Abstract: A graphics processing system with a data store includes processing units for processing tasks. A check unit forms a signature which is characteristic of an output from processing a task on a processing unit, and a fault detection unit compares signatures formed at the check unit. Each task is processed first and second times at the processing units to generate first and second processed outputs. The graphics processing system write outs the first processed output to the data store, reads back the first processed output from the data store and forms at the check unit a first signature characteristic of the first processed output as read back from the data store; forms at the check unit a second signature characteristic of the second processed output, compares the first and second signatures at the fault detection unit, and raises a fault signal if the signatures do not match.

Abstract: Methods and apparatus relating to loop driven region based frontend translation control for performant and secure data-space guided micro-sequencing are described. In an embodiment, Data-space Translation Logic (DTL) circuitry receives a static input and a dynamic input and generates one or more outputs based at least in part on the static input and the dynamic input. A frontend counter generates a count value for the dynamic input based at least in part on an incremented/decremented counter value and a next counter value from the DTL circuitry. The DTL circuitry is capable to receive a new dynamic input prior to consumption of the one or more outputs. Other embodiments are also disclosed and claimed.

Abstract: Systems and methods for checking program code are disclosed. The method includes executing a first check of the program code, and making a first determination about the first check. Metadata is stored based on making the first determination. A second check of the program code is run based on the metadata, and a result of the second check is stored in memory. The program code is executed based on the second check. In some embodiments, the method for checking program code includes running a simulation of a check of the program code by a processing device; storing a result of the simulation in memory; executing the check of the program code outside of the simulation; making a determination about the check outside of the simulation; retrieving a result stored in the memory based on the determination; and executing the program code based on the check.

Abstract: Electronic system level (ESL) design and verification of the present disclosure is utilized to provide an electronic simulation and modeling of function safety and fault management of an electronic device. A method for simulating a safety circuit includes providing an electronic architectural design to perform one or more functional behaviors of the electronic device in accordance with an electronic design specification. The method further includes modeling the safety circuit of the electronic architectural design and one or more other electronic circuits of the electronic architectural design that communicate with the safety circuit. The method further includes simulating, using the modeling, operation of the safety circuit while the electronic architectural design is performing the one or more functional behaviors. The method also includes determining whether the simulated operation of the safety circuit satisfies the electronic design specification.

Abstract: A computing system performs shared cache allocation to allocate cache resources to groups of tasks. The computing system monitors the bandwidth at a memory hierarchy device that is at a next level to the cache in a memory hierarchy of the computing system. The computing system estimates a change in dynamic power from a corresponding change in the bandwidth before and after the cache resources are allocated. The allocation of the cache resources are adjusted according to an allocation policy that receives inputs including the estimated change in the dynamic power and a performance indication of task execution.

Abstract: The present disclosure relates to a method for authenticating instructions and operands in an electronic system comprising a controller. The method includes extracting instructions and operands via a first circuit of the controller from at least a first memory internal to the controller using a matrix bus of the controller, collecting, on the matrix bus, via a second circuit internal to the controller, instructions and operands during their transmission to the first circuit, and generating a word representative of the instructions and operands.

Abstract: A processor with efficient instruction translation is shown, which uses a microcode device that has a first storage device with a micro-operation bit width. The first storage device stores a fast translation table that records micro-operations corresponding to a particular complex instruction. When determining that a received macro instruction is the particular complex instruction, an instruction translator operates a register alias table hardware to enable the microcode device to query the fast translation table to obtain and return the micro-operations corresponding to the particular complex instruction to the register alias table hardware, for running execution units of the processor.

Abstract: A processor includes a time counter, a vector coprocessor, and an extended vector register file for executing vector instructions and extending the data width of vector registers. The processor statically dispatches vector instructions with preset execution times based on a write time of a register in a coprocessor register scoreboard and a time counter provided to a vector execution pipeline.

Abstract: In some examples, a system includes a plurality of processors and a kernel scheduler. The kernel scheduler associates each respective processor of the plurality of processors with a collection of clusters, wherein each cluster of the collection of clusters represents a respective different subset of the plurality of processors, and the respective processor is a member of each cluster of the collection of clusters. For each corresponding cluster of the collection of clusters, the kernel scheduler maintains a data structure associated with a ready queue of the kernel scheduler, the data structure comprising elements representing thread priorities, wherein an element of the data structure is associated with an ordered list of threads in the ready queue.

