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

* 'for-linus' of git://linux-nfs.org/~bfields/linux: (100 commits)
  SUNRPC: RPC program information is stored in unsigned integers
  SUNRPC: Move exported symbol definitions after function declaration part 2
  NLM: tear down RPC clients in nlm_shutdown_hosts
  SUNRPC: spin svc_rqst initialization to its own function
  nfsd: more careful input validation in nfsctl write methods
  lockd: minor log message fix
  knfsd: don't bother mapping putrootfh enoent to eperm
  rdma: makefile
  rdma: ONCRPC RDMA protocol marshalling
  rdma: SVCRDMA sendto
  rdma: SVCRDMA recvfrom
  rdma: SVCRDMA Core Transport Services
  rdma: SVCRDMA Transport Module
  rdma: SVCRMDA Header File
  svc: Add svc_xprt_names service to replace svc_sock_names
  knfsd: Support adding transports by writing portlist file
  svc: Add svc API that queries for a transport instance
  svc: Add /proc/sys/sunrpc/transport files
  svc: Add transport hdr size for defer/revisit
  svc: Move the xprt independent code to the svc_xprt.c file
  ...

include/linux/sunrpc/svcsock.h

@@ -10,42 +10,16 @@
 #define SUNRPC_SVCSOCK_H
 
 #include <linux/sunrpc/svc.h>
+#include <linux/sunrpc/svc_xprt.h>
 
 /*
  * RPC server socket.
  */
 struct svc_sock {
-	struct list_head	sk_ready;	/* list of ready sockets */
-	struct list_head	sk_list;	/* list of all sockets */
+	struct svc_xprt		sk_xprt;
 	struct socket *		sk_sock;	/* berkeley socket layer */
 	struct sock *		sk_sk;		/* INET layer */
 
-	struct svc_pool *	sk_pool;	/* current pool iff queued */
-	struct svc_serv *	sk_server;	/* service for this socket */
-	atomic_t		sk_inuse;	/* use count */
-	unsigned long		sk_flags;
-#define	SK_BUSY		0			/* enqueued/receiving */
-#define	SK_CONN		1			/* conn pending */
-#define	SK_CLOSE	2			/* dead or dying */
-#define	SK_DATA		3			/* data pending */
-#define	SK_TEMP		4			/* temp (TCP) socket */
-#define	SK_DEAD		6			/* socket closed */
-#define	SK_CHNGBUF	7			/* need to change snd/rcv buffer sizes */
-#define	SK_DEFERRED	8			/* request on sk_deferred */
-#define	SK_OLD		9			/* used for temp socket aging mark+sweep */
-#define	SK_DETACHED	10			/* detached from tempsocks list */
-
-	atomic_t    	    	sk_reserved;	/* space on outq that is reserved */
-
-	spinlock_t		sk_lock;	/* protects sk_deferred and
-						 * sk_info_authunix */
-	struct list_head	sk_deferred;	/* deferred requests that need to
-						 * be revisted */
-	struct mutex		sk_mutex;	/* to serialize sending data */
-
-	int			(*sk_recvfrom)(struct svc_rqst *rqstp);
-	int			(*sk_sendto)(struct svc_rqst *rqstp);
-
 	/* We keep the old state_change and data_ready CB's here */
 	void			(*sk_ostate)(struct sock *);
 	void			(*sk_odata)(struct sock *, int bytes);
@@ -54,21 +28,12 @@ struct svc_sock {
 	/* private TCP part */
 	int			sk_reclen;	/* length of record */
 	int			sk_tcplen;	/* current read length */
-	time_t			sk_lastrecv;	/* time of last received request */
-
-	/* cache of various info for TCP sockets */
-	void			*sk_info_authunix;
-
-	struct sockaddr_storage	sk_local;	/* local address */
-	struct sockaddr_storage	sk_remote;	/* remote peer's address */
-	int			sk_remotelen;	/* length of address */
 };
 
 /*
  * Function prototypes.
  */
-int		svc_makesock(struct svc_serv *, int, unsigned short, int flags);
-void		svc_force_close_socket(struct svc_sock *);
+void		svc_close_all(struct list_head *);
 int		svc_recv(struct svc_rqst *, long);
 int		svc_send(struct svc_rqst *);
 void		svc_drop(struct svc_rqst *);
@@ -78,6 +43,8 @@ int		svc_addsock(struct svc_serv *serv,
 			    int fd,
 			    char *name_return,
 			    int *proto);
+void		svc_init_xprt_sock(void);
+void		svc_cleanup_xprt_sock(void);
 
 /*
  * svc_makesock socket characteristics

include/linux/sunrpc/xdr.h

@@ -112,7 +112,8 @@ struct xdr_buf {
 __be32 *xdr_encode_opaque_fixed(__be32 *p, const void *ptr, unsigned int len);
 __be32 *xdr_encode_opaque(__be32 *p, const void *ptr, unsigned int len);
 __be32 *xdr_encode_string(__be32 *p, const char *s);
-__be32 *xdr_decode_string_inplace(__be32 *p, char **sp, int *lenp, int maxlen);
+__be32 *xdr_decode_string_inplace(__be32 *p, char **sp, unsigned int *lenp,
+			unsigned int maxlen);
 __be32 *xdr_encode_netobj(__be32 *p, const struct xdr_netobj *);
 __be32 *xdr_decode_netobj(__be32 *p, struct xdr_netobj *);
 

net/sunrpc/Makefile

@@ -11,6 +11,7 @@ sunrpc-y := clnt.o xprt.o socklib.o xprtsock.o sched.o \
 	    auth.o auth_null.o auth_unix.o \
 	    svc.o svcsock.o svcauth.o svcauth_unix.o \
 	    rpcb_clnt.o timer.o xdr.o \
-	    sunrpc_syms.o cache.o rpc_pipe.o
+	    sunrpc_syms.o cache.o rpc_pipe.o \
+	    svc_xprt.o
 sunrpc-$(CONFIG_PROC_FS) += stats.o
 sunrpc-$(CONFIG_SYSCTL) += sysctl.o

net/sunrpc/auth_gss/svcauth_gss.c

@@ -224,38 +224,34 @@ static int rsi_parse(struct cache_detail *cd,
 
 	/* major/minor */
 	len = qword_get(&mesg, buf, mlen);
-	if (len < 0)
+	if (len <= 0)
 		goto out;
-	if (len == 0) {
+	rsii.major_status = simple_strtoul(buf, &ep, 10);
+	if (*ep)
+		goto out;
+	len = qword_get(&mesg, buf, mlen);
+	if (len <= 0)
+		goto out;
+	rsii.minor_status = simple_strtoul(buf, &ep, 10);
+	if (*ep)
 		goto out;
-	} else {
-		rsii.major_status = simple_strtoul(buf, &ep, 10);
-		if (*ep)
-			goto out;
-		len = qword_get(&mesg, buf, mlen);
-		if (len <= 0)
-			goto out;
-		rsii.minor_status = simple_strtoul(buf, &ep, 10);
-		if (*ep)
-			goto out;
 
-		/* out_handle */
-		len = qword_get(&mesg, buf, mlen);
-		if (len < 0)
-			goto out;
-		status = -ENOMEM;
-		if (dup_to_netobj(&rsii.out_handle, buf, len))
-			goto out;
+	/* out_handle */
+	len = qword_get(&mesg, buf, mlen);
+	if (len < 0)
+		goto out;
+	status = -ENOMEM;
+	if (dup_to_netobj(&rsii.out_handle, buf, len))
+		goto out;
 
-		/* out_token */
-		len = qword_get(&mesg, buf, mlen);
-		status = -EINVAL;
-		if (len < 0)
-			goto out;
-		status = -ENOMEM;
-		if (dup_to_netobj(&rsii.out_token, buf, len))
-			goto out;
-	}
+	/* out_token */
+	len = qword_get(&mesg, buf, mlen);
+	status = -EINVAL;
+	if (len < 0)
+		goto out;
+	status = -ENOMEM;
+	if (dup_to_netobj(&rsii.out_token, buf, len))
+		goto out;
 	rsii.h.expiry_time = expiry;
 	rsip = rsi_update(&rsii, rsip);
 	status = 0;
@@ -975,6 +971,7 @@ static int svcauth_gss_handle_init(struct svc_rqst *rqstp,
 	struct kvec *resv = &rqstp->rq_res.head[0];
 	struct xdr_netobj tmpobj;
 	struct rsi *rsip, rsikey;
+	int ret;
 
