Merge branch 'for-4.0' of git://linux-nfs.org/~bfields/linux

Pull nfsd fixes from Bruce Fields:
 "Two main issues:

   - We found that turning on pNFS by default (when it's configured at
     build time) was too aggressive, so we want to switch the default
     before the 4.0 release.

   - Recent client changes to increase open parallelism uncovered a
     serious bug lurking in the server's open code.

  Also fix a krb5/selinux regression.

  The rest is mainly smaller pNFS fixes"

* 'for-4.0' of git://linux-nfs.org/~bfields/linux:
  sunrpc: make debugfs file creation failure non-fatal
  nfsd: require an explicit option to enable pNFS
  NFSD: Fix bad update of layout in nfsd4_return_file_layout
  NFSD: Take care the return value from nfsd4_encode_stateid
  NFSD: Printk blocklayout length and offset as format 0x%llx
  nfsd: return correct lockowner when there is a race on hash insert
  nfsd: return correct openowner when there is a race to put one in the hash
  NFSD: Put exports after nfsd4_layout_verify fail
  NFSD: Error out when register_shrinker() fail
  NFSD: Take care the return value from nfsd4_decode_stateid
  NFSD: Check layout type when returning client layouts
  NFSD: restore trace event lost in mismerge

13 files changed, 78 insertions, 64 deletions

diff --git a/fs/nfsd/blocklayout.c b/fs/nfsd/blocklayout.c
index cdbc78c7254218..03d647bf195d78 100644
--- a/fs/nfsd/blocklayout.c
+++ b/fs/nfsd/blocklayout.c
@@ -137,7 +137,7 @@ nfsd4_block_proc_layoutget(struct inode *inode, const struct svc_fh *fhp,
 	seg->offset = iomap.offset;
 	seg->length = iomap.length;
 
-	dprintk("GET: %lld:%lld %d\n", bex->foff, bex->len, bex->es);
+	dprintk("GET: 0x%llx:0x%llx %d\n", bex->foff, bex->len, bex->es);
 	return 0;
 
 out_error:
diff --git a/fs/nfsd/blocklayoutxdr.c b/fs/nfsd/blocklayoutxdr.c
index 9da89fddab338f..9aa2796da90d91 100644
--- a/fs/nfsd/blocklayoutxdr.c
+++ b/fs/nfsd/blocklayoutxdr.c
@@ -122,19 +122,19 @@ nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
 
 		p = xdr_decode_hyper(p, &bex.foff);
 		if (bex.foff & (block_size - 1)) {
-			dprintk("%s: unaligned offset %lld\n",
+			dprintk("%s: unaligned offset 0x%llx\n",
 				__func__, bex.foff);
 			goto fail;
 		}
 		p = xdr_decode_hyper(p, &bex.len);
 		if (bex.len & (block_size - 1)) {
-			dprintk("%s: unaligned length %lld\n",
+			dprintk("%s: unaligned length 0x%llx\n",
 				__func__, bex.foff);
 			goto fail;
 		}
 		p = xdr_decode_hyper(p, &bex.soff);
 		if (bex.soff & (block_size - 1)) {
-			dprintk("%s: unaligned disk offset %lld\n",
+			dprintk("%s: unaligned disk offset 0x%llx\n",
 				__func__, bex.soff);
 			goto fail;
 		}
diff --git a/fs/nfsd/nfs4layouts.c b/fs/nfsd/nfs4layouts.c
index 1028a062954357..6904213a436368 100644
--- a/fs/nfsd/nfs4layouts.c
+++ b/fs/nfsd/nfs4layouts.c
@@ -118,7 +118,7 @@ void nfsd4_setup_layout_type(struct svc_export *exp)
 {
 	struct super_block *sb = exp->ex_path.mnt->mnt_sb;
 
