From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756311AbZBIVTa (ORCPT ); Mon, 9 Feb 2009 16:19:30 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754104AbZBIVRF (ORCPT ); Mon, 9 Feb 2009 16:17:05 -0500 Received: from mx2.redhat.com ([66.187.237.31]:36438 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754109AbZBIVRE (ORCPT ); Mon, 9 Feb 2009 16:17:04 -0500 From: Eric Paris Subject: [PATCH -v1 09/11] fsnotify: fsnotify marks on inodes pin them in core To: linux-kernel@vger.kernel.org Cc: viro@zeniv.linux.org.uk, hch@infradead.org, alan@lxorguk.ukuu.org.uk, sfr@canb.auug.org.au, john@johnmccutchan.com, rlove@rlove.org, malware-list@lists.printk.net, akpm@linux-foundation.org Date: Mon, 09 Feb 2009 16:16:09 -0500 Message-ID: <20090209211609.8985.24420.stgit@paris.rdu.redhat.com> In-Reply-To: <20090209211525.8985.13887.stgit@paris.rdu.redhat.com> References: <20090209211525.8985.13887.stgit@paris.rdu.redhat.com> User-Agent: StGIT/0.14.3 MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org This patch pins any inodes with an fsnotify mark in core. The idea is that as soon as the mark is removed from the inode->fsnotify_mark_entries list the inode will be iput. In reality is doesn't quite work exactly this way. The igrab will happen when the mark is added to an inode, but the iput will happen when the inode pointer is NULL'd inside the mark. It's possible that 2 racing things will try to remove the mark from different directions. One may try to remove the mark because of an explicit request and one might try to remove it because the inode was deleted. It's possible that the removal because of inode deletion will remove the mark from the inode's list, but the removal by explicit request will actually set entry->inode == NULL; and call the iput. This is safe. Signed-off-by: Eric Paris --- fs/notify/inode_mark.c | 25 ++++++++++++++++++++----- 1 files changed, 20 insertions(+), 5 deletions(-) diff --git a/fs/notify/inode_mark.c b/fs/notify/inode_mark.c index 840bd91..ff65e62 100644 --- a/fs/notify/inode_mark.c +++ b/fs/notify/inode_mark.c @@ -108,13 +108,16 @@ void fsnotify_clear_marks_by_group(struct fsnotify_group *group) spin_lock(&inode->i_lock); list_del_init(&entry->i_list); - entry->inode = NULL; list_del_init(&entry->g_list); - entry->group = NULL; - entry->freeme = 1; fsnotify_recalc_inode_mask_locked(inode); spin_unlock(&inode->i_lock); + + entry->group = NULL; + entry->freeme = 1; + entry->inode = NULL; + iput(inode); + spin_unlock(&entry->lock); fsnotify_put_mark(entry); @@ -137,14 +140,17 @@ void fsnotify_destroy_mark_by_entry(struct fsnotify_mark_entry *entry) spin_lock(&inode->i_lock); list_del_init(&entry->i_list); - entry->inode = NULL; list_del_init(&entry->g_list); + + entry->inode = NULL; entry->group = NULL; entry->freeme = 1; if (inode) { fsnotify_recalc_inode_mask_locked(inode); spin_unlock(&inode->i_lock); + + iput(inode); } if (group) spin_unlock(&group->mark_lock); @@ -173,6 +179,11 @@ void fsnotify_clear_marks_by_inode(struct inode *inode, unsigned int flags) * try to grab entry->inode->i_lock without a problem. */ list_for_each_entry_safe(entry, lentry, &free_list, free_i_list) { + spin_lock(&entry->lock); + if (entry->inode) + iput(entry->inode); + entry->inode = NULL; + spin_unlock(&entry->lock); entry->group->ops->mark_clear_inode(entry, inode, flags); fsnotify_put_mark(entry); } @@ -209,9 +220,13 @@ int fsnotify_add_mark(struct fsnotify_mark_entry *entry) { struct fsnotify_mark_entry *lentry; struct fsnotify_group *group = entry->group; - struct inode *inode = entry->inode; + struct inode *inode; int ret = 0; + inode = igrab(entry->inode); + if (unlikely(!inode)) + return -EINVAL; + /* * LOCKING ORDER!!!! * entry->lock