An apparatus comprises a plurality of redundant processing units (4) to perform data processing redundantly in lockstep; common mode fault detection circuitry *6, 22) to detect an event indicative of a potential common mode fault affecting each of the plurality of redundant processing units; a memory (10) shared between the plurality of redundant processing units; and memory checking circuitry (30) to perform a memory scanning operation to scan at least part of the memory for errors; in which the memory checking circuitry (30) performs the memory scanning operation in response to a common mode fault signal generated by the common mode fault detection circuitry (6, 22) indicating that the event indicative of a potential common mode fault has been detected.

An apparatus has first processing circuitry and second processing circuity. The second processing circuitry has at least one hardware mechanism providing a greater level of fault protection or fault detection than is provided for the first processing circuitry. Coherency control circuitry controls access to data from at least part of a shared address space by the first and second processing circuitry according to an asymmetric coherency protocol in which a local-only update of data in a local cache of the first processing circuitry is restricted in comparison to a local-only update of data in a local cache of the second processing circuitry.

An embodiment device comprises a first processing unit configured to process an initial data line and deliver a first processed data line, a first delay unit coupled to the output of the first processing unit and configured to deliver a delayed first processed data line delayed by a first delay, a second delay unit configured to deliver the delayed initial data line delayed by a second delay, a second processing unit coupled to the output of the second delay unit and configured to process the delayed initial data line and deliver a delayed second processed data line, and a comparison unit configured to compare the contents of the delayed first and second processed data lines and deliver a non-authentication signal if the contents are not identical, the first and second delays being equal to a variable value.

Provided is an image recognition processor. The image recognition processor includes a plurality of nano cores each configured to perform a pattern recognition operation and arranged in rows and columns, an instruction memory configured to provide instructions to the plurality of nano cores in a row unit, a feature memory configured to provide input features to the plurality of nano cores in a row unit, a kernel memory configured to provide a kernel coefficient to the plurality of nano cores in a column unit, and a difference checker configured to receive a result of the pattern recognition operation of each of the plurality of nano cores, detect whether there is an error by referring to the received result, and provide a fault tolerance function that allows an error below a predefined level.

According to one embodiment, the first processing unit receives a first clock and outputs, at its first output node, data obtained by first processing of data at an input node. The second processing unit receives a first clock and outputs, at its second output node, data obtained by the first processing of the data at the input node. The third processing receives a second clock, outputs, from its third output nodes, data obtained by the first processing of the data at the input node, and outputs, from its fourth output nodes, data obtained by the first processing of the data at the input node. The determination unit outputs a first signal based on data at the fifth to eighth nodes respectively coupled to the first to fourth output nodes.

A resilient system implementation in a network-on-ship with at least one functional logic unit and at least one duplicated logic unit. A resilient system and method, in accordance with the invention, are disclosed for detecting a fault or an uncorrectable error and isolating the fault. Isolation of the fault prevents further propagation of the fault throughout the system. The resilient system includes isolation logic or an isolation unit that isolates the fault.

An apparatus has two or more processing elements to redundantly process a same processing workload; and divergence detection circuitry to detect divergence between the plurality of processing elements. When a correctable error is detected by error detection circuitry of an erroneous processing element, the erroneous processing element signals detection of the correctable error to another processing element, to control the other processing element to delay processing to maintain a predetermined time offset between the erroneous processing element and the other processing element.

A method, executed by a computer, includes pairing a first core with a second core to form a first core group, wherein each core of the group has a plurality of functional units, transferring instructions received by the first core to the second core for execution via a first inter-core communication bus, and executing the instructions on the second core. A computer system and computer program product corresponding to the above method are also disclosed herein.

A method to check for redundancy in two or more data lines comprises receiving data on a first data line, computing a first cyclic redundancy check (CRC) value on the data of the first data line, performing an exclusive OR (XOR) function on the first CRC value with a stored memory value, and updating the stored memory value with a result of the XOR function, and repeating on additional data lines until a last line is processed such that an error is indicated if a final stored memory value is not zero. An apparatus to check that two cores are operating in lockstep comprises a first core comprising a first data checker, a second core comprising a second data checker, and a lockstep checker to compare an output of the first data checker with an output of the second data checker.

Examples of the present disclosure relate to a method for anomaly response in a system on chip. The method comprises measuring a magnitude of a transient anomaly event in an operating condition of the system on chip. Based on the magnitude it is determined, for each of a plurality of components of the system on chip, an indication of susceptibility of that component to an anomaly event of the measured magnitude. Based on the determined indications of susceptibility for each of the plurality of components, an anomaly response action is determined. The method then comprises performing the anomaly response action.