Intel Corporation

Aaron R Kunze

Erik J Johnson

David M Putzolu

Christian Dreke

James L Jason, Jr

Jim W Edwards

Teresa M Morrison

Harrick M Vin

Ravi L Sahita

Mark B Rosenbluth

Andrew Brownsword

James W Edwards

Hermann Gartler

Todd A Anderson

Frederick William Strahm

Xiao-Feng Li

Dz-ching (Roy) Ju

Walter K Hazzard

Dillon Sharlet

Erik Johnson

According to some embodiments, the workgroup divisibility requirement may be dispensed with on a selective or permanent basis, i.e. in all cases, particular cases or at particular times and/or under particular conditions. An application programming interface implementation may be allowed to launch workgroups with non-uniform local sizes. Two different local sizes may be used in a case of a one-dimensional workload.

Range selection for data parallel programming environments

Methods and apparatuses to reduce memory bandwidth consumption when executing a program that uses integral images are provided. A first integral image representation of a grayscale image may be computed using a first number of bits per pixel. A second integral image representation of the grayscale image may be computed using a second number of bits per pixel. Additionally, an image processing algorithm may be executed on the grayscale image, wherein the size of a rectangle to be used by the image processing algorithm can be determined. Based on the size of the rectangle to be used by the image processing algorithm, it may be determined whether to use the first integral image representation or second integral image representation for the image processing algorithm.

Reducing memory bandwidth consumption when executing a program that uses integral images

In some embodiments, the invention involves a framework for using virtualization technology to efficiently support a domain-specific run-time environment. In at least one embodiment, a framework is utilized to take advantage of virtualization technology (VT) to partition performance critical and non-performance critical tasks of the same domain-specific application. An embodiment of the invention utilizes a general-purpose operating system to execute non-performance critical aspects of a domain, and uses a privileged VT-root mode to execute performance critical aspects of the domain. Another embodiment uses one or more guest VMs to execute the performance critical aspects of the domain-specific run-time environment. Other embodiments are described and claimed.

Framework for domain-specific run-time environment acceleration using virtualization technology

In one embodiment, the present invention includes a method for identifying a deformable object of a scene of a computer game that is visible by an artificial intelligence (AI) character of the game, requesting a speculative physics simulation associated with the deformable object to determine a result of an action to the deformable object by the AI character, and selecting an action to be performed by the AI character, where the selection is based at least in part on the speculative physics simulation. Other embodiments are described and claimed.

Combining speculative physics modeling with goal-based artificial intelligence

In some embodiments, the invention involves a novel combination of techniques for prefetching data and passing messages between and among cores in a multi-processor/multi-core platform. In an embodiment, a receiving core has a message queue and a message prefetcher. Incoming messages are simultaneously written to the message queue and the message prefetcher. The prefetcher speculatively fetches data referenced in the received message so that the data is available when the message is executed in the execution pipeline, or shortly thereafter. Other embodiments are described and claimed.

Method and apparatus for speculative prefetching in a multi-processor/multi-core message-passing machine

A connection is established between a node on a first network and at least one node on a second network lacking a valid address on the first network by receiving at a gateway a connection request from the node on the first network, sending information related to the connection request to a plurality of nodes on the second network, receiving responses to the sent information from at least one of the nodes on the second network, and establishing at least one connection between at least one of the nodes on the second network from which responses are received, and the node on the first network, according to the order of the received responses or according to priority weighting information or according to priority weighting information concerning the nodes from which the responses are received.

Connections of nodes on different networks

In general, in one aspect, the disclosure describes a method of determining if a first query for data related to a protocol data unit in a first table is a query to a table merged into a combination table formed from multiple tables. If so, the method can generate a second query for the first query for data stored by the combination table.

Merger of tables storing protocol data unit related data

Techniques that may be utilized in a multiprocessor system are described. In one embodiment, one or more signals are generated to indicate that a breakpoint instruction is executed by one of the plurality of processors in the multiprocessor system. For example, a signal may be generated to indicate whether a processor is to be halted once it receives the a signal that indicates the breakpoint instruction. Other embodiments are also disclosed.

Multiprocessor breakpoint

Embodiments of the present invention relate to a method and system for automatically configuring network processing software to reduce memory latency associated with parallel processing using a plurality of processing elements.

Method and system for configuring network processing software to exploit packet flow data locality

A system and method of transmitting network packets between network processing elements through links are disclosed. One or more configuration entities may allocate one or more fixed length slots to be appended to network packets forwarded on a link between network processing elements in-band of the link.

System and method of annotating network packets

A method for employing a driver in a router to communicate between the router's operating system and network processor. The driver appears to the operating system as a device driver for a conventional network interface. The driver also registers with the network processor and appears itself as a forwarding interface. Control information received the network processor is forwarded to the operating system by way of the driver using the network processor's existing forwarding mechanisms, and control information from the operating system is sent to the network processor by way of the driver.

Network routing using a driver that is registered with both operating system and network processor

In one embodiment, a method is provided. The method of this embodiment includes generating, by a processor that includes a plurality of processing engines capable of executing program instructions, a packet. The method of this embodiment also includes transmitting the packet to at least one of the processing engines. Additionally, the method of this embodiment also includes, in response, at least in part to receipt of the packet by the at least one of the processing engines, modifying at least in part, by the at least one of the processing engines, a set of program instructions that the at least one processing engine is capable of executing. Of course, many modifications, variations, and alternatives are possible without departing from this embodiment.

Passing a received packet for modifying pipelining processing engines' routine instructions

Processing Internet Protocol security (IPsec) traffic includes determining at a first location if a classification parameter is available for the IPsec traffic that indicates a route for the IPsec traffic and forwarding the IPsec traffic based on the classification parameter. If a classification parameter is not available, processing IPsec traffic includes decrypting the IPsec traffic at a second location if the IPsec traffic is encrypted and determining the classification parameter for the IPsec traffic at the second location.

Growing community of inventors

Portland, OR, United States of America

Aaron R Kunze