Abstract: Various aspects provide for optimizing memory mappings associated with network nodes. For example, a system can include a first network node and a second network node. The first network node generates a memory page request in response to an invalid memory access associated with a virtual central processing unit of the first network node. The second network node receives the memory page request in response to a determination that the second network node comprises a memory space associated with the memory page request. The first network node also maps a memory page associated with the memory page request based on a set of memory page mappings stored by the first network node.

Abstract: Techniques efficiently improve circuit design to reduce its sensitivity to random device variation. A characterizer component can identify a subset of cells for an integrated circuit that can be representative of respective other cells of a set of cells. The characterizer component can analyze the representative cells of the subset to generate a variation profile, and can map the representative cells to physical cells used in the design of the circuit. A cell library comprising cells that are usable, have limited usage, and/or have general usage can be generated based on analysis results from the mapped cells. The circuit can be reconstructed based on the list of available cells using the cell library. The reconstructed circuit can be analyzed, and in case of a cell(s) violating a constraint, the cell(s) can be modified or enhanced to achieve target performance criteria.

Abstract: Techniques efficiently improve an integrated circuit design by simultaneously analyzing timing paths of the circuit design. A design management component can access data relating to the integrated circuit design from a design database. The design management component can perform a static timing analysis of the integrated circuit design and generate a timing path distribution, filtered analytics, and/or a probability density function associated with the integrated circuit design, wherein all of the timing paths of the integrated circuit design can be evaluated. The design management component can determine a modification to make to a cell, device, interconnection between cells or devices, or another element(s) of the integrated circuit design, based at least in part on the static timing analysis, the timing path distribution, the filtered analytics, and/or the probability density function, to generate a modified integrated circuit design, in accordance with defined design criteria.

Abstract: Systems and methods that facilitate multi-word atomic operation support for systems on chip are described. One method involves: receiving an instruction associated with a calling process, and determining a first memory width associated with execution of the instruction based on an operator of the instruction and a width of at least one operand of the instruction. The instruction can be associated with an atomic operation. In some embodiments, the instruction contains a message having a first field identifying the operator and a second field identifying the operand.

Abstract: Providing for address translation in a virtualized system environment is disclosed herein. By way of example, a memory management apparatus is provided that comprises a shared translation look-aside buffer (TLB) that includes a plurality of translation types, each supporting a plurality of page sizes, one or more processors, and a memory management controller configured to work with the one or more processors. The memory management controller includes logic configured for caching virtual address to physical address translations and intermediate physical address to physical address translations in the shared TLB, logic configured to receive a virtual address for translation from a requester, logic configured to conduct a table walk of a translation table in the shared TLB to determine a translated physical address in accordance with the virtual address, and logic configured to transmit the translated physical address to the requester.

Abstract: An error injection system of a built-in self-repairable memory system renders the redundant spare columns of the repairable memory accessible to built-in self-test (BIST) read and write operations. To this end, the error injection system selectively injects fault data at one or more locations of the main memory during a BIST sequence, causing the BIST controller to issue a repair instruction that allocates one or more spare columns as replacement memory areas for the presumed faulty main memory locations. Thereafter, BIST read/write operations directed to the main memory locations will be performed on the allocated spare columns, thereby allowing the spare columns to be validated as part of the BIST. Injection of fault data to the main memory locations in this manner can also facilitate validation of the built-in self-repair logic by verifying the repair instruction codes that are generated in response to the injected faults.

Abstract: Provided is an end-to-end flow control management for a system on chip interface. As tokens are injected into agents arranged in a computer network the input point for the token is dynamically changed such that tokens are not always injected into the same agent. Additionally or alternatively, as tokens are injected into a token ring, the tokens are initially not activated until a predetermined event occurs (e.g., after a specific number of hops). Additionally or alternatively, also provided is a free pool manager that can keep at least some high priority slots available by consuming lower priority slots first.

Abstract: Proving for a framework for propagating priorities to a memory subsystem in a computing system environment is disclosed herein. By way of example, a memory access handler is provided for managing memory access requests and determining associated priorities. The memory access handler includes logic configured for propagating memory requests and the associated priorities to lower levels of a computer hierarchy. A memory subsystem receives the memory access requests and the priorities.

Abstract: Various aspects provide for detecting voltage droops. For example, a system can include a voltage calibrator component and a comparator component. The voltage calibrator component can convert a first supply voltage associated with a power distribution network of an integrated circuit to a second supply voltage via a resistance ladder circuit. The comparator component can generate a comparison output signal in response to a determination that a comparison between the second supply voltage and a reference voltage satisfies a defined criterion.

Abstract: A self-referenced on-die voltage droop detector generates a reference voltage from the supply voltage of an integrated circuit's power distribution network, and compares this reference voltage to the transient supply voltage in order to detect voltage droops. The detector responds to detected occurrences of voltage droop with low latency by virtue of being located on-die. Also, by generating the reference voltage from the integrated circuit's power domain rather than using a separate reference voltage source, the detector does not introduce noise and distortion associated with a separate power domain.

