[PATCH 3.16 36/47] NFSv4: Handle the special Linux file open access mode

[Ben Hutchings]

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

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

From: Trond Myklebust <trond.myklebust@xxxxxxxxxxxxxxx>

commit 44942b4e457beda00981f616402a1a791e8c616e upstream.

According to the open() manpage, Linux reserves the access mode 3
to mean "check for read and write permission on the file and return
a file descriptor that can't be used for reading or writing."

Currently, the NFSv4 code will ask the server to open the file,
and will use an incorrect share access mode of 0. Since it has
an incorrect share access mode, the client later forgets to send
a corresponding close, meaning it can leak stateids on the server.

Fixes: ce4ef7c0a8a05 ("NFS: Split out NFS v4 file operations")
Signed-off-by: Trond Myklebust <trond.myklebust@xxxxxxxxxxxxxxx>
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
---
 fs/nfs/inode.c    | 1 +
 fs/nfs/nfs4file.c | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -893,6 +893,7 @@ int nfs_open(struct inode *inode, struct
 	nfs_fscache_open_file(inode, filp);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(nfs_open);
 
 int nfs_release(struct inode *inode, struct file *filp)
 {
--- a/fs/nfs/nfs4file.c
+++ b/fs/nfs/nfs4file.c
@@ -34,7 +34,7 @@ nfs4_file_open(struct inode *inode, stru
 	dprintk("NFS: open file(%pd2)\n", dentry);
 
 	if ((openflags & O_ACCMODE) == 3)
-		openflags--;
+		return nfs_open(inode, filp);
 
 	/* We can't create new files here */
 	openflags &= ~(O_CREAT|O_EXCL);