Techniques related to executing a plurality of instructions by a processor comprising receiving a first instruction for execution on an instruction execution pipeline, beginning execution of the first instruction, receiving one or more second instructions for execution on the instruction execution pipeline, the one or more second instructions associated with a higher priority task than the first instruction, storing a register state associated with the execution of the first instruction in one or more registers of a capture queue associated with the instruction execution pipeline, copying the register state from the capture queue to a memory, determining that the one or more second instructions have been executed, copying the register state from the memory to the one or more registers of the capture queue, and restoring the register state to the instruction execution pipeline from the capture queue.

Techniques related to executing a plurality of instructions by a processor comprising receiving a first instruction for execution on an instruction execution pipeline, beginning execution of the first instruction, receiving one or more second instructions for execution on the instruction execution pipeline, the one or more second instructions associated with a higher priority task than the first instruction, storing a register state associated with the execution of the first instruction in one or more registers of a capture queue associated with the instruction execution pipeline, copying the register state from the capture queue to a memory, determining that the one or more second instructions have been executed, copying the register state from the memory to the one or more registers of the capture queue, and restoring the register state to the instruction execution pipeline from the capture queue.

A streaming engine employed in a digital data processor specifies a fixed read only data stream defined by plural nested loops. An address generator produces addresses of data elements. A steam head register stores data elements next to be supplied to functional units for use as operands. Stream metadata is stored in response to a stream store instruction. Stored stream metadata is restored to the stream engine in response to a stream restore instruction. An interrupt changes an open stream to a frozen state discarding stored stream data. A return from interrupt changes a frozen stream to an active state.

A streaming engine employed in a digital data processor specifies a fixed read only data stream defined by plural nested loops. An address generator produces addresses of data elements. A steam head register stores data elements next to be supplied to functional units for use as operands. Stream metadata is stored in response to a stream store instruction. Stored stream metadata is restored to the stream engine in response to a stream restore instruction. An interrupt changes an open stream to a frozen state discarding stored stream data. A return from interrupt changes a frozen stream to an active state.

A processor achieving a zero-overhead loop, includes instruction stream control circuitry and loop control circuitry. The loop control circuitry includes loop address detecting circuitry and loop end determining circuitry. By combining instructions and hardware, the loop control circuitry eliminates additional control instructions required b each loop iteration and can achieve loop acceleration with zero overhead, thereby improving the loop execution efficiency.

A data processing device is described including one or more processors implementing a plurality of data processing entities, one or more software interrupt nodes and an access register for each software interrupt node. The access register specifies which one or more data processing entities of the plurality of data processing entities is/are each allowed to, as interrupt source data processing entity, trigger an interrupt service request on the software interrupt node for another one of the plurality of data processing entities as an interrupt target processing entity. Each software interrupt node is configured to forward an interrupt service request triggered by an interrupt source data processing entity which is allowed to trigger an interrupt service request on the software interrupt node to an interrupt target processing entity.

An apparatus includes an array processor to process at least one array. The apparatus further includes a memory coupled to the array processor. The at least one array is stored in memory with programmable per-dimension size and stride values.

A memory device in accordance with a described method of operation includes a read only memory (ROM) address controller and a suspend signal generator. The ROM address controller is configured to sequentially output a plurality of operation ROM addresses at which ROM codes to be executed in response to an operation command are stored, and to suspend output of the plurality of operation ROM addresses in response to a suspend signal. The suspend signal generator is configured to generate the suspend signal that is activated during a preset period depending on whether a suspend ROM address is identical to an operation ROM address, among the plurality of operation ROM addresses, currently being output. The suspend ROM address is an address at which a ROM code, execution of which is to be suspended, among the ROM codes, is stored.

A memory device in accordance with a described method of operation includes a read only memory (ROM) address controller and a suspend signal generator. The ROM address controller is configured to sequentially output a plurality of operation ROM addresses at which ROM codes to be executed in response to an operation command are stored, and to suspend output of the plurality of operation ROM addresses in response to a suspend signal. The suspend signal generator is configured to generate the suspend signal that is activated during a preset period depending on whether a suspend ROM address is identical to an operation ROM address, among the plurality of operation ROM addresses, currently being output. The suspend ROM address is an address at which a ROM code, execution of which is to be suspended, among the ROM codes, is stored.

An embodiment of an integrated circuit may comprise a return stack buffer (RSB), a speculative return stack buffer (SRSB), and circuitry coupled to the RSB and the SRSB, the circuitry to track a count until the SRSB is empty at a time of a prediction by a branch prediction unit, and return an output from the branch prediction unit that corresponds to one of the RSB and the SRSB based at least in part on the count until the SRSB is empty. Other embodiments are disclosed and claimed.