Intel Corporation

Herbert H J Hum

James R Goodman

Aaron T Spink

Stephan Jourdan

Robert H Beers

Robert G Blankenship

Per H Hammarlund

Ali-Reza Adl-Tabatabai

Robert J Greiner

Chris J Newburn

Ram Huggahalli

Anwar Ghuloum

Deborrah Marr

Rajnish Ghughal

David I Poisner

Andrew V Anderson

Kenneth C Creta

David Lawrence Hill

Raymond Scott Tetrick

Frank Binns

Keshavan K Tiruvallur

Buderya S Acharya

In one embodiment, the present invention includes a method for receiving a first portion of a first packet at a first agent and determining whether the first portion is an interleaved portion based on a value of an interleave indicator. The interleave indicator may be sent as part of the first portion. In such manner, interleaved packets may be sent within transmission of another packet, such as a lengthy data packet, providing improved processing capabilities. Other embodiments are described and claimed.

Interleaving data packets in a packet-based communication system

A conflict resolution technique provides consistency such that all conflicts can be detected by at least one of the conflicting requestors if each node monitors all requests after that node has made its own request. If a line is in the Exclusive, Modified or Forward state, conflicts are resolved at the node holding the unique copy. The winner of the conflict resolution, and possibly the losers, report the conflict to the home node, which pairs conflict reports and issues forwarding instructions to assure that all requesting nodes eventually receive the requested data. If a requested cache line is either uncached or present only in the Shared state, the home node provides a copy of the cache node and resolves conflicts. In one embodiment, a blackout period after all responses until an acknowledgement message has been received allows all conflicting nodes to be aware of conflicts in which they are involved.

Speculative distributed conflict resolution for a cache coherency protocol

Systems and methods of managing transactions provide for receiving a first flush command at a first I/O hub, wherein the first flush command is dedicated to non-posted transactions. One embodiment further provides for halting an inbound ordering queue of the first I/O hub with regard to non-posted transactions in response to the first flush command and flushing a non-posted transaction from an outgoing buffer of the first I/O hub to a second I/O hub while the inbound ordering queue is halted with regard to non-posted transactions.

Multi-node chipset lock flow with peer-to-peer non-posted I/O requests

In one embodiment of the present invention, a method includes identifying a transaction from a first processor to a second processor of a system with a transaction identifier. The transaction identifier may have a value that is less than or equal to a maximum number of outstanding transactions between the two processors. In such manner, a transaction field for the transaction identifier may be limited to n bits, where the maximum number of outstanding transactions is less than or equal to 2. In various embodiments, such a transaction identifier combined with a source identifier and a home node identifier may form a globally unique transaction identifier.

Globally unique transaction identifiers

Architectures and techniques that allow legacy pin functionality to be replaced with a 'virtual wire' that may communicate information that would otherwise be communicated by a wired interface. A message may be passed between a system controller and a processor that includes a virtual wire value and a virtual wire change indicator. The virtual wire value may include a signal corresponding to one or more pins that have been eliminated from the physical interface and the virtual wire change value may include an indication of whether the virtual wire value has changed. The combination of the virtual wire value and the virtual wire change indicator may allow multiple physical pins to be replaced by message values.

Virtualization of pin functionality in a point-to-point interface

In one embodiment, the present invention includes an apparatus that has multiple buffers, including a first buffer dedicated to a first virtual channel of a first virtual network and a second buffer shared among virtual channels of a second virtual network. The shared buffer may be implemented as a shared adaptive buffer, and the buffers can be controlled using different flow control schemes, such as on a packet and flit basis. Other embodiments are described and claimed.

Buffering data packets according to multiple flow control schemes

A memory controller is described that comprises a compression map cache. The compression map cache is to store information that identifies a cache line's worth of information that has been compressed with another cache line's worth of information. A processor and a memory controller integrated on a same semiconductor die is also described. The memory controller comprises a compression map cache. The compression map cache is to store information that identifies a cache line's worth of information that has been compressed with another cache line's worth of information.

Processor and memory controller capable of use in computing system that employs compressed cache lines' worth of information

Use of an import cache and/or an export directory with an agent within to respond to requests for data. The import cache stores data that has been imported through the agent. The export directory stores information related to data that has been exported through the agent. Because the import cache and the export directory only store data that has passed through the agent, not all data transferred within a system are tracked by a single import cache or export directory.

Hierarchical directories for cache coherency in a multiprocessor system

Non-speculative distributed conflict resolution for a cache coherency protocol

The cache coherency protocol described herein can be used to maintain a virtual model of a system, where the virtual model does not change as the system configuration changes. In general, the virtual model is based on the assumption that each node in the system can directly communicate with some number of other nodes in the system. In one embodiment, for each cache line, the address of the cache line is used to designate a node as the 'home' node and all other nodes as 'peer' nodes. The protocol specifies one set of messages for communication with the line's home node and another set of messages for communication with the line's peer nodes.

Hierarchical virtual model of a cache hierarchy in a multiprocessor system

Cache coherency rules for a multi-processor computing system that is capable of working with compressed cache lines' worth of information are described. A multi-processor computing system that is capable of working with compressed cache lines' worth of information is also described. The multi-processor computing system includes a plurality of hubs for communicating with various computing system components and for compressing/decompressing cache lines' worth of information. A processor that is capable of labeling cache lines' worth of information in accordance with the cache coherency rules is described. A processor that includes a hub as described above is also described.

Growing community of inventors

Portland, OR, United States of America

Herbert H J Hum