From 48db892356d6cb80f6942885545de4a6dd8d2a29 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 19 Feb 2026 16:50:16 -0500 Subject: [PATCH 1/4] NFSD: Defer sub-object cleanup in export put callbacks svc_export_put() calls path_put() and auth_domain_put() immediately when the last reference drops, before the RCU grace period. RCU readers in e_show() and c_show() access both ex_path (via seq_path/d_path) and ex_client->name (via seq_escape) without holding a reference. If cache_clean removes the entry and drops the last reference concurrently, the sub-objects are freed while still in use, producing a NULL pointer dereference in d_path. Commit 2530766492ec ("nfsd: fix UAF when access ex_uuid or ex_stats") moved kfree of ex_uuid and ex_stats into the call_rcu callback, but left path_put() and auth_domain_put() running before the grace period because both may sleep and call_rcu callbacks execute in softirq context. Replace call_rcu/kfree_rcu with queue_rcu_work(), which defers the callback until after the RCU grace period and executes it in process context where sleeping is permitted. This allows path_put() and auth_domain_put() to be moved into the deferred callback alongside the other resource releases. Apply the same fix to expkey_put(), which has the identical pattern with ek_path and ek_client. A dedicated workqueue scopes the shutdown drain to only NFSD export release work items; flushing the shared system_unbound_wq would stall on unrelated work from other subsystems. nfsd_export_shutdown() uses rcu_barrier() followed by flush_workqueue() to ensure all deferred release callbacks complete before the export caches are destroyed. Reported-by: Misbah Anjum N Closes: https://lore.kernel.org/linux-nfs/dcd371d3a95815a84ba7de52cef447b8@linux.ibm.com/ Fixes: c224edca7af0 ("nfsd: no need get cache ref when protected by rcu") Fixes: 1b10f0b603c0 ("SUNRPC: no need get cache ref when protected by rcu") Cc: stable@vger.kernel.org Reviwed-by: Jeff Layton Reviewed-by: NeilBrown Tested-by: Olga Kornievskaia Signed-off-by: Chuck Lever --- fs/nfsd/export.c | 63 +++++++++++++++++++++++++++++++++++++++++------- fs/nfsd/export.h | 7 ++++-- fs/nfsd/nfsctl.c | 8 +++++- 3 files changed, 66 insertions(+), 12 deletions(-) diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 2a1499f2ad19..18933ca407be 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -36,19 +36,30 @@ * second map contains a reference to the entry in the first map. */ +static struct workqueue_struct *nfsd_export_wq; + #define EXPKEY_HASHBITS 8 #define EXPKEY_HASHMAX (1 << EXPKEY_HASHBITS) #define EXPKEY_HASHMASK (EXPKEY_HASHMAX -1) -static void expkey_put(struct kref *ref) +static void expkey_release(struct work_struct *work) { - struct svc_expkey *key = container_of(ref, struct svc_expkey, h.ref); + struct svc_expkey *key = container_of(to_rcu_work(work), + struct svc_expkey, ek_rwork); if (test_bit(CACHE_VALID, &key->h.flags) && !test_bit(CACHE_NEGATIVE, &key->h.flags)) path_put(&key->ek_path); auth_domain_put(key->ek_client); - kfree_rcu(key, ek_rcu); + kfree(key); +} + +static void expkey_put(struct kref *ref) +{ + struct svc_expkey *key = container_of(ref, struct svc_expkey, h.ref); + + INIT_RCU_WORK(&key->ek_rwork, expkey_release); + queue_rcu_work(nfsd_export_wq, &key->ek_rwork); } static int expkey_upcall(struct cache_detail *cd, struct cache_head *h) @@ -353,11 +364,13 @@ static void export_stats_destroy(struct export_stats *stats) EXP_STATS_COUNTERS_NUM); } -static void svc_export_release(struct rcu_head *rcu_head) +static void svc_export_release(struct work_struct *work) { - struct svc_export *exp = container_of(rcu_head, struct svc_export, - ex_rcu); + struct svc_export *exp = container_of(to_rcu_work(work), + struct svc_export, ex_rwork); + path_put(&exp->ex_path); + auth_domain_put(exp->ex_client); nfsd4_fslocs_free(&exp->ex_fslocs); export_stats_destroy(exp->ex_stats); kfree(exp->ex_stats); @@ -369,9 +382,8 @@ static void svc_export_put(struct kref *ref) { struct svc_export *exp = container_of(ref, struct