9 files changed, 135 insertions, 97 deletions

diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index cadfc2bae60e..caa695c06efb 100644
--- a/fs/nfsd/export.c
+++ b/fs/nfsd/export.c
@@ -402,7 +402,7 @@ 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 *);
 
-static int check_export(struct path *path, int *flags, unsigned char *uuid)
+static int check_export(const struct path *path, int *flags, unsigned char *uuid)
 {
 	struct inode *inode = d_inode(path->dentry);
 
@@ -1181,7 +1181,7 @@ denied:
  * use exp_get_by_name() or exp_find().
  */
 struct svc_export *
-rqst_exp_get_by_name(struct svc_rqst *rqstp, struct path *path)
+rqst_exp_get_by_name(struct svc_rqst *rqstp, const struct path *path)
 {
 	struct svc_export *gssexp, *exp = ERR_PTR(-ENOENT);
 	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
diff --git a/fs/nfsd/export.h b/fs/nfsd/export.h
index b9c0adb3ce09..cb36e6cce829 100644
--- a/fs/nfsd/export.h
+++ b/fs/nfsd/export.h
@@ -111,7 +111,7 @@ int			nfsd_export_init(struct net *);
 void			nfsd_export_shutdown(struct net *);
 void			nfsd_export_flush(struct net *);
 struct svc_export *	rqst_exp_get_by_name(struct svc_rqst *,
-					     struct path *);
+					     const struct path *);
 struct svc_export *	rqst_exp_parent(struct svc_rqst *,
 					struct path *);
 struct svc_export *	rqst_find_fsidzero_export(struct svc_rqst *);
diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c
index 85ca663c052c..e010d90aeb27 100644
--- a/fs/nfsd/filecache.c
+++ b/fs/nfsd/filecache.c
@@ -231,6 +231,9 @@ nfsd_file_alloc(struct net *net, struct inode *inode, unsigned char need,
 	refcount_set(&nf->nf_ref, 1);
 	nf->nf_may = need;
 	nf->nf_mark = NULL;
+	nf->nf_dio_mem_align = 0;
+	nf->nf_dio_offset_align = 0;
+	nf->nf_dio_read_offset_align = 0;
 	return nf;
 }
 
@@ -1070,6 +1073,35 @@ nfsd_file_is_cached(struct inode *inode)
 }
 
 static __be32
+nfsd_file_get_dio_attrs(const struct svc_fh *fhp, struct nfsd_file *nf)
+{
+	struct inode *inode = file_inode(nf->nf_file);
+	struct kstat stat;
+	__be32 status;
+
+	/* Currently only need to get DIO alignment info for regular files */
+	if (!S_ISREG(inode->i_mode))
+		return nfs_ok;
+
+	status = fh_getattr(fhp, &stat);
+	if (status != nfs_ok)
+		return status;
+
+	trace_nfsd_file_get_dio_attrs(inode, &stat);
+
+	if (stat.result_mask & STATX_DIOALIGN) {
+		nf->nf_dio_mem_align = stat.dio_mem_align;
+		nf->nf_dio_offset_align = stat.dio_offset_align;
+	}
+	if (stat.result_mask & STATX_DIO_READ_ALIGN)
+		nf->nf_dio_read_offset_align = stat.dio_read_offset_align;
+	else
+		nf->nf_dio_read_offset_align = nf->nf_dio_offset_align;
+
+	return nfs_ok;
+}
+
+static __be32
 nfsd_file_do_acquire(struct svc_rqst *rqstp, struct net *net,
 		     struct svc_cred *cred,
 		     struct auth_domain *client,
@@ -1187,6 +1219,8 @@ open_file:
 			}
 			status = nfserrno(ret);
 			trace_nfsd_file_open(nf, status);
+			if (status == nfs_ok)
+				status = nfsd_file_get_dio_attrs(fhp, nf);
 		}
 	} else
 		status = nfserr_jukebox;
diff --git a/fs/nfsd/filecache.h b/fs/nfsd/filecache.h
index 722b26c71e45..237a05c74211 100644
--- a/fs/nfsd/filecache.h
+++ b/fs/nfsd/filecache.h
@@ -54,6 +54,10 @@ struct nfsd_file {
 	struct list_head	nf_gc;
 	struct rcu_head		nf_rcu;
 	ktime_t			nf_birthtime;
+
+	u32			nf_dio_mem_align;
+	u32			nf_dio_offset_align;
+	u32			nf_dio_read_offset_align;
 };
 
