mirror of
https://github.com/git/git.git
synced 2026-03-12 18:09:46 +01:00
c30871b91d
On Debian-alikes, POSIX sh has a hardcoded recursion depth
of 1000. This limit operates like bash's `$FUNCNEST` [1], but
it does not actually respect `$FUNCNEST`. This is non-standard
behavior. On other distros, the sh recursion depth is limited
only by the available stack size.
With certain history graphs, subtree splits are recursive—with
one recursion per commit. Attempting to split complex repos that
have thousands of commits, like [2], may fail on these distros.
Reduce the amount of recursion required by eagerly discovering
the complete range of commits to process.
The recursion is a side-effect of the rejoin-finder in
`find_existing_splits`. Rejoin mode, as in
git subtree split --rejoin -b hax main ...
improves the speed of later splits by merging the split history
back into `main`. This gives the splitting algorithm a stopping
point. The rejoin maps one commit on `main` to one split commit
on `hax`. If we encounter this commit, we know that it maps to
`hax`.
But this is only a single point in the history. Many splits
require history from before the rejoin. See patch content for
examples.
If pre-rejoin history is required, `check_parents` recursively
discovers each individual parent, with one recursion per commit.
The recursion deepens the entire tree, even if an older rejoin
is available. This quickly overwhelms the Debian sh stack.
Instead of recursively processing each commit, process *all* the
commits back to the next obvious starting point: i.e., either the
next-oldest --rejoin or the beginning of history. This is where the
recursion is likely to stop anyway.
While this still requires recursion, it is *considerably* less
recursive.
[1]: https://www.gnu.org/software/bash/manual/html_node/Bash-Variables.html#index-FUNCNEST
[2]: https://github.com/christian-heusel/aur.git
Signed-off-by: Colin Stagner <ask+git@howdoi.land>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Contributed Software Although these pieces are available as part of the official git source tree, they are in somewhat different status. The intention is to keep interesting tools around git here, maybe even experimental ones, to give users an easier access to them, and to give tools wider exposure, so that they can be improved faster. I am not expecting to touch these myself that much. As far as my day-to-day operation is concerned, these subdirectories are owned by their respective primary authors. I am willing to help if users of these components and the contrib/ subtree "owners" have technical/design issues to resolve, but the initiative to fix and/or enhance things _must_ be on the side of the subtree owners. IOW, I won't be actively looking for bugs and rooms for enhancements in them as the git maintainer -- I may only do so just as one of the users when I want to scratch my own itch. If you have patches to things in contrib/ area, the patch should be first sent to the primary author, and then the primary author should ack and forward it to me (git pull request is nicer). This is the same way as how I have been treating gitk, and to a lesser degree various foreign SCM interfaces, so you know the drill. I expect things that start their life in the contrib/ area to graduate out of contrib/ once they mature, either by becoming projects on their own, or moving to the toplevel directory. On the other hand, I expect I'll be proposing removal of disused and inactive ones from time to time. If you have new things to add to this area, please first propose it on the git mailing list, and after a list discussion proves there is general interest (it does not have to be a list-wide consensus for a tool targeted to a relatively narrow audience -- for example I do not work with projects whose upstream is svn, so I have no use for git-svn myself, but it is of general interest for people who need to interoperate with SVN repositories in a way git-svn works better than git-svnimport), submit a patch to create a subdirectory of contrib/ and put your stuff there. -jc