Cavium, Inc.

Cavium Networks, Inc.

Oavium Networks, Inc.

Robert A Sanzone

Richard Eugene Kessler

Wilson Parkhurst Snyder, Ii

Muhammad Raghib Hussain

Michael S Bertone

Thomas Frederick Hummel

Daniel A Katz

David Asher

Faisal Masood

Imran Badr

David Albert Carlson

Gregg Alan Bouchard

Khaja E Ahmed

Michael D Varga

Philip H Dickinson

Aseem Maheshwari

A data processor includes an input/output bridge that provides enforcement of a security status on transactions between devices across the bridge. The bridge includes circuitry to parse a received request to obtain one or more identifiers, and compare the identifiers against one or more programmable lookup tables. Based on this comparison, the bridge can determine the security status of the transaction, as well as selectively forward the transaction based on the security status.

Programmable validation of transaction requests

An arbiter circuit manages and enforces arbitration and quality of service (QOS) among multiple devices accessing a resource, such as a memory. The arbiter circuit receives requests from a number of devices to use resources of a bridge connecting to a memory, and maintains a count of bridge resources available on a per-device and per-bus basis. The arbiter circuit operates to select a next one of the requests to grant a bridge resource based on the device originating the request, a count of the per-device resources available, and a count of the resources available to the bus connecting the device to the bridge.

Arbitrated access to resources among multiple devices

A circuit manages and controls access requests to a register, such as a control and status register (CSR) among a number of devices. In particular, the circuit selectively forwards or suspends off-chip access requests and forwards on-chip access requests independent of the status of off-chip requests. The circuit receives access requests at a plurality of buses, one or more of which can be dedicated to exclusively on-chip requests and/or exclusively off-chip requests. Based on the completion status of previous off-chip access requests, further off-chip access requests are selectively forwarded or suspended, while on-chip access request are sent independently of off-chip request status.

Register access control among multiple devices

In one embodiment, a system includes a packet reception unit. The packet reception unit is configured to receive a packet, create a header indicating scheduling of the packet in a plurality of cores and concatenate the header and the packet. The header is based on the content of the packet. In one embodiment, a system includes a transmit silo configured to store a multiple fragments of a packet, the fragments having been sent to a destination and the transmit silo having not received an acknowledgement of receipt of the fragments from the destination. The system further includes a restriction verifier coupled with the transmit silo. The restriction verifier is configured to receive the fragments and determine whether the fragments can be sent and stored in the transmit silo.

Messaging with flexible transmit ordering

An input/output bridge controls access to a memory by a number of devices. The bridge enforces ordering of access requests according to a register storing an order configuration, which can be programmed to accommodate a given application. When suspending an access request as a result of enforcing an order configuration, the bridge may also cause a prefetch at the memory for the suspended access request. Subsequently, following the completion of a previous access request meeting the order configuration, the suspended access request is released. Due to the prefetch, an access operation can be completed with minimal delay.

Programmable ordering and prefetch

Work submitted to a co-processor enters through one of multiple input queues, used to provide various quality of service levels. In-memory linked-lists store work to be performed by a network services processor in response to lack of processing resources in the network services processor. The work is moved back from the in-memory inked-lists to the network services processor in response to availability of processing resources in the network services processor.

Processor with efficient work queuing

In an embodiment, a method for managing access to memory includes receiving requests for access to a memory from one or more devices, each particular request associated with one of a plurality of virtual channels. A tag is assigned to each request received. Each tag assigned is added to a linked list associated with the corresponding virtual channel. Each request received with the assigned tag is transmitted to the memory. Responses to the requests are received from the memory, each response having an associated tag, and the responses received are sent to the one or more devices based on the corresponding linked list and the corresponding tag.

Method and apparatus for memory access management

In one embodiment, a system includes reassembly stores configured to store a fragment of a packet in a particular reassembly store corresponding with the packet, and when the particular reassembly store contains fragments of the packet representing the packet as a whole, forward the packet to a plurality of cores. The system further includes a packet reception unit configured to store the fragment in one of a plurality of memories within the reassembly stores, and, when the one of the plurality of memories is filled, copy the at least one fragment to a memory external to the packet reception unit.

A network services processor includes an input/output bridge that avoids unnecessary updates to memory when cache blocks storing processed packet data are no longer required. The input/output bridge monitors requests to free buffers in memory received from cores and IO units in the network services processor. Instead of writing the cache block back to the buffer in memory that will be freed, the input/output bridge issues don't write back commands to a cache controller to clear the dirty bit for the selected cache block, thus avoiding wasteful write-backs from cache to memory. After the dirty bit is cleared, the buffer in memory is freed, that is, made available for allocation to store data for another packet.

Method and apparatus for managing write back cache

In one embodiment, a system includes a memory and a first bridge unit for processor access with the memory coupled with an input-output bus and the memory. The first bridge unit is configured to receive requests from the input-output bus to read or write data receive requests from the MFNU to free memory and choose among the requests to send to the memory on a first memory bus. The system also includes a second bridge unit for packet data access with the memory coupled with a packet input unit, packet output unit, and the memory. The second bridge unit is configured to receive requests to write packet data from the packet input unit, receive requests to read packet data from the packet output unit, and choose among the requests from the packet input unit and the packet output unit to send to the memory on a second memory bus.

Input output bridging

In one embodiment, a system comprises a memory, and a first bridge unit for processor access with the memory. The first bridge unit comprises a first arbitration unit that is coupled with an input-output bus, a memory free notification unit ('MFNU'), and the memory, and is configured to receive requests from the input-output bus and receive requests from the MFNU and choose among the requests to send to the memory on a first memory bus. The system further comprises a second bridge unit for packet data access with the memory that includes a second arbitration unit that is coupled with a packet input unit, a packet output unit, and the memory and is configured to receive requests from the packet input unit and receive requests from the packet output unit, and choose among the requests to send to the memory on a second memory bus.

A network application executing on a host system provides a list of application buffers in host memory stored in a queue to a network services processor coupled to the host system. The application buffers are used for storing data transferred on a socket established between the network application and a remote network application executing in a remote host system. Using the application buffers, data received by the network services processor over the network is transferred between the network services processor and the application buffers. After the transfer, a completion notification is written to one of the two control queues in the host system. The completion notification includes the size of the data transferred and an identifier associated with the socket. The identifier identifies a thread associated with the transferred data and the location of the data in the host system.

Method and apparatus for reducing host overhead in a socket server implementation

A network transport layer accelerator accelerates processing of packets so that packets can be forwarded at wire-speed. To accelerate processing of packets, the accelerator performs pre-processing on a network transport layer header encapsulated in a packet for a connection and performs in-line network transport layer checksum insertion prior to transmitting a packet. A timer unit in the accelerator schedules processing of the received packets. The accelerator also includes a free pool allocator which manages buffers for storing the received packets and a packet order unit which synchronizes processing of received packets for a same connection.

Growing community of inventors

Hudson, MA, United States of America

Robert A Sanzone