Abstract: In described examples, a coherent memory system includes a central processing unit (CPU) and first and second level caches. The memory system can include a pipeline for accessing data stored in one of the caches. Requestors can access the data stored in one of the caches by sending requests at a same time that can be arbitrated by the pipeline.

Abstract: A memory system includes a host circuit and a memory circuit. The host circuit controls a bandwidth of a command-address signal based on data driving cycle information. The memory circuit performs an input/output operation based on the command-address signal.

Abstract: A networking device/storage device direct write system includes a chassis that houses a Solid State Drive (SSD) storage device coupled to a Smart Network Interface Controller (SmartNIC) networking device. The SmartNIC networking device receives data via a network, stores the data in a SmartNIC buffer memory subsystem that is included in the SmartNIC networking device, and then perform a Direct Memory Access (DMA) operation to transfer the data stored in the SmartNIC buffer memory subsystem to an addressable memory subsystem that is included in the SSD storage device. If the addressable memory subsystem in the SSD storage device is a volatile memory subsystem, the SmartNIC networking device then transmits a persistent storage instruction to the SSD storage device that causes the SSD storage device to transfer the data stored in the addressable memory subsystem to a persistent memory subsystem in the SSD storage device.

Abstract: Systems, devices, computer program products, and methods include determining by a connection manager that a connected device can be enhanced by an asymmetrical multi-lane link. The connection manager can use system parameters, including bandwidth information, to switch a direction of one or more lanes of the multi-lane link. The connection manager can use register setting instructions to change register settings on the host side and on the device side to switch the direction of one or more lanes of the multi-lane link.

Abstract: Bandwidth for information transfers between devices is dynamically changed to accommodate transitions between power modes employed in a system. The bandwidth is changed by selectively enabling and disabling individual control links and data links that carry the information. During a highest bandwidth mode for the system, all of the data and control links are enabled to provide maximum information throughout. During one or more lower bandwidth modes for the system, at least one data link and/or at least one control link is disabled to reduce the power consumption of the devices. At least one data link and at least one control link remain enabled during each low bandwidth mode. For these links, the same signaling rate is used for both bandwidth modes to reduce latency that would otherwise be caused by changing signaling rates. Also, calibration information is generated for disabled links so that these links may be quickly brought back into service.

Abstract: Aspects of the embodiments are directed to a port comprising hardware to support the multi-lane link, the link comprising a lane that comprises a first differential signal pair and a second differential signal pair. Link configuration logic, implemented at least in part in hardware circuitry, can determine that the port comprises hardware to support one or both of receiving data on the first differential signal pair or transmitting data on the second differential signal pair, and reconfigure the first differential signal pair to receive data with the second differential signal pair or reconfigure the second differential signal pair to transmit data with the first differential signal pair; and wherein the port is to transmit data or receive data based on reconfiguration of one or both the first differential signal pair and the second differential signal pair.

Abstract: An arrangement is described used to throttle data in a connected computer device having a device configured to transmit and receive data from a host, the device comprising, a device controller configured to interact with at least memory array and a data transfer throttling arrangement, the data transfer throttling arrangement configured to measure a bandwidth threshold for the device controller and pass data through the device controller when a bandwidth of the device controller is one of at and below a threshold.

Abstract: According to one embodiment, in a first state, a control circuit determines, based on first information and second information, information on a request that includes a setting of a transmission circuit of a host to be set as an initial setting in a second state. The first state is a state of communicating with a host at a first communication speed conforming to a first specification. The second state is a state of communicating with the host at a second communication speed conforming to a second specification. The second communication speed is different from the first communication speed. The first information is information on a request of a setting of the transmission circuit of the host. The second information is information on a quality of a signal received by a reception circuit, which has been transmitted from the transmission circuit of the host.

Abstract: An electronic device includes a transmit buffer, a receive buffer, a communication port, and a controller. The controller is to: communicate with a target device via a data link established via the communication port; determine a throughput ratio between the transmit buffer and the receive buffer; in response to a determination that the throughput ratio is above a threshold, transmit a request to the target device to change an aspect of the data link, where the request includes a payload size indicating an amount of data to be transmitted from the electronic device to the target device; and in response to receiving a grant message associated with the request, increase an amount of transmit lanes within the data link from the electronic device to the target device.

