Intel Corporation

Michael W Leddige

James A McCall

Bryce D Horine

John Thomas Sprietsma

Kuljit Singh Bains

Chaitanya Sreerama

Stephen H Hall, Jr

Olufemi B Oluwafemi

Peter D MacWilliams

Randy M Bonella

Hing Yan To

Maynard C Falconer

Stephen H Hall

Jason A Mix

Paul G Huray

David J Harriman

Mikal Hunsaker

Xiang Li

Su Wei Lim

Howard L Heck

Lily Pao Looi

Thomas J Holman

Kuan-Yu Chen

Michael J Espig

Ronald W Swartz

Yun Ling

Mark B Trobough

Zhen Fang

Vui Yong Liew

Rohit R Verma

Bryan L Spry

Mohiuddin M Mazumder

Sadagopan Srinivasan

Bin Li

Ven R Holalkere

Brad Larson

Alok Tripathi

Antonio Zenteno Ramirez

Kai Wang

Earl J Wight

Jose A Sanchez Sanchez

Pascal C Meier

Ajit Deosthali

Kuan-yu Chen

Earl Wight

Paul Huray

An apparatus is described herein. The apparatus includes a plurality of conductors, wherein at least one conductor is a common-mode conductor. The apparatus also includes an encoder to encode data to be transmitted on the plurality of conductors, wherein a data speed of the common-mode conductor is limited and a data speed of other conductors is maximized according to an encoding matrix.

Mode selective balanced encoded interconnect

Techniques and mechanisms for exchanging single-ended communications with a protocol stack of an integrated circuit package. In an embodiment, an integrated circuit (IC) chip includes a protocol stack comprising a transaction layer which performs operations compatible with a Peripheral Component Interconnect Express&#x2122; (PCIe&#x2122;) specification. Transaction layer packets, exchanged between the transaction layer and a link layer of the protocol stack, are compatible with a PCIe&#x2122; format. In another embodiment, a physical layer of the protocol stack is to couple the IC chip to another IC chip for an exchange of the transaction layer packets via single-ended communications. A packaged device includes both of the IC chips.

Method, apparatus and system for single-ended communication of transaction layer packets

Techniques for decoding encoded data are described herein. An example of a device in accordance with the present techniques includes a receiving signaling module coupled to a plurality of signal lines. The signaling module includes a receiver to receive a plurality of encoded line voltages or currents on the plurality of signal lines of a bus, wherein each one of the plurality of encoded line voltages corresponds to a weighted sum of data. The signaling module includes a comparator to determine the voltage level of each line at a unit interval and convert the voltage level to a digital value. The signaling module includes a lookup table correlating the digital value with a digital bit stream.

Crosstalk aware decoding for a data bus

Methods and systems to support input output (IO) communications may include an IO connector having a housing with surfaces defining a paddle card region, and a set of compressible contacts extending vertically through the housing into the paddle card region. In addition, an IO interconnect can include a cable portion and at least one end portion coupled to the cable portion. The end portion may include a paddle card having a circuit board with a set of contacts disposed on a bottom surface of the circuit board. The end portion can also include an asymmetric metal shell having a configuration that encloses at least a portion of the paddle card and exposes the set of contacts.

High bandwidth connector for internal and external IO interfaces

Techniques for encoding data are described herein. An example of a device in accordance with the present techniques includes a signaling module coupled to a plurality of digital inputs. The signaling module is to encode data received at the plurality of digital inputs to generate encoded data. Based on the encoded data, the signaling module can drive line voltages on a plurality of signal lines of a bus. Each one of the plurality of line voltages corresponds to a weighted sum of the data received at the plurality of digital inputs.

Crosstalk aware encoding for a data bus

A memory module has an array of connections. The array of connections is arranged in rows and columns such that there are first and second outer columns. Connections in the first and second outer columns can be interchanged to optimize double-side module placement on a substrate. Other embodiments are also disclosed and claimed.

