Abstract: A system and method for virtualizing a hardware device node for a network interface circuit or other device. In a clone driver's configuration file, multiple virtual or clone nodes are defined. Their major numbers match the major number of the driver, but a virtual node's minor number for a Style 2 driver is calculated from the hardware driver's major number and the instance or port number of the corresponding hardware node. By creating multiple clone nodes, multiple paths through the protocol stack are created. Each one can be configured differently (e.g., for different MTU sizes), and each may be assigned a different address (e.g., MAC address). The separate addresses may be registered with hardware filters to filter incoming packets at the hardware level. A traffic classifier created in the device driver manages each virtual node's association with a hardware filter.
Abstract: Systems, devices and methods for improving network performance are disclosed. In particular, in one embodiment, a method of processing data packets in an electronic network is provided. The method includes receiving data packets at a receive stack of an electronic network device and creating a first packet from at least a portion of each of the set of data packets. The first packet includes an indicator that the first packet is a first packet type. The creation of the first packet includes saving a first header of the plurality of data packets, modifying the first header with information from each of the set of data packets, and concatenating the set of data packets.
Abstract: Systems, devices and methods for improving network performance are disclosed. In particular, in one embodiment, a method of processing data packets in an electronic network is provided. The method includes receiving data packets at a receive stack of an electronic network device and creating a first packet from at least a portion of each of the set of data packets. The first packet includes an indicator that the first packet is a first packet type. Additionally, the method includes passing the first packet to one of a bridge and IP layer of the receive stack. A first packet inhibitor at the one of the bridge and IP layer is bypassed upon recognizing the first packet as a first packet type. The first packet is routed to a transmit stack of the electronic network device from the one of the bridge and IP layer.
Abstract: Systems, devices and methods for improving network performance are disclosed. In particular, in one embodiment, a method of processing data packets in an electronic network is provided. The method includes receiving data packets at a receive stack of an electronic network device and creating a first packet from at least a portion of each of the set of data packets. The first packet includes an indicator that the first packet is a first packet type. Additionally, the method includes passing the first packet to one of a bridge and IP layer of the receive stack. A first packet inhibitor at the one of the bridge and IP layer is bypassed upon recognizing the first packet as a first packet type. The first packet is routed to a transmit stack of the electronic network device from the one of the bridge and IP layer.
Abstract: Systems, devices and methods for improving network performance are disclosed. In particular, in one embodiment, a method of processing data packets in an electronic network is provided. The method includes receiving data packets at a receive stack of an electronic network device and creating a first packet from at least a portion of each of the set of data packets. The first packet includes an indicator that the first packet is a first packet type. Additionally, the method includes passing the first packet to one of a bridge and IP layer of the receive stack. A first packet inhibitor at the one of the bridge and IP layer is bypassed upon recognizing the first packet as a first packet type. The first packet is routed to a transmit stack of the electronic network device from the one of the bridge and IP layer.
Abstract: Systems, devices and methods for improving network performance are disclosed. In particular, in one embodiment, a method of processing data packets in an electronic network is provided. The method includes receiving data packets at a receive stack of an electronic network device and creating a first packet from at least a portion of each of the set of data packets. The first packet includes an indicator that the first packet is a first packet type. Additionally, the method includes passing the first packet to one of a bridge and IP layer of the receive stack. A first packet inhibitor at the one of the bridge and IP layer is bypassed upon recognizing the first packet as a first packet type. The first packet is routed to a transmit stack of the electronic network device from the one of the bridge and IP layer.
Abstract: Systems, devices and methods for improving network performance are disclosed. In particular, in one embodiment, a method of processing data packets in an electronic network is provided. The method includes receiving data packets at a receive stack of an electronic network device and creating a first packet from at least a portion of each of the set of data packets. The first packet includes an indicator that the first packet is a first packet type. The creation of the first packet includes saving a first header of the plurality of data packets, modifying the first header with information from each of the set of data packets, and concatenating the set of data packets.
Abstract: A system and method for improving the efficiency with which data communications are passed through a protocol stack. Protocol layer modules within a protocol stack built upon the Streams framework are modified to establish tight embraces between adjacent modules. A tight embrace may be established by passing from one, upper, protocol layer module to its adjacent lower protocol layer module a pointer or reference to the upper module's functionality (e.g., rput) for reading a data communication into the upper module from the lower module. Similarly, the lower module passes the upper module a pointer or reference to its functionality (e.g., wput) for writing a data communication into the lower module from the upper module. After a tight embrace is established, the protocol layer modules can directly invoke each other's functionality, without incurring the overhead associated with the Streams framework's “canputnext” and “putnext” messages.
Abstract: An apparatus and method for expeditiously steering a received communication (e.g., packet, cell, frame) to one or more communication streams. When a new communication stream is opened on a network interface circuit or other Primary Point of Attachment (PPA), the protocol to which the stream is bound is used as a Service Access Point (SAP). A hierarchy of SAP nodes (e.g., a binary tree) is constructed, with each node corresponding to a unique SAP. Each node includes its corresponding SAP value, a count of the number of communication streams bound to the associated protocol, and a reference to those streams. When a communication is received, its SAP is extracted and the hierarchy is searched for a matching node. The communication is then forwarded to each referenced stream.
Abstract: A method and apparatus for facilitating transfer of packets from communication hardware to a host computing device or software. After receiving s set of packets at a communication interface, the system writes in a hybrid buffer a type II completion line configured to identify a payload buffer in which payloads of packets are stored. For each packet, the system writes a type I completion line configured to indicate the length of a packet's header in the hybrid buffer and the length and/or offset of the packet's payload in the payload buffer. A type 0 completion line indicates that no more packets are stored in the payload buffer. The system then signals the host computing device by configuring a single completion descriptor to identify the hybrid buffer in which the completion lines are stored. The host computing device reads the single completion descriptor to process the packets.
