Hewlett-packard Development Company, L.p.

Hewlett Packard Enterprise Development LP

David Plaquin

Jeffrey Kevin Jeansonne

Maugan Villatel

Boris Balacheff

Ronny Chevalier

Chris I Dalton

Graeme John Proudler

Adrian John Baldwin

Christopher Ian Dalton

Guillaume Hiet

Valiuddin Y Ali

Joshua Serratelli Schiffman

Vali Ali

Rosilet Retnamoni Braduke

Christopher L Dalton

Yolanta Beresnevichiene

Pierre Belgarric

Serdar Cabuk

William Burton

Dirk Kuhlmann

Gregg Bernard Lesartre

Joseph Foster

Liqun Chen

James M Mann

Chris Richard Gunning

Richard A Bramley, Jr

James Robert Waldron

Daniel Cameron Ellam

Simon Kai-Ying Shiu

Adrian Shaw

Eugene Cohen

Gurchetan Grewal

Mark Ryan

Taram Sierra Devitt-Carolan

Carey Huscroft

Marco Ricca

Ijlal Loutfi

In an example, an apparatus is described. The apparatus comprises processing circuitry comprising a control module. The control module determines whether a computing device communicatively coupled to the control module is in a specified state for executing a machine learning model controlled by a third party entity. In response to determining that the computing device is in the specified state, the control module is to send, to an attestation module in a data processing pipeline associated with the computing device, an indication that the computing device is in the specified state.

Controlling execution of machine learning models

In an example, a method includes receiving an indication of an electrical parameter associated with at least part of a computing device. The method further includes determining whether or not the indication is indicative of an expected electric field distribution associated with a specified hardware configuration for the computing device.

Determining electric field distributions

In an example there is provided a method for receiving notification of an intrusion event in relation to an application from an intrusion detection system, accessing state data in relation to a state of the application prior to the intrusion event, the state data having been stored on the basis of a change of state of the application, accessing a policy to be applied to the state data in response to the intrusion event, modifying the state data on the basis of the policy, and restoring the application on the basis of the modified state data.

Intrusion resilient applications

An example method is disclosed, for example a method of executing a software module in a computing system, the method comprising executing, in a first processing device of the computing system, a first software module to verify a second software module and to cause a second processing device of the computing system to execute the second software module, executing, in the second processing device, the second software module to execute, in the second processing device, a third software module and to provide a first key of a key pair to the third software module, and protecting, by the second processing device, a memory space associated with the third software module, wherein the memory space contains the first key of the key pair, wherein the first processing device contains a second key of the key pair.

Executing software

A method for monitoring control-flow integrity in a low-level execution environment, the method comprising receiving, at a monitor, a message from the execution environment indicating that the execution environment has entered a controlled mode of operation, receiving, at the monitor, a data packet representing execution of a selected portion of a control-flow process at the execution environment, identifying, using the data packet, a pathway corresponding to the selected portion of the control-flow process from a set of permissible control-flow pathways and determining whether the identified pathway corresponds to an expected control-flow behaviour.

Monitoring control-flow integrity

A method is provided, comprising actively testing the access control policy of a software target using a probing logic. The method further comprises determining whether an intrusion in the software target has occurred based on monitored side effects. According to the method, the probing logic is to execute at least one operation that is forbidden by the access control policy. The probing logic is further to create at least one predetermined observable side effect based on the successful execution of the operation.

Active testing of access control policy

An example system with a pre-OS (Operating System) environment, the pre-OS environment includes a private memory that is isolated from a processor of the system. The pre-OS environment also includes an embedded controller (EC) coupled to the private memory, where the EC includes an embedded key. The EC is to execute instructions to generate an encryption key based on the embedded key; generate a signature key; obtain data; produce an integrity-verification tag based on a hash of the obtained data, where the hash employs the signature key; encrypt the obtained data based on the encryption key; store the encrypted data in the private memory; and store the integrity-verification tag in the private memory in association with the stored encrypted data.

Data protection in a pre-operation system environment based on an embedded key of an embedded controller

An apparatus includes a solid-state a solid-state non-volatile computer memory; and a controller coupled to the memory. The controller to: generate a data set including a tag that indicates that the data set is valid; write the data set into a block of the memory, wherein the block includes multiple addressable locations set to a common first binary value before the write; generate a subsequent data set including a tag that indicates that the subsequent data set is valid; update the tag of the written data set to indicate that the written data set is invalid, wherein the update includes setting an addressable location corresponding to the tag to second binary value different from the first binary value; write the subsequent data set to addressable locations in the block of memory other than the addressable locations of the invalid data set.

Validity of data sets stored in memory

A method for monitoring access to a user account comprises receiving a user account login status from a target service in response to a user login request, comparing the user account login status with an expected status value at a user apparatus, and on the basis of the comparison, performing at least one of: synchronising the status value at the user apparatus with the user account login status from the target service, and executing a user login update process at the user apparatus.

Monitoring access

In an example, there is provided a method for attesting to a management state of a device platform configuration, based on a signed data structure that represents a device management status and a time-based value from a secure cryptoprocessor of the device. The signed data structure is sent to a management system for validation based on a comparison to the state of the cryptoprocessor at the time of signing the data structure.

Platform configurations

An intrusion detection system, comprising a monitor to receive messages from a target over a low-latency communication link comprising a controlled access memory structure logically positioned between the target and the monitor using point-to-point interconnects, the controlled access memory structure to receive a message from the target indicating that the target has entered a controlled mode of operation.

Intrusion detection systems

An IC comprising functional circuit to perform primary functions of the IC is provided. The functional circuit is to enable electrical signals to propagate through it within a timing constraint of the functional circuit. The IC comprises at least one canary circuit used for detecting glitch attacks on the circuit. Electrical signals are to propagate through the canary circuit(s) within a defined timing constraint of the canary circuit(s). The canary circuit is to provide a signal path designed such that in the event of a timing constraint of the functional circuit(s) is violated due to a glitch attack, also the timing constraint of the canary circuit(s) is violated.

Integrated circuit(s) with anti-glitch canary circuit(s)

Examples herein disclose monitoring an expected functionality upon execution of a system management mode (SMM) code. The examples detect whether a change has occurred to the SMM code based on the monitoring of the expected functionality. The change indicates that the SMM code is compromised.

Detecting a change to system management mode bios code

A method for restricting write access to a non-volatile memory. The method includes receiving a request to write to a protected location in the non-volatile memory and determining whether the protected location is in a write-protected state. If the protected location is not in a write-protected state, the method includes writing data indicated by the request to the protected location. If the protected location is in a write-protected state, the method includes rejecting the request. The protected location stores a validation key to validate the contents of another portion of the non-volatile memory.

Systems and methods for restricting write access to non-volatile memory

In an example there is provided a method of applying a mitigation action to a computing system. The method comprises receiving notification of an intrusion event on a computing system. The notification identifies one or more of data, and a process affected by the intrusion event. The method comprises accessing state data corresponding to a state of the computing system prior to the intrusion event, accessing a policy specifying one or more mitigation actions to be applied to the one or more of data, and a process in response to an intrusion event, restoring the one or more of data, and the process on the basis of the state data, and applying a mitigation action according to the policy.

Growing community of inventors

Bristol, United Kingdom

David Plaquin