Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Mike Christie <michael.christie@oracle.com>
To:     stefanha@redhat.com, jasowang@redhat.com, mst@redhat.com,
        sgarzare@redhat.com, virtualization@lists.linux-foundation.org,
        christian.brauner@ubuntu.com, axboe@kernel.dk,
        linux-kernel@vger.kernel.org
Cc:     Mike Christie <michael.christie@oracle.com>
Subject: [PATCH 3/8] fork: add option to not clone or dup files
Date:   Thu, 16 Sep 2021 16:20:46 -0500
Message-Id: <20210916212051.6918-4-michael.christie@oracle.com>
In-Reply-To: <20210916212051.6918-1-michael.christie@oracle.com>
References: <20210916212051.6918-1-michael.christie@oracle.com>
Content-Transfer-Encoding: 8bit
Content-Type: text/plain
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: =?us-ascii?Q?9yqXJLj9aL2ToiGlJs7EgJM9XmZV/wmBS6vFtNOwb4KloorrsO2DPIWWmph9?=
 =?us-ascii?Q?/IGyRTrLgAyO1Sj/zIzmROHSI/vkDwP9gcFz0aOaTYoRXYsmsL6KgDLGgliF?=
 =?us-ascii?Q?Wm269MR4k3Ld3ziGmd0B+E2EBsM6U0qPnI/D/zDkmFlUe6/3BpVgRiWvaMrN?=
 =?us-ascii?Q?9ZvMrpNQW99fodfAA5GAw9HKTLdO31iAaAq6EZWR8lUq4py3LjhMtCfMAyx7?=
 =?us-ascii?Q?u2SbqE5O5uTpjGnXL4xq8OSRKUNtncYXxyG+dJQWQDx6TiHSxHLSxzAFSWkd?=
 =?us-ascii?Q?oaSA5PxsWSo8IFaqYUvsQF+DTUmh212FMhiAxTAtEjb0XtOeF+jUlXpqxVWW?=
 =?us-ascii?Q?2EcuUJyOLpg/+iTD/m84QqOAPIMr85qsga6+mauXJO0eECBZTsk7gMGWfdf1?=
 =?us-ascii?Q?BA4T6N9PzMWejVPUgN8R12y1oC8IUbiDbLm7VQkGKwkpsQW7vN/0SZUoiLAL?=
 =?us-ascii?Q?jHBZ33ClDNphDNi/pHgl7mDGVF3nXkUhCpghXlenK5fRDcIlXgbCb9ROy0+8?=
 =?us-ascii?Q?Ct+xnemAHa34lJztJv1VelxXjVBmMW5ArQOknHUCZQPStvcIIO/zjAdeIa78?=
 =?us-ascii?Q?1NVi/c8C4T045vmlf7W3uD+p0M54+HfyXf82wZXQD6Z12nl+88l+eFRn3vz3?=
 =?us-ascii?Q?JcjTxk800RTXlYnzWN7EM7NGLY9EUB9Oyib/xl9m4Sa4MPFEyhTe7I30PsEt?=
 =?us-ascii?Q?Dgvnw1yMK9gU5leNT4cj6SRfGKj6/f9PQHd8VPz/FhIa1G0S2kwhfigvEqR+?=
 =?us-ascii?Q?HRQ+3F/4wcWNLPbxqdpTVh6+c3xndxG6I8+jvREfXJgKfg6jV1n3kvajFe2X?=
 =?us-ascii?Q?XT06/aTASvH0NAglbGAGWJD0YfwlK2vfUgYe+lm5DNfn7X8MHwnS2owqVjkl?=
 =?us-ascii?Q?aLf7OnrBsv7yFxGYRdUtmOd2fAyjH14gQ8Mx87Nvn4bUL0XqPRebWr5X3y1Y?=
 =?us-ascii?Q?RAPMwC5N2zqiKg/b//FIx58xtru/MoSezoq2zgjN9LWUejoXHLH4WV7awJFk?=
 =?us-ascii?Q?7PidC6UOI9K1zqJw4CCwYtDVSYUolrj8ik/0BIGuKJSl6XSVRWDTRiVyLXPc?=
 =?us-ascii?Q?4TH86wdEtrw9kWDpnXDVzcJNZYBDaEtwN3sxTu7MD28qFG9v+str+NJ1XmuT?=
 =?us-ascii?Q?8IjUoxG+x1SQTXFQcpt2eAJr82Vd0nBuXGfx5fvrVjAsbOgIuWTd2ZLx99Vd?=
 =?us-ascii?Q?ado1iXAy3Xj2Gst5Q2SA+kuhqFHsAHLlbZhJj+2QvczjSjfSGg8NNoLEZ1Kk?=
 =?us-ascii?Q?LpqSPSOiQiLB6Ai5u19CW8aCIaRtSxtqKWi5c6JCvH1ppw8xk9D2NbUFzaP5?=
 =?us-ascii?Q?mI2l7CAkhwlZzezeHFPc4i8O?=
X-OriginatorOrg: oracle.com
X-MS-Exchange-CrossTenant-Network-Message-Id: ece1d2c9-27fb-49b9-102b-08d97957e42b
X-MS-Exchange-CrossTenant-AuthSource: BYAPR10MB3573.namprd10.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 16 Sep 2021 21:21:04.8867
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 4e2c6054-71cb-48f1-bd6c-3a9705aca71b
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: FlBxfrSeqhWyXwc/oA5UF1FvQOVZzdYQkK1Ob+mppTt9+56QRXgnhVln8Qi6aBcNi9dxITsblp8lSgM8e5P6Ci8ejMhKch0ZYTej+g1VcRk=
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BYAPR10MB2935
X-Proofpoint-Virus-Version: vendor=nai engine=6300 definitions=10109 signatures=668682
X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 mlxscore=0 adultscore=0 phishscore=0
 mlxlogscore=999 suspectscore=0 spamscore=0 bulkscore=0 malwarescore=0
 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.12.0-2109030001
 definitions=main-2109160122
X-Proofpoint-ORIG-GUID: yKkJOJtFfxS4tGmaR2e-SFvk4tmAD-99
X-Proofpoint-GUID: yKkJOJtFfxS4tGmaR2e-SFvk4tmAD-99
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