Abstract: A system and method are provided for using multiple transmit descriptor rings to transmit packets from a computer system. A device driver for a communication interface (e.g., a NIC) receives a packet (e.g., from an upper layer protocol), selects one of the multiple rings and places the packet on the ring. Because the rings are managed in a mutually exclusive manner, packets can be placed on more than one ring at the same time (e.g., by different processors), thus allowing them to be populated in parallel, rather than serially. To select a ring, a packet's destination address, destination port or other characteristic may be hashed, or a modulo of that characteristic over the number of rings may be calculated. Illustratively, all packets in one connection or flow are transmitted through the same ring.
Abstract: A system and method for describing multiple packets to a communication module or device without requiring descriptors for each packet. A device driver for the communication apparatus receives a set of packet headers and a set of packet payloads from an upper layer protocol. The headers may be received in a header buffer, and the payloads may be received in a separate data buffer. Or, the packets may be received, in assembled form, in a single buffer. The device driver configures a single descriptor to describe the multiple packets to the communication apparatus. The descriptor includes base address(es) of the buffer(s) in which the headers, payloads or assembled packets are stored, and the lengths of the individual headers, payloads or assembled packets. If the buffer contents are not packed, the descriptor may also include offsets of the headers, payloads or packets.
Abstract: A system and method for distributing a portion of the processing of a received packet among a plurality of service threads. When an ISR or similar process retrieves a packet from a communication interface via a receive descriptor ring, it places the packet on one of a plurality of service queues. Each queue has an associated service thread or process that initiates upper layer protocol processing for queued packets. The ISR may select a particular service queue based on the packet's communication flow or connection. Alternatively, the ISR may use a processor identifier provided by the communication interface to select a queue (e.g., in a multi-processor computer system). Or, other information provided by the interface may be used.
Abstract: An apparatus and method for expeditiously steering a received communication (e.g., packet, cell, frame) to one or more communication streams. When a new communication stream is opened on a network interface circuit or other Primary Point of Attachment (PPA), the protocol to which the stream is bound is used as a Service Access Point (SAP). A hierarchy of SAP nodes (e.g., a binary tree) is constructed, with each node corresponding to a unique SAP. Each node includes its corresponding SAP value, a count of the number of communication streams bound to the associated protocol, and a reference to those streams. When a communication is received, its SAP is extracted and the hierarchy is searched for a matching node. The communication is then forwarded to each referenced stream.
Abstract: A system and method for distributing a portion of the processing of a received packet among a plurality of service threads. When an ISR or similar process retrieves a packet from a communication interface via a receive descriptor ring, it places the packet on one of a plurality of service queues. Each queue has an associated service thread or process that initiates upper layer protocol processing for queued packets. The ISR may select a particular service queue based on the packet's communication flow or connection. Alternatively, the ISR may use a processor identifier provided by the communication interface to select a queue (e.g., in a multi-processor computer system). Or, other information provided by the interface may be used.
Abstract: A system and method for performing non-intrusive loopback testing in a communication device. When a loopback mode of testing is requested for the communication device (e.g., from a diagnostic application), and one or more communication streams are active or bound to the device, the streams are suspended instead of terminated. In a list of the active streams, maintained in the device's information data structure, a device driver or the application modifies each of the streams (e.g., by setting a flag). While a stream is suspended, any traffic for the stream is dropped. After the loopback testing is completed, the streams are reactivated.
Abstract: A method and apparatus for determining a communication device's loopback capabilities by querying a device driver of the device. The device's loopback capabilities identify locations (e.g., internal modules, protocol layers) in the device, and/or data rates, at which loopback testing may be performed. Instead of embedding those capabilities in a diagnostic application and modifying the application every time a device changes or is upgraded, the application queries the device driver. The device driver sends the application a data structure or message identifying the capabilities, including identifiers. The application specifies a loopback capability to be exercised by returning a corresponding identifier to the driver.
Abstract: A configuration state of a networked system is automatically adjusted to achieve optimal performance. The networked system is defined by a plurality of data exchange entities having one or more tunable parameters. Networked system performance is monitored and performance metrics are stored to a database. The performance metrics are analyzed along with network and system rules and constraints, and the particular configuration of the networked system. The tunable parameters are automatically adjusted to achieve optimal networked system performance.
Abstract: A system and method for reclaiming descriptors in a separate process (e.g., thread) from one in which packets are posted to a descriptor ring. The packet-posting process need not be interrupted to reclaim used descriptors, thereby reducing the delay in transmitting packets through a communication interface. The descriptor reclamation process, which may execute on a separate processor from the packet-posting process, may remain quiescent as long as the ring is less than half full or there are no descriptors to reclaim. Then, both processes may operate in parallel.
Type:
Application
Filed:
February 25, 2003
Publication date:
September 9, 2004
Inventors:
Francesco DiMambro, Charles Suresh, Jian Huang
Abstract: A kernel data transfer method and system for transmitting multiple packets of data in a single block of data presented by application programs to the kernel's network subsystem for processing in accordance with data transfer parameters set by the application program. The multi-packet transmit system includes logic that allows header information of the multiple packets of data to be generated in a single buffer and appended to a second buffer containing the data packets to be transmitted through the network stack. The multi-data transmit system allows a device driver to amortize the input/output memory management related overhead across a number of packets. With some assistance from the network stack, the device driver needs to only perform the necessary IOMMU operations on two contiguous memory blocks representing the header information and the data payload of multiple packets during each transmit call.
Type:
Grant
Filed:
May 9, 2003
Date of Patent:
March 31, 2009
Assignee:
Sun Microsystems, Inc.
Inventors:
Francesco R. DiMambro, Cahya Adi Masputra