Torvalds wanted a distributed system that he could use like BitKeeper, but none of the available free systems met his needs, particularly in terms of performance. Torvalds took an example of an SCM system requiring thirty seconds to apply a patch and update all associated metadata, and noted that this would not scale to the needs of Linux kernel development, where syncing with fellow maintainers could require 250 such actions at a time. He wanted patching to take three seconds, and had several other design criteria in mind:
very strong safeguards against corruption, either accidental or malicious.
These three criteria eliminated every then-existing version control system, except for Monotone. Considering performance excluded this too. So immediately after the 2.6.12-rc2 Linux kernel development release, Torvalds set out to write his own.
Torvalds has quipped about the name git, which is British English slang roughly equivalent to "unpleasant person". Torvalds said: "I'm an egotistical bastard, and I name all my projects after myself. First 'Linux', now 'git'." The man page describes git as "the stupid content tracker".
The development of Git began on 3 April 2005. The project was announced on 6 April, and became self-hosting as of 7 April. The first merge of multiple branches was done on 18 April. Torvalds achieved his performance goals; on 29 April, the nascent Git was benchmarked recording patches to the Linux kernel tree at the rate of 6.7 per second. On 16 June Git managed the kernel 2.6.12 release. Torvalds turned over maintenance on 26 July 2005 to Junio Hamano, a major contributor to the project. Hamano was responsible for the 1.0 release on 21 December 2005, and remains the project's maintainer.
Git's design was inspired by BitKeeper and Monotone. Git was originally designed as a low-level version control system engine on top of which others could write front ends, such as Cogito or StGIT. The core Git project has since become a complete version control system that is usable directly. While strongly influenced by BitKeeper, Torvalds deliberately attempted to avoid conventional approaches, leading to a unique design.
Git's design is a synthesis of Torvalds's experience with Linux in maintaining a large distributed development project, along with his intimate knowledge of file system performance gained from the same project and the urgent need to produce a working system in short order. These influences led to the following implementation choices:
Strong support for non-linear development
Git supports rapid branching and merging, and includes specific tools for visualizing and navigating a non-linear development history. A core assumption in Git is that a change will be merged more often than it is written, as it is passed around various reviewers. Branches in git are very lightweight: A branch in git is only a reference to a single commit. With its parental commits, the full branch structure can be constructed.
Like Darcs, BitKeeper, Mercurial, SVK, Bazaar and Monotone, Git gives each developer a local copy of the entire development history, and changes are copied from one such repository to another. These changes are imported as additional development branches, and can be merged in the same way as a locally developed branch.
Compatibility with existing systems/protocols
Repositories can be published via HTTP, FTP, rsync, or a Git protocol over either a plain socket, or ssh. Git also has a CVS server emulation, which enables the use of existing CVS clients and IDE plugins to access Git repositories. Subversion and svk repositories can be used directly with git-svn.
Efficient handling of large projects
Torvalds has described Git as being very fast and scalable, and performance tests done by Mozilla showed it was an order of magnitude faster than some version-control systems, and fetching version history from a locally stored repository can be one hundred times faster than fetching it from the remote server.
Cryptographic authentication of history
The Git history is stored in such a way that the ID of a particular version (a commit in Git terms) depends upon the complete development history leading up to that commit. Once it is published, it is not possible to change the old versions without it being noticed. The structure is similar to a Merkle tree, but with additional data at the nodes as well as the leaves. (Mercurial and Monotone also have this property.)
Git was designed as a set of programs written in C, and a number of shell scripts that provide wrappers around those programs. Although most of those scripts have since been rewritten in C for speed and portability, the design remains, and it is easy to chain the components together.
Pluggable merge strategies
As part of its toolkit design, Git has a well-defined model of an incomplete merge, and it has multiple algorithms for completing it, culminating in telling the user that it is unable to complete the merge automatically and that manual editing is required.
