of the channel address, channel data, and channel control registers are saved to permit page fault recovery.

Mechanism for performing data references to storage in parallel with

Growing community of inventors

Lyngby, Denmark

Ole H Moller