Abstract: A processor includes a plurality of execution pipes and a distributed scheduler coupled to the plurality of execution pipes. The distributed scheduler includes a first queue to buffer instruction operations from a front end of an instruction pipeline of the processor and a plurality of second queues, wherein each second queue is to buffer instruction operations allocated from the first queue for a corresponding separate subset of execution pipes of the plurality of execution pipes. The distributed scheduler further includes a queue controller to select an allocation mode from a plurality of allocation modes based on whether at least one indicator of an imbalance at the distributed scheduler is detected, and further to control the distributed scheduler to allocate instruction operations from the first queue among the plurality of second queues in accordance with the selected allocation mode.

Abstract: A compiler device, for generating an instruction sequence to be executed by an arithmetic processing device, includes at least one memory and at least one processor. The at least one processor is configured to receive a first instruction sequence for a first process and a second instruction sequence for a second process to be executed after the first process; generate third instructions, each third instruction being generated by merging a first instruction included in the first instruction sequence and a second instruction included in the second instruction sequence; and generate a third instruction sequence by concatenating the third instructions, instructions included in the first instruction sequence that are not merged into the third instructions, and instructions other than the second instruction among the plurality of instructions included in the second instruction sequence that are not merged into the one or more third instructions.

Abstract: In an example embodiment, to improve performance, a solution is provided that removes the bottleneck associated with a single events table. Specifically, the solution provides for three events tables rather than one table. Two of the three tables are domain event outbox tables that are parallel and equivalent to each other. The third table is a domain event indication table, which records which of the domain event outbox tables is in a serve mode versus a maintain mode. At any point in time, one of the two domain event outbox tables is in serve mode while the other is in maintain mode. Serve mode is used for event insert, update, and query, while maintain mode is used for event purge.

Abstract: The present disclosure relates to a system and method for monitoring system calls to an operating system kernel. A performance monitoring unit is used to monitor system calls and to gather information about each system call. The information is gathered upon interrupting the system call and can include system call type, parameters, and information about the calling thread/process, in order to determine whether the system call was generated by malicious software code. Potentially malicious software code is nullified by a malicious code counter-attack module.

Abstract: An instruction execution method for a microprocessor is provided. The microprocessor includes a model specific register (MSR). And, the instruction execution method includes the following steps. A target instruction is received using an instruction cache. The target instruction is decoded using an instruction translator to determine whether the target instruction is a specific instruction is a specific instruction. When the target instruction is the specific instruction, a model specific register index of the target instruction is obtained to directly read or write the model specific register.

Abstract: An instruction configuration and execution method includes the following steps. A target instruction is received through an instruction cache. The target instruction is decoded by an instruction translator. It is determined whether the target instruction has the authority to read or write the model specific register in an unprivileged state. It is determined whether the model specific register index of the specific instruction corresponds to a specific model specific register, so as to order the microprocessor to perform an instruction serialization operation.

Abstract: A power supply and a power saving method thereof are provided. The power saving method includes: utilizing a monitoring circuit within the power supply to generate a monitoring signal according to operating loading of the power supply; and utilizing a control signal generating circuit within the power supply to generate a control signal according to the monitoring signal, to drive at least two switch transistors within a circuitry within the power supply, wherein the switch transistors are connected in parallel.

Abstract: Techniques are disclosed that relate to executing fused instructions. A processor may include a decoder circuit and a load/store circuit. The decoder circuit may detect a load/store instruction to load a value from a memory and detect a non-load/store instruction that depends on the value to be loaded. The decoder circuit may fuse the load/store instruction and the non-load/store instruction such that one or more operations that the non-load/store instruction is defined to perform are to be executed within the load/store circuit. The load/store circuit may receive an indication of the fused load/store and non-load/store instructions and then execute one or more operations of the load/store instruction and the one or more operations of the non-load/store instruction using a circuit included in the load/store circuit.