Interchangeable connection arrays for double-sided DIMM placement

According to some embodiments, first and second processing elements may be provided on a die, and there may be a plurality of potential communication links between the first and second processing elements. Moreover, control logic may be provided on the die to dynamically activate at least some of the potential communication links (e.g., based on a current bandwidth appropriate between the first and second processing elements).

Dynamic configuration of potential links between processing elements

In one embodiment, the present invention includes a method for receiving a memory request from a device coupled to an input/output (IO) interconnect, accessing a mapping table associated with the IO interconnect to determine if an address range including an address of the memory request is coherent, and if so, sending the memory request and a coherency indicator to indicate the coherent state of data at the address, otherwise sending the memory request and the coherency indicator to indicate a non-coherent state. Other embodiments are described and claimed.

Providing address range coherency capability to a device

A memory module has an array of connections. The array of connections is arranged in rows and columns such that there are first and second outer columns. Connections in the first and second outer columns can be interchanged to optimize double-side module placement on a substrate.

A stackable die mounting system with an efficient interconnect is disclosed that can have a base chip carrier to interconnect a base integrated circuit die to a circuit board on a first side and to a second stacked integrated circuit on a second side. The second side can include a first region having a pad out configuration of a first input output (I/O) to transmit data to be stored by the stacked integrated circuit die. The base chip carrier can have a second region with a pad out of a second I/O that is configured to receive data transmitted by the stacked integrated circuit die wherein the pad out of the second port is translated and rotated about an axis from the pad out of the first region such that a busses with different functions can be vertically integrated from the circuit board.

Stacking integrated circuit dies

An integrated circuit interconnection system is disclosed. The system can include a first integrated circuit die having a first electrode configuration and a second integrated die having the same or a substantially similar electrode configuration. The system can also include a multilayer flexible cable having a first side and a second side that has substantially parallel conductors running along the cable. At least a portion of one of the parallel conductors can be exposed on the first side and/or the second side, such that the first and second integrated circuit die can be connected to both the first side and the second side of the multilayer flexible cable. The cable can be folded to provide a dense interconnect for stacked memory configurations.

Systems and arrangements for interconnecting integrated circuit dies

An apparatus includes a printed circuit board (PCB) and a first flexible conductive cable ('flex cable') secured to the PCB. The apparatus also includes a daughter card having an end adjacent to the PCB and a second flex cable secured to the daughter card. The apparatus further includes a connector which provides an electrically conductive connection between the first flex cable and the second flex cable. The connector is positioned to sandwich a portion of the first flex cable between the connector and the PCB.

Technique for blind-mating daughtercard to mainboard

Embodiments of the invention provide a memory command and address (CA) bus architecture that can accommodate higher CA data output frequencies with reduced signal degradation. For one embodiment of the invention the CA is divided on the motherboard and a CA signal component routed to each of two dual in-line memory modules (DIMMs) of a two-DIMM/channel memory bus design. The CA signal component on each DIMM is then routed sequentially through each dynamic random access memory (DRAM) chip on the respective DIMM. In one embodiment, after routing through each DRAM, the CA signal is terminated on the DIMM. In an alternative embodiment, the CA signal is terminated on the die at the last DRAM of each respective DIMM.

Memory system and method to reduce reflection and signal degradation

Embodiments of the invention provide a memory command and address (CA) bus architecture that can accommodate higher CA data output frequencies with reduced signal degradation. For one embodiment of the invention the CA signal is routed to a first of two dual in-line memory modules (DIMMs) of a two-DIMM/channel memory bus design. The CA signal is then divided into components, with each CA signal component routed serially through a group of dynamic random access memory (DRAM) chips on the first DIMM. The CA signal components are then recombined and routed to the second DIMM. The recombined CA signal is then divided again into components, with each CA signal component routed serially through a group of dynamic random access memory (DRAM) chips on the first DIMM and the CA signal components are then recombined. In one embodiment, after routing through each DRAM, the CA signal is terminated on the DIMM.

Circulator chain memory command and address bus topology

Growing community of inventors

Beaverton, OR, United States of America

Michael W Leddige