1GIT-CHECKOUT(1) Git Manual GIT-CHECKOUT(1)
2
6 git-checkout - Switch branches or restore working tree files
7
9 git checkout [-q] [-f] [-m] [<branch>]
10 git checkout [-q] [-f] [-m] --detach [<branch>]
11 git checkout [-q] [-f] [-m] [--detach] <commit>
12 git checkout [-q] [-f] [-m] [[-b|-B|--orphan] <new-branch>] [<start-point>]
13 git checkout [-f] <tree-ish> [--] <pathspec>...
14 git checkout [-f] <tree-ish> --pathspec-from-file=<file> [--pathspec-file-nul]
15 git checkout [-f|--ours|--theirs|-m|--conflict=<style>] [--] <pathspec>...
16 git checkout [-f|--ours|--theirs|-m|--conflict=<style>] --pathspec-from-file=<file> [--pathspec-file-nul]
17 git checkout (-p|--patch) [<tree-ish>] [--] [<pathspec>...]
18
20 Updates files in the working tree to match the version in the index or
21 the specified tree. If no pathspec was given, git checkout will also
22 update HEAD to set the specified branch as the current branch.
23 git checkout [<branch>]
25 To prepare for working on <branch>, switch to it by updating the
26 index and the files in the working tree, and by pointing HEAD at
27 the branch. Local modifications to the files in the working tree
28 are kept, so that they can be committed to the <branch>.
29 If <branch> is not found but there does exist a tracking branch in
31 exactly one remote (call it <remote>) with a matching name and
32 --no-guess is not specified, treat as equivalent to
33 $ git checkout -b <branch> --track <remote>/<branch>
35 You could omit <branch>, in which case the command degenerates to
37 "check out the current branch", which is a glorified no-op with
38 rather expensive side-effects to show only the tracking
39 information, if it exists, for the current branch.
40 git checkout -b|-B <new-branch> [<start-point>]
42 Specifying -b causes a new branch to be created as if git-branch(1)
43 were called and then checked out. In this case you can use the
44 --track or --no-track options, which will be passed to git branch.
45 As a convenience, --track without -b implies branch creation; see
46 the description of --track below.
47 If -B is given, <new-branch> is created if it doesn’t exist;
49 otherwise, it is reset. This is the transactional equivalent of
50 $ git branch -f <branch> [<start-point>]
52 $ git checkout <branch>
53 that is to say, the branch is not reset/created unless "git
55 checkout" is successful.
56 git checkout --detach [<branch>], git checkout [--detach] <commit>
58 Prepare to work on top of <commit>, by detaching HEAD at it (see
59 "DETACHED HEAD" section), and updating the index and the files in
60 the working tree. Local modifications to the files in the working
61 tree are kept, so that the resulting working tree will be the state
62 recorded in the commit plus the local modifications.
63 When the <commit> argument is a branch name, the --detach option
65 can be used to detach HEAD at the tip of the branch (git checkout
66 <branch> would check out that branch without detaching HEAD).
67 Omitting <branch> detaches HEAD at the tip of the current branch.
69 git checkout [-f|--ours|--theirs|-m|--conflict=<style>] [<tree-ish>]
71 [--] <pathspec>..., git checkout
72 [-f|--ours|--theirs|-m|--conflict=<style>] [<tree-ish>]
73 --pathspec-from-file=<file> [--pathspec-file-nul]
74 Overwrite the contents of the files that match the pathspec. When
75 the <tree-ish> (most often a commit) is not given, overwrite
76 working tree with the contents in the index. When the <tree-ish> is
77 given, overwrite both the index and the working tree with the
78 contents at the <tree-ish>.
79 The index may contain unmerged entries because of a previous failed
81 merge. By default, if you try to check out such an entry from the
82 index, the checkout operation will fail and nothing will be checked
83 out. Using -f will ignore these unmerged entries. The contents from
84 a specific side of the merge can be checked out of the index by
85 using --ours or --theirs. With -m, changes made to the working tree
86 file can be discarded to re-create the original conflicted merge
87 result.
88 git checkout (-p|--patch) [<tree-ish>] [--] [<pathspec>...]
90 This is similar to the previous mode, but lets you use the
91 interactive interface to show the "diff" output and choose which
92 hunks to use in the result. See below for the description of
93 --patch option.
94
96 -q, --quiet
97 Quiet, suppress feedback messages.
98 --progress, --no-progress
100 Progress status is reported on the standard error stream by default
101 when it is attached to a terminal, unless --quiet is specified.
