proc: don't (ab)use ->group_leader in proc_task_readdir() paths

proc_task_readdir() does not really need "leader", first_tid() has to
revalidate it anyway.  Just pass proc_pid(inode) to first_tid() instead,
it can do pid_task(PIDTYPE_PID) itself and read ->group_leader only if
necessary.

The patch also extracts the "inode is dead" code from
pid_delete_dentry(dentry) into the new trivial helper,
proc_inode_is_dead(inode), proc_task_readdir() uses it to return -ENOENT
if this dir was removed.

This is a bit racy, but the race is very inlikely and the getdents() after
openndir() can see the empty "." + ".." dir only once.

Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Sameer Nanda <snanda@chromium.org>
Cc: Sergey Dyasly <dserrg@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/fs/proc/base.c b/fs/proc/base.c
index be8e17cabfc75c53ad2e3ad6d6ec3c46d63afc83..9b423fec973226b31e29189574e863b80dd9ae0c 100644 (file)
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -1658,13 +1658,18 @@ int pid_revalidate(struct dentry *dentry, unsigned int flags)
        return 0;
 }
 
+static inline bool proc_inode_is_dead(struct inode *inode)
+{
+       return !proc_pid(inode)->tasks[PIDTYPE_PID].first;
+}
+
 int pid_delete_dentry(const struct dentry *dentry)
 {
        /* Is the task we represent dead?
         * If so, then don't put the dentry on the lru list,
         * kill it immediately.
         */
-       return !proc_pid(dentry->d_inode)->tasks[PIDTYPE_PID].first;
+       return proc_inode_is_dead(dentry->d_inode);
 }
 
 const struct dentry_operations pid_dentry_operations =
@@ -3092,34 +3097,35 @@ out_no_task:
  * In the case of a seek we start with the leader and walk nr
  * threads past it.
  */
-static struct task_struct *first_tid(struct task_struct *leader,
-               int tid, int nr, struct pid_namespace *ns)
+static struct task_struct *first_tid(struct pid *pid, int tid,
+                                       int nr, struct pid_namespace *ns)
 {
-       struct task_struct *pos;
+       struct task_struct *pos, *task;
 
        rcu_read_lock();
-       /* Attempt to start with the pid of a thread */
+       task = pid_task(pid, PIDTYPE_PID);
+       if (!task)
+               goto fail;
+
+       /* Attempt to start with the tid of a thread */
        if (tid && (nr > 0)) {
                pos = find_task_by_pid_ns(tid, ns);
-               if (pos && (pos->group_leader == leader))
+               if (pos && same_thread_group(pos, task))
                        goto found;
        }
 
        /* If nr exceeds the number of threads there is nothing todo */
-       if (nr && nr >= get_nr_threads(leader))
-               goto fail;
-       /* It could be unhashed before we take rcu lock */
-       if (!pid_alive(leader))
+       if (nr && nr >= get_nr_threads(task))
                goto fail;
 
        /* If we haven't found our starting place yet start
         * with the leader and walk nr threads forward.
         */
-       pos = leader;
+       pos = task = task->group_leader;
        do {
                if (nr-- <= 0)
                        goto found;
-       } while_each_thread(leader, pos);
+       } while_each_thread(task, pos);
 fail:
        pos = NULL;
        goto out;
@@ -3155,25 +3161,16 @@ static struct task_struct *next_tid(struct task_struct *start)
 /* for the /proc/TGID/task/ directories */
 static int proc_task_readdir(struct file *file, struct dir_context *ctx)
 {
-       struct task_struct *leader = NULL;
-       struct task_struct *task = get_proc_task(file_inode(file));
+       struct inode *inode = file_inode(file);
+       struct task_struct *task;
        struct pid_namespace *ns;
        int tid;
 
-       if (!task)
-               return -ENOENT;
-       rcu_read_lock();
-       if (pid_alive(task)) {
-               leader = task->group_leader;
-               get_task_struct(leader);
-       }
-       rcu_read_unlock();
-       put_task_struct(task);
-       if (!leader)
+       if (proc_inode_is_dead(inode))
                return -ENOENT;
 
        if (!dir_emit_dots(file, ctx))
-               goto out;
+               return 0;
 
        /* f_version caches the tgid value that the last readdir call couldn't
         * return. lseek aka telldir automagically resets f_version to 0.
@@ -3181,7 +3178,7 @@ static int proc_task_readdir(struct file *file, struct dir_context *ctx)
        ns = file->f_dentry->d_sb->s_fs_info;
        tid = (int)file->f_version;
        file->f_version = 0;
-       for (task = first_tid(leader, tid, ctx->pos - 2, ns);
+       for (task = first_tid(proc_pid(inode), tid, ctx->pos - 2, ns);
             task;
             task = next_tid(task), ctx->pos++) {
                char name[PROC_NUMBUF];
@@ -3197,8 +3194,7 @@ static int proc_task_readdir(struct file *file, struct dir_context *ctx)
                        break;
                }
        }
-out:
-       put_task_struct(leader);
+
        return 0;
 }