Stmicroelectronics Gmbh

Siemens Industry Software Gmbh

Rambus Inc.

Ultrasoc Technologies Limited

Other

Intel Corporation

Mentor Graphics Corporation

Clearspeed Technology Limited

Gajinder Singh Panesar

Peter William Thompson

John A Carey

Neil Richards

Iain Craig Robertson

Ray McConnell

Paul Winser

Ken Cameron

Gordon Faulds

Anthony Spencer

Jeff Bond

John Rhoades

Simon McIntosh-Smith

Matthias Dejaegher

Danny Halamish

Callum Stewart

Melvin Cheah

Marcin Hlond

Hanan Moller

William P Robbins

Andrew Duller

Anthony Peter John Claydon

Peter Claydon

Andrew Kuligowski

Alex Orr

Olfat Younis

Rupert Baines

A method of monitoring messages from a sensor using an integrated circuit is provided. The messages include data measured by that sensor. The method includes reading a first message from interconnect circuitry of the integrated circuit. The interconnect circuitry connects the sensor to one or more core devices configured to process the messages. A first hash value is calculated for the first message. The first hash value is compared to one or more prior hash values stored in a hash store. Each prior hash value of the one or more prior hash values corresponds to a message that was read from the interconnect circuitry prior to the first message. A corrective action is performed when a difference between the first hash value and at least one of the prior hash values stored in the hash store is below a predetermined threshold.

Hardware-based sensor analysis

A method of detecting anomalous latencies in communications between components on an integrated circuit (IC) chip. The method includes: (i) monitoring communications between a first component of the IC chip and other components of the IC chip, each communication comprising a command sent from the first component to another component, and a response received by the first component from that other component, the monitoring comprising: measuring the number of communications in each of a series of monitored time windows, and measuring the latency of each communication in the series of monitored time windows; (ii) calculating a maximum tolerable latency for each operational time window of the first component from the number of communications in that operational time window, an available stall time of the first component in that operational time window, and a latency penalty factor for that operational time window; and (iii) determining a measured latency to be anomalous if the measured latency is greater than the maximum tolerable latency.

Detecting anomalous latent communications in an integrated circuit chip

An integrated circuit (IC) chip includes system circuitry having system memory, and a master processor and a checker processor configured to operate in lockstep; and monitoring circuitry comprising an internal lockstep monitor, a master tracer and a checker tracer. The internal lockstep monitor is configured to: observe states of internal signals of the master processor and the checker processor, compare corresponding observed states of the master processor and the checker processor, and if the corresponding observed states differ: trigger the master tracer to output stored master trace data recorded from the output of the master processor, and trigger the checker tracer to output stored checker trace data recorded from the output of the checker processor.

Monitoring processors operating in lockstep

A method of identifying a cause of an anomalous feature measured from system circuitry on an integrated circuit (IC) chip, the IC chip comprising the system circuitry and monitoring circuitry for monitoring the system circuitry by measuring features of the system circuitry in each window of a series of windows, the method comprising: (i) from a set of windows prior to the anomalous window comprising the anomalous feature, identifying a candidate window set in which to search for the cause of the anomalous feature; (ii) for each of the measured features of the system circuitry: (a) calculating a first feature probability distribution of that measured feature for the candidate window set; (b) calculating a second feature probability distribution of that measured feature for window(s) not in the candidate window set; (c) comparing the first and second feature probability distributions; and (d) identifying that measured feature in the timeframe of the candidate window set as a cause of the anomalous feature if the first and second feature probability distributions differ by more than a threshold value; (iii) iterating steps (i) and (ii) for further candidate window sets from the set of windows prior to the anomalous window; and (iv) outputting a signal indicating those measured feature(s) of step (ii)(d) identified as a cause of the anomalous feature.

Identifying causes of anomalies observed in an integrated circuit chip

A supervisory unit configured to supervise interconnect messages passing to or from an interconnect is provided. The supervisory unit is configured to, on receiving an interconnect message: store the interconnect message in a data store; compare the interconnect message to predetermined filter criteria; and select, in dependence on that comparison, one or more actions to be taken with respect to the interconnect message. The one or more actions are selected from the group including: permitting the interconnect message to pass unaltered; blocking the interconnect message from passing and permitting the interconnect message to pass in an altered state; and performing the one or more selected actions with respect to the interconnect message.

