Merge branch 'for-3.9' of git://linux-nfs.org/~bfields/linux

Pull nfsd changes from J Bruce Fields:
 "Miscellaneous bugfixes, plus:

   - An overhaul of the DRC cache by Jeff Layton.  The main effect is
     just to make it larger.  This decreases the chances of intermittent
     errors especially in the UDP case.  But we'll need to watch for any
     reports of performance regressions.

   - Containerized nfsd: with some limitations, we now support
     per-container nfs-service, thanks to extensive work from Stanislav
     Kinsbursky over the last year."

Some notes about conflicts, since there were *two* non-data semantic
conflicts here:

 - idr_remove_all() had been added by a memory leak fix, but has since
   become deprecated since idr_destroy() does it for us now.

 - xs_local_connect() had been added by this branch to make AF_LOCAL
   connections be synchronous, but in the meantime Trond had changed the
   calling convention in order to avoid a RCU dereference.

There were a couple of more obvious actual source-level conflicts due to
the hlist traversal changes and one just due to code changes next to
each other, but those were trivial.

* 'for-3.9' of git://linux-nfs.org/~bfields/linux: (49 commits)
  SUNRPC: make AF_LOCAL connect synchronous
  nfsd: fix compiler warning about ambiguous types in nfsd_cache_csum
  svcrpc: fix rpc server shutdown races
  svcrpc: make svc_age_temp_xprts enqueue under sv_lock
  lockd: nlmclnt_reclaim(): avoid stack overflow
  nfsd: enable NFSv4 state in containers
  nfsd: disable usermode helper client tracker in container
  nfsd: use proper net while reading "exports" file
  nfsd: containerize NFSd filesystem
  nfsd: fix comments on nfsd_cache_lookup
  SUNRPC: move cache_detail->cache_request callback call to cache_read()
  SUNRPC: remove "cache_request" argument in sunrpc_cache_pipe_upcall() function
  SUNRPC: rework cache upcall logic
  SUNRPC: introduce cache_detail->cache_request callback
  NFS: simplify and clean cache library
  NFS: use SUNRPC cache creation and destruction helper for DNS cache
  nfsd4: free_stid can be static
  nfsd: keep a checksum of the first 256 bytes of request
  sunrpc: trim off trailing checksum before returning decrypted or integrity authenticated buffer
  sunrpc: fix comment in struct xdr_buf definition
  ...

fs/lockd/clntlock.c

@@ -11,7 +11,7 @@
 #include <linux/slab.h>
 #include <linux/time.h>
 #include <linux/nfs_fs.h>
-#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/svc.h>
 #include <linux/lockd/lockd.h>
 #include <linux/kthread.h>
@@ -220,10 +220,19 @@ reclaimer(void *ptr)
 {
 	struct nlm_host	  *host = (struct nlm_host *) ptr;
 	struct nlm_wait	  *block;
+	struct nlm_rqst   *req;
 	struct file_lock *fl, *next;
 	u32 nsmstate;
 	struct net *net = host->net;
 
+	req = kmalloc(sizeof(*req), GFP_KERNEL);
+	if (!req) {
+		printk(KERN_ERR "lockd: reclaimer unable to alloc memory."
+				" Locks for %s won't be reclaimed!\n",
+				host->h_name);
+		return 0;
+	}
+
 	allow_signal(SIGKILL);
 
 	down_write(&host->h_rwsem);
@@ -253,7 +262,7 @@ restart:
 		 */
 		if (signalled())
 			continue;
-		if (nlmclnt_reclaim(host, fl) != 0)
+		if (nlmclnt_reclaim(host, fl, req) != 0)
 			continue;
 		list_add_tail(&fl->fl_u.nfs_fl.list, &host->h_granted);
 		if (host->h_nsmstate != nsmstate) {
@@ -279,5 +288,6 @@ restart:
 	/* Release host handle after use */
 	nlmclnt_release_host(host);
 	lockd_down(net);
+	kfree(req);
 	return 0;
 }

fs/lockd/clntproc.c

@@ -618,17 +618,15 @@ out_unlock:
  * RECLAIM: Try to reclaim a lock
  */
 int
-nlmclnt_reclaim(struct nlm_host *host, struct file_lock *fl)
+nlmclnt_reclaim(struct nlm_host *host, struct file_lock *fl,
+		struct nlm_rqst *req)
 {
-	struct nlm_rqst reqst, *req;
 	int		status;
 
-	req = &reqst;
 	memset(req, 0, sizeof(*req));
 	locks_init_lock(&req->a_args.lock.fl);
 	locks_init_lock(&req->a_res.lock.fl);
 	req->a_host  = host;
-	req->a_flags = 0;
 
 	/* Set up the argument struct */
 	nlmclnt_setlockargs(req, fl);

fs/lockd/host.c

@@ -13,6 +13,7 @@
 #include <linux/in.h>
 #include <linux/in6.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/svc.h>
 #include <linux/lockd/lockd.h>
 #include <linux/mutex.h>

fs/lockd/mon.c

@@ -12,6 +12,7 @@
 #include <linux/slab.h>
 
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/xprtsock.h>
 #include <linux/sunrpc/svc.h>
 #include <linux/lockd/lockd.h>

fs/lockd/svcsubs.c

@@ -13,7 +13,7 @@
 #include <linux/slab.h>
 #include <linux/mutex.h>
 #include <linux/sunrpc/svc.h>
-#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/nfsd/nfsfh.h>
 #include <linux/nfsd/export.h>
 #include <linux/lockd/lockd.h>

fs/nfs/cache_lib.c

@@ -128,10 +128,13 @@ int nfs_cache_register_net(struct net *net, struct cache_detail *cd)
 	struct super_block *pipefs_sb;
 	int ret = 0;
 
+	sunrpc_init_cache_detail(cd);
 	pipefs_sb = rpc_get_sb_net(net);
 	if (pipefs_sb) {
 		ret = nfs_cache_register_sb(pipefs_sb, cd);
 		rpc_put_sb_net(net);
+		if (ret)
+			sunrpc_destroy_cache_detail(cd);
 	}
 	return ret;
 }
@@ -151,14 +154,5 @@ void nfs_cache_unregister_net(struct net *net, struct cache_detail *cd)
 		nfs_cache_unregister_sb(pipefs_sb, cd);
 		rpc_put_sb_net(net);
 	}
-}
-
-void nfs_cache_init(struct cache_detail *cd)
-{
-	sunrpc_init_cache_detail(cd);
-}
-
-void nfs_cache_destroy(struct cache_detail *cd)
-{
 	sunrpc_destroy_cache_detail(cd);
 }

fs/nfs/cache_lib.h

@@ -23,8 +23,6 @@ extern struct nfs_cache_defer_req *nfs_cache_defer_req_alloc(void);
 extern void nfs_cache_defer_req_put(struct nfs_cache_defer_req *dreq);
 extern int nfs_cache_wait_for_upcall(struct nfs_cache_defer_req *dreq);
 
-extern void nfs_cache_init(struct cache_detail *cd);
-extern void nfs_cache_destroy(struct cache_detail *cd);
 extern int nfs_cache_register_net(struct net *net, struct cache_detail *cd);
 extern void nfs_cache_unregister_net(struct net *net, struct cache_detail *cd);
 extern int nfs_cache_register_sb(struct super_block *sb,

fs/nfs/dns_resolve.c

@@ -10,6 +10,7 @@
 
 #include <linux/module.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/dns_resolver.h>
 #include "dns_resolve.h"
 
@@ -42,6 +43,7 @@ EXPORT_SYMBOL_GPL(nfs_dns_resolve_name);
 #include <linux/seq_file.h>
 #include <linux/inet.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/cache.h>
 #include <linux/sunrpc/svcauth.h>
 #include <linux/sunrpc/rpc_pipe_fs.h>
@@ -142,7 +144,7 @@ static int nfs_dns_upcall(struct cache_detail *cd,
 
 	ret = nfs_cache_upcall(cd, key->hostname);
 	if (ret)
-		ret = sunrpc_cache_pipe_upcall(cd, ch, nfs_dns_request);
+		ret = sunrpc_cache_pipe_upcall(cd, ch);
 	return ret;
 }
 
@@ -351,60 +353,47 @@ ssize_t nfs_dns_resolve_name(struct net *net, char *name,
 }
 EXPORT_SYMBOL_GPL(nfs_dns_resolve_name);
 
+static struct cache_detail nfs_dns_resolve_template = {
+	.owner		= THIS_MODULE,
+	.hash_size	= NFS_DNS_HASHTBL_SIZE,
+	.name		= "dns_resolve",
+	.cache_put	= nfs_dns_ent_put,
+	.cache_upcall	= nfs_dns_upcall,
+	.cache_request	= nfs_dns_request,
+	.cache_parse	= nfs_dns_parse,
+	.cache_show	= nfs_dns_show,
+	.match		= nfs_dns_match,
+	.init		= nfs_dns_ent_init,
+	.update		= nfs_dns_ent_update,
+	.alloc		= nfs_dns_ent_alloc,
+};
+
+
 int nfs_dns_resolver_cache_init(struct net *net)
 {
-	int err = -ENOMEM;
+	int err;
 	struct nfs_net *nn = net_generic(net, nfs_net_id);
-	struct cache_detail *cd;
-	struct cache_head **tbl;
 
-	cd = kzalloc(sizeof(struct cache_detail), GFP_KERNEL);
-	if (cd == NULL)
-		goto err_cd;
-
-	tbl = kzalloc(NFS_DNS_HASHTBL_SIZE * sizeof(struct cache_head *),
-			GFP_KERNEL);
-	if (tbl == NULL)
-		goto err_tbl;
-
-	cd->owner = THIS_MODULE,
-	cd->hash_size = NFS_DNS_HASHTBL_SIZE,
-	cd->hash_table = tbl,
-	cd->name = "dns_resolve",
-	cd->cache_put = nfs_dns_ent_put,
-	cd->cache_upcall = nfs_dns_upcall,
-	cd->cache_parse = nfs_dns_parse,
-	cd->cache_show = nfs_dns_show,
-	cd->match = nfs_dns_match,
-	cd->init = nfs_dns_ent_init,
-	cd->update = nfs_dns_ent_update,
-	cd->alloc = nfs_dns_ent_alloc,
-
-	nfs_cache_init(cd);
-	err = nfs_cache_register_net(net, cd);
+	nn->nfs_dns_resolve = cache_create_net(&nfs_dns_resolve_template, net);
+	if (IS_ERR(nn->nfs_dns_resolve))
+		return PTR_ERR(nn->nfs_dns_resolve);
+
+	err = nfs_cache_register_net(net, nn->nfs_dns_resolve);
 	if (err)
 		goto err_reg;
-	nn->nfs_dns_resolve = cd;
 	return 0;
 
 err_reg:
-	nfs_cache_destroy(cd);
-	kfree(cd->hash_table);
-err_tbl:
-	kfree(cd);
-err_cd:
+	cache_destroy_net(nn->nfs_dns_resolve, net);
 	return err;
 }
 
 void nfs_dns_resolver_cache_destroy(struct net *net)
 {
 	struct nfs_net *nn = net_generic(net, nfs_net_id);
-	struct cache_detail *cd = nn->nfs_dns_resolve;
 
-	nfs_cache_unregister_net(net, cd);
-	nfs_cache_destroy(cd);
-	kfree(cd->hash_table);
-	kfree(cd);
+	nfs_cache_unregister_net(net, nn->nfs_dns_resolve);
+	cache_destroy_net(nn->nfs_dns_resolve, net);
 }
 
 static int rpc_pipefs_event(struct notifier_block *nb, unsigned long event,

fs/nfs/nfs4client.c

@@ -6,6 +6,7 @@
 #include <linux/nfs_fs.h>
 #include <linux/nfs_idmap.h>
 #include <linux/nfs_mount.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/auth.h>
 #include <linux/sunrpc/xprt.h>
 #include <linux/sunrpc/bc_xprt.h>

fs/nfs/nfs4filelayoutdev.c

@@ -31,6 +31,7 @@
 #include <linux/nfs_fs.h>
 #include <linux/vmalloc.h>
 #include <linux/module.h>
+#include <linux/sunrpc/addr.h>
 
