Abstract: Packets received non-contiguously from a network are processed by a network interface controller by coalescing received packet payload into receive buffers on a receive buffer queue and writing descriptors associated with the receive buffers for a same flow consecutively in a receive completion queue. System performance is optimized by reusing a small working set of provisioned receive buffers to minimize the memory footprint of memory allocated to store packet data. The remainder of the provisioned buffers are in an overflow queue and can be assigned to the network interface controller if the small working set of receive buffers is not sufficient to keep up with the received packet rate. The receive buffer queue can be refilled based on either timers or when the number of buffers in the receive buffer queue is below a configurable low watermark.

Abstract: An apparatus, a method and a computer program for generating data packets according to a transport protocol from an application buffer comprising a plurality of data streams is provided. The apparatus comprises an input circuit configured to receive metadata comprising at least one of information about data packet types supported by the transport protocol, information about an offset and a length of the supported data packet types, and information about possible stream header start positions, possible payload start positions and possible offsets in the data streams. Further, the apparatus comprises a parsing circuit configured to identify offsets in an application buffer as possible segmentation points based on the metadata, to segment the application buffer at the possible segmentation points into segments for data packets, and to generate data packets according to the transport protocol based on the segments.

Abstract: Generally, this disclosure provides systems, methods and computer readable media for management of sockets and device queues for reduced latency packet processing. The method may include maintaining a unique-list comprising entries identifying device queues and an associated unique socket for each of the device queues, the unique socket selected from a plurality of sockets configured to receive packets; busy-polling the device queues on the unique-list; receiving a packet from one of the plurality of sockets; and updating the unique-list in response to detecting that the received packet was provided by an interrupt processing module. The updating may include identifying a device queue associated with the received packet; identifying a socket associated with the received packet; and if the identified device queue is not on one of the entries on the unique-list, creating a new entry on the unique-list, the new entry comprising the identified device queue and the identified socket.

Abstract: A lower latency communications path is provided with checkpointing to verify a packet transmission is permitted. When a client initiates communication with the lower latency path, the client uses the unique tag in a packet to be transmitted. The network interface of the transmitter device can verify that the packet is an acceptable format and formed in an accepted manner. If the packet is verified, the network interface can transmit the packet to a next node according to the end-to-end configuration. The next node can read the packet's unique tag and verify the packet is an accepted format using context information associated with the unique tag. Each device in the path can perform a verification based on the tag in the packet before allow progress to a next prescribed step. A destination device can perform a verification based on the tag in the packet before allow progress to the destination receive queue.

Abstract: Technologies for I/O device virtualization include a computing device with an I/O device that includes a physical function, multiple virtual functions, and multiple assignable resources, such as I/O queues. The physical function assigns an assignable resource to a virtual function. The computing device configures a page table mapping from a virtual function memory page located in a configuration space of the virtual function to a physical function memory page located in a configuration space of the physical function. The virtual function memory page includes a control register for the assignable resource, and the physical function memory page includes another control register for the assignable resource. A value may be written to the control register in the virtual function memory page. A processor of the computing device translates the virtual function memory page to the physical function memory page using the page mapping. Other embodiments are described and claimed.

Abstract: Techniques are described that can be used to enable a transfer of an operating system from one machine to another. The transfer permits the operating system to be available to the target machine at buffers that are accessible to one or more application or other logic. In some implementations, information related to an operating system migration is stored in a buffer that is accessible to an application that is to use the information and thereby avoids a copy of such information from an intermediate buffer to an application buffer.

Abstract: Disclosed herein are systems, devices, and methods for simultaneous multithreading (SMT) with context associations. For example, in some embodiments, a computing device may include: one or more physical cores; and SMT logic to manage multiple logical cores per physical core such that operations of a first computing context are to be executed by a first logical core associated with the first computing context and operations of a second computing context are to be executed by a second logical core associated with the second computing context, wherein the first logical core and the second logical core share a common physical core.

Abstract: Technologies for control plane separation at a network interface controller (NIC) of a compute device configured to transmit, by a resource of the compute device, commands to a physical function managed by a network interface controller (NIC) of the compute device. The NIC is further to establish a data plane separate from a control plane, wherein the control plane comprises one of the trusted control path and the untrusted control path. Additionally, the resource is configured to transmit the commands via one of the trusted control path or the untrusted control path based on a trust level associated with the physical function. Other embodiments are described herein.

Abstract: A network switch circuit associated with a communication network is disclosed. The network switch circuit comprises a switching circuit configured to receive a switch configuration request comprising a configuration command comprising information on a switching rule associated with the network switch circuit; and a configuration template comprising a real packet header comprising real values for one or more fields required for the execution of the switching rule. In some embodiments, the configuration template is indicative of a real network data packet format. In some embodiments, the switching circuit is further configured to process the switch configuration request, in order to apply the switching rule to an incoming network data packet.

Abstract: The present disclosure is directed to a unified device interface for a multi-bus system. In at least one embodiment, a system may comprise more than one data bus. Each data bus may be to convey data between an operating system (OS) and at least one device in the system, wherein a plurality of driver instances may facilitate interaction between the OS and a device via one or more of the data buses. In one embodiment, a main driver instance may be determined from the plurality of driver instances to present the device to the OS and coordinate operation of other driver instances. The other driver instances may map addresses in the memory of processing entities corresponding to each of the data buses and report these mappings to the main driver instance. Alternatively, a supervisory driver may be loaded to present the device and to control operation of the driver instances.