102 This flag enables progress reporting even if not attached to a
103 terminal, regardless of --quiet.
104 -f, --force
106 When switching branches, proceed even if the index or the working
107 tree differs from HEAD, and even if there are untracked files in
108 the way. This is used to throw away local changes and any untracked
109 files or directories that are in the way.
110 When checking out paths from the index, do not fail upon unmerged
112 entries; instead, unmerged entries are ignored.
113 --ours, --theirs
115 When checking out paths from the index, check out stage #2 (ours)
116 or #3 (theirs) for unmerged paths.
117 Note that during git rebase and git pull --rebase, ours and theirs
119 may appear swapped; --ours gives the version from the branch the
120 changes are rebased onto, while --theirs gives the version from the
121 branch that holds your work that is being rebased.
122 This is because rebase is used in a workflow that treats the
124 history at the remote as the shared canonical one, and treats the
125 work done on the branch you are rebasing as the third-party work to
126 be integrated, and you are temporarily assuming the role of the
127 keeper of the canonical history during the rebase. As the keeper of
128 the canonical history, you need to view the history from the remote
129 as ours (i.e. "our shared canonical history"), while what you did
130 on your side branch as theirs (i.e. "one contributor’s work on top
131 of it").
132 -b <new-branch>
134 Create a new branch named <new-branch>, start it at <start-point>,
135 and check the resulting branch out; see git-branch(1) for details.
136 -B <new-branch>
138 Creates the branch <new-branch>, start it at <start-point>; if it
139 already exists, then reset it to <start-point>. And then check the
140 resulting branch out. This is equivalent to running "git branch"
141 with "-f" followed by "git checkout" of that branch; see git-
142 branch(1) for details.
143 -t, --track[=(direct|inherit)]
145 When creating a new branch, set up "upstream" configuration. See
146 "--track" in git-branch(1) for details.
147 If no -b option is given, the name of the new branch will be
149 derived from the remote-tracking branch, by looking at the local
150 part of the refspec configured for the corresponding remote, and
151 then stripping the initial part up to the "*". This would tell us
152 to use hack as the local branch when branching off of origin/hack
153 (or remotes/origin/hack, or even refs/remotes/origin/hack). If the
154 given name has no slash, or the above guessing results in an empty
155 name, the guessing is aborted. You can explicitly give a name with
156 -b in such a case.
157 --no-track
159 Do not set up "upstream" configuration, even if the
160 branch.autoSetupMerge configuration variable is true.
161 --guess, --no-guess
163 If <branch> is not found but there does exist a tracking branch in
164 exactly one remote (call it <remote>) with a matching name, treat
165 as equivalent to
166 $ git checkout -b <branch> --track <remote>/<branch>
168 If the branch exists in multiple remotes and one of them is named
170 by the checkout.defaultRemote configuration variable, we’ll use
171 that one for the purposes of disambiguation, even if the <branch>
172 isn’t unique across all remotes. Set it to e.g.
173 checkout.defaultRemote=origin to always checkout remote branches
174 from there if <branch> is ambiguous but exists on the origin
175 remote. See also checkout.defaultRemote in git-config(1).
176 --guess is the default behavior. Use --no-guess to disable it.
178 The default behavior can be set via the checkout.guess
180 configuration variable.
181 -l
183 Create the new branch’s reflog; see git-branch(1) for details.
184 -d, --detach
186 Rather than checking out a branch to work on it, check out a commit
187 for inspection and discardable experiments. This is the default
188 behavior of git checkout <commit> when <commit> is not a branch
189 name. See the "DETACHED HEAD" section below for details.
190 --orphan <new-branch>
192 Create a new orphan branch, named <new-branch>, started from
193 <start-point> and switch to it. The first commit made on this new
194 branch will have no parents and it will be the root of a new
195 history totally disconnected from all the other branches and
196 commits.
197 The index and the working tree are adjusted as if you had
199 previously run git checkout <start-point>. This allows you to start
200 a new history that records a set of paths similar to <start-point>
201 by easily running git commit -a to make the root commit.
202 This can be useful when you want to publish the tree from a commit
204 without exposing its full history. You might want to do this to
205 publish an open source branch of a project whose current tree is
206 "clean", but whose full history contains proprietary or otherwise
207 encumbered bits of code.
208 If you want to start a disconnected history that records a set of
210 paths that is totally different from the one of <start-point>, then
211 you should clear the index and the working tree right after
212 creating the orphan branch by running git rm -rf . from the top
213 level of the working tree. Afterwards you will be ready to prepare
214 your new files, repopulating the working tree, by copying them from
215 elsewhere, extracting a tarball, etc.