 #include "internal.h"
 #include "nfs4session.h"

fs/nfs/nfs4namespace.c

@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/string.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/vfs.h>
 #include <linux/inet.h>
 #include "internal.h"

fs/nfs/super.c

@@ -31,6 +31,7 @@
 #include <linux/errno.h>
 #include <linux/unistd.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/stats.h>
 #include <linux/sunrpc/metrics.h>
 #include <linux/sunrpc/xprtsock.h>

fs/nfsd/cache.h

@@ -12,6 +12,10 @@
 
 /*
  * Representation of a reply cache entry.
+ *
+ * Note that we use a sockaddr_in6 to hold the address instead of the more
+ * typical sockaddr_storage. This is for space reasons, since sockaddr_storage
+ * is much larger than a sockaddr_in6.
  */
 struct svc_cacherep {
 	struct hlist_node	c_hash;
@@ -20,11 +24,13 @@ struct svc_cacherep {
 	unsigned char		c_state,	/* unused, inprog, done */
 				c_type,		/* status, buffer */
 				c_secure : 1;	/* req came from port < 1024 */
-	struct sockaddr_in	c_addr;
+	struct sockaddr_in6	c_addr;
 	__be32			c_xid;
 	u32			c_prot;
 	u32			c_proc;
 	u32			c_vers;
+	unsigned int		c_len;
+	__wsum			c_csum;
 	unsigned long		c_timestamp;
 	union {
 		struct kvec	u_vec;
@@ -46,8 +52,7 @@ enum {
 enum {
 	RC_DROPIT,
 	RC_REPLY,
-	RC_DOIT,
-	RC_INTR
+	RC_DOIT
 };
 
 /*
@@ -67,6 +72,12 @@ enum {
  */
 #define RC_DELAY		(HZ/5)
 
+/* Cache entries expire after this time period */
+#define RC_EXPIRE		(120 * HZ)
+
+/* Checksum this amount of the request */
+#define RC_CSUMLEN		(256U)
+
 int	nfsd_reply_cache_init(void);
 void	nfsd_reply_cache_shutdown(void);
 int	nfsd_cache_lookup(struct svc_rqst *);

fs/nfsd/export.c

@@ -67,11 +67,6 @@ static void expkey_request(struct cache_detail *cd,
 	(*bpp)[-1] = '\n';
 }
 
-static int expkey_upcall(struct cache_detail *cd, struct cache_head *h)
-{
-	return sunrpc_cache_pipe_upcall(cd, h, expkey_request);
-}
-
 static struct svc_expkey *svc_expkey_update(struct cache_detail *cd, struct svc_expkey *new,
 					    struct svc_expkey *old);
 static struct svc_expkey *svc_expkey_lookup(struct cache_detail *cd, struct svc_expkey *);
@@ -245,7 +240,7 @@ static struct cache_detail svc_expkey_cache_template = {
 	.hash_size	= EXPKEY_HASHMAX,
 	.name		= "nfsd.fh",
 	.cache_put	= expkey_put,
-	.cache_upcall	= expkey_upcall,
+	.cache_request	= expkey_request,
 	.cache_parse	= expkey_parse,
 	.cache_show	= expkey_show,
 	.match		= expkey_match,
@@ -315,6 +310,7 @@ static void svc_export_put(struct kref *ref)
 	path_put(&exp->ex_path);
 	auth_domain_put(exp->ex_client);
 	nfsd4_fslocs_free(&exp->ex_fslocs);
+	kfree(exp->ex_uuid);
 	kfree(exp);
 }
 
@@ -337,11 +333,6 @@ static void svc_export_request(struct cache_detail *cd,
 	(*bpp)[-1] = '\n';
 }
 
-static int svc_export_upcall(struct cache_detail *cd, struct cache_head *h)
-{
-	return sunrpc_cache_pipe_upcall(cd, h, svc_export_request);
-}
-
 static struct svc_export *svc_export_update(struct svc_export *new,
 					    struct svc_export *old);
 static struct svc_export *svc_export_lookup(struct svc_export *);
@@ -674,6 +665,7 @@ static void svc_export_init(struct cache_head *cnew, struct cache_head *citem)
 	new->ex_fslocs.locations = NULL;
 	new->ex_fslocs.locations_count = 0;
 	new->ex_fslocs.migrated = 0;
+	new->ex_uuid = NULL;
 	new->cd = item->cd;
 }
 
@@ -715,7 +707,7 @@ static struct cache_detail svc_export_cache_template = {
 	.hash_size	= EXPORT_HASHMAX,
 	.name		= "nfsd.export",
 	.cache_put	= svc_export_put,
-	.cache_upcall	= svc_export_upcall,
+	.cache_request	= svc_export_request,
 	.cache_parse	= svc_export_parse,
 	.cache_show	= svc_export_show,
 	.match		= svc_export_match,

fs/nfsd/fault_inject.c

@@ -9,7 +9,7 @@
 #include <linux/debugfs.h>
 #include <linux/module.h>
 #include <linux/nsproxy.h>
-#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <asm/uaccess.h>
 
 #include "state.h"

fs/nfsd/nfs4idmap.c

@@ -139,12 +139,6 @@ idtoname_request(struct cache_detail *cd, struct cache_head *ch, char **bpp,
 	(*bpp)[-1] = '\n';
 }
 
-static int
-idtoname_upcall(struct cache_detail *cd, struct cache_head *ch)
-{
-	return sunrpc_cache_pipe_upcall(cd, ch, idtoname_request);
-}
-
 static int
 idtoname_match(struct cache_head *ca, struct cache_head *cb)
 {
@@ -192,7 +186,7 @@ static struct cache_detail idtoname_cache_template = {
 	.hash_size	= ENT_HASHMAX,
 	.name		= "nfs4.idtoname",
 	.cache_put	= ent_put,
-	.cache_upcall	= idtoname_upcall,
+	.cache_request	= idtoname_request,
 	.cache_parse	= idtoname_parse,
 	.cache_show	= idtoname_show,
 	.warn_no_listener = warn_no_idmapd,
@@ -320,12 +314,6 @@ nametoid_request(struct cache_detail *cd, struct cache_head *ch, char **bpp,
 	(*bpp)[-1] = '\n';
 }
 
-static int
-nametoid_upcall(struct cache_detail *cd, struct cache_head *ch)
-{
-	return sunrpc_cache_pipe_upcall(cd, ch, nametoid_request);
-}
-
 static int
 nametoid_match(struct cache_head *ca, struct cache_head *cb)
 {
@@ -365,7 +353,7 @@ static struct cache_detail nametoid_cache_template = {
 	.hash_size	= ENT_HASHMAX,
 	.name		= "nfs4.nametoid",
 	.cache_put	= ent_put,
-	.cache_upcall	= nametoid_upcall,
+	.cache_request	= nametoid_request,
 	.cache_parse	= nametoid_parse,
 	.cache_show	= nametoid_show,
 	.warn_no_listener = warn_no_idmapd,

fs/nfsd/nfs4proc.c

@@ -993,14 +993,15 @@ _nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (!buf)
 		return nfserr_jukebox;
 
+	p = buf;
 	status = nfsd4_encode_fattr(&cstate->current_fh,
 				    cstate->current_fh.fh_export,
-				    cstate->current_fh.fh_dentry, buf,
-				    &count, verify->ve_bmval,
+				    cstate->current_fh.fh_dentry, &p,
+				    count, verify->ve_bmval,
 				    rqstp, 0);
 
 	/* this means that nfsd4_encode_fattr() ran out of space */
-	if (status == nfserr_resource && count == 0)
+	if (status == nfserr_resource)
 		status = nfserr_not_same;
 	if (status)
 		goto out_kfree;

fs/nfsd/nfs4recover.c

@@ -1185,6 +1185,12 @@ bin_to_hex_dup(const unsigned char *src, int srclen)
 static int
 nfsd4_umh_cltrack_init(struct net __attribute__((unused)) *net)
 {
+	/* XXX: The usermode helper s not working in container yet. */
+	if (net != &init_net) {
+		WARN(1, KERN_ERR "NFSD: attempt to initialize umh client "
+			"tracking in a container!\n");
+		return -EINVAL;
+	}
 	return nfsd4_umh_cltrack_upcall("init", NULL, NULL);
 }
 

fs/nfsd/nfs4state.c

@@ -40,7 +40,7 @@
 #include <linux/pagemap.h>
 #include <linux/ratelimit.h>
 #include <linux/sunrpc/svcauth_gss.h>
-#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include "xdr4.h"
 #include "vfs.h"
 #include "current_stateid.h"
@@ -261,33 +261,46 @@ static inline int get_new_stid(struct nfs4_stid *stid)
 	return new_stid;
 }
 
-static void init_stid(struct nfs4_stid *stid, struct nfs4_client *cl, unsigned char type)
+static struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, struct
+kmem_cache *slab)
 {
-	stateid_t *s = &stid->sc_stateid;
+	struct idr *stateids = &cl->cl_stateids;
+	static int min_stateid = 0;
+	struct nfs4_stid *stid;
 	int new_id;
 
-	stid->sc_type = type;
+	stid = kmem_cache_alloc(slab, GFP_KERNEL);
+	if (!stid)
+		return NULL;
+
+	if (!idr_pre_get(stateids, GFP_KERNEL))
+		goto out_free;
+	if (idr_get_new_above(stateids, stid, min_stateid, &new_id))
+		goto out_free;
 	stid->sc_client = cl;
-	s->si_opaque.so_clid = cl->cl_clientid;
-	new_id = get_new_stid(stid);
-	s->si_opaque.so_id = (u32)new_id;
+	stid->sc_type = 0;
+	stid->sc_stateid.si_opaque.so_id = new_id;
+	stid->sc_stateid.si_opaque.so_clid = cl->cl_clientid;
 	/* Will be incremented before return to client: */
-	s->si_generation = 0;
-}
-
-static struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, struct kmem_cache *slab)
-{
-	struct idr *stateids = &cl->cl_stateids;
+	stid->sc_stateid.si_generation = 0;
 
-	if (!idr_pre_get(stateids, GFP_KERNEL))
-		return NULL;
 	/*
-	 * Note: if we fail here (or any time between now and the time
-	 * we actually get the new idr), we won't need to undo the idr
-	 * preallocation, since the idr code caps the number of
-	 * preallocated entries.
+	 * It shouldn't be a problem to reuse an opaque stateid value.
+	 * I don't think it is for 4.1.  But with 4.0 I worry that, for
+	 * example, a stray write retransmission could be accepted by
+	 * the server when it should have been rejected.  Therefore,
+	 * adopt a trick from the sctp code to attempt to maximize the
+	 * amount of time until an id is reused, by ensuring they always
+	 * "increase" (mod INT_MAX):
 	 */
-	return kmem_cache_alloc(slab, GFP_KERNEL);
+
+	min_stateid = new_id+1;
+	if (min_stateid == INT_MAX)
+		min_stateid = 0;
+	return stid;
+out_free:
+	kfree(stid);
+	return NULL;
 }
 
 static struct nfs4_ol_stateid * nfs4_alloc_stateid(struct nfs4_client *clp)
@@ -316,7 +329,7 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_ol_stateid *stp, struct sv
 	dp = delegstateid(nfs4_alloc_stid(clp, deleg_slab));
 	if (dp == NULL)
 		return dp;
-	init_stid(&dp->dl_stid, clp, NFS4_DELEG_STID);
+	dp->dl_stid.sc_type = NFS4_DELEG_STID;
 	/*
 	 * delegation seqid's are never incremented.  The 4.1 special
 	 * meaning of seqid 0 isn't meaningful, really, but let's avoid
@@ -337,13 +350,21 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_ol_stateid *stp, struct sv
 	return dp;
 }
 
+static void free_stid(struct nfs4_stid *s, struct kmem_cache *slab)
+{
+	struct idr *stateids = &s->sc_client->cl_stateids;
+
+	idr_remove(stateids, s->sc_stateid.si_opaque.so_id);
+	kmem_cache_free(slab, s);
+}
+
 void
 nfs4_put_delegation(struct nfs4_delegation *dp)
 {
 	if (atomic_dec_and_test(&dp->dl_count)) {
 		dprintk("NFSD: freeing dp %p\n",dp);
 		put_nfs4_file(dp->dl_file);
-		kmem_cache_free(deleg_slab, dp);
+		free_stid(&dp->dl_stid, deleg_slab);
 		num_delegations--;
 	}
 }
@@ -360,9 +381,7 @@ static void nfs4_put_deleg_lease(struct nfs4_file *fp)
 
 static void unhash_stid(struct nfs4_stid *s)
 {
-	struct idr *stateids = &s->sc_client->cl_stateids;
-
-	idr_remove(stateids, s->sc_stateid.si_opaque.so_id);
+	s->sc_type = 0;
 }
 
