Advanced Micro Devices Corporation

Globalfoundries Inc.

Ati Technologies Ulc

Chen-Ping Yang

Vydhyanathan Kalyanasundharam

William Alexander Hughes

Greggory D Donley

Michael K Fertig

Amit P Apte

Alexander J Branover

Benjamin Tsien

John P Petry

Eric Christopher Morton

Elizabeth Morrow Cooper

Rostyslav Kyrychynskyi

A processing system that includes a shared data fabric resets a first client processor while operating a second client processor. The first client processor is instructed to stop making requests to one or more devices of the shared data fabric. Status communications are blocked between the first client processor and a memory controller, the second client processor, or both, such that the first client processor enters a temporary offline state. The first client processor is indicated as being non-coherent. Accordingly, when the processor is reset some errors and efficiency losses due messages sent during or prior to the reset are prevented.

Shared data fabric processing client reset system and method

Systems, apparatuses, and methods for routing traffic between clients and system memory are disclosed. A computing system includes system memory and one or more clients, each capable of generating memory access requests. The computing system also includes a communication fabric for transferring traffic between the clients and the system memory. The fabric includes master units for interfacing with clients and grouping write requests with a same target together. The fabric also includes slave units for interfacing with memory controllers and for sending a single write response when each write request in a group has been serviced. When the master unit receives the single write response for the group, it sends a respective acknowledgment response for each of the multiple write requests in the group to clients that generated the multiple write requests.

Method to reduce write responses to improve bandwidth and efficiency

Systems, apparatuses, and methods for implementing a cancel and replay mechanism for ordered requests are disclosed. A system includes at least an ordering master, a memory controller, a coherent slave coupled to the memory controller, and an interconnect fabric coupled to the ordering master and the coherent slave. The ordering master generates a write request which is forwarded to the coherent slave on the path to memory. The coherent slave sends invalidating probes to all processing nodes and then sends an indication that the write request is globally visible to the ordering master when all cached copies of the data targeted by the write request have been invalidated. In response to receiving the globally visible indication, the ordering master starts a timer. If the timer expires before all older requests have become globally visible, then the write request is cancelled and replayed to ensure forward progress in the fabric and avoid a potential deadlock scenario.

Cancel and replay protocol scheme to improve ordered bandwidth

In an embodiment, a node comprises a packet scheduler configured to schedule packets to be transmitted on a link and an interface circuit coupled to the packet scheduler and configured to transmit the packets on the link. The interface circuit is configured to generate error detection data covering the packets, wherein the error detection data is transmitted between packets on the link. The interface circuit is configured to cover up to N packets with one transmission of error detection data, where N is an integer &#x3e;=2. The number of packets covered with one transmission of error detection data is determined by the interface circuit dependent on an availability of packets to transmit. In another embodiment, the interface circuit is configured to dynamically vary a frequency of transmission of the error detection data on the link based on an amount of bandwidth being consumed on the link.

Command packet packing to mitigate CRC overhead

In an embodiment, a node comprises a packet scheduler configured to schedule a packet to be transmitted on the link, the packet comprising a command and associated packet data. Coupled to the packet scheduler and configured to transmit the packet on the link, and interface circuit is configured to generate error detection data covering the packet. The interface circuit is configured to transmit the error detection data covering the packet at an end of the packet, and is further configured to insert at least one partial error detection data within the packet. The partial error detection data covers a portion of the packet that precedes the partial error detection data. A receiver is configured to receive the data and forward the data based on partial CRC check.

Partial CRC insertion in data packets for early forwarding

Growing community of inventors

Fremont, CA, United States of America

Chen-Ping Yang