public inbox for git@vger.kernel.org 
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox•com>
To: mkoegler@auto•tuwien.ac.at (Martin Koegler)
Cc: Nicolas Pitre <nico@cam•org>, Simon Hausmann <simon@lst•de>,
	Git Mailing List <git@vger•kernel.org>
Subject: Re: git fsck segmentation fault
Date: Wed, 10 Dec 2008 18:33:20 -0800	[thread overview]
Message-ID: <7vljunwidr.fsf@gitster.siamese.dyndns.org> (raw)
In-Reply-To: <20081210075338.GA7776@auto.tuwien.ac.at> (Martin Koegler's message of "Wed, 10 Dec 2008 08:53:38 +0100")

mkoegler@auto•tuwien.ac.at (Martin Koegler) writes:

> Maybe something like this could help:

>>From 32be177cbb0825fc019200b172f3d79117b28140 Mon Sep 17 00:00:00 2001
> From: Martin Koegler <mkoegler@auto•tuwien.ac.at>
> Date: Wed, 10 Dec 2008 08:42:08 +0100
> Subject: [PATCH] fsck: use fewer stack
>
> This patch moves the state while traversing the tree
> from the stack to the heap.

Hmm, after the change:

	* mark_object() marks the object as reachable, and pushes the
	  objects to the objectstack;

	* mark_object_reachable() marks the object using mark_object(),
          and repeatedly calls mark_child_object() until the objectstack
          is fully drained;

	* mark_child_object() inspects the object taken from the
          objectstack, calls fsck_walk() on it, with mark_object as the
          callback;

	  * fsck_walk() calls the callback function (i.e. mark_object) on
            the object given, and the objects immediately reachable from
            it;

            * mark_object() does not recurse, so these immediately
              reachable objects are left in the objectstack, without a
              deep recursion.
        
That seems to be what is going on, and this should be a good fix.

A similar change would be needed for other callers of fsck_walk(), no?
There seem to be one in builtin-unpack-objects.c (check_object calls
fsck_walk as itself as the callback). 

Another caller is in index-pack.c (sha1_object() calls fsck_walk with
mark_link as the callback), but I do not think it would  recurse for the
depth of the history, so we are safe there.

I initially expected that the fix would be to introduce this "userspace
work queue" (i.e. your objectstack) to be maintained on the
fsck.c:fsck_walk() side (perhaps as an extra parameter to an actual queue
for reentrancy), not by making the callee not to recurse, though.

  reply	other threads:[~2008-12-11  2:34 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-11-27 17:14 git fsck segmentation fault Simon Hausmann
2008-11-27 17:47 ` Nicolas Pitre
2008-11-27 19:10   ` Simon Hausmann
2008-11-27 19:21     ` Simon Hausmann
2008-11-27 19:57       ` Nicolas Pitre
2008-11-28  8:19         ` Simon Hausmann
2008-12-09 19:09           ` Nicolas Pitre
2008-12-09 21:57             ` Martin Koegler
2008-12-10  7:53             ` Martin Koegler
2008-12-11  2:33               ` Junio C Hamano [this message]
2008-12-11  6:27                 ` Martin Koegler
2008-12-11  6:42                   ` Junio C Hamano

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=7vljunwidr.fsf@gitster.siamese.dyndns.org \
    --to=gitster@pobox$(echo .)com \
    --cc=git@vger$(echo .)kernel.org \
    --cc=mkoegler@auto$(echo .)tuwien.ac.at \
    --cc=nico@cam$(echo .)org \
    --cc=simon@lst$(echo .)de \
    /path/to/YOUR_REPLY

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