 	/* Read the verifier; should be NULL: */
 	*authp = rpc_autherr_badverf;
@@ -1014,23 +1011,27 @@ static int svcauth_gss_handle_init(struct svc_rqst *rqstp,
 		/* No upcall result: */
 		return SVC_DROP;
 	case 0:
+		ret = SVC_DROP;
 		/* Got an answer to the upcall; use it: */
 		if (gss_write_init_verf(rqstp, rsip))
-			return SVC_DROP;
+			goto out;
 		if (resv->iov_len + 4 > PAGE_SIZE)
-			return SVC_DROP;
+			goto out;
 		svc_putnl(resv, RPC_SUCCESS);
 		if (svc_safe_putnetobj(resv, &rsip->out_handle))
-			return SVC_DROP;
+			goto out;
 		if (resv->iov_len + 3 * 4 > PAGE_SIZE)
-			return SVC_DROP;
+			goto out;
 		svc_putnl(resv, rsip->major_status);
 		svc_putnl(resv, rsip->minor_status);
 		svc_putnl(resv, GSS_SEQ_WIN);
 		if (svc_safe_putnetobj(resv, &rsip->out_token))
-			return SVC_DROP;
+			goto out;
 	}
-	return SVC_COMPLETE;
+	ret = SVC_COMPLETE;
+out:
+	cache_put(&rsip->h, &rsi_cache);
+	return ret;
 }
 
 /*
@@ -1125,6 +1126,7 @@ svcauth_gss_accept(struct svc_rqst *rqstp, __be32 *authp)
 	case RPC_GSS_PROC_DESTROY:
 		if (gss_write_verf(rqstp, rsci->mechctx, gc->gc_seq))
 			goto auth_err;
+		rsci->h.expiry_time = get_seconds();
 		set_bit(CACHE_NEGATIVE, &rsci->h.flags);
 		if (resv->iov_len + 4 > PAGE_SIZE)
 			goto drop;
@@ -1386,19 +1388,26 @@ int
 gss_svc_init(void)
 {
 	int rv = svc_auth_register(RPC_AUTH_GSS, &svcauthops_gss);
-	if (rv == 0) {
-		cache_register(&rsc_cache);
-		cache_register(&rsi_cache);
-	}
+	if (rv)
+		return rv;
+	rv = cache_register(&rsc_cache);
+	if (rv)
+		goto out1;
+	rv = cache_register(&rsi_cache);
+	if (rv)
+		goto out2;
+	return 0;
+out2:
+	cache_unregister(&rsc_cache);
+out1:
+	svc_auth_unregister(RPC_AUTH_GSS);
 	return rv;
 }
 
 void
 gss_svc_shutdown(void)
 {
-	if (cache_unregister(&rsc_cache))
-		printk(KERN_ERR "auth_rpcgss: failed to unregister rsc cache\n");
-	if (cache_unregister(&rsi_cache))
-		printk(KERN_ERR "auth_rpcgss: failed to unregister rsi cache\n");
+	cache_unregister(&rsc_cache);
+	cache_unregister(&rsi_cache);
 	svc_auth_unregister(RPC_AUTH_GSS);
 }

net/sunrpc/cache.c

@@ -245,6 +245,7 @@ int cache_check(struct cache_detail *detail,
 		cache_put(h, detail);
 	return rv;
 }
+EXPORT_SYMBOL(cache_check);
 
 /*
  * caches need to be periodically cleaned.
@@ -290,44 +291,78 @@ static const struct file_operations cache_flush_operations;
 static void do_cache_clean(struct work_struct *work);
 static DECLARE_DELAYED_WORK(cache_cleaner, do_cache_clean);
 
-void cache_register(struct cache_detail *cd)
+static void remove_cache_proc_entries(struct cache_detail *cd)
 {
-	cd->proc_ent = proc_mkdir(cd->name, proc_net_rpc);
-	if (cd->proc_ent) {
-		struct proc_dir_entry *p;
-		cd->proc_ent->owner = cd->owner;
-		cd->channel_ent = cd->content_ent = NULL;
+	if (cd->proc_ent == NULL)
+		return;
+	if (cd->flush_ent)
+		remove_proc_entry("flush", cd->proc_ent);
+	if (cd->channel_ent)
+		remove_proc_entry("channel", cd->proc_ent);
+	if (cd->content_ent)
+		remove_proc_entry("content", cd->proc_ent);
+	cd->proc_ent = NULL;
+	remove_proc_entry(cd->name, proc_net_rpc);
+}
 
-		p = create_proc_entry("flush", S_IFREG|S_IRUSR|S_IWUSR,
-				      cd->proc_ent);
-		cd->flush_ent =  p;
-		if (p) {
-			p->proc_fops = &cache_flush_operations;
-			p->owner = cd->owner;
-			p->data = cd;
-		}
+#ifdef CONFIG_PROC_FS
+static int create_cache_proc_entries(struct cache_detail *cd)
+{
+	struct proc_dir_entry *p;
 
-		if (cd->cache_request || cd->cache_parse) {
-			p = create_proc_entry("channel", S_IFREG|S_IRUSR|S_IWUSR,
-					      cd->proc_ent);
-			cd->channel_ent = p;
-			if (p) {
-				p->proc_fops = &cache_file_operations;
-				p->owner = cd->owner;
-				p->data = cd;
-			}
-		}
-		if (cd->cache_show) {
-			p = create_proc_entry("content", S_IFREG|S_IRUSR|S_IWUSR,
-					      cd->proc_ent);
-			cd->content_ent = p;
-			if (p) {
-				p->proc_fops = &content_file_operations;
-				p->owner = cd->owner;
-				p->data = cd;
-			}
-		}
+	cd->proc_ent = proc_mkdir(cd->name, proc_net_rpc);
+	if (cd->proc_ent == NULL)
+		goto out_nomem;
+	cd->proc_ent->owner = cd->owner;
+	cd->channel_ent = cd->content_ent = NULL;
+
+	p = create_proc_entry("flush", S_IFREG|S_IRUSR|S_IWUSR, cd->proc_ent);
+	cd->flush_ent = p;
+	if (p == NULL)
+		goto out_nomem;
+	p->proc_fops = &cache_flush_operations;
+	p->owner = cd->owner;
+	p->data = cd;
+
+	if (cd->cache_request || cd->cache_parse) {
+		p = create_proc_entry("channel", S_IFREG|S_IRUSR|S_IWUSR,
+				      cd->proc_ent);
+		cd->channel_ent = p;
+		if (p == NULL)
+			goto out_nomem;
+		p->proc_fops = &cache_file_operations;
+		p->owner = cd->owner;
+		p->data = cd;
 	}
+	if (cd->cache_show) {
+		p = create_proc_entry("content", S_IFREG|S_IRUSR|S_IWUSR,
+				      cd->proc_ent);
+		cd->content_ent = p;
+		if (p == NULL)
+			goto out_nomem;
+		p->proc_fops = &content_file_operations;
+		p->owner = cd->owner;
+		p->data = cd;
+	}
+	return 0;
+out_nomem:
+	remove_cache_proc_entries(cd);
+	return -ENOMEM;
+}
+#else /* CONFIG_PROC_FS */
+static int create_cache_proc_entries(struct cache_detail *cd)
+{
+	return 0;
+}
+#endif
+
+int cache_register(struct cache_detail *cd)
+{
+	int ret;
+
+	ret = create_cache_proc_entries(cd);
+	if (ret)
+		return ret;
 	rwlock_init(&cd->hash_lock);
 	INIT_LIST_HEAD(&cd->queue);
 	spin_lock(&cache_list_lock);
@@ -341,9 +376,11 @@ void cache_register(struct cache_detail *cd)
 
 	/* start the cleaning process */
 	schedule_delayed_work(&cache_cleaner, 0);
+	return 0;
 }
+EXPORT_SYMBOL(cache_register);
 
