Abstract: The disclosed device includes a data path having multiple transmission drivers. The device also includes a controller that is configured to tune each of the transmission drivers to a signal speed of a reference transmission driver. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: An electronic device having a frame for coupling a plurality of thermal management devices to the printed circuit board is provided. The electronic device includes a first chip package mounted to the PCB and a second chip package mounted to the PCB. The frame is secured to the PCB, and the frame has a first aperture disposed over the first chip package and a second aperture disposed over the second chip package. The plurality of thermal management devices coupled to the frame includes a first thermal management device contacting an IC die of the first chip package through the first aperture and a second thermal management device contacting an IC die of the second chip package through the second aperture.

Abstract: Embodiments herein describe creating multiple packet fragments from a large data chunk that, for example, exceeds a maximum transmission unit (MTU) supported by a network. In one embodiment, a network interface controller or card (NIC) receives a direct memory access (DMA) from a connected host to transmit an IP packet or data using remote direct memory access (RDMA) technologies. The NIC can evaluate the data chunk associated with the DMA request and determine whether it exceeds the MTU for the network. Assuming it does, the NIC determines how many fragments to divide the data chunk into, and can fragment any portion of the data at flexible packet/payload offsets. The NIC can then retrieve the data chunk from host memory fragment-by-fragment, rather than reading the data chunk all at once, generating headers for the fragments, and then transmit them as packet fragments.

Abstract: A processing system executes a specialized wavefront, referred to as a “garbage collecting wavefront” or GCWF, to identify and deallocate resources such as, for example, scalar registers, vector registers, and local data share space, that are no longer being used by wavefronts of a workgroup executing at the processing system (i.e., dead resources). In some embodiments, the GCWF is programmed to have compiler information regarding the resource requirements of the other wavefronts of the workgroup and specifies the program counter after which there will be a permanent drop in resource requirements for the other wavefronts. In other embodiments, the standard compute wavefronts signal the GCWF when they have completed using resources. The GCWF sends a command to deallocate the dead resources so the dead resources can be made available for additional wavefronts.

Abstract: A memory includes a read clock state machine and a read clock driver circuit. The read clock state machine has a first input for receiving a read command signal, a second input for receiving a read clock mode signal, and an output for providing a drive enable signal. The read clock driver circuit has an output for providing a read clock signal in response to a clock signal when the drive enable signal is active. When the read clock mode signal indicates a read-only mode, the read clock state machine starts toggling the read clock signal during a read preamble period before a data transmission of a first read command, and continues toggling the read clock signal for at least a read postamble period following the data transmission of the first read command.

Abstract: Methods, devices, and systems for managing performance of a processor having multiple compute units. An effective number of the multiple compute units may be determined to designate as having priority. On a condition that the effective number is nonzero, the effective number of the multiple compute units may each be designated as a priority compute unit. Priority compute units may have access to a shared cache whereas non-priority compute units may not. Workgroups may be preferentially dispatched to priority compute units. Memory access requests from priority compute units may be served ahead of requests from non-priority compute units.

Abstract: A network of thermal sensors can be integrated within a semiconductor chip in a manner effective to provide local temperature monitoring and dynamic control of an associated device or system. The thermal sensors can include small area thermal ring oscillators located proximate to the core of a central processing unit (CPU), for example, and can be disposed on the chip at locations based on a designed output power density and attendant thermal gradients encountered during operation. In certain implementations, the presently-disclosed sensor configuration can be used to measure deviation from set threshold temperatures. Closed-loop control can be implemented to mitigate performance loss while adjusting the clock speed of the CPU independent of the system management unit.

Abstract: A method for collapsing operations into super operations in a computing system includes dispatching a super operation corresponding to a collapsible sequence of operations to a scheduler, performing a lookup in a super operation table for the collapsible sequence of operations in response to the super operation being picked from the scheduler, and multi-pumping the collapsible sequence of operations to a pipe operationally coupled to the scheduler. For example, the multi-pumped collapsible sequence of operations may then be sequentially executed by an execution unit. The collapsible sequence of operations may be identified as collapsible according to a set of rules.

Abstract: A disclosed method for cooling an integrated circuit can include positioning a first blower to draw air through a first air inlet in a first side of a housing. The method can additionally include positioning a second blower to draw air through a second air inlet in a second side of the housing that is opposite the first side, wherein the second blower is mirror-stacked with the first blower. Various other methods and systems are also disclosed.

Abstract: Supply chain security for chiplets is described. In accordance with the described techniques, a chiplet manufacturing interface obtains first test results, and stores an encrypted version of the first test results in a database accessible by the chiplet manufacturing interface and a chiplet integration interface. The chiplet integration interface obtains second test results from at least one chiplet, retrieves, from the database, the encrypted version of the first test results, decrypts the encrypted version of the first test results to obtain a first hash of the first test results, and selectively integrates the at least one chiplet into an integrated circuit based on a comparison of the first test results and the second test results and a comparison of the first hash and a second hash of the second test results generated by the chiplet integration interface.