216 --ignore-skip-worktree-bits
218 In sparse checkout mode, git checkout -- <paths> would update only
219 entries matched by <paths> and sparse patterns in
220 $GIT_DIR/info/sparse-checkout. This option ignores the sparse
221 patterns and adds back any files in <paths>.
222 -m, --merge
224 When switching branches, if you have local modifications to one or
225 more files that are different between the current branch and the
226 branch to which you are switching, the command refuses to switch
227 branches in order to preserve your modifications in context.
228 However, with this option, a three-way merge between the current
229 branch, your working tree contents, and the new branch is done, and
230 you will be on the new branch.
231 When a merge conflict happens, the index entries for conflicting
233 paths are left unmerged, and you need to resolve the conflicts and
234 mark the resolved paths with git add (or git rm if the merge should
235 result in deletion of the path).
236 When checking out paths from the index, this option lets you
238 recreate the conflicted merge in the specified paths. This option
239 cannot be used when checking out paths from a tree-ish.
240 When switching branches with --merge, staged changes may be lost.
242 --conflict=<style>
244 The same as --merge option above, but changes the way the
245 conflicting hunks are presented, overriding the merge.conflictStyle
246 configuration variable. Possible values are "merge" (default),
247 "diff3", and "zdiff3".
248 -p, --patch
250 Interactively select hunks in the difference between the <tree-ish>
251 (or the index, if unspecified) and the working tree. The chosen
252 hunks are then applied in reverse to the working tree (and if a
253 <tree-ish> was specified, the index).
254 This means that you can use git checkout -p to selectively discard
256 edits from your current working tree. See the “Interactive Mode”
257 section of git-add(1) to learn how to operate the --patch mode.
258 Note that this option uses the no overlay mode by default (see also
260 --overlay), and currently doesn’t support overlay mode.
261 --ignore-other-worktrees
263 git checkout refuses when the wanted ref is already checked out by
264 another worktree. This option makes it check the ref out anyway. In
265 other words, the ref can be held by more than one worktree.
266 --overwrite-ignore, --no-overwrite-ignore
268 Silently overwrite ignored files when switching branches. This is
269 the default behavior. Use --no-overwrite-ignore to abort the
270 operation when the new branch contains ignored files.
271 --recurse-submodules, --no-recurse-submodules
273 Using --recurse-submodules will update the content of all active
274 submodules according to the commit recorded in the superproject. If
275 local modifications in a submodule would be overwritten the
276 checkout will fail unless -f is used. If nothing (or
277 --no-recurse-submodules) is used, submodules working trees will not
278 be updated. Just like git-submodule(1), this will detach HEAD of
279 the submodule.
280 --overlay, --no-overlay
282 In the default overlay mode, git checkout never removes files from
283 the index or the working tree. When specifying --no-overlay, files
284 that appear in the index and working tree, but not in <tree-ish>
285 are removed, to make them match <tree-ish> exactly.
286 --pathspec-from-file=<file>
288 Pathspec is passed in <file> instead of commandline args. If <file>
289 is exactly - then standard input is used. Pathspec elements are
290 separated by LF or CR/LF. Pathspec elements can be quoted as
291 explained for the configuration variable core.quotePath (see git-
292 config(1)). See also --pathspec-file-nul and global
293 --literal-pathspecs.
294 --pathspec-file-nul
296 Only meaningful with --pathspec-from-file. Pathspec elements are
297 separated with NUL character and all other characters are taken
298 literally (including newlines and quotes).
299 <branch>
301 Branch to checkout; if it refers to a branch (i.e., a name that,
302 when prepended with "refs/heads/", is a valid ref), then that
303 branch is checked out. Otherwise, if it refers to a valid commit,
304 your HEAD becomes "detached" and you are no longer on any branch
305 (see below for details).
306 You can use the @{-N} syntax to refer to the N-th last
308 branch/commit checked out using "git checkout" operation. You may
309 also specify - which is synonymous to @{-1}.
310 As a special case, you may use A...B as a shortcut for the merge
312 base of A and B if there is exactly one merge base. You can leave
313 out at most one of A and B, in which case it defaults to HEAD.
314 <new-branch>
316 Name for the new branch.
317 <start-point>
319 The name of a commit at which to start the new branch; see git-
320 branch(1) for details. Defaults to HEAD.