Abstract: Systems, apparatuses, and methods related to channel architecture for memory devices are described. Various applications can access data from a memory device via a plurality of channels. The channels can be selectively enabled or disabled based on the behavior of the applications. For instance, an apparatus in the form of a memory system can include an interface coupled to a controller and a plurality of channels. The controller can be configured to determine an aggregate amount of bandwidth used by a plurality of applications accessing data from a memory device coupled to the controller via the plurality of channels and disable one or more channels of the plurality of channels based, at least in part, on the aggregate amount of bandwidth used by the plurality of applications.

Abstract: A configurable logic platform may include a physical interconnect for connecting the platform to a processor, a reconfigurable logic region having logic blocks configured based on configuration data, a configuration port for applying configuration data to the reconfigurable logic region, a reconfiguration logic function accessible via transactions of the physical interconnect and in communication with the configuration port, the reconfiguration logic function providing restricted access to the configuration port from the physical interconnect, and an interface function accessible via transactions of the physical interconnect and providing an interface to the reconfigurable logic region which allows information to be transmitted over the physical interconnect and prevents the reconfigurable logic region from directly accessing the physical interconnect. The reconfiguration logic function may be implemented in the reconfigurable logic region.

Abstract: Aspects of the invention include receiving a request for data. The request is received from a computing element implementing a first bus protocol, and the data is accessible via a reduced instruction set computer (RISC) system implementing a plurality of bus protocols. A type of the received request is determined. A bus protocol is selected from the plurality of bus protocols based at least in part on the type of the received request. The received request is translated into a format that is compatible with the selected bus protocol and transmitted to the RISC system. Data is received from the RISC system in response to transmitting the translated request.

Abstract: Method and apparatus for improving continuous read operations with expanded serial interface are provided. In one aspect, a device comprises: a memory configured to store data; a buffer configured to receive data from outside of the device and transfer the received data to the memory; a plurality of input pins configured to be coupled to an expanded serial peripheral interface (xSPI); and a processor configured to: select a slave device, through the xSPI, from a plurality of slave devices, send instruction data to the slave device for data reading, receive data, through the xSPI, from the selected slave device, and receive a signal on a data strobe line of the xSPI and determine data reading operations based on the received signal.

Abstract: A port of a computing device includes protocol circuitry to implement a particular interconnect protocol, where the protocol circuitry is to generate a set of ordered sets defined according to the particular interconnect protocol. The set of ordered sets is generated for a link to couple a first device to a second device and the set of ordered sets comprises link information for the link. Translation layer circuitry is provided to: generate, from the set of ordered sets, at least one data structure to comprise at least a portion of the link information, and cause the data structure to be sent from the first device to the second device on the link in lieu of the set of ordered sets.

Abstract: A processor includes a plurality of first processing units. The processor further includes a direct memory access unit coupled to a first processing unit of the plurality of first processing units. The processor further includes a data storage unit. The processor further includes a second processing unit adapted to process data transferred from the data storage unit, wherein the direct memory access unit is configured to transfer data stored in a memory to the data storage unit, the second processing unit is separate from the plurality of first processing units and the direct memory access unit, and the first processing unit of the plurality of first processing units and the second processing unit are configured to work in parallel. The processor further includes a first register configured to store data corresponding to an interrupt request related to the second processing unit or the data storage unit.

Abstract: An integrated circuit includes a plurality of tiles. Each tile includes a processor, a switch including switching circuitry to forward data over data paths from other tiles to the processor and to switches of other tiles, and a switch memory that stores instruction streams that are able to operate independently for respective output ports of the switch. Also disclosed is a direct memory access (DMA) scheme in which sizes of DMA transfers are limited according to whether a cache miss has occurred.

Abstract: Embodiments may include systems and methods for communication including a communication port with a first lane and a second lane, a first power controller and a second power controller coupled to the communication port. The first power controller is to control, at a first time instance, the first lane to operate in a first power state selected from a first set of power states for the first lane. The second power controller is to control, at a second time instance, the second lane to operate in a second power state selected from a second set of power states for the second lane, wherein the first power state is different from the second power state. Other embodiments may be described and/or claimed.

