Abstract:
A processor is provided that is capable of concurrently processing a sequence of instructions for a plurality of threads achieving the retry success rate equivalent to the success rate in processors that process a sequence of instructions for a single thread. An arithmetic device 200 is provided with an instruction execution circuit 201 for executing a plurality of threads, or an execution control circuit 202 for controlling the execution state or rerunning of the threads.
Abstract:
Embodiments include a system, an apparatus, a device, and a method. An apparatus includes a synchronous circuit including a first subcircuit powered by a first power plane having a first power plane voltage and a second subcircuit powered by a second power plane having a second power plane voltage. The apparatus also includes an error detector operable to detect an incidence of a computational error occurring in the first subcircuit. The apparatus includes a controller operable to change the first power plane voltage based upon the detected incidence of a computational error. The apparatus also includes a power supply configured to electrically couple with a portable power source and operable to provide a selected one of at least two voltages to the first power plane in response to the controller.
Abstract:
A processor core includes an instruction decode unit that may dispatch a same integer instruction stream to a plurality of integer execution units and may consecutively dispatch a same floating-point instruction stream to a floating-point unit. The integer execution units may operate in lock-step such that during each clock cycle, each respective integer execution unit executes the same integer instruction. The floating-point unit may execute the same floating-point instruction stream twice. Prior to the integer instructions retiring, compare logic may detect a mismatch between execution results from each of the integer execution units. In addition, prior to the results of the floating-point instruction stream transferring out of the floating-point unit, the compare logic may also detect a mismatch between results of execution of each consecutive floating-point instruction stream. Further, in response to detecting any mismatch, the compare logic may cause instructions causing the mismatch to be re-executed.
Abstract:
Split hardware transaction techniques may support execution of serial and parallel nesting of code within an atomic block to an arbitrary nesting depth. An atomic block including child code sequences nested within a parent code sequence may be executed using separate hardware transactions for each child, but the execution of the parent code sequence, the child code sequences, and other code within the atomic block may appear to have been executed as a single transaction. If a child transaction fails, it may be retried without retrying the parent code sequence or other child code sequences. Before a child transaction is executed, a determination of memory consistency may be made. If a memory inconsistency is detected, the child transaction may be retried or control may be returned to its parent. Memory inconsistencies between parallel child transactions may be resolved by serializing their execution before retrying at least one of them.
Abstract:
Embodiments include a system, a device, and a method. A computing system includes a synchronous circuit. The synchronous circuit includes a first subcircuit powered by a first power plane having a first power plane voltage and a second subcircuit powered by a second power plane having a second power plane voltage. The system also includes an error detector operable to detect an incidence of a computational error occurring in the first subcircuit. The system further includes a controller operable to change the first power plane voltage based upon the detected incidence of a computational error. The system may include a power supply operable to provide a selected one of at least two voltages to the first power plane in response to the controller.
Abstract:
A cosmic ray detector includes a cantilever with a first tip. The detector also includes a second tip and circuitry to provide a signal indicative of a distance between the first and second tips being such as would be caused by a cosmic ray interaction event.
Abstract:
Embodiments include a system, a device, and a method. A computing system includes a synchronous circuit. The synchronous circuit includes a first subcircuit powered by a first power plane having a first power plane voltage and a second subcircuit powered by a second power plane having a second power plane voltage. The system also includes an error detector operable to detect an incidence of a computational error occurring in the first subcircuit. The system further includes a controller operable to change the first power plane voltage based upon the detected incidence of a computational error. The system may include a power supply operable to provide a selected one of at least two voltages to the first power plane in response to the controller.
Abstract:
Embodiments include a system, an apparatus, a device, and a method. An apparatus includes a synchronous circuit including a first subcircuit powered by a first power plane having a first power plane voltage and a second subcircuit powered by a second power plane having a second power plane voltage. The apparatus also includes an error detector operable to detect an incidence of a computational error occurring in the first subcircuit. The apparatus includes a controller operable to change the first power plane voltage based upon the detected incidence of a computational error. The apparatus also includes a power supply configured to electrically couple with a portable power source and operable to provide a selected one of at least two voltages to the first power plane in response to the controller.
Abstract:
Disclosed is a coherence preservation method of duplicated data in RAID subsystem. The method comprises the steps of: if a read operation is requested by the RAID subsystem, selecting an arbitrary target disk among disks having duplicated data, and implementing the read operation on the selected disk; implementing a read error handling to determine whether an error occurs in the disks selected upon the read operation, and if the error occurs in the read operation of the selected target disk, repeating the read error handling until the read operation of the data succeeds by circulating the remaining metering-treated disks in turns; after the read error handling, if the read operation on the remaining disk in all succeeds, completing the read error handling, and if the read operation on the remaining disk in all fails, returning the error to the disk, and completing the read error handling; if a write operation is requested by the RAID subsystem, implementing the write operation on the disks having duplicated data; after implementing the write operation, if a temporary error occurs in the duplicated data, implementing a temporary error handling to maintain the data coherence in the disks having the duplicated data and to recover the temporary error; and after implementing the write operation, if a permanent error occurs in the duplicated data, implementing a permanent error handling to maintain the data coherence in the disks having the duplicated data and to recover the permanent error.
Abstract:
A technique to detect and correct corruption of instructions by soft errors. A parity bit is propagated with an instruction through the instruction flow path and checked at selected places. When a parity error is detected, a replay circuit is used to perform a replay to reload the instruction to remove the corrupted instruction.