An integrated circuit includes a first input/output lane comprising first external terminals and first driver circuits. The first driver circuits exchange signals with a first external device through the first external terminals as part of a first external interface. The first input/output lane is part of a sub-bank in an input/output bank that implements at least a part of the first external interface. The integrated circuit includes a second input/output lane comprising second external terminals and second driver circuits. The second driver circuits exchange signals with a second external device through the second external terminals as part of a second external interface. The second input/output lane is part of the sub-bank in the input/output bank that implements at least a part of the second external interface.

A writing-block for writing data to a memory-buffer, wherein the memory-buffer comprises an ordered sequence of elements and the writing-block is configured to: receive an input-data-stream; and write the input-data-stream to the memory-buffer in a successive manner from a first-element of the ordered sequence to a predetermined-element of the ordered sequence. Following writing to the predetermined-element the writing-block is configured to continue to write the input-data-stream to the memory-buffer in a successive manner restarting at the first-element. In response to writing the predetermined-element, the writing-block is configured to also continue to write the input-data-stream to the memory-buffer in a successive manner from an element immediately following the predetermined element until a second predetermined-element of the memory-buffer.

An endpoint in a network may make posted or non-posted write requests to another endpoint in the network. For a non-posted write request, the target endpoint provides a response to the requesting endpoint indicating that the write request has been serviced. For a posted write request, the target endpoint does not provide such an acknowledgment. Hence, posted write requests have lower overhead, but they suffer from potential synchronization and resiliency issues. While non-posted write requests do not have those issues, they cause increased load on the network because such requests require the target endpoint to acknowledge each write request. Introduced herein is a network operation technique that uses non-posted transactions while maintaining a load overhead of the network as a manageable level. The introduced technique reduces the load overhead of the non-posted write requests by collapsing and reducing a number of the responses.

Systems, apparatuses, and methods for efficient data transfer in a computing system are disclosed. A source generates packets to send across a communication fabric (or fabric) to a destination. The source generates partition enable signals for the partitions of payload data. The source negates an enable signal for a particular partition when the source determines the packet type indicates the particular partition should have an associated asserted enable signal in the packet, but the source also determines the particular partition includes a particular data pattern. Routing components of the fabric disable clock signals to storage elements assigned to store the particular partition. The destination inserts the particular data pattern for the particular partition in the payload data.

A method and apparatus for scrambling and descrambling data in a computer system includes transmitting non-scrambled data from a first high speed inter chip (IP) link circuit located on a first chip to a first serializer/deserializer (SERDES) physical (PHY) circuit located on the first chip, the first high speed link IP indicating the data is not scrambled. The received non-scrambled data is scrambled by the first SERDES PHY circuit and transmitted to a second chip. The received scrambled data is descrambled by a second SERDES PHY circuit located on the second chip. The non-scrambled data is transmitted by the second SERDES PHY circuit to a second high speed link IP circuit located on the second chip to a third circuit for further processing or transmission.

In one embodiment, a Light Emitting Diode (LED) driving device for driving a plurality of LEDs has a switching matrix utilizing a plurality of one of a turn off thyristors or turn off triacs coupled to the plurality of LEDs. A controller is coupled to the switching matrix responsive to a voltage of a rectified AC halfwave, wherein combinations of the plurality of LEDs are altered to ensure a maximum operating voltage of the plurality of LEDs is not exceeded. A current limiting device is coupled to the combinations of the plurality of LED to regulate current. In a second embodiment an offline charge pump utilizes a switching matrix to recombine capacitors in accordance with the voltage on the AC half wave and then in, accordance with a desired output voltage to feed a load, such that said recombinations occur at a frequency much higher than the frequency of the AC rectified half wave such that charge is pumped from the input at one voltage to the output at another voltage through the AC halfwave while, providing a constant output voltage to the load.

An endpoint in a network may make posted or non-posted write requests to another endpoint in the network. For a non-posted write request, the target endpoint provides a response to the requesting endpoint indicating that the write request has been serviced. For a posted write request, the target endpoint does not provide such an acknowledgment. Hence, posted write requests have lower overhead, but they suffer from potential synchronization and resiliency issues. While non-posted write requests do not have those issues, they cause increased load on the network because such requests require the target endpoint to acknowledge each write request. Introduced herein is a network operation technique that uses non-posted transactions while maintaining a load overhead of the network as a manageable level. The introduced technique reduces the load overhead of the non-posted write requests by collapsing and reducing a number of the responses.

An information handling system includes a first device having a first data communication interface connected to a first socket area of a socket. A second device includes a second data communication interface connected to a second socket area of the socket. A host processor includes a third data communication interface connected to a third socket area of the socket. When an interposer is installed into the socket in a first orientation, the interposer connects the first data communication interface to the third data communication interface. When the interposer is installed into the socket in a second orientation, the interposer connects the first data communication interface to the second data communication interface.

A multi-endpoint adapter device includes a splitter device that is coupled to a network port and a plurality of endpoint subsystems that are each coupled to a processing subsystem. The splitter device receives, via the network port, a first data payload, and identifies both a first data sub-payload that is included in the first data payload and that is associated with a first endpoint subsystem included in the plurality of endpoint subsystems and a second data sub-payload that is included in the first data payload and that is associated with a second endpoint subsystem included in the plurality of endpoint subsystems. The splitter device then splits the first data payload into the first data sub-payload and the second data sub-payload, and forwards both the first data sub-payload to the first endpoint subsystem and the second data sub-payload to the second endpoint subsystem.

A processing element includes an input zero detector to detect whether the input from the neighbor processing element contains a zero. When the input from the neighbor processing element contains the zero, a zero disable circuit controls the input from the neighbor processing element and respective data of the memory to both appear as unchanged to the arithmetic logic unit for the operation. A controller of an array of processing elements adds a row of error-checking values to a matrix of coefficients, each error-checking value of the row of error-checking values being a negative sum of a respective column of the matrix of coefficients. The controller controls a processing element to perform an operation with the matrix of coefficients and an input vector to accumulate a result vector. Owing to the error-checking values, when a sum of elements of the result vector is non-zero, an error is detected.