svc_export, h.ref); - path_put(&exp->ex_path); - auth_domain_put(exp->ex_client); - call_rcu(&exp->ex_rcu, svc_export_release); + INIT_RCU_WORK(&exp->ex_rwork, svc_export_release); + queue_rcu_work(nfsd_export_wq, &exp->ex_rwork); } static int svc_export_upcall(struct cache_detail *cd, struct cache_head *h) @@ -1478,6 +1490,36 @@ const struct seq_operations nfs_exports_op = { .show = e_show, }; +/** + * nfsd_export_wq_init - allocate the export release workqueue + * + * Called once at module load. The workqueue runs deferred svc_export and + * svc_expkey release work scheduled by queue_rcu_work() in the cache put + * callbacks. + * + * Return values: + * %0: workqueue allocated + * %-ENOMEM: allocation failed + */ +int nfsd_export_wq_init(void) +{ + nfsd_export_wq = alloc_workqueue("nfsd_export", WQ_UNBOUND, 0); + if (!nfsd_export_wq) + return -ENOMEM; + return 0; +} + +/** + * nfsd_export_wq_shutdown - drain and free the export release workqueue + * + * Called once at module unload. Per-namespace teardown in + * nfsd_export_shutdown() has already drained all deferred work. + */ +void nfsd_export_wq_shutdown(void) +{ + destroy_workqueue(nfsd_export_wq); +} + /* * Initialize the exports module. */ @@ -1539,6 +1581,9 @@ nfsd_export_shutdown(struct net *net) cache_unregister_net(nn->svc_expkey_cache, net); cache_unregister_net(nn->svc_export_cache, net); + /* Drain deferred export and expkey release work. */ + rcu_barrier(); + flush_workqueue(nfsd_export_wq); cache_destroy_net(nn->svc_expkey_cache, net); cache_destroy_net(nn->svc_export_cache, net); svcauth_unix_purge(net); diff --git a/fs/nfsd/export.h b/fs/nfsd/export.h index d2b09cd76145..b05399374574 100644 --- a/fs/nfsd/export.h +++ b/fs/nfsd/export.h @@ -7,6 +7,7 @@ #include #include +#include #include #include @@ -75,7 +76,7 @@ struct svc_export { u32 ex_layout_types; struct nfsd4_deviceid_map *ex_devid_map; struct cache_detail *cd; - struct rcu_head ex_rcu; + struct rcu_work ex_rwork; unsigned long ex_xprtsec_modes; struct export_stats *ex_stats; }; @@ -92,7 +93,7 @@ struct svc_expkey { u32 ek_fsid[6]; struct path ek_path; - struct rcu_head ek_rcu; + struct rcu_work ek_rwork; }; #define EX_ISSYNC(exp) (!((exp)->ex_flags & NFSEXP_ASYNC)) @@ -110,6 +111,8 @@ __be32 check_nfsd_access(struct svc_export *exp, struct svc_rqst *rqstp, /* * Function declarations */ +int nfsd_export_wq_init(void); +void nfsd_export_wq_shutdown(void); int nfsd_export_init(struct net *); void nfsd_export_shutdown(struct net *); void nfsd_export_flush(struct net *); diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 369da69d5efe..fe3b3f206aa9 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -2259,9 +2259,12 @@ static int __init init_nfsd(void) if (retval) goto out_free_pnfs; nfsd_lockd_init(); /* lockd->nfsd callbacks */ + retval = nfsd_export_wq_init(); + if (retval) + goto out_free_lockd; retval = register_pernet_subsys(&nfsd_net_ops); if (retval < 0) - goto out_free_lockd; + goto out_free_export_wq; retval = register_cld_notifier(); if (retval) goto out_free_subsys; @@ -2290,6 +2293,8 @@ out_free_cld: unregister_cld_notifier(); out_free_subsys: unregister_pernet_subsys(&nfsd_net_ops); +out_free_export_wq: + nfsd_export_wq_shutdown(); out_free_lockd: nfsd_lockd_shutdown(); nfsd_drc_slab_free(); @@ -2310,6 +2315,7 @@ static void __exit exit_nfsd(void) nfsd4_destroy_laundry_wq(); unregister_cld_notifier(); unregister_pernet_subsys(&nfsd_net_ops); + nfsd_export_wq_shutdown(); nfsd_drc_slab_free(); nfsd_lockd_shutdown(); nfsd4_free_slabs(); From e7fcf179b82d3a3730fd8615da01b087cc654d0b Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 19 Feb 2026 16:50:17 -0500 Subject: [PATCH 2/4] NFSD: Hold net reference for the lifetime of /proc/fs/nfs/exports fd The /proc/fs/nfs/exports proc entry is created at module init and persists for the module's lifetime. exports_proc_open() captures the caller's current network namespace and stores its svc_export_cache in seq->private, but takes no reference on the namespace. If the namespace is subsequently torn down (e.g. container destruction after the opener does setns() to a different namespace), nfsd_net_exit() calls nfsd_export_shutdown() which frees the