From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Content-Type: text/plain; charset="UTF-8" Content-Disposition: inline Content-Transfer-Encoding: 8bit MIME-Version: 1.0 From: Ben Hutchings To: linux-kernel@vger.kernel.org, stable@vger.kernel.org CC: akpm@linux-foundation.org, "Trond Myklebust" , "Jorge Mora" , "Weston Andros Adamson" Date: Sun, 29 Dec 2013 03:08:43 +0100 Message-ID: Subject: [PATCH 3.2 002/185] NFSv4: Fix a use-after-free situation in _nfs4_proc_getlk() In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-ID: 3.2.54-rc1 review patch. If anyone has any objections, please let me know. ------------------ From: Trond Myklebust commit a6f951ddbdfb7bd87d31a44f61abe202ed6ce57f upstream. In nfs4_proc_getlk(), when some error causes a retry of the call to _nfs4_proc_getlk(), we can end up with Oopses of the form BUG: unable to handle kernel NULL pointer dereference at 0000000000000134 IP: [] _raw_spin_lock+0xe/0x30 Call Trace: [] _atomic_dec_and_lock+0x4d/0x70 [] nfs4_put_lock_state+0x32/0xb0 [nfsv4] [] nfs4_fl_release_lock+0x15/0x20 [nfsv4] [] _nfs4_proc_getlk.isra.40+0x146/0x170 [nfsv4] [] nfs4_proc_lock+0x399/0x5a0 [nfsv4] The problem is that we don't clear the request->fl_ops after the first try and so when we retry, nfs4_set_lock_state() exits early without setting the lock stateid. Regression introduced by commit 70cc6487a4e08b8698c0e2ec935fb48d10490162 (locks: make ->lock release private data before returning in GETLK case) Reported-by: Weston Andros Adamson Reported-by: Jorge Mora Signed-off-by: Trond Myklebust Signed-off-by: Ben Hutchings --- fs/nfs/nfs4proc.c | 1 + 1 file changed, 1 insertion(+) --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -4052,6 +4052,7 @@ static int _nfs4_proc_getlk(struct nfs4_ status = 0; } request->fl_ops->fl_release_private(request); + request->fl_ops = NULL; out: return status; }