-	if (exp->ex_flags & NFSEXP_NOPNFS)
+	if (!(exp->ex_flags & NFSEXP_PNFS))
 		return;
 
 	if (sb->s_export_op->get_uuid &&
@@ -440,15 +440,14 @@ nfsd4_return_file_layout(struct nfs4_layout *lp, struct nfsd4_layout_seg *seg,
 			list_move_tail(&lp->lo_perstate, reaplist);
 			return;
 		}
-		end = seg->offset;
+		lo->offset = layout_end(seg);
 	} else {
 		/* retain the whole layout segment on a split. */
 		if (layout_end(seg) < end) {
 			dprintk("%s: split not supported\n", __func__);
 			return;
 		}
-
-		lo->offset = layout_end(seg);
+		end = seg->offset;
 	}
 
 	layout_update_len(lo, end);
@@ -513,6 +512,9 @@ nfsd4_return_client_layouts(struct svc_rqst *rqstp,
 
 	spin_lock(&clp->cl_lock);
 	list_for_each_entry_safe(ls, n, &clp->cl_lo_states, ls_perclnt) {
+		if (ls->ls_layout_type != lrp->lr_layout_type)
+			continue;
+
 		if (lrp->lr_return_type == RETURN_FSID &&
 		    !fh_fsid_match(&ls->ls_stid.sc_file->fi_fhandle,
 				   &cstate->current_fh.fh_handle))
@@ -587,6 +589,8 @@ nfsd4_cb_layout_fail(struct nfs4_layout_stateid *ls)
 
 	rpc_ntop((struct sockaddr *)&clp->cl_addr, addr_str, sizeof(addr_str));
 
+	trace_layout_recall_fail(&ls->ls_stid.sc_stateid);
+
 	printk(KERN_WARNING
 		"nfsd: client %s failed to respond to layout recall. "
 		"  Fencing..\n", addr_str);
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index d30bea8d0277ab..92b9d97aff4f1a 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -1237,8 +1237,8 @@ nfsd4_getdeviceinfo(struct svc_rqst *rqstp,
 		nfserr = ops->proc_getdeviceinfo(exp->ex_path.mnt->mnt_sb, gdp);
 
 	gdp->gd_notify_types &= ops->notify_types;
-	exp_put(exp);
 out:
+	exp_put(exp);
 	return nfserr;
 }
 
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index d2f2c37dc2dbd2..8ba1d888f1e624 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -3221,7 +3221,7 @@ alloc_init_open_stateowner(unsigned int strhashval, struct nfsd4_open *open,
 	} else
 		nfs4_free_openowner(&oo->oo_owner);
 	spin_unlock(&clp->cl_lock);
-	return oo;
+	return ret;
 }
 
 static void init_open_stateid(struct nfs4_ol_stateid *stp, struct nfs4_file *fp, struct nfsd4_open *open) {
@@ -5062,7 +5062,7 @@ alloc_init_lock_stateowner(unsigned int strhashval, struct nfs4_client *clp,
 	} else
 		nfs4_free_lockowner(&lo->lo_owner);
 	spin_unlock(&clp->cl_lock);
-	return lo;
+	return ret;
 }
 
 static void
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index df5e66caf100ca..5fb7e78169a6b2 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1562,7 +1562,11 @@ nfsd4_decode_layoutget(struct nfsd4_compoundargs *argp,
 	p = xdr_decode_hyper(p, &lgp->lg_seg.offset);
 	p = xdr_decode_hyper(p, &lgp->lg_seg.length);
 	p = xdr_decode_hyper(p, &lgp->lg_minlength);
-	nfsd4_decode_stateid(argp, &lgp->lg_sid);
+
+	status = nfsd4_decode_stateid(argp, &lgp->lg_sid);
+	if (status)
+		return status;
+
 	READ_BUF(4);
 	lgp->lg_maxcount = be32_to_cpup(p++);
 
