Files
linux/fs/nfsd/export.h
Chuck Lever 48db892356 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 2530766492 ("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 <misanjum@linux.ibm.com>
Closes: https://lore.kernel.org/linux-nfs/dcd371d3a95815a84ba7de52cef447b8@linux.ibm.com/
Fixes: c224edca7a ("nfsd: no need get cache ref when protected by rcu")
Fixes: 1b10f0b603 ("SUNRPC: no need get cache ref when protected by rcu")
Cc: stable@vger.kernel.org
Reviwed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: NeilBrown <neil@brown.name>
Tested-by: Olga Kornievskaia <okorniev@redhat.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2026-03-14 11:34:25 -04:00

143 lines
3.8 KiB
C

/* SPDX-License-Identifier: GPL-2.0 */
/*
* Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
*/
#ifndef NFSD_EXPORT_H
#define NFSD_EXPORT_H
#include <linux/sunrpc/cache.h>
#include <linux/percpu_counter.h>
#include <linux/workqueue.h>
#include <uapi/linux/nfsd/export.h>
#include <linux/nfs4.h>
struct knfsd_fh;
struct svc_fh;
struct svc_rqst;
/*
* FS Locations
*/
#define MAX_FS_LOCATIONS 128
struct nfsd4_fs_location {
char *hosts; /* colon separated list of hosts */
char *path; /* slash separated list of path components */
};
struct nfsd4_fs_locations {
uint32_t locations_count;
struct nfsd4_fs_location *locations;
/* If we're not actually serving this data ourselves (only providing a
* list of replicas that do serve it) then we set "migrated": */
int migrated;
};
/*
* We keep an array of pseudoflavors with the export, in order from most
* to least preferred. For the foreseeable future, we don't expect more
* than the eight pseudoflavors null, unix, krb5, krb5i, krb5p, skpm3,
* spkm3i, and spkm3p (and using all 8 at once should be rare).
*/
#define MAX_SECINFO_LIST 8
#define EX_UUID_LEN 16
struct exp_flavor_info {
u32 pseudoflavor;
u32 flags;
};
/* Per-export stats */
enum {
EXP_STATS_FH_STALE,
EXP_STATS_IO_READ,
EXP_STATS_IO_WRITE,
EXP_STATS_COUNTERS_NUM
};
struct export_stats {
time64_t start_time;
struct percpu_counter counter[EXP_STATS_COUNTERS_NUM];
};
struct svc_export {
struct cache_head h;
struct auth_domain * ex_client;
int ex_flags;
int ex_fsid;
struct path ex_path;
kuid_t ex_anon_uid;
kgid_t ex_anon_gid;
unsigned char * ex_uuid; /* 16 byte fsid */
struct nfsd4_fs_locations ex_fslocs;
uint32_t ex_nflavors;
struct exp_flavor_info ex_flavors[MAX_SECINFO_LIST];
u32 ex_layout_types;
struct nfsd4_deviceid_map *ex_devid_map;
struct cache_detail *cd;
struct rcu_work ex_rwork;
unsigned long ex_xprtsec_modes;
struct export_stats *ex_stats;
};
/* an "export key" (expkey) maps a filehandlefragement to an
* svc_export for a given client. There can be several per export,
* for the different fsid types.
*/
struct svc_expkey {
struct cache_head h;
struct auth_domain * ek_client;
u8 ek_fsidtype;
u32 ek_fsid[6];
struct path ek_path;
struct rcu_work ek_rwork;
};
#define EX_ISSYNC(exp) (!((exp)->ex_flags & NFSEXP_ASYNC))
#define EX_NOHIDE(exp) ((exp)->ex_flags & NFSEXP_NOHIDE)
#define EX_WGATHER(exp) ((exp)->ex_flags & NFSEXP_GATHERED_WRITES)
struct svc_cred;
int nfsexp_flags(struct svc_cred *cred, struct svc_export *exp);
__be32 check_xprtsec_policy(struct svc_export *exp, struct svc_rqst *rqstp);
__be32 check_security_flavor(struct svc_export *exp, struct svc_rqst *rqstp,
bool may_bypass_gss);
__be32 check_nfsd_access(struct svc_export *exp, struct svc_rqst *rqstp,
bool may_bypass_gss);
/*
* 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 *);
struct svc_export * rqst_exp_get_by_name(struct svc_rqst *,
const struct path *);
struct svc_export * rqst_exp_parent(struct svc_rqst *,
struct path *);
struct svc_export * rqst_find_fsidzero_export(struct svc_rqst *);
int exp_rootfh(struct net *, struct auth_domain *,
char *path, struct knfsd_fh *, int maxsize);
__be32 exp_pseudoroot(struct svc_rqst *, struct svc_fh *);
static inline void exp_put(struct svc_export *exp)
{
cache_put(&exp->h, exp->cd);
}
static inline struct svc_export *exp_get(struct svc_export *exp)
{
cache_get(&exp->h);
return exp;
}
struct svc_export *rqst_exp_find(struct cache_req *reqp, struct net *net,
struct auth_domain *cl, struct auth_domain *gsscl,
int fsid_type, u32 *fsidv);
#endif /* NFSD_EXPORT_H */