Abstract: Systems, methods, and techniques dynamically utilize load balancing for workgroup assignments between a group of shader engines by a command processor of a graphics processing unit (GPU). Based on one or more commands received for execution, a plurality of workgroups is generated for assignment to a plurality of shader engines for processing, each shader engine including a respective quantity of active compute units. Each workgroup of the plurality of workgroups is dynamically assigned to a respective shader engine for execution based at least in part on indications of available resources respectively associated with each of the shader engines. In various embodiments, the indications of available resources may include physical parameters regarding each shader engine, as well as current status information regarding the processing of workgroups assigned to each shader engine.

Abstract: An approach is provided for implementing register based single instruction, multiple data (SIMD) lookup table operations. According to the approach, an instruction set architecture (ISA) can support one or more SIMD instructions that enable vectors or multiple values in source data registers to be processed in parallel using a lookup table or truth table stored in one or more function registers. The SIMD instructions can be flexibly configured to support functions with inputs and outputs of various sizes and data formats. Various approaches are also described for supporting very large lookup tables that span multiple registers.

Abstract: A voltage regulator includes an input power supply node, an output regulated power supply node, a flipped voltage follower circuit, and a compensation capacitor. The flipped voltage follower circuit includes an output transistor configured as a common-source amplifier circuit. A source terminal of the output transistor is coupled to the input power supply node and a drain terminal of the output transistor is coupled to the output regulated power supply node. The flipped voltage follower circuit includes a folded cascode feedback circuit. The folded cascode feedback circuit includes a folding node. The folded cascode feedback circuit is configured to receive an output regulated voltage on the output regulated power supply node and to provide a feedback signal to a gate terminal of the output transistor. The compensation capacitor is coupled to the output regulated power supply node and the folding node.

Abstract: A method for dimensioning a land grid array pad can include forming an initial landing area of a land grid array pad, wherein the initial landing area is dimensioned to cause at least a majority of a landing surface of one or more socket pins to land off of the initial landing area prior to actuation of one or more sockets including the one or more socket pins. The method can also include forming a final landing area of the land grid array pad, wherein the final landing area is dimensioned to maintain electrical contact with the landing surface of the one or more socket pins after actuation of the one or more sockets including the one or more socket pins. Various other methods and systems are also disclosed.

Abstract: Error alert encoding for improved error mitigation is described. In one or more implementations, a system includes a processor configured to receive an encoded signal indicating a type of an error detected in a memory, and output one or more mitigation commands to mitigate the type of the error detected in the memory based on the encoded signal. In one or more implementations, a memory system includes a memory and a buffer. The buffer is configured to output an encoded signal indicating a type of an error detected in the memory.

Abstract: A first frame of a video stream rendered at a first resolution is obtained. A second frame of the video stream upscaled to a second higher resolution is also obtained. The first plurality of pixels is upscaled to the second resolution. The upsampling generates upsampled color data for the upsampled first plurality of pixels. The upsampled color data is accumulated with a second set of color data associated with a second plurality of pixels defining the second frame to generate final color data for the upsampled first plurality of pixels. Color data of the second set of color data associated with a pixel lock contributes more to the final color data than corresponding color data of the upsampled color data. The upsampled first plurality of pixels is stored with the final color data as an upscaled frame representing the first frame at the second resolution.

Abstract: A method and a processing device for performing rendering are disclosed. The method comprises generating a base hierarchy tree comprising data representing a first object and generating a second hierarchy tree representing a second object comprising shared data of the base hierarchy tree and the second hierarchy tree and difference data. The method further comprises storing the difference data in the memory without storing the shared data, and generating an overlay hierarchy tree comprising the shared data, the difference data, and indication information indicating nodes of the overlay hierarchy tree that comprise the difference data. The method further comprises rendering the first object using the data stored for the base hierarchy tree, and rendering the second object using any one or a combination of the shared data, the difference data, and the indication information.

Abstract: A method and apparatus of managing memory includes storing a first memory page at a shared memory location in response to the first memory page including data shared between a first virtual machine and a second virtual machine. A second memory page is stored at a memory location unique to the first virtual machine in response to the second memory page including data unique to the first virtual machine. The first memory page is accessed by the first virtual machine and the second virtual machine, and the second memory page is accessed by the first virtual machine and not the second virtual machine.

Abstract: A memory controller includes a command queue stage for storing decoded memory access requests, a first arbiter operable to select first decoded memory access requests for a first pseudo channel from the command queue stage, and a second arbiter operable to select second decoded memory access requests for a second pseudo channel from the command queue stage. Each of the first arbiter and the second arbiter is operable to select a first streak of a first type of accesses, and to change to selecting a second streak of a second type of accesses in response to the first arbiter and the second arbiter meeting a cross mode condition.

Abstract: A link controller includes a Peripheral Component Interconnect Express (PCIe) physical layer circuit for coupling to a communication link and providing a data path over the communication link, a first data link layer controller which operates according to a PCIe protocol, and a second data link layer controller which operates according to a non-PCIe protocol. A multiplexer-demultiplexer selectively connects both data link layer controllers to the PCIe physical layer circuit. A protocol translation circuit is coupled between the multiplexer-demultiplexer and the second data link layer controller, the protocol translation circuit receiving traffic data from the second data link layer controller in a non-PCIe format, encapsulating the non-PCIe format in a PCIe format, and passing traffic data to the multiplexer-demultiplexer circuit.