-int cache_unregister(struct cache_detail *cd)
+void cache_unregister(struct cache_detail *cd)
 {
 	cache_purge(cd);
 	spin_lock(&cache_list_lock);
@@ -351,30 +388,23 @@ int cache_unregister(struct cache_detail *cd)
 	if (cd->entries || atomic_read(&cd->inuse)) {
 		write_unlock(&cd->hash_lock);
 		spin_unlock(&cache_list_lock);
-		return -EBUSY;
+		goto out;
 	}
 	if (current_detail == cd)
 		current_detail = NULL;
 	list_del_init(&cd->others);
 	write_unlock(&cd->hash_lock);
 	spin_unlock(&cache_list_lock);
-	if (cd->proc_ent) {
-		if (cd->flush_ent)
-			remove_proc_entry("flush", cd->proc_ent);
-		if (cd->channel_ent)
-			remove_proc_entry("channel", cd->proc_ent);
-		if (cd->content_ent)
-			remove_proc_entry("content", cd->proc_ent);
-
-		cd->proc_ent = NULL;
-		remove_proc_entry(cd->name, proc_net_rpc);
-	}
+	remove_cache_proc_entries(cd);
 	if (list_empty(&cache_list)) {
 		/* module must be being unloaded so its safe to kill the worker */
 		cancel_delayed_work_sync(&cache_cleaner);
 	}
-	return 0;
+	return;
+out:
+	printk(KERN_ERR "nfsd: failed to unregister %s cache\n", cd->name);
 }
+EXPORT_SYMBOL(cache_unregister);
 
 /* clean cache tries to find something to clean
  * and cleans it.
@@ -489,6 +519,7 @@ void cache_flush(void)
 	while (cache_clean() != -1)
 		cond_resched();
 }
+EXPORT_SYMBOL(cache_flush);
 
 void cache_purge(struct cache_detail *detail)
 {
@@ -497,7 +528,7 @@ void cache_purge(struct cache_detail *detail)
 	cache_flush();
 	detail->flush_time = 1;
 }
-
+EXPORT_SYMBOL(cache_purge);
 
 
 /*
@@ -634,13 +665,13 @@ void cache_clean_deferred(void *owner)
 /*
  * communicate with user-space
  *
- * We have a magic /proc file - /proc/sunrpc/cache
- * On read, you get a full request, or block
- * On write, an update request is processed
- * Poll works if anything to read, and always allows write
+ * We have a magic /proc file - /proc/sunrpc/<cachename>/channel.
+ * On read, you get a full request, or block.
+ * On write, an update request is processed.
+ * Poll works if anything to read, and always allows write.
  *
  * Implemented by linked list of requests.  Each open file has
- * a ->private that also exists in this list.  New request are added
+ * a ->private that also exists in this list.  New requests are added
  * to the end and may wakeup and preceding readers.
  * New readers are added to the head.  If, on read, an item is found with
  * CACHE_UPCALLING clear, we free it from the list.
@@ -963,6 +994,7 @@ void qword_add(char **bpp, int *lp, char *str)
 	*bpp = bp;
 	*lp = len;
 }
+EXPORT_SYMBOL(qword_add);
 
 void qword_addhex(char **bpp, int *lp, char *buf, int blen)
 {
@@ -991,6 +1023,7 @@ void qword_addhex(char **bpp, int *lp, char *buf, int blen)
 	*bpp = bp;
 	*lp = len;
 }
+EXPORT_SYMBOL(qword_addhex);
 
 static void warn_no_listener(struct cache_detail *detail)
 {
@@ -1113,6 +1146,7 @@ int qword_get(char **bpp, char *dest, int bufsize)
 	*dest = '\0';
 	return len;
 }
+EXPORT_SYMBOL(qword_get);
 
 
 /*
@@ -1244,18 +1278,18 @@ static ssize_t read_flush(struct file *file, char __user *buf,
 	struct cache_detail *cd = PDE(file->f_path.dentry->d_inode)->data;
 	char tbuf[20];
 	unsigned long p = *ppos;
-	int len;
+	size_t len;
 
 	sprintf(tbuf, "%lu\n", cd->flush_time);
 	len = strlen(tbuf);
 	if (p >= len)
 		return 0;
 	len -= p;
-	if (len > count) len = count;
+	if (len > count)
+		len = count;
 	if (copy_to_user(buf, (void*)(tbuf+p), len))
-		len = -EFAULT;
-	else
-		*ppos += len;
+		return -EFAULT;
+	*ppos += len;
 	return len;
 }
 

net/sunrpc/stats.c

@@ -33,7 +33,7 @@ struct proc_dir_entry	*proc_net_rpc = NULL;
 static int rpc_proc_show(struct seq_file *seq, void *v) {
 	const struct rpc_stat	*statp = seq->private;
 	const struct rpc_program *prog = statp->program;
-	int		i, j;
+	unsigned int i, j;
 
 	seq_printf(seq,
 		"net %u %u %u %u\n",
@@ -81,7 +81,7 @@ void svc_seq_show(struct seq_file *seq, const struct svc_stat *statp) {
 	const struct svc_program *prog = statp->program;
 	const struct svc_procedure *proc;
 	const struct svc_version *vers;
-	int		i, j;
+	unsigned int i, j;
 
 	seq_printf(seq,
 		"net %u %u %u %u\n",
@@ -106,6 +106,7 @@ void svc_seq_show(struct seq_file *seq, const struct svc_stat *statp) {
 		seq_putc(seq, '\n');
 	}
 }
+EXPORT_SYMBOL(svc_seq_show);
 
 /**
  * rpc_alloc_iostats - allocate an rpc_iostats structure
@@ -255,12 +256,14 @@ svc_proc_register(struct svc_stat *statp, const struct file_operations *fops)
 {
 	return do_register(statp->program->pg_name, statp, fops);
 }
+EXPORT_SYMBOL(svc_proc_register);
 
 void
 svc_proc_unregister(const char *name)
 {
 	remove_proc_entry(name, proc_net_rpc);
 }
+EXPORT_SYMBOL(svc_proc_unregister);
 
 void
 rpc_proc_init(void)

net/sunrpc/sunrpc_syms.c

@@ -22,48 +22,6 @@
 #include <linux/sunrpc/rpc_pipe_fs.h>
 #include <linux/sunrpc/xprtsock.h>
 
-/* RPC server stuff */
-EXPORT_SYMBOL(svc_create);
-EXPORT_SYMBOL(svc_create_thread);
-EXPORT_SYMBOL(svc_create_pooled);
-EXPORT_SYMBOL(svc_set_num_threads);
-EXPORT_SYMBOL(svc_exit_thread);
-EXPORT_SYMBOL(svc_destroy);
-EXPORT_SYMBOL(svc_drop);
-EXPORT_SYMBOL(svc_process);
-EXPORT_SYMBOL(svc_recv);
-EXPORT_SYMBOL(svc_wake_up);
-EXPORT_SYMBOL(svc_makesock);
-EXPORT_SYMBOL(svc_reserve);
-EXPORT_SYMBOL(svc_auth_register);
-EXPORT_SYMBOL(auth_domain_lookup);
-EXPORT_SYMBOL(svc_authenticate);
-EXPORT_SYMBOL(svc_set_client);
-
-/* RPC statistics */
-#ifdef CONFIG_PROC_FS
-EXPORT_SYMBOL(svc_proc_register);
-EXPORT_SYMBOL(svc_proc_unregister);
-EXPORT_SYMBOL(svc_seq_show);
-#endif
-
-/* caching... */
-EXPORT_SYMBOL(auth_domain_find);
-EXPORT_SYMBOL(auth_domain_put);
-EXPORT_SYMBOL(auth_unix_add_addr);
-EXPORT_SYMBOL(auth_unix_forget_old);
-EXPORT_SYMBOL(auth_unix_lookup);
-EXPORT_SYMBOL(cache_check);
-EXPORT_SYMBOL(cache_flush);
-EXPORT_SYMBOL(cache_purge);
-EXPORT_SYMBOL(cache_register);
-EXPORT_SYMBOL(cache_unregister);
-EXPORT_SYMBOL(qword_add);
-EXPORT_SYMBOL(qword_addhex);
-EXPORT_SYMBOL(qword_get);
-EXPORT_SYMBOL(svcauth_unix_purge);
-EXPORT_SYMBOL(unix_domain_find);
-
 extern struct cache_detail ip_map_cache, unix_gid_cache;
 
 static int __init
@@ -85,7 +43,8 @@ init_sunrpc(void)
 #endif
 	cache_register(&ip_map_cache);
 	cache_register(&unix_gid_cache);
-	init_socket_xprt();
+	svc_init_xprt_sock();	/* svc sock transport */
+	init_socket_xprt();	/* clnt sock transport */
 	rpcauth_init_module();
 out:
 	return err;
@@ -96,12 +55,11 @@ cleanup_sunrpc(void)
 {
 	rpcauth_remove_module();
 	cleanup_socket_xprt();
+	svc_cleanup_xprt_sock();
 	unregister_rpc_pipefs();
 	rpc_destroy_mempool();
-	if (cache_unregister(&ip_map_cache))
-		printk(KERN_ERR "sunrpc: failed to unregister ip_map cache\n");
-	if (cache_unregister(&unix_gid_cache))
-	      printk(KERN_ERR "sunrpc: failed to unregister unix_gid cache\n");
+	cache_unregister(&ip_map_cache);
+	cache_unregister(&unix_gid_cache);
 #ifdef RPC_DEBUG
 	rpc_unregister_sysctl();
 #endif

net/sunrpc/svc.c

@@ -364,7 +364,7 @@ __svc_create(struct svc_program *prog, unsigned int bufsize, int npools,
 	   void (*shutdown)(struct svc_serv *serv))
 {
 	struct svc_serv	*serv;
-	int vers;
+	unsigned int vers;
 	unsigned int xdrsize;
 	unsigned int i;
 