cache. Subsequent reads on the still-open fd dereference the freed cache_detail, walking a freed hash table. Hold a reference on the struct net for the lifetime of the open file descriptor. This prevents nfsd_net_exit() from running -- and thus prevents nfsd_export_shutdown() from freeing the cache -- while any exports fd is open. cache_detail already stores its net pointer (cd->net, set by cache_create_net()), so exports_release() can retrieve it without additional per-file storage. Reported-by: Misbah Anjum N Closes: https://lore.kernel.org/linux-nfs/dcd371d3a95815a84ba7de52cef447b8@linux.ibm.com/ Fixes: 96d851c4d28d ("nfsd: use proper net while reading "exports" file") Cc: stable@vger.kernel.org Reviewed-by: Jeff Layton Reviewed-by: NeilBrown Tested-by: Olga Kornievskaia Signed-off-by: Chuck Lever --- fs/nfsd/nfsctl.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index fe3b3f206aa9..d67c169526d0 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -149,9 +149,19 @@ static int exports_net_open(struct net *net, struct file *file) seq = file->private_data; seq->private = nn->svc_export_cache; + get_net(net); return 0; } +static int exports_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = file->private_data; + struct cache_detail *cd = seq->private; + + put_net(cd->net); + return seq_release(inode, file); +} + static int exports_nfsd_open(struct inode *inode, struct file *file) { return exports_net_open(inode->i_sb->s_fs_info, file); @@ -161,7 +171,7 @@ static const struct file_operations exports_nfsd_operations = { .open = exports_nfsd_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release, + .release = exports_release, }; static int export_features_show(struct seq_file *m, void *v) @@ -1376,7 +1386,7 @@ static const struct proc_ops exports_proc_ops = { .proc_open = exports_proc_open, .proc_read = seq_read, .proc_lseek = seq_lseek, - .proc_release = seq_release, + .proc_release = exports_release, }; static int create_proc_exports_entry(void) From 17ad31b3a43b72aec3a3d83605891e1397d0d065 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Mon, 23 Feb 2026 12:09:58 -0500 Subject: [PATCH 3/4] sunrpc: fix cache_request leak in cache_release When a reader's file descriptor is closed while in the middle of reading a cache_request (rp->offset != 0), cache_release() decrements the request's readers count but never checks whether it should free the request. In cache_read(), when readers drops to 0 and CACHE_PENDING is clear, the cache_request is removed from the queue and freed along with its buffer and cache_head reference. cache_release() lacks this cleanup. The only other path that frees requests with readers == 0 is cache_dequeue(), but it runs only when CACHE_PENDING transitions from set to clear. If that transition already happened while readers was still non-zero, cache_dequeue() will have skipped the request, and no subsequent call will clean it up. Add the same cleanup logic from cache_read() to cache_release(): after decrementing readers, check if it reached 0 with CACHE_PENDING clear, and if so, dequeue and free the cache_request. Reported-by: NeilBrown Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@kernel.org Signed-off-by: Jeff Layton Signed-off-by: Chuck Lever --- net/sunrpc/cache.c | 26 +++++++++++++++++++++----- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index 131090f31e6a..6f6e0d4928af 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -1061,14 +1061,25 @@ static int cache_release(struct inode *inode, struct file *filp, struct cache_reader *rp = filp->private_data; if (rp) { + struct cache_request *rq = NULL; + spin_lock(&queue_lock); if (rp->offset) { struct cache_queue *cq; - for (cq= &rp->q; &cq->list != &cd->queue; - cq = list_entry(cq->list.next, struct cache_queue, list)) + for (cq = &rp->q; &cq->list != &cd->queue; + cq = list_entry(cq->list.next, + struct cache_queue, list)) if (!cq->reader) { - container_of(cq, struct cache_request, q) - ->readers--; + struct cache_request *cr = + container_of(cq, + struct cache_request, q); + cr->readers--; + if (cr->readers == 0 && + !test_bit(CACHE_PENDING, + &cr->item->flags)) { + list_del(&cr->q.list); + rq = cr; + } break; } rp->offset = 0; @@ -1076,9 +1087,14 @@ static int cache_release(struct inode *inode, struct file *filp, list_del(&rp->q.list); spin_unlock(&queue_lock); + if (rq) { + cache_put(rq->item, cd); + kfree(rq->buf); + kfree(rq); + } + filp->private_data = NULL; kfree(rp); - } if (filp->f_mode & FMODE_WRITE) { atomic_dec(&cd->writers); From 5133b61aaf437e5f25b1b396b14242a6bb0508e2 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 24 Feb 2026 11:33:35 -0500 Subject: [PATCH 4/4] nfsd: fix heap overflow in NFSv4.0 LOCK replay cache The NFSv4.0 replay cache uses a fixed 112-byte inline buffer (rp_ibuf[NFSD4_REPLAY_ISIZE]) to store encoded operation responses. This size was calculated based on OPEN responses and does not account for LOCK denied responses, which include the conflicting lock owner as a variable-length field up to 1024 bytes (NFS4_OPAQUE_LIMIT). When a LOCK operation is denied due to a conflict with an existing lock that has a large owner, nfsd4_encode_operation() copies the full encoded response into the undersized replay buffer via read_bytes_from_xdr_buf() with no bounds check. This results in a slab-out-of-bounds write of up to 944 bytes past the end of the buffer, corrupting adjacent heap memory. This can be triggered remotely by an unauthenticated attacker with two cooperating NFSv4.0 clients: one sets a lock with a large owner string, then the other requests a conflicting lock to provoke the denial. We could fix this by increasing NFSD4_REPLAY_ISIZE to allow for a full opaque, but that would increase the size of every stateowner, when most lockowners are not that large. Instead, fix this by checking the encoded response length against NFSD4_REPLAY_ISIZE before copying into the replay buffer. If the response is too large, set rp_buflen to 0 to skip caching the replay payload. The status is still cached, and the client already received the correct response on the original request. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@kernel.org Reported-by: Nicholas Carlini Tested-by: Nicholas Carlini Signed-off-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/nfsd/nfs4xdr.c | 9 +++++++-- fs/nfsd/state.h | 17 ++++++++++++----- 2 files changed, 19 insertions(+), 7 deletions(-) diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 5172dbd0cb05..fa16b34fae50 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -6281,9 +6281,14 @@ nfsd4_encode_operation(struct nfsd4_compoundres *resp, struct nfsd4_op *op) int len = xdr->buf->len - (op_status_offset + XDR_UNIT); so->so_replay.rp_status = op->status; - so->so_replay.rp_buflen = len; - read_bytes_from_xdr_buf(xdr->buf, op_status_offset + XDR_UNIT, + if (len <= NFSD4_REPLAY_ISIZE) { + so->so_replay.rp_buflen = len; + read_bytes_from_xdr_buf(xdr->buf, + op_status_offset + XDR_UNIT, so->so_replay.rp_buf, len); + } else { + so->so_replay.rp_buflen = 0; + } } status: op->status = nfsd4_map_status(op->status, diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index 6fcbf1e427d4..c0ca115c3b74 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -541,11 +541,18 @@ struct nfs4_client_reclaim { struct xdr_netobj cr_princhash; }; -/* A reasonable value for REPLAY_ISIZE was estimated as follows: - * The OPEN response, typically the largest, requires - * 4(status) + 8(stateid) + 20(changeinfo) + 4(rflags) + 8(verifier) + - * 4(deleg. type) + 8(deleg. stateid) + 4(deleg. recall flag) + - * 20(deleg. space limit) + ~32(deleg. ace) = 112 bytes +/* + * REPLAY_ISIZE is sized for an OPEN response with delegation: + * 4(status) + 8(stateid) + 20(changeinfo) + 4(rflags) + + * 8(verifier) + 4(deleg. type) + 8(deleg. stateid) + + * 4(deleg. recall flag) + 20(deleg. space limit) + + * ~32(deleg. ace) = 112 bytes + * + * Some responses can exceed this. A LOCK denial includes the conflicting + * lock owner, which can be up to 1024 bytes (NFS4_OPAQUE_LIMIT). Responses + * larger than REPLAY_ISIZE are not cached in rp_ibuf; only rp_status is + * saved. Enlarging this constant increases the size of every + * nfs4_stateowner. */ #define NFSD4_REPLAY_ISIZE 112