mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Alex Riesen <raa.lkml@gmail.com>
To: linux-kernel <linux-kernel@vger.kernel.org>
Cc: fs_other@kernel-bugs.osdl.org,
	Russell Bell <russellbell@gmail.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Al Viro <viro@zeniv.linux.org.uk>,
	josef@redhat.com, Eric Paris <eparis@redhat.com>,
	Robert Love <rlove@rlove.org>,
	John McCutchan <john@johnmccutchan.com>
Subject: [PATCH] inotify: fix the running away counter of user watches
Date: Mon, 27 Jul 2009 19:25:42 +0200	[thread overview]
Message-ID: <20090727172542.GA5104@blimp.localdomain> (raw)

The counter of userspace inotify watches is incremented when a watch is
added, but was never decremented. This patch uses the fsnotify callback
(freeing_marks, which in turn calls inotify_ignored_and_remove_idr of
inotify userspace interface) called when an fsnotify mark is removed.

The bug has been reported in

    http://bugzilla.kernel.org/show_bug.cgi?id=12754

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
---

Sorry for long Cc:, I'm a little confused about "fs_other@" (and the
list of participants) in the bug report.

The patch works in my limited testing. I just did:

    # echo 10 >/proc/sys/fs/inotify/max_user_watches
    # inotify-watch -r /lib # to max the counter
    Ctrl-C
    # inotify-watch /bin/true # must work

in a clean environment (init=/bin/bash).

But I don't know whether it safe to access the counter in this
callback, nor am I sure the patch is enough.

 fs/notify/inotify/inotify_user.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index ff27a29..c2d94df 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -404,6 +404,7 @@ skip_send_ignore:
 
 	/* removed from idr, drop that reference */
 	fsnotify_put_mark(entry);
+	atomic_dec(&group->inotify_data.user->inotify_watches);
 }
 
 /* ding dong the mark is dead */
-- 
1.6.4.rc3.56.g32a7



             reply	other threads:[~2009-07-27 18:50 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-07-27 17:25 Alex Riesen [this message]
2009-07-28  6:08 ` Alex Riesen

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=20090727172542.GA5104@blimp.localdomain \
    --to=raa.lkml@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=eparis@redhat.com \
    --cc=fs_other@kernel-bugs.osdl.org \
    --cc=john@johnmccutchan.com \
    --cc=josef@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rlove@rlove.org \
    --cc=russellbell@gmail.com \
    --cc=viro@zeniv.linux.org.uk \
    /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

all inboxes | Powered by JetHome®