@@ -433,6 +433,7 @@ svc_create(struct svc_program *prog, unsigned int bufsize,
 {
 	return __svc_create(prog, bufsize, /*npools*/1, shutdown);
 }
+EXPORT_SYMBOL(svc_create);
 
 struct svc_serv *
 svc_create_pooled(struct svc_program *prog, unsigned int bufsize,
@@ -452,6 +453,7 @@ svc_create_pooled(struct svc_program *prog, unsigned int bufsize,
 
 	return serv;
 }
+EXPORT_SYMBOL(svc_create_pooled);
 
 /*
  * Destroy an RPC service.  Should be called with the BKL held
@@ -459,9 +461,6 @@ svc_create_pooled(struct svc_program *prog, unsigned int bufsize,
 void
 svc_destroy(struct svc_serv *serv)
 {
-	struct svc_sock	*svsk;
-	struct svc_sock *tmp;
-
 	dprintk("svc: svc_destroy(%s, %d)\n",
 				serv->sv_program->pg_name,
 				serv->sv_nrthreads);
@@ -476,14 +475,12 @@ svc_destroy(struct svc_serv *serv)
 
 	del_timer_sync(&serv->sv_temptimer);
 
-	list_for_each_entry_safe(svsk, tmp, &serv->sv_tempsocks, sk_list)
-		svc_force_close_socket(svsk);
+	svc_close_all(&serv->sv_tempsocks);
 
 	if (serv->sv_shutdown)
 		serv->sv_shutdown(serv);
 
-	list_for_each_entry_safe(svsk, tmp, &serv->sv_permsocks, sk_list)
-		svc_force_close_socket(svsk);
+	svc_close_all(&serv->sv_permsocks);
 
 	BUG_ON(!list_empty(&serv->sv_permsocks));
 	BUG_ON(!list_empty(&serv->sv_tempsocks));
@@ -498,6 +495,7 @@ svc_destroy(struct svc_serv *serv)
 	kfree(serv->sv_pools);
 	kfree(serv);
 }
+EXPORT_SYMBOL(svc_destroy);
 
 /*
  * Allocate an RPC server's buffer space.
@@ -536,31 +534,17 @@ svc_release_buffer(struct svc_rqst *rqstp)
 			put_page(rqstp->rq_pages[i]);
 }
 
-/*
- * Create a thread in the given pool.  Caller must hold BKL.
- * On a NUMA or SMP machine, with a multi-pool serv, the thread
- * will be restricted to run on the cpus belonging to the pool.
- */
-static int
-__svc_create_thread(svc_thread_fn func, struct svc_serv *serv,
-		    struct svc_pool *pool)
+struct svc_rqst *
+svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool)
 {
 	struct svc_rqst	*rqstp;
-	int		error = -ENOMEM;
-	int		have_oldmask = 0;
-	cpumask_t	oldmask;
 
 	rqstp = kzalloc(sizeof(*rqstp), GFP_KERNEL);
 	if (!rqstp)
-		goto out;
+		goto out_enomem;
 
 	init_waitqueue_head(&rqstp->rq_wait);
 
-	if (!(rqstp->rq_argp = kmalloc(serv->sv_xdrsize, GFP_KERNEL))
-	 || !(rqstp->rq_resp = kmalloc(serv->sv_xdrsize, GFP_KERNEL))
-	 || !svc_init_buffer(rqstp, serv->sv_max_mesg))
-		goto out_thread;
-
 	serv->sv_nrthreads++;
 	spin_lock_bh(&pool->sp_lock);
 	pool->sp_nrthreads++;
@@ -569,6 +553,45 @@ __svc_create_thread(svc_thread_fn func, struct svc_serv *serv,
 	rqstp->rq_server = serv;
 	rqstp->rq_pool = pool;
 
+	rqstp->rq_argp = kmalloc(serv->sv_xdrsize, GFP_KERNEL);
+	if (!rqstp->rq_argp)
+		goto out_thread;
+
+	rqstp->rq_resp = kmalloc(serv->sv_xdrsize, GFP_KERNEL);
+	if (!rqstp->rq_resp)
+		goto out_thread;
+
+	if (!svc_init_buffer(rqstp, serv->sv_max_mesg))
+		goto out_thread;
+
+	return rqstp;
+out_thread:
+	svc_exit_thread(rqstp);
+out_enomem:
+	return ERR_PTR(-ENOMEM);
+}
+EXPORT_SYMBOL(svc_prepare_thread);
+
+/*
+ * Create a thread in the given pool.  Caller must hold BKL.
+ * On a NUMA or SMP machine, with a multi-pool serv, the thread
+ * will be restricted to run on the cpus belonging to the pool.
+ */
+static int
+__svc_create_thread(svc_thread_fn func, struct svc_serv *serv,
+		    struct svc_pool *pool)
+{
+	struct svc_rqst	*rqstp;
+	int		error = -ENOMEM;
+	int		have_oldmask = 0;
+	cpumask_t	oldmask;
+
+	rqstp = svc_prepare_thread(serv, pool);
+	if (IS_ERR(rqstp)) {
+		error = PTR_ERR(rqstp);
+		goto out;
+	}
+
 	if (serv->sv_nrpools > 1)
 		have_oldmask = svc_pool_map_set_cpumask(pool->sp_id, &oldmask);
 
@@ -597,6 +620,7 @@ svc_create_thread(svc_thread_fn func, struct svc_serv *serv)
 {
 	return __svc_create_thread(func, serv, &serv->sv_pools[0]);
 }
+EXPORT_SYMBOL(svc_create_thread);
 
 /*
  * Choose a pool in which to create a new thread, for svc_set_num_threads
@@ -700,6 +724,7 @@ svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
 
 	return error;
 }
+EXPORT_SYMBOL(svc_set_num_threads);
 
 /*
  * Called from a server thread as it's exiting.  Caller must hold BKL.
@@ -726,6 +751,7 @@ svc_exit_thread(struct svc_rqst *rqstp)
 	if (serv)
 		svc_destroy(serv);
 }
+EXPORT_SYMBOL(svc_exit_thread);
 
 /*
  * Register an RPC service with the local portmapper.
@@ -737,7 +763,8 @@ svc_register(struct svc_serv *serv, int proto, unsigned short port)
 {
 	struct svc_program	*progp;
 	unsigned long		flags;
-	int			i, error = 0, dummy;
+	unsigned int		i;
+	int			error = 0, dummy;
 
 	if (!port)
 		clear_thread_flag(TIF_SIGPENDING);
@@ -840,9 +867,9 @@ svc_process(struct svc_rqst *rqstp)
 	rqstp->rq_res.tail[0].iov_len = 0;
 	/* Will be turned off only in gss privacy case: */
 	rqstp->rq_splice_ok = 1;
-	/* tcp needs a space for the record length... */
-	if (rqstp->rq_prot == IPPROTO_TCP)
-		svc_putnl(resv, 0);
+
+	/* Setup reply header */
+	rqstp->rq_xprt->xpt_ops->xpo_prep_reply_hdr(rqstp);
 
 	rqstp->rq_xid = svc_getu32(argv);
 	svc_putu32(resv, rqstp->rq_xid);
@@ -1049,16 +1076,15 @@ err_bad:
 	svc_putnl(resv, ntohl(rpc_stat));
 	goto sendit;
 }
+EXPORT_SYMBOL(svc_process);
 