 /* Called under the state lock. */
@@ -519,7 +538,7 @@ static void close_generic_stateid(struct nfs4_ol_stateid *stp)
 
 static void free_generic_stateid(struct nfs4_ol_stateid *stp)
 {
-	kmem_cache_free(stateid_slab, stp);
+	free_stid(&stp->st_stid, stateid_slab);
 }
 
 static void release_lock_stateid(struct nfs4_ol_stateid *stp)
@@ -905,7 +924,7 @@ static struct nfsd4_session *alloc_session(struct nfsd4_channel_attrs *fchan,
 
 	new = __alloc_session(slotsize, numslots);
 	if (!new) {
-		nfsd4_put_drc_mem(slotsize, fchan->maxreqs);
+		nfsd4_put_drc_mem(slotsize, numslots);
 		return NULL;
 	}
 	init_forechannel_attrs(&new->se_fchannel, fchan, numslots, slotsize, nn);
@@ -1048,7 +1067,7 @@ static struct nfs4_client *alloc_client(struct xdr_netobj name)
 static inline void
 free_client(struct nfs4_client *clp)
 {
-	struct nfsd_net *nn = net_generic(clp->net, nfsd_net_id);
+	struct nfsd_net __maybe_unused *nn = net_generic(clp->net, nfsd_net_id);
 
 	lockdep_assert_held(&nn->client_lock);
 	while (!list_empty(&clp->cl_sessions)) {
@@ -1060,6 +1079,7 @@ free_client(struct nfs4_client *clp)
 	}
 	free_svc_cred(&clp->cl_cred);
 	kfree(clp->cl_name.data);
+	idr_destroy(&clp->cl_stateids);
 	kfree(clp);
 }
 
@@ -1258,7 +1278,12 @@ static void gen_confirm(struct nfs4_client *clp)
 
 static struct nfs4_stid *find_stateid(struct nfs4_client *cl, stateid_t *t)
 {
-	return idr_find(&cl->cl_stateids, t->si_opaque.so_id);
+	struct nfs4_stid *ret;
+
+	ret = idr_find(&cl->cl_stateids, t->si_opaque.so_id);
+	if (!ret || !ret->sc_type)
+		return NULL;
+	return ret;
 }
 
 static struct nfs4_stid *find_stateid_by_type(struct nfs4_client *cl, stateid_t *t, char typemask)
@@ -1844,11 +1869,12 @@ nfsd4_create_session(struct svc_rqst *rqstp,
 
 	/* cache solo and embedded create sessions under the state lock */
 	nfsd4_cache_create_session(cr_ses, cs_slot, status);
-out:
 	nfs4_unlock_state();
+out:
 	dprintk("%s returns %d\n", __func__, ntohl(status));
 	return status;
 out_free_conn:
+	nfs4_unlock_state();
 	free_conn(conn);
 out_free_session:
 	__free_session(new);
@@ -2443,9 +2469,8 @@ alloc_init_open_stateowner(unsigned int strhashval, struct nfs4_client *clp, str
 
 static void init_open_stateid(struct nfs4_ol_stateid *stp, struct nfs4_file *fp, struct nfsd4_open *open) {
 	struct nfs4_openowner *oo = open->op_openowner;
-	struct nfs4_client *clp = oo->oo_owner.so_client;
 
-	init_stid(&stp->st_stid, clp, NFS4_OPEN_STID);
+	stp->st_stid.sc_type = NFS4_OPEN_STID;
 	INIT_LIST_HEAD(&stp->st_lockowners);
 	list_add(&stp->st_perstateowner, &oo->oo_owner.so_stateids);
 	list_add(&stp->st_perfile, &fp->fi_stateids);
@@ -4031,7 +4056,7 @@ alloc_init_lock_stateid(struct nfs4_lockowner *lo, struct nfs4_file *fp, struct
 	stp = nfs4_alloc_stateid(clp);
 	if (stp == NULL)
 		return NULL;
-	init_stid(&stp->st_stid, clp, NFS4_LOCK_STID);
+	stp->st_stid.sc_type = NFS4_LOCK_STID;
 	list_add(&stp->st_perfile, &fp->fi_stateids);
 	list_add(&stp->st_perstateowner, &lo->lo_owner.so_stateids);
 	stp->st_stateowner = &lo->lo_owner;
@@ -4913,16 +4938,6 @@ nfs4_state_start_net(struct net *net)
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 	int ret;
 
-	/*
-	 * FIXME: For now, we hang most of the pernet global stuff off of
-	 * init_net until nfsd is fully containerized. Eventually, we'll
-	 * need to pass a net pointer into this function, take a reference
-	 * to that instead and then do most of the rest of this on a per-net
-	 * basis.
-	 */
-	if (net != &init_net)
-		return -EINVAL;
-
 	ret = nfs4_state_create_net(net);
 	if (ret)
 		return ret;

fs/nfsd/nfs4xdr.c

@@ -2024,12 +2024,11 @@ static int get_parent_attributes(struct svc_export *exp, struct kstat *stat)
  * Note: @fhp can be NULL; in this case, we might have to compose the filehandle
  * ourselves.
  *
- * @countp is the buffer size in _words_; upon successful return this becomes
- * replaced with the number of words written.
+ * countp is the buffer size in _words_
  */
 __be32
 nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp,
-		struct dentry *dentry, __be32 *buffer, int *countp, u32 *bmval,
+		struct dentry *dentry, __be32 **buffer, int count, u32 *bmval,
 		struct svc_rqst *rqstp, int ignore_crossmnt)
 {
 	u32 bmval0 = bmval[0];
@@ -2038,12 +2037,12 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp,
 	struct kstat stat;
 	struct svc_fh tempfh;
 	struct kstatfs statfs;
-	int buflen = *countp << 2;
+	int buflen = count << 2;
 	__be32 *attrlenp;
 	u32 dummy;
 	u64 dummy64;
 	u32 rdattr_err = 0;
-	__be32 *p = buffer;
+	__be32 *p = *buffer;
 	__be32 status;
 	int err;
 	int aclsupport = 0;
@@ -2447,7 +2446,7 @@ out_acl:
 	}
 
 	*attrlenp = htonl((char *)p - (char *)attrlenp - 4);
-	*countp = p - buffer;
+	*buffer = p;
 	status = nfs_ok;
 
 out:
@@ -2459,7 +2458,6 @@ out_nfserr:
 	status = nfserrno(err);
 	goto out;
 out_resource:
-	*countp = 0;
 	status = nfserr_resource;
 	goto out;
 out_serverfault:
@@ -2478,7 +2476,7 @@ static inline int attributes_need_mount(u32 *bmval)
 
 static __be32
 nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd,
-		const char *name, int namlen, __be32 *p, int *buflen)
+		const char *name, int namlen, __be32 **p, int buflen)
 {
 	struct svc_export *exp = cd->rd_fhp->fh_export;
 	struct dentry *dentry;
@@ -2584,10 +2582,9 @@ nfsd4_encode_dirent(void *ccdv, const char *name, int namlen,
 	p = xdr_encode_hyper(p, NFS_OFFSET_MAX);    /* offset of next entry */
 	p = xdr_encode_array(p, name, namlen);      /* name length & name */
 
-	nfserr = nfsd4_encode_dirent_fattr(cd, name, namlen, p, &buflen);
+	nfserr = nfsd4_encode_dirent_fattr(cd, name, namlen, &p, buflen);
 	switch (nfserr) {
 	case nfs_ok:
-		p += buflen;
 		break;
 	case nfserr_resource:
 		nfserr = nfserr_toosmall;
@@ -2714,10 +2711,8 @@ nfsd4_encode_getattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4
 
 	buflen = resp->end - resp->p - (COMPOUND_ERR_SLACK_SPACE >> 2);
 	nfserr = nfsd4_encode_fattr(fhp, fhp->fh_export, fhp->fh_dentry,
-				    resp->p, &buflen, getattr->ga_bmval,
+				    &resp->p, buflen, getattr->ga_bmval,
 				    resp->rqstp, 0);
-	if (!nfserr)
-		resp->p += buflen;
 	return nfserr;
 }
 

fs/nfsd/nfscache.c

@@ -9,22 +9,22 @@
  */
 
 #include <linux/slab.h>
+#include <linux/sunrpc/addr.h>
+#include <linux/highmem.h>
+#include <net/checksum.h>
 
 #include "nfsd.h"
 #include "cache.h"
 
-/* Size of reply cache. Common values are:
- * 4.3BSD:	128
- * 4.4BSD:	256
- * Solaris2:	1024
- * DEC Unix:	512-4096
- */
-#define CACHESIZE		1024
+#define NFSDDBG_FACILITY	NFSDDBG_REPCACHE
+
 #define HASHSIZE		64
 
 static struct hlist_head *	cache_hash;
 static struct list_head 	lru_head;
-static int			cache_disabled = 1;
+static struct kmem_cache	*drc_slab;
+static unsigned int		num_drc_entries;
+static unsigned int		max_drc_entries;
 
 /*
  * Calculate the hash index from an XID.
@@ -37,6 +37,14 @@ static inline u32 request_hash(u32 xid)
 }
 
 static int	nfsd_cache_append(struct svc_rqst *rqstp, struct kvec *vec);
+static void	cache_cleaner_func(struct work_struct *unused);
+static int 	nfsd_reply_cache_shrink(struct shrinker *shrink,
+					struct shrink_control *sc);
+
+struct shrinker nfsd_reply_cache_shrinker = {
+	.shrink	= nfsd_reply_cache_shrink,
+	.seeks	= 1,
+};
 
 /*
  * locking for the reply cache:
@@ -44,30 +52,86 @@ static int	nfsd_cache_append(struct svc_rqst *rqstp, struct kvec *vec);
  * Otherwise, it when accessing _prev or _next, the lock must be held.
  */
 static DEFINE_SPINLOCK(cache_lock);
+static DECLARE_DELAYED_WORK(cache_cleaner, cache_cleaner_func);
 
-int nfsd_reply_cache_init(void)
+/*
+ * Put a cap on the size of the DRC based on the amount of available
+ * low memory in the machine.
+ *
+ *  64MB:    8192
+ * 128MB:   11585
+ * 256MB:   16384
+ * 512MB:   23170
+ *   1GB:   32768
+ *   2GB:   46340
+ *   4GB:   65536
+ *   8GB:   92681
+ *  16GB:  131072
+ *
+ * ...with a hard cap of 256k entries. In the worst case, each entry will be
+ * ~1k, so the above numbers should give a rough max of the amount of memory
+ * used in k.
+ */
+static unsigned int
+nfsd_cache_size_limit(void)
+{
+	unsigned int limit;
+	unsigned long low_pages = totalram_pages - totalhigh_pages;
+
+	limit = (16 * int_sqrt(low_pages)) << (PAGE_SHIFT-10);
+	return min_t(unsigned int, limit, 256*1024);
+}
+
+static struct svc_cacherep *
+nfsd_reply_cache_alloc(void)
 {
 	struct svc_cacherep	*rp;
-	int			i;
 
-	INIT_LIST_HEAD(&lru_head);
-	i = CACHESIZE;
-	while (i) {
-		rp = kmalloc(sizeof(*rp), GFP_KERNEL);
-		if (!rp)
-			goto out_nomem;
-		list_add(&rp->c_lru, &lru_head);
+	rp = kmem_cache_alloc(drc_slab, GFP_KERNEL);
+	if (rp) {
 		rp->c_state = RC_UNUSED;
 		rp->c_type = RC_NOCACHE;
+		INIT_LIST_HEAD(&rp->c_lru);
 		INIT_HLIST_NODE(&rp->c_hash);
-		i--;
 	}
+	return rp;
+}
 
-	cache_hash = kcalloc (HASHSIZE, sizeof(struct hlist_head), GFP_KERNEL);
+static void
+nfsd_reply_cache_free_locked(struct svc_cacherep *rp)
+{
+	if (rp->c_type == RC_REPLBUFF)
+		kfree(rp->c_replvec.iov_base);
+	hlist_del(&rp->c_hash);
+	list_del(&rp->c_lru);
+	--num_drc_entries;
+	kmem_cache_free(drc_slab, rp);
+}
+
+static void
+nfsd_reply_cache_free(struct svc_cacherep *rp)
+{
+	spin_lock(&cache_lock);
+	nfsd_reply_cache_free_locked(rp);
+	spin_unlock(&cache_lock);
+}
+
+int nfsd_reply_cache_init(void)
+{
+	register_shrinker(&nfsd_reply_cache_shrinker);
+	drc_slab = kmem_cache_create("nfsd_drc", sizeof(struct svc_cacherep),
+					0, 0, NULL);
+	if (!drc_slab)
+		goto out_nomem;
+
+	cache_hash = kcalloc(HASHSIZE, sizeof(struct hlist_head), GFP_KERNEL);
 	if (!cache_hash)
 		goto out_nomem;
 
