Patch (computing)

A patch is data for modifying an existing software resource such as a program or a file, often to fix bugs and security vulnerabilities.^[1][2] To patch is also the process of applying the data to the existing resource. Patching a system involves applying a patch. A patch may be created to improve functionality, usability, or performance. A patch may be created manually, but commonly it is created via a tool that compares two versions of the resource and generates data that can be used to transform one to the other.

Typically, a patch needs to be applied to the specific version of the resource it is intended to modify, although there are exceptions. Some patching tools can detect the version of the existing resource and apply the appropriate patch, even if it supports multiple versions. As more patches are released, their cumulative size can grow significantly, sometimes exceeding the size of the resource itself. To manage this, the number of supported versions may be limited, or a complete copy of the resource might be provided instead.

Patching allows for modifying a binaryexecutable. Although this can be technically challenging (requires a thorough understanding of the workings of the executable), it may be feasible when the source code is unavailable to build a full executable, and it allows for a smaller distribution which can be more economical than distributing full files.

Although often intended to fix problems, a patch can introduce new problems – a scenario called software regression. In some cases, an update intentionally disables functionality, for instance, by removing aspects for which the consumer is no longer licensed. Patch management is a part of lifecycle management, and involves a strategy and planning of what patches should be applied to which systems and at what times. Typically, a patch is applied in a permanent way (i.e. to storage), but in some cases, a patch is applied to memory (i.e. via a tool such as a debugger) in which case the change is lost when the resource is reloaded from storage.

Software update is sometimes conflated with patch even though they are not synonyms. An update can be implemented using patch files and the patching process. Also, some may contend that patching is not limited to modifying file content – that adding, removing and replacing whole files is patching. Typically, patch connotates a relatively small change, so a patch that is large in size or scope may be called the more general software update or another more specific name such as service pack. Windows NT and its successors (including Windows 2000, Windows XP, Windows Vista and Windows 7) use service pack.^[3] Historically, IBM used the terms FixPak and Corrective Service Diskette for such updates.^[4]

Historically, software suppliers distributed patches on paper tape or on punched cards, expecting the recipient to cut out the indicated part of the original tape (or deck), and patch in (hence the name) the replacement segment. Later patch distributions used magnetic tape. Then, after the invention of removable disk drives, patches came from the software developer via a disk or, later, CD-ROM via mail. With widely available Internet access, downloading patches from the developer's web site or through automated software updates became often available to the end-users. Starting with Apple's Mac OS 9 and Microsoft's Windows ME, PC operating systems gained the ability to get automatic software updates via the Internet.

Computer programs can often coordinate patches to update a target program. Automation simplifies the end-user's task, allowing them to only execute an update program, whereupon that program makes sure that updating the target takes place completely and correctly. Service packs for Microsoft Windows NT and its successors and for many commercial software products adopt such automated strategies.

Some programs can update themselves via the Internet with very little or no intervention on the part of users. The maintenance of server software and of operating systems often takes place in this manner. In situations where system administrators control a number of computers, this sort of automation helps to maintain consistency. The application of security patches commonly occurs in this manner.

With the advent of larger storage media and higher Internet bandwidth, it became common to replace entire files (or even all of a program's files) rather than modifying existing files, especially for smaller programs.

• Automatic bug fixing – Automatic repair of software bugs
• Backporting – Developing an update for older software based on an update to a newer version
• Delta encoding – Type of data transmission method
• Dribbleware
• Monkey patch – Technique to dynamically modify runtime code
• Patch (Unix) – Command for applying changes to text files
• Porting – Process of adapting software to run in a different context
• Software maintenance – Modification of software after delivery
• Software release life cycle – Stages in development and support of computer software
• SMP/E – Software management utility for mainframes
• Upgrade – Process of improving something by replacing part of it or adding additional parts
• Vulnerability database – List of computer security vulnerabilities
• White hat (computer security) – Computer hacker who hacks ethically

• The Jargon File entry for patch
• Official Linux kernel patch format
• 0-Day Patch – Exposing Vendors (In)security Performance. Archived 2014-01-31 at the Wayback Machine. A metric comparing patch performance of Microsoft and Apple.