Advanced Micro Devices Corporation

Hewlett-packard Company

Josef A Dvorak

Ronny Lee Arnold

Kent R Townley

Douglas R Williams

Edward Chang

Paul L Rogers

David B Hollenbeck

Mahadev S Deshpande

Ricky L Pettit

A prefetch device and method are disclosed that determines from which addresses to speculatively fetch data based on information collected regarding previous cache-miss addresses. A historical record showing a propensity to experience cache-misses at a particular address-offset from a prior cache-miss address within a region of memory provides an indication that data needed by future instructions has an increased likelihood to be located at a similar offset from a current cache-miss address. The prefetch device disclosed herein maintains a record of the relationship between a cache-miss address and subsequent cache-miss addresses for the most recent sixty-four unique data manipulation instructions that resulted in a cache-miss. The record includes a weighted confidence value indicative of how many cache-misses previously occurred at each of a selection of offsets from a particular cache-miss address.

Region prefetcher and methods thereof

A multiplexer (MUX) circuit with balanced select line loading is provided. The MUX circuit includes a plurality of 2:1 MUX units coupled together in a multistage cascading arrangement, along with a selection module coupled to the MUX units. The MUX units are arranged in an initial MUX stage, at least one intermediate MUX stage coupled to and following the initial MUX stage, and a final MUX stage coupled to and following the at least one intermediate MUX stage. Each MUX unit is controlled with a respective select bit input value provided by the selection module. The selection module controls the operation of the MUX units in the initial MUX stage with a first plurality of different select bits, controls the operation of the MUX units in the at least one intermediate MUX stage with a second plurality of different select bits, and controls the operation of the final MUX stage with a devoted select bit.

Multiplexer circuit with load balanced fanout characteristics

An apparatus for producing one or more clock signals comprises a plurality of delay elements sequentially connected and logic circuitry connected to several of the plurality of delay elements. A clock signal fed through the plurality of delay elements produces multiple delayed versions of the clock signal. Logic circuitry selects and combines the delayed clock signal versions to produce one or more output clock signals, each having a frequency that is a selected fraction of the input clock signal. An associated method delays the input clock signal N times sequentially for a natural number N. then selects a series of time splices of the delayed clock signals to produce an output clock signal. In some implementations the input clock signal can be referenced to a reference clock signal. The output clock signal frequency can be set to (N/M)&times;f , for a natural number M and reference clock signal frequency f . The apparatus and associated method can flexibly produce a large variety of output clock frequencies and frequency ratios, lock to f with a dynamic response independent of the output frequency range and can be optimized to a single reference frequency, need not relock to change output frequency, and reduce clock skew.

Input clock delayed by a plurality of elements that are connected to logic circuitry to produce a clock frequency having a rational multiple less than one

Growing community of inventors

Fort Collins, CO, United States of America

Josef A Dvorak