Aborting operations or backing out changes will leave useless dangling objects in the database. These are generally a small fraction of the continuously growing history of wanted objects. Git will automatically perform garbage collection when enough loose objects have been created in the repository. Garbage collection can be called explicitly using git gc --prune.
Periodic explicit object packing
Git stores each newly created object as a separate file. Although individually compressed, this takes a great deal of space and is inefficient. This is solved by the use of packs that store a large number of objects in a single file (or network byte stream) called packfile, delta-compressed among themselves. Packs are compressed using the heuristic that files with the same name are probably similar, but do not depend on it for correctness. A corresponding index file is created for each packfile, telling the offset of each object in the packfile. Newly created objects (newly added history) are still stored singly, and periodic repacking is required to maintain space efficiency. The process of packing the repository can be very computationally expensive. By allowing objects to exist in the repository in a loose, but quickly generated format, git allows the expensive pack operation to be deferred until later when time does not matter (e.g. the end of the work day). Git does periodic repacking automatically but manual repacking is also possible with the git gc command. For data integrity, both packfile and its index have SHA-1 checksum inside, and also the file name of packfile contains a SHA-1 checksum. To check integrity, run the git fsck command.
Another property of Git is that it snapshots directory trees of files. The earliest systems for tracking versions of source code, SCCS and RCS, worked on individual files and emphasized the space savings to be gained from interleaved deltas (SCCS) or delta encoding (RCS) the (mostly similar) versions. Later revision control systems maintained this notion of a file having an identity across multiple revisions of a project. However, Torvalds rejected this concept. Consequently, Git does not explicitly record file revision relationships at any level below the source code tree.
Implicit revision relationships have some significant consequences:
It is slightly more expensive to examine the change history of a single file than the whole project. To obtain a history of changes affecting a given file, Git must walk the global history and then determine whether each change modified that file. This method of examining history does, however, let Git produce with equal efficiency a single history showing the changes to an arbitrary set of files. For example, a subdirectory of the source tree plus an associated global header file is a very common case.
Renames are handled implicitly rather than explicitly. A common complaint with CVS is that it uses the name of a file to identify its revision history, so moving or renaming a file is not possible without either interrupting its history, or renaming the history and thereby making the history inaccurate. Most post-CVS revision control systems solve this by giving a file a unique long-lived name (a sort of inode number) that survives renaming. Git does not record such an identifier, and this is claimed as an advantage.Source code files are sometimes split or merged as well as simply renamed, and recording this as a simple rename would freeze an inaccurate description of what happened in the (immutable) history. Git addresses the issue by detecting renames while browsing the history of snapshots rather than recording it when making the snapshot. (Briefly, given a file in revision N, a file of the same name in revision N−1 is its default ancestor. However, when there is no like-named file in revision N−1, Git searches for a file that existed only in revision N−1 and is very similar to the new file.) However, it does require more CPU-intensive work every time history is reviewed, and a number of options to adjust the heuristics. This mechanism does not always work; sometimes a file that is renamed with changes in the same commit is read as a deletion of the old file and the creation of a new file. Developers can work around this limitation by committing the rename and changes separately.
Git implements several merging strategies; a non-default can be selected at merge time:
recursive: This is the default when pulling or merging one branch, and is a variant of the three-way merge algorithm.
When there are more than one common ancestors that can be used for three-way merge, it creates a merged tree of the common ancestors and uses that as the reference tree for the three-way merge. This has been reported to result in fewer merge conflicts without causing mis-merges by tests done on actual merge commits taken from Linux 2.6 kernel development history. Additionally this can detect and handle merges involving renames.
In many ways you can just see git as a filesystem — it is content-addressable, and it has a notion of versioning, but I really really designed it coming at the problem from the viewpoint of a filesystem person (hey, kernels is what I do), and I actually have absolutely zero interest in creating a traditional SCM system.
From this initial design approach, Git has developed the full set of features expected of a traditional SCM, with features mostly being created as needed, then refined and extended over time.
Some data flows and storage levels in the Git revision control system.
