Disclosed in some examples are systems, methods, devices, and machine-readable mediums to detect and terminate programmable atomic transactions that are stuck in an infinite loop. In order to detect and terminate these transactions, the programmable atomic unit may use an instruction counter that increments each time an instruction is executed during execution of a programmable atomic transaction. If the instruction counter meets or exceeds a threshold instruction execution limit without reaching the termination instruction, the programmable atomic transaction may be terminated, all resources used (e.g., memory locks) may be freed, and a response may be sent to a calling processor.

A method of synchronizing a group of scheduled tasks within a parallel processing unit into a known state is described. The method uses a synchronization instruction in a scheduled task which triggers, in response to decoding of the instruction, an instruction decoder to place the scheduled task into a non-active state and forward the decoded synchronization instruction to an atomic ALU for execution. When the atomic ALU executes the decoded synchronization instruction, the atomic ALU performs an operation and check on data assigned to the group ID of the scheduled task and if the check is passed, all scheduled tasks having the particular group ID are removed from the non-active state.

Disclosed embodiments relate to spatial and temporal merging of remote atomic operations. In one example, a system includes an RAO instruction queue stored in a memory and having entries grouped by destination cache line, each entry to enqueue an RAO instruction including an opcode, a destination identifier, and source data, optimization circuitry to receive an incoming RAO instruction, scan the RAO instruction queue to detect a matching enqueued RAO instruction identifying a same destination cache line as the incoming RAO instruction, the optimization circuitry further to, responsive to no matching enqueued RAO instruction being detected, enqueue the incoming RAO instruction; and, responsive to a matching enqueued RAO instruction being detected, determine whether the incoming and matching RAO instructions have a same opcode to non-overlapping cache line elements, and, if so, spatially combine the incoming and matching RAO instructions by enqueuing both RAO instructions in a same group of cache line queue entries at different offsets.

A radio frequency (RF) system may include an RF transceiver and a baseband engine, application specific integrated circuit (ASIC) coupled to the RF transceiver and configured to perform a given baseband engine operation from among different baseband engine operations. The baseband engine ASIC may include a memory and a state machine coupled thereto and configured to store a respective set of programming instructions for each of different baseband engine operations and to permit selection of the given set of programming instructions. The baseband engine may also include a programmable processing circuit coupled to the memory and the state machine and configured to perform block coding computations responsive to the given set of programming instructions.

A machine learning network is implemented by executing a computer program of instructions on a machine learning accelerator (MLA) comprising a plurality of interconnected storage elements (SEs) and processing elements (PEs). The instructions are partitioned into blocks, which are retrieved from off-chip memory. The block includes a set of deterministic instructions (MLA instructions) to be executed by on-chip storage elements and/or processing elements according to a static schedule from a compiler. The MLA instructions may require data retrieved from off-chip memory by memory access instructions contained in prior blocks. The compiler also schedules the memory access instructions in a manner that avoids contention for access to the off-chip memory. By avoiding contention, the execution time of off-chip memory accesses becomes predictable enough and short enough that the memory access instructions may be scheduled so that they are known to complete before the retrieved data is required.

A digital data processor includes an instruction memory storing instructions specifying data processing operations and a data operand field, an instruction decoder coupled to the instruction memory for recalling instructions from the instruction memory and determining the operation and the data operand, and an operational unit coupled to a data register file and an instruction decoder to perform an operation upon an operand corresponding to an instruction decoded by the instruction decoder and storing results of the operation. The operational unit is configured to perform a table recall in response to a look up table read instruction by recalling data elements from a specified location and adjacent location to the specified location, in a specified number of at least one table and storing the recalled data elements in successive slots in a destination register. Recalled data elements include at least one interpolated data element in the adjacent location.

A neural processing unit (NPU) includes a controller including a scheduler, the controller configured to receive from a compiler a machine code of an artificial neural network (ANN) including a fusion ANN, the machine code including data locality information of the fusion ANN, and receive heterogeneous sensor data from a plurality of sensors corresponding to the fusion ANN; at least one processing element configured to perform fusion operations of the fusion ANN including a convolution operation and at least one special function operation; a special function unit (SFU) configured to perform a special function operation of the fusion ANN; and an on-chip memory configured to store operation data of the fusion ANN, wherein the schedular is configured to control the at least one processing element and the on-chip memory such that all operations of the fusion ANN are processed in a predetermined sequence according to the data locality information.

According to some example embodiments of the present disclosure, in a method for a memory lookup mechanism in a high-bandwidth memory system, the method includes: using a memory die to conduct a multiplication operation using a lookup table (LUT) methodology by accessing a LUT, which includes floating point operation results, stored on the memory die; sending, by the memory die, a result of the multiplication operation to a logic die including a processor and a buffer; and conducting, by the logic die, a matrix multiplication operation using computation units.

An electronic cigarette comprises a battery portion which is operable to provide power to a heating element of the electronic cigarette, and a cartomizer coupled with the battery portion. The cartomizer comprises a liquid container which provides a liquid toward the heating element wherein the liquid is used for producing and flavoring a vapor of the electronic cigarette when the liquid is heated by the heating element, and a memory which is operable to record and store an amount of the liquid remaining in the liquid container.

An apparatus, and corresponding method, for input/output (I/O) value determination, generates an I/O instruction for an I/O device, the I/O device including a state machine with state transition logic. The apparatus comprises a controller that includes a simplified state machine with a reduced version of the state transition logic of the state machine of the I/O device. The controller is configured to improve instruction execution performance of a processor core by employing the simplified state machine to predict at least one state value of at least one I/O device true state value to be affected by the I/O instruction at the I/O device.