 /*
  * Return (transport-specific) limit on the rpc payload.
  */
 u32 svc_max_payload(const struct svc_rqst *rqstp)
 {
-	int max = RPCSVC_MAXPAYLOAD_TCP;
+	u32 max = rqstp->rq_xprt->xpt_class->xcl_max_payload;
 
-	if (rqstp->rq_sock->sk_sock->type == SOCK_DGRAM)
-		max = RPCSVC_MAXPAYLOAD_UDP;
 	if (rqstp->rq_server->sv_max_payload < max)
 		max = rqstp->rq_server->sv_max_payload;
 	return max;

net/sunrpc/svcauth.c

@@ -57,11 +57,13 @@ svc_authenticate(struct svc_rqst *rqstp, __be32 *authp)
 	rqstp->rq_authop = aops;
 	return aops->accept(rqstp, authp);
 }
+EXPORT_SYMBOL(svc_authenticate);
 
 int svc_set_client(struct svc_rqst *rqstp)
 {
 	return rqstp->rq_authop->set_client(rqstp);
 }
+EXPORT_SYMBOL(svc_set_client);
 
 /* A request, which was authenticated, has now executed.
  * Time to finalise the credentials and verifier
@@ -93,6 +95,7 @@ svc_auth_register(rpc_authflavor_t flavor, struct auth_ops *aops)
 	spin_unlock(&authtab_lock);
 	return rv;
 }
+EXPORT_SYMBOL(svc_auth_register);
 
 void
 svc_auth_unregister(rpc_authflavor_t flavor)
@@ -129,6 +132,7 @@ void auth_domain_put(struct auth_domain *dom)
 		spin_unlock(&auth_domain_lock);
 	}
 }
+EXPORT_SYMBOL(auth_domain_put);
 
 struct auth_domain *
 auth_domain_lookup(char *name, struct auth_domain *new)
@@ -153,8 +157,10 @@ auth_domain_lookup(char *name, struct auth_domain *new)
 	spin_unlock(&auth_domain_lock);
 	return new;
 }
+EXPORT_SYMBOL(auth_domain_lookup);
 
 struct auth_domain *auth_domain_find(char *name)
 {
 	return auth_domain_lookup(name, NULL);
 }
+EXPORT_SYMBOL(auth_domain_find);

net/sunrpc/svcauth_unix.c

@@ -63,6 +63,7 @@ struct auth_domain *unix_domain_find(char *name)
 		rv = auth_domain_lookup(name, &new->h);
 	}
 }
+EXPORT_SYMBOL(unix_domain_find);
 
 static void svcauth_unix_domain_release(struct auth_domain *dom)
 {
@@ -340,6 +341,7 @@ int auth_unix_add_addr(struct in_addr addr, struct auth_domain *dom)
 	else
 		return -ENOMEM;
 }
+EXPORT_SYMBOL(auth_unix_add_addr);
 
 int auth_unix_forget_old(struct auth_domain *dom)
 {
@@ -351,6 +353,7 @@ int auth_unix_forget_old(struct auth_domain *dom)
 	udom->addr_changes++;
 	return 0;
 }
+EXPORT_SYMBOL(auth_unix_forget_old);
 
 struct auth_domain *auth_unix_lookup(struct in_addr addr)
 {
@@ -375,50 +378,56 @@ struct auth_domain *auth_unix_lookup(struct in_addr addr)
 	cache_put(&ipm->h, &ip_map_cache);
 	return rv;
 }
+EXPORT_SYMBOL(auth_unix_lookup);
 
 void svcauth_unix_purge(void)
 {
 	cache_purge(&ip_map_cache);
 }
+EXPORT_SYMBOL(svcauth_unix_purge);
 
 static inline struct ip_map *
 ip_map_cached_get(struct svc_rqst *rqstp)
 {
-	struct ip_map *ipm;
-	struct svc_sock *svsk = rqstp->rq_sock;
-	spin_lock(&svsk->sk_lock);
-	ipm = svsk->sk_info_authunix;
-	if (ipm != NULL) {
-		if (!cache_valid(&ipm->h)) {
-			/*
-			 * The entry has been invalidated since it was
-			 * remembered, e.g. by a second mount from the
-			 * same IP address.
-			 */
-			svsk->sk_info_authunix = NULL;
-			spin_unlock(&svsk->sk_lock);
-			cache_put(&ipm->h, &ip_map_cache);
-			return NULL;
+	struct ip_map *ipm = NULL;
+	struct svc_xprt *xprt = rqstp->rq_xprt;
+
+	if (test_bit(XPT_CACHE_AUTH, &xprt->xpt_flags)) {
+		spin_lock(&xprt->xpt_lock);
+		ipm = xprt->xpt_auth_cache;
+		if (ipm != NULL) {
+			if (!cache_valid(&ipm->h)) {
+				/*
+				 * The entry has been invalidated since it was
+				 * remembered, e.g. by a second mount from the
+				 * same IP address.
+				 */
+				xprt->xpt_auth_cache = NULL;
+				spin_unlock(&xprt->xpt_lock);
+				cache_put(&ipm->h, &ip_map_cache);
+				return NULL;
+			}
+			cache_get(&ipm->h);
 		}
-		cache_get(&ipm->h);
+		spin_unlock(&xprt->xpt_lock);
 	}
-	spin_unlock(&svsk->sk_lock);
 	return ipm;
 }
 
 static inline void
 ip_map_cached_put(struct svc_rqst *rqstp, struct ip_map *ipm)
 {
-	struct svc_sock *svsk = rqstp->rq_sock;
+	struct svc_xprt *xprt = rqstp->rq_xprt;
 
-	spin_lock(&svsk->sk_lock);
-	if (svsk->sk_sock->type == SOCK_STREAM &&
-	    svsk->sk_info_authunix == NULL) {
-		/* newly cached, keep the reference */
-		svsk->sk_info_authunix = ipm;
-		ipm = NULL;
+	if (test_bit(XPT_CACHE_AUTH, &xprt->xpt_flags)) {
+		spin_lock(&xprt->xpt_lock);
+		if (xprt->xpt_auth_cache == NULL) {
+			/* newly cached, keep the reference */
+			xprt->xpt_auth_cache = ipm;
+			ipm = NULL;
+		}
+		spin_unlock(&xprt->xpt_lock);
 	}
-	spin_unlock(&svsk->sk_lock);
 	if (ipm)
 		cache_put(&ipm->h, &ip_map_cache);
 }

net/sunrpc/sysctl.c

@@ -18,6 +18,7 @@
 #include <linux/sunrpc/types.h>
 #include <linux/sunrpc/sched.h>
 #include <linux/sunrpc/stats.h>
+#include <linux/sunrpc/svc_xprt.h>
 
 /*
  * Declare the debug flags here
@@ -55,6 +56,30 @@ rpc_unregister_sysctl(void)
 	}
 }
 
+static int proc_do_xprt(ctl_table *table, int write, struct file *file,
+			void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+	char tmpbuf[256];
+	int len;
+	if ((*ppos && !write) || !*lenp) {
+		*lenp = 0;
+		return 0;
+	}
+	if (write)
+		return -EINVAL;
+	else {
+		len = svc_print_xprts(tmpbuf, sizeof(tmpbuf));
+		if (!access_ok(VERIFY_WRITE, buffer, len))
+			return -EFAULT;
+
+		if (__copy_to_user(buffer, tmpbuf, len))
+			return -EFAULT;
+	}
+	*lenp -= len;
+	*ppos += len;
+	return 0;
+}
+
 static int
 proc_dodebug(ctl_table *table, int write, struct file *file,
 				void __user *buffer, size_t *lenp, loff_t *ppos)
@@ -147,6 +172,12 @@ static ctl_table debug_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dodebug
 	},
+	{
+		.procname	= "transports",
+		.maxlen		= 256,
+		.mode		= 0444,
+		.proc_handler	= &proc_do_xprt,
+	},
 	{ .ctl_name = 0 }
 };
 

net/sunrpc/xdr.c

@@ -96,11 +96,13 @@ xdr_encode_string(__be32 *p, const char *string)
 EXPORT_SYMBOL(xdr_encode_string);
 
 __be32 *
-xdr_decode_string_inplace(__be32 *p, char **sp, int *lenp, int maxlen)
+xdr_decode_string_inplace(__be32 *p, char **sp,
+			  unsigned int *lenp, unsigned int maxlen)
 {
-	unsigned int	len;
+	u32 len;
 
