231

I received a git checkout from someone else and am trying to commit the unstaged changes to the local repository. However, a lot (if not every) file appears as modified even though the contents are exactly the same.

I already set core.fileMode to false and also set core.autocrlf to false, without success.

Worth mentioning is that the Git repo I received was from someone using Windows, while I use Linux.

What can I do to commit the actual changes?

EDIT: output of git config -l:

user.name=Aron Rotteveel
user.email=<removed>
color.diff=auto
color.status=auto
color.branch=auto
color.interactive=auto
color.ui=true
color.pager=true
color.branch.current=yellow reverse
color.branch.local=yellow
color.branch.remote=green
color.diff.meta=yellow bold
color.diff.frag=magenta bold
color.diff.old=red bold
color.diff.new=green bold
color.status.added=yellow
color.status.changed=green
color.status.untracked=cyan
core.pager=less -FRSX
core.whitespace=fix,-indent-with-non-tab,trailing-space,cr-at-eol
alias.co=checkout
core.repositoryformatversion=0
core.filemode=false
core.bare=false
core.logallrefupdates=true
core.symlinks=false
core.ignorecase=true
core.hidedotfiles=dotGitOnly
core.autocrlf=false
remote.origin.url=<removed>
remote.origin.fetch=+refs/heads/*:refs/remotes/origin/*

Update: added some random example files. These files are just plaintext, so are the easiest to include.

Original files are located here: https://gist.github.com/c3c5302430935155ef3d. Hexdumps definately indicate that the files are different, but I have no clue what causes this, and how to fix it.

HEAD version:

0000000: 4854 4d4c 2e53 6166 654f 626a 6563 740d  HTML.SafeObject.
0000010: 0a54 5950 453a 2062 6f6f 6c0d 0a56 4552  .TYPE: bool..VER
0000020: 5349 4f4e 3a20 332e 312e 310d 0a44 4546  SION: 3.1.1..DEF
0000030: 4155 4c54 3a20 6661 6c73 650d 0a2d 2d44  AULT: false..--D
0000040: 4553 4352 4950 5449 4f4e 2d2d 0d0a 3c70  ESCRIPTION--..<p
0000050: 3e0d 0a20 2020 2057 6865 7468 6572 206f  >..    Whether o
0000060: 7220 6e6f 7420 746f 2070 6572 6d69 7420  r not to permit 
0000070: 6f62 6a65 6374 2074 6167 7320 696e 2064  object tags in d
0000080: 6f63 756d 656e 7473 2c20 7769 7468 2061  ocuments, with a
0000090: 206e 756d 6265 7220 6f66 2065 7874 7261   number of extra
00000a0: 0d0a 2020 2020 7365 6375 7269 7479 2066  ..    security f
00000b0: 6561 7475 7265 7320 6164 6465 6420 746f  eatures added to
00000c0: 2070 7265 7665 6e74 2073 6372 6970 7420   prevent script 
00000d0: 6578 6563 7574 696f 6e2e 2054 6869 7320  execution. This 
00000e0: 6973 2073 696d 696c 6172 2074 6f0d 0a20  is similar to.. 
00000f0: 2020 2077 6861 7420 7765 6273 6974 6573     what websites
0000100: 206c 696b 6520 4d79 5370 6163 6520 646f   like MySpace do
0000110: 2074 6f20 6f62 6a65 6374 2074 6167 732e   to object tags.
0000120: 2020 596f 7520 7368 6f75 6c64 2061 6c73    You should als
0000130: 6f20 656e 6162 6c65 0d0a 2020 2020 254f  o enable..    %O
0000140: 7574 7075 742e 466c 6173 6843 6f6d 7061  utput.FlashCompa
0000150: 7420 696e 206f 7264 6572 2074 6f20 6765  t in order to ge
0000160: 6e65 7261 7465 2049 6e74 6572 6e65 7420  nerate Internet 
0000170: 4578 706c 6f72 6572 0d0a 2020 2020 636f  Explorer..    co
0000180: 6d70 6174 6962 696c 6974 7920 636f 6465  mpatibility code
0000190: 2066 6f72 2079 6f75 7220 6f62 6a65 6374   for your object
00001a0: 2074 6167 732e 0d0a 3c2f 703e 0d0a 2d2d   tags...</p>..--
00001b0: 2320 7669 6d3a 2065 7420 7377 3d34 2073  # vim: et sw=4 s
00001c0: 7473 3d34 0d0a                           ts=4..