Git has two data structures: a mutable index (also called stage or cache) that caches information about the working directory and the next revision to be committed; and an immutable, append-only object database.
The object database contains four types of objects:
A tree object is the equivalent of a directory. It contains a list of file names, each with some type bits and the name of a blob or tree object that is that file, symbolic link, or directory's contents. This object describes a snapshot of the source tree.
A commit object links tree objects together into a history. It contains the name of a tree object (of the top-level source directory), a time stamp, a log message, and the names of zero or more parent commit objects.
A tag object is a container that contains reference to another object and can hold additional meta-data related to another object. Most commonly, it is used to store a digital signature of a commit object corresponding to a particular release of the data being tracked by Git.
The index serves as connection point between the object database and the working tree.
Each object is identified by a SHA-1hash of its contents. Git computes the hash, and uses this value for the object's name. The object is put into a directory matching the first two characters of its hash. The rest of the hash is used as the file name for that object.
Git stores each revision of a file as a unique blob. The relationships between the blobs can be found through examining the tree and commit objects. Newly added objects are stored in their entirety using zlib compression. This can consume a large amount of disk space quickly, so objects can be combined into packs, which use delta compression to save space, storing blobs as their changes relative to other blobs.
Every object in the git database which is not referred to may be cleaned up by using a garbage collection command, or automatically. An object may be referenced by another object, or an explicit reference. Git knows different types of references. The commands to create, move, and delete references vary. "git show-ref" lists all references. Some types are:
refers to an object locally.
refers to an object which exists in a remote repository.
The JGit implementation of Git is a pure Java software library, designed to be embedded in any Java application. JGit is used in the Gerrit code review tool and in EGit, a Git client for the Eclipse IDE.
The Dulwich implementation of Git is a pure Python software component for Python 2.
The libgit2 implementation of Git is an ANSI C software library with no other dependencies, which can be built on multiple platforms including Microsoft Windows, Linux, Mac OS X, and BSD. It has bindings for many programming languages, including Ruby, Python and Haskell.
As git is a distributed version control system, it can be used as server out of the box. Dedicated git server software helps, amongst other features, to add access control, display the contents of a git repository via web, and help managing multiple repositories.
Remote file store and shell access
A git repository can be cloned to a shared file system, and accessed by other persons. It can also be accessed via remote shell just by having the git software installed and allowing a user to log in.
Git daemon, instaweb
Git daemon allows users to share their own repository to colleagues quickly. Git instaweb allows users to provide web view to the repository. As of 2014-04 instaweb does not work on Windows. Both can be seen in the line of Mercurial's "hg serve".
Gitolite is an access control layer on top of git, providing fine access control to git repositories. It relies on other software to remotely view the repositories on the server.
Gerrit provides two out of three functionalities: access control, and managing repositories. It uses jGit. To view repositories it is combined e.g. with Gitiles or GitBlit.
Gitblit can provide all three functions, but is in larger installations used as repository browser installed with gerrit for access control and management of repositories.
Gitiles is a simple repository browser, usually used together with gerrit.
Bonobo Git Server
Bonobo Git Server is a simple git server for Windows implemented as an ASP.NET gateway. It relies on the authentication mechanisms provided by Windows Internet Information Services, thus it does not support SSH access but can be easily integrated with Active Directory.
The Eclipse Foundation reported in its annual community survey that as of May 2014, Git is now the most widely used source code management tool, with 33.3% of professional software developers reporting that they use Git or GitHub as their primary source control system compared with 30.3% in 2013, 27.6% in 2012 and 12.8% in 2011. Open source directory Black Duck Open Hub reports a similar uptake among open source projects.
^"After controversy, Torvalds begins work on 'git'". PC World. 2012-07-14. "Torvalds seemed aware that his decision to drop BitKeeper would also be controversial. When asked why he called the new software, "git", British slang meaning "a rotten person", he said. "I'm an egotistical bastard, so I name all my projects after myself. First Linux, now git""