-	if ((len = ntohl(*p++)) > maxlen)
+	len = ntohl(*p++);
+	if (len > maxlen)
 		return NULL;
 	*lenp = len;
 	*sp = (char *) p;

net/sunrpc/xprtrdma/Makefile

 obj-$(CONFIG_SUNRPC_XPRT_RDMA) += xprtrdma.o
 
 xprtrdma-y := transport.o rpc_rdma.o verbs.o
+
+obj-$(CONFIG_SUNRPC_XPRT_RDMA) += svcrdma.o
+
+svcrdma-y := svc_rdma.o svc_rdma_transport.o \
+	svc_rdma_marshal.o svc_rdma_sendto.o svc_rdma_recvfrom.o

net/sunrpc/xprtrdma/svc_rdma.c

+/*
+ * Copyright (c) 2005-2006 Network Appliance, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the BSD-type
+ * license below:
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ *      Redistributions of source code must retain the above copyright
+ *      notice, this list of conditions and the following disclaimer.
+ *
+ *      Redistributions in binary form must reproduce the above
+ *      copyright notice, this list of conditions and the following
+ *      disclaimer in the documentation and/or other materials provided
+ *      with the distribution.
+ *
+ *      Neither the name of the Network Appliance, Inc. nor the names of
+ *      its contributors may be used to endorse or promote products
+ *      derived from this software without specific prior written
+ *      permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * Author: Tom Tucker <tom@opengridcomputing.com>
+ */
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <linux/sysctl.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/sched.h>
+#include <linux/sunrpc/svc_rdma.h>
+
+#define RPCDBG_FACILITY	RPCDBG_SVCXPRT
+
+/* RPC/RDMA parameters */
+unsigned int svcrdma_ord = RPCRDMA_ORD;
+static unsigned int min_ord = 1;
+static unsigned int max_ord = 4096;
+unsigned int svcrdma_max_requests = RPCRDMA_MAX_REQUESTS;
+static unsigned int min_max_requests = 4;
+static unsigned int max_max_requests = 16384;
+unsigned int svcrdma_max_req_size = RPCRDMA_MAX_REQ_SIZE;
+static unsigned int min_max_inline = 4096;
+static unsigned int max_max_inline = 65536;
+
+atomic_t rdma_stat_recv;
+atomic_t rdma_stat_read;
+atomic_t rdma_stat_write;
+atomic_t rdma_stat_sq_starve;
+atomic_t rdma_stat_rq_starve;
+atomic_t rdma_stat_rq_poll;
+atomic_t rdma_stat_rq_prod;
+atomic_t rdma_stat_sq_poll;
+atomic_t rdma_stat_sq_prod;
+
+/*
+ * This function implements reading and resetting an atomic_t stat
+ * variable through read/write to a proc file. Any write to the file
+ * resets the associated statistic to zero. Any read returns it's
+ * current value.
+ */
+static int read_reset_stat(ctl_table *table, int write,
+			   struct file *filp, void __user *buffer, size_t *lenp,
+			   loff_t *ppos)
+{
+	atomic_t *stat = (atomic_t *)table->data;
+
+	if (!stat)
+		return -EINVAL;
+
+	if (write)
+		atomic_set(stat, 0);
+	else {
+		char str_buf[32];
+		char *data;
+		int len = snprintf(str_buf, 32, "%d\n", atomic_read(stat));
+		if (len >= 32)
+			return -EFAULT;
+		len = strlen(str_buf);
+		if (*ppos > len) {
+			*lenp = 0;
+			return 0;
+		}
+		data = &str_buf[*ppos];
+		len -= *ppos;
+		if (len > *lenp)
+			len = *lenp;
+		if (len && copy_to_user(buffer, str_buf, len))
+			return -EFAULT;
+		*lenp = len;
+		*ppos += len;
+	}
+	return 0;
+}
+
+static struct ctl_table_header *svcrdma_table_header;
+static ctl_table svcrdma_parm_table[] = {
+	{
+		.procname	= "max_requests",
+		.data		= &svcrdma_max_requests,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &min_max_requests,
+		.extra2		= &max_max_requests
+	},
+	{
+		.procname	= "max_req_size",
+		.data		= &svcrdma_max_req_size,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &min_max_inline,
+		.extra2		= &max_max_inline
+	},
+	{
+		.procname	= "max_outbound_read_requests",
+		.data		= &svcrdma_ord,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &min_ord,
+		.extra2		= &max_ord,
+	},
+
+	{
+		.procname	= "rdma_stat_read",
+		.data		= &rdma_stat_read,
+		.maxlen		= sizeof(atomic_t),
+		.mode		= 0644,
+		.proc_handler	= &read_reset_stat,
+	},
+	{
+		.procname	= "rdma_stat_recv",
+		.data		= &rdma_stat_recv,
+		.maxlen		= sizeof(atomic_t),
+		.mode		= 0644,
+		.proc_handler	= &read_reset_stat,
+	},
+	{
+		.procname	= "rdma_stat_write",
+		.data		= &rdma_stat_write,
+		.maxlen		= sizeof(atomic_t),
+		.mode		= 0644,
+		.proc_handler	= &read_reset_stat,
+	},
+	{
+		.procname	= "rdma_stat_sq_starve",
+		.data		= &rdma_stat_sq_starve,
+		.maxlen		= sizeof(atomic_t),
+		.mode		= 0644,
+		.proc_handler	= &read_reset_stat,
+	},
+	{
+		.procname	= "rdma_stat_rq_starve",
+		.data		= &rdma_stat_rq_starve,
+		.maxlen		= sizeof(atomic_t),
+		.mode		= 0644,
+		.proc_handler	= &read_reset_stat,
+	},
+	{
+		.procname	= "rdma_stat_rq_poll",
+		.data		= &rdma_stat_rq_poll,
+		.maxlen		= sizeof(atomic_t),
+		.mode		= 0644,
+		.proc_handler	= &read_reset_stat,
+	},
+	{
+		.procname	= "rdma_stat_rq_prod",
+		.data		= &rdma_stat_rq_prod,
+		.maxlen		= sizeof(atomic_t),
+		.mode		= 0644,
+		.proc_handler	= &read_reset_stat,
+	},
+	{
+		.procname	= "rdma_stat_sq_poll",
+		.data		= &rdma_stat_sq_poll,
+		.maxlen		= sizeof(atomic_t),
+		.mode		= 0644,
+		.proc_handler	= &read_reset_stat,
+	},
+	{
+		.procname	= "rdma_stat_sq_prod",
+		.data		= &rdma_stat_sq_prod,
+		.maxlen		= sizeof(atomic_t),
+		.mode		= 0644,
+		.proc_handler	= &read_reset_stat,
+	},
+	{
+		.ctl_name = 0,
+	},
+};
+
+static ctl_table svcrdma_table[] = {
+	{
+		.procname	= "svc_rdma",
+		.mode		= 0555,
+		.child		= svcrdma_parm_table
+	},
+	{
+		.ctl_name = 0,
+	},
+};
+
+static ctl_table svcrdma_root_table[] = {
+	{
+		.ctl_name	= CTL_SUNRPC,
+		.procname	= "sunrpc",
+		.mode		= 0555,
+		.child		= svcrdma_table
+	},
+	{
+		.ctl_name = 0,
+	},
+};
+
+void svc_rdma_cleanup(void)
+{
+	dprintk("SVCRDMA Module Removed, deregister RPC RDMA transport\n");
+	if (svcrdma_table_header) {
+		unregister_sysctl_table(svcrdma_table_header);
+		svcrdma_table_header = NULL;
+	}
+	svc_unreg_xprt_class(&svc_rdma_class);
+}
+
+int svc_rdma_init(void)
+{
+	dprintk("SVCRDMA Module Init, register RPC RDMA transport\n");
+	dprintk("\tsvcrdma_ord      : %d\n", svcrdma_ord);
+	dprintk("\tmax_requests     : %d\n", svcrdma_max_requests);
+	dprintk("\tsq_depth         : %d\n",
+		svcrdma_max_requests * RPCRDMA_SQ_DEPTH_MULT);
+	dprintk("\tmax_inline       : %d\n", svcrdma_max_req_size);
+	if (!svcrdma_table_header)
+		svcrdma_table_header =
+			register_sysctl_table(svcrdma_root_table);
+
+	/* Register RDMA with the SVC transport switch */
+	svc_reg_xprt_class(&svc_rdma_class);
+	return 0;
+}
+MODULE_AUTHOR("Tom Tucker <tom@opengridcomputing.com>");
+MODULE_DESCRIPTION("SVC RDMA Transport");
+MODULE_LICENSE("Dual BSD/GPL");
+module_init(svc_rdma_init);
+module_exit(svc_rdma_cleanup);