@@ -1580,7 +1584,11 @@ nfsd4_decode_layoutcommit(struct nfsd4_compoundargs *argp,
 	p = xdr_decode_hyper(p, &lcp->lc_seg.offset);
 	p = xdr_decode_hyper(p, &lcp->lc_seg.length);
 	lcp->lc_reclaim = be32_to_cpup(p++);
-	nfsd4_decode_stateid(argp, &lcp->lc_sid);
+
+	status = nfsd4_decode_stateid(argp, &lcp->lc_sid);
+	if (status)
+		return status;
+
 	READ_BUF(4);
 	lcp->lc_newoffset = be32_to_cpup(p++);
 	if (lcp->lc_newoffset) {
@@ -1628,7 +1636,11 @@ nfsd4_decode_layoutreturn(struct nfsd4_compoundargs *argp,
 		READ_BUF(16);
 		p = xdr_decode_hyper(p, &lrp->lr_seg.offset);
 		p = xdr_decode_hyper(p, &lrp->lr_seg.length);
-		nfsd4_decode_stateid(argp, &lrp->lr_sid);
+
+		status = nfsd4_decode_stateid(argp, &lrp->lr_sid);
+		if (status)
+			return status;
+
 		READ_BUF(4);
 		lrp->lrf_body_len = be32_to_cpup(p++);
 		if (lrp->lrf_body_len > 0) {
@@ -4123,7 +4135,7 @@ nfsd4_encode_layoutreturn(struct nfsd4_compoundres *resp, __be32 nfserr,
 		return nfserr_resource;
 	*p++ = cpu_to_be32(lrp->lrs_present);
 	if (lrp->lrs_present)
-		nfsd4_encode_stateid(xdr, &lrp->lr_sid);
+		return nfsd4_encode_stateid(xdr, &lrp->lr_sid);
 	return nfs_ok;
 }
 #endif /* CONFIG_NFSD_PNFS */
diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c
index 83a9694ec485b0..46ec934f5dee8c 100644
--- a/fs/nfsd/nfscache.c
+++ b/fs/nfsd/nfscache.c
@@ -165,13 +165,17 @@ int nfsd_reply_cache_init(void)
 {
 	unsigned int hashsize;
 	unsigned int i;
+	int status = 0;
 
 	max_drc_entries = nfsd_cache_size_limit();
 	atomic_set(&num_drc_entries, 0);
 	hashsize = nfsd_hashsize(max_drc_entries);
 	maskbits = ilog2(hashsize);
 
-	register_shrinker(&nfsd_reply_cache_shrinker);
+	status = register_shrinker(&nfsd_reply_cache_shrinker);
+	if (status)
+		return status;
+
 	drc_slab = kmem_cache_create("nfsd_drc", sizeof(struct svc_cacherep),
 					0, 0, NULL);
 	if (!drc_slab)
diff --git a/include/linux/sunrpc/debug.h b/include/linux/sunrpc/debug.h
index c57d8ea0716cdd..59a7889e15db51 100644
--- a/include/linux/sunrpc/debug.h
+++ b/include/linux/sunrpc/debug.h
@@ -60,17 +60,17 @@ struct rpc_xprt;
 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 void		rpc_register_sysctl(void);
 void		rpc_unregister_sysctl(void);
-int		sunrpc_debugfs_init(void);
+void		sunrpc_debugfs_init(void);
 void		sunrpc_debugfs_exit(void);
-int		rpc_clnt_debugfs_register(struct rpc_clnt *);
+void		rpc_clnt_debugfs_register(struct rpc_clnt *);
 void		rpc_clnt_debugfs_unregister(struct rpc_clnt *);
-int		rpc_xprt_debugfs_register(struct rpc_xprt *);
+void		rpc_xprt_debugfs_register(struct rpc_xprt *);
 void		rpc_xprt_debugfs_unregister(struct rpc_xprt *);
 #else
-static inline int
+static inline void
 sunrpc_debugfs_init(void)
 {
-	return 0;
+	return;
 }
 
 static inline void
@@ -79,10 +79,10 @@ sunrpc_debugfs_exit(void)
 	return;
 }
 
-static inline int
+static inline void
 rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
 {
-	return 0;
+	return;
 }
 
 static inline void
@@ -91,10 +91,10 @@ rpc_clnt_debugfs_unregister(struct rpc_clnt *clnt)
 	return;
 }
 