Abstract: A method of converting a data stored in a memory from a first format to a second format is disclosed. The method includes extending a number of bits in the data stored in a double data rate (DDR) memory by one bit to form an extended data. The method further includes determining whether the data stored in the DDR is signed or unsigned data. Moreover, responsive to determining that the data is signed, a sign value is added to the most significant bit of the extended data and the data is copied to lower order bits of the extended data. Responsive to determining that the data is unsigned, the data is copied to lower order bits of the extended data and the most significant bit is set to an unsigned value, e.g., zero. The extended data is stored in an on-chip memory (OCM) of a processing tile of a machine learning computer array.

Abstract: In one embodiment, a processor includes: a plurality of cores each comprising a multi-threaded core to concurrently execute a plurality of threads; and a control circuit to concurrently enable at least one of the plurality of cores to operate in a single-threaded mode and at least one other of the plurality of cores to operate in a multi-threaded mode. Other embodiments are described and claimed.

Abstract: A field programmable gate array (FPGA) including a configurable interconnect fabric connecting a plurality of logic blocks, the configurable interconnect fabric and the logic blocks being configured to implement a data masking circuit configured to: receive input data including data values at a plurality of indices of the input data; select between a data value of the data values and an alternative value using a masking multiplexer to generate masked data, the masking multiplexer being controlled by a mask value of a plurality of mask values at indices corresponding to the indices of the input data; and output the masked data. In some examples, the configurable interconnect fabric and the logic blocks are further configured to implement a mask generation circuit configured to generate the mask values. In some examples, the mask values are received from external memory.

Abstract: A method for executing new instructions is provided. The method is used in a processor and includes: receiving an instruction; when the received instruction is an unknown instruction, executing a conversion program by an operating system, wherein the conversion program executes the following steps: determining whether the received instruction is a new instruction; converting the received instruction into at least one old instruction when the received instruction is a new instruction; and executing the at least one old instruction.

Abstract: Implementing a camouflage of current traces generated by a hardware component having one or more set of digital elements defining a plurality of operational datapaths comprises adjusting (761) one or more working condition(s) of the hardware component, measuring (762) a reaction of the hardware component to the working condition(s) by a logic test circuit through processing data operations along a reference datapath having a minimum duration corresponding to at least the longest of the operational datapaths, and in response to detecting an error (763) along the reference datapath, modifying (764) the working condition(s) so that the error generated by the logic test circuit is cancelled. Applications to countermeasures to side-channel attacks.

Abstract: A system may determine that two instructions may be combined based on a processing power of the processor and a size of the instructions, fuse the two instructions into a pair, map the two instructions with a single register tag, write the register tag into a mapper with bits indicating that the register tag is for a first instruction of the two instructions, write the register tag into the mapper with bits indicating that the register tag is for a second instruction of the two instructions, write the fused instruction pair into an issue queue, issue the fused instruction pair to a vector-scalar transformation units (VSU), and execute the two instructions.

Abstract: A method for storing data in a storage system having solid-state memory is provided. The method includes determining portions of the solid-state memory that have a faster access rate and portions of the solid-state memory that have a slower access rate, relative to each other or to a threshold. The method includes writing data bits of erasure coded data to the portions of the solid-state memory having the faster access rate, and writing one or more parity bits of the erasure coded data to the portions of the solid-state memory having the slower access rate. A storage system is also provided.

Abstract: The technology disclosed herein determining one or more vulnerable instructions in workload code and determining one or more additional instructions to be inserted in the workload code based at least in part on a power model of a system bus of a processor, when a power model of a processor is dependent on an order of instructions of workload code, inserting the one or more additional instructions with dependency to the workload code to produce complementary power consumption of the system bus to power consumption of the system bus from executing the one or more vulnerable instructions; and when the power model is not dependent on the order of instructions of workload code, inserting the one or more additional instructions without dependency to the workload code to produce complementary power consumption of the system bus to power consumption of the system bus from executing the one or more vulnerable instructions.

Abstract: A method for a controller to execute a program comprising a sequence of functions on an accelerator with a pipelined architecture comprising a microcode buffer. The method comprises executing a function of the program as a sequence of operations, wherein the sequence of operations is represented by a sequence of templates, determining whether the template is non-colliding with previously inserted templates in the microcode buffer, determining whether data in local memory will be referenced before all previously inserted templates have taken effect, determining whether registers will be referenced before all previously inserted templates in the microcode buffer have taken effect, when it is determined that the template fits, that resources are available, that local data memory accesses will not collide, and that register accesses will not collide: creating a sequence of microcode instructions in the template, and inserting the template into the microcode buffer.