net/sunrpc/xprtrdma/svc_rdma_marshal.c

+/*
+ * Copyright (c) 2005-2006 Network Appliance, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the BSD-type
+ * license below:
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ *      Redistributions of source code must retain the above copyright
+ *      notice, this list of conditions and the following disclaimer.
+ *
+ *      Redistributions in binary form must reproduce the above
+ *      copyright notice, this list of conditions and the following
+ *      disclaimer in the documentation and/or other materials provided
+ *      with the distribution.
+ *
+ *      Neither the name of the Network Appliance, Inc. nor the names of
+ *      its contributors may be used to endorse or promote products
+ *      derived from this software without specific prior written
+ *      permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * Author: Tom Tucker <tom@opengridcomputing.com>
+ */
+
+#include <linux/sunrpc/xdr.h>
+#include <linux/sunrpc/debug.h>
+#include <asm/unaligned.h>
+#include <linux/sunrpc/rpc_rdma.h>
+#include <linux/sunrpc/svc_rdma.h>
+
+#define RPCDBG_FACILITY	RPCDBG_SVCXPRT
+
+/*
+ * Decodes a read chunk list. The expected format is as follows:
+ *    descrim  : xdr_one
+ *    position : u32 offset into XDR stream
+ *    handle   : u32 RKEY
+ *    . . .
+ *  end-of-list: xdr_zero
+ */
+static u32 *decode_read_list(u32 *va, u32 *vaend)
+{
+	struct rpcrdma_read_chunk *ch = (struct rpcrdma_read_chunk *)va;
+
+	while (ch->rc_discrim != xdr_zero) {
+		u64 ch_offset;
+
+		if (((unsigned long)ch + sizeof(struct rpcrdma_read_chunk)) >
+		    (unsigned long)vaend) {
+			dprintk("svcrdma: vaend=%p, ch=%p\n", vaend, ch);
+			return NULL;
+		}
+
+		ch->rc_discrim = ntohl(ch->rc_discrim);
+		ch->rc_position = ntohl(ch->rc_position);
+		ch->rc_target.rs_handle = ntohl(ch->rc_target.rs_handle);
+		ch->rc_target.rs_length = ntohl(ch->rc_target.rs_length);
+		va = (u32 *)&ch->rc_target.rs_offset;
+		xdr_decode_hyper(va, &ch_offset);
+		put_unaligned(ch_offset, (u64 *)va);
+		ch++;
+	}
+	return (u32 *)&ch->rc_position;
+}
+
+/*
+ * Determine number of chunks and total bytes in chunk list. The chunk
+ * list has already been verified to fit within the RPCRDMA header.
+ */
+void svc_rdma_rcl_chunk_counts(struct rpcrdma_read_chunk *ch,
+			       int *ch_count, int *byte_count)
+{
+	/* compute the number of bytes represented by read chunks */
+	*byte_count = 0;
+	*ch_count = 0;
+	for (; ch->rc_discrim != 0; ch++) {
+		*byte_count = *byte_count + ch->rc_target.rs_length;
+		*ch_count = *ch_count + 1;
+	}
+}
+
+/*
+ * Decodes a write chunk list. The expected format is as follows:
+ *    descrim  : xdr_one
+ *    nchunks  : <count>
+ *       handle   : u32 RKEY              ---+
+ *       length   : u32 <len of segment>     |
+ *       offset   : remove va                + <count>
+ *       . . .                               |
+ *                                        ---+
+ */
+static u32 *decode_write_list(u32 *va, u32 *vaend)
+{
+	int ch_no;
+	struct rpcrdma_write_array *ary =
+		(struct rpcrdma_write_array *)va;
+
+	/* Check for not write-array */
+	if (ary->wc_discrim == xdr_zero)
+		return (u32 *)&ary->wc_nchunks;
+
+	if ((unsigned long)ary + sizeof(struct rpcrdma_write_array) >
+	    (unsigned long)vaend) {
+		dprintk("svcrdma: ary=%p, vaend=%p\n", ary, vaend);
+		return NULL;
+	}
+	ary->wc_discrim = ntohl(ary->wc_discrim);
+	ary->wc_nchunks = ntohl(ary->wc_nchunks);
+	if (((unsigned long)&ary->wc_array[0] +
+	     (sizeof(struct rpcrdma_write_chunk) * ary->wc_nchunks)) >
+	    (unsigned long)vaend) {
+		dprintk("svcrdma: ary=%p, wc_nchunks=%d, vaend=%p\n",
+			ary, ary->wc_nchunks, vaend);
+		return NULL;
+	}
+	for (ch_no = 0; ch_no < ary->wc_nchunks; ch_no++) {
+		u64 ch_offset;
+
+		ary->wc_array[ch_no].wc_target.rs_handle =
+			ntohl(ary->wc_array[ch_no].wc_target.rs_handle);
+		ary->wc_array[ch_no].wc_target.rs_length =
+			ntohl(ary->wc_array[ch_no].wc_target.rs_length);
+		va = (u32 *)&ary->wc_array[ch_no].wc_target.rs_offset;
+		xdr_decode_hyper(va, &ch_offset);
+		put_unaligned(ch_offset, (u64 *)va);
+	}
+
+	/*
+	 * rs_length is the 2nd 4B field in wc_target and taking its
+	 * address skips the list terminator
+	 */
+	return (u32 *)&ary->wc_array[ch_no].wc_target.rs_length;
+}
+
+static u32 *decode_reply_array(u32 *va, u32 *vaend)
+{
+	int ch_no;
+	struct rpcrdma_write_array *ary =
+		(struct rpcrdma_write_array *)va;
+
+	/* Check for no reply-array */
+	if (ary->wc_discrim == xdr_zero)
+		return (u32 *)&ary->wc_nchunks;
+
+	if ((unsigned long)ary + sizeof(struct rpcrdma_write_array) >
+	    (unsigned long)vaend) {
+		dprintk("svcrdma: ary=%p, vaend=%p\n", ary, vaend);
+		return NULL;
+	}
+	ary->wc_discrim = ntohl(ary->wc_discrim);
+	ary->wc_nchunks = ntohl(ary->wc_nchunks);
+	if (((unsigned long)&ary->wc_array[0] +
+	     (sizeof(struct rpcrdma_write_chunk) * ary->wc_nchunks)) >
+	    (unsigned long)vaend) {
+		dprintk("svcrdma: ary=%p, wc_nchunks=%d, vaend=%p\n",
+			ary, ary->wc_nchunks, vaend);
+		return NULL;
+	}
+	for (ch_no = 0; ch_no < ary->wc_nchunks; ch_no++) {
+		u64 ch_offset;
+
+		ary->wc_array[ch_no].wc_target.rs_handle =
+			ntohl(ary->wc_array[ch_no].wc_target.rs_handle);
+		ary->wc_array[ch_no].wc_target.rs_length =
+			ntohl(ary->wc_array[ch_no].wc_target.rs_length);
+		va = (u32 *)&ary->wc_array[ch_no].wc_target.rs_offset;
+		xdr_decode_hyper(va, &ch_offset);
+		put_unaligned(ch_offset, (u64 *)va);
+	}
+
+	return (u32 *)&ary->wc_array[ch_no];
+}
+
+int svc_rdma_xdr_decode_req(struct rpcrdma_msg **rdma_req,
+			    struct svc_rqst *rqstp)
+{
+	struct rpcrdma_msg *rmsgp = NULL;
+	u32 *va;
+	u32 *vaend;
+	u32 hdr_len;
+
+	rmsgp = (struct rpcrdma_msg *)rqstp->rq_arg.head[0].iov_base;
+
+	/* Verify that there's enough bytes for header + something */
+	if (rqstp->rq_arg.len <= RPCRDMA_HDRLEN_MIN) {
+		dprintk("svcrdma: header too short = %d\n",
