Intel Corporation

Other

Tuong P Trieu

Zohar Bogin

Anoop Mukker

Aditya Sreenivas

Wishwesh Anil Gandhi

David D Lent

Ashish S Gadagkar

James P Kardach

Steve John Clohset

David E Freker

Aditya Navale

Thomas A Piazza

Trung Am Diep

Robert J Johnston

Sachidanandan Sambandan

Vincent E VonBokern

Sarath Kotamreddy

David Sastry

Surya Kareenahalli

A method and apparatus for dedicating cache entries to certain streams for performance optimization are disclosed. The method according to the present techniques comprises partitioning a cache array into one or more special-purpose entries and one or more general-purpose entries, wherein special-purpose entries are only allocated for one or more streams having a particular stream ID.

Method and apparatus for dedicating cache entries to certain streams for performance optimization

Embodiments of the present invention relate to accessing a first pair of adjacent data blocks using a first channel of a dual channel memory device; and simultaneously accessing a second pair of adjacent data blocks using a second channel of the memory device, the second pair being spaced apart from the first pair by a predetermined interval.

Optimized memory addressing

A method to deterministically shut down memory devices in response to a system warm reset has been disclosed. One embodiment of the method includes causing a first type of reset in a number of memory devices in a system in response to a second type of reset in the system being initiated if the memory devices are not initialized and enabling a deterministic shutdown mode in a memory controller, which is coupled to the memory devices, after the memory devices have been initialized. Other embodiments are described and claimed.

Deterministic shut down of memory devices in response to a system warm reset

According to one embodiment a chipset is disclosed. The chipset includes a graphics accelerator, a memory controller and a queue mechanism. The queue mechanism includes a first functional unit block (FUB) coupled to the graphics accelerator, and a second FUB coupled to the memory controller.

Queue partitioning mechanism

A method and apparatus for improving read latency for processor to system memory read transactions is disclosed. One embodiment of a system logic device includes logic that assumes a transfer size of a predetermined length. In this manner, the system logic device can issue a read transaction request to system memory as soon as the read request address is delivered by the processor rather than waiting for the processor to deliver information indicating the transfer length. Once the actual transfer length information is delivered from the processor to the system logic device, the system logic device determines whether any of the data returned by the system memory needs to be purged before returning the requested data to the processor.

Method and apparatus for improving read latency for processor to system memory read transactions

According to one embodiment, a computer system is disclosed that includes a memory and a memory controller coupled to the memory. The memory controller includes an arbitration unit that may be programmed to operate according to a first arbitration mode or a second arbitration mode. The computer system also includes a first device and a second device coupled to the arbitration unit. According to a further embodiment, the first device is assigned a higher priority classification than the second device for accessing the memory while the arbitration unit is operating according to the first arbitration mode. In addition, the first device and the second device are assigned equal priority classifications for accessing the memory while the arbitration unit is operating according to the second arbitration mode.

Arbitration mechanism for a computer system having a unified memory architecture

A computer system is provided. The computer system includes a host processor (HP), a system memory (SM), and an input/output (I/O) master device to perform a read of a continuous stream of data to the SM. The computer system also includes a bridge coupled to the HP, SM, and I/O master device. The bridge reads ahead to the SM when the I/O master device reads a continuous stream of data from the SM. The bridge aborts read ahead accesses to the SM, prior to an access commit point to the SM, responsive to disengagement of the I/O master device.

Abort of DRAM read ahead when PCI read multiple has ended

In a bus resource having an outbound pipe for processing both non-posted and posted transactions in a FIFO manner, a rejected non-posted transaction at the head of the outbound pipe is moved aside and into an auxiliary buffer to avoid a potential blockage of the outbound pipe. The auxiliary buffer is for holding transaction information and return data of the rejected non-posted transaction. The rejected transaction is eventually completed from the auxiliary buffer as determined by an arbiter.

Avoiding deadlock by storing non-posted transactions in an auxiliary buffer when performing posted and non-posted bus transactions from an outbound pipe

An arbiter comprising logic circuitry configured to delay granting bus ownership to an agent X in response to receiving a first signal which indicates that a device coupled to the bus is not available to service transactions directed at the device over the bus. For instance, the device may be a bridge and the grant is delayed if an inbound pipe of the bridge is full. The arbiter may provide a borrowed grant to an outbound pipe of the device for performing a transaction on the bus while waiting for an inbound pipe of the device to become available.

Avoiding livelock when performing a long stream of transactions

A slave device having clock rate compensation circuitry for low frequency operation. The slave device is coupled to a bus having a first operating frequency yet uses a slave clock signal having a frequency less than the first operating frequency. The slave device includes a bus clock driver circuit coupled to a bus clock interface for a bus clock signal. A slave controller state machine is clocked by the slave clock signal and accordingly operates at less than the first operating frequency. The clock rate compensation circuitry receives the bus clock signal, a data signal, and the slave clock signal, and synchronizes bus events for the state machine. The clock rate compensation circuitry also asynchronously begins a bus clock signal stretching period.

Clock rate compensation for a low frequency slave device

An output buffer circuit including different arrangements of output devices selectable by circuitry which tests the load characteristics to provide different buffer drive strengths, and different paths having different current carrying characteristics for enabling the output devices selectable by the circuitry which tests the load characteristics to vary the slew rate of the output devices to best match the load.

Output buffer with current paths having different current carrying

An address shadow feature and methods of using the same. A slave controller of the present invention includes an address register coupled to receive a device address from a secondary bus interface. A match circuit is coupled to the address register and compares the device address to a shadow address, generating a match signal upon detection of a match between the shadow address and the device address. An interrupt generation circuit generates an interrupt signal in response to the match signal.

Growing community of inventors

Folsom, CA, United States of America

Tuong P Trieu