Abstract: A method for executing new instructions is provided. The method is used in a processor and includes: receiving an instruction; when the received instruction is an unknown instruction, the processor executes the following steps through a conversion program: determining whether the received instruction is a new instruction; and converting the received instruction into at least one old instruction when the received instruction is a new instruction; and simulating the execution of the received instruction by executing the at least one old instruction.

Abstract: A method performed by a solid state drive is described. The method includes, on a channel that internally couples a controller of the solid state drive to a storage device of the solid state drive, sending write data for a program operation to be performed by one of the storage device's logical units in separate chunks over the channel. The method also includes inserting higher priority traffic items of other logical units of the storage device in between the separate chunks.

Abstract: A superscalar processor has a thread mode of operation for supporting multiple instruction execution threads which are full data path wide instructions, and a micro-thread mode of operation where each thread supports two micro-threads which independently execute instructions. An executed instruction sets a micro-thread mode and an executed instruction sets the thread mode.

Abstract: A coprocessor with register renaming is disclosed. An apparatus includes a plurality of processors and a coprocessor respectively configured to execute processor instructions and coprocessor instructions. The coprocessor receives coprocessor instructions from ones of the processors. The coprocessor includes an array of processing elements and a result register set comprising storage elements respectively distributed within the array of processing elements. For a given member of the array of processing elements, a corresponding storage element is configured to store coprocessor instruction results generated by the given member. The result register set implements a plurality of contexts to store respective coprocessor states corresponding to coprocessor instructions received from different processors.

Abstract: A method for executing new instructions is provided. The method is used in a processor and includes: receiving an instruction; when the received instruction is an unknown instruction, executing a conversion program by an operating system, wherein the conversion program executes the following steps: determining whether the received instruction is a new instruction; converting the received instruction into at least one old instruction when the received instruction is a new instruction; and executing the at least one old instruction.

Abstract: Techniques are described for metadata processing that can be used to encode an arbitrary number of security policies for code running on a processor. Metadata may be added to every word in the system and a metadata processing unit may be used that works in parallel with data flow to enforce an arbitrary set of policies. In one aspect, the metadata may be characterized as unbounded and software programmable to be applicable to a wide range of metadata processing policies. Techniques and policies have a wide range of uses including, for example, safety, security, and synchronization. Additionally, described are aspects and techniques in connection with metadata processing in an embodiment based on the RISC-V architecture.

Abstract: The present disclosure includes a feedback framework that receives feedback for a component of an information technology platform. The component includes the feedback framework, the information technology platform, a software application, a web browser, a client device, a client instance, or a virtual server. The feedback framework obtains context information associated with the feedback. The context information includes a system log, a screenshot, a web address of a web browser of the client device, version information, and/or the like. The feedback framework also determines an identity of the component by executing a handler. The feedback framework then determines a notification to send based on the identity of the component, and sends the notification with the feedback and the context information.

Abstract: Disclosed herein are systems and method for instruction tightly-coupled memory (iTIM) and instruction cache (iCache) access prediction. A processor may use a predictor to enable access to the iTIM or the iCache and a particular way (a memory structure) based on a location state and program counter value. The predictor may determine whether to stay in an enabled memory structure, move to and enable a different memory structure, or move to and enable both memory structures. Stay and move predictions may be based on whether a memory structure boundary crossing has occurred due to sequential instruction processing, branch or jump instruction processing, branch resolution, and cache miss processing. The program counter and a location state indicator may use feedback and be updated each instruction-fetch cycle to determine which memory structure(s) needs to be enabled for the next instruction fetch.

Abstract: The technology disclosed relates to runtime execution of configuration files on reconfigurable processors with varying configuration granularity. In particular, the technology disclosed relates to a runtime logic that is configured to receive a set of configuration files for an application, and load and execute a first subset of configuration files in the set of configuration files and associated application data on a first reconfigurable processor. The first reconfigurable processor has a first level of configurable granularity. The runtime logic is further configured to load and execute a second subset of configuration files in the set of configuration files and associated application data on a second reconfigurable processor. The second reconfigurable processor has a second level of configurable granularity that is different from the first level of configurable granularity.

