6 git-reset - Reset current HEAD to the specified state
11 'git reset' [-q] [<tree-ish>] [--] <paths>...
12 'git reset' (--patch | -p) [<tree-ish>] [--] [<paths>...]
13 'git reset' [--soft | --mixed [-N] | --hard | --merge | --keep] [-q] [<commit>]
17 In the first and second form, copy entries from `<tree-ish>` to the index.
18 In the third form, set the current branch head (`HEAD`) to `<commit>`,
19 optionally modifying index and working tree to match.
20 The `<tree-ish>`/`<commit>` defaults to `HEAD` in all forms.
22 'git reset' [-q] [<tree-ish>] [--] <paths>...::
23 This form resets the index entries for all `<paths>` to their
24 state at `<tree-ish>`. (It does not affect the working tree or
27 This means that `git reset <paths>` is the opposite of `git add
28 <paths>`. This command is equivalent to
29 `git restore [--source=<tree-ish>] --staged <paths>...`.
31 After running `git reset <paths>` to update the index entry, you can
32 use linkgit:git-restore[1] to check the contents out of the index to
33 the working tree. Alternatively, using linkgit:git-restore[1]
34 and specifying a commit with `--source`, you
35 can copy the contents of a path out of a commit to the index and to the
36 working tree in one go.
38 'git reset' (--patch | -p) [<tree-ish>] [--] [<paths>...]::
39 Interactively select hunks in the difference between the index
40 and `<tree-ish>` (defaults to `HEAD`). The chosen hunks are applied
41 in reverse to the index.
43 This means that `git reset -p` is the opposite of `git add -p`, i.e.
44 you can use it to selectively reset hunks. See the ``Interactive Mode''
45 section of linkgit:git-add[1] to learn how to operate the `--patch` mode.
47 'git reset' [<mode>] [<commit>]::
48 This form resets the current branch head to `<commit>` and
49 possibly updates the index (resetting it to the tree of `<commit>`) and
50 the working tree depending on `<mode>`. If `<mode>` is omitted,
51 defaults to `--mixed`. The `<mode>` must be one of the following:
55 Does not touch the index file or the working tree at all (but
56 resets the head to `<commit>`, just like all modes do). This leaves
57 all your changed files "Changes to be committed", as `git status`
61 Resets the index but not the working tree (i.e., the changed files
62 are preserved but not marked for commit) and reports what has not
63 been updated. This is the default action.
65 If `-N` is specified, removed paths are marked as intent-to-add (see
69 Resets the index and working tree. Any changes to tracked files in the
70 working tree since `<commit>` are discarded.
73 Resets the index and updates the files in the working tree that are
74 different between `<commit>` and `HEAD`, but keeps those which are
75 different between the index and working tree (i.e. which have changes
76 which have not been added).
77 If a file that is different between `<commit>` and the index has
78 unstaged changes, reset is aborted.
80 In other words, `--merge` does something like a `git read-tree -u -m <commit>`,
81 but carries forward unmerged index entries.
84 Resets index entries and updates files in the working tree that are
85 different between `<commit>` and `HEAD`.
86 If a file that is different between `<commit>` and `HEAD` has local
87 changes, reset is aborted.
90 See "Reset, restore and revert" in linkgit:git[1] for the differences
91 between the three commands.
100 Be quiet, only report errors. The default behavior is set by the
101 `reset.quiet` config option. `--quiet` and `--no-quiet` will
102 override the default behavior.
112 $ git add frotz.c filfre.c
115 $ git pull git://info.example.com/ nitfol <4>
118 <1> You are happily working on something, and find the changes
119 in these files are in good order. You do not want to see them
120 when you run `git diff`, because you plan to work on other files
121 and changes with these files are distracting.
122 <2> Somebody asks you to pull, and the changes sound worthy of merging.
123 <3> However, you already dirtied the index (i.e. your index does
124 not match the `HEAD` commit). But you know the pull you are going
125 to make does not affect `frotz.c` or `filfre.c`, so you revert the
126 index changes for these two files. Your changes in working tree
128 <4> Then you can pull and merge, leaving `frotz.c` and `filfre.c`
129 changes still in the working tree.
131 Undo a commit and redo::
135 $ git reset --soft HEAD^ <1>
137 $ git commit -a -c ORIG_HEAD <3>
140 <1> This is most often done when you remembered what you
141 just committed is incomplete, or you misspelled your commit
142 message, or both. Leaves working tree as it was before "reset".
143 <2> Make corrections to working tree files.
