Advanced Micro Devices Corporation

Ati Technologies Inc.

Ati Technologies Ulc

Other

Joseph D Macri

Lin Chen

Ming-Ju Edward Lee

Stephen L Morein

Claude R Gauthier

Osamu Nagashima

Grigori Temkine

Olge Drapkin

Daniel L Bouvier

Steven Morein

Maurice B Steinman

Gerald Robert Talbot

Gordon Caruk

Gin S Yee

Arvind Bomdica

Siji Menokki Kandiyil

Edward Lee

Sam Huynh

Dan Shimizu

Mark Frankovich

Hiok-Tiaq Ng

John Ling

Jennifer Ho

Siji MK

Gin Yee

Lee E Ming-Ju

Chen Lin

Ahmad Sagheer

Eric Scott

A link controller includes a Peripheral Component Interconnect Express (PCIe) physical layer circuit for coupling to a communication link and providing a data path over the communication link, a first data link layer controller which operates according to a PCIe protocol, and a second data link layer controller which operates according to a non-PCIe protocol. A multiplexer-demultiplexer selectively connects both data link layer controllers to the PCIe physical layer circuit. A protocol translation circuit is coupled between the multiplexer-demultiplexer and the second data link layer controller, the protocol translation circuit receiving traffic data from the second data link layer controller in a non-PCIe format, encapsulating the non-PCIe format in a PCIe format, and passing traffic data to the multiplexer-demultiplexer circuit.

Alternative protocol over physical layer

A link controller includes a Peripheral Component Interconnect Express (PCIe) physical layer circuit for coupling to a communication link and providing a data path over the communication link, a first data link layer controller which operates according to a PCIe protocol, and a second data link layer controller which operates according to a Gen-Z protocol. A multiplexer-demultiplexer selectively connects both data link layer controllers to the PCIe physical layer circuit. A protocol translation circuit is coupled between the multiplexer-demultiplexer and the second data link layer controller, the protocol translation circuit receiving traffic data from the second data link layer controller in a Gen-Z format, encapsulating the Gen-Z format in a PCIe format, and passing traffic data to the multiplexer-demultiplexer circuit.

A computer system includes a first processor, a second processor, and a common memory connected to the second processor. The computer system is switched from a high performance mode, in which at least a portion of the first processor and at least a portion of components on the second processor are active, to a low power mode, in which at least a portion of the first processor is active and the components on the second processor are inactive. All central processing unit (CPU) cores on the second processor are quiesced. Traffic from the second processor to the common memory is quiesced. Paths used by the first processor to access the common memory are switched from a first path across the second processor to a second path across the second processor.

Switching a computer system from a high performance mode to a low power mode

A system and method for detecting errors in high-speed asymmetric interfaces are described. Embodiments include transmitting digital data between a first system component and a second system component over a bidirectional interface, wherein the first component is significantly more intelligent than the second component. The first component receives a signature from the second component over a line of the interface concurrent with READ and WRITE operations over the interface. The latency associated with transmission of a signature from the second component to the first component is the time taken for the second component to compute a signature. The signature received is compared to a signature stored by the first component. Both signatures correspond to a particular READ or WRITE command. Based on the comparison, the first component determines whether the READ or WRITE operation was successful, and directs the second component as necessary.

Error detection in high-speed asymmetric interfaces utilizing dedicated interface lines

A single interconnect is provided between a first processor and a second processor, such that the first processor may access a common memory through the second processor while the second processor can be mostly powered off. The first processor accesses the memory through a memory controller using a standard dynamic random access memory (DRAM) bus protocol. Instead of the memory controller directly connecting to the memory, the access path is through the second processor to the memory. Additionally, a bidirectional communication protocol bus is mapped to the existing DRAM bus signals. When both the first processor and the second processor are active, the bus protocol between the processors switches from the DRAM protocol to the bidirectional communication protocol. This enables the necessary chip-to-chip transaction semantics without requiring the additional cost burden of a dedicated interface for the bidirectional communication protocol.

Polymorphous signal interface between processing units

A method and system for performing byte-writes are described, where byte-writes involve writing only particular bytes of a multiple byte write operation. Embodiments include mask data that indicates which bytes are to be written in a byte-write operation. No dedicated mask pin(s) or dedicated mask line(s) are used. In one embodiment, the mask data is transmitted on data lines and store in response to a write_mask command. In one embodiment, the mask data is transmitted as part of the write command.

Write data mask method and system

A system and method for detecting errors in high-speed asymmetric interfaces are described. Embodiments include transmitting digital data between a first system component and a second system component over a bidirectional interface, wherein the first component is significantly more intelligent than the second component. The first component controls many operations of the second component, including receiving a signature from the second component over an existing line of the interface. The signature received is compared to a signature stored by the first component. Both signatures correspond to a transaction over the interface. Based on the comparison, the first component determines whether the transaction was successful, and directs the second component as necessary.

Error detection in high-speed asymmetric interfaces

An asymmetrical IO method and system are described. In one embodiment, a host device includes shared resources for data synchronization of the host device and a client device. The shared resources include a shared phase interpolator. In an embodiment, data lines between the host and client are also used to transmit phase information from the client device to the host device, obviating the need for additional, dedicated lines or pins.

System and apparatus for transmitting phase information from a client to a host between read and write operations

A system includes a first communication device and a second communication device. The first communication device includes a programmable region. The programmable region of the first communication device is programmed so that an associated signal includes a number of preamble cycles. The second communication device also can include a programmable region. The programmable region of the second communication device can be programmed so that an associated signal includes a number of preamble cycles. The number of preamble cycles can be based on a variety of factors, such as the topology or implementation of the system. In an embodiment, the number of preamble cycles is associated with a data strobe signal, and data is not read or written in response to the data strobe signal until all of the preamble cycles have been transmitted and received.

Programmable preamble system and method

The present invention is a method of writing information to a synchronous memory device by examining a present word of N bits to be written, where each bit has a high or low value. The present word is compared to a previous word also having N bits to identify the number of bit transitions from a low value to a high value of vice versa. The present bit is inverted when the number of transitions is greater than N/2. To avoid the need for having an extra bit accompany data bytes to indicate the presence or absence of inversion, the present invention takes advantage of a data mask pin that is normally unused during writing operations to carry the inversion bit. Non-inverted data is written directly into the memory device while inverted data is first inverted again before writing to storage locations, so that true data is stored in the memory device.

Method and apparatus for data inversion in memory device

Embodiments of a power consumption reduction process for memory interfaces are described. A power management process reduces the amount of time that current flows in a high or low terminated, current or voltage mode unipolar bus interface by reducing the amount of time the bus remains in a logic state that requires current flow.

Low power mode unipolar current/voltage mode interface

A dynamic bus inversion (DBI) method and system are described. In various embodiments, a transmitter transmits data over a multi-bit high-speed bus to a receiver. In an embodiment, the transmitter determines whether to invert the bus based on the number of data bits that will be transitioning to a new value. If it is determined that the bus is to be inverted, the transmitter encodes a DBI signal on a shared line of the bus. In an embodiment, the shared line is used for different purposes at different times, obviating the need for a dedicated line or pin for the encoded DBI signal. The receiver receives and decodes the DBI signal and, in response, appropriately decodes the received data.

Growing community of inventors

San Francisco, CA, United States of America

Joseph D Macri