-static inline int
+static inline void
 rpc_xprt_debugfs_register(struct rpc_xprt *xprt)
 {
-	return 0;
+	return;
 }
 
 static inline void
diff --git a/include/uapi/linux/nfsd/export.h b/include/uapi/linux/nfsd/export.h
index 4742f2cb42f2bd..d3bd6ffec04101 100644
--- a/include/uapi/linux/nfsd/export.h
+++ b/include/uapi/linux/nfsd/export.h
@@ -47,7 +47,7 @@
  * exported filesystem.
  */
 #define	NFSEXP_V4ROOT		0x10000
-#define NFSEXP_NOPNFS		0x20000
+#define NFSEXP_PNFS		0x20000
 
 /* All flags that we claim to support.  (Note we don't support NOACL.) */
 #define NFSEXP_ALLFLAGS		0x3FE7F
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 612aa73bbc60c9..e6ce1517367f88 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -303,9 +303,7 @@ static int rpc_client_register(struct rpc_clnt *clnt,
 	struct super_block *pipefs_sb;
 	int err;
 
-	err = rpc_clnt_debugfs_register(clnt);
-	if (err)
-		return err;
+	rpc_clnt_debugfs_register(clnt);
 
 	pipefs_sb = rpc_get_sb_net(net);
 	if (pipefs_sb) {
diff --git a/net/sunrpc/debugfs.c b/net/sunrpc/debugfs.c
index e811f390f9f67c..82962f7e6e888f 100644
--- a/net/sunrpc/debugfs.c
+++ b/net/sunrpc/debugfs.c
@@ -129,48 +129,52 @@ static const struct file_operations tasks_fops = {
 	.release	= tasks_release,
 };
 
-int
+void
 rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
 {
-	int len, err;
+	int len;
 	char name[24]; /* enough for "../../rpc_xprt/ + 8 hex digits + NULL */
+	struct rpc_xprt *xprt;
 
 	/* Already registered? */
-	if (clnt->cl_debugfs)
-		return 0;
+	if (clnt->cl_debugfs || !rpc_clnt_dir)
+		return;
 
 	len = snprintf(name, sizeof(name), "%x", clnt->cl_clid);
 	if (len >= sizeof(name))
-		return -EINVAL;
+		return;
 
 	/* make the per-client dir */
 	clnt->cl_debugfs = debugfs_create_dir(name, rpc_clnt_dir);
 	if (!clnt->cl_debugfs)
-		return -ENOMEM;
+		return;
 
 	/* make tasks file */
-	err = -ENOMEM;
 	if (!debugfs_create_file("tasks", S_IFREG | S_IRUSR, clnt->cl_debugfs,
 				 clnt, &tasks_fops))
 		goto out_err;
 
-	err = -EINVAL;
 	rcu_read_lock();
+	xprt = rcu_dereference(clnt->cl_xprt);
+	/* no "debugfs" dentry? Don't bother with the symlink. */
+	if (!xprt->debugfs) {
+		rcu_read_unlock();
+		return;
+	}
 	len = snprintf(name, sizeof(name), "../../rpc_xprt/%s",
-			rcu_dereference(clnt->cl_xprt)->debugfs->d_name.name);
+			xprt->debugfs->d_name.name);
 	rcu_read_unlock();
+
 	if (len >= sizeof(name))
 		goto out_err;
 
-	err = -ENOMEM;
 	if (!debugfs_create_symlink("xprt", clnt->cl_debugfs, name))
 		goto out_err;
 