 int nfsd_file_cache_init(void);
diff --git a/fs/nfsd/localio.c b/fs/nfsd/localio.c
index cb237f1b902a..9e0a37cd29d8 100644
--- a/fs/nfsd/localio.c
+++ b/fs/nfsd/localio.c
@@ -117,6 +117,16 @@ nfsd_open_local_fh(struct net *net, struct auth_domain *dom,
 	return localio;
 }
 
+static void nfsd_file_dio_alignment(struct nfsd_file *nf,
+				    u32 *nf_dio_mem_align,
+				    u32 *nf_dio_offset_align,
+				    u32 *nf_dio_read_offset_align)
+{
+	*nf_dio_mem_align = nf->nf_dio_mem_align;
+	*nf_dio_offset_align = nf->nf_dio_offset_align;
+	*nf_dio_read_offset_align = nf->nf_dio_read_offset_align;
+}
+
 static const struct nfsd_localio_operations nfsd_localio_ops = {
 	.nfsd_net_try_get  = nfsd_net_try_get,
 	.nfsd_net_put  = nfsd_net_put,
@@ -124,6 +134,7 @@ static const struct nfsd_localio_operations nfsd_localio_ops = {
 	.nfsd_file_put_local = nfsd_file_put_local,
 	.nfsd_file_get_local = nfsd_file_get_local,
 	.nfsd_file_file = nfsd_file_file,
+	.nfsd_file_dio_alignment = nfsd_file_dio_alignment,
 };
 
 void nfsd_localio_ops_init(void)
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index bc6b776fc657..2b79129703d5 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -1103,89 +1103,48 @@ static ssize_t write_v4_end_grace(struct file *file, char *buf, size_t size)
  *	populating the filesystem.
  */
 
-/* Basically copying rpc_get_inode. */
 static struct inode *nfsd_get_inode(struct super_block *sb, umode_t mode)
 {
 	struct inode *inode = new_inode(sb);
-	if (!inode)
-		return NULL;
-	/* Following advice from simple_fill_super documentation: */
-	inode->i_ino = iunique(sb, NFSD_MaxReserved);
-	inode->i_mode = mode;
-	simple_inode_init_ts(inode);
-	switch (mode & S_IFMT) {
-	case S_IFDIR:
-		inode->i_fop = &simple_dir_operations;
-		inode->i_op = &simple_dir_inode_operations;
-		inc_nlink(inode);
-		break;
-	case S_IFLNK:
-		inode->i_op = &simple_symlink_inode_operations;
-		break;
-	default:
-		break;
+	if (inode) {
+		/* Following advice from simple_fill_super documentation: */
+		inode->i_ino = iunique(sb, NFSD_MaxReserved);
+		inode->i_mode = mode;
+		simple_inode_init_ts(inode);
 	}
 	return inode;
 }
 
-static int __nfsd_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode, struct nfsdfs_client *ncl)
+static struct dentry *nfsd_mkdir(struct dentry *parent, struct nfsdfs_client *ncl, char *name)
 {
+	struct inode *dir = parent->d_inode;
+	struct dentry *dentry;
 	struct inode *inode;
 