Abstract: An interconnect controller includes a data link layer controller coupled to a transaction layer, wherein the data link layer controller selectively receives data packets from and sends data packets to the transaction layer, and a physical layer controller coupled to the data link layer controller and to a communication link. The physical layer controller selectively operates at a first predetermined link speed. The physical layer controller has an enhanced speed mode, wherein in response to performing a link initialization, the interconnect controller queries a data processing platform to determine whether the enhanced speed mode is permitted, performs at least one setup operation to select an enhanced speed, wherein the enhanced speed is greater than the first predetermined link speed, and subsequently operates the communication link using the enhanced speed.

Abstract: The present invention is directed to a system and method which employ two memory access paths: 1) a cache-access path in which block data is fetched from main memory for loading to a cache, and 2) a direct-access path in which individually-addressed data is fetched from main memory. The system may comprise one or more processor cores that utilize the cache-access path for accessing data. The system may further comprise at least one heterogeneous functional unit that is operable to utilize the direct-access path for accessing data. In certain embodiments, the one or more processor cores, cache, and the at least one heterogeneous functional unit may be included on a common semiconductor die (e.g., as part of an integrated circuit). Embodiments of the present invention enable improved system performance by selectively employing the cache-access path for certain instructions while selectively employing the direct-access path for other instructions.

Abstract: According to one embodiment, a data buffer is described. The data buffer comprises a first input/output circuit configured to receive and provide a first signal encoded according to a first communications protocol, a second input/output circuit configured to receive and provide a second signal encoded according to a second communications protocol, and a conversion circuit coupled to the first and second input/output circuits and configured to convert the first signal to the second signal and to convert the second signal to the first signal.

Abstract: Methods and apparatus to implement efficient communications between components of computing systems are disclosed. An example apparatus includes a message generator to: add a first value associated with a first field of a message to a shift register based on a first push operation, the message including multiple fields, at least two of the fields having different bit widths; and add a second value associated with a second field of the message to the shift register based on a second push operation, the second value to be adjacent the first value in the shift register in accordance with a structure of the message. The example apparatus further includes a communications interface to transmit content stored in the shift register to a hardware device via a bus having a width corresponding to a width of the shift register, the content including the message.

Abstract: A random-access memory (RAM) controller is connected with multiple memories. The random-access memory controller selectively boots at least one memory of the multiple memories based on booting-related information about the multiple memories.

Abstract: A queue manager apparatus converts inbound commands of a first width into scalar format commands to be queued in a command queue. Furthermore, the queue manager converts the scalar format commands residing in the command queue into outbound commands of a second width for transmission. Converting inbound commands to scalar format commands and then converting the scalar format commands to a target width for transmission allows the queue manager to advantageously provide efficient and programmable command transmission between arbitrary processing units, regardless of potentially mismatched native command widths.

Abstract: In one embodiment, an apparatus includes: a transmitter to send first data to a device coupled to the apparatus via a physical link; a receiver to receive second data from the device via the physical link; and a control circuit to control the transmitter to send the first data at a first effective rate during a link activation interval of a data transfer interval and to control the receiver to receive the second data at a second effective rate during the link activation interval, the second effective rate different than the first effective rate. Other embodiments are described and claimed.

Abstract: Disclosed is a device for maintaining consistency between a host system cache and a main memory in a general-purpose computing system equipped with a hardware accelerator for processing main memory data. The device for maintaining data consistency between a hardware accelerator and a host system, which is at least temporarily implemented by a computer, includes a determination unit responsible for determining whether an address which the hardware accelerator should access is present in a cache, and a processing unit responsible for selectively performing write-back on data corresponding to the address when the address is present in the cache based on the determined result.

Abstract: Memory devices with half-width data path or data buses clocked by double-pumped strobe signals are disclosed herein. The methods and devices may employ a single delay chain (e.g., a column access strobe (CAS) chain) to perform the double-pumped operations. The delay chain may include multiple delay elements that may generate one or two pulses based on the double-pumped strobe signals. Methods for interfacing, such as read and write methods are also disclosed.