Copied version:

0000000: 4854 4d4c 2e53 6166 654f 626a 6563 740a  HTML.SafeObject.
0000010: 5459 5045 3a20 626f 6f6c 0a56 4552 5349  TYPE: bool.VERSI
0000020: 4f4e 3a20 332e 312e 310a 4445 4641 554c  ON: 3.1.1.DEFAUL
0000030: 543a 2066 616c 7365 0a2d 2d44 4553 4352  T: false.--DESCR
0000040: 4950 5449 4f4e 2d2d 0a3c 703e 0a20 2020  IPTION--.<p>.   
0000050: 2057 6865 7468 6572 206f 7220 6e6f 7420   Whether or not 
0000060: 746f 2070 6572 6d69 7420 6f62 6a65 6374  to permit object
0000070: 2074 6167 7320 696e 2064 6f63 756d 656e   tags in documen
0000080: 7473 2c20 7769 7468 2061 206e 756d 6265  ts, with a numbe
0000090: 7220 6f66 2065 7874 7261 0a20 2020 2073  r of extra.    s
00000a0: 6563 7572 6974 7920 6665 6174 7572 6573  ecurity features
00000b0: 2061 6464 6564 2074 6f20 7072 6576 656e   added to preven
00000c0: 7420 7363 7269 7074 2065 7865 6375 7469  t script executi
00000d0: 6f6e 2e20 5468 6973 2069 7320 7369 6d69  on. This is simi
00000e0: 6c61 7220 746f 0a20 2020 2077 6861 7420  lar to.    what 
00000f0: 7765 6273 6974 6573 206c 696b 6520 4d79  websites like My
0000100: 5370 6163 6520 646f 2074 6f20 6f62 6a65  Space do to obje
0000110: 6374 2074 6167 732e 2020 596f 7520 7368  ct tags.  You sh
0000120: 6f75 6c64 2061 6c73 6f20 656e 6162 6c65  ould also enable
0000130: 0a20 2020 2025 4f75 7470 7574 2e46 6c61  .    %Output.Fla
0000140: 7368 436f 6d70 6174 2069 6e20 6f72 6465  shCompat in orde
0000150: 7220 746f 2067 656e 6572 6174 6520 496e  r to generate In
0000160: 7465 726e 6574 2045 7870 6c6f 7265 720a  ternet Explorer.
0000170: 2020 2020 636f 6d70 6174 6962 696c 6974      compatibilit
0000180: 7920 636f 6465 2066 6f72 2079 6f75 7220  y code for your 
0000190: 6f62 6a65 6374 2074 6167 732e 0a3c 2f70  object tags..</p
00001a0: 3e0a 2d2d 2320 7669 6d3a 2065 7420 7377  >.--# vim: et sw
00001b0: 3d34 2073 7473 3d34 0a                   =4 sts=4.
Aron Rotteveel
  • 73,917
  • 17
  • 100
  • 128
  • 1
    If you have `core.filemode` unset, or set to `true`, is the output different? – Mark Longair Apr 26 '11 at 09:20
  • The other important bit of information that would help is the output of `git --version` – Mark Longair Apr 26 '11 at 11:10
  • 2
    @AronRotteveel: That is easy: the first file has CRLF line-ends (windows), the second LF (Unix) – sehe Nov 23 '11 at 14:36
  • 5
    git 2.8 (March 2016) introduces `git ls-files --eol`, to quickly see if eol is involved. See [my answer below](http://stackoverflow.com/a/35204436/6309) – VonC Feb 04 '16 at 15:03
  • 2
    The person on windows can run **git config --global core.autocrlf true** to solve this issue. – Inyoka Mar 21 '18 at 07:02
  • @Inyoka autocrlf=true can prevent this problem if used from the beginning, but would it actually fix the issue after the CRLFs are already in there? https://stackoverflow.com/a/37749424/8340554 suggests it wouldn't. – Kalinda Pride Jan 06 '21 at 01:17
  • @KalindaPride, I think this issue with Git was been patched by Johannes Schindelin. Most of the mentions of this problem are several years old. – Inyoka Jan 14 '21 at 11:37
  • @Inyoka This past month I cloned a repo onto a new computer (newest version of git) and encountered this problem, with every file showing as modified due to line endings. If the issue has been patched as you say, maybe the problem was caused by previous committers because they were running older versions of git? Or did you mean that I could have updated git on the old (Windows) computer and then run "git config --global core.autocrlf true" there, and then the problem would be fixed? – Kalinda Pride Jan 15 '21 at 17:14
  • @KalindaPride you still need to deal with the underlying problem of how you want to deal with Line Endings on different systems. You might need a global gitattributes file or separate policies for *nix and windows. This answer might help https://stackoverflow.com/a/11199598/792015, otherwise it would be better to post a Question with more details on your situation. – Inyoka Jan 17 '21 at 10:13
  • I got rid of the issue using: `git add --renormalize .` For details see [configuring-git-to-handle-line-endings](https://docs.github.com/en/github/using-git/configuring-git-to-handle-line-endings) – mihca Mar 04 '21 at 19:10

