selftests/bpf: Reject graph kptr use after RCU unlockgrafted

Add a sleepable verifier test that loads a graph-node local kptr in an
explicit RCU read-side critical section, then passes its node to
bpf_rbtree_remove() after the section ends.

Before the verifier fix, the stale NON_OWN_REF flag makes the node look like
a live borrowed reference and the program is accepted. After the fix, the
pointer is demoted without NON_OWN_REF and the graph kfunc argument is
rejected.

Also exercise a graph kptr loaded while a spin lock provides implicit RCU
protection. The pointer must be invalidated when the lock is released, which
guards the required ordering between non-owning-reference invalidation and
RCU demotion.

Update the existing fault-protected load test state description. The
post-unlock pointer no longer carries NON_OWN_REF, but remains readable
because the load is rewritten to use BPF_PROBE_MEM.

Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20260904084325.52250-7-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

1 files changed, 247 insertions, 0 deletions

diff --git a/fs/nfsd/netns.h b/fs/nfsd/netns.h
new file mode 100644
index 000000000..27da1a3ed
--- /dev/null
+++ b/fs/nfsd/netns.h
@@ -0,0 +1,247 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * per net namespace data structures for nfsd
+ *
+ * Copyright (C) 2012, Jeff Layton <jlayton@redhat.com>
+ */
+
+#ifndef __NFSD_NETNS_H__
+#define __NFSD_NETNS_H__
+
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+#include <linux/filelock.h>
+#include <linux/nfs4.h>
+#include <linux/percpu_counter.h>
+#include <linux/percpu-refcount.h>
+#include <linux/siphash.h>
+#include <linux/sunrpc/stats.h>
+
+/* Hash tables for nfs4_clientid state */
+#define CLIENT_HASH_BITS                 4
+#define CLIENT_HASH_SIZE                (1 << CLIENT_HASH_BITS)
+#define CLIENT_HASH_MASK                (CLIENT_HASH_SIZE - 1)
+
+#define SESSION_HASH_SIZE	512
+
+struct cld_net;
+struct nfsd_net_cb;
+struct nfsd4_client_tracking_ops;
+
+enum {
+	/* cache misses due only to checksum comparison failures */
+	NFSD_STATS_PAYLOAD_MISSES,
+	/* amount of memory (in bytes) currently consumed by the DRC */
+	NFSD_STATS_DRC_MEM_USAGE,
+	NFSD_STATS_RC_HITS,		/* repcache hits */
+	NFSD_STATS_RC_MISSES,		/* repcache misses */
+	NFSD_STATS_RC_NOCACHE,		/* uncached reqs */
+	NFSD_STATS_FH_STALE,		/* FH stale error */
+	NFSD_STATS_IO_READ,		/* bytes returned to read requests */
+	NFSD_STATS_IO_WRITE,		/* bytes passed in write requests */
+#ifdef CONFIG_NFSD_V4
+	NFSD_STATS_FIRST_NFS4_OP,	/* count of individual nfsv4 operations */
+	NFSD_STATS_LAST_NFS4_OP = NFSD_STATS_FIRST_NFS4_OP + LAST_NFS4_OP,
+#define NFSD_STATS_NFS4_OP(op)	(NFSD_STATS_FIRST_NFS4_OP + (op))
+	NFSD_STATS_WDELEG_GETATTR,	/* count of getattr conflict with wdeleg */
+#endif
+	NFSD_STATS_COUNTERS_NUM
+};
+
+/*
+ * Represents a nfsd "container". With respect to nfsv4 state tracking, the
+ * fields of interest are the *_id_hashtbls and the *_name_tree. These track
+ * the nfs4_client objects by either short or long form clientid.
+ *
+ * Each nfsd_net runs a nfs4_laundromat workqueue job when necessary to clean
+ * up expired clients and delegations within the container.
+ */
+struct nfsd_net {
+	struct cld_net *cld_net;
+
+	struct cache_detail *svc_expkey_cache;
+	struct cache_detail *svc_export_cache;
+
+	struct cache_detail *idtoname_cache;
+	struct cache_detail *nametoid_cache;
+
+	struct lock_manager nfsd4_manager;
+	bool grace_ended;
+	bool grace_end_forced;
+	time64_t boot_time;
+
+	struct dentry *nfsd_client_dir;
+
+	/*
+	 * reclaim_str_hashtbl[] holds known client info from previous reset/reboot
+	 * used in reboot/reset lease grace period processing
+	 *
+	 * conf_id_hashtbl[], and conf_name_tree hold confirmed
+	 * setclientid_confirmed info.
+	 *
+	 * unconf_str_hastbl[] and unconf_name_tree hold unconfirmed
+	 * setclientid info.
+	 */
+	struct list_head *reclaim_str_hashtbl;
+	int reclaim_str_hashtbl_size;
+	struct list_head *conf_id_hashtbl;
+	struct rb_root conf_name_tree;
+	struct list_head *unconf_id_hashtbl;
+	struct rb_root unconf_name_tree;
+	struct list_head *sessionid_hashtbl;
+	/*
+	 * client_lru holds client queue ordered by nfs4_client.cl_time
