[PATCH 3.16 278/366] smb3: directory sync should not return an error

[Ben Hutchings]

3.16.60-rc1 review patch.  If anyone has any objections, please let me know.

------------------

From: Steve French <smfrench@xxxxxxxxx>

commit 6e70c267e68d77679534dcf4aaf84e66f2cf1425 upstream.

As with NFS, which ignores sync on directory handles,
fsync on a directory handle is a noop for CIFS/SMB3.
Do not return an error on it.  It breaks some database
apps otherwise.

Signed-off-by: Steve French <smfrench@xxxxxxxxx>
Reviewed-by: Ronnie Sahlberg <lsahlber@xxxxxxxxxx>
Reviewed-by: Pavel Shilovsky <pshilov@xxxxxxxxxxxxx>
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
---
 fs/cifs/cifsfs.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -899,6 +899,18 @@ const struct inode_operations cifs_symli
 #endif
 };
 
+/*
+ * Directory operations under CIFS/SMB2/SMB3 are synchronous, so fsync()
+ * is a dummy operation.
+ */
+static int cifs_dir_fsync(struct file *file, loff_t start, loff_t end, int datasync)
+{
+	cifs_dbg(FYI, "Sync directory - name: %pD datasync: 0x%x\n",
+		 file, datasync);
+
+	return 0;
+}
+
 const struct file_operations cifs_file_ops = {
 	.read = new_sync_read,
 	.write = new_sync_write,
@@ -1018,6 +1030,7 @@ const struct file_operations cifs_dir_op
 	.read    = generic_read_dir,
 	.unlocked_ioctl  = cifs_ioctl,
 	.llseek = generic_file_llseek,
+	.fsync = cifs_dir_fsync,
 };
 
 static void