Abstract: An apparatus comprises capability checking circuitry 86 to perform a capability validity checking operation to determine whether use of a capability satisfies one or more use-limiting conditions. The capability comprises a pointer and pointer-use-limiting information specifying the one or more use-limiting conditions. The one or more use-limiting conditions comprise at least an allowable range of addresses for the pointer. In response to a capability write request requesting that a capability is written to a memory location associated with a capability write target address, when capability write address tracking is enabled, capability write address tracking circuitry 200 updates a capability write address tracking structure 100 based on the capability write target address.

Abstract: Techniques for system recovery using a failover processor are disclosed. A first processor, with a first instruction set, is configured to execute operations of a first type; and a second processor, with a second instruction set different from the first instruction set, is configured to execute operations of a second type. A determination is made that the second processor has failed to execute at least one operation of the second type within a particular period of time. Responsive to determining that the second processor has failed to execute at least one operation of the second type within the particular period of time, the first processor is configured to execute both the operations of the first type and the operations of the second type.

Abstract: A method of an escape reorder mode for neural network model compression, is performed by at least one processor, and includes determining whether a frequency count of a codebook index included in a predicted codebook is less than a predetermined value, the codebook index corresponding to a neural network. The method further includes, based on the frequency count of the codebook index being determined to be greater than the predetermined value, maintaining the codebook index, and based on the frequency count of the codebook index being determined to be less than the predetermined value, assigning the codebook index to be an escape index of 0 or a predetermined number. The method further includes encoding the codebook index, and transmitting the encoded codebook index.

Abstract: The present disclosure provides a chip fault diagnosis method, which includes: determining an interrupt flag of an interrupt flag register based on first data identifying an interrupt state in the interrupt flag register; and determining a fault state of chip interrupt corresponding to the interrupt flag based on the interrupt flag. By adopting the technical solution provided by the present disclosure, a fault of the interrupt can be diagnosed in time, and the interrupt can be processed in time.

Abstract: A method of converting a data stored in a memory from a first format to a second format is disclosed. The method includes extending a number of bits in the data stored in a double data rate (DDR) memory by one bit to form an extended data. The method further includes determining whether the data stored in the DDR is signed or unsigned data. Moreover, responsive to determining that the data is signed, a sign value is added to the most significant bit of the extended data and the data is copied to lower order bits of the extended data. Responsive to determining that the data is unsigned, the data is copied to lower order bits of the extended data and the most significant bit is set to an unsigned value, e.g., zero. The extended data is stored in an on-chip memory (OCM) of a processing tile of a machine learning computer array.

Abstract: Techniques are described for metadata processing that can be used to encode an arbitrary number of security policies for code running on a processor. Metadata may be added to every word in the system and a metadata processing unit may be used that works in parallel with data flow to enforce an arbitrary set of policies. In one aspect, the metadata may be characterized as unbounded and software programmable to be applicable to a wide range of metadata processing policies. Techniques and policies have a wide range of uses including, for example, safety, security, and synchronization. Additionally, described are aspects and techniques in connection with metadata processing in an embodiment based on the RISC-V architecture.

Abstract: Techniques for signal handling between programs associated with different addressing modes in a computer system are described herein. An aspect includes, based on a signal occurring during execution of a first program in a first runtime environment, wherein the first program and the first runtime environment are associated with a first addressing mode, invoking a first signal exit routine associated with the first addressing mode. Another aspect includes allocating a signal information area (SIA) by the first signal exit routine. Another aspect includes calling a second signal exit routine associated with a second addressing mode that is different from the first addressing mode with an address of the SIA. Another aspect includes allocating a mirror SIA by the second signal exit routine. Another aspect includes handling the signal, and resuming execution based on the handling of the signal.

Abstract: A simulation method and a simulation system are provided. The simulation system may be divided into an execution model and a processor model based on a JIT emulation engine. The execution model can call the JIT emulation engine to execute instructions, and obtain influence of instructions on a processor architectural status. The processor model may simulate an internal process of a target processor and determine whether to start/end a speculation. The execution model and the processor model may interact through a specific protocol. After the speculation is started, the simulation method may store an application running scene when the speculation is started, and redirect influence of speculation instructions on a memory to a memory snapshot. After the speculation is ended, the simulation method may also restore the application running scene to a status before the speculation is started, and cancel influence of the speculation instructions on the memory.