321 As a special case, you may use "A...B" as a shortcut for the merge
323 base of A and B if there is exactly one merge base. You can leave
324 out at most one of A and B, in which case it defaults to HEAD.
325 <tree-ish>
327 Tree to checkout from (when paths are given). If not specified, the
328 index will be used.
329 As a special case, you may use "A...B" as a shortcut for the merge
331 base of A and B if there is exactly one merge base. You can leave
332 out at most one of A and B, in which case it defaults to HEAD.
333 --
335 Do not interpret any more arguments as options.
336 <pathspec>...
338 Limits the paths affected by the operation.
339 For more details, see the pathspec entry in gitglossary(7).
341
343 HEAD normally refers to a named branch (e.g. master). Meanwhile, each
344 branch refers to a specific commit. Let’s look at a repo with three
345 commits, one of them tagged, and with branch master checked out:
346 HEAD (refers to branch 'master')
348 |
349 v
350 a---b---c branch 'master' (refers to commit 'c')
351 ^
352 |
353 tag 'v2.0' (refers to commit 'b')
354 When a commit is created in this state, the branch is updated to refer
356 to the new commit. Specifically, git commit creates a new commit d,
357 whose parent is commit c, and then updates branch master to refer to
358 new commit d. HEAD still refers to branch master and so indirectly now
359 refers to commit d:
360 $ edit; git add; git commit
362 HEAD (refers to branch 'master')
364 |
365 v
366 a---b---c---d branch 'master' (refers to commit 'd')
367 ^
368 |
369 tag 'v2.0' (refers to commit 'b')
370 It is sometimes useful to be able to checkout a commit that is not at
372 the tip of any named branch, or even to create a new commit that is not
373 referenced by a named branch. Let’s look at what happens when we
374 checkout commit b (here we show two ways this may be done):
375 $ git checkout v2.0 # or
377 $ git checkout master^^
378 HEAD (refers to commit 'b')
380 |
381 v
382 a---b---c---d branch 'master' (refers to commit 'd')
383 ^
384 |
385 tag 'v2.0' (refers to commit 'b')
386 Notice that regardless of which checkout command we use, HEAD now
388 refers directly to commit b. This is known as being in detached HEAD
389 state. It means simply that HEAD refers to a specific commit, as
390 opposed to referring to a named branch. Let’s see what happens when we
391 create a commit:
392 $ edit; git add; git commit
394 HEAD (refers to commit 'e')
396 |
397 v
398 e
399 /
400 a---b---c---d branch 'master' (refers to commit 'd')
401 ^
402 |
403 tag 'v2.0' (refers to commit 'b')
404 There is now a new commit e, but it is referenced only by HEAD. We can
406 of course add yet another commit in this state:
407 $ edit; git add; git commit
409 HEAD (refers to commit 'f')
411 |
412 v
413 e---f
414 /
415 a---b---c---d branch 'master' (refers to commit 'd')
416 ^
417 |
418 tag 'v2.0' (refers to commit 'b')
419 In fact, we can perform all the normal Git operations. But, let’s look
421 at what happens when we then checkout master:
422 $ git checkout master
424 HEAD (refers to branch 'master')
426 e---f |
427 / v
428 a---b---c---d branch 'master' (refers to commit 'd')
429 ^
430 |
431 tag 'v2.0' (refers to commit 'b')
432 It is important to realize that at this point nothing refers to commit
434 f. Eventually commit f (and by extension commit e) will be deleted by
435 the routine Git garbage collection process, unless we create a
436 reference before that happens. If we have not yet moved away from
437 commit f, any of these will create a reference to it:
438 $ git checkout -b foo # or "git switch -c foo" [1m(1)
440 $ git branch foo [1m(2)
441 $ git tag foo [1m(3)
442 1. creates a new branch foo, which refers to commit f, and
445 then updates HEAD to refer to branch foo. In other words,
446 we’ll no longer be in detached HEAD state after this
447 command.
448 2. similarly creates a new branch foo, which refers to commit
449 f, but leaves HEAD detached.
450 3. creates a new tag foo, which refers to commit f, leaving
451 HEAD detached.
452 If we have moved away from commit f, then we must first recover its
454 object name (typically by using git reflog), and then we can create a
455 reference to it. For example, to see the last two commits to which HEAD
456 referred, we can use either of these commands:
457 $ git reflog -2 HEAD # or
459 $ git log -g -2 HEAD
460
462 When there is only one argument given and it is not -- (e.g. git
463 checkout abc), and when the argument is both a valid <tree-ish> (e.g. a
464 branch abc exists) and a valid <pathspec> (e.g. a file or a directory
465 whose name is "abc" exists), Git would usually ask you to disambiguate.
