Intel Corporation

Ching-Hua Chu

James Nadir

Alper Ilkbahar

David N Harris

Sunny C Huang

Jason C Stinson

An arbitration circuit and method for a multiport high speed memory in a computer microprocessor. A plurality of addresses are provided to a plurality of ports. The addresses are decoded in a plurality of decoders. The decoded output lines are compared in a comparison circuitry to determine if one or more of the ports is requesting access to the same memory line, and a comparison bit indicative of a match is outputted. If asserted, the comparison bit disables a line driver so that only one of the wordlines in a particular memory line is driven at any one time.

Multiport high speed memory having contention arbitration capability

An opportunistic time-borrowing domino logic includes a domino pipeline having a plurality of logic gates coupled in series and controlled by first, second, third and fourth clock signals. The first domino gate in a half-cycle is clocked by either the first or the second clock signals, wherein the last domino gate in a half-cycle is clocked by either the third or the fourth clock cycles. The second clock signal is an inverse of the first clock signal, and the third and fourth clock signals have local delayed clock phases in which the falling edges of the third and fourth clock signals are delayed relative to the falling edges of the respective first and second clock signals. In a first half-cycle, a first type of domino gate is controlled by the first clock signal, with subsequent domino gates of the same type being controlled by the third clock signal. Odd-numbered half-cycles begin with a domino gate of the second type controlled by the second clock signal, followed by domino gates of the first type controlled by the fourth clock signal.

Opportunistic time-borrowing domino logic

A cache circuit for a computer microprocessor and a method for performing cache operations (e.g., read and write) in a single, short cycle using overlapped clocking. The cache includes a tag array, a status array, and a data array. Parity information is generated and checked to verify data and tag integrity. The parity field is stored in a status array physically separate from the tag array. The status array is offset in timing so that it lags behind the tag array for both read and write operations. Therefore, fields in the status array can be written in the early part of the next clock cycle without affecting the tag array or another operation that may be scheduled for the next time cycle.

Method and apparatus for overlapped timing of cache operations including

A set select circuit and method for selecting a set in a set associative cache in a microprocessor. The set select circuit, responsive to a main clock, includes an input latch coupled to receive select data before the main clock cycle. The input latch is transparent to set select data so that predecoding can begin before the main clock. The input latch latches the set select data on the initial clock edge and holds the set select data during the first half of the main clock cycle. A pre-decoder is coupled to the input latch for receiving and predecoding the set select data, and a decoder is coupled to the predecoder for receiving and decoding the pre-decoded set select data to supply an output to an output latch. The output latch is also coupled to a clock inverter to receive the inverted delayed clock signal. The output latch is transparent during the second half of an inverted delayed clock cycle. The output latch latches the selected set on the initial inverted delayed clock edge and holds the selected set during the first half of the inverted delayed clock cycle.

Growing community of inventors

San Jose, CA, United States of America

Ching-Hua Chu