144 <3> "reset" copies the old head to `.git/ORIG_HEAD`; redo the
145 commit by starting with its log message. If you do not need to
146 edit the message further, you can give `-C` option instead.
148 See also the `--amend` option to linkgit:git-commit[1].
150 Undo a commit, making it a topic branch::
153 $ git branch topic/wip <1>
154 $ git reset --hard HEAD~3 <2>
155 $ git switch topic/wip <3>
158 <1> You have made some commits, but realize they were premature
159 to be in the `master` branch. You want to continue polishing
160 them in a topic branch, so create `topic/wip` branch off of the
162 <2> Rewind the master branch to get rid of those three commits.
163 <3> Switch to `topic/wip` branch and keep working.
165 Undo commits permanently::
169 $ git reset --hard HEAD~3 <1>
172 <1> The last three commits (`HEAD`, `HEAD^`, and `HEAD~2`) were bad
173 and you do not want to ever see them again. Do *not* do this if
174 you have already given these commits to somebody else. (See the
175 "RECOVERING FROM UPSTREAM REBASE" section in linkgit:git-rebase[1]
176 for the implications of doing so.)
178 Undo a merge or pull::
183 CONFLICT (content): Merge conflict in nitfol
184 Automatic merge failed; fix conflicts and then commit the result.
185 $ git reset --hard <2>
186 $ git pull . topic/branch <3>
187 Updating from 41223... to 13134...
189 $ git reset --hard ORIG_HEAD <4>
192 <1> Try to update from the upstream resulted in a lot of
193 conflicts; you were not ready to spend a lot of time merging
194 right now, so you decide to do that later.
195 <2> "pull" has not made merge commit, so `git reset --hard`
196 which is a synonym for `git reset --hard HEAD` clears the mess
197 from the index file and the working tree.
198 <3> Merge a topic branch into the current branch, which resulted
200 <4> But you decided that the topic branch is not ready for public
201 consumption yet. "pull" or "merge" always leaves the original
202 tip of the current branch in `ORIG_HEAD`, so resetting hard to it
203 brings your index file and the working tree back to that state,
204 and resets the tip of the branch to that commit.
206 Undo a merge or pull inside a dirty working tree::
211 Merge made by recursive.
212 nitfol | 20 +++++----
214 $ git reset --merge ORIG_HEAD <2>
217 <1> Even if you may have local modifications in your
218 working tree, you can safely say `git pull` when you know
219 that the change in the other branch does not overlap with
221 <2> After inspecting the result of the merge, you may find
222 that the change in the other branch is unsatisfactory. Running
223 `git reset --hard ORIG_HEAD` will let you go back to where you
224 were, but it will discard your local changes, which you do not
225 want. `git reset --merge` keeps your local changes.
228 Interrupted workflow::
230 Suppose you are interrupted by an urgent fix request while you
231 are in the middle of a large change. The files in your
232 working tree are not in any shape to be committed yet, but you
233 need to get to the other branch for a quick bugfix.
236 $ git switch feature ;# you were working in "feature" branch and
237 $ work work work ;# got interrupted
238 $ git commit -a -m "snapshot WIP" <1>
241 $ git commit ;# commit with real log
243 $ git reset --soft HEAD^ ;# go back to WIP state <2>
247 <1> This commit will get blown away so a throw-away log message is OK.
248 <2> This removes the 'WIP' commit from the commit history, and sets
249 your working tree to the state just before you made that snapshot.
250 <3> At this point the index file still has all the WIP changes you
251 committed as 'snapshot WIP'. This updates the index to show your
252 WIP files as uncommitted.
254 See also linkgit:git-stash[1].
256 Reset a single file in the index::
258 Suppose you have added a file to your index, but later decide you do not
259 want to add it to your commit. You can remove the file from the index
260 while keeping your changes with git reset.
263 $ git reset -- frotz.c <1>
264 $ git commit -m "Commit files in index" <2>
265 $ git add frotz.c <3>
268 <1> This removes the file from the index while keeping it in the working
270 <2> This commits all other changes in the index.
271 <3> Adds the file to the index again.
273 Keep changes in working tree while discarding some previous commits::
275 Suppose you are working on something and you commit it, and then you
276 continue working a bit more, but now you think that what you have in
277 your working tree should be in another branch that has nothing to do
278 with what you committed previously. You can start a new branch and
279 reset it while keeping the changes in your working tree.
283 $ git switch -c branch1
287 $ git switch -c branch2 <2>
288 $ git reset --keep start <3>
291 <1> This commits your first edits in `branch1`.