466 Because checking out a branch is so common an operation, however, git
467 checkout abc takes "abc" as a <tree-ish> in such a situation. Use git
468 checkout -- <pathspec> if you want to checkout these paths out of the
469 index.
470
472 1. Paths
473 The following sequence checks out the master branch, reverts the
474 Makefile to two revisions back, deletes hello.c by mistake, and gets it
475 back from the index.
476 $ git checkout master [1m(1)
478 $ git checkout master~2 Makefile [1m(2)
479 $ rm -f hello.c
480 $ git checkout hello.c [1m(3)
481 1. switch branch
484 2. take a file out of another commit
485 3. restore hello.c from the index
486 If you want to check out all C source files out of the index, you can
488 say
489 $ git checkout -- '*.c'
491 Note the quotes around *.c. The file hello.c will also be checked out,
493 even though it is no longer in the working tree, because the file
494 globbing is used to match entries in the index (not in the working tree
495 by the shell).
496 If you have an unfortunate branch that is named hello.c, this step
498 would be confused as an instruction to switch to that branch. You
499 should instead write:
500 $ git checkout -- hello.c
502 2. Merge
504 After working in the wrong branch, switching to the correct branch
505 would be done using:
506 $ git checkout mytopic
508 However, your "wrong" branch and correct mytopic branch may differ in
510 files that you have modified locally, in which case the above checkout
511 would fail like this:
512 $ git checkout mytopic
514 error: You have local changes to 'frotz'; not switching branches.
515 You can give the -m flag to the command, which would try a three-way
517 merge:
518 $ git checkout -m mytopic
520 Auto-merging frotz
521 After this three-way merge, the local modifications are not registered
523 in your index file, so git diff would show you what changes you made
524 since the tip of the new branch.
525 3. Merge conflict
527 When a merge conflict happens during switching branches with the -m
528 option, you would see something like this:
529 $ git checkout -m mytopic
531 Auto-merging frotz
532 ERROR: Merge conflict in frotz
533 fatal: merge program failed
534 At this point, git diff shows the changes cleanly merged as in the
536 previous example, as well as the changes in the conflicted files. Edit
537 and resolve the conflict and mark it resolved with git add as usual:
538 $ edit frotz
540 $ git add frotz
541
543 Everything below this line in this section is selectively included from
544 the git-config(1) documentation. The content is the same as what’s
545 found there:
546 checkout.defaultRemote
548 When you run git checkout <something> or git switch <something> and
549 only have one remote, it may implicitly fall back on checking out
550 and tracking e.g. origin/<something>. This stops working as soon
551 as you have more than one remote with a <something> reference. This
552 setting allows for setting the name of a preferred remote that
553 should always win when it comes to disambiguation. The typical
554 use-case is to set this to origin.
555 Currently this is used by git-switch(1) and git-checkout(1) when
557 git checkout <something> or git switch <something> will checkout
558 the <something> branch on another remote, and by git-worktree(1)
559 when git worktree add refers to a remote branch. This setting might
560 be used for other checkout-like commands or functionality in the
561 future.
562 checkout.guess
564 Provides the default value for the --guess or --no-guess option in
565 git checkout and git switch. See git-switch(1) and git-checkout(1).
566 checkout.workers
568 The number of parallel workers to use when updating the working
569 tree. The default is one, i.e. sequential execution. If set to a
570 value less than one, Git will use as many workers as the number of
571 logical cores available. This setting and
572 checkout.thresholdForParallelism affect all commands that perform
573 checkout. E.g. checkout, clone, reset, sparse-checkout, etc.
574 Note: Parallel checkout usually delivers better performance for
576 repositories located on SSDs or over NFS. For repositories on
577 spinning disks and/or machines with a small number of cores, the
578 default sequential checkout often performs better. The size and
579 compression level of a repository might also influence how well the
580 parallel version performs.
581 checkout.thresholdForParallelism
583 When running parallel checkout with a small number of files, the
584 cost of subprocess spawning and inter-process communication might
585 outweigh the parallelization gains. This setting allows you to
586 define the minimum number of files for which parallel checkout
587 should be attempted. The default is 100.
588
590 git-switch(1), git-restore(1)
591
593 Part of the git(1) suite
594Git 2.43.0 11/20/2023 GIT-CHECKOUT(1)