Abstract: An error correcting system is provided. The error correcting system includes an error correcting code (ECC) circuit and a control circuit. The ECC circuit is configured to encode input data received from M input terminals to generate encoded data in response to a write operation, and output the encoded data. The input data includes write data associated with the write operation, and the encoded data includes the input data and associated parity data. The control circuit is coupled to at least one of the M input terminals. When the write operation is directed to a memory device having a data bit width less than M bits, the write data is inputted to a portion of the M input terminals, the control circuit is configured to provide reference data to another portion of the M input terminals, and the write data and the reference data serve as the input data.

Abstract: Described herein are various embodiments of reducing dynamic power consumption within a processor device. One embodiment provides a technique for dynamic link width reduction based on the instantaneous throughput demand for client of an interconnect fabric. One embodiment provides for a parallel processor comprising an interconnect fabric including a dynamic bus module to configure a bus width for a client of the interconnect fabric based on throughput demand from the client.

Abstract: A PCIe capable semiconductor device includes; ports respectively configured to transmit and receive data in a PCIe environment, and a PCIe controller configured to set a link between the PCIe capable semiconductor device and another PCIe capable semiconductor device. The link includes at least one lane implemented over at least one of the ports. The PCIe controller includes a link training and status state machine (LTSSM) configured to perform a first lane number negotiation according to a first ordering of the ports and a second lane number negotiation according to a second ordering of the ports different from the first ordering of the ports, and determine an optimized link width for the link according to the results of the first lane number negotiation and the second lane number negotiation.

Abstract: There is provided an integrated circuit comprising at least one logic path, comprising a plurality of sequential logic elements operably coupled into a scan chain to form at least one scan chain under test, at least one IR drop sensor operably coupled to the integrated circuit power supply, operable to output a first logic state when a sensed supply voltage is below a first predefined value and to output a second logic state when the sensed supply voltage is above the first predefined value, at least one memory buffer operably coupled to a scan test data load-in input and a scan test data output of the at least one scan chain under test, and control logic operable to gate logic activity including the scan shift operation inside the integrated circuit for a single cycle when the at least one IR drop sensor outputs the first logic state and to allow normal scan test flow when the at least one IR drop sensor outputs the second logic state.

Abstract: A memory controller interfaces with one or more memory devices having configurable width data buses and configurable connectivity between data pins of the memory devices and data pins of the memory controller. Upon initialization of the memory devices, the memory controller automatically discovers the connectivity configuration of the one or more memory devices, including both individually selected and jointly selected devices. After discovering connectivity of the connected devices, the memory controller configures the memory devices according to the discovered connectivity and assigns unique addresses to jointly selected devices.

Abstract: Mechanisms and techniques for configuring a configurable slave device using a high speed serial link where a different number of lanes of the high speed serial link are used to send data between the slave device and a master device, depending on whether the slave device is in configuration mode or in normal operations mode, are provided.

Abstract: A modem identifies an idle condition associated with a data network to be accessed and determines an opportunity to enter a first one of a set of low power device states based on the idle condition, where the set of low power device states further includes a second low power device state, and a host device consumes less power in the first low power device state than in the second low power device state. A notification is sent to an application processor of the host device that the modem is to enter a sleep state, where the notification identifies the first low power device state, and a low power link state is entered corresponding to the first low power device state based on a signal from the application processor. The low power link state applies to a link coupling a communications processor of the modem to the application processor.

Abstract: In a very long instruction word (VLIW) central processing unit instructions are grouped into execute packets that execute in parallel. A constant may be specified or extended by bits in a constant extension instruction in the same execute packet. If an instruction includes an indication of constant extension, the decoder employs bits of a constant extension instruction to extend the constant of an immediate field. Two or more constant extension slots are permitted in each execute packet, each extending constants for a different predetermined subset of functional unit instructions. In an alternative embodiment, more than one functional unit may have constants extended from the same constant extension instruction employing the same extended bits. A long extended constant may be formed using the extension bits of two constant extension instructions.