18 Answers18

174

I have resolved this problem using following steps

  1. Remove every file from Git's index.

    git rm --cached -r .

  2. Rewrite the Git index to pick up all the new line endings.

    git reset --hard

Note that step 2 may remove your local changes. Solution was part of steps described on git site https://help.github.com/articles/dealing-with-line-endings/

avolkmann
  • 2,204
  • 1
  • 14
  • 23
Jacek Szybisz
  • 3,633
  • 3
  • 8
  • 3
  • 1
    In my case I had made a bunch of notes to some files, then copied the repo without the .git folder to a new computer. When I realized I was missing my .git package, it was too late to go back and retrieve it. So I: 1. Checked out a new copy of the whole repo 2. Replaced the vanilla files with the files with my changes 3. Ran step one in the OP's post: `git rm --cached -r .` 4. This staged my changes (and any other files that got replaced) so I unstaged them. At this point my repo was back to normal. – cody Aug 27 '18 at 18:26
  • 3
    Brilliant. Thanks for this. – rupi Feb 01 '19 at 12:08
  • 12
    Gotta be careful as you will lose any other changes if you do. – Mohy Eldeen May 06 '19 at 19:00
  • I had this problem after copy&pasting my repo folder from Windows to Ubuntu. There were no local changes, but for some reason, git thought every single file has changed. This solution resolved this problem. – Linek Dec 06 '19 at 11:00
  • An excellent solution for those using Windows, but moving their development environment into Linux, or WSL (which is how I landed here). – Joshua Schlichting Jul 23 '20 at 15:21
  • I lost all my changes... – DarioN1 Apr 20 '21 at 15:26
  • I kinda wish this was the official answer as it's a more general solution. CRLF is not the only cause of this. In my case I've gotten it several times from copying an entire repository (Linux --> Linux), and this was the only workable solution for me. – Brendano257 May 04 '21 at 11:23
123

Have you changed the mode of the files? I did it on my machine and the local dev machine had 777 given to all the files whereas the repo had 755 which showed every file as modified. I did git diff and it showed the old mode and new mode are different. If that is the problem then you can easily ignore them by git config core.filemode false
Cheers

trainoasis
  • 5,531
  • 10
  • 43
  • 74
itsandy
  • 1,638
  • 2
  • 12
  • 14
  • 3
    Using Windows 10 lxss (i.e. Ubuntu Bash) with a repo in Windows filesystem with git in Linux, I had thought it was a line ending issue. Hadn't even considered it could have been the way file modes are different as well. – Matt L Mar 02 '18 at 17:25
  • This worked for me when I changed my local O/S from Ubuntu to Windows. Cheers – Mark Bucknell Apr 26 '18 at 02:59
  • @MattL and itsandy I usually develop on Win 10 with Git Bash, but today I'm on Mac and see that git thinks `.gitignore` files have changed when they haven't. On this Mac, `git config core.filemode` responds with `true`. I changed it to `false`, but that didn't help. – Ryan Dec 30 '18 at 15:19
  • you are the real God – gursahib.singh.sahni Jan 22 '21 at 19:03
  • Why `core.filemode` is enabled by default? This probably ruined many clean repository histories. – Martin Braun May 28 '21 at 04:38
69

Update: as per the comment on this question, the problem has been solved:

That is easy: the first file has CRLF line-ends (windows), the second LF (Unix). The file util (available in git\usr\bin) will show you that (file a b will reply something like a: ASCII text, with CRLF line terminators b: ASCII text)

Original answer below:


The diff you show does not show a single different line. Can you post .git/config (or better git config -l).

