Differential file generation method and upgrade method and system based on differential file

Abstract

The invention provides a differential file generation method and an upgrade method and system based on a differential file, and belongs to the technical field of terminal upgrade. The differential file generation method comprises the steps that an index is established for an old upgrade file and used for searching for a reusable data segment; a new upgrade file and the old upgrade file are analyzed, and a perfectly-matched area of the new and old upgrade files is obtained; with the perfectly-matched area as an interval and a set difference as a standard, areas with a certain proportion of similar content are expanded leftwards and rightwards respectively and are merged and recorded with the perfectly-matched area into a DIFF area, and other parts are recorded into an EXTRA area; the difference file is generated, wherein the difference file comprises a file description header and three compressed data segments. The invention further provides the upgrade method and system based on the differential file. The differential file generation method and the upgrade method and system based on the differential file have the advantages that by adopting a high-compression differential algorithmand just downloading a difference part, firmware upgrade can be completed, the upgrade flow consumption is reduced, and the upgrade speed is greatly increased.

Description

technical field [0001] The present invention relates to a terminal upgrade technology, in particular to a method for generating a differential file, and also to an upgrade method and system based on the differential file. Background technique [0002] In the process of embedded project development, it is often impossible to achieve the final design goal at one time, and it is necessary to continuously fix bugs or add some new functions during the running process, so online upgrades are inevitable. However, due to cost considerations, the embedded software and hardware environment cannot be very powerful. The traditional upgrade scheme requires a complete backup of the firmware, and the required program storage space often exceeds the system capacity limit, and a higher-end CPU has to be selected. Increases the hardware cost; in most application environments, the network speed is slow and the traffic cost is high. It takes a long time to download the firmware when performing ...

AI Technical Summary

Problems solved by technology

[0002] In the development process of embedded projects, it is often impossible to achieve the final design goal at one time, and it is necessary to continuously fix bugs or add some new functions during the running process, so online upgrades are inevitable

However, limited by cost considerations, the embedded software and hardware environment cannot be very powerful. The traditional upgrade solution requires a complete backup of the firmware, and the required program storage space often exceeds the system capacity limit, so a higher-end CPU has to be selected, which greatly Increased hardware costs; in most application environments, the network speed is slow, the traffic fee is high, and it takes a long time to download the firmware when performing a complete upgrade, which will have a relatively large impact on the business itself

[0003] In the prior art, the difference between the new and old versions of software is mainly the change of code or data, and the size of the difference information is usually much smaller than the size of the new version of the software, but due to the optimization factors of the compiler, the code or data segment may be shifted Or there are very small but a large number of scattered differences, so the difference algorithm needs to solve two main problems: one is how to efficiently find as many data segments as possible in the old version file that can be used to construct the new file version file, and the other is how to Minimize the number of bytes required to describe the conversion from an old version file to a new version file as much as possible

[0006] As can be seen from the above table, DicDiff has a great advantage in terms of compression rate. Using the suffix sorting algorithm for data matching can efficiently find out as much old data as possible for reuse. However, in actual embedded applications, the code Upgrade changes often introduce a large number of discrete small changes, such as figure 1 As shown, it is difficult to generate large continuous reusable data segments, therefore, it is difficult to obtain satisfactory results based on the DicDiff algorithm based on complete matching

Images