Abstract: An operating system monitors a performance metric of a direct memory access (DMA) engine on an I/O adapter to update a translation table used during DMA operations. The translation table is used during a DMA operation to map a virtual address provided by the I/O adapter to a physical address of a data page in the memory modules. If the DMA engine is being underutilized, the operating system updates the translation table such that a virtual address maps to physical address corresponding to a memory location in a more energy efficient memory module. However, if the DMA engine is over-utilized, the operating system may update the translation table such that the data used in the DMA engine is stored in memory modules that provide quicker access times—e.g., the operating system may map virtual addresses to physical addresses in DRAM rather than phase change memory.

Abstract: Differential data access. A method for storing and reading data elements to and from a memory is provided. The method includes storing a data element as a base word in a first precision, storing at least one delta word including additional information related to a second precision version of the stored data element, and reading the base word and the at least one delta word of the stored data element to access the data element in the second precision.

Abstract: Differential data access. A method for storing and reading data elements to and from a memory is provided. The method includes storing a data element as a base word in a first precision, storing at least one delta word including additional information related to a second precision version of the stored data element, and reading the base word and the at least one delta word of the stored data element to access the data element in the second precision.

Abstract: A memory access control device includes: a memory configured to hold data from a host device; and a processor coupled to the memory, the processor: detects an overlapping portion of addresses of data transfer sources of a plurality of memory access requests; merges, in a case where the overlapping portion is detected, read accesses to the data transfer sources for the overlapping portion of the plurality of memory access requests collectively to generate a merged memory access request; executes a data transfer in accordance with the merged memory access request; and instructs writing of data transferred in the data transfer to a plurality of addresses of data transfer destinations of the plurality of memory access requests.

Abstract: Embodiments including systems, methods, and apparatuses associated providing an interface between a north complex and a south complex of a system on a chip (SoC). In embodiments, the north complex may include a microcontroller in an input signal requirement. A power-on control block may be coupled with the microcontroller, and the power-on control block may be configured to receive a control signal from a component of the south complex, and alter the control signal based at least in part on the input signal requirement of the microcontroller.

Abstract: A memory subsystem is provided, including a memory controller integrated circuit (IC), a memory bus and a memory IC, all which use fewer signals than common DDR type memory of the same peak bandwidth. Using no more than 22 switching signals, the subsystem can transfer data over 3000 Megabytes/second across the bus interconnecting the ICs. Signal count reduction is attained by time-multiplexing address/control commands onto at least some of the same signals used for data transfer. A single bus signal is used to initiate bus operation, and once in operation the single signal can transfer addressing and control information to the memory IC concurrent with data transfer via a serial protocol based on 16 bit samples of this single bus signal. Bus bandwidth can be scaled by adding additional data and data strobe IO signals. These additional data bus signals might be used only for data and data mask transport.

Abstract: A data management method and system for migrating a data volume from a source array to a destination array includes receiving an indication of a source volume to migrate from the source array to the destination array wherein the source volume comprises a clustered volume associated with a plurality of cluster hosts. A path-flip is performed by the host multipath module, to switch the paths between source and destination array. This switch is done in a seamless manner, without disrupting host I/O. The path-flip includes transferring associated metadata from source to destination and can be performed even in clustered and/or multi-host environments. Sanity timers are built-in to ensure that the path-flip completes within a stipulated time-frame and, if not, the path-flip process is aborted and I/O is resumed without any disruption.

Abstract: A data transfer system, a method of data transfer and a corresponding transmitter and receiver are disclosed. A communication protocol between the transmitter and receiver is defined using a set of valid transmission states for communication from the transmitter to the receiver and a set of valid acknowledgement states for transmission from the receiver to the transmitter. A Hamming distance between patterns of zeroes, and between patterns of ones, in valid states of each of these sets is at least one and the transmitter is arranged to transition between a number of transmission states in response to the reception of an acknowledgement state from the receiver which matches a transmission state it has sent to the receiver on a request bus. A communication protocol which is robust across a multi-voltage and/or clock domain interface is thus provided.

