Commit bd81ccea authored by Linus Torvalds's avatar Linus Torvalds
Browse files

Merge branch 'for-3.7' of git://linux-nfs.org/~bfields/linux

Pull nfsd update from J Bruce Fields:
 "Another relatively quiet cycle.  There was some progress on my
  remaining 4.1 todo's, but a couple of them were just of the form
  "check that we do X correctly", so didn't have much affect on the
  code.

  Other than that, a bunch of cleanup and some bugfixes (including an
  annoying NFSv4.0 state leak and a busy-loop in the server that could
  cause it to peg the CPU without making progress)."

* 'for-3.7' of git://linux-nfs.org/~bfields/linux: (46 commits)
  UAPI: (Scripted) Disintegrate include/linux/sunrpc
  UAPI: (Scripted) Disintegrate include/linux/nfsd
  nfsd4: don't allow reclaims of expired clients
  nfsd4: remove redundant callback probe
  nfsd4: expire old client earlier
  nfsd4: separate session allocation and initialization
  nfsd4: clean up session allocation
  nfsd4: minor free_session cleanup
  nfsd4: new_conn_from_crses should only allocate
  nfsd4: separate connection allocation and initialization
  nfsd4: reject bad forechannel attrs earlier
  nfsd4: enforce per-client sessions/no-sessions distinction
  nfsd4: set cl_minorversion at create time
  nfsd4: don't pin clientids to pseudoflavors
  nfsd4: fix bind_conn_to_session xdr comment
  nfsd4: cast readlink() bug argument
  NFSD: pass null terminated buf to kstrtouint()
  nfsd: remove duplicate init in nfsd4_cb_recall
  nfsd4: eliminate redundant nfs4_free_stateid
  fs/nfsd/nfs4idmap.c: adjust inconsistent IS_ERR and PTR_ERR
  ...
parents 98260daa a9ca4043
Administrative interfaces for nfsd
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Note that normally these interfaces are used only by the utilities in
nfs-utils.
nfsd is controlled mainly by pseudofiles under the "nfsd" filesystem,
which is normally mounted at /proc/fs/nfsd/.
The server is always started by the first write of a nonzero value to
nfsd/threads.
Before doing that, NFSD can be told which sockets to listen on by
writing to nfsd/portlist; that write may be:
- an ascii-encoded file descriptor, which should refer to a
bound (and listening, for tcp) socket, or
- "transportname port", where transportname is currently either
"udp", "tcp", or "rdma".
If nfsd is started without doing any of these, then it will create one
udp and one tcp listener at port 2049 (see nfsd_init_socks).
On startup, nfsd and lockd grace periods start.
nfsd is shut down by a write of 0 to nfsd/threads. All locks and state
are thrown away at that point.
Between startup and shutdown, the number of threads may be adjusted up
or down by additional writes to nfsd/threads or by writes to
nfsd/pool_threads.
For more detail about files under nfsd/ and what they control, see
fs/nfsd/nfsctl.c; most of them have detailed comments.
Implementation notes
^^^^^^^^^^^^^^^^^^^^
Note that the rpc server requires the caller to serialize addition and
removal of listening sockets, and startup and shutdown of the server.
For nfsd this is done using nfsd_mutex.
......@@ -126,7 +126,7 @@ static void restart_grace(void)
static int
lockd(void *vrqstp)
{
int err = 0, preverr = 0;
int err = 0;
struct svc_rqst *rqstp = vrqstp;
/* try_to_freeze() is called from svc_recv() */
......@@ -165,21 +165,8 @@ lockd(void *vrqstp)
* recvfrom routine.
*/
err = svc_recv(rqstp, timeout);
if (err == -EAGAIN || err == -EINTR) {
preverr = err;
if (err == -EAGAIN || err == -EINTR)
continue;
}
if (err < 0) {
if (err != preverr) {
printk(KERN_WARNING "%s: unexpected error "
"from svc_recv (%d)\n", __func__, err);
preverr = err;
}
schedule_timeout_interruptible(HZ);
continue;
}
preverr = err;
dprintk("lockd: request from %s\n",
svc_print_addr(rqstp, buf, sizeof(buf)));
......