-	cache_disabled = 0;
+	INIT_LIST_HEAD(&lru_head);
+	max_drc_entries = nfsd_cache_size_limit();
+	num_drc_entries = 0;
+
 	return 0;
 out_nomem:
 	printk(KERN_ERR "nfsd: failed to allocate reply cache\n");
@@ -79,27 +143,33 @@ void nfsd_reply_cache_shutdown(void)
 {
 	struct svc_cacherep	*rp;
 
+	unregister_shrinker(&nfsd_reply_cache_shrinker);
+	cancel_delayed_work_sync(&cache_cleaner);
+
 	while (!list_empty(&lru_head)) {
 		rp = list_entry(lru_head.next, struct svc_cacherep, c_lru);
-		if (rp->c_state == RC_DONE && rp->c_type == RC_REPLBUFF)
-			kfree(rp->c_replvec.iov_base);
-		list_del(&rp->c_lru);
-		kfree(rp);
+		nfsd_reply_cache_free_locked(rp);
 	}
 
-	cache_disabled = 1;
-
 	kfree (cache_hash);
 	cache_hash = NULL;
+
+	if (drc_slab) {
+		kmem_cache_destroy(drc_slab);
+		drc_slab = NULL;
+	}
 }
 
 /*
- * Move cache entry to end of LRU list
+ * Move cache entry to end of LRU list, and queue the cleaner to run if it's
+ * not already scheduled.
  */
 static void
 lru_put_end(struct svc_cacherep *rp)
 {
+	rp->c_timestamp = jiffies;
 	list_move_tail(&rp->c_lru, &lru_head);
+	schedule_delayed_work(&cache_cleaner, RC_EXPIRE);
 }
 
 /*
@@ -112,82 +182,214 @@ hash_refile(struct svc_cacherep *rp)
 	hlist_add_head(&rp->c_hash, cache_hash + request_hash(rp->c_xid));
 }
 
+static inline bool
+nfsd_cache_entry_expired(struct svc_cacherep *rp)
+{
+	return rp->c_state != RC_INPROG &&
+	       time_after(jiffies, rp->c_timestamp + RC_EXPIRE);
+}
+
+/*
+ * Walk the LRU list and prune off entries that are older than RC_EXPIRE.
+ * Also prune the oldest ones when the total exceeds the max number of entries.
+ */
+static void
+prune_cache_entries(void)
+{
+	struct svc_cacherep *rp, *tmp;
+
+	list_for_each_entry_safe(rp, tmp, &lru_head, c_lru) {
+		if (!nfsd_cache_entry_expired(rp) &&
+		    num_drc_entries <= max_drc_entries)
+			break;
+		nfsd_reply_cache_free_locked(rp);
+	}
+
+	/*
+	 * Conditionally rearm the job. If we cleaned out the list, then
+	 * cancel any pending run (since there won't be any work to do).
+	 * Otherwise, we rearm the job or modify the existing one to run in
+	 * RC_EXPIRE since we just ran the pruner.
+	 */
+	if (list_empty(&lru_head))
+		cancel_delayed_work(&cache_cleaner);
+	else
+		mod_delayed_work(system_wq, &cache_cleaner, RC_EXPIRE);
+}
+
+static void
+cache_cleaner_func(struct work_struct *unused)
+{
+	spin_lock(&cache_lock);
+	prune_cache_entries();
+	spin_unlock(&cache_lock);
+}
+
+static int
+nfsd_reply_cache_shrink(struct shrinker *shrink, struct shrink_control *sc)
+{
+	unsigned int num;
+
+	spin_lock(&cache_lock);
+	if (sc->nr_to_scan)
+		prune_cache_entries();
+	num = num_drc_entries;
+	spin_unlock(&cache_lock);
+
+	return num;
+}
+
+/*
+ * Walk an xdr_buf and get a CRC for at most the first RC_CSUMLEN bytes
+ */
+static __wsum
+nfsd_cache_csum(struct svc_rqst *rqstp)
+{
+	int idx;
+	unsigned int base;
+	__wsum csum;
+	struct xdr_buf *buf = &rqstp->rq_arg;
+	const unsigned char *p = buf->head[0].iov_base;
+	size_t csum_len = min_t(size_t, buf->head[0].iov_len + buf->page_len,
+				RC_CSUMLEN);
+	size_t len = min(buf->head[0].iov_len, csum_len);
+
+	/* rq_arg.head first */
+	csum = csum_partial(p, len, 0);
+	csum_len -= len;
+
+	/* Continue into page array */
+	idx = buf->page_base / PAGE_SIZE;
+	base = buf->page_base & ~PAGE_MASK;
+	while (csum_len) {
+		p = page_address(buf->pages[idx]) + base;
+		len = min_t(size_t, PAGE_SIZE - base, csum_len);
+		csum = csum_partial(p, len, csum);
+		csum_len -= len;
+		base = 0;
+		++idx;
+	}
+	return csum;
+}
+
+/*
+ * Search the request hash for an entry that matches the given rqstp.
+ * Must be called with cache_lock held. Returns the found entry or
+ * NULL on failure.
+ */
+static struct svc_cacherep *
+nfsd_cache_search(struct svc_rqst *rqstp, __wsum csum)
+{
+	struct svc_cacherep	*rp;
+	struct hlist_head 	*rh;
+	__be32			xid = rqstp->rq_xid;
+	u32			proto =  rqstp->rq_prot,
+				vers = rqstp->rq_vers,
+				proc = rqstp->rq_proc;
+
+	rh = &cache_hash[request_hash(xid)];
+	hlist_for_each_entry(rp, rh, c_hash) {
+		if (xid == rp->c_xid && proc == rp->c_proc &&
+		    proto == rp->c_prot && vers == rp->c_vers &&
+		    rqstp->rq_arg.len == rp->c_len && csum == rp->c_csum &&
+		    rpc_cmp_addr(svc_addr(rqstp), (struct sockaddr *)&rp->c_addr) &&
+		    rpc_get_port(svc_addr(rqstp)) == rpc_get_port((struct sockaddr *)&rp->c_addr))
+			return rp;
+	}
+	return NULL;
+}
+
 /*
  * Try to find an entry matching the current call in the cache. When none
- * is found, we grab the oldest unlocked entry off the LRU list.
- * Note that no operation within the loop may sleep.
+ * is found, we try to grab the oldest expired entry off the LRU list. If
+ * a suitable one isn't there, then drop the cache_lock and allocate a
+ * new one, then search again in case one got inserted while this thread
+ * didn't hold the lock.
  */
 int
 nfsd_cache_lookup(struct svc_rqst *rqstp)
 {
-	struct hlist_head 	*rh;
-	struct svc_cacherep	*rp;
+	struct svc_cacherep	*rp, *found;
 	__be32			xid = rqstp->rq_xid;
 	u32			proto =  rqstp->rq_prot,
 				vers = rqstp->rq_vers,
 				proc = rqstp->rq_proc;
+	__wsum			csum;
 	unsigned long		age;
 	int type = rqstp->rq_cachetype;
 	int rtn;
 
 	rqstp->rq_cacherep = NULL;
-	if (cache_disabled || type == RC_NOCACHE) {
+	if (type == RC_NOCACHE) {
 		nfsdstats.rcnocache++;
 		return RC_DOIT;
 	}
 
+	csum = nfsd_cache_csum(rqstp);
+
 	spin_lock(&cache_lock);
 	rtn = RC_DOIT;
 
-	rh = &cache_hash[request_hash(xid)];
-	hlist_for_each_entry(rp, rh, c_hash) {
-		if (rp->c_state != RC_UNUSED &&
-		    xid == rp->c_xid && proc == rp->c_proc &&
-		    proto == rp->c_prot && vers == rp->c_vers &&
-		    time_before(jiffies, rp->c_timestamp + 120*HZ) &&
-		    memcmp((char*)&rqstp->rq_addr, (char*)&rp->c_addr, sizeof(rp->c_addr))==0) {
-			nfsdstats.rchits++;
-			goto found_entry;
+	rp = nfsd_cache_search(rqstp, csum);
+	if (rp)
+		goto found_entry;
+
+	/* Try to use the first entry on the LRU */
+	if (!list_empty(&lru_head)) {
+		rp = list_first_entry(&lru_head, struct svc_cacherep, c_lru);
+		if (nfsd_cache_entry_expired(rp) ||
+		    num_drc_entries >= max_drc_entries) {
+			lru_put_end(rp);
+			prune_cache_entries();
+			goto setup_entry;
 		}
 	}
-	nfsdstats.rcmisses++;
 
-	/* This loop shouldn't take more than a few iterations normally */
-	{
-	int	safe = 0;
-	list_for_each_entry(rp, &lru_head, c_lru) {
-		if (rp->c_state != RC_INPROG)
-			break;
-		if (safe++ > CACHESIZE) {
-			printk("nfsd: loop in repcache LRU list\n");
-			cache_disabled = 1;
-			goto out;
-		}
+	/* Drop the lock and allocate a new entry */
+	spin_unlock(&cache_lock);
+	rp = nfsd_reply_cache_alloc();
+	if (!rp) {
+		dprintk("nfsd: unable to allocate DRC entry!\n");
+		return RC_DOIT;
 	}
+	spin_lock(&cache_lock);
+	++num_drc_entries;
+
+	/*
+	 * Must search again just in case someone inserted one
+	 * after we dropped the lock above.
+	 */
+	found = nfsd_cache_search(rqstp, csum);
+	if (found) {
+		nfsd_reply_cache_free_locked(rp);
+		rp = found;
+		goto found_entry;
 	}
 
-	/* All entries on the LRU are in-progress. This should not happen */
-	if (&rp->c_lru == &lru_head) {
-		static int	complaints;
-
-		printk(KERN_WARNING "nfsd: all repcache entries locked!\n");
-		if (++complaints > 5) {
-			printk(KERN_WARNING "nfsd: disabling repcache.\n");
-			cache_disabled = 1;
-		}
-		goto out;
-	}
+	/*
+	 * We're keeping the one we just allocated. Are we now over the
+	 * limit? Prune one off the tip of the LRU in trade for the one we
+	 * just allocated if so.
+	 */
+	if (num_drc_entries >= max_drc_entries)
+		nfsd_reply_cache_free_locked(list_first_entry(&lru_head,
+						struct svc_cacherep, c_lru));
 
+setup_entry:
+	nfsdstats.rcmisses++;
 	rqstp->rq_cacherep = rp;
 	rp->c_state = RC_INPROG;
 	rp->c_xid = xid;
 	rp->c_proc = proc;
-	memcpy(&rp->c_addr, svc_addr_in(rqstp), sizeof(rp->c_addr));
+	rpc_copy_addr((struct sockaddr *)&rp->c_addr, svc_addr(rqstp));
+	rpc_set_port((struct sockaddr *)&rp->c_addr, rpc_get_port(svc_addr(rqstp)));
 	rp->c_prot = proto;
 	rp->c_vers = vers;
-	rp->c_timestamp = jiffies;
+	rp->c_len = rqstp->rq_arg.len;
+	rp->c_csum = csum;
 
 	hash_refile(rp);
+	lru_put_end(rp);
 
 	/* release any buffer */
 	if (rp->c_type == RC_REPLBUFF) {
@@ -200,9 +402,9 @@ nfsd_cache_lookup(struct svc_rqst *rqstp)
 	return rtn;
 
 found_entry:
+	nfsdstats.rchits++;
 	/* We found a matching entry which is either in progress or done. */
 	age = jiffies - rp->c_timestamp;
-	rp->c_timestamp = jiffies;
 	lru_put_end(rp);
 
 	rtn = RC_DROPIT;
@@ -231,7 +433,7 @@ found_entry:
 		break;
 	default:
 		printk(KERN_WARNING "nfsd: bad repcache type %d\n", rp->c_type);
-		rp->c_state = RC_UNUSED;
+		nfsd_reply_cache_free_locked(rp);
 	}
 
 	goto out;
@@ -256,11 +458,11 @@ found_entry:
 void
 nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, __be32 *statp)
 {
-	struct svc_cacherep *rp;
+	struct svc_cacherep *rp = rqstp->rq_cacherep;
 	struct kvec	*resv = &rqstp->rq_res.head[0], *cachv;
 	int		len;
 
-	if (!(rp = rqstp->rq_cacherep) || cache_disabled)
+	if (!rp)
 		return;
 
 	len = resv->iov_len - ((char*)statp - (char*)resv->iov_base);
@@ -268,7 +470,7 @@ nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, __be32 *statp)
 