Abstract: Techniques are described that can be used to enable a transfer of an operating system from one machine to another. The transfer permits the operating system to be available to the target machine at buffers that are accessible to one or more application or other logic. In some implementations, information related to an operating system migration is stored in a buffer that is accessible to an application that is to use the information and thereby avoids a copy of such information from an intermediate buffer to an application buffer.

Abstract: Disclosed is a mechanism for maintaining a single lookup table entry for symmetric/bidirectional flows. Multiple recipes are stored for each flow. A recipe is employed to select address information from an incoming packet header based on the packet's direction. The address information and an index are employed to generate a lookup key to find the single lookup table entry with the pertinent switching information. The recipe further indicates action pointers in the lookup table entry that are specific to direction. The action pointers point to an address in an action table that contains instructions for actions that are applied to the packet during switching based on the packet's direction.

Abstract: Methods for implementing mini-mezzanine Open Compute Project (OCP) plug-and-play Network PHY Cards and associated apparatus. In accordance with one aspect, the MAC (Media Access Channel) and PHY (Physical) layer functions in one or more communication protocol stacks are split between a MAC block in a Platform Controller Hub (PCH) or processor SoC and a PHY card installed in a mezzanine slot of a platform and including one or more ports. During platform initialization operations, configuration parameters are read from the PHY card including a PHY card ID, and a corresponding configuration script is selected and executed to configure the PHY card for use in the platform. The configuration parameters are also used to enumerate PCIe devices associated with physical functions and ports supported by the PHY card.

Abstract: Examples are disclosed for exchanging a key between an input/output device for network device and a first processing element operating on the network device. Data having a destination associated with the first processing element may be received by the input/output device. The exchanged key may be used to encrypt the received data. The encrypted data may then be sent to a buffer maintained at least in part in a memory for the network device. The memory may be arranged to enable sharing of the buffer with at least a second processing element operating on the network device. Examples are also disclosed for the processing element to receive an indication of the storing of the encrypted data in the buffer. The processing element may then obtain the encrypted data from the buffer and decrypt the data using the exchanged key.

Abstract: A collection of techniques allow for the detection of covert malware that attempts to hide its existence on a system by leveraging both trusted hardware event counters and the particular memory addresses (as well as the sequences of such addresses) of the instructions that are generating the suspected malicious activity. By monitoring the address distribution's specific patterns over time, one can build a behavioral model (i.e., “fingerprint”) of a particular process—and later attempt to match suspected malicious processes to the stored behavioral models. Whenever the actual measured behavior of a suspected malicious process fails to match said stored behavioral models, the system or system administrator may attempt to perform rehabilitative actions on the computer system to locate and remove the malware hiding on the system.

Abstract: The present disclosure is directed to a unified device interface for a multi-bus system. In at least one embodiment, a system may comprise more than one data bus. Each data bus may be to convey data between an operating system (OS) and at least one device in the system, wherein a plurality of driver instances may facilitate interaction between the OS and a device via one or more of the data buses. In one embodiment, a main driver instance may be determined from the plurality of driver instances to present the device to the OS and coordinate operation of other driver instances. The other driver instances may map addresses in the memory of processing entities corresponding to each of the data buses and report these mappings to the main driver instance. Alternatively, a supervisory driver may be loaded to present the device and to control operation of the driver instances.

Abstract: Generally, this disclosure describes techniques for buffer management based on link status. A host platform may include a Baseboard Management Controller (BMC) and a network controller that includes a buffer used by the BMC. When a network controller is in a lower power link state, the BMC may attempt to send data to the link partner which causes the network controller to transition out of the low power state. However, this transition may take longer than the buffer's ability to buffer the incoming flow from the BMC. Accordingly, to avoid the need for larger buffer space, a buffer manager is used to provide flow control management of the buffer based on link status.

Abstract: An embodiment may include circuitry that may be capable of performing operations that may include generating, at least in part, at least one message to announce that at least one network node (1) is requesting, at least in part, that one or more transmissions to the at least one network node be postponed, at least in part, and/or (2) is entering, at least in part after issuance of the at least one message, a relatively lower power state relative to a relatively higher power state. Additionally or alternatively, the operations may include, in response, at least in part, to the at least one message, postponing, at least in part, at least one intermediate node at least one transmission (received by the at least one intermediate node) to the at least one network node. Many alternatives, variations, and/or modifications are possible without departing from this embodiment.

Abstract: Generally, this disclosure provides systems, methods and computer readable media for management of sockets and device queues for reduced latency packet processing. The method may include maintaining a unique-list comprising entries identifying device queues and an associated unique socket for each of the device queues, the unique socket selected from a plurality of sockets configured to receive packets; busy-polling the device queues on the unique-list; receiving a packet from one of the plurality of sockets; and updating the unique-list in response to detecting that the received packet was provided by an interrupt processing module. The updating may include identifying a device queue associated with the received packet; identifying a socket associated with the received packet; and if the identified device queue is not on one of the entries on the unique-list, creating a new entry on the unique-list, the new entry comprising the identified device queue and the identified socket.

Abstract: The present disclosure is directed to a unified device interface for a multi-bus system. In at least one embodiment, a system may comprise more than one data bus. Each data bus may be to convey data between an operating system (OS) and at least one device in the system, wherein a plurality of driver instances may facilitate interaction between the OS and a device via one or more of the data buses. In one embodiment, a main driver instance may be determined from the plurality of driver instances to present the device to the OS and coordinate operation of other driver instances. The other driver instances may map addresses in the memory of processing entities corresponding to each of the data buses and report these mappings to the main driver instance. Alternatively, a supervisory driver may be loaded to present the device and to control operation of the driver instances.