Commit 81907739 authored by Oleg Nesterov's avatar Oleg Nesterov Committed by Linus Torvalds

kernel/fork.c:copy_process(): don't add the uninitialized child to thread/task/pid lists

copy_process() adds the new child to thread_group/init_task.tasks list and
then does attach_pid(child, PIDTYPE_PID).  This means that the lockless
next_thread() or next_task() can see this thread with the wrong pid.  Say,
"ls /proc/pid/task" can list the same inode twice.

We could move attach_pid(child, PIDTYPE_PID) up, but in this case
find_task_by_vpid() can find the new thread before it was fully
initialized.

And this is already true for PIDTYPE_PGID/PIDTYPE_SID, With this patch
copy_process() initializes child->pids[*].pid first, then calls
attach_pid() to insert the task into the pid->tasks list.

attach_pid() no longer need the "struct pid*" argument, it is always
called after pid_link->pid was already set.
Signed-off-by: default avatarOleg Nesterov <oleg@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Pavel Emelyanov <xemul@parallels.com>
Cc: Sergey Dyasly <dserrg@gmail.com>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 80628ca0
...@@ -86,11 +86,9 @@ extern struct task_struct *get_pid_task(struct pid *pid, enum pid_type); ...@@ -86,11 +86,9 @@ extern struct task_struct *get_pid_task(struct pid *pid, enum pid_type);
extern struct pid *get_task_pid(struct task_struct *task, enum pid_type type); extern struct pid *get_task_pid(struct task_struct *task, enum pid_type type);
/* /*
* attach_pid() and detach_pid() must be called with the tasklist_lock * these helpers must be called with the tasklist_lock write-held.
* write-held.
*/ */
extern void attach_pid(struct task_struct *task, enum pid_type type, extern void attach_pid(struct task_struct *task, enum pid_type);
struct pid *pid);
extern void detach_pid(struct task_struct *task, enum pid_type); extern void detach_pid(struct task_struct *task, enum pid_type);
extern void change_pid(struct task_struct *task, enum pid_type, extern void change_pid(struct task_struct *task, enum pid_type,
struct pid *pid); struct pid *pid);
......
...@@ -1121,6 +1121,12 @@ static void posix_cpu_timers_init(struct task_struct *tsk) ...@@ -1121,6 +1121,12 @@ static void posix_cpu_timers_init(struct task_struct *tsk)
INIT_LIST_HEAD(&tsk->cpu_timers[2]); INIT_LIST_HEAD(&tsk->cpu_timers[2]);
} }
static inline void
init_task_pid(struct task_struct *task, enum pid_type type, struct pid *pid)
{
task->pids[type].pid = pid;
}
/* /*
* This creates a new process as a copy of the old one, * This creates a new process as a copy of the old one,
* but does not actually start it yet. * but does not actually start it yet.
...@@ -1449,7 +1455,11 @@ static struct task_struct *copy_process(unsigned long clone_flags, ...@@ -1449,7 +1455,11 @@ static struct task_struct *copy_process(unsigned long clone_flags,
if (likely(p->pid)) { if (likely(p->pid)) {
ptrace_init_task(p, (clone_flags & CLONE_PTRACE) || trace); ptrace_init_task(p, (clone_flags & CLONE_PTRACE) || trace);
init_task_pid(p, PIDTYPE_PID, pid);
if (thread_group_leader(p)) { if (thread_group_leader(p)) {
init_task_pid(p, PIDTYPE_PGID, task_pgrp(current));
init_task_pid(p, PIDTYPE_SID, task_session(current));
if (is_child_reaper(pid)) { if (is_child_reaper(pid)) {
ns_of_pid(pid)->child_reaper = p; ns_of_pid(pid)->child_reaper = p;
p->signal->flags |= SIGNAL_UNKILLABLE; p->signal->flags |= SIGNAL_UNKILLABLE;
...@@ -1457,10 +1467,10 @@ static struct task_struct *copy_process(unsigned long clone_flags, ...@@ -1457,10 +1467,10 @@ static struct task_struct *copy_process(unsigned long clone_flags,
p->signal->leader_pid = pid; p->signal->leader_pid = pid;
p->signal->tty = tty_kref_get(current->signal->tty); p->signal->tty = tty_kref_get(current->signal->tty);
attach_pid(p, PIDTYPE_PGID, task_pgrp(current));
attach_pid(p, PIDTYPE_SID, task_session(current));
list_add_tail(&p->sibling, &p->real_parent->children); list_add_tail(&p->sibling, &p->real_parent->children);
list_add_tail_rcu(&p->tasks, &init_task.tasks); list_add_tail_rcu(&p->tasks, &init_task.tasks);
attach_pid(p, PIDTYPE_PGID);
attach_pid(p, PIDTYPE_SID);
__this_cpu_inc(process_counts); __this_cpu_inc(process_counts);
} else { } else {
current->signal->nr_threads++; current->signal->nr_threads++;
...@@ -1470,7 +1480,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, ...@@ -1470,7 +1480,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
list_add_tail_rcu(&p->thread_group, list_add_tail_rcu(&p->thread_group,
&p->group_leader->thread_group); &p->group_leader->thread_group);
} }
attach_pid(p, PIDTYPE_PID, pid); attach_pid(p, PIDTYPE_PID);
nr_threads++; nr_threads++;
} }
......
...@@ -373,14 +373,10 @@ EXPORT_SYMBOL_GPL(find_vpid); ...@@ -373,14 +373,10 @@ EXPORT_SYMBOL_GPL(find_vpid);
/* /*
* attach_pid() must be called with the tasklist_lock write-held. * attach_pid() must be called with the tasklist_lock write-held.
*/ */
void attach_pid(struct task_struct *task, enum pid_type type, void attach_pid(struct task_struct *task, enum pid_type type)
struct pid *pid)
{ {
struct pid_link *link; struct pid_link *link = &task->pids[type];
hlist_add_head_rcu(&link->node, &link->pid->tasks[type]);
link = &task->pids[type];
link->pid = pid;
hlist_add_head_rcu(&link->node, &pid->tasks[type]);
} }
static void __change_pid(struct task_struct *task, enum pid_type type, static void __change_pid(struct task_struct *task, enum pid_type type,
...@@ -412,7 +408,7 @@ void change_pid(struct task_struct *task, enum pid_type type, ...@@ -412,7 +408,7 @@ void change_pid(struct task_struct *task, enum pid_type type,
struct pid *pid) struct pid *pid)
{ {
__change_pid(task, type, pid); __change_pid(task, type, pid);
attach_pid(task, type, pid); attach_pid(task, type);
} }
/* transfer_pid is an optimization of attach_pid(new), detach_pid(old) */ /* transfer_pid is an optimization of attach_pid(new), detach_pid(old) */
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment