sched_show_task: fix unsafe usage of ->real_parent
authorOleg Nesterov <oleg@redhat.com>
Wed, 10 Dec 2014 23:45:21 +0000 (15:45 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 11 Dec 2014 01:41:09 +0000 (17:41 -0800)
rcu_read_lock() can not protect p->real_parent if release_task(p) was
already called, change sched_show_task() to check pis_alive() like other
users do.

Note: we need some helpers to cleanup the code like this.  And it seems
that that the usage of cpu_curr(cpu) in dump_cpu_task() is not safe too.

Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Aaron Tomlin <atomlin@redhat.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>,
Cc: Sterling Alexander <stalexan@redhat.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Roland McGrath <roland@hack.frob.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
kernel/sched/core.c

index bb398c0..b5797b7 100644 (file)
@@ -4527,8 +4527,10 @@ void sched_show_task(struct task_struct *p)
 #ifdef CONFIG_DEBUG_STACK_USAGE
        free = stack_not_used(p);
 #endif
+       ppid = 0;
        rcu_read_lock();
-       ppid = task_pid_nr(rcu_dereference(p->real_parent));
+       if (pid_alive(p))
+               ppid = task_pid_nr(rcu_dereference(p->real_parent));
        rcu_read_unlock();
        printk(KERN_CONT "%5lu %5d %6d 0x%08lx\n", free,
                task_pid_nr(p), ppid,