You might have some whitespace ignores activated

You should try to disable core.whitespace=fix,-indent-with-non-tab,trailing-space,cr-at-eol;

also

git show HEAD:myfile|md5sum
md5sum myfile

could be used to verify that the files are in fact different. Using external diff could work as well

git show HEAD:myfile > /tmp/myfile.HEAD

diff -u myfile /tmp/myfile.HEAD

# or if you prefer an interactive tool like e.g.:
vim -d myfile /tmp/myfile.HEAD
Michael Freidgeim
  • 21,559
  • 15
  • 127
  • 153
sehe
  • 328,274
  • 43
  • 416
  • 565
  • The strange thing is that the md5sums for both files are different but I cannot find *ANY* whitespace difference. (I assume this is the case, but I simple do not see it). Any help would be welcome. – Aron Rotteveel Nov 23 '11 at 14:12
  • Just upload an example somewhere (gist.github.com would be appropriate for this). I assume it is with newline on the last line, a [byte-order-mark](http://en.wikipedia.org/wiki/Byte_order_mark), or non-canonical UTF8 encodings in general. You can always look at `xxd` or `bdiff` to do binary diffs too – sehe Nov 23 '11 at 14:14
  • thanks for the tip. `xdd` was new to me. Great tool! I updated my post with an example. – Aron Rotteveel Nov 23 '11 at 14:27
  • 2
    @AronRotteveel: That is easy: the first file has CRLF line-ends (windows), the second LF (Unix). **Edit** The `file` util will show you that (`file a b` will reply something like `a: ASCII text, with CRLF line terminators b: ASCII text`) – sehe Nov 23 '11 at 14:41
  • shouldn't those actually show as ^M in VIM? (I don't see it). Obviously, I believe you, but I'm really interested in how you actually noticed this :) – Aron Rotteveel Nov 23 '11 at 14:45
  • @AronRotteveel: no, Vim detects the lineend style (do `:se fileformat` to see the value: `unix` or `dos`). Use that to convert as well: `:se ff=unix | w`. For completeness, vim shows the ugly ducklings (^M) _iff_ you have _mixed_ line-endings, in which case it chooses `unix` and the redundant `^M` show up – sehe Nov 23 '11 at 14:56
  • 1
    I'm my case the hash is the same, but git still indicates a change – tofutim Jun 26 '15 at 20:31
  • @tofutim Did the modes change (permissions?) – sehe Jun 26 '15 at 22:56
  • `git diff filename` wasn't printing anything and they had the same md5 checksums. I just reverted the change ‍♂️. – Arye Eidelman Jul 24 '20 at 17:04
46

i had the same problem. after win->lin copy i've got all files modified.
i used fromdos to fix line endings
and then

git add -uv 

to add changes.
it added 3 files (not all of them), which i actually modified. and after that git status shows only 3 modified files. after git commit everything is ok with git status

Demo_S
  • 629
  • 5
  • 7
  • 2
    Thanks, this was what I needed. I used @sehe's suggestion of comparing md5sum's to verify the files were identical (in my case they were). After that using the -u flag correctly filtered those files with no actual differences from those with changes. – STW Jan 28 '14 at 21:26
  • Thanks, it was helpful. I had this problem after doing `npm i` in the root of my project. Somehow, many files were getting a different line endings, which was creating this issue. – Server Khalilov Jul 03 '19 at 10:52
  • this is what i expected...the changes which have done remains and rest got undo – Deepak Ganachari Sep 25 '19 at 12:45
  • I struggled in the same situation. I used `find . -print0 | xargs -0 dos2unix -ic0 | xargs -0 dos2unix -b` to update line breaks from Windows to Linux and `git add .` to fix status. After it `git status` showed me all up to date status. It looks like a bug in the git, it shouldn't be necessary to execute `git add .` to fix `git status`. – Konard Nov 28 '20 at 11:02
31

In my case the files were appeared as modified after changing the files permissions.

To make git ignore permission changes, do the following :

# For the current repository
git config core.filemode false   

# Globally
git config --global core.filemode false
Sakis
  • 504
  • 4
  • 11
30

I was able to fix the problems on Windows machine by changing core.autocrlf from false to core.autocrlf=input

git config core.autocrlf input

as it's suggested in https://stackoverflow.com/a/1112313/52277

Community
  • 1
  • 1
Michael Freidgeim
  • 21,559
  • 15
  • 127
  • 153
28

However, a lot (if not every) file appears as modified even though the contents are exactly the same.

