Git segfault when called via Google Repo [Was: [ANNOUNCEMENT] Updated: git-2.0.4-1]

Adam Dinwoodie adam@dinwoodie.org
Sat Aug 30 10:31:00 GMT 2014


On Wed, Aug 27, 2014 at 02:23:02PM +0800, Robert Bu wrote:
> If you remember, I've changed CYGWIN=winsymlinks:nativestrict to
> CYGWIN=winsymlinks:lnk to satisfy repo to create the symlink.
> However, I still get a problem in Cygwin-64 on Win7. It seems to be
> a git problem now. The error message is:
> 
> From ssh://xxxx:20000/vm/xxx
>  * [new branch]      master     -> origin/master
> Fetching projects: 100% (1/1), done.
> Traceback (most recent call last):
>   File
> "/cygdrive/d/cygwin/home/robert_bu/repo/vm/.repo/repo/main.py", line
> 500, in <module>
>     _Main(sys.argv[1:])
>   File
> "/cygdrive/d/cygwin/home/robert_bu/repo/vm/.repo/repo/main.py", line
> 476, in _Main
>     result = repo._Run(argv) or 0
>   File
> "/cygdrive/d/cygwin/home/robert_bu/repo/vm/.repo/repo/main.py", line
> 155, in _Run
>     result = cmd.Execute(copts, cargs)
>   File "/cygdrive/d/cygwin/home/robert_bu/repo/vm/.repo/repo/subcmds/sync.py",
> line 675, in Execute
>     project.Sync_LocalHalf(syncbuf)
>   File
> "/cygdrive/d/cygwin/home/robert_bu/repo/vm/.repo/repo/project.py",
> line 1170, in Sync_LocalHa
> lf
>     self._InitWorkTree()
>   File
> "/cygdrive/d/cygwin/home/robert_bu/repo/vm/.repo/repo/project.py",
> line 2230, in _InitWorkTre
> e
>     raise GitError("cannot initialize work tree")
> error.GitError: cannot initialize work tree
> 
> 
> git seems to crash. I've attached the stackdump file of git.

Hi Robert,

I've been trying to reproduce this bug myself, but I haven't been able
to.  Using CYGWIN=winsymlinks:lnk seems to work absolutely fine in all
the tests I've done (namely running the entire Git test suite, which
includes a variety of tests that use symlinks).

To make any progress here, I really need a simple test case that I can
use to reproduce the problem, ideally one that doesn't involve Repo.
The simpler the test case, the easier it'll be to work out what's going
wrong (and thus the more likely it'll be that I can actually find time
to do something about it).

I don't think you answered my previous question about whether this is
only a problem with v2.0.4-1 or if it also affected the previous version
of Git on Cygwin, v1.7.9-1.  I don't think that'll make a significant
difference to isolating the problem, but it may be an interesting
datapoint.

Adam

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple



More information about the Cygwin mailing list