......@@ -1289,7 +1289,7 @@ EXPORT_SYMBOL(__break_lease);
void lease_get_mtime(struct inode *inode, struct timespec *time)
{
struct file_lock *flock = inode->i_flock;
if (flock && IS_LEASE(flock) && (flock->fl_type & F_WRLCK))
if (flock && IS_LEASE(flock) && (flock->fl_type == F_WRLCK))
*time = current_fs_time(inode->i_sb);
else
*time = inode->i_mtime;
......@@ -2185,8 +2185,8 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl,
} else {
seq_printf(f, "%s ",
(lease_breaking(fl))
? (fl->fl_type & F_UNLCK) ? "UNLCK" : "READ "
: (fl->fl_type & F_WRLCK) ? "WRITE" : "READ ");
? (fl->fl_type == F_UNLCK) ? "UNLCK" : "READ "
: (fl->fl_type == F_WRLCK) ? "WRITE" : "READ ");
}
if (inode) {
#ifdef WE_CAN_BREAK_LSLK_NOW
......
......@@ -72,7 +72,7 @@ static int nfs4_callback_up_net(struct svc_serv *serv, struct net *net)
static int
nfs4_callback_svc(void *vrqstp)
{
int err, preverr = 0;
int err;
struct svc_rqst *rqstp = vrqstp;
set_freezable();
......@@ -82,20 +82,8 @@ nfs4_callback_svc(void *vrqstp)
* Listen for a request on the socket
*/
err = svc_recv(rqstp, MAX_SCHEDULE_TIMEOUT);
if (err == -EAGAIN || err == -EINTR) {
preverr = err;
if (err == -EAGAIN || err == -EINTR)
continue;
}
if (err < 0) {
if (err != preverr) {
printk(KERN_WARNING "NFS: %s: unexpected error "
"from svc_recv (%d)\n", __func__, err);
preverr = err;
}
schedule_timeout_uninterruptible(HZ);
continue;
}
preverr = err;
svc_process(rqstp);
}
return 0;
......
......@@ -218,8 +218,7 @@ static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p,
* There must be an encoding function for void results so svc_process
* will work properly.
*/
int
nfsaclsvc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy)
static int nfsaclsvc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy)
{
return xdr_ressize_check(rqstp, p);
}
......
......@@ -247,7 +247,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp,
/* Now create the file and set attributes */
nfserr = do_nfsd_create(rqstp, dirfhp, argp->name, argp->len,
attr, newfhp,
argp->createmode, argp->verf, NULL, NULL);
argp->createmode, (u32 *)argp->verf, NULL, NULL);
RETURN_STATUS(nfserr);
}
......
......@@ -1028,7 +1028,6 @@ void nfsd4_cb_recall(struct nfs4_delegation *dp)
cb->cb_msg.rpc_cred = callback_cred;
cb->cb_ops = &nfsd4_cb_recall_ops;
dp->dl_retries = 1;
INIT_LIST_HEAD(&cb->cb_per_client);
cb->cb_done = true;
......
......@@ -478,7 +478,7 @@ nfsd_idmap_init(struct net *net)
goto destroy_idtoname_cache;
nn->nametoid_cache = cache_create_net(&nametoid_cache_template, net);
if (IS_ERR(nn->nametoid_cache)) {
rv = PTR_ERR(nn->idtoname_cache);
rv = PTR_ERR(nn->nametoid_cache);
goto unregister_idtoname_cache;
}
rv = cache_register_net(nn->nametoid_cache, net);
......@@ -598,7 +598,7 @@ numeric_name_to_id(struct svc_rqst *rqstp, int type, const char *name, u32 namel
/* Just to make sure it's null-terminated: */
memcpy(buf, name, namelen);
buf[namelen] = '\0';
ret = kstrtouint(name, 10, id);
ret = kstrtouint(buf, 10, id);
return ret == 0;
}
......