Abstract: Techniques efficiently assist in performing write operations in memories with resistive bit lines. A memory can comprise memory cells associated with respective word lines and bit lines. A write assist component can be associated with a subset of the memory cells associated with a bit line. Configuration of the write assist component can be based on the type of transistors employed by write circuits associated with the memory cells. During a write operation, the write assist component adds an additional current path to the ground, or the power supply, or both, at or in proximity to the far end of the write bit line when an appropriate write polarity is applied to the bit line by the driver at the other end of the bit line. This mitigates the effects of resistance of the bit line, which mitigates IR loss of the write signal.

Abstract: A mapping technique sets coalescing latency values for computing systems that use multiple data queues having a shared base timer. A computing system having at least one receive queue and at least one transmit queue receives user-provided coalescing latency values for the respective queues, and converts these user-provided latencies to coalescing latency hardware register values as well as a base timer register value for the shared base timer. The hardware register values for the coalescing latencies together with the shared base timer register value determine the coalescing latencies for the respective queues. This mapping technique allows a user to conveniently set coalescing latencies for multi-queue processing systems while shielding the user settings from hardware complexity.

Abstract: Various aspects facilitate implementing a memory translation table associated with key-based indexing. A table component is configured for generating a memory translation table and a key component is configured for allocating a key associated with a memory access based on a virtual address and a set of access permissions. A descriptor component is configured for generating a descriptor associated with the memory translation table that comprises at least the set of access permissions and a portion of the key.

Abstract: Techniques efficiently assist in performing write operations in memories with resistive bit lines. A memory can comprise memory cells associated with respective word lines and bit lines. A write assist component can be associated with a subset of the memory cells associated with a bit line. Configuration of the write assist component can be based on the type of transistors employed by write circuits associated with the memory cells. During a write operation, the write assist component adds an additional current path to the ground, or the power supply, or both, at or in proximity to the far end of the write bit line when an appropriate write polarity is applied to the bit line by the driver at the other end of the bit line. This mitigates the effects of resistance of the bit line, which mitigates IR loss of the write signal.

Abstract: A memory bus comprising a plurality of latches arranged sequentially between a source node and a destination node of a channel of the memory bus; and a pulse generator. The pulse generator is operable to generate a sequence of pulses, each sequential pulse to be simultaneously received by the plurality of latches. A pulse is generated for each edge of a clock signal. A first latch of the plurality of latches is operable to pass on a first data sample while a first pulse is received by the first latch of the plurality of latches. A second latch of the plurality of latches is operable to pass on a second data sample towards the first latch of the plurality of latches while the first pulse is simultaneously received by the first and second latches of the plurality of latches.

Abstract: A semiconductor chip allows for a selected amount of on-die decoupling capacitance to be connected to a very-large-scale integrated circuit (VLSI) system after the circuit design is complete. The semiconductor chip comprises an integrated circuit disposed on a packaging substrate, and a power distribution network that is electrically connectable to the integrated circuit via a programmable connectivity array via the packaging substrate.

Abstract: Various embodiments provide for a system that integrates 64 bit ARM cores and a switch on a single chip. The RISC style processors use highly optimized sets of instructions rather than the specialized set of instructions found in other architectures (e.g., x86). The system also includes multiple high bandwidth ports that enable multi-ported virtual appliances to be built using a single chip. The virtual appliances are software implemented versions of the physical appliances that are installed with servers to provide network services such routing and switching services, firewall, VPN, SSL, and other security services, as well as load balancing. The virtual appliances are implemented in software and the system can add new virtual appliances, or change the functions performed by existing virtual appliances flexibly without having to install or remove physical hardware.

Abstract: Various embodiments provide for a system that prefetches data from a main memory to a cache and then evicts unused data to a lower level cache. The prefetching system will prefetch data from a main memory to a cache, and data that is not immediately useable or is part of a data set which is too large to fit in the cache can be tagged for eviction to a lower level cache, which keeps the data available with a shorter latency than if the data had to be loaded from main memory again. This lowers the cost of prefetching useable data too far ahead and prevents cache trashing.

Abstract: A mapping technique sets coalescing latency values for computing systems that use multiple data queues having a shared base timer. A computing system having at least one receive queue and at least one transmit queue receives user-provided coalescing latency values for the respective queues, and converts these user-provided latencies to coalescing latency hardware register values as well as a base timer register value for the shared base timer. The hardware register values for the coalescing latencies together with the shared base timer register value determine the coalescing latencies for the respective queues. This mapping technique allows a user to conveniently set coalescing latencies for multi-queue processing systems while shielding the user settings from hardware complexity.

Abstract: Proving for a framework for propagating priorities to a memory subsystem in a computing system environment is disclosed herein. By way of example, a memory access handler is provided for managing memory access requests and determining associated priorities. The memory access handler includes logic configured for propagating memory requests and the associated priorities to lower levels of a computer hierarchy. A memory subsystem receives the memory access requests and the priorities.