Abstract: There is provided a method for saving fault wave data, in which when a fault wave generated in a system is saved, a size of wave data is calculated, and the saving of the wave data is performed according to the size of the wave data, so that it is possible to implement the use of spaces of a buffer and a memory and the simultaneous saving of wave data simultaneously or subsequently generated. Accordingly, it is possible to minimize a delay in the saving of the wave data. Also, it is possible to record all accurate wave data even when consecutive faults occur. Also, it is possible to efficiently use the space of the memory.

Abstract: Described is an apparatus which comprises: a first electrical path comprising at least one driver and receiver; and a second electrical path comprising at least one driver and receiver, wherein the first and second electrical paths are to receive a same input signal, wherein the first electrical path and the second electrical path are parallel to one another and have substantially the same propagation delays, and wherein the second electrical path is enabled during a first operation mode and disabled during a second operation mode.

Abstract: A processor with an accumulator. An event is selected to produce one or more selected events. A reset signal to the accumulator is generated responsive to the selected event. Responsive to the reset signal, the accumulator is reset to zero or another initial value while avoiding breaking pipelined execution of the processor.

Abstract: A method of sending information between first and second modules connected by a signal bus comprises generating a clock signal in the first module, and imposing the clock signal on a first line of the bus. A first pattern of bit values is transmitted from the second module to the first module on a second line of the bus, during first half-periods of each period of said clock signal. A second pattern of bit values is transmitted from the first module to the second module on the second line of the bus, during second half-periods of each period of said clock signal, wherein the second half-periods of each period of said clock signal are different from the first half-periods of each period of said clock signal. Information can then be transmitted from the first module to the second module by altering the second pattern of bit values; and information can be transmitted from the second module to the first module by altering the first pattern of bit values.

Abstract: In one aspect, a processor has a register file, a private Level 1 (L1) cache, and an interface to a shared memory hierarchy (e.g., an Level 2 (L2) cache and so on). The processor has a Load Store Unit (LSU) that handles decoded load and store instructions. The processor may support out of order and multi-threaded execution. As store instructions arrive at the LSU for processing, the LSU determines whether a counter, from a set of counters, is allocated to a cache line affected by each store. If not, the LSU allocates a counter. If so, then the LSU updates the counter. Also, in response to a store instruction, affecting a cache line neighboring a cache line that has a counter that meets a criteria, the LSU characterizes that store instruction as one to be effected without obtaining ownership of the effected cache line, and provides that store to be serviced by an element of the shared memory hierarchy.

Abstract: The embodiments relate to dynamically allocating lanes of a computer bus. A computer system is configured with a plurality of connectors in communication with a module, with each connector configured to receive a respective adapter. The module detects a presence of each primary and backup adapter present, and controls an initial allocation of lanes to each detected primary adapter for maximizing adapter functionality. After the initial allocation and in response to detecting a failure of at least one primary adapter, the module dynamically switches lanes from the failed adapter to at least one of the one or more remaining primary adapters and the backup adapter.

Abstract: Described are memory modules that support dynamic point-to-point extensibility using fixed-width memory die. The memory modules include data-width translators that allow the modules to vary the effective width of their external memory interfaces without varying the width of the internal memory interfaces extending between the translators and associated fixed-width dies. The data-width translators use a data-mask signal to selectively prevent memory accesses to subsets of physical addresses. This data masking divides the physical address locations into two or more temporal subsets of the physical address locations, effectively increasing the number of uniquely addressable locations in a given module. Reading temporal addresses in write order can introduce undesirable read latency. Some embodiments reorder read data to reduce this latency.

Abstract: An integrated circuit may have configurable storage blocks. A configurable storage block may include a memory array, a processing circuit, and a configurable control circuit. The configurable storage block may receive an instruction which may be decoded in the control block to identify a command. The command may be associated with a pre-defined sequence of operations that the control block executes by directing the memory array to perform memory access operations and the processing circuit to execute data processing operations. These data processing operations may be executed on data retrieved during memory access operations, data received subsequent to receiving the instruction, or previously computed data. The processed data may be provided for further processing outside the configurable storage block or stored in the memory array. The configurable storage block may further have delay blocks to allow for delayed memory access to the memory array.