 	/* Don't cache excessive amounts of data and XDR failures */
 	if (!statp || len > (256 >> 2)) {
-		rp->c_state = RC_UNUSED;
+		nfsd_reply_cache_free(rp);
 		return;
 	}
 
@@ -282,21 +484,21 @@ nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, __be32 *statp)
 		cachv = &rp->c_replvec;
 		cachv->iov_base = kmalloc(len << 2, GFP_KERNEL);
 		if (!cachv->iov_base) {
-			spin_lock(&cache_lock);
-			rp->c_state = RC_UNUSED;
-			spin_unlock(&cache_lock);
+			nfsd_reply_cache_free(rp);
 			return;
 		}
 		cachv->iov_len = len << 2;
 		memcpy(cachv->iov_base, statp, len << 2);
 		break;
+	case RC_NOCACHE:
+		nfsd_reply_cache_free(rp);
+		return;
 	}
 	spin_lock(&cache_lock);
 	lru_put_end(rp);
 	rp->c_secure = rqstp->rq_secure;
 	rp->c_type = cachetype;
 	rp->c_state = RC_DONE;
-	rp->c_timestamp = jiffies;
 	spin_unlock(&cache_lock);
 	return;
 }

fs/nfsd/nfsctl.c

@@ -10,7 +10,7 @@
 
 #include <linux/sunrpc/svcsock.h>
 #include <linux/lockd/lockd.h>
-#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/gss_api.h>
 #include <linux/sunrpc/gss_krb5_enctypes.h>
 #include <linux/sunrpc/rpc_pipe_fs.h>
@@ -125,11 +125,11 @@ static const struct file_operations transaction_ops = {
 	.llseek		= default_llseek,
 };
 
-static int exports_open(struct inode *inode, struct file *file)
+static int exports_net_open(struct net *net, struct file *file)
 {
 	int err;
 	struct seq_file *seq;
-	struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id);
+	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 
 	err = seq_open(file, &nfs_exports_op);
 	if (err)
@@ -140,8 +140,26 @@ static int exports_open(struct inode *inode, struct file *file)
 	return 0;
 }
 
-static const struct file_operations exports_operations = {
-	.open		= exports_open,
+static int exports_proc_open(struct inode *inode, struct file *file)
+{
+	return exports_net_open(current->nsproxy->net_ns, file);
+}
+
+static const struct file_operations exports_proc_operations = {
+	.open		= exports_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+	.owner		= THIS_MODULE,
+};
+
+static int exports_nfsd_open(struct inode *inode, struct file *file)
+{
+	return exports_net_open(inode->i_sb->s_fs_info, file);
+}
+
+static const struct file_operations exports_nfsd_operations = {
+	.open		= exports_nfsd_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
 	.release	= seq_release,
@@ -220,6 +238,7 @@ static ssize_t write_unlock_ip(struct file *file, char *buf, size_t size)
 	struct sockaddr *sap = (struct sockaddr *)&address;
 	size_t salen = sizeof(address);
 	char *fo_path;
+	struct net *net = file->f_dentry->d_sb->s_fs_info;
 
 	/* sanity check */
 	if (size == 0)
@@ -232,7 +251,7 @@ static ssize_t write_unlock_ip(struct file *file, char *buf, size_t size)
 	if (qword_get(&buf, fo_path, size) < 0)
 		return -EINVAL;
 
-	if (rpc_pton(&init_net, fo_path, size, sap, salen) == 0)
+	if (rpc_pton(net, fo_path, size, sap, salen) == 0)
 		return -EINVAL;
 
 	return nlmsvc_unlock_all_by_ip(sap);
@@ -317,6 +336,7 @@ static ssize_t write_filehandle(struct file *file, char *buf, size_t size)
 	int len;
 	struct auth_domain *dom;
 	struct knfsd_fh fh;
+	struct net *net = file->f_dentry->d_sb->s_fs_info;
 
 	if (size == 0)
 		return -EINVAL;
@@ -352,7 +372,7 @@ static ssize_t write_filehandle(struct file *file, char *buf, size_t size)
 	if (!dom)
 		return -ENOMEM;
 
-	len = exp_rootfh(&init_net, dom, path, &fh,  maxsize);
+	len = exp_rootfh(net, dom, path, &fh,  maxsize);
 	auth_domain_put(dom);
 	if (len)
 		return len;
@@ -396,7 +416,7 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size)
 {
 	char *mesg = buf;
 	int rv;
-	struct net *net = &init_net;
+	struct net *net = file->f_dentry->d_sb->s_fs_info;
 
 	if (size > 0) {
 		int newthreads;
@@ -447,7 +467,7 @@ static ssize_t write_pool_threads(struct file *file, char *buf, size_t size)
 	int len;
 	int npools;
 	int *nthreads;
-	struct net *net = &init_net;
+	struct net *net = file->f_dentry->d_sb->s_fs_info;
 
 	mutex_lock(&nfsd_mutex);
 	npools = nfsd_nrpools(net);
@@ -510,7 +530,7 @@ static ssize_t __write_versions(struct file *file, char *buf, size_t size)
 	unsigned minor;
 	ssize_t tlen = 0;
 	char *sep;
-	struct net *net = &init_net;
+	struct net *net = file->f_dentry->d_sb->s_fs_info;
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 
 	if (size>0) {
@@ -534,7 +554,7 @@ static ssize_t __write_versions(struct file *file, char *buf, size_t size)
 			else
 				num = simple_strtol(vers, &minorp, 0);
 			if (*minorp == '.') {
-				if (num < 4)
+				if (num != 4)
 					return -EINVAL;
 				minor = simple_strtoul(minorp+1, NULL, 0);
 				if (minor == 0)
@@ -792,7 +812,7 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size,
 static ssize_t write_ports(struct file *file, char *buf, size_t size)
 {
 	ssize_t rv;
-	struct net *net = &init_net;
+	struct net *net = file->f_dentry->d_sb->s_fs_info;
 
 	mutex_lock(&nfsd_mutex);
 	rv = __write_ports(file, buf, size, net);
@@ -827,7 +847,7 @@ int nfsd_max_blksize;
 static ssize_t write_maxblksize(struct file *file, char *buf, size_t size)
 {
 	char *mesg = buf;
-	struct net *net = &init_net;
+	struct net *net = file->f_dentry->d_sb->s_fs_info;
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 
 	if (size > 0) {
@@ -923,7 +943,8 @@ static ssize_t nfsd4_write_time(struct file *file, char *buf, size_t size,
  */
 static ssize_t write_leasetime(struct file *file, char *buf, size_t size)
 {
-	struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id);
+	struct net *net = file->f_dentry->d_sb->s_fs_info;
+	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 	return nfsd4_write_time(file, buf, size, &nn->nfsd4_lease, nn);
 }
 
@@ -939,7 +960,8 @@ static ssize_t write_leasetime(struct file *file, char *buf, size_t size)
  */
 static ssize_t write_gracetime(struct file *file, char *buf, size_t size)
 {
-	struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id);
+	struct net *net = file->f_dentry->d_sb->s_fs_info;
+	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 	return nfsd4_write_time(file, buf, size, &nn->nfsd4_grace, nn);
 }
 
@@ -995,7 +1017,8 @@ static ssize_t __write_recoverydir(struct file *file, char *buf, size_t size,
 static ssize_t write_recoverydir(struct file *file, char *buf, size_t size)
 {
 	ssize_t rv;
-	struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id);
+	struct net *net = file->f_dentry->d_sb->s_fs_info;
+	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 
 	mutex_lock(&nfsd_mutex);
 	rv = __write_recoverydir(file, buf, size, nn);
@@ -1013,7 +1036,7 @@ static ssize_t write_recoverydir(struct file *file, char *buf, size_t size)
 static int nfsd_fill_super(struct super_block * sb, void * data, int silent)
 {
 	static struct tree_descr nfsd_files[] = {
-		[NFSD_List] = {"exports", &exports_operations, S_IRUGO},
+		[NFSD_List] = {"exports", &exports_nfsd_operations, S_IRUGO},
 		[NFSD_Export_features] = {"export_features",
 					&export_features_operations, S_IRUGO},
 		[NFSD_FO_UnlockIP] = {"unlock_ip",
@@ -1037,20 +1060,35 @@ static int nfsd_fill_super(struct super_block * sb, void * data, int silent)
 #endif
 		/* last one */ {""}
 	};
-	return simple_fill_super(sb, 0x6e667364, nfsd_files);
+	struct net *net = data;
+	int ret;
+
+	ret = simple_fill_super(sb, 0x6e667364, nfsd_files);
+	if (ret)
+		return ret;
+	sb->s_fs_info = get_net(net);
+	return 0;
 }
 
 static struct dentry *nfsd_mount(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *data)
 {
-	return mount_single(fs_type, flags, data, nfsd_fill_super);
+	return mount_ns(fs_type, flags, current->nsproxy->net_ns, nfsd_fill_super);
+}
+
+static void nfsd_umount(struct super_block *sb)
+{
+	struct net *net = sb->s_fs_info;
+
+	kill_litter_super(sb);
+	put_net(net);
 }
 
 static struct file_system_type nfsd_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "nfsd",
 	.mount		= nfsd_mount,
-	.kill_sb	= kill_litter_super,
+	.kill_sb	= nfsd_umount,
 };
 
 #ifdef CONFIG_PROC_FS
@@ -1061,7 +1099,8 @@ static int create_proc_exports_entry(void)
 	entry = proc_mkdir("fs/nfs", NULL);
 	if (!entry)
 		return -ENOMEM;
-	entry = proc_create("exports", 0, entry, &exports_operations);
+	entry = proc_create("exports", 0, entry,
+				 &exports_proc_operations);
 	if (!entry)
 		return -ENOMEM;
 	return 0;

fs/nfsd/nfssvc.c

@@ -652,7 +652,6 @@ nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp)
 
 	/* Check whether we have this call in the cache. */
 	switch (nfsd_cache_lookup(rqstp)) {
-	case RC_INTR:
 	case RC_DROPIT:
 		return 0;
 	case RC_REPLY:
@@ -703,8 +702,7 @@ nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp)
 int nfsd_pool_stats_open(struct inode *inode, struct file *file)
 {
 	int ret;
-	struct net *net = &init_net;
-	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
+	struct nfsd_net *nn = net_generic(inode->i_sb->s_fs_info, nfsd_net_id);
 
 	mutex_lock(&nfsd_mutex);
 	if (nn->nfsd_serv == NULL) {
@@ -721,7 +719,7 @@ int nfsd_pool_stats_open(struct inode *inode, struct file *file)
 int nfsd_pool_stats_release(struct inode *inode, struct file *file)
 {
 	int ret = seq_release(inode, file);
-	struct net *net = &init_net;
+	struct net *net = inode->i_sb->s_fs_info;
 
 	mutex_lock(&nfsd_mutex);
 	/* this function really, really should have been called svc_put() */

fs/nfsd/xdr4.h

@@ -563,7 +563,7 @@ __be32 nfsd4_check_resp_size(struct nfsd4_compoundres *, u32);
 void nfsd4_encode_operation(struct nfsd4_compoundres *, struct nfsd4_op *);
 void nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op);
 __be32 nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp,
-		       struct dentry *dentry, __be32 *buffer, int *countp,
+		       struct dentry *dentry, __be32 **buffer, int countp,
 		       u32 *bmval, struct svc_rqst *, int ignore_crossmnt);
 extern __be32 nfsd4_setclientid(struct svc_rqst *rqstp,
 		struct nfsd4_compound_state *,

include/linux/lockd/lockd.h

@@ -212,7 +212,8 @@ int		  nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout)
 __be32		  nlmclnt_grant(const struct sockaddr *addr,
 				const struct nlm_lock *lock);
 void		  nlmclnt_recovery(struct nlm_host *);
-int		  nlmclnt_reclaim(struct nlm_host *, struct file_lock *);
+int		  nlmclnt_reclaim(struct nlm_host *, struct file_lock *,
+				  struct nlm_rqst *);
 void		  nlmclnt_next_cookie(struct nlm_cookie *);
 
 /*

include/linux/sunrpc/addr.h

@@ -0,0 +1,170 @@
+/*
+ * linux/include/linux/sunrpc/addr.h
+ *
+ * Various routines for copying and comparing sockaddrs and for
+ * converting them to and from presentation format.
+ */
+#ifndef _LINUX_SUNRPC_ADDR_H
+#define _LINUX_SUNRPC_ADDR_H
+
+#include <linux/socket.h>
+#include <linux/in.h>
+#include <linux/in6.h>
+#include <net/ipv6.h>
+
+size_t		rpc_ntop(const struct sockaddr *, char *, const size_t);
+size_t		rpc_pton(struct net *, const char *, const size_t,
+			 struct sockaddr *, const size_t);
+char *		rpc_sockaddr2uaddr(const struct sockaddr *, gfp_t);
+size_t		rpc_uaddr2sockaddr(struct net *, const char *, const size_t,
+				   struct sockaddr *, const size_t);
+
+static inline unsigned short rpc_get_port(const struct sockaddr *sap)
+{
+	switch (sap->sa_family) {
+	case AF_INET:
+		return ntohs(((struct sockaddr_in *)sap)->sin_port);
+	case AF_INET6:
+		return ntohs(((struct sockaddr_in6 *)sap)->sin6_port);
+	}
+	return 0;
+}
+
+static inline void rpc_set_port(struct sockaddr *sap,
+				const unsigned short port)
+{
+	switch (sap->sa_family) {
+	case AF_INET:
+		((struct sockaddr_in *)sap)->sin_port = htons(port);
+		break;
+	case AF_INET6:
+		((struct sockaddr_in6 *)sap)->sin6_port = htons(port);
+		break;
+	}
+}
+
+#define IPV6_SCOPE_DELIMITER		'%'
+#define IPV6_SCOPE_ID_LEN		sizeof("%nnnnnnnnnn")
+
+static inline bool __rpc_cmp_addr4(const struct sockaddr *sap1,
+				   const struct sockaddr *sap2)
+{
+	const struct sockaddr_in *sin1 = (const struct sockaddr_in *)sap1;
+	const struct sockaddr_in *sin2 = (const struct sockaddr_in *)sap2;
+
+	return sin1->sin_addr.s_addr == sin2->sin_addr.s_addr;
+}
+
+static inline bool __rpc_copy_addr4(struct sockaddr *dst,
+				    const struct sockaddr *src)
+{
+	const struct sockaddr_in *ssin = (struct sockaddr_in *) src;
+	struct sockaddr_in *dsin = (struct sockaddr_in *) dst;
+
+	dsin->sin_family = ssin->sin_family;
+	dsin->sin_addr.s_addr = ssin->sin_addr.s_addr;
+	return true;
+}
+
+#if IS_ENABLED(CONFIG_IPV6)
+static inline bool __rpc_cmp_addr6(const struct sockaddr *sap1,
+				   const struct sockaddr *sap2)
+{
+	const struct sockaddr_in6 *sin1 = (const struct sockaddr_in6 *)sap1;
+	const struct sockaddr_in6 *sin2 = (const struct sockaddr_in6 *)sap2;
+
+	if (!ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr))
+		return false;
+	else if (ipv6_addr_type(&sin1->sin6_addr) & IPV6_ADDR_LINKLOCAL)
+		return sin1->sin6_scope_id == sin2->sin6_scope_id;
+
+	return true;
+}
+
+static inline bool __rpc_copy_addr6(struct sockaddr *dst,
+				    const struct sockaddr *src)
+{
+	const struct sockaddr_in6 *ssin6 = (const struct sockaddr_in6 *) src;
+	struct sockaddr_in6 *dsin6 = (struct sockaddr_in6 *) dst;
+
+	dsin6->sin6_family = ssin6->sin6_family;
+	dsin6->sin6_addr = ssin6->sin6_addr;
+	dsin6->sin6_scope_id = ssin6->sin6_scope_id;
+	return true;
+}
+#else	/* !(IS_ENABLED(CONFIG_IPV6) */
+static inline bool __rpc_cmp_addr6(const struct sockaddr *sap1,
+				   const struct sockaddr *sap2)
+{
+	return false;
+}
+
+static inline bool __rpc_copy_addr6(struct sockaddr *dst,
+				    const struct sockaddr *src)
+{
+	return false;
+}
+#endif	/* !(IS_ENABLED(CONFIG_IPV6) */
+
+/**
+ * rpc_cmp_addr - compare the address portion of two sockaddrs.
+ * @sap1: first sockaddr
+ * @sap2: second sockaddr
+ *
+ * Just compares the family and address portion. Ignores port, but
+ * compares the scope if it's a link-local address.
+ *
+ * Returns true if the addrs are equal, false if they aren't.
+ */
+static inline bool rpc_cmp_addr(const struct sockaddr *sap1,
+				const struct sockaddr *sap2)
+{
+	if (sap1->sa_family == sap2->sa_family) {
+		switch (sap1->sa_family) {
+		case AF_INET:
+			return __rpc_cmp_addr4(sap1, sap2);
+		case AF_INET6:
+			return __rpc_cmp_addr6(sap1, sap2);
+		}
+	}
+	return false;
+}
+
+/**
+ * rpc_copy_addr - copy the address portion of one sockaddr to another
+ * @dst: destination sockaddr
+ * @src: source sockaddr
+ *
+ * Just copies the address portion and family. Ignores port, scope, etc.
+ * Caller is responsible for making certain that dst is large enough to hold
+ * the address in src. Returns true if address family is supported. Returns
+ * false otherwise.
+ */
+static inline bool rpc_copy_addr(struct sockaddr *dst,
+				 const struct sockaddr *src)
+{
+	switch (src->sa_family) {
+	case AF_INET:
+		return __rpc_copy_addr4(dst, src);
+	case AF_INET6:
+		return __rpc_copy_addr6(dst, src);
+	}
+	return false;
+}
+
+/**
+ * rpc_get_scope_id - return scopeid for a given sockaddr
+ * @sa: sockaddr to get scopeid from
+ *
+ * Returns the value of the sin6_scope_id for AF_INET6 addrs, or 0 if
+ * not an AF_INET6 address.
+ */
+static inline u32 rpc_get_scope_id(const struct sockaddr *sa)
+{
+	if (sa->sa_family != AF_INET6)
+		return 0;
+
+	return ((struct sockaddr_in6 *) sa)->sin6_scope_id;
+}
+
+#endif /* _LINUX_SUNRPC_ADDR_H */

include/linux/sunrpc/cache.h

@@ -83,6 +83,10 @@ struct cache_detail {
 	int			(*cache_upcall)(struct cache_detail *,
 						struct cache_head *);
 
+	void			(*cache_request)(struct cache_detail *cd,
+						 struct cache_head *ch,
+						 char **bpp, int *blen);
+
 	int			(*cache_parse)(struct cache_detail *,
 					       char *buf, int len);
 
@@ -157,11 +161,7 @@ sunrpc_cache_update(struct cache_detail *detail,
 		    struct cache_head *new, struct cache_head *old, int hash);
 
 extern int
-sunrpc_cache_pipe_upcall(struct cache_detail *detail, struct cache_head *h,
-		void (*cache_request)(struct cache_detail *,
-				      struct cache_head *,
-				      char **,
-				      int *));
+sunrpc_cache_pipe_upcall(struct cache_detail *detail, struct cache_head *h);
 
 
 extern void cache_clean_deferred(void *owner);

include/linux/sunrpc/clnt.h

@@ -165,157 +165,5 @@ size_t		rpc_peeraddr(struct rpc_clnt *, struct sockaddr *, size_t);
 const char	*rpc_peeraddr2str(struct rpc_clnt *, enum rpc_display_format_t);
 int		rpc_localaddr(struct rpc_clnt *, struct sockaddr *, size_t);
 
-size_t		rpc_ntop(const struct sockaddr *, char *, const size_t);
-size_t		rpc_pton(struct net *, const char *, const size_t,
-			 struct sockaddr *, const size_t);
-char *		rpc_sockaddr2uaddr(const struct sockaddr *, gfp_t);
-size_t		rpc_uaddr2sockaddr(struct net *, const char *, const size_t,
-				   struct sockaddr *, const size_t);
-
-static inline unsigned short rpc_get_port(const struct sockaddr *sap)
-{
-	switch (sap->sa_family) {
-	case AF_INET:
-		return ntohs(((struct sockaddr_in *)sap)->sin_port);
-	case AF_INET6:
-		return ntohs(((struct sockaddr_in6 *)sap)->sin6_port);
-	}
-	return 0;
-}
-
-static inline void rpc_set_port(struct sockaddr *sap,
-				const unsigned short port)
-{
-	switch (sap->sa_family) {
-	case AF_INET:
-		((struct sockaddr_in *)sap)->sin_port = htons(port);
-		break;
-	case AF_INET6:
-		((struct sockaddr_in6 *)sap)->sin6_port = htons(port);
-		break;
-	}
-}
-
-#define IPV6_SCOPE_DELIMITER		'%'
-#define IPV6_SCOPE_ID_LEN		sizeof("%nnnnnnnnnn")
-
-static inline bool __rpc_cmp_addr4(const struct sockaddr *sap1,
-				   const struct sockaddr *sap2)
-{
-	const struct sockaddr_in *sin1 = (const struct sockaddr_in *)sap1;
-	const struct sockaddr_in *sin2 = (const struct sockaddr_in *)sap2;
-
-	return sin1->sin_addr.s_addr == sin2->sin_addr.s_addr;
-}
-
-static inline bool __rpc_copy_addr4(struct sockaddr *dst,
-				    const struct sockaddr *src)
-{
-	const struct sockaddr_in *ssin = (struct sockaddr_in *) src;
-	struct sockaddr_in *dsin = (struct sockaddr_in *) dst;
-
-	dsin->sin_family = ssin->sin_family;
-	dsin->sin_addr.s_addr = ssin->sin_addr.s_addr;
-	return true;
-}
-
-#if IS_ENABLED(CONFIG_IPV6)
-static inline bool __rpc_cmp_addr6(const struct sockaddr *sap1,
-				   const struct sockaddr *sap2)
-{
-	const struct sockaddr_in6 *sin1 = (const struct sockaddr_in6 *)sap1;
-	const struct sockaddr_in6 *sin2 = (const struct sockaddr_in6 *)sap2;
-
-	if (!ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr))
-		return false;
-	else if (ipv6_addr_type(&sin1->sin6_addr) & IPV6_ADDR_LINKLOCAL)
-		return sin1->sin6_scope_id == sin2->sin6_scope_id;
-
-	return true;
-}
-
-static inline bool __rpc_copy_addr6(struct sockaddr *dst,
-				    const struct sockaddr *src)
-{
-	const struct sockaddr_in6 *ssin6 = (const struct sockaddr_in6 *) src;
-	struct sockaddr_in6 *dsin6 = (struct sockaddr_in6 *) dst;
-
-	dsin6->sin6_family = ssin6->sin6_family;
-	dsin6->sin6_addr = ssin6->sin6_addr;
-	return true;
-}
-#else	/* !(IS_ENABLED(CONFIG_IPV6) */
-static inline bool __rpc_cmp_addr6(const struct sockaddr *sap1,
-				   const struct sockaddr *sap2)
-{
-	return false;
-}
-
-static inline bool __rpc_copy_addr6(struct sockaddr *dst,
-				    const struct sockaddr *src)
-{
-	return false;
-}
-#endif	/* !(IS_ENABLED(CONFIG_IPV6) */
-
-/**
- * rpc_cmp_addr - compare the address portion of two sockaddrs.
- * @sap1: first sockaddr
- * @sap2: second sockaddr
- *
- * Just compares the family and address portion. Ignores port, scope, etc.
- * Returns true if the addrs are equal, false if they aren't.
- */
-static inline bool rpc_cmp_addr(const struct sockaddr *sap1,
-				const struct sockaddr *sap2)
-{
-	if (sap1->sa_family == sap2->sa_family) {
-		switch (sap1->sa_family) {
-		case AF_INET:
-			return __rpc_cmp_addr4(sap1, sap2);
-		case AF_INET6:
-			return __rpc_cmp_addr6(sap1, sap2);
-		}
-	}
-	return false;
-}
-
-/**
- * rpc_copy_addr - copy the address portion of one sockaddr to another
- * @dst: destination sockaddr
- * @src: source sockaddr
- *
- * Just copies the address portion and family. Ignores port, scope, etc.
- * Caller is responsible for making certain that dst is large enough to hold
- * the address in src. Returns true if address family is supported. Returns
- * false otherwise.
- */
-static inline bool rpc_copy_addr(struct sockaddr *dst,
-				 const struct sockaddr *src)
-{
-	switch (src->sa_family) {
-	case AF_INET:
-		return __rpc_copy_addr4(dst, src);
-	case AF_INET6:
-		return __rpc_copy_addr6(dst, src);
-	}
-	return false;
-}
-
-/**
- * rpc_get_scope_id - return scopeid for a given sockaddr
- * @sa: sockaddr to get scopeid from
- *
- * Returns the value of the sin6_scope_id for AF_INET6 addrs, or 0 if
- * not an AF_INET6 address.
- */
-static inline u32 rpc_get_scope_id(const struct sockaddr *sa)
-{
-	if (sa->sa_family != AF_INET6)
-		return 0;
-
-	return ((struct sockaddr_in6 *) sa)->sin6_scope_id;
-}
-
 #endif /* __KERNEL__ */
 #endif /* _LINUX_SUNRPC_CLNT_H */

include/linux/sunrpc/svc.h

@@ -50,6 +50,7 @@ struct svc_pool {
 	unsigned int		sp_nrthreads;	/* # of threads in pool */
 	struct list_head	sp_all_threads;	/* all server threads */
 	struct svc_pool_stats	sp_stats;	/* statistics on pool operation */
+	int			sp_task_pending;/* has pending task */
 } ____cacheline_aligned_in_smp;
 
 /*

include/linux/sunrpc/xdr.h

@@ -56,7 +56,7 @@ struct xdr_buf {
 	struct kvec	head[1],	/* RPC header + non-page data */
 			tail[1];	/* Appended after page data */
 
-	struct page **	pages;		/* Array of contiguous pages */
+	struct page **	pages;		/* Array of pages */
 	unsigned int	page_base,	/* Start of page data */
 			page_len,	/* Length of page data */
 			flags;		/* Flags for data disposition */
@@ -152,6 +152,7 @@ xdr_adjust_iovec(struct kvec *iov, __be32 *p)
 extern void xdr_shift_buf(struct xdr_buf *, size_t);
 extern void xdr_buf_from_iov(struct kvec *, struct xdr_buf *);
 extern int xdr_buf_subsegment(struct xdr_buf *, struct xdr_buf *, unsigned int, unsigned int);
+extern void xdr_buf_trim(struct xdr_buf *, unsigned int);
 extern int xdr_buf_read_netobj(struct xdr_buf *, struct xdr_netobj *, unsigned int);
 extern int read_bytes_from_xdr_buf(struct xdr_buf *, unsigned int, void *, unsigned int);
 extern int write_bytes_to_xdr_buf(struct xdr_buf *, unsigned int, void *, unsigned int);

net/sunrpc/addr.c

@@ -17,7 +17,8 @@
  */
 
 #include <net/ipv6.h>
-#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
+#include <linux/sunrpc/msg_prot.h>
 #include <linux/slab.h>
 #include <linux/export.h>
 

net/sunrpc/auth_gss/gss_krb5_wrap.c

@@ -574,6 +574,8 @@ gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 	buf->head[0].iov_len -= GSS_KRB5_TOK_HDR_LEN + headskip;
 	buf->len -= GSS_KRB5_TOK_HDR_LEN + headskip;
 
+	/* Trim off the checksum blob */
+	xdr_buf_trim(buf, GSS_KRB5_TOK_HDR_LEN + tailskip);
 	return GSS_S_COMPLETE;
 }
 

net/sunrpc/auth_gss/svcauth_gss.c

@@ -182,12 +182,6 @@ static void rsi_request(struct cache_detail *cd,
 	(*bpp)[-1] = '\n';
 }
 
-static int rsi_upcall(struct cache_detail *cd, struct cache_head *h)
-{
-	return sunrpc_cache_pipe_upcall(cd, h, rsi_request);
-}
-
-
 static int rsi_parse(struct cache_detail *cd,
 		    char *mesg, int mlen)
 {
@@ -275,7 +269,7 @@ static struct cache_detail rsi_cache_template = {
 	.hash_size	= RSI_HASHMAX,
 	.name           = "auth.rpcsec.init",
 	.cache_put      = rsi_put,
-	.cache_upcall   = rsi_upcall,
+	.cache_request  = rsi_request,
 	.cache_parse    = rsi_parse,
 	.match		= rsi_match,
 	.init		= rsi_init,
@@ -825,13 +819,17 @@ read_u32_from_xdr_buf(struct xdr_buf *buf, int base, u32 *obj)
  *	The server uses base of head iovec as read pointer, while the
  *	client uses separate pointer. */
 static int
-unwrap_integ_data(struct xdr_buf *buf, u32 seq, struct gss_ctx *ctx)
+unwrap_integ_data(struct svc_rqst *rqstp, struct xdr_buf *buf, u32 seq, struct gss_ctx *ctx)
 {
 	int stat = -EINVAL;
 	u32 integ_len, maj_stat;
 	struct xdr_netobj mic;
 	struct xdr_buf integ_buf;
 
+	/* Did we already verify the signature on the original pass through? */
+	if (rqstp->rq_deferred)
+		return 0;
+
 	integ_len = svc_getnl(&buf->head[0]);
 	if (integ_len & 3)
 		return stat;
@@ -854,6 +852,8 @@ unwrap_integ_data(struct xdr_buf *buf, u32 seq, struct gss_ctx *ctx)
 		goto out;
 	if (svc_getnl(&buf->head[0]) != seq)
 		goto out;
+	/* trim off the mic at the end before returning */
+	xdr_buf_trim(buf, mic.len + 4);
 	stat = 0;
 out:
 	kfree(mic.data);
@@ -1198,7 +1198,7 @@ svcauth_gss_accept(struct svc_rqst *rqstp, __be32 *authp)
 			/* placeholders for length and seq. number: */
 			svc_putnl(resv, 0);
 			svc_putnl(resv, 0);
-			if (unwrap_integ_data(&rqstp->rq_arg,
+			if (unwrap_integ_data(rqstp, &rqstp->rq_arg,
 					gc->gc_seq, rsci->mechctx))
 				goto garbage_args;
 			break;

net/sunrpc/cache.c

@@ -196,9 +196,9 @@ EXPORT_SYMBOL_GPL(sunrpc_cache_update);
 
 static int cache_make_upcall(struct cache_detail *cd, struct cache_head *h)
 {
-	if (!cd->cache_upcall)
-		return -EINVAL;
-	return cd->cache_upcall(cd, h);
+	if (cd->cache_upcall)
+		return cd->cache_upcall(cd, h);
+	return sunrpc_cache_pipe_upcall(cd, h);
 }
 
 static inline int cache_is_valid(struct cache_detail *detail, struct cache_head *h)
@@ -750,6 +750,18 @@ struct cache_reader {
 	int			offset;	/* if non-0, we have a refcnt on next request */
 };
 
+static int cache_request(struct cache_detail *detail,
+			       struct cache_request *crq)
+{
+	char *bp = crq->buf;
+	int len = PAGE_SIZE;
+
+	detail->cache_request(detail, crq->item, &bp, &len);
+	if (len < 0)
+		return -EAGAIN;
+	return PAGE_SIZE - len;
+}
+
 static ssize_t cache_read(struct file *filp, char __user *buf, size_t count,
 			  loff_t *ppos, struct cache_detail *cd)
 {
@@ -784,6 +796,13 @@ static ssize_t cache_read(struct file *filp, char __user *buf, size_t count,
 		rq->readers++;
 	spin_unlock(&queue_lock);
 
+	if (rq->len == 0) {
+		err = cache_request(cd, rq);
+		if (err < 0)
+			goto out;
+		rq->len = err;
+	}
+
 	if (rp->offset == 0 && !test_bit(CACHE_PENDING, &rq->item->flags)) {
 		err = -EAGAIN;
 		spin_lock(&queue_lock);
@@ -1140,17 +1159,14 @@ static bool cache_listeners_exist(struct cache_detail *detail)
  *
  * Each request is at most one page long.
  */
-int sunrpc_cache_pipe_upcall(struct cache_detail *detail, struct cache_head *h,
-		void (*cache_request)(struct cache_detail *,
-				      struct cache_head *,
-				      char **,
-				      int *))
+int sunrpc_cache_pipe_upcall(struct cache_detail *detail, struct cache_head *h)
 {
 
 	char *buf;
 	struct cache_request *crq;
-	char *bp;
-	int len;
+
+	if (!detail->cache_request)
+		return -EINVAL;
 
 	if (!cache_listeners_exist(detail)) {
 		warn_no_listener(detail);
@@ -1167,19 +1183,10 @@ int sunrpc_cache_pipe_upcall(struct cache_detail *detail, struct cache_head *h,
 		return -EAGAIN;
 	}
 
-	bp = buf; len = PAGE_SIZE;
-
-	cache_request(detail, h, &bp, &len);
-
-	if (len < 0) {
-		kfree(buf);
-		kfree(crq);
-		return -EAGAIN;
-	}
 	crq->q.reader = 0;
 	crq->item = cache_get(h);
 	crq->buf = buf;
-	crq->len = PAGE_SIZE - len;
+	crq->len = 0;
 	crq->readers = 0;
 	spin_lock(&queue_lock);
 	list_add_tail(&crq->q.list, &detail->queue);
@@ -1605,7 +1612,7 @@ static int create_cache_proc_entries(struct cache_detail *cd, struct net *net)
 	if (p == NULL)
 		goto out_nomem;
 
-	if (cd->cache_upcall || cd->cache_parse) {
+	if (cd->cache_request || cd->cache_parse) {
 		p = proc_create_data("channel", S_IFREG|S_IRUSR|S_IWUSR,
 				     cd->u.procfs.proc_ent,
 				     &cache_file_operations_procfs, cd);
@@ -1614,7 +1621,7 @@ static int create_cache_proc_entries(struct cache_detail *cd, struct net *net)
 			goto out_nomem;
 	}
 	if (cd->cache_show) {
-		p = proc_create_data("content", S_IFREG|S_IRUSR|S_IWUSR,
+		p = proc_create_data("content", S_IFREG|S_IRUSR,
 				cd->u.procfs.proc_ent,
 				&content_file_operations_procfs, cd);
 		cd->u.procfs.content_ent = p;

net/sunrpc/clnt.c

@@ -33,6 +33,7 @@
 #include <linux/rcupdate.h>
 
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/rpc_pipe_fs.h>
 #include <linux/sunrpc/metrics.h>
 #include <linux/sunrpc/bc_xprt.h>

net/sunrpc/rpcb_clnt.c

@@ -26,6 +26,7 @@
 #include <net/ipv6.h>
 
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/sched.h>
 #include <linux/sunrpc/xprtsock.h>
 

net/sunrpc/svc.c

@@ -515,15 +515,6 @@ EXPORT_SYMBOL_GPL(svc_create_pooled);
 
 void svc_shutdown_net(struct svc_serv *serv, struct net *net)
 {
-	/*
-	 * The set of xprts (contained in the sv_tempsocks and
-	 * sv_permsocks lists) is now constant, since it is modified
-	 * only by accepting new sockets (done by service threads in
-	 * svc_recv) or aging old ones (done by sv_temptimer), or
-	 * configuration changes (excluded by whatever locking the
-	 * caller is using--nfsd_mutex in the case of nfsd).  So it's
-	 * safe to traverse those lists and shut everything down:
-	 */
 	svc_close_net(serv, net);
 
 	if (serv->sv_shutdown)
@@ -1042,6 +1033,7 @@ static void svc_unregister(const struct svc_serv *serv, struct net *net)
 /*
  * dprintk the given error with the address of the client that caused it.
  */
+#ifdef RPC_DEBUG
 static __printf(2, 3)
 void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...)
 {
@@ -1058,6 +1050,9 @@ void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...)
 
 	va_end(args);
 }
+#else
+static __printf(2,3) void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...) {}
+#endif
 
 /*
  * Common routine for processing the RPC request.

net/sunrpc/svc_xprt.c

@@ -499,7 +499,8 @@ void svc_wake_up(struct svc_serv *serv)
 			rqstp->rq_xprt = NULL;
 			 */
 			wake_up(&rqstp->rq_wait);
-		}
+		} else
+			pool->sp_task_pending = 1;
 		spin_unlock_bh(&pool->sp_lock);
 	}
 }
@@ -634,7 +635,13 @@ struct svc_xprt *svc_get_next_xprt(struct svc_rqst *rqstp, long timeout)
 		 * long for cache updates.
 		 */
 		rqstp->rq_chandle.thread_wait = 1*HZ;
+		pool->sp_task_pending = 0;
 	} else {
+		if (pool->sp_task_pending) {
+			pool->sp_task_pending = 0;
+			spin_unlock_bh(&pool->sp_lock);
+			return ERR_PTR(-EAGAIN);
+		}
 		/* No data pending. Go to sleep */
 		svc_thread_enqueue(pool, rqstp);
 
@@ -856,7 +863,6 @@ static void svc_age_temp_xprts(unsigned long closure)
 	struct svc_serv *serv = (struct svc_serv *)closure;
 	struct svc_xprt *xprt;
 	struct list_head *le, *next;
-	LIST_HEAD(to_be_aged);
 
 	dprintk("svc_age_temp_xprts\n");
 
