nfs: disintegrate UAPI for nfs

This is to complete part of the Userspace API (UAPI) disintegration for which
the preparatory patches were pulled recently.  After these patches, userspace
headers will be segregated into:

        include/uapi/linux/.../foo.h

for the userspace interface stuff, and:

        include/linux/.../foo.h

for the strictly kernel internal stuff.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>

8 files changed, 67 insertions, 36 deletions

diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 452e71a1b753..6690269f5dde 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -205,7 +205,7 @@ static int readpage_nounlock(struct file *filp, struct page *page)
 	dout("readpage inode %p file %p page %p index %lu\n",
 	     inode, filp, page, page->index);
 	err = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout,
-				  page->index << PAGE_CACHE_SHIFT, &len,
+				  (u64) page_offset(page), &len,
 				  ci->i_truncate_seq, ci->i_truncate_size,
 				  &page, 1, 0);
 	if (err == -ENOENT)
@@ -286,7 +286,7 @@ static int start_read(struct inode *inode, struct list_head *page_list, int max)
 	int nr_pages = 0;
 	int ret;
 
-	off = page->index << PAGE_CACHE_SHIFT;
+	off = (u64) page_offset(page);
 
 	/* count pages */
 	next_index = page->index;
@@ -308,8 +308,8 @@ static int start_read(struct inode *inode, struct list_head *page_list, int max)
 				    NULL, 0,
 				    ci->i_truncate_seq, ci->i_truncate_size,
 				    NULL, false, 1, 0);
-	if (!req)
-		return -ENOMEM;
+	if (IS_ERR(req))
+		return PTR_ERR(req);
 
 	/* build page vector */
 	nr_pages = len >> PAGE_CACHE_SHIFT;
@@ -426,7 +426,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 	struct ceph_inode_info *ci;
 	struct ceph_fs_client *fsc;
 	struct ceph_osd_client *osdc;
-	loff_t page_off = page->index << PAGE_CACHE_SHIFT;
+	loff_t page_off = page_offset(page);
 	int len = PAGE_CACHE_SIZE;
 	loff_t i_size;
 	int err = 0;
@@ -817,8 +817,7 @@ get_more_pages:
 			/* ok */
 			if (locked_pages == 0) {
 				/* prepare async write request */
-				offset = (unsigned long long)page->index
-					<< PAGE_CACHE_SHIFT;
+				offset = (u64) page_offset(page);
 				len = wsize;
 				req = ceph_osdc_new_request(&fsc->client->osdc,
 					    &ci->i_layout,
@@ -832,8 +831,8 @@ get_more_pages:
 					    ci->i_truncate_size,
 					    &inode->i_mtime, true, 1, 0);
 
-				if (!req) {
-					rc = -ENOMEM;
+				if (IS_ERR(req)) {
+					rc = PTR_ERR(req);
 					unlock_page(page);
 					break;
 				}
@@ -1180,7 +1179,7 @@ static int ceph_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 	struct inode *inode = vma->vm_file->f_dentry->d_inode;
 	struct page *page = vmf->page;
 	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
-	loff_t off = page->index << PAGE_CACHE_SHIFT;
+	loff_t off = page_offset(page);
 	loff_t size, len;
 	int ret;
 
@@ -1225,6 +1224,7 @@ out:
 static struct vm_operations_struct ceph_vmops = {
 	.fault		= filemap_fault,
 	.page_mkwrite	= ceph_page_mkwrite,
+	.remap_pages	= generic_file_remap_pages,
 };
 
 int ceph_mmap(struct file *file, struct vm_area_struct *vma)
@@ -1235,6 +1235,5 @@ int ceph_mmap(struct file *file, struct vm_area_struct *vma)
 		return -ENOEXEC;
 	file_accessed(file);
 	vma->vm_ops = &ceph_vmops;
-	vma->vm_flags |= VM_CAN_NONLINEAR;
 	return 0;
 }
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 620daad201db..3251e9cc6401 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -1005,7 +1005,7 @@ static void __queue_cap_release(struct ceph_mds_session *session,
 
 	BUG_ON(msg->front.iov_len + sizeof(*item) > PAGE_CACHE_SIZE);
 	head = msg->front.iov_base;
-	head->num = cpu_to_le32(le32_to_cpu(head->num) + 1);
+	le32_add_cpu(&head->num, 1);
 	item = msg->front.iov_base + msg->front.iov_len;
 	item->ino = cpu_to_le64(ino);
 	item->cap_id = cpu_to_le64(cap_id);
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index fb962efdacee..6d59006bfa27 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -201,6 +201,7 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
 	int err = -ENOMEM;
 
 	dout("ceph_fs_debugfs_init\n");
+	BUG_ON(!fsc->client->debugfs_dir);
 	fsc->debugfs_congestion_kb =
 		debugfs_create_file("writeback_congestion_kb",
 				    0600,
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index ecebbc09bfc7..5840d2aaed15 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -536,8 +536,8 @@ more:
 				    do_sync,
 				    ci->i_truncate_seq, ci->i_truncate_size,
 				    &mtime, false, 2, page_align);
-	if (!req)
-		return -ENOMEM;
+	if (IS_ERR(req))
+		return PTR_ERR(req);
 
 	if (file->f_flags & O_DIRECT) {
 		pages = ceph_get_direct_page_vector(data, num_pages, false);
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 9fff9f3b17e4..ba95eea201bf 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -992,11 +992,15 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 	if (rinfo->head->is_dentry) {
 		struct inode *dir = req->r_locked_dir;
 
-		err = fill_inode(dir, &rinfo->diri, rinfo->dirfrag,
-				 session, req->r_request_started, -1,
-				 &req->r_caps_reservation);
-		if (err < 0)
-			return err;
+		if (dir) {
+			err = fill_inode(dir, &rinfo->diri, rinfo->dirfrag,
+					 session, req->r_request_started, -1,
+					 &req->r_caps_reservation);
+			if (err < 0)
+				return err;
+		} else {
+			WARN_ON_ONCE(1);
+		}
 	}
 
 	/*
@@ -1004,6 +1008,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 	 * will have trouble splicing in the virtual snapdir later
 	 */
 	if (rinfo->head->is_dentry && !req->r_aborted &&
+	    req->r_locked_dir &&
 	    (rinfo->head->is_target || strncmp(req->r_dentry->d_name.name,
 					       fsc->mount_options->snapdir_name,
 					       req->r_dentry->d_name.len))) {
@@ -1099,7 +1104,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 				pr_err("fill_trace bad get_inode "
 				       "%llx.%llx\n", vino.ino, vino.snap);
 				err = PTR_ERR(in);
-				d_delete(dn);
+				d_drop(dn);
 				goto done;
 			}
 			dn = splice_dentry(dn, in, &have_lease, true);
@@ -1272,7 +1277,7 @@ retry_lookup:
 			in = ceph_get_inode(parent->d_sb, vino);
 			if (IS_ERR(in)) {
 				dout("new_inode badness\n");
-				d_delete(dn);
+				d_drop(dn);
 				dput(dn);
 				err = PTR_ERR(in);
 				goto out;
diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c
index 8e3fb69fbe62..36549a46e311 100644
--- a/fs/ceph/ioctl.c
+++ b/fs/ceph/ioctl.c
@@ -42,7 +42,8 @@ static long __validate_layout(struct ceph_mds_client *mdsc,
 	/* validate striping parameters */
 	if ((l->object_size & ~PAGE_MASK) ||
 	    (l->stripe_unit & ~PAGE_MASK) ||
-	    ((unsigned)l->object_size % (unsigned)l->stripe_unit))
+	    (l->stripe_unit != 0 &&
+	     ((unsigned)l->object_size % (unsigned)l->stripe_unit)))
 		return -EINVAL;
 
 	/* make sure it's a valid data pool */
@@ -186,14 +187,18 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg)
 	u64 tmp;
 	struct ceph_object_layout ol;
 	struct ceph_pg pgid;
+	int r;
 
 	/* copy and validate */
 	if (copy_from_user(&dl, arg, sizeof(dl)))
 		return -EFAULT;
 
 	down_read(&osdc->map_sem);
-	ceph_calc_file_object_mapping(&ci->i_layout, dl.file_offset, &len,
-				      &dl.object_no, &dl.object_offset, &olen);
+	r = ceph_calc_file_object_mapping(&ci->i_layout, dl.file_offset, &len,
+					  &dl.object_no, &dl.object_offset,
+					  &olen);
+	if (r < 0)
+		return -EIO;
 	dl.file_offset -= dl.object_offset;
 	dl.object_size = ceph_file_layout_object_size(ci->i_layout);
 	dl.block_size = ceph_file_layout_su(ci->i_layout);
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index a5a735422aa7..1bcf712655d9 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -2625,7 +2625,8 @@ static void check_new_map(struct ceph_mds_client *mdsc,
 		     ceph_mdsmap_is_laggy(newmap, i) ? " (laggy)" : "",
 		     session_state_name(s->s_state));
 
-		if (memcmp(ceph_mdsmap_get_addr(oldmap, i),
+		if (i >= newmap->m_max_mds ||
+		    memcmp(ceph_mdsmap_get_addr(oldmap, i),
 			   ceph_mdsmap_get_addr(newmap, i),
 			   sizeof(struct ceph_entity_addr))) {
 			if (s->s_state == CEPH_MDS_SESSION_OPENING) {
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index b982239f38f9..2eb43f211325 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -307,7 +307,10 @@ static int parse_mount_options(struct ceph_mount_options **pfsopt,
 {
 	struct ceph_mount_options *fsopt;
 	const char *dev_name_end;
-	int err = -ENOMEM;
+	int err;
+
+	if (!dev_name || !*dev_name)
+		return -EINVAL;
 
 	fsopt = kzalloc(sizeof(*fsopt), GFP_KERNEL);
 	if (!fsopt)
@@ -328,21 +331,33 @@ static int parse_mount_options(struct ceph_mount_options **pfsopt,
 	fsopt->max_readdir_bytes = CEPH_MAX_READDIR_BYTES_DEFAULT;
 	fsopt->congestion_kb = default_congestion_kb();
 
-	/* ip1[:port1][,ip2[:port2]...]:/subdir/in/fs */
+	/*
+	 * Distinguish the server list from the path in "dev_name".
+	 * Internally we do not include the leading '/' in the path.
+	 *
+	 * "dev_name" will look like:
+	 *     <server_spec>[,<server_spec>...]:[<path>]
+	 * where
+	 *     <server_spec> is <ip>[:<port>]
+	 *     <path> is optional, but if present must begin with '/'
+	 */
+	dev_name_end = strchr(dev_name, '/');
+	if (dev_name_end) {
+		/* skip over leading '/' for path */
+		*path = dev_name_end + 1;
+	} else {
+		/* path is empty */
+		dev_name_end = dev_name + strlen(dev_name);
+		*path = dev_name_end;
+	}
 	err = -EINVAL;
-	if (!dev_name)
-		goto out;
-	*path = strstr(dev_name, ":/");
-	if (*path == NULL) {
-		pr_err("device name is missing path (no :/ in %s)\n",
+	dev_name_end--;		/* back up to ':' separator */
+	if (*dev_name_end != ':') {
+		pr_err("device name is missing path (no : separator in %s)\n",
 				dev_name);
 		goto out;
 	}
-	dev_name_end = *path;
 	dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name);
-
-	/* path on server */
-	*path += 2;
 	dout("server path '%s'\n", *path);
 
 	*popt = ceph_parse_options(options, dev_name, dev_name_end,
@@ -603,6 +618,11 @@ bad_cap:
 
 static void destroy_caches(void)
 {
+	/*
+	 * Make sure all delayed rcu free inodes are flushed before we
+	 * destroy cache.
+	 */
+	rcu_barrier();
 	kmem_cache_destroy(ceph_inode_cachep);
 	kmem_cache_destroy(ceph_cap_cachep);
 	kmem_cache_destroy(ceph_dentry_cachep);