With git 2.8 (March 2016), you will able to quickly check if those changes are eol-related.

If that is the case, Devin G Rhode adds in the comments:

I solved this again by adding a .gitattributes file to a repo with just * text=auto inside of it

For the check, see commit a7630bd (16 Jan 2016) by Torsten Bögershausen (tboegi).
(Merged by Junio C Hamano -- gitster -- in commit 05f1539, 03 Feb 2016)

ls-files: add eol diagnostics

When working in a cross-platform environment, a user may want to check if text files are stored normalized in the repository and if .gitattributes are set appropriately.

Make it possible to let Git show the line endings in the index and in the working tree and the effective text/eol attributes.

The end of line ("eolinfo") are shown like this:

"-text"        binary (or with bare CR) file
"none"         text file without any EOL
"lf"           text file with LF
"crlf"         text file with CRLF
"mixed"        text file with mixed line endings.

The effective text/eol attribute is one of these:

"", "-text", "text", "text=auto", "text eol=lf", "text eol=crlf"

git ls-files --eol gives an output like this:

i/none   w/none   attr/text=auto      t/t5100/empty
i/-text  w/-text  attr/-text          t/test-binary-2.png
i/lf     w/lf     attr/text eol=lf    t/t5100/rfc2047-info-0007
i/lf     w/crlf   attr/text eol=crlf  doit.bat
i/mixed  w/mixed  attr/               locale/XX.po

to show what eol convention is used in the data in the index ('i'), and in the working tree ('w'), and what attribute is in effect, for each path that is shown.

VonC
  • 1,042,979
  • 435
  • 3,649
  • 4,283
  • 2
    Although this shows the problem, it does not fix it. – Greeso Feb 11 '20 at 21:56
  • This provided the best clue in my situation. I had one gradlew.bat file that was modified, and no matter what I did to reset the working directory (git reset --hard && git status and git reset --hard && rm gradlew.bat && git status && git checkout . && git status) the file remained modified. I have a global .gitattributes from here: https://github.com/alexkaratarakis/gitattributes/blob/master/Web.gitattributes with a few extra bits from here: https://rehansaeed.com/gitattributes-best-practices/ – Devin Rhode Sep 02 '20 at 15:55
  • `git ls-files --eol` showed my gradlew.bat was one of/the only file that did not have `text=auto` set. In my ~/.gitattributes_global I had `*.bat text eol=crlf` so I tried `text eol=lf` but that didn't work. So then I tried `text=auto` in the .gitattributes_global and it fixed the issue. – Devin Rhode Sep 02 '20 at 15:57
  • 1
    @DevinGRhode Well done! `ls-files --eol` is handy indeed. – VonC Sep 02 '20 at 16:12
  • 2
    I solved this again by adding a .gitattributes file to a repo with just `* text=auto` inside of it :) - maybe put this in around the top of your answer? – Devin Rhode Oct 03 '20 at 02:14
  • 1
    @DevinGRhode Thank you. I have included your comment in the answer for more visibility. – VonC Oct 03 '20 at 05:17
8

Here is how I fixed the problem on Linux while I cloned a project that was created on Windows:

on linux in order for things to work properly you have to have this setting: core.autocrlf=input

this is how to set it: git config --global core.autocrlf input

Then clone the project again from github.

truthseeker
  • 369
  • 1
  • 4
  • 7
  • This worked for me in a scenario where I'm using Visual Studio on Windows, but I am doing commits and additional checks in WSL on the command line. On command line, every file showed as changed, but in Visual Studio, only a few files changed. I added autoclrf = input in my .gitconfig file and it fixed it instantly. Thank you! – Big Data Brian Mar 10 '20 at 13:41
  • Thanks. I had also moved from windows to linux. Your solution worked for me. – Noor Ahmed Sep 10 '20 at 05:28
6

The Git FAQ has an answer that might be relevant, although I've never come across this before:

Why does git diff sometimes list a file that has no changes?

git diff and other git operations is optimized so it does not even look at files whose status (size, modification time etc) on disk and in git's index are different. This makes git diff extremely fast for small changes. If the file has been touched somehow, git diff has to look at the content of and compare it which is a much slower operation even when there is in fact no change. git diff lists the files as a reminder that it is not used optimally. Running git status will not only show status, but will also update the index with status for unchanged files disk making subsequent operations, not only diff, much faster. A typical case that causes many files to be listed by diff is running mass editing commands like perl -pi -e '...'.

