git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Shawn O. Pearce" <spearce@spearce.org>
To: Andreas Gruenbacher <agruen@suse.de>
Cc: git@vger.kernel.org
Subject: Re: [PATCH 0/9] Multiple remotes without conflicts
Date: Thu, 18 Mar 2010 12:07:12 -0700	[thread overview]
Message-ID: <20100318190712.GD10981@spearce.org> (raw)
In-Reply-To: <cover.1268913163.git.agruen@suse.de>

Andreas Gruenbacher <agruen@suse.de> wrote:
> I'm still trying to find a simple and painless way of sharing the object
> store among multiple repositories: the idea is to have a "parent"
> repository which contains the actual object store, and a number of
> "child" repositories which link to that object store.  The obvious
> problem is garbage collection: we can only garbage collect the parent
> once it has all refs of all its children.
> 
> One way of ensuring that is to make each child a "remote" of the parent,
> and to fetch all remotes first.  This works for branches, but not for
> tags or for the reflog.

This just feels like the wrong solution.

Why can't we have a "$GIT_DIR/children" subdirectory with a symlink
or file-containing-path to each child repository.  Modify the fsck
and gc paths to include these additional reference and reflog spaces,
and that's that.

Child registration is then just a matter of installing the symlink
in the parent, or removing it, and gc/fsck never needs to worry
about a fetch up front in order for it to be accurate.
 
-- 
Shawn.

  parent reply	other threads:[~2010-03-18 19:07 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-03-18 11:52 [PATCH 0/9] Multiple remotes without conflicts Andreas Gruenbacher
2010-03-13 17:17 ` [PATCH 1/9] fetch: Check for a "^{}" suffix with suffixcmp() Andreas Gruenbacher
2010-03-12 22:27   ` [PATCH 2/9] fetch: Properly initialize refspec on stack Andreas Gruenbacher
2010-03-15 22:18     ` [PATCH 3/9] fetch: Fix minor memory leak Andreas Gruenbacher
2010-03-16 15:45       ` [PATCH 4/9] fetch: Move deepening fetch check into builtin/fetch.c Andreas Gruenbacher
2010-03-16 17:49         ` [PATCH 5/9] fetch: Move loop checking which refs we have already Andreas Gruenbacher
2010-03-16 18:48           ` [PATCH 6/9] fetch: Check if all objects exist after fetching Andreas Gruenbacher
2010-03-17 21:42             ` [PATCH 7/9] fetch: Use the same ref map for all branches and tags Andreas Gruenbacher
2010-03-17 23:02               ` [PATCH 8/9] fetch: Don't fetch tags twice Andreas Gruenbacher
2010-03-17 23:59                 ` [PATCH 9/9] fetch: Make automatic tag following work with arbitrary refspecs Andreas Gruenbacher
2010-03-18 19:08             ` [PATCH 6/9] fetch: Check if all objects exist after fetching Shawn O. Pearce
2010-03-18 19:36               ` Andreas Gruenbacher
2010-03-18 19:07 ` Shawn O. Pearce [this message]
2010-03-18 19:42   ` [PATCH 0/9] Multiple remotes without conflicts Andreas Gruenbacher

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20100318190712.GD10981@spearce.org \
    --to=spearce@spearce.org \
    --cc=agruen@suse.de \
    --cc=git@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).