Abstract: A method and apparatus for processing data received and transmitted on a TCP connection is described. An offload unit processes received data for which a special case does not exist, to produce payload data, which is uploaded directly to application memory. The offload unit partially processes received data for which a special case does exist and uploads the partially processed received data to a buffer stored in system memory. The partially processed received data is then further processed by a TCP stack to produce payload data, which is copied to application memory.

Abstract: A method and apparatus for storing and accessing connection information is described. A delegated connection table stores an entry for each connection delegated by a TCP stack for processing by an offload unit. A portion of the delegated connection table storing receive buffer information is accessed by the TCP stack without disrupting receive or transmit traffic. The offload unit offloads some TCP processing from a host processor and processes data received on connections not stored in the delegated connection table while accepting incoming data.

Abstract: A system and methods for wireless computing devices to become mesh member nodes within a self-configuring mesh network includes mechanisms for neighbor discovery and sharing of a common topology database including mesh topology and mesh network information. Each mesh node may use the topology database to determine optimized routing paths within the mesh network. Mesh member nodes are configured to detect and communicate topology changes and measured mesh network attributes to other members of the self-configuring wireless network.

Abstract: One embodiment of the present invention is a method for identifying a faulty NIC in a team of NICs using a minimum number of packets transmitted in a round-robin scheme. Relative to prior art schemes, the disclosed method advantageously reduces the number of keep-alive packets necessary to monitor the NICs within a team and, to the extent there is a failure, enables the faulty NIC to be identified.

Abstract: A system and methods for wireless computing devices to become mesh member nodes within a self-configuring mesh network includes mechanisms for neighbor discovery and sharing of a common topology database including mesh topology and mesh network information. Each mesh node may use the topology database to determine optimized routing paths within the mesh network. Mesh member nodes are configured to detect and communicate topology changes and measured mesh network attributes to other members of the self-configuring wireless network.

Abstract: An alternate checksum option for transmitting TCP frame data is used that does not require computing a TCP checksum. Instead, an integrity check value is computed and placed in the ESP authentication portion near the end of the frame. Transmission of a portion of the TCP frame data may begin before all of the TCP frame data is received by a TCP checksum offload device because the integrity check value is inserted at the end of the frame. Furthermore, the integrity check value is computed using a static key or a dynamic key may be computed to cover a greater portion of the frame compared with ESP authentication specified by the IPsec standard.

Abstract: A system and methods for wireless computing devices to become mesh member nodes within a self-configuring mesh network includes mechanisms for neighbor discovery and sharing of a common topology database including mesh topology and mesh network information. Each mesh node may use the topology database to determine optimized routing paths within the mesh network. Mesh member nodes are configured to detect and communicate topology changes and measured mesh network attributes to other members of the self-configuring wireless network.

Abstract: A firewall identifies unsolicited messages having an address resolution for a network protocol address different than cached address resolution information. The accuracy of the unsolicited messages is checked by requesting network elements to report address resolution information for the network protocol address.

Abstract: A method and apparatus for filtering a packet on a connection within a computing system. In one embodiment, the method includes: receiving the packet; delegating the packet to an offload unit for filtering the packet; and determining, by the offload unit, whether the connection is a delegated connection.

Abstract: A method of setting up a delegated connection for processing by an offload unit is described. The method comprises establishing a TCP connection and determining whether or not to delegate the TCP connection for processing by the offload unit, producing a delegated connection, and setting up the delegated connection by creating a delegated connection table entry. When frames are received on the delegated connection by the offload unit, the offload unit determines if user buffers are available. When user buffers are available, the offload unit uploads payload data to the user buffers. When user buffers are not available, the offload unit uploads a portion of the payload data to a buffer allocated in Operating System memory space.

Abstract: A method for autonomously and dynamically optimizing transmission power of an endpoint in a wireless network includes the step of monitoring a received signal strength, a received signal quality and a transmission error rate of a signal transmitted between an access point in the wireless network and the endpoint at a given transmission power and transmission speed. The method also includes the steps of reducing the transmission power when the received signal strength, the received signal quality and the transmission error rate are at respectively acceptable operating levels and then monitoring the transmission error rate of the signal transmitted at the reduced transmission power level. The method further includes the step of adjusting one of the transmission power or the transmission speed based on whether the transmission error rate of the signal transmitted at the reduced transmission power is still at its respective acceptable operating level.

Abstract: A method and apparatus for editing outbound frames and generating acknowledgements for a TCP connection is described. Acknowledgements are automatically generated and included in outbound frames during data transmissions with minimal processor intervention.

Abstract: Method and apparatus for reassembling a packet from fragments. The fragments of the packet are obtained by a device, such as a firewalling device. The fragments are sorted according to the packet and order of the fragments. The fragments are stored in association with the packet and in order. Once all the fragments to reconstitute the packet have been collected, the fragments are assembled in order to reconstitute the packet.

Abstract: Method and apparatus for routing and bridging are described. An address from a packet is obtained and used to find an index in a table. If the address is found in the table, the index stored in association with the address obtained. The index is stored in a data structure associated with the packet.

Abstract: A method and apparatus for filtering a packet on a connection within a computing system. In one embodiment, the method includes: receiving the packet; delegating the packet to an offload unit for filtering the packet; and determining, by the offload unit, whether the connection is a delegated connection.

Abstract: Method and apparatus for deflecting connection flooding attacks. Specifically, the stateful firewall allows all connection attempts to flow into the destination host, but monitors the connection attempts to ensure that only legitimate connections are allowed. If the firewall detects that a connection is half-open for longer than a certain timer threshold, it will instruct the destination host to tear down the half-open connection, thereby freeing up resources in the destination host for other connection attempts. The timer threshold can be dynamically adjusted if a connection flooding attack is detected.

Abstract: Embodiments related to wireless access point repeaters are disclosed.

Abstract: An alternate checksum option for transmitting TCP frame data is used that does not require computing a TCP checksum. Instead, an integrity check value is computed and placed in the ESP authentication portion near the end of the frame. Transmission of a portion of the TCP frame data may begin before all of the TCP frame data is received by a TCP checksum offload device because the integrity check value is inserted at the end of the frame. Furthermore, the integrity check value is computed using a static key or a dynamic key may be computed to cover a greater portion of the frame compared with ESP authentication specified by the IPsec standard.

Abstract: A method for autonomously and dynamically optimizing transmission power of an endpoint in a wireless network includes the step of monitoring a received signal strength, a received signal quality and a transmission error rate of a signal transmitted between an access point in the wireless network and the endpoint at a given transmission power and transmission speed. The method also includes the steps of reducing the transmission power when the received signal strength, the received signal quality and the transmission error rate are at respectively acceptable operating levels and then monitoring the transmission error rate of the signal transmitted at the reduced transmission power level. The method further includes the step of adjusting one of the transmission power or the transmission speed based on whether the transmission error rate of the signal transmitted at the reduced transmission power is still at its respective acceptable operating level.

Abstract: A method for autonomously and dynamically optimizing transmission power of an endpoint in a wireless network includes the step of monitoring the signal quality associated with data transfers between an access point in the wireless network and the endpoint at a first transmission power and a first transmission speed. The method also includes the steps of determining whether the signal quality is acceptable and adjusting one of the first transmission power or the first transmission speed based on whether the signal quality is acceptable.