-	return 0;
+	return;
 out_err:
 	debugfs_remove_recursive(clnt->cl_debugfs);
 	clnt->cl_debugfs = NULL;
-	return err;
 }
 
 void
@@ -226,33 +230,33 @@ static const struct file_operations xprt_info_fops = {
 	.release	= xprt_info_release,
 };
 
-int
+void
 rpc_xprt_debugfs_register(struct rpc_xprt *xprt)
 {
 	int len, id;
 	static atomic_t	cur_id;
 	char		name[9]; /* 8 hex digits + NULL term */
 
+	if (!rpc_xprt_dir)
+		return;
+
 	id = (unsigned int)atomic_inc_return(&cur_id);
 
 	len = snprintf(name, sizeof(name), "%x", id);
 	if (len >= sizeof(name))
-		return -EINVAL;
+		return;
 
 	/* make the per-client dir */
 	xprt->debugfs = debugfs_create_dir(name, rpc_xprt_dir);
 	if (!xprt->debugfs)
-		return -ENOMEM;
+		return;
 
 	/* make tasks file */
 	if (!debugfs_create_file("info", S_IFREG | S_IRUSR, xprt->debugfs,
 				 xprt, &xprt_info_fops)) {
 		debugfs_remove_recursive(xprt->debugfs);
 		xprt->debugfs = NULL;
-		return -ENOMEM;
 	}
-
-	return 0;
 }
 
 void
@@ -266,14 +270,17 @@ void __exit
 sunrpc_debugfs_exit(void)
 {
 	debugfs_remove_recursive(topdir);
+	topdir = NULL;
+	rpc_clnt_dir = NULL;
+	rpc_xprt_dir = NULL;
 }
 
-int __init
+void __init
 sunrpc_debugfs_init(void)
 {
 	topdir = debugfs_create_dir("sunrpc", NULL);
 	if (!topdir)
-		goto out;
+		return;
 
 	rpc_clnt_dir = debugfs_create_dir("rpc_clnt", topdir);
 	if (!rpc_clnt_dir)
@@ -283,10 +290,9 @@ sunrpc_debugfs_init(void)
 	if (!rpc_xprt_dir)
 		goto out_remove;
 
-	return 0;
+	return;
 out_remove:
 	debugfs_remove_recursive(topdir);
 	topdir = NULL;
-out:
-	return -ENOMEM;
+	rpc_clnt_dir = NULL;
 }
diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index e37fbed879568d..ee5d3d253102bf 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -98,10 +98,7 @@ init_sunrpc(void)
 	if (err)
 		goto out4;
 
-	err = sunrpc_debugfs_init();
-	if (err)
-		goto out5;
-
+	sunrpc_debugfs_init();
 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	rpc_register_sysctl();
 #endif
@@ -109,8 +106,6 @@ init_sunrpc(void)
 	init_socket_xprt();	/* clnt sock transport */
 	return 0;
 
-out5:
-	unregister_rpc_pipefs();
 out4:
 	unregister_pernet_subsys(&sunrpc_net_ops);
 out3:
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index e3015aede0d944..9949722d99cebf 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -1331,7 +1331,6 @@ static void xprt_init(struct rpc_xprt *xprt, struct net *net)
  */
 struct rpc_xprt *xprt_create_transport(struct xprt_create *args)
 {
-	int err;
 	struct rpc_xprt	*xprt;
 	struct xprt_class *t;
 
@@ -1372,11 +1371,7 @@ found:
 		return ERR_PTR(-ENOMEM);
 	}
 
-	err = rpc_xprt_debugfs_register(xprt);
-	if (err) {
-		xprt_destroy(xprt);
-		return ERR_PTR(err);
-	}
+	rpc_xprt_debugfs_register(xprt);
 
 	dprintk("RPC:       created transport %p with %u slots\n", xprt,
 			xprt->max_reqs);