Intel Corporation

William H Clifford

Brent S Baxter

Prashant Sethi

Clifford D Hall

James E Akiyama

Randy B Osborne

Paul M Brown

A method and apparatus for matching parent processor address translations to media processors' address translations and providing concurrent memory access to a plurality of media processors through separate translation table information. In particular, a page directory for a given media application is copied to a media processor's page directory when the media application allocates memory that is to be shared by a media application running on the parent processor and media processors.

System and methods exchanging data between processors through concurrent shared memory

A method and apparatus for matching parent processor address translations to media processors'address translations and providing concurrent memory access to a plurality of media processors through separate translation table information. In particular, a page directory for a given media application is copied to a media processor's page directory when the media application allocates memory that is to be shared by a media application running on the parent processor and media processors.

Systems and methods exchanging data between processors through concurrent shared memory

An embodiment of the present invention is a technique to control memory access. An address pre-swizzle circuit conditions address bits provided by a processor according to access control signals. A data steering circuit connects to N sub-channels of memory to dynamically steer data for a memory access type including tiled and untiled memory accesses according to the access control signals, the conditioned address bits, and sub-channel identifiers associated with the N sub-channels. The tiled memory access includes horizontally and vertically tiled memory accesses. An address post-swizzle circuit generates sub-channel address bits to the N sub-channels using the conditioned address bits and according to the access control signals and the sub-channel identifiers.

Accessing memory using multi-tiling

According to one embodiment, a memory controller is disclosed. The memory controller includes assignment logic, a reorder table and a transaction assembler. The assignment logic receives a request to access a memory channel and assigns the request to access one of two or more independently addressable subchannels within the channel. The reorder table includes two or more table elements. Each table element includes a shared address component and an independent address component corresponding to each of the two or more independently addressable subchannels. The transaction assembler combines the shared and independent address components in a reorder table element and issue a single memory transaction.

Memory micro-tiling request reordering

According to one embodiment, a memory controller is disclosed. The memory controller includes assignment logic and a transaction assembler. The assignment logic receives a request to access a memory channel. The transaction assembler combines the request into one or more additional requests to access two or more independently addressable subchannels within the channel.

Memory micro-tiling

According to one embodiment, a memory controller is disclosed. The memory controller includes assignment logic and a transaction assembler. The assignment logic receives a request to access a memory channel and assigns the request to access one of two or more independently addressable subchannels within the channel. The transaction assembler combines the request with one or more additional requests to access the two or more independently addressable subchannels within the channel and facilitates a speculative return of data from a subchannel for which a subchannel request is not available.

Mechanism for assembling memory access requests while speculatively returning data

Media memory system and method for providing concurrent memory access to a plurality of processors through separate translation table information

In some embodiments, a method includes generating blurred copies of an object by applying multi-texturing to the object during one pass through a graphics processing pipeline. In some embodiments, a graphics pipeline includes a texture memory and a graphics processor coupled to the texture memory. The texture memory provides a location to store texture information. The graphics processor provides processor to process the texture information by shifting and blending the texture information in one pass through the graphics processor to obtain shifted and blended texture information.

Growing community of inventors

Gig Harbor, WA, United States of America

William H Clifford