-	inode = nfsd_get_inode(dir->i_sb, mode);
+	inode = nfsd_get_inode(parent->d_sb, S_IFDIR | 0600);
 	if (!inode)
-		return -ENOMEM;
+		return ERR_PTR(-ENOMEM);
+
+	dentry = simple_start_creating(parent, name);
+	if (IS_ERR(dentry)) {
+		iput(inode);
+		return dentry;
+	}
+	inode->i_fop = &simple_dir_operations;
+	inode->i_op = &simple_dir_inode_operations;
+	inc_nlink(inode);
 	if (ncl) {
 		inode->i_private = ncl;
 		kref_get(&ncl->cl_ref);
 	}
-	d_add(dentry, inode);
+	d_instantiate(dentry, inode);
 	inc_nlink(dir);
 	fsnotify_mkdir(dir, dentry);
-	return 0;
-}
-
-static struct dentry *nfsd_mkdir(struct dentry *parent, struct nfsdfs_client *ncl, char *name)
-{
-	struct inode *dir = parent->d_inode;
-	struct dentry *dentry;
-	int ret = -ENOMEM;
-
-	inode_lock(dir);
-	dentry = d_alloc_name(parent, name);
-	if (!dentry)
-		goto out_err;
-	ret = __nfsd_mkdir(d_inode(parent), dentry, S_IFDIR | 0600, ncl);
-	if (ret)
-		goto out_err;
-out:
 	inode_unlock(dir);
 	return dentry;
-out_err:
-	dput(dentry);
-	dentry = ERR_PTR(ret);
-	goto out;
 }
 
 #if IS_ENABLED(CONFIG_SUNRPC_GSS)
-static int __nfsd_symlink(struct inode *dir, struct dentry *dentry,
-			  umode_t mode, const char *content)
-{
-	struct inode *inode;
-
-	inode = nfsd_get_inode(dir->i_sb, mode);
-	if (!inode)
-		return -ENOMEM;
-
-	inode->i_link = (char *)content;
-	inode->i_size = strlen(content);
-
-	d_add(dentry, inode);
-	inc_nlink(dir);
-	fsnotify_create(dir, dentry);
-	return 0;
-}
-
 /*
  * @content is assumed to be a NUL-terminated string that lives
  * longer than the symlink itself.
@@ -1194,17 +1153,25 @@ static void _nfsd_symlink(struct dentry *parent, const char *name,
 			  const char *content)
 {
 	struct inode *dir = parent->d_inode;
+	struct inode *inode;
 	struct dentry *dentry;
-	int ret;
 
-	inode_lock(dir);
-	dentry = d_alloc_name(parent, name);
-	if (!dentry)
-		goto out;
-	ret = __nfsd_symlink(d_inode(parent), dentry, S_IFLNK | 0777, content);
-	if (ret)
-		dput(dentry);
-out:
+	inode = nfsd_get_inode(dir->i_sb, S_IFLNK | 0777);
+	if (!inode)
+		return;
+
+	dentry = simple_start_creating(parent, name);
+	if (IS_ERR(dentry)) {
+		iput(inode);
+		return;
+	}
+
+	inode->i_op = &simple_symlink_inode_operations;
+	inode->i_link = (char *)content;
+	inode->i_size = strlen(content);
+
+	d_instantiate(dentry, inode);
+	fsnotify_create(dir, dentry);
 	inode_unlock(dir);
 }
 #else
@@ -1240,40 +1207,34 @@ struct nfsdfs_client *get_nfsdfs_client(struct inode *inode)
 
 /* XXX: cut'n'paste from simple_fill_super; figure out if we could share
  * code instead. */
-static  int nfsdfs_create_files(struct dentry *root,
+static int nfsdfs_create_files(struct dentry *root,
 				const struct tree_descr *files,
 				struct nfsdfs_client *ncl,
 				struct dentry **fdentries)
 {
 	struct inode *dir = d_inode(root);
-	struct inode *inode;
 	struct dentry *dentry;
-	int i;
 
-	inode_lock(dir);
-	for (i = 0; files->name && files->name[0]; i++, files++) {
-		dentry = d_alloc_name(root, files->name);
-		if (!dentry)
-			goto out;
-		inode = nfsd_get_inode(d_inode(root)->i_sb,
-					S_IFREG | files->mode);
-		if (!inode) {
-			dput(dentry);
-			goto out;
+	for (int i = 0; files->name && files->name[0]; i++, files++) {
+		struct inode *inode = nfsd_get_inode(root->d_sb,
+						     S_IFREG | files->mode);
+		if (!inode)
+			return -ENOMEM;
+		dentry = simple_start_creating(root, files->name);
+		if (IS_ERR(dentry)) {
+			iput(inode);
+			return PTR_ERR(dentry);
 		}
 		kref_get(&ncl->cl_ref);
 		inode->i_fop = files->ops;
 		inode->i_private = ncl;
-		d_add(dentry, inode);
+		d_instantiate(dentry, inode);
 		fsnotify_create(dir, dentry);
 		if (fdentries)
 			fdentries[i] = dentry;
+		inode_unlock(dir);
 	}
-	inode_unlock(dir);
 	return 0;
-out:
-	inode_unlock(dir);
-	return -ENOMEM;
 }
 
 /* on success, returns positive number unique to that client. */
@@ -1993,7 +1954,7 @@ int nfsd_nl_listener_set_doit(struct sk_buff *skb, struct genl_info *info)
 	 * remaining listeners and recreate the list.
 	 */
 	if (delete)
-		svc_xprt_destroy_all(serv, net);
+		svc_xprt_destroy_all(serv, net, false);
 
 	/* walk list of addrs again, open any that still don't exist */
 	nlmsg_for_each_attr_type(attr, NFSD_A_SERVER_SOCK_ADDR, info->nlhdr,
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 82b0111ac469..7057ddd7a0a8 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -535,16 +535,13 @@ void nfsd_destroy_serv(struct net *net)
 #endif
 	}
 
-	svc_xprt_destroy_all(serv, net);
-
 	/*
 	 * write_ports can create the server without actually starting
-	 * any threads--if we get shut down before any threads are
+	 * any threads.  If we get shut down before any threads are
 	 * started, then nfsd_destroy_serv will be run before any of this
 	 * other initialization has been done except the rpcb information.
 	 */
-	svc_rpcb_cleanup(serv, net);
-
+	svc_xprt_destroy_all(serv, net, true);
 	nfsd_shutdown_net(net);
 	svc_destroy(&serv);
 }
diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h
index a664fdf1161e..6e2c8e2aab10 100644
--- a/fs/nfsd/trace.h
+++ b/fs/nfsd/trace.h
@@ -1133,6 +1133,33 @@ TRACE_EVENT(nfsd_file_alloc,
 	)
 );
 
+TRACE_EVENT(nfsd_file_get_dio_attrs,
+	TP_PROTO(
+		const struct inode *inode,
+		const struct kstat *stat
+	),
+	TP_ARGS(inode, stat),
+	TP_STRUCT__entry(
+		__field(const void *, inode)
+		__field(unsigned long, mask)
+		__field(u32, mem_align)
+		__field(u32, offset_align)
+		__field(u32, read_offset_align)
+	),
+	TP_fast_assign(
+		__entry->inode = inode;
+		__entry->mask = stat->result_mask;
+		__entry->mem_align = stat->dio_mem_align;
+		__entry->offset_align = stat->dio_offset_align;
+		__entry->read_offset_align = stat->dio_read_offset_align;
+	),
+	TP_printk("inode=%p flags=%s mem_align=%u offset_align=%u read_offset_align=%u",
+		__entry->inode, show_statx_mask(__entry->mask),
+		__entry->mem_align, __entry->offset_align,
+		__entry->read_offset_align
+	)
+);
+
 TRACE_EVENT(nfsd_file_acquire,
 	TP_PROTO(
 		const struct svc_rqst *rqstp,
diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h
index eff04959606f..fde3e0c11dba 100644
--- a/fs/nfsd/vfs.h
+++ b/fs/nfsd/vfs.h
@@ -185,6 +185,10 @@ static inline __be32 fh_getattr(const struct svc_fh *fh, struct kstat *stat)
 	u32 request_mask = STATX_BASIC_STATS;
 	struct path p = {.mnt = fh->fh_export->ex_path.mnt,
 			 .dentry = fh->fh_dentry};
+	struct inode *inode = d_inode(p.dentry);
+
+	if (S_ISREG(inode->i_mode))
+		request_mask |= (STATX_DIOALIGN | STATX_DIO_READ_ALIGN);
 
 	if (fh->fh_maxsize == NFS4_FHSIZE)
 		request_mask |= (STATX_BTIME | STATX_CHANGE_COOKIE);