[PATCH 4.3 019/200] proc: actually make proc_fd_permission() thread-friendly

[Greg Kroah-Hartman]

4.3-stable review patch.  If anyone has any objections, please let me know.

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

From: Oleg Nesterov <oleg@xxxxxxxxxx>

commit 54708d2858e79a2bdda10bf8a20c80eb96c20613 upstream.

The commit 96d0df79f264 ("proc: make proc_fd_permission() thread-friendly")
fixed the access to /proc/self/fd from sub-threads, but introduced another
problem: a sub-thread can't access /proc/<tid>/fd/ or /proc/thread-self/fd
if generic_permission() fails.

Change proc_fd_permission() to check same_thread_group(pid_task(), current).

Fixes: 96d0df79f264 ("proc: make proc_fd_permission() thread-friendly")
Reported-by: "Jin, Yihua" <yihua.jin@xxxxxxxxx>
Signed-off-by: Oleg Nesterov <oleg@xxxxxxxxxx>
Cc: "Eric W. Biederman" <ebiederm@xxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
 fs/proc/fd.c |   14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

--- a/fs/proc/fd.c
+++ b/fs/proc/fd.c
@@ -291,11 +291,19 @@ static struct dentry *proc_lookupfd(stru
  */
 int proc_fd_permission(struct inode *inode, int mask)
 {
-	int rv = generic_permission(inode, mask);
+	struct task_struct *p;
+	int rv;
+
+	rv = generic_permission(inode, mask);
 	if (rv == 0)
-		return 0;
-	if (task_tgid(current) == proc_pid(inode))
+		return rv;
+
+	rcu_read_lock();
+	p = pid_task(proc_pid(inode), PIDTYPE_PID);
+	if (p && same_thread_group(p, current))
 		rv = 0;
+	rcu_read_unlock();
+
 	return rv;
 }