A device includes a receiver to receive one or more training sequences during a training of a link, where the link connects two devices. The device may include agent logic to determine, from the one or more training sequences, a number of extension devices on the link between the two devices, and determine that the number of extension devices exceeds a threshold number. The device may include a transmitter to send a plurality of clock compensation ordered sets on the link based on determining that the number of extension devices exceeds a threshold number.

A device includes a receiver to receive one or more training sequences during a training of a link, where the link connects two devices. The device may include agent logic to determine, from the one or more training sequences, a number of extension devices on the link between the two devices, and determine that the number of extension devices exceeds a threshold number. The device may include a transmitter to send a plurality of clock compensation ordered sets on the link based on determining that the number of extension devices exceeds a threshold number.

Disabling communication in a multiprocessor fabric. The multiprocessor fabric may include a plurality of processors and a plurality of communication elements and each of the plurality of communication elements may include a memory. A configuration may be received for the multiprocessor fabric, which specifies disabling of communication paths between one or more of: one or more processors and one or more communication elements; one or more processors and one or more other processors; or one or more communication elements and one or more other communication elements. Accordingly, the multiprocessor fabric may be automatically configured in hardware to disable the communication paths specified by the configuration. The multiprocessor fabric may be operated to execute a software application according to the configuration.

An information handling system includes processors disposed in sockets, and interconnect links providing point-to-point links between the sockets. One of the processors determines an arrangement of the processors, memories and the interconnect links, and determines a value for each of the processors, each of the memories, and each of the interconnect links. The processor calculates interconnect link bandwidth values for each of the interconnect links based at least in part on the determined value and the arrangement of the processors, the memories and the interconnect links. The processor also populates an interconnect bandwidth table using the interconnect link bandwidth values.

Hardware accelerators and methods for out-of-order processing are described. In one embodiment, a processor includes a hardware accelerator having a plurality of processing elements coupled to form a plurality of logical rows of a multidimensional processing array and a plurality of logical columns of the multidimensional processing array, wherein a processing element of the plurality of processing elements includes a switch to selectively source, from either of an output for a first dataset from an upstream processing element of the plurality of processing elements or a boundary condition value for a second dataset stored in the processing element, based on a switch control value provided to the processing element; and a core coupled to the hardware accelerator.

Disabling communication in a multiprocessor fabric. The multiprocessor fabric may include a plurality of processors and a plurality of communication elements and each of the plurality of communication elements may include a memory. A configuration may be received for the multiprocessor fabric, which specifies disabling of communication paths between one or more of: one or more processors and one or more communication elements; one or more processors and one or more other processors; or one or more communication elements and one or more other communication elements. Accordingly, the multiprocessor fabric may be automatically configured in hardware to disable the communication paths specified by the configuration. The multiprocessor fabric may be operated to execute a software application according to the configuration.

The flow of data in an integrated circuit is controlled. The integrated circuit comprising a plurality of tiles, each tile comprising a processor, a switch including switching circuitry to forward data over data paths from other tiles to the processor and to switches of other tiles, and a receive buffer to store data from the switch. At a first tile, a count is maintained of data that has been sent to a second tile without receiving an acknowledgement up to a credit limit. At the second tile, data that arrives from the first tile when the receive buffer is full is sent to a memory outside of the tile.

Embodiments are provided herein for component composition of a disaggregated computing system. A plurality of general purpose links connecting a computing element to other hardware elements are provided within the disaggregated computing system. Each of the plurality of general purpose links comprise a point-to-point connection to at least one of the other hardware elements such that the plurality of general purpose links conform to a configuration used by the other hardware elements regardless of a type of data being transferred through the plurality of general purpose links.

The present invention relates to cache coherent node controllers for scale-up shared memory systems. In particular it is disclosed a computer system at least comprising a first group of CPU modules connected to at least one first FPGA Node Controller configured to execute transactions directly or through a first interconnect switch to at least one second FPGA Node Controller connected to a second group of CPU modules running a single instance of an operating system.

A node controller for a first processor socket group may include a node memory storing a coherence directory and logic. Logic may cause the node controller to: receive a memory operation request directly from a second processor socket group, follow a coherence protocol based on the memory operation request and the coherence directory and directly access a socket group memory of the first processor socket group based on the request.