From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bob Peterson Date: Thu, 29 Oct 2015 15:47:52 -0400 (EDT) Subject: [Cluster-devel] [GFS2 PATCH v2] GFS2: Protect freeing directory hash table with i_lock spin_lock In-Reply-To: <1744548592.67923378.1446147842477.JavaMail.zimbra@redhat.com> Message-ID: <93040260.67926194.1446148072277.JavaMail.zimbra@redhat.com> List-Id: To: cluster-devel.redhat.com MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hi, This is a re-spin of a patch I posted earlier today. Andy Price pointed out that my comment was wrong: The race concerns gfs2_evict_inode, not gfs2_clear_inode (as the previous version's comment indicated). Function gfs2_clear_inode exists in RHEL6 and earlier, but not in upstream. Revised patch description: This patch changes function gfs2_dir_hash_inval so it uses the i_lock spin_lock to protect the in-core hash table, i_hash_cache. This will prevent double-frees due to a race between gfs2_evict_inode and inode invalidation. Signed-off-by: Bob Peterson --- diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index 487527b..ad8a5b7 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -388,8 +388,13 @@ static __be64 *gfs2_dir_get_hash_table(struct gfs2_inode *ip) */ void gfs2_dir_hash_inval(struct gfs2_inode *ip) { - __be64 *hc = ip->i_hash_cache; + __be64 *hc; + + spin_lock(&ip->i_inode.i_lock); + hc = ip->i_hash_cache; ip->i_hash_cache = NULL; + spin_unlock(&ip->i_inode.i_lock); + kvfree(hc); }