......@@ -370,7 +370,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
break;
case NFS4_OPEN_CLAIM_PREVIOUS:
open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
status = nfs4_check_open_reclaim(&open->op_clientid);
status = nfs4_check_open_reclaim(&open->op_clientid, cstate->minorversion);
if (status)
goto out;
case NFS4_OPEN_CLAIM_FH:
......@@ -1054,8 +1054,8 @@ struct nfsd4_operation {
char *op_name;
/* Try to get response size before operation */
nfsd4op_rsize op_rsize_bop;
stateid_setter op_get_currentstateid;
stateid_getter op_set_currentstateid;
stateid_getter op_get_currentstateid;
stateid_setter op_set_currentstateid;
};
static struct nfsd4_operation nfsd4_ops[];
......
This diff is collapsed.
......@@ -2659,7 +2659,7 @@ static __be32 nfsd4_encode_bind_conn_to_session(struct nfsd4_compoundres *resp,
RESERVE_SPACE(NFS4_MAX_SESSIONID_LEN + 8);
WRITEMEM(bcts->sessionid.data, NFS4_MAX_SESSIONID_LEN);
WRITE32(bcts->dir);
/* XXX: ? */
/* Sorry, we do not yet support RDMA over 4.1: */
WRITE32(0);
ADJUST_ARGS();
}
......
......@@ -406,7 +406,7 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size)
return rv;
if (newthreads < 0)
return -EINVAL;
rv = nfsd_svc(NFS_PORT, newthreads);
rv = nfsd_svc(newthreads);
if (rv < 0)
return rv;
} else
......@@ -682,25 +682,6 @@ static ssize_t __write_ports_addfd(char *buf)
return err;
}
/*
* A '-' followed by the 'name' of a socket means we close the socket.
*/
static ssize_t __write_ports_delfd(char *buf)
{
char *toclose;
int len = 0;
toclose = kstrdup(buf + 1, GFP_KERNEL);
if (toclose == NULL)
return -ENOMEM;
if (nfsd_serv != NULL)
len = svc_sock_names(nfsd_serv, buf,
SIMPLE_TRANSACTION_LIMIT, toclose);
kfree(toclose);
return len;
}
/*
* A transport listener is added by writing it's transport name and
* a port number.
......@@ -712,7 +693,7 @@ static ssize_t __write_ports_addxprt(char *buf)
int port, err;
struct net *net = &init_net;
if (sscanf(buf, "%15s %4u", transport, &port) != 2)
if (sscanf(buf, "%15s %5u", transport, &port) != 2)
return -EINVAL;
if (port < 1 || port > USHRT_MAX)
......@@ -746,31 +727,6 @@ static ssize_t __write_ports_addxprt(char *buf)
return err;
}
/*
* A transport listener is removed by writing a "-", it's transport
* name, and it's port number.
*/
static ssize_t __write_ports_delxprt(char *buf)
{
struct svc_xprt *xprt;
char transport[16];
int port;
if (sscanf(&buf[1], "%15s %4u", transport, &port) != 2)
return -EINVAL;
if (port < 1 || port > USHRT_MAX || nfsd_serv == NULL)
return -EINVAL;
xprt = svc_find_xprt(nfsd_serv, transport, &init_net, AF_UNSPEC, port);
if (xprt == NULL)
return -ENOTCONN;
svc_close_xprt(xprt);
svc_xprt_put(xprt);
return 0;
}
static ssize_t __write_ports(struct file *file, char *buf, size_t size)
{
if (size == 0)
......@@ -779,15 +735,9 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
if (isdigit(buf[0]))
return __write_ports_addfd(buf);
if (buf[0] == '-' && isdigit(buf[1]))
return __write_ports_delfd(buf);
if (isalpha(buf[0]))
return __write_ports_addxprt(buf);
if (buf[0] == '-' && isalpha(buf[1]))
return __write_ports_delxprt(buf);
return -EINVAL;
}
......@@ -825,21 +775,6 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
* OR
*
* Input:
* buf: C string containing a "-" followed
* by an integer value representing a
* previously passed in socket file
* descriptor
* size: non-zero length of C string in @buf
* Output:
* On success: NFS service no longer listens on that socket;
* passed-in buffer filled with a '\n'-terminated C
* string containing a unique name of the listener;
* return code is the size in bytes of the string
* On error: return code is a negative errno value
*
* OR
*
* Input:
* buf: C string containing a transport
* name and an unsigned integer value
* representing the port to listen on,
......@@ -848,19 +783,6 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
* Output:
* On success: returns zero; NFS service is started
* On error: return code is a negative errno value
*
* OR
*
* Input:
* buf: C string containing a "-" followed
* by a transport name and an unsigned
* integer value representing the port
* to listen on, separated by whitespace
* size: non-zero length of C string in @buf
* Output:
* On success: returns zero; NFS service no longer listens
* on that transport
* On error: return code is a negative errno value
*/
static ssize_t write_ports(struct file *file, char *buf, size_t size)
{
......@@ -1008,8 +930,6 @@ static ssize_t write_gracetime(struct file *file, char *buf, size_t size)
return nfsd4_write_time(file, buf, size, &nfsd4_grace);
}
extern char *nfs4_recoverydir(void);
static ssize_t __write_recoverydir(struct file *file, char *buf, size_t size)
{
char *mesg = buf;
......
......@@ -65,7 +65,7 @@ extern const struct seq_operations nfs_exports_op;
/*
* Function prototypes.
*/
int nfsd_svc(unsigned short port, int nrservs);
int nfsd_svc(int nrservs);
int nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp);
int nfsd_nrthreads(void);
......@@ -124,6 +124,7 @@ int nfs4_state_start(void);
void nfs4_state_shutdown(void);
void nfs4_reset_lease(time_t leasetime);
int nfs4_reset_recoverydir(char *recdir);
char * nfs4_recoverydir(void);
#else
static inline void nfs4_state_init(void) { }
static inline int nfsd4_init_slabs(void) { return 0; }
......@@ -132,6 +133,7 @@ static inline int nfs4_state_start(void) { return 0; }
static inline void nfs4_state_shutdown(void) { }
static inline void nfs4_reset_lease(time_t leasetime) { }
static inline int nfs4_reset_recoverydir(char *recdir) { return 0; }
static inline char * nfs4_recoverydir(void) {return NULL; }
#endif
/*
......
......@@ -183,18 +183,18 @@ int nfsd_nrthreads(void)
return rv;
}
static int nfsd_init_socks(int port)
static int nfsd_init_socks(void)
{
int error;
if (!list_empty(&nfsd_serv->sv_permsocks))
return 0;
error = svc_create_xprt(nfsd_serv, "udp", &init_net, PF_INET, port,
error = svc_create_xprt(nfsd_serv, "udp", &init_net, PF_INET, NFS_PORT,
SVC_SOCK_DEFAULTS);
if (error < 0)
return error;
error = svc_create_xprt(nfsd_serv, "tcp", &init_net, PF_INET, port,
error = svc_create_xprt(nfsd_serv, "tcp", &init_net, PF_INET, NFS_PORT,
SVC_SOCK_DEFAULTS);
if (error < 0)
return error;
......@@ -204,7 +204,7 @@ static int nfsd_init_socks(int port)
static bool nfsd_up = false;
static int nfsd_startup(unsigned short port, int nrservs)
static int nfsd_startup(int nrservs)
{
int ret;
......@@ -218,7 +218,7 @@ static int nfsd_startup(unsigned short port, int nrservs)
ret = nfsd_racache_init(2*nrservs);
if (ret)
return ret;
ret = nfsd_init_socks(port);
ret = nfsd_init_socks();
if (ret)
goto out_racache;
ret = lockd_up(&init_net);
......@@ -436,7 +436,7 @@ int nfsd_set_nrthreads(int n, int *nthreads)
* this is the first time nrservs is nonzero.
*/
int
nfsd_svc(unsigned short port, int nrservs)
nfsd_svc(int nrservs)
{
int error;
bool nfsd_up_before;
......@@ -458,7 +458,7 @@ nfsd_svc(unsigned short port, int nrservs)
nfsd_up_before = nfsd_up;
error = nfsd_startup(port, nrservs);
error = nfsd_startup(nrservs);
if (error)
goto out_destroy;
error = svc_set_num_threads(nfsd_serv, NULL, nrservs);
......@@ -487,7 +487,7 @@ static int
nfsd(void *vrqstp)
{
struct svc_rqst *rqstp = (struct svc_rqst *) vrqstp;
int err, preverr = 0;
int err;
/* Lock module and set up kernel thread */
mutex_lock(&nfsd_mutex);
......@@ -534,16 +534,6 @@ nfsd(void *vrqstp)
;
if (err == -EINTR)
break;
else if (err < 0) {
if (err != preverr) {
printk(KERN_WARNING "%s: unexpected error "
"from svc_recv (%d)\n", __func__, -err);
preverr = err;
}
schedule_timeout_uninterruptible(HZ);
continue;
}
validate_process_creds();
svc_process(rqstp);
validate_process_creds();
......
......@@ -373,11 +373,7 @@ static inline struct nfs4_lockowner * lockowner(struct nfs4_stateowner *so)
return container_of(so, struct nfs4_lockowner, lo_owner);
}
/*
* nfs4_file: a file opened by some number of (open) nfs4_stateowners.
* o fi_perfile list is used to search for conflicting
* share_acces, share_deny on the file.
*/
/* nfs4_file: a file opened by some number of (open) nfs4_stateowners. */
struct nfs4_file {
atomic_t fi_ref;
struct list_head fi_hash; /* hash by "struct inode *" */
......@@ -459,7 +455,7 @@ extern void nfs4_unlock_state(void);
extern int nfs4_in_grace(void);
extern void nfs4_release_reclaim(void);
extern struct nfs4_client_reclaim *nfsd4_find_reclaim_client(struct nfs4_client *crp);
extern __be32 nfs4_check_open_reclaim(clientid_t *clid);
extern __be32 nfs4_check_open_reclaim(clientid_t *clid, bool sessions);
extern void nfs4_free_openowner(struct nfs4_openowner *);
extern void nfs4_free_lockowner(struct nfs4_lockowner *);
extern int set_callback_cred(void);
......
......@@ -1581,7 +1581,7 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp)
*/
oldfs = get_fs(); set_fs(KERNEL_DS);
host_err = inode->i_op->readlink(path.dentry, buf, *lenp);
host_err = inode->i_op->readlink(path.dentry, (char __user *)buf, *lenp);
set_fs(oldfs);
if (host_err < 0)
......
header-y += cld.h
header-y += debug.h
header-y += export.h
header-y += nfsfh.h
header-y += stats.h
......@@ -5,44 +5,15 @@
*
* Copyright (C) 1995 Olaf Kirch <okir@monad.swb.de>
*/
#ifndef LINUX_NFSD_DEBUG_H
#define LINUX_NFSD_DEBUG_H
#include <linux/sunrpc/debug.h>
#include <uapi/linux/nfsd/debug.h>
/*
* Enable debugging for nfsd.
* Requires RPC_DEBUG.
*/
#ifdef RPC_DEBUG
# define NFSD_DEBUG 1
#endif
/*
* knfsd debug flags
*/
#define NFSDDBG_SOCK 0x0001
#define NFSDDBG_FH 0x0002
#define NFSDDBG_EXPORT 0x0004
#define NFSDDBG_SVC 0x0008
#define NFSDDBG_PROC 0x0010
#define NFSDDBG_FILEOP 0x0020
#define NFSDDBG_AUTH 0x0040
#define NFSDDBG_REPCACHE 0x0080
#define NFSDDBG_XDR 0x0100
#define NFSDDBG_LOCKD 0x0200
#define NFSDDBG_ALL 0x7FFF
#define NFSDDBG_NOCHANGE 0xFFFF
#ifdef __KERNEL__
# undef ifdebug
# ifdef NFSD_DEBUG
# define ifdebug(flag) if (nfsd_debug & NFSDDBG_##flag)
# else
# define ifdebug(flag) if (0)
# endif
#endif /* __KERNEL__ */
#endif /* LINUX_NFSD_DEBUG_H */
......@@ -6,58 +6,11 @@
*
* Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
*/
#ifndef NFSD_EXPORT_H
#define NFSD_EXPORT_H
# include <linux/types.h>
#ifdef __KERNEL__
# include <linux/nfsd/nfsfh.h>
#endif
/*
* Important limits for the exports stuff.
*/
#define NFSCLNT_IDMAX 1024
#define NFSCLNT_ADDRMAX 16
#define NFSCLNT_KEYMAX 32
/*
* Export flags.
*/
#define NFSEXP_READONLY 0x0001
#define NFSEXP_INSECURE_PORT 0x0002
#define NFSEXP_ROOTSQUASH 0x0004
#define NFSEXP_ALLSQUASH 0x0008
#define NFSEXP_ASYNC 0x0010
#define NFSEXP_GATHERED_WRITES 0x0020
/* 40 80 100 currently unused */
#define NFSEXP_NOHIDE 0x0200
#define NFSEXP_NOSUBTREECHECK 0x0400
#define NFSEXP_NOAUTHNLM 0x0800 /* Don't authenticate NLM requests - just trust */
#define NFSEXP_MSNFS 0x1000 /* do silly things that MS clients expect; no longer supported */
#define NFSEXP_FSID 0x2000
#define NFSEXP_CROSSMOUNT 0x4000
#define NFSEXP_NOACL 0x8000 /* reserved for possible ACL related use */
/*
* The NFSEXP_V4ROOT flag causes the kernel to give access only to NFSv4
* clients, and only to the single directory that is the root of the
* export; further lookup and readdir operations are treated as if every
* subdirectory was a mountpoint, and ignored if they are not themselves
* exported. This is used by nfsd and mountd to construct the NFSv4
* pseudofilesystem, which provides access only to paths leading to each
* exported filesystem.
*/
#define NFSEXP_V4ROOT 0x10000
/* All flags that we claim to support. (Note we don't support NOACL.) */
#define NFSEXP_ALLFLAGS 0x17E3F
/* The flags that may vary depending on security flavor: */
#define NFSEXP_SECINFO_FLAGS (NFSEXP_READONLY | NFSEXP_ROOTSQUASH \
| NFSEXP_ALLSQUASH \
| NFSEXP_INSECURE_PORT)
#ifdef __KERNEL__
#include <uapi/linux/nfsd/export.h>
/*
* FS Locations
......@@ -154,7 +107,4 @@ static inline void exp_get(struct svc_export *exp)
}
struct svc_export * rqst_exp_find(struct svc_rqst *, int, u32 *);
#endif /* __KERNEL__ */
#endif /* NFSD_EXPORT_H */
......@@ -10,117 +10,11 @@
*
* Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de>
*/
#ifndef _LINUX_NFSD_FH_H
#define _LINUX_NFSD_FH_H
#include <linux/types.h>
#include <linux/nfs.h>
#include <linux/nfs2.h>
#include <linux/nfs3.h>
#include <linux/nfs4.h>
#ifdef __KERNEL__
# include <linux/sunrpc/svc.h>
#endif
/*
* This is the old "dentry style" Linux NFSv2 file handle.
*
* The xino and xdev fields are currently used to transport the
* ino/dev of the exported inode.
*/
struct nfs_fhbase_old {
__u32 fb_