Message monitoring

A method of performing error detection within an integrated circuit chip analyses transactions communicated over interconnect circuitry of the integrated circuit chip to detect whether a message contains a data error. A memory of the integrated circuit chip coupled to the interconnect circuitry is scanned to detect whether there is a data error stored in the memory, and in response to detecting a data error in a transaction communicated over the interconnect circuitry and/or a data error stored in the memory, a dedicated action indicative of a data error is performed.

Error detection within an integrated circuit chip

An apparatus comprising: a processing unit configured to execute a plurality of threads; a profiling unit configured to: profile the operation of the processing unit over a time period to generate an activity profile indicating when each of the plurality of threads is executed by the processing unit over the time period; analyse the generated activity profile to determine whether a signature of the processing unit's thread execution for the time period matches a signature indicating a baseline of thread execution for the processing unit; output an alert signal if the signature of the processing unit's thread execution for the time period does not match the signature indicating a baseline of thread execution for the processing unit.

Performance profiling for a multithreaded processor

An integrated circuit chip comprising: system circuitry comprising interconnect circuitry for transporting transactions; and monitoring circuitry configured to: monitor transactions from the interconnect circuitry comprising transactions between an entity and a specified region of the integrated circuit chip, the entity being associated with a set of one or more access rights for accessing the specified region of the integrated circuit chip; determine from the monitored transactions values of one or more parameters associated with the access to the specified region by the entity to identify whether the entity has breached its access rights; and perform a dedicated action indicative of a breach of the access rights in response to determining from the parameter values that the entity has breached its access rights.

Monitoring accesses to a region of an integrated circuit chip

An integrated circuit, having a security supervision system, comprising a plurality of functional circuit blocks interconnected to collectively performing data processing tasks, one or more communication adaptors, having: (i) a hardware interconnection to the functional circuit blocks, whereby the communication adaptor senses the state and/or activity of the functional circuit block; (ii) memory storing definitions of state and/or activity of functional circuit block and actions for each definition; and (iii) processing circuitry comparing the state and/or activity of the functional block with each definition, such that when state and/or activity of the functional block corresponding to a stored definition is detected, perform the corresponding action. The memory stores a definition of state and/or activity characteristic of insecure operation of the functional circuit block and a corresponding action to partially disabling the functional circuit block and/or (ii) causing a message to be transmitted to a destination off the integrated circuit.

Integrated circuit security

A LIW processor comprises multiple execution units. The multiple execution units of the processor are divided into groups, and an input instruction word can contain instructions for one execution unit in each of the groups. The processor is optimized for use in signal processing operations, in that the multiple execution units of the processor are divided into groups which do not place significant restrictions on the desirable uses of the processor, because it has been determined that, in signal processing applications, it is not usually necessary for certain execution units to operate simultaneously. These execution units can therefore be grouped together, in such a way that only one of them can operate at a particular time, without significantly impacting on the operation of the device. An array is formed from multiple interconnected processors of this type.

Processor architecture for processing variable length instruction words

A data processing architecture includes an input device that receives an incoming stream of data packets. A plurality of processing elements are operable to process data received from the input device. The input device is operable to distribute data packets in whole or in part to the processing elements in dependence upon the data processing bandwidth of the processing elements.

SIMD array operable to process different respective packet protocols simultaneously while executing a single common instruction stream

Data is transmitted from a sending processor over a network to one or more receiving processor in a forward direction during an allocated slot, and acknowledge signals are sent in a reverse direction during the same allocated slot, to indicate whether the receiving processor is able to receive data If one or more of the receiving processors indicates that it is unable to receive the data, the data is retransmitted during the next allocated slot. This means that the sending processor is able to determine within the slot period whether a retransmission is necessary, but that the slot period only needs to be long enough for one-way communication.

Communications in a processor array

An input/output system transfers data packets to and from a SIMD array of processing elements (PEs) such that different sizes of data packets are transferred to respective ones of the PEs. The packets are transferred in batches to respective different addresses in the array under the control of the PEs. Transfer to or from the array may be carried out when either a batch or part of a batch is ready for transfer. The decision to transfer either full or part batches is made in dependence upon the speed of the PEs and the speed and intermittency of the data packets.

Data processing architectures for packet handling using a SIMD array

A data processing architecture comprising:

Data processing architectures for packet handling wherein batches of data packets of unpredictable size are distributed across processing elements arranged in a SIMD array operable to process different respective packet protocols at once while executing a single common instruction stream

Growing community of inventors

Montpelier Bristol, United Kingdom

Gajinder Singh Panesar