Browse Prior Art Database

Method and apparatus for restoring and debugging the embedded application context after system failure

IP.com Disclosure Number: IPCOM000240578D
Publication Date: 2015-Feb-10
Document File: 2 page(s) / 655K

Publishing Venue

The IP.com Prior Art Database

Abstract

The presented approach relates to a method and apparatus for offline restoring and debugging of an application and target context after system failure. Such method is proved to be useful for devices deployed on the field where normal debugging procedures in case of failure are very hard or even impossible to be mimicked. This approach addresses the limitation in respect with read-only resource restoration (e.g.: status registers or read-only register bit fields). Current proposal is a method/apparatus that will use code injection to change the value of the read only resources. It’s a complete method that can present to the user an accurate state of the target before any fault occurred. The ultimate goal is to bring any target at anytime and anywhere, under the same conditions that caused the failure of the original equipment and allows the lab technicians to perform post mortem analysis. The method can be applied to any DSP processor that has status resources read only. Using this method, the user will be able to browse live the target state before the fault occurred. Also, it offers the possibility to locate the root cause of the failure, fix the failure and validate the solution, all in the same debug session.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Method and apparatus for restoring and debugging the embedded application context after system failure

The presented approach relates to a method and apparatus for offline restoring and debugging of an application and target context after system failure. Such method is proved to be useful for devices deployed on the field where normal debugging procedures in case of failure are very hard or even impossible to be mimicked.

This approach addresses the limitation in respect with read-only resource restoration (e.g.: status registers or read-only register bit fields). Current proposal is a method/apparatus that will use code injection to change the value of the read only resources. It’s a complete method that can present to the user an accurate state of the target before anyfault occurred. The ultimate goal is to bring any target at anytime and anywhere, under the same conditions that caused the failure of the original equipment and allows the lab technicians to perform post mortem analysis.

The method can be applied to any DSP processor that has status resources read only. Using this method, the user will be able to browse live the target state before the fault occurred. Also, it offers the possibility to locate the root cause of the failure, fix the failure and validate the solution, all in the same debug session.

This paper proposes a solution to restore target and application context to after system failure. Furthermore, it offers the user the possibility to identify the cause of the failure, fix it and validate the solution in the same debug session.

The challenge when trying to reproduce the target state is to restore the read-only registers. The solution proposed is based on debugger and compiler capabilities to reproduce a certain value i...