Abstract: An integrated circuit such as, for example a graphics processing unit (GPU), includes a dynamic power controller for adjusting operating voltage and/or frequency. The controller may receive current power used by the integrated circuit and a predicted power determined based on instructions pending in a plurality of processors. The controller determines adjustments that need to be made to the operating voltage and/or frequency to minimize the difference between the current power and the predicted power. An in-system reinforced learning mechanism is included to self-tune parameters of the controller.

Abstract: An instruction execution method includes the following steps: translating a macro-instruction into a first micro-instruction and a second micro-instruction, and marking first binding information on the first micro-instruction, and marking second binding information on the second micro-instruction; and simultaneously retiring the first micro-instruction and the second micro-instruction according to the first binding information and the second binding information. The first micro-instruction and the second micro-instruction are adjacent to one another in the micro-instruction storage space.

Abstract: Methods, apparatus and computer program products are provided for the visualization of software execution. One method comprises: generating, using a static code analysis, a data flow diagram illustrating a plurality of possible code execution paths of software code; marking, using an execution profiler, a particular execution path through the data flow diagram for a specific execution scenario; and enriching one or more nodes in the data flow diagram marked with the particular execution path for the specific execution scenario with information extracted from one or more application logs. The marked data flow diagram may be enriched, for example, by extracting a particular portion of a log file and tagging a corresponding portion of the data flow diagram with the extracted particular portion of the log file. The extracted log information may comprise, for example, a timestamp, a log category, a software module name, a function name and/or contextual information.

Abstract: A non-volatile memory device having at least one non-volatile flash memory formatted with physical addresses to read and write data that is organized into blocks of data, wherein the blocks of data are organized into pages of data, and wherein the pages of data are organized into cells of data. The non-volatile memory device includes a non-volatile memory controller to direct read and write requests to the non-volatile flash memory for the storage and retrieval of data. The non-volatile memory controller includes a flash translation layer to correlate read and write requests for data having a logical address between the reading and writing the data to physical address location of the non-volatile flash memory. The flash translation layer, when writing to a physical address location, chooses between a wear-leveling circuit and a wear-limiting circuit to select the physical address location.

Abstract: In accordance with embodiments of the invention, a set of Microinstructions define a set of primitives or instructions to execute a business process modeling language in accordance with the semantics defined by the language. The Microinstructions define atomic operations that can be executed as part of a running process. The Microinstructions can be combined into Microprograms which map to business processes of the business process modeling language, for example BPMN. The Microprograms can be compiled using a suitable compiler into any execution language. The Microinstructions thereby provide flexibility for executing business process modeling languages in any business process execution language.

Abstract: A microcontroller is provided and includes a reset pin, a reset circuit, and a first logical circuit. A first reset signal is generated at the reset pin when the microcontroller is powered on. The reset circuit receives the first reset signal and generates a second reset signal. The reset circuit includes a plurality of flipflops. After the microcontroller is powered on, the reset circuit switches a state of the second reset signal according to the first reset signal when an output combination of a plurality of output values of the plurality of flipflops is not a specific value. The first logical circuit performs a first initialization operation when the state of the second reset signal is switched. When the second reset signal is switched, the reset circuit sets the output combination of the plurality of output values of the plurality of flipflops to the specific value.

Abstract: A multi-branch determination syntax optimization apparatus includes: a memory that retains multi-branch determination syntax including tokens; a database that retains (1) CPU performance information being a parameter depending on a CPU incorporated in the multi-branch determination syntax optimization apparatus and set based on time required for multi-branch determination processing and (2) frequently-appearing token table representing types and rates of appearance of tokens sorted in order of appearance frequency in a query that statically analyzes a source code and performs lexical analysis in advance; and a conversion section that executes determination for the multi-branch determination syntax by referring to the CPU performance information and the frequently-appearing token table, and creates a branch code converted to make a speculatively executable branch for the token type having a high frequency of appearance and to make a branch using the jump table for the token type having a low frequency of appea