+			rqstp->rq_arg.len);
+		return -EINVAL;
+	}
+
+	/* Decode the header */
+	rmsgp->rm_xid = ntohl(rmsgp->rm_xid);
+	rmsgp->rm_vers = ntohl(rmsgp->rm_vers);
+	rmsgp->rm_credit = ntohl(rmsgp->rm_credit);
+	rmsgp->rm_type = ntohl(rmsgp->rm_type);
+
+	if (rmsgp->rm_vers != RPCRDMA_VERSION)
+		return -ENOSYS;
+
+	/* Pull in the extra for the padded case and bump our pointer */
+	if (rmsgp->rm_type == RDMA_MSGP) {
+		int hdrlen;
+		rmsgp->rm_body.rm_padded.rm_align =
+			ntohl(rmsgp->rm_body.rm_padded.rm_align);
+		rmsgp->rm_body.rm_padded.rm_thresh =
+			ntohl(rmsgp->rm_body.rm_padded.rm_thresh);
+
+		va = &rmsgp->rm_body.rm_padded.rm_pempty[4];
+		rqstp->rq_arg.head[0].iov_base = va;
+		hdrlen = (u32)((unsigned long)va - (unsigned long)rmsgp);
+		rqstp->rq_arg.head[0].iov_len -= hdrlen;
+		if (hdrlen > rqstp->rq_arg.len)
+			return -EINVAL;
+		return hdrlen;
+	}
+
+	/* The chunk list may contain either a read chunk list or a write
+	 * chunk list and a reply chunk list.
+	 */
+	va = &rmsgp->rm_body.rm_chunks[0];
+	vaend = (u32 *)((unsigned long)rmsgp + rqstp->rq_arg.len);
+	va = decode_read_list(va, vaend);
+	if (!va)
+		return -EINVAL;
+	va = decode_write_list(va, vaend);
+	if (!va)
+		return -EINVAL;
+	va = decode_reply_array(va, vaend);
+	if (!va)
+		return -EINVAL;
+
+	rqstp->rq_arg.head[0].iov_base = va;
+	hdr_len = (unsigned long)va - (unsigned long)rmsgp;
+	rqstp->rq_arg.head[0].iov_len -= hdr_len;
+
+	*rdma_req = rmsgp;
+	return hdr_len;
+}
+
+int svc_rdma_xdr_decode_deferred_req(struct svc_rqst *rqstp)
+{
+	struct rpcrdma_msg *rmsgp = NULL;
+	struct rpcrdma_read_chunk *ch;
+	struct rpcrdma_write_array *ary;
+	u32 *va;
+	u32 hdrlen;
+
+	dprintk("svcrdma: processing deferred RDMA header on rqstp=%p\n",
+		rqstp);
+	rmsgp = (struct rpcrdma_msg *)rqstp->rq_arg.head[0].iov_base;
+
+	/* Pull in the extra for the padded case and bump our pointer */
+	if (rmsgp->rm_type == RDMA_MSGP) {
+		va = &rmsgp->rm_body.rm_padded.rm_pempty[4];
+		rqstp->rq_arg.head[0].iov_base = va;
+		hdrlen = (u32)((unsigned long)va - (unsigned long)rmsgp);
+		rqstp->rq_arg.head[0].iov_len -= hdrlen;
+		return hdrlen;
+	}
+
+	/*
+	 * Skip all chunks to find RPC msg. These were previously processed
+	 */
+	va = &rmsgp->rm_body.rm_chunks[0];
+
+	/* Skip read-list */
+	for (ch = (struct rpcrdma_read_chunk *)va;
+	     ch->rc_discrim != xdr_zero; ch++);
+	va = (u32 *)&ch->rc_position;
+
+	/* Skip write-list */
+	ary = (struct rpcrdma_write_array *)va;
+	if (ary->wc_discrim == xdr_zero)
+		va = (u32 *)&ary->wc_nchunks;
+	else
+		/*
+		 * rs_length is the 2nd 4B field in wc_target and taking its
+		 * address skips the list terminator
+		 */
+		va = (u32 *)&ary->wc_array[ary->wc_nchunks].wc_target.rs_length;
+
+	/* Skip reply-array */
+	ary = (struct rpcrdma_write_array *)va;
+	if (ary->wc_discrim == xdr_zero)
+		va = (u32 *)&ary->wc_nchunks;
+	else
+		va = (u32 *)&ary->wc_array[ary->wc_nchunks];
+
+	rqstp->rq_arg.head[0].iov_base = va;
+	hdrlen = (unsigned long)va - (unsigned long)rmsgp;
+	rqstp->rq_arg.head[0].iov_len -= hdrlen;
+
+	return hdrlen;
+}
+
+int svc_rdma_xdr_encode_error(struct svcxprt_rdma *xprt,
+			      struct rpcrdma_msg *rmsgp,
+			      enum rpcrdma_errcode err, u32 *va)
+{
+	u32 *startp = va;
+
+	*va++ = htonl(rmsgp->rm_xid);
+	*va++ = htonl(rmsgp->rm_vers);
+	*va++ = htonl(xprt->sc_max_requests);
+	*va++ = htonl(RDMA_ERROR);
+	*va++ = htonl(err);
+	if (err == ERR_VERS) {
+		*va++ = htonl(RPCRDMA_VERSION);
+		*va++ = htonl(RPCRDMA_VERSION);
+	}
+
+	return (int)((unsigned long)va - (unsigned long)startp);
+}
+
+int svc_rdma_xdr_get_reply_hdr_len(struct rpcrdma_msg *rmsgp)
+{
+	struct rpcrdma_write_array *wr_ary;
+
+	/* There is no read-list in a reply */
+
+	/* skip write list */
+	wr_ary = (struct rpcrdma_write_array *)
+		&rmsgp->rm_body.rm_chunks[1];
+	if (wr_ary->wc_discrim)
+		wr_ary = (struct rpcrdma_write_array *)
+			&wr_ary->wc_array[ntohl(wr_ary->wc_nchunks)].
+			wc_target.rs_length;
+	else
+		wr_ary = (struct rpcrdma_write_array *)
+			&wr_ary->wc_nchunks;
+
+	/* skip reply array */
+	if (wr_ary->wc_discrim)
+		wr_ary = (struct rpcrdma_write_array *)
+			&wr_ary->wc_array[ntohl(wr_ary->wc_nchunks)];
+	else
+		wr_ary = (struct rpcrdma_write_array *)
+			&wr_ary->wc_nchunks;
+
+	return (unsigned long) wr_ary - (unsigned long) rmsgp;
+}
+
+void svc_rdma_xdr_encode_write_list(struct rpcrdma_msg *rmsgp, int chunks)
+{
+	struct rpcrdma_write_array *ary;
+
+	/* no read-list */
+	rmsgp->rm_body.rm_chunks[0] = xdr_zero;
+
+	/* write-array discrim */
+	ary = (struct rpcrdma_write_array *)
+		&rmsgp->rm_body.rm_chunks[1];
+	ary->wc_discrim = xdr_one;
+	ary->wc_nchunks = htonl(chunks);
+
+	/* write-list terminator */
+	ary->wc_array[chunks].wc_target.rs_handle = xdr_zero;
+
+	/* reply-array discriminator */
+	ary->wc_array[chunks].wc_target.rs_length = xdr_zero;
+}
+
+void svc_rdma_xdr_encode_reply_array(struct rpcrdma_write_array *ary,
+				 int chunks)
+{
+	ary->wc_discrim = xdr_one;
+	ary->wc_nchunks = htonl(chunks);
+}
+
+void svc_rdma_xdr_encode_array_chunk(struct rpcrdma_write_array *ary,
+				     int chunk_no,
+				     u32 rs_handle, u64 rs_offset,
+				     u32 write_len)
+{
+	struct rpcrdma_segment *seg = &ary->wc_array[chunk_no].wc_target;
+	seg->rs_handle = htonl(rs_handle);
+	seg->rs_length = htonl(write_len);
+	xdr_encode_hyper((u32 *) &seg->rs_offset, rs_offset);
+}
+
+void svc_rdma_xdr_encode_reply_header(struct svcxprt_rdma *xprt,
+				  struct rpcrdma_msg *rdma_argp,
+				  struct rpcrdma_msg *rdma_resp,
+				  enum rpcrdma_proc rdma_type)
+{
+	rdma_resp->rm_xid = htonl(rdma_argp->rm_xid);
+	rdma_resp->rm_vers = htonl(rdma_argp->rm_vers);
+	rdma_resp->rm_credit = htonl(xprt->sc_max_requests);
+	rdma_resp->rm_type = htonl(rdma_type);
+
+	/* Encode <nul> chunks lists */
+	rdma_resp->rm_body.rm_chunks[0] = xdr_zero;
+	rdma_resp->rm_body.rm_chunks[1] = xdr_zero;
+	rdma_resp->rm_body.rm_chunks[2] = xdr_zero;
+}