From: Chuck Lever <cel@kernel.org>
To: Jeff Layton <jlayton@kernel.org>, NeilBrown <neil@brown.name>,
Olga Kornievskaia <okorniev@redhat.com>,
Dai Ngo <Dai.Ngo@oracle.com>, Tom Talpey <tom@talpey.com>
Cc: linux-nfs@vger.kernel.org, Chuck Lever <cel@kernel.org>
Subject: [PATCH v2 2/6] NFSD: Prevent client use-after-free during delegation revoke
Date: Sun, 05 Jul 2026 21:25:38 -0400 [thread overview]
Message-ID: <20260705-cel-v2-2-d88c3b68e8bc@kernel.org> (raw)
In-Reply-To: <20260705-cel-v2-0-d88c3b68e8bc@kernel.org>
A delegation stateid stores only a bare pointer to its owning
nfs4_client; a reference on the stateid does not keep the client
alive. The client outlives its stateids solely because
__destroy_client() drains every delegation from cl_delegations and
cl_revoked before free_client() runs.
nfs4_laundromat() breaks that invariant. It unhashes an expired
delegation from cl_delegations under deleg_lock, drops the lock, and
calls revoke_delegation(), which reacquires clp->cl_lock and links the
delegation onto clp->cl_revoked. Between the unhash and the revoke the
delegation is on neither client-reachable list, so client_has_state()
can report no remaining state.
Every path that can free a client holding delegations first requires
cl_rpc_users to be zero: DESTROY_CLIENTID and a superseding EXCHANGE_ID
gate on mark_client_expired_locked(), and the admin "expire" write
waits in force_expire_client(). The laundromat holds no such
reference. A client whose recalled delegation has just timed out -- a
rebooted client whose new incarnation supersedes the old one, say --
can therefore run __destroy_client() to completion and free the client
while revoke_delegation() is still about to dereference clp->cl_lock
and clp->cl_revoked, a use-after-free.
Pin the client with cl_rpc_users across the revoke so the teardown
paths block until it completes and then reap it from cl_revoked
themselves. A client already expiring reaps its own delegations, so
skip it and leave the delegation on del_recall_lru for
__destroy_client() to handle.
Fixes: 3bd64a5ba171 ("nfsd4: implement SEQ4_STATUS_RECALLABLE_STATE_REVOKED")
Signed-off-by: Chuck Lever <cel@kernel.org>
---
fs/nfsd/netns.h | 6 ++++--
fs/nfsd/nfs4state.c | 23 +++++++++++++++++++++++
2 files changed, 27 insertions(+), 2 deletions(-)
diff --git a/fs/nfsd/netns.h b/fs/nfsd/netns.h
index 03724bef10a7..a7bd7b67fa4f 100644
--- a/fs/nfsd/netns.h
+++ b/fs/nfsd/netns.h
@@ -115,7 +115,8 @@ struct nfsd_net {
struct list_head client_lru;
struct list_head close_lru;
- /* protects del_recall_lru and delegation hash/unhash */
+ /* protects del_recall_lru and delegation hash/unhash;
+ * nests outside client_lock */
spinlock_t deleg_lock ____cacheline_aligned;
struct list_head del_recall_lru;
@@ -124,7 +125,8 @@ struct nfsd_net {
struct delayed_work laundromat_work;
- /* client_lock protects the client lru list and session hash table */
+ /* client_lock protects the client lru list and session hash
+ * table; nests inside deleg_lock */
spinlock_t client_lock;
/* protects blocked_locks_lru */
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index e000ed3e96e9..efeb2a2e9c8f 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -7457,6 +7457,7 @@ nfs4_laundromat(struct nfsd_net *nn)
.new_timeo = nn->nfsd4_lease
};
struct nfs4_cpntf_state *cps;
+ struct nfs4_client *clp;
copy_stateid_t *cps_t;
int i;
@@ -7485,6 +7486,18 @@ nfs4_laundromat(struct nfsd_net *nn)
dp = list_entry (pos, struct nfs4_delegation, dl_recall_lru);
if (!state_expired(<, dp->dl_time))
break;
+ clp = dp->dl_stid.sc_client;
+ spin_lock(&nn->client_lock);
+ if (is_client_expired(clp)) {
+ spin_unlock(&nn->client_lock);
+ continue;
+ }
+ /*
+ * Pin without reviving: get_client_locked() would
+ * flip a courtesy client back to NFSD4_ACTIVE.
+ */
+ atomic_inc(&clp->cl_rpc_users);
+ spin_unlock(&nn->client_lock);
refcount_inc(&dp->dl_stid.sc_count);
unhash_delegation_locked(dp, SC_STATUS_REVOKED);
list_add(&dp->dl_recall_lru, &reaplist);
@@ -7493,8 +7506,18 @@ nfs4_laundromat(struct nfsd_net *nn)
while (!list_empty(&reaplist)) {
dp = list_first_entry(&reaplist, struct nfs4_delegation,
dl_recall_lru);
+ clp = dp->dl_stid.sc_client;
list_del_init(&dp->dl_recall_lru);
revoke_delegation(dp);
+ /*
+ * Unpin without renewing: put_client_renew() would
+ * renew the reaped client's lease.
+ */
+ if (atomic_dec_and_lock(&clp->cl_rpc_users, &nn->client_lock)) {
+ if (is_client_expired(clp))
+ wake_up_all(&expiry_wq);
+ spin_unlock(&nn->client_lock);
+ }
}
spin_lock(&nn->client_lock);
--
2.54.0
next prev parent reply other threads:[~2026-07-06 1:26 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-06 1:25 [PATCH v2 0/6] NFSD: Fix UAFs in client teardown and state revocation Chuck Lever
2026-07-06 1:25 ` [PATCH v2 1/6] NFSD: Prevent lock owner use-after-free during client teardown Chuck Lever
2026-07-06 1:25 ` Chuck Lever [this message]
2026-07-06 17:29 ` [PATCH v2 2/6] NFSD: Prevent client use-after-free during delegation revoke Jeff Layton
2026-07-06 1:25 ` [PATCH v2 3/6] NFSD: Prevent client use-after-free during admin state revocation Chuck Lever
2026-07-06 1:25 ` [PATCH v2 4/6] NFSD: Prevent client use-after-free during export " Chuck Lever
2026-07-06 1:25 ` [PATCH v2 5/6] NFSD: Prevent client use-after-free during NFSv4.0 revoked-state cleanup Chuck Lever
2026-07-06 1:25 ` [PATCH v2 6/6] NFSD: Consolidate the revocation-path client unpin Chuck Lever
2026-07-06 3:53 ` [PATCH v2 0/6] NFSD: Fix UAFs in client teardown and state revocation NeilBrown
2026-07-06 16:42 ` Jeff Layton
2026-07-06 17:14 ` Chuck Lever
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=20260705-cel-v2-2-d88c3b68e8bc@kernel.org \
--to=cel@kernel.org \
--cc=Dai.Ngo@oracle.com \
--cc=jlayton@kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=neil@brown.name \
--cc=okorniev@redhat.com \
--cc=tom@talpey.com \
/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