292 <2> In the ideal world, you could have realized that the earlier
293 commit did not belong to the new topic when you created and switched
294 to `branch2` (i.e. `git switch -c branch2 start`), but nobody is
296 <3> But you can use `reset --keep` to remove the unwanted commit after
297 you switched to `branch2`.
299 Split a commit apart into a sequence of commits::
301 Suppose that you have created lots of logically separate changes and committed
302 them together. Then, later you decide that it might be better to have each
303 logical chunk associated with its own commit. You can use git reset to rewind
304 history without changing the contents of your local files, and then successively
305 use `git add -p` to interactively select which hunks to include into each commit,
306 using `git commit -c` to pre-populate the commit message.
309 $ git reset -N HEAD^ <1>
311 $ git diff --cached <3>
312 $ git commit -c HEAD@{1} <4>
315 $ git diff --cached <7>
319 <1> First, reset the history back one commit so that we remove the original
320 commit, but leave the working tree with all the changes. The -N ensures
321 that any new files added with `HEAD` are still marked so that `git add -p`
323 <2> Next, we interactively select diff hunks to add using the `git add -p`
324 facility. This will ask you about each diff hunk in sequence and you can
325 use simple commands such as "yes, include this", "No don't include this"
326 or even the very powerful "edit" facility.
327 <3> Once satisfied with the hunks you want to include, you should verify what
328 has been prepared for the first commit by using `git diff --cached`. This
329 shows all the changes that have been moved into the index and are about
331 <4> Next, commit the changes stored in the index. The `-c` option specifies to
332 pre-populate the commit message from the original message that you started
333 with in the first commit. This is helpful to avoid retyping it. The
334 `HEAD@{1}` is a special notation for the commit that `HEAD` used to be at
335 prior to the original reset commit (1 change ago).
336 See linkgit:git-reflog[1] for more details. You may also use any other
337 valid commit reference.
338 <5> You can repeat steps 2-4 multiple times to break the original code into
339 any number of commits.
340 <6> Now you've split out many of the changes into their own commits, and might
341 no longer use the patch mode of `git add`, in order to select all remaining
343 <7> Once again, check to verify that you've included what you want to. You may
344 also wish to verify that git diff doesn't show any remaining changes to be
346 <8> And finally create the final commit.
352 The tables below show what happens when running:
355 git reset --option target
358 to reset the `HEAD` to another commit (`target`) with the different
359 reset options depending on the state of the files.
361 In these tables, `A`, `B`, `C` and `D` are some different states of a
362 file. For example, the first line of the first table means that if a
363 file is in state `A` in the working tree, in state `B` in the index, in
364 state `C` in `HEAD` and in state `D` in the target, then `git reset --soft
365 target` will leave the file in the working tree in state `A` and in the
366 index in state `B`. It resets (i.e. moves) the `HEAD` (i.e. the tip of
367 the current branch, if you are on one) to `target` (which has the file
371 working index HEAD target working index HEAD
372 ----------------------------------------------------
381 working index HEAD target working index HEAD
382 ----------------------------------------------------
391 working index HEAD target working index HEAD
392 ----------------------------------------------------
401 working index HEAD target working index HEAD
402 ----------------------------------------------------
411 working index HEAD target working index HEAD
412 ----------------------------------------------------
421 working index HEAD target working index HEAD
422 ----------------------------------------------------
430 `reset --merge` is meant to be used when resetting out of a conflicted
431 merge. Any mergy operation guarantees that the working tree file that is
432 involved in the merge does not have a local change with respect to the index
433 before it starts, and that it writes the result out to the working tree. So if
434 we see some difference between the index and the target and also
435 between the index and the working tree, then it means that we are not
436 resetting out from a state that a mergy operation left after failing
437 with a conflict. That is why we disallow `--merge` option in this case.
439 `reset --keep` is meant to be used when removing some of the last
440 commits in the current branch while keeping changes in the working
441 tree. If there could be conflicts between the changes in the commit we
442 want to remove and the changes in the working tree we want to keep,
443 the reset is disallowed. That's why it is disallowed if there are both
444 changes between the working tree and `HEAD`, and between `HEAD` and the
445 target. To be safe, it is also disallowed when there are unmerged
448 The following tables show what happens when there are unmerged
452 working index HEAD target working index HEAD
453 ----------------------------------------------------
454 X U A B --soft (disallowed)
462 working index HEAD target working index HEAD
463 ----------------------------------------------------
464 X U A A --soft (disallowed)
471 `X` means any state and `U` means an unmerged index.
475 Part of the linkgit:git[1] suite