@@ -877,25 +883,15 @@ static void svc_age_temp_xprts(unsigned long closure)
 		if (atomic_read(&xprt->xpt_ref.refcount) > 1 ||
 		    test_bit(XPT_BUSY, &xprt->xpt_flags))
 			continue;
-		svc_xprt_get(xprt);
-		list_move(le, &to_be_aged);
+		list_del_init(le);
 		set_bit(XPT_CLOSE, &xprt->xpt_flags);
 		set_bit(XPT_DETACHED, &xprt->xpt_flags);
-	}
-	spin_unlock_bh(&serv->sv_lock);
-
-	while (!list_empty(&to_be_aged)) {
-		le = to_be_aged.next;
-		/* fiddling the xpt_list node is safe 'cos we're XPT_DETACHED */
-		list_del_init(le);
-		xprt = list_entry(le, struct svc_xprt, xpt_list);
-
 		dprintk("queuing xprt %p for closing\n", xprt);
 
 		/* a thread will dequeue and close it soon */
 		svc_xprt_enqueue(xprt);
-		svc_xprt_put(xprt);
 	}
+	spin_unlock_bh(&serv->sv_lock);
 
 	mod_timer(&serv->sv_temptimer, jiffies + svc_conn_age_period * HZ);
 }
@@ -959,21 +955,24 @@ void svc_close_xprt(struct svc_xprt *xprt)
 }
 EXPORT_SYMBOL_GPL(svc_close_xprt);
 
-static void svc_close_list(struct svc_serv *serv, struct list_head *xprt_list, struct net *net)
+static int svc_close_list(struct svc_serv *serv, struct list_head *xprt_list, struct net *net)
 {
 	struct svc_xprt *xprt;
+	int ret = 0;
 
 	spin_lock(&serv->sv_lock);
 	list_for_each_entry(xprt, xprt_list, xpt_list) {
 		if (xprt->xpt_net != net)
 			continue;
+		ret++;
 		set_bit(XPT_CLOSE, &xprt->xpt_flags);
-		set_bit(XPT_BUSY, &xprt->xpt_flags);
+		svc_xprt_enqueue(xprt);
 	}
 	spin_unlock(&serv->sv_lock);
+	return ret;
 }
 
-static void svc_clear_pools(struct svc_serv *serv, struct net *net)
+static struct svc_xprt *svc_dequeue_net(struct svc_serv *serv, struct net *net)
 {
 	struct svc_pool *pool;
 	struct svc_xprt *xprt;
@@ -988,42 +987,46 @@ static void svc_clear_pools(struct svc_serv *serv, struct net *net)
 			if (xprt->xpt_net != net)
 				continue;
 			list_del_init(&xprt->xpt_ready);
+			spin_unlock_bh(&pool->sp_lock);
+			return xprt;
 		}
 		spin_unlock_bh(&pool->sp_lock);
 	}
+	return NULL;
 }
 
-static void svc_clear_list(struct svc_serv *serv, struct list_head *xprt_list, struct net *net)
+static void svc_clean_up_xprts(struct svc_serv *serv, struct net *net)
 {
 	struct svc_xprt *xprt;
-	struct svc_xprt *tmp;
-	LIST_HEAD(victims);
 
-	spin_lock(&serv->sv_lock);
-	list_for_each_entry_safe(xprt, tmp, xprt_list, xpt_list) {
-		if (xprt->xpt_net != net)
-			continue;
-		list_move(&xprt->xpt_list, &victims);
-	}
-	spin_unlock(&serv->sv_lock);
-
-	list_for_each_entry_safe(xprt, tmp, &victims, xpt_list)
+	while ((xprt = svc_dequeue_net(serv, net))) {
+		set_bit(XPT_CLOSE, &xprt->xpt_flags);
 		svc_delete_xprt(xprt);
+	}
 }
 
+/*
+ * Server threads may still be running (especially in the case where the
+ * service is still running in other network namespaces).
+ *
+ * So we shut down sockets the same way we would on a running server, by
+ * setting XPT_CLOSE, enqueuing, and letting a thread pick it up to do
+ * the close.  In the case there are no such other threads,
+ * threads running, svc_clean_up_xprts() does a simple version of a
+ * server's main event loop, and in the case where there are other
+ * threads, we may need to wait a little while and then check again to
+ * see if they're done.
+ */
 void svc_close_net(struct svc_serv *serv, struct net *net)
 {
-	svc_close_list(serv, &serv->sv_tempsocks, net);
-	svc_close_list(serv, &serv->sv_permsocks, net);
+	int delay = 0;
 
-	svc_clear_pools(serv, net);
-	/*
-	 * At this point the sp_sockets lists will stay empty, since
-	 * svc_xprt_enqueue will not add new entries without taking the
-	 * sp_lock and checking XPT_BUSY.
-	 */
-	svc_clear_list(serv, &serv->sv_tempsocks, net);
-	svc_clear_list(serv, &serv->sv_permsocks, net);
+	while (svc_close_list(serv, &serv->sv_permsocks, net) +
+	       svc_close_list(serv, &serv->sv_tempsocks, net)) {
+
+		svc_clean_up_xprts(serv, net);
+		msleep(delay++);
+	}
 }
 
 /*

net/sunrpc/svcauth_unix.c

@@ -6,6 +6,7 @@
 #include <linux/sunrpc/svcsock.h>
 #include <linux/sunrpc/svcauth.h>
 #include <linux/sunrpc/gss_api.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/err.h>
 #include <linux/seq_file.h>
 #include <linux/hash.h>
@@ -17,7 +18,6 @@
 #include <linux/user_namespace.h>
 #define RPCDBG_FACILITY	RPCDBG_AUTH
 
-#include <linux/sunrpc/clnt.h>
 
 #include "netns.h"
 
@@ -157,11 +157,6 @@ static void ip_map_request(struct cache_detail *cd,
 	(*bpp)[-1] = '\n';
 }
 
-static int ip_map_upcall(struct cache_detail *cd, struct cache_head *h)
-{
-	return sunrpc_cache_pipe_upcall(cd, h, ip_map_request);
-}
-
 static struct ip_map *__ip_map_lookup(struct cache_detail *cd, char *class, struct in6_addr *addr);
 static int __ip_map_update(struct cache_detail *cd, struct ip_map *ipm, struct unix_domain *udom, time_t expiry);
 
@@ -475,11 +470,6 @@ static void unix_gid_request(struct cache_detail *cd,
 	(*bpp)[-1] = '\n';
 }
 
-static int unix_gid_upcall(struct cache_detail *cd, struct cache_head *h)
-{
-	return sunrpc_cache_pipe_upcall(cd, h, unix_gid_request);
-}
-
 static struct unix_gid *unix_gid_lookup(struct cache_detail *cd, kuid_t uid);
 
 static int unix_gid_parse(struct cache_detail *cd,
@@ -586,7 +576,7 @@ static struct cache_detail unix_gid_cache_template = {
 	.hash_size	= GID_HASHMAX,
 	.name		= "auth.unix.gid",
 	.cache_put	= unix_gid_put,
-	.cache_upcall	= unix_gid_upcall,
+	.cache_request	= unix_gid_request,
 	.cache_parse	= unix_gid_parse,
 	.cache_show	= unix_gid_show,
 	.match		= unix_gid_match,
@@ -885,7 +875,7 @@ static struct cache_detail ip_map_cache_template = {
 	.hash_size	= IP_HASHMAX,
 	.name		= "auth.unix.ip",
 	.cache_put	= ip_map_put,
-	.cache_upcall	= ip_map_upcall,
+	.cache_request	= ip_map_request,
 	.cache_parse	= ip_map_parse,
 	.cache_show	= ip_map_show,
 	.match		= ip_map_match,

net/sunrpc/xdr.c

@@ -879,6 +879,47 @@ xdr_buf_subsegment(struct xdr_buf *buf, struct xdr_buf *subbuf,
 }
 EXPORT_SYMBOL_GPL(xdr_buf_subsegment);
 
+/**
+ * xdr_buf_trim - lop at most "len" bytes off the end of "buf"
+ * @buf: buf to be trimmed
+ * @len: number of bytes to reduce "buf" by
+ *
+ * Trim an xdr_buf by the given number of bytes by fixing up the lengths. Note
+ * that it's possible that we'll trim less than that amount if the xdr_buf is
+ * too small, or if (for instance) it's all in the head and the parser has
+ * already read too far into it.
+ */
+void xdr_buf_trim(struct xdr_buf *buf, unsigned int len)
+{
+	size_t cur;
+	unsigned int trim = len;
+
+	if (buf->tail[0].iov_len) {
+		cur = min_t(size_t, buf->tail[0].iov_len, trim);
+		buf->tail[0].iov_len -= cur;
+		trim -= cur;
+		if (!trim)
+			goto fix_len;
+	}
+
+	if (buf->page_len) {
+		cur = min_t(unsigned int, buf->page_len, trim);
+		buf->page_len -= cur;
+		trim -= cur;
+		if (!trim)
+			goto fix_len;
+	}
+
+	if (buf->head[0].iov_len) {
+		cur = min_t(size_t, buf->head[0].iov_len, trim);
+		buf->head[0].iov_len -= cur;
+		trim -= cur;
+	}
+fix_len:
+	buf->len -= (len - trim);
+}
+EXPORT_SYMBOL_GPL(xdr_buf_trim);
+
 static void __read_bytes_from_xdr_buf(struct xdr_buf *subbuf, void *obj, unsigned int len)
 {
 	unsigned int this_len;

net/sunrpc/xprtrdma/transport.c

@@ -51,6 +51,7 @@
 #include <linux/init.h>
 #include <linux/slab.h>
 #include <linux/seq_file.h>
+#include <linux/sunrpc/addr.h>
 
 #include "xprt_rdma.h"
 

net/sunrpc/xprtsock.c

@@ -33,6 +33,7 @@
 #include <linux/udp.h>
 #include <linux/tcp.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/sched.h>
 #include <linux/sunrpc/svcsock.h>
 #include <linux/sunrpc/xprtsock.h>
@@ -1867,13 +1868,9 @@ static int xs_local_finish_connecting(struct rpc_xprt *xprt,
  * @xprt: RPC transport to connect
  * @transport: socket transport to connect
  * @create_sock: function to create a socket of the correct type
- *
- * Invoked by a work queue tasklet.
  */
-static void xs_local_setup_socket(struct work_struct *work)
+static int xs_local_setup_socket(struct sock_xprt *transport)
 {
-	struct sock_xprt *transport =
-		container_of(work, struct sock_xprt, connect_worker.work);
 	struct rpc_xprt *xprt = &transport->xprt;
 	struct socket *sock;
 	int status = -EIO;
@@ -1918,6 +1915,30 @@ out:
 	xprt_clear_connecting(xprt);
 	xprt_wake_pending_tasks(xprt, status);
 	current->flags &= ~PF_FSTRANS;
+	return status;
+}
+
+static void xs_local_connect(struct rpc_xprt *xprt, struct rpc_task *task)
+{
+	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
+	int ret;
+
+	 if (RPC_IS_ASYNC(task)) {
+		/*
+		 * We want the AF_LOCAL connect to be resolved in the
+		 * filesystem namespace of the process making the rpc
+		 * call.  Thus we connect synchronously.
+		 *
+		 * If we want to support asynchronous AF_LOCAL calls,
+		 * we'll need to figure out how to pass a namespace to
+		 * connect.
+		 */
+		rpc_exit(task, -ENOTCONN);
+		return;
+	}
+	ret = xs_local_setup_socket(transport);
+	if (ret && !RPC_IS_SOFTCONN(task))
+		msleep_interruptible(15000);
 }
 
 #ifdef CONFIG_SUNRPC_SWAP
@@ -2455,7 +2476,7 @@ static struct rpc_xprt_ops xs_local_ops = {
 	.alloc_slot		= xprt_alloc_slot,
 	.rpcbind		= xs_local_rpcbind,
 	.set_port		= xs_local_set_port,
-	.connect		= xs_connect,
+	.connect		= xs_local_connect,
 	.buf_alloc		= rpc_malloc,
 	.buf_free		= rpc_free,
 	.send_request		= xs_local_send_request,
@@ -2628,8 +2649,6 @@ static struct rpc_xprt *xs_setup_local(struct xprt_create *args)
 			goto out_err;
 		}
 		xprt_set_bound(xprt);
-		INIT_DELAYED_WORK(&transport->connect_worker,
-					xs_local_setup_socket);
 		xs_format_peer_addresses(xprt, "local", RPCBIND_NETID_LOCAL);
 		break;
 	default: