Commit 30c71233 authored by Jeff Layton's avatar Jeff Layton Committed by Ilya Dryomov

ceph: clean up unsafe d_parent access in __choose_mds

__choose_mds exists to pick an MDS to use when issuing a call. Doing
that typically involves picking an inode and using the authoritative
MDS for it. In most cases, that's pretty straightforward, as we are
using an inode to which we hold a reference (usually represented by
r_dentry or r_inode in the request).

In the case of a snapshotted directory however, we need to fetch
the non-snapped parent, which involves walking back up the parents
in the tree. The dentries in the snapshot dir are effectively frozen
but the overall parent is _not_, and could vanish if a concurrent
rename were to occur.

Clean this code up and take special care to ensure the validity of
the entries we're working with. First, try to use the inode in
r_locked_dir if one exists. If not and all we have is r_dentry,
then we have to walk back up the tree. Use the rcu_read_lock for
this so we can ensure that any d_parent we find won't go away, and
take extra care to deal with the possibility that the dentries could
go negative.

Change get_nonsnap_parent to return an inode, and take a reference to
that inode before returning (if any). Change all of the other places
where we set "inode" in __choose_mds to also take a reference, and then
call iput on that inode before exiting the function.

Link: http://tracker.ceph.com/issues/18148Signed-off-by: default avatarJeff Layton <jlayton@redhat.com>
Reviewed-by: default avatarYan, Zheng <zyan@redhat.com>
Signed-off-by: default avatarIlya Dryomov <idryomov@gmail.com>
parent c470abd4
...@@ -667,6 +667,27 @@ static void __unregister_request(struct ceph_mds_client *mdsc, ...@@ -667,6 +667,27 @@ static void __unregister_request(struct ceph_mds_client *mdsc,
ceph_mdsc_put_request(req); ceph_mdsc_put_request(req);
} }
/*
* Walk back up the dentry tree until we hit a dentry representing a
* non-snapshot inode. We do this using the rcu_read_lock (which must be held
* when calling this) to ensure that the objects won't disappear while we're
* working with them. Once we hit a candidate dentry, we attempt to take a
* reference to it, and return that as the result.
*/
static struct inode *get_nonsnap_parent(struct dentry *dentry) { struct inode
*inode = NULL;
while (dentry && !IS_ROOT(dentry)) {
inode = d_inode_rcu(dentry);
if (!inode || ceph_snap(inode) == CEPH_NOSNAP)
break;
dentry = dentry->d_parent;
}
if (inode)
inode = igrab(inode);
return inode;
}
/* /*
* Choose mds to send request to next. If there is a hint set in the * Choose mds to send request to next. If there is a hint set in the
* request (e.g., due to a prior forward hint from the mds), use that. * request (e.g., due to a prior forward hint from the mds), use that.
...@@ -675,19 +696,6 @@ static void __unregister_request(struct ceph_mds_client *mdsc, ...@@ -675,19 +696,6 @@ static void __unregister_request(struct ceph_mds_client *mdsc,
* *
* Called under mdsc->mutex. * Called under mdsc->mutex.
*/ */
static struct dentry *get_nonsnap_parent(struct dentry *dentry)
{
/*
* we don't need to worry about protecting the d_parent access
* here because we never renaming inside the snapped namespace
* except to resplice to another snapdir, and either the old or new
* result is a valid result.
*/
while (!IS_ROOT(dentry) && ceph_snap(d_inode(dentry)) != CEPH_NOSNAP)
dentry = dentry->d_parent;
return dentry;
}
static int __choose_mds(struct ceph_mds_client *mdsc, static int __choose_mds(struct ceph_mds_client *mdsc,
struct ceph_mds_request *req) struct ceph_mds_request *req)
{ {