Each vhost device gets a thread that is used to perform IO and management
operations. Instead of a thread that is accessing a device, the thread is
part of the device, so when it calls kernel_copy_process we can't dup or
clone the parent's (Qemu thread that does the VHOST_SET_OWNER ioctl)
files/FDS because it would do an extra increment on ourself.

Later, when we do:

Qemu process exits:
	do_exit -> exit_files -> put_files_struct -> close_files

we would leak the device's resources because of that extra refcount
on the fd or file_struct.

This patch adds a no_files option so these worker threads can prevent
taking an extra refcount on themselves.

Signed-off-by: Mike Christie <michael.christie@oracle.com>
---
 include/linux/sched/task.h |  3 ++-
 kernel/fork.c              | 14 +++++++++++---
 2 files changed, 13 insertions(+), 4 deletions(-)

diff --git a/include/linux/sched/task.h b/include/linux/sched/task.h
index c55f1eb69d41..d0b0872f56cc 100644
--- a/include/linux/sched/task.h
+++ b/include/linux/sched/task.h
@@ -32,6 +32,7 @@ struct kernel_clone_args {
 	size_t set_tid_size;
 	int cgroup;
 	int io_thread;
+	int no_files;
 	struct cgroup *cgrp;
 	struct css_set *cset;
 };
@@ -86,7 +87,7 @@ extern pid_t kernel_clone(struct kernel_clone_args *kargs);
 struct task_struct *create_io_thread(int (*fn)(void *), void *arg, int node);
 struct task_struct *kernel_copy_process(int (*fn)(void *), void *arg, int node,
 					unsigned long clone_flags,
-					int io_thread);
+					int io_thread, int no_files);
 struct task_struct *fork_idle(int);
 struct mm_struct *copy_init_mm(void);
 extern pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
diff --git a/kernel/fork.c b/kernel/fork.c
index cec7b6011beb..a0468e30b27e 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1532,7 +1532,8 @@ static int copy_fs(unsigned long clone_flags, struct task_struct *tsk)
 	return 0;
 }
 
-static int copy_files(unsigned long clone_flags, struct task_struct *tsk)
+static int copy_files(unsigned long clone_flags, struct task_struct *tsk,
+		      int no_files)
 {
 	struct files_struct *oldf, *newf;
 	int error = 0;
@@ -1544,6 +1545,11 @@ static int copy_files(unsigned long clone_flags, struct task_struct *tsk)
 	if (!oldf)
 		goto out;
 
+	if (no_files) {
+		tsk->files = NULL;
+		goto out;
+	}
+
 	if (clone_flags & CLONE_FILES) {
 		atomic_inc(&oldf->count);
 		goto out;
@@ -2179,7 +2185,7 @@ static __latent_entropy struct task_struct *copy_process(
 	retval = copy_semundo(clone_flags, p);
 	if (retval)
 		goto bad_fork_cleanup_security;
-	retval = copy_files(clone_flags, p);
+	retval = copy_files(clone_flags, p, args->no_files);
 	if (retval)
 		goto bad_fork_cleanup_semundo;
 	retval = copy_fs(clone_flags, p);
@@ -2539,6 +2545,7 @@ struct task_struct *create_io_thread(int (*fn)(void *), void *arg, int node)
  * @node: numa node to allocate task from
  * @clone_flags: CLONE flags
  * @io_thread: 1 if this will be a PF_IO_WORKER else 0.
+ * @no_files: Do not duplicate or copy the parent's open files.
  *
  * This returns a created task, or an error pointer. The returned task is
  * inactive, and the caller must fire it up through wake_up_new_task(p). If
@@ -2546,7 +2553,7 @@ struct task_struct *create_io_thread(int (*fn)(void *), void *arg, int node)
  */
 struct task_struct *kernel_copy_process(int (*fn)(void *), void *arg, int node,
 					unsigned long clone_flags,
-					int io_thread)
+					int io_thread, int no_files)
 {
 	struct kernel_clone_args args = {
 		.flags		= ((lower_32_bits(clone_flags) | CLONE_VM |
@@ -2555,6 +2562,7 @@ struct task_struct *kernel_copy_process(int (*fn)(void *), void *arg, int node,
 		.stack		= (unsigned long)fn,
 		.stack_size	= (unsigned long)arg,
 		.io_thread	= io_thread,
+		.no_files	= no_files,
 	};
 
 	return copy_process(NULL, 0, node, &args);
-- 
2.25.1