+	 * for lease renewal.
+	 *
+	 * close_lru holds (open) stateowner queue ordered by nfs4_stateowner.so_time
+	 * for last close replay.
+	 *
+	 * All of the above fields are protected by the client_mutex.
+	 */
+	struct list_head client_lru;
+	struct list_head close_lru;
+
+	/* protects del_recall_lru and delegation hash/unhash */
+	spinlock_t deleg_lock ____cacheline_aligned;
+	struct list_head del_recall_lru;
+
+	/* protected by blocked_locks_lock */
+	struct list_head blocked_locks_lru;
+
+	struct delayed_work laundromat_work;
+
+	/* client_lock protects the client lru list and session hash table */
+	spinlock_t client_lock;
+
+	/* protects blocked_locks_lru */
+	spinlock_t blocked_locks_lock;
+
+	struct file *rec_file;
+	bool in_grace;
+	const struct nfsd4_client_tracking_ops *client_tracking_ops;
+
+	time64_t nfsd4_lease;
+	time64_t nfsd4_grace;
+	bool somebody_reclaimed;
+
+	bool track_reclaim_completes;
+	atomic_t nr_reclaim_complete;
+
+	bool nfsd_net_up;
+	bool lockd_up;
+
+	seqlock_t writeverf_lock;
+	unsigned char writeverf[8];
+
+	/*
+	 * Minimum number of threads to run per pool.  If 0 then the
+	 * min == max requested number of threads.
+	 */
+	unsigned int min_threads;
+
+	u32 clientid_base;
+	u32 clientid_counter;
+	u32 clverifier_counter;
+
+	struct svc_info nfsd_info;
+#define nfsd_serv nfsd_info.serv
+
+	struct percpu_ref nfsd_net_ref;
+	struct completion nfsd_net_confirm_done;
+	struct completion nfsd_net_free_done;
+
+	/*
+	 * clientid and stateid data for construction of net unique COPY
+	 * stateids.
+	 */
+	u32		s2s_cp_cl_id;
+	struct idr	s2s_cp_stateids;
+	spinlock_t	s2s_cp_lock;
+	atomic_t	pending_async_copies;
+
+	/*
+	 * Version information
+	 */
+	bool nfsd_versions[NFSD_MAXVERS + 1];
+	bool nfsd4_minorversions[NFSD_SUPPORTED_MINOR_VERSION + 1];
+
+	/*
+	 * Duplicate reply cache
+	 */
+	struct nfsd_drc_bucket   *drc_hashtbl;
+
+	/* max number of entries allowed in the cache */
+	unsigned int             max_drc_entries;
+
+	/* number of significant bits in the hash value */
+	unsigned int             maskbits;
+	unsigned int             drc_hashsize;
+
+	/*
+	 * Stats and other tracking of on the duplicate reply cache.
+	 * The longest_chain* fields are modified with only the per-bucket
+	 * cache lock, which isn't really safe and should be fixed if we want
+	 * these statistics to be completely accurate.
+	 */
+
+	/* total number of entries */
+	atomic_t                 num_drc_entries;
+
+	/* Per-netns stats counters */
+	struct percpu_counter    counter[NFSD_STATS_COUNTERS_NUM];
+
+	/* sunrpc svc stats */
+	struct svc_stat          nfsd_svcstats;
+
+	/* longest hash chain seen */
+	unsigned int             longest_chain;
+
+	/* size of cache when we saw the longest hash chain */
+	unsigned int             longest_chain_cachesize;
+
+	struct shrinker		*nfsd_reply_cache_shrinker;
+
+	/* tracking server-to-server copy mounts */
+	spinlock_t              nfsd_ssc_lock;
+	struct list_head        nfsd_ssc_mount_list;
+	wait_queue_head_t       nfsd_ssc_waitq;
+
+	/* utsname taken from the process that starts the server */
+	char			nfsd_name[UNX_MAXNODENAME+1];
+
+	struct nfsd_fcache_disposal *fcache_disposal;
+
+	siphash_key_t		siphash_key;
+
+	atomic_t		nfs4_client_count;
+	int			nfs4_max_clients;
+
+	atomic_t		nfsd_courtesy_clients;
+	struct shrinker		*nfsd_client_shrinker;
+	struct work_struct	nfsd_shrinker_work;
+
+	/* last time an admin-revoke happened for NFSv4.0 */
+	time64_t		nfs40_last_revoke;
+
+#if IS_ENABLED(CONFIG_NFS_LOCALIO)
+	/* Local clients to be invalidated when net is shut down */
+	spinlock_t              local_clients_lock;
+	struct list_head	local_clients;
+#endif
+	siphash_key_t		*fh_key;
+
+	struct nfsd_net_cb	*nfsd_cb;
+};
+
+/* Simple check to find out if a given net was properly initialized */
+#define nfsd_netns_ready(nn) ((nn)->sessionid_hashtbl)
+
+extern bool nfsd_support_version(int vers);
+extern unsigned int nfsd_net_id;
+
+bool nfsd_net_try_get(struct net *net);
+void nfsd_net_put(struct net *net);
+
+void nfsd_copy_write_verifier(__be32 verf[2], struct nfsd_net *nn);
+void nfsd_reset_write_verifier(struct nfsd_net *nn);
+#endif /* __NFSD_NETNS_H__ */