...@@ -717,30 +725,39 @@ static int __choose_mds(struct ceph_mds_client *mdsc, ...@@ -717,30 +725,39 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
inode = NULL; inode = NULL;
if (req->r_inode) { if (req->r_inode) {
inode = req->r_inode; inode = req->r_inode;
ihold(inode);
} else if (req->r_dentry) { } else if (req->r_dentry) {
/* ignore race with rename; old or new d_parent is okay */ /* ignore race with rename; old or new d_parent is okay */
struct dentry *parent = req->r_dentry->d_parent; struct dentry *parent;
struct inode *dir = d_inode(parent); struct inode *dir;
rcu_read_lock();
parent = req->r_dentry->d_parent;
dir = req->r_locked_dir ? : d_inode_rcu(parent);
if (dir->i_sb != mdsc->fsc->sb) { if (!dir || dir->i_sb != mdsc->fsc->sb) {
/* not this fs! */ /* not this fs or parent went negative */
inode = d_inode(req->r_dentry); inode = d_inode(req->r_dentry);
if (inode)
ihold(inode);
} else if (ceph_snap(dir) != CEPH_NOSNAP) { } else if (ceph_snap(dir) != CEPH_NOSNAP) {
/* direct snapped/virtual snapdir requests /* direct snapped/virtual snapdir requests
* based on parent dir inode */ * based on parent dir inode */
struct dentry *dn = get_nonsnap_parent(parent); inode = get_nonsnap_parent(parent);
inode = d_inode(dn);
dout("__choose_mds using nonsnap parent %p\n", inode); dout("__choose_mds using nonsnap parent %p\n", inode);
} else { } else {
/* dentry target */ /* dentry target */
inode = d_inode(req->r_dentry); inode = d_inode(req->r_dentry);
if (!inode || mode == USE_AUTH_MDS) { if (!inode || mode == USE_AUTH_MDS) {
/* dir + name */ /* dir + name */
inode = dir; inode = igrab(dir);
hash = ceph_dentry_hash(dir, req->r_dentry); hash = ceph_dentry_hash(dir, req->r_dentry);
is_hash = true; is_hash = true;
} else {
ihold(inode);
} }
} }
rcu_read_unlock();
} }
dout("__choose_mds %p is_hash=%d (%d) mode %d\n", inode, (int)is_hash, dout("__choose_mds %p is_hash=%d (%d) mode %d\n", inode, (int)is_hash,
...@@ -769,7 +786,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc, ...@@ -769,7 +786,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
(int)r, frag.ndist); (int)r, frag.ndist);
if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >= if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
CEPH_MDS_STATE_ACTIVE) CEPH_MDS_STATE_ACTIVE)
return mds; goto out;
} }
/* since this file/dir wasn't known to be /* since this file/dir wasn't known to be
...@@ -784,7 +801,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc, ...@@ -784,7 +801,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
inode, ceph_vinop(inode), frag.frag, mds); inode, ceph_vinop(inode), frag.frag, mds);
if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >= if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
CEPH_MDS_STATE_ACTIVE) CEPH_MDS_STATE_ACTIVE)
return mds; goto out;
} }
} }
} }
...@@ -797,6 +814,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc, ...@@ -797,6 +814,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
cap = rb_entry(rb_first(&ci->i_caps), struct ceph_cap, ci_node); cap = rb_entry(rb_first(&ci->i_caps), struct ceph_cap, ci_node);
if (!cap) { if (!cap) {
spin_unlock(&ci->i_ceph_lock); spin_unlock(&ci->i_ceph_lock);
iput(inode);
goto random; goto random;
} }
mds = cap->session->s_mds; mds = cap->session->s_mds;
...@@ -804,6 +822,8 @@ static int __choose_mds(struct ceph_mds_client *mdsc, ...@@ -804,6 +822,8 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
inode, ceph_vinop(inode), mds, inode, ceph_vinop(inode), mds,
cap == ci->i_auth_cap ? "auth " : "", cap); cap == ci->i_auth_cap ? "auth " : "", cap);
spin_unlock(&ci->i_ceph_lock); spin_unlock(&ci->i_ceph_lock);
out:
iput(inode);
return mds; return mds;
random: random:
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment