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
  ...

Documentation/filesystems/nfs/nfsd-admin-interfaces.txt

+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.

fs/lockd/svc.c

@@ -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)));
 

fs/locks.c

@@ -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

fs/nfs/callback.c

@@ -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;

fs/nfsd/nfs2acl.c

@@ -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);
 }

fs/nfsd/nfs3proc.c

@@ -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);
 }

fs/nfsd/nfs4callback.c

@@ -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;

fs/nfsd/nfs4idmap.c

@@ -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;
 }
 

fs/nfsd/nfs4proc.c

@@ -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[];

fs/nfsd/nfs4xdr.c

@@ -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();
 	}

fs/nfsd/nfsctl.c

@@ -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;

fs/nfsd/nfsd.h

@@ -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
 
 /*

fs/nfsd/nfssvc.c

@@ -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();

fs/nfsd/state.h

@@ -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);

fs/nfsd/vfs.c

@@ -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)

include/linux/nfsd/Kbuild

-header-y += cld.h
-header-y += debug.h
-header-y += export.h
-header-y += nfsfh.h
-header-y += stats.h

include/linux/nfsd/debug.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 */

include/linux/nfsd/export.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 */
-

include/linux/nfsd/nfsfh.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_dcookie;	/* dentry cookie - always 0xfeebbaca */
-	__u32		fb_ino;		/* our inode number */
-	__u32		fb_dirino;	/* dir inode number, 0 for directories */
-	__u32		fb_dev;		/* our device */
-	__u32		fb_xdev;
-	__u32		fb_xino;
-	__u32		fb_generation;
-};
-
-/*
- * This is the new flexible, extensible style NFSv2/v3 file handle.
- * by Neil Brown <neilb@cse.unsw.edu.au> - March 2000
- *
- * The file handle starts with a sequence of four-byte words.
- * The first word contains a version number (1) and three descriptor bytes
- * that tell how the remaining 3 variable length fields should be handled.
- * These three bytes are auth_type, fsid_type and fileid_type.
- *
- * All four-byte values are in host-byte-order.
- *
- * The auth_type field specifies how the filehandle can be authenticated
- * This might allow a file to be confirmed to be in a writable part of a
- * filetree without checking the path from it up to the root.
- * Current values:
- *     0  - No authentication.  fb_auth is 0 bytes long
- * Possible future values:
- *     1  - 4 bytes taken from MD5 hash of the remainer of the file handle
- *          prefixed by a secret and with the important export flags.
- *
- * The fsid_type identifies how the filesystem (or export point) is
- *    encoded.
- *  Current values:
- *     0  - 4 byte device id (ms-2-bytes major, ls-2-bytes minor), 4byte inode number
- *        NOTE: we cannot use the kdev_t device id value, because kdev_t.h
- *              says we mustn't.  We must break it up and reassemble.
- *     1  - 4 byte user specified identifier
- *     2  - 4 byte major, 4 byte minor, 4 byte inode number - DEPRECATED
- *     3  - 4 byte device id, encoded for user-space, 4 byte inode number
- *     4  - 4 byte inode number and 4 byte uuid
- *     5  - 8 byte uuid
- *     6  - 16 byte uuid
- *     7  - 8 byte inode number and 16 byte uuid
- *
- * The fileid_type identified how the file within the filesystem is encoded.
- * This is (will be) passed to, and set by, the underlying filesystem if it supports
- * filehandle operations.  The filesystem must not use the value '0' or '0xff' and may
- * only use the values 1 and 2 as defined below:
- *  Current values:
- *    0   - The root, or export point, of the filesystem.  fb_fileid is 0 bytes.
- *    1   - 32bit inode number, 32 bit generation number.
- *    2   - 32bit inode number, 32 bit generation number, 32 bit parent directory inode number.
- *
- */
-struct nfs_fhbase_new {
-	__u8		fb_version;	/* == 1, even => nfs_fhbase_old */
-	__u8		fb_auth_type;
-	__u8		fb_fsid_type;
-	__u8		fb_fileid_type;
-	__u32		fb_auth[1];
-/*	__u32		fb_fsid[0]; floating */
-/*	__u32		fb_fileid[0]; floating */
-};
-
-struct knfsd_fh {
-	unsigned int	fh_size;	/* significant for NFSv3.
-					 * Points to the current size while building
-					 * a new file handle
-					 */
-	union {
-		struct nfs_fhbase_old	fh_old;
-		__u32			fh_pad[NFS4_FHSIZE/4];
-		struct nfs_fhbase_new	fh_new;
-	} fh_base;
-};
-
-#define ofh_dcookie		fh_base.fh_old.fb_dcookie
-#define ofh_ino			fh_base.fh_old.fb_ino
-#define ofh_dirino		fh_base.fh_old.fb_dirino
-#define ofh_dev			fh_base.fh_old.fb_dev
-#define ofh_xdev		fh_base.fh_old.fb_xdev
-#define ofh_xino		fh_base.fh_old.fb_xino
-#define ofh_generation		fh_base.fh_old.fb_generation
-
-#define	fh_version		fh_base.fh_new.fb_version
-#define	fh_fsid_type		fh_base.fh_new.fb_fsid_type
-#define	fh_auth_type		fh_base.fh_new.fb_auth_type
-#define	fh_fileid_type		fh_base.fh_new.fb_fileid_type
-#define	fh_auth			fh_base.fh_new.fb_auth
-#define	fh_fsid			fh_base.fh_new.fb_auth
-
-#ifdef __KERNEL__
+#include <uapi/linux/nfsd/nfsfh.h>
 
 static inline __u32 ino_t_to_u32(ino_t ino)
 {
@@ -166,7 +60,4 @@ typedef struct svc_fh {
 
 } svc_fh;
 
-#endif /* __KERNEL__ */
-
-
 #endif /* _LINUX_NFSD_FH_H */

include/linux/nfsd/stats.h

@@ -5,16 +5,11 @@
  *
  * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
  */
-
 #ifndef LINUX_NFSD_STATS_H
 #define LINUX_NFSD_STATS_H
 
-#include <linux/nfs4.h>
-
-/* thread usage wraps very million seconds (approx one fortnight) */
-#define	NFSD_USAGE_WRAP	(HZ*1000000)
+#include <uapi/linux/nfsd/stats.h>
 
-#ifdef __KERNEL__
 
 struct nfsd_stats {
 	unsigned int	rchits;		/* repcache hits */
@@ -47,5 +42,4 @@ extern struct svc_stat		nfsd_svcstats;
 void	nfsd_stat_init(void);
 void	nfsd_stat_shutdown(void);
 
-#endif /* __KERNEL__ */
 #endif /* LINUX_NFSD_STATS_H */

include/linux/sunrpc/Kbuild

-header-y += debug.h

include/linux/sunrpc/debug.h

@@ -5,28 +5,11 @@
  *
  * Copyright (C) 1996, Olaf Kirch <okir@monad.swb.de>
  */
-
 #ifndef _LINUX_SUNRPC_DEBUG_H_
 #define _LINUX_SUNRPC_DEBUG_H_
 
-/*
- * RPC debug facilities
- */
-#define RPCDBG_XPRT		0x0001
-#define RPCDBG_CALL		0x0002
-#define RPCDBG_DEBUG		0x0004
-#define RPCDBG_NFS		0x0008
-#define RPCDBG_AUTH		0x0010
-#define RPCDBG_BIND		0x0020
-#define RPCDBG_SCHED		0x0040
-#define RPCDBG_TRANS		0x0080
-#define RPCDBG_SVCXPRT		0x0100
-#define RPCDBG_SVCDSP		0x0200
-#define RPCDBG_MISC		0x0400
-#define RPCDBG_CACHE		0x0800
-#define RPCDBG_ALL		0x7fff
+#include <uapi/linux/sunrpc/debug.h>
 
-#ifdef __KERNEL__
 
 /*
  * Enable RPC debugging/profiling.
@@ -87,24 +70,4 @@ void		rpc_register_sysctl(void);
 void		rpc_unregister_sysctl(void);
 #endif
 
-#endif /* __KERNEL__ */
-
-/*
- * Declarations for the sysctl debug interface, which allows to read or
- * change the debug flags for rpc, nfs, nfsd, and lockd. Since the sunrpc
- * module currently registers its sysctl table dynamically, the sysctl path
- * for module FOO is <CTL_SUNRPC, CTL_FOODEBUG>.
- */
-
-enum {
-	CTL_RPCDEBUG = 1,
-	CTL_NFSDEBUG,
-	CTL_NFSDDEBUG,
-	CTL_NLMDEBUG,
-	CTL_SLOTTABLE_UDP,
-	CTL_SLOTTABLE_TCP,
-	CTL_MIN_RESVPORT,
-	CTL_MAX_RESVPORT,
-};
-
 #endif /* _LINUX_SUNRPC_DEBUG_H_ */

include/linux/sunrpc/svc_xprt.h

@@ -114,7 +114,6 @@ void	svc_xprt_init(struct net *, struct svc_xprt_class *, struct svc_xprt *,
 int	svc_create_xprt(struct svc_serv *, const char *, struct net *,
 			const int, const unsigned short, int);
 void	svc_xprt_enqueue(struct svc_xprt *xprt);
-void	svc_xprt_received(struct svc_xprt *);
 void	svc_xprt_put(struct svc_xprt *xprt);
 void	svc_xprt_copy_addrs(struct svc_rqst *rqstp, struct svc_xprt *xprt);
 void	svc_close_xprt(struct svc_xprt *xprt);
@@ -124,6 +123,7 @@ struct	svc_xprt *svc_find_xprt(struct svc_serv *serv, const char *xcl_name,
 			struct net *net, const sa_family_t af,
 			const unsigned short port);
 int	svc_xprt_names(struct svc_serv *serv, char *buf, const int buflen);
+void	svc_add_new_perm_xprt(struct svc_serv *serv, struct svc_xprt *xprt);
 
 static inline void svc_xprt_get(struct svc_xprt *xprt)
 {
@@ -166,8 +166,7 @@ static inline size_t svc_addr_len(const struct sockaddr *sa)
 	case AF_INET6:
 		return sizeof(struct sockaddr_in6);
 	}
-
-	return 0;
+	BUG();
 }
 
 static inline unsigned short svc_xprt_local_port(const struct svc_xprt *xprt)

include/linux/sunrpc/svcsock.h

@@ -39,9 +39,6 @@ int		svc_recv(struct svc_rqst *, long);
 int		svc_send(struct svc_rqst *);
 void		svc_drop(struct svc_rqst *);
 void		svc_sock_update_bufs(struct svc_serv *serv);
-int		svc_sock_names(struct svc_serv *serv, char *buf,
-					const size_t buflen,
-					const char *toclose);
 int		svc_addsock(struct svc_serv *serv, const int fd,
 					char *name_return, const size_t len);
 void		svc_init_xprt_sock(void);

include/uapi/linux/nfsd/Kbuild

 # UAPI Header export list
+header-y += cld.h
+header-y += debug.h
+header-y += export.h
+header-y += nfsfh.h
+header-y += stats.h

include/uapi/linux/nfsd/debug.h

+/*
+ * linux/include/linux/nfsd/debug.h
+ *
+ * Debugging-related stuff for nfsd
+ *
+ * Copyright (C) 1995 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _UAPILINUX_NFSD_DEBUG_H
+#define _UAPILINUX_NFSD_DEBUG_H
+
+#include <linux/sunrpc/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
+
+
+
+#endif /* _UAPILINUX_NFSD_DEBUG_H */

include/uapi/linux/nfsd/export.h

+/*
+ * include/linux/nfsd/export.h
+ * 
+ * Public declarations for NFS exports. The definitions for the
+ * syscall interface are in nfsctl.h
+ *
+ * Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _UAPINFSD_EXPORT_H
+#define _UAPINFSD_EXPORT_H
+
+# include <linux/types.h>
+
+/*
+ * 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)
+
+
+#endif /* _UAPINFSD_EXPORT_H */

include/uapi/linux/nfsd/nfsfh.h

+/*
+ * include/linux/nfsd/nfsfh.h
+ *
+ * This file describes the layout of the file handles as passed
+ * over the wire.
+ *
+ * Earlier versions of knfsd used to sign file handles using keyed MD5
+ * or SHA. I've removed this code, because it doesn't give you more
+ * security than blocking external access to port 2049 on your firewall.
+ *
+ * Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _UAPI_LINUX_NFSD_FH_H
+#define _UAPI_LINUX_NFSD_FH_H
+
+#include <linux/types.h>
+#include <linux/nfs.h>
+#include <linux/nfs2.h>
+#include <linux/nfs3.h>
+#include <linux/nfs4.h>
+
+/*
+ * 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_dcookie;	/* dentry cookie - always 0xfeebbaca */
+	__u32		fb_ino;		/* our inode number */
+	__u32		fb_dirino;	/* dir inode number, 0 for directories */
+	__u32		fb_dev;		/* our device */
+	__u32		fb_xdev;
+	__u32		fb_xino;
+	__u32		fb_generation;
+};
+
+/*
+ * This is the new flexible, extensible style NFSv2/v3 file handle.
+ * by Neil Brown <neilb@cse.unsw.edu.au> - March 2000
+ *
+ * The file handle starts with a sequence of four-byte words.
+ * The first word contains a version number (1) and three descriptor bytes
+ * that tell how the remaining 3 variable length fields should be handled.
+ * These three bytes are auth_type, fsid_type and fileid_type.
+ *
+ * All four-byte values are in host-byte-order.
+ *
+ * The auth_type field specifies how the filehandle can be authenticated
+ * This might allow a file to be confirmed to be in a writable part of a
+ * filetree without checking the path from it up to the root.
+ * Current values:
+ *     0  - No authentication.  fb_auth is 0 bytes long
+ * Possible future values:
+ *     1  - 4 bytes taken from MD5 hash of the remainer of the file handle
+ *          prefixed by a secret and with the important export flags.
+ *
+ * The fsid_type identifies how the filesystem (or export point) is
+ *    encoded.
+ *  Current values:
+ *     0  - 4 byte device id (ms-2-bytes major, ls-2-bytes minor), 4byte inode number
+ *        NOTE: we cannot use the kdev_t device id value, because kdev_t.h
+ *              says we mustn't.  We must break it up and reassemble.
+ *     1  - 4 byte user specified identifier
+ *     2  - 4 byte major, 4 byte minor, 4 byte inode number - DEPRECATED
+ *     3  - 4 byte device id, encoded for user-space, 4 byte inode number
+ *     4  - 4 byte inode number and 4 byte uuid
+ *     5  - 8 byte uuid
+ *     6  - 16 byte uuid
+ *     7  - 8 byte inode number and 16 byte uuid
+ *
+ * The fileid_type identified how the file within the filesystem is encoded.
+ * This is (will be) passed to, and set by, the underlying filesystem if it supports
+ * filehandle operations.  The filesystem must not use the value '0' or '0xff' and may
+ * only use the values 1 and 2 as defined below:
+ *  Current values:
+ *    0   - The root, or export point, of the filesystem.  fb_fileid is 0 bytes.
+ *    1   - 32bit inode number, 32 bit generation number.
+ *    2   - 32bit inode number, 32 bit generation number, 32 bit parent directory inode number.
+ *
+ */
+struct nfs_fhbase_new {
+	__u8		fb_version;	/* == 1, even => nfs_fhbase_old */
+	__u8		fb_auth_type;
+	__u8		fb_fsid_type;
+	__u8		fb_fileid_type;
+	__u32		fb_auth[1];
+/*	__u32		fb_fsid[0]; floating */
+/*	__u32		fb_fileid[0]; floating */
+};
+
+struct knfsd_fh {
+	unsigned int	fh_size;	/* significant for NFSv3.
+					 * Points to the current size while building
+					 * a new file handle
+					 */
+	union {
+		struct nfs_fhbase_old	fh_old;
+		__u32			fh_pad[NFS4_FHSIZE/4];
+		struct nfs_fhbase_new	fh_new;
+	} fh_base;
+};
+
+#define ofh_dcookie		fh_base.fh_old.fb_dcookie
+#define ofh_ino			fh_base.fh_old.fb_ino
+#define ofh_dirino		fh_base.fh_old.fb_dirino
+#define ofh_dev			fh_base.fh_old.fb_dev
+#define ofh_xdev		fh_base.fh_old.fb_xdev
+#define ofh_xino		fh_base.fh_old.fb_xino
+#define ofh_generation		fh_base.fh_old.fb_generation
+
+#define	fh_version		fh_base.fh_new.fb_version
+#define	fh_fsid_type		fh_base.fh_new.fb_fsid_type
+#define	fh_auth_type		fh_base.fh_new.fb_auth_type
+#define	fh_fileid_type		fh_base.fh_new.fb_fileid_type
+#define	fh_auth			fh_base.fh_new.fb_auth
+#define	fh_fsid			fh_base.fh_new.fb_auth
+
+
+
+#endif /* _UAPI_LINUX_NFSD_FH_H */

include/uapi/linux/nfsd/stats.h

+/*
+ * linux/include/linux/nfsd/stats.h
+ *
+ * Statistics for NFS server.
+ *
+ * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _UAPILINUX_NFSD_STATS_H
+#define _UAPILINUX_NFSD_STATS_H
+
+#include <linux/nfs4.h>
+
+/* thread usage wraps very million seconds (approx one fortnight) */
+#define	NFSD_USAGE_WRAP	(HZ*1000000)
+
+#endif /* _UAPILINUX_NFSD_STATS_H */

include/uapi/linux/sunrpc/Kbuild

 # UAPI Header export list
+header-y += debug.h

include/uapi/linux/sunrpc/debug.h

+/*
+ * linux/include/linux/sunrpc/debug.h
+ *
+ * Debugging support for sunrpc module
+ *
+ * Copyright (C) 1996, Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _UAPI_LINUX_SUNRPC_DEBUG_H_
+#define _UAPI_LINUX_SUNRPC_DEBUG_H_
+
+/*
+ * RPC debug facilities
+ */
+#define RPCDBG_XPRT		0x0001
+#define RPCDBG_CALL		0x0002
+#define RPCDBG_DEBUG		0x0004
+#define RPCDBG_NFS		0x0008
+#define RPCDBG_AUTH		0x0010
+#define RPCDBG_BIND		0x0020
+#define RPCDBG_SCHED		0x0040
+#define RPCDBG_TRANS		0x0080
+#define RPCDBG_SVCXPRT		0x0100
+#define RPCDBG_SVCDSP		0x0200
+#define RPCDBG_MISC		0x0400
+#define RPCDBG_CACHE		0x0800
+#define RPCDBG_ALL		0x7fff
+
+
+/*
+ * Declarations for the sysctl debug interface, which allows to read or
+ * change the debug flags for rpc, nfs, nfsd, and lockd. Since the sunrpc
+ * module currently registers its sysctl table dynamically, the sysctl path
+ * for module FOO is <CTL_SUNRPC, CTL_FOODEBUG>.
+ */
+
+enum {
+	CTL_RPCDEBUG = 1,
+	CTL_NFSDEBUG,
+	CTL_NFSDDEBUG,
+	CTL_NLMDEBUG,
+	CTL_SLOTTABLE_UDP,
+	CTL_SLOTTABLE_TCP,
+	CTL_MIN_RESVPORT,
+	CTL_MAX_RESVPORT,
+};
+
+#endif /* _UAPI_LINUX_SUNRPC_DEBUG_H_ */

net/sunrpc/svcsock.c

@@ -59,7 +59,7 @@
 
 
 static struct svc_sock *svc_setup_socket(struct svc_serv *, struct socket *,
-					 int *errp, int flags);
+					 int flags);
 static void		svc_udp_data_ready(struct sock *, int);
 static int		svc_udp_recvfrom(struct svc_rqst *);
 static int		svc_udp_sendto(struct svc_rqst *);
@@ -305,57 +305,6 @@ static int svc_one_sock_name(struct svc_sock *svsk, char *buf, int remaining)
 	return len;
 }
 
-/**
- * svc_sock_names - construct a list of listener names in a string
- * @serv: pointer to RPC service
- * @buf: pointer to a buffer to fill in with socket names
- * @buflen: size of the buffer to be filled
- * @toclose: pointer to '\0'-terminated C string containing the name
- *		of a listener to be closed
- *
- * Fills in @buf with a '\n'-separated list of names of listener
- * sockets.  If @toclose is not NULL, the socket named by @toclose
- * is closed, and is not included in the output list.
- *
- * Returns positive length of the socket name string, or a negative
- * errno value on error.
- */
-int svc_sock_names(struct svc_serv *serv, char *buf, const size_t buflen,
-		   const char *toclose)
-{
-	struct svc_sock *svsk, *closesk = NULL;
-	int len = 0;
-
-	if (!serv)
-		return 0;
-
-	spin_lock_bh(&serv->sv_lock);
-	list_for_each_entry(svsk, &serv->sv_permsocks, sk_xprt.xpt_list) {
-		int onelen = svc_one_sock_name(svsk, buf + len, buflen - len);
-		if (onelen < 0) {
-			len = onelen;
-			break;
-		}
-		if (toclose && strcmp(toclose, buf + len) == 0) {
-			closesk = svsk;
-			svc_xprt_get(&closesk->sk_xprt);
-		} else
-			len += onelen;
-	}
-	spin_unlock_bh(&serv->sv_lock);
-
-	if (closesk) {
-		/* Should unregister with portmap, but you cannot
-		 * unregister just one protocol...
-		 */
-		svc_close_xprt(&closesk->sk_xprt);
-		svc_xprt_put(&closesk->sk_xprt);
-	} else if (toclose)
-		return -ENOENT;
-	return len;
-}
-EXPORT_SYMBOL_GPL(svc_sock_names);
-
 /*
  * Check input queue length
  */
@@ -598,11 +547,9 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 			dprintk("svc: recvfrom returned error %d\n", -err);
 			set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 		}
-		return -EAGAIN;
+		return 0;
 	}
 	len = svc_addr_len(svc_addr(rqstp));
-	if (len == 0)
-		return -EAFNOSUPPORT;
 	rqstp->rq_addrlen = len;
 	if (skb->tstamp.tv64 == 0) {
 		skb->tstamp = ktime_get_real();
@@ -620,10 +567,7 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 	if (!svc_udp_get_dest_address(rqstp, cmh)) {
 		net_warn_ratelimited("svc: received unknown control message %d/%d; dropping RPC reply datagram\n",
 				     cmh->cmsg_level, cmh->cmsg_type);
-out_free:
-		trace_kfree_skb(skb, svc_udp_recvfrom);
-		skb_free_datagram_locked(svsk->sk_sk, skb);
-		return 0;
+		goto out_free;
 	}
 	rqstp->rq_daddrlen = svc_addr_len(svc_daddr(rqstp));
 
@@ -662,6 +606,10 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 		serv->sv_stats->netudpcnt++;
 
 	return len;
+out_free:
+	trace_kfree_skb(skb, svc_udp_recvfrom);
+	skb_free_datagram_locked(svsk->sk_sk, skb);
+	return 0;
 }
 
 static int
@@ -900,8 +848,9 @@ static struct svc_xprt *svc_tcp_accept(struct svc_xprt *xprt)
 	 */
 	newsock->sk->sk_sndtimeo = HZ*30;
 
-	if (!(newsvsk = svc_setup_socket(serv, newsock, &err,
-				 (SVC_SOCK_ANONYMOUS | SVC_SOCK_TEMPORARY))))
+	newsvsk = svc_setup_socket(serv, newsock,
+				 (SVC_SOCK_ANONYMOUS | SVC_SOCK_TEMPORARY));
+	if (IS_ERR(newsvsk))
 		goto failed;
 	svc_xprt_set_remote(&newsvsk->sk_xprt, sin, slen);
 	err = kernel_getsockname(newsock, sin, &slen);
@@ -1174,13 +1123,13 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
 	if (len != -EAGAIN)
 		goto err_other;
 	dprintk("RPC: TCP recvfrom got EAGAIN\n");
-	return -EAGAIN;
+	return 0;
 err_other:
 	printk(KERN_NOTICE "%s: recvfrom returned errno %d\n",
 	       svsk->sk_xprt.xpt_server->sv_name, -len);
 	set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
 err_noclose:
-	return -EAGAIN;	/* record not complete */
+	return 0;	/* record not complete */
 }
 
 /*
@@ -1383,29 +1332,29 @@ EXPORT_SYMBOL_GPL(svc_sock_update_bufs);
  */
 static struct svc_sock *svc_setup_socket(struct svc_serv *serv,
 						struct socket *sock,
-						int *errp, int flags)
+						int flags)
 {
 	struct svc_sock	*svsk;
 	struct sock	*inet;
 	int		pmap_register = !(flags & SVC_SOCK_ANONYMOUS);
+	int		err = 0;
 
 	dprintk("svc: svc_setup_socket %p\n", sock);
-	if (!(svsk = kzalloc(sizeof(*svsk), GFP_KERNEL))) {
-		*errp = -ENOMEM;
-		return NULL;
-	}
+	svsk = kzalloc(sizeof(*svsk), GFP_KERNEL);
+	if (!svsk)
+		return ERR_PTR(-ENOMEM);
 
 	inet = sock->sk;
 
 	/* Register socket with portmapper */
-	if (*errp >= 0 && pmap_register)
-		*errp = svc_register(serv, sock_net(sock->sk), inet->sk_family,
+	if (pmap_register)
+		err = svc_register(serv, sock_net(sock->sk), inet->sk_family,
 				     inet->sk_protocol,
 				     ntohs(inet_sk(inet)->inet_sport));
 
-	if (*errp < 0) {
+	if (err < 0) {
 		kfree(svsk);
-		return NULL;
+		return ERR_PTR(err);
 	}
 
 	inet->sk_user_data = svsk;
@@ -1450,42 +1399,38 @@ int svc_addsock(struct svc_serv *serv, const int fd, char *name_return,
 	int err = 0;
 	struct socket *so = sockfd_lookup(fd, &err);
 	struct svc_sock *svsk = NULL;
+	struct sockaddr_storage addr;
+	struct sockaddr *sin = (struct sockaddr *)&addr;
+	int salen;
 
 	if (!so)
 		return err;
+	err = -EAFNOSUPPORT;
 	if ((so->sk->sk_family != PF_INET) && (so->sk->sk_family != PF_INET6))
-		err =  -EAFNOSUPPORT;
-	else if (so->sk->sk_protocol != IPPROTO_TCP &&
+		goto out;
+	err =  -EPROTONOSUPPORT;
+	if (so->sk->sk_protocol != IPPROTO_TCP &&
 	    so->sk->sk_protocol != IPPROTO_UDP)
-		err =  -EPROTONOSUPPORT;
-	else if (so->state > SS_UNCONNECTED)
-		err = -EISCONN;
-	else {
-		if (!try_module_get(THIS_MODULE))
-			err = -ENOENT;
-		else
-			svsk = svc_setup_socket(serv, so, &err,
-						SVC_SOCK_DEFAULTS);
-		if (svsk) {
-			struct sockaddr_storage addr;
-			struct sockaddr *sin = (struct sockaddr *)&addr;
-			int salen;
-			if (kernel_getsockname(svsk->sk_sock, sin, &salen) == 0)
-				svc_xprt_set_local(&svsk->sk_xprt, sin, salen);
-			clear_bit(XPT_TEMP, &svsk->sk_xprt.xpt_flags);
-			spin_lock_bh(&serv->sv_lock);
-			list_add(&svsk->sk_xprt.xpt_list, &serv->sv_permsocks);
-			spin_unlock_bh(&serv->sv_lock);
-			svc_xprt_received(&svsk->sk_xprt);
-			err = 0;
-		} else
-			module_put(THIS_MODULE);
-	}
-	if (err) {
-		sockfd_put(so);
-		return err;
+		goto out;
+	err = -EISCONN;
+	if (so->state > SS_UNCONNECTED)
+		goto out;
+	err = -ENOENT;
+	if (!try_module_get(THIS_MODULE))
+		goto out;
+	svsk = svc_setup_socket(serv, so, SVC_SOCK_DEFAULTS);
+	if (IS_ERR(svsk)) {
+		module_put(THIS_MODULE);
+		err = PTR_ERR(svsk);
+		goto out;
 	}
+	if (kernel_getsockname(svsk->sk_sock, sin, &salen) == 0)
+		svc_xprt_set_local(&svsk->sk_xprt, sin, salen);
+	svc_add_new_perm_xprt(serv, &svsk->sk_xprt);
 	return svc_one_sock_name(svsk, name_return, len);
+out:
+	sockfd_put(so);
+	return err;
 }
 EXPORT_SYMBOL_GPL(svc_addsock);
 
@@ -1563,11 +1508,13 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv,
 			goto bummer;
 	}
 
-	if ((svsk = svc_setup_socket(serv, sock, &error, flags)) != NULL) {
-		svc_xprt_set_local(&svsk->sk_xprt, newsin, newlen);
-		return (struct svc_xprt *)svsk;
+	svsk = svc_setup_socket(serv, sock, flags);
+	if (IS_ERR(svsk)) {
+		error = PTR_ERR(svsk);
+		goto bummer;
 	}
-
+	svc_xprt_set_local(&svsk->sk_xprt, newsin, newlen);
+	return (struct svc_xprt *)svsk;
 bummer:
 	dprintk("svc: svc_create_socket error = %d\n", -error);
 	sock_release(sock);

net/sunrpc/xprtrdma/svc_rdma_transport.c

@@ -578,10 +578,6 @@ static void handle_connect_req(struct rdma_cm_id *new_cma_id, size_t client_ird)
 	list_add_tail(&newxprt->sc_accept_q, &listen_xprt->sc_accept_q);
 	spin_unlock_bh(&listen_xprt->sc_lock);
 
-	/*
-	 * Can't use svc_xprt_received here because we are not on a
-	 * rqstp thread
-	*/
 	set_bit(XPT_CONN, &listen_xprt->sc_xprt.xpt_flags);
 	svc_xprt_enqueue(&listen_xprt->sc_xprt);
 }