What does git status show for you?

Mark Longair
  • 385,867
  • 66
  • 394
  • 320
4

I just set filemode = false in .git/config and it worked for me.

filemode = false
Felipe Santa
  • 169
  • 1
  • 4
3

In my case, files were exactly the same (content, line endings, filemodes, md5 checksum, everything), yet were still visible in git status.

What helped was

# Garbage collect stuffs from .git
git gc

# commit stuff
git add . 
git commit -m "temporary" 

# this doesn't change contents
git reset HEAD^1 --soft    

After that, git status showed only one editted file that I editted.

Danon
  • 1,793
  • 14
  • 28
2

After copying my local repository and working copy to another folder (on Windows by the way), I had four files that kept showing up as changed and tried every suggestion listed in the other answers. In the end what fixed it for me was deleting the local branch and downloading it again from the remote. In my case I guess it had something to do with copying a local repository rather than cloning.

Nate Cook
  • 7,577
  • 5
  • 44
  • 36
2

So, I tried just about everything here and want to contribute one more solution that fixed all of my problems. My issues was not with line endings or actual permissions or anything like that. It was because I had installed Cygwin and whole host of stuff that comes with that, which unbeknownst to me also installed its own version of git. I never noticed this, just that I was having strange issues with users and files being marked as changed (because of perms changes).

It turns out that I figured this out because I thought I should just update Git to the latest version, which I did, but running git --version returned the old version number. After the ensuing hunt for why, I found the cygwin bin directory root in my environment path, which contained a git executable, running at the old version number. Go figure.

This was also hard to find because I have TortoiseGit installed. My command line tools would use the cygwin version due to path fallbacks, and TortoiseGit was configured to use the windows version, making it even more confusing.

Hope this helps somebody.

dudewad
  • 10,029
  • 4
  • 31
  • 41
  • 1
    Good catch. +1. That is why I always set my PATH myself, as in https://stackoverflow.com/a/44351065/6309 – VonC Jun 06 '17 at 06:54
  • I often will let an installer set the PATH but then check it manually, which I did here. My previous install had git 2.4.x (x86) and I installed git 2.13.x (x64). You can imagine my confusion when I removed the x86 path reference and still got 2.4!! That's when I saw cygwin's bin directory and that seemed the next logical place to look. So FWIW setting or even checking your PATH visually might not solve this if the Cygwin bin directory is listed first! – dudewad Jun 06 '17 at 17:00
2

I recently moved my local repo from one Windows x64 system to another. The first time I use it half my files appear to be changed. Thanks to Jacek Szybisz for sending me to Configuring Git to handle line endings where I found the following one-liner that removed all the no-change files from Gitkraken's change queue:

git config --global core.autocrlf true
CAK2
  • 1,452
  • 1
  • 11
  • 15
1

The only suspect entry in your config looks to me to be core.ignorecase. You could try unsetting that with:

  git config --unset core.ignorecase

... and see if the output from git status or git diff is different.

Mark Longair
  • 385,867
  • 66
  • 394
  • 320
0

For me it was because 2 linux VMs were both mapped to the same home file system. One VM was running git-1.7.1 an the other was running git-2.14

The VM running git-1.7.1 would always show 4 files as changed (even thought the contents and line endings were identical).

Once 'git status' was run on the VM running g-2.14, then both VMs would start reporting the repository as clean. 'git status' has side effects. It is not an immutable operation. And git-1.7.1 does not understand the world in the same way that git-2+ does.

William
  • 19,567
  • 8
  • 45
  • 88
0

For me the issue was a case difference in the file.

I renamed the file and committed the delete, It showed both the upper and lower case versions of the file to be deleted.

After the delete was committed I renamed the file back to its original name and pushed.

  • Welcome to Stack Overflow. This is a nine year old question with an accepted answer and a lot of existing answers. When adding another answer it is important to point out what new aspect of the question your answer addresses. It appears that your answer is answering a different git issue than the one in the question (file name case versus line endings). – Jason Aller Jul 30 '20 at 01:26
0

In my case, git was cloning slow, so I just copied the cloned repo from my Mac to Windows and none of these helped. What did help is doing a clone of the copied repo and used that instead and that seems to have fixed it. I am assuming there are some options set by default that differ cross platform that could cause the issue.

BrainStorm.exe
  • 1,287
  • 3
  • 21
  • 37