Re: [RFC/PATCH] Re: git submodule -b ... of current HEAD fails

Jonathan Nieder [off-list ref] writes:

	git submodule add -b $branch $repository

fails when HEAD already points to $branch in $repository.

I was reviewing the overall picture before tagging 1.7.4-rc0 and started
wondering if this was a good change.  If repository already had branch
checked out, and if it was pointing at a commit that was different from
whatever was taken from origin, shouldn't the command _fail_ to prevent
the divergent commits from getting lost?