Patchwork connector: Fix sid connector (was: Badness at kernel/softirq.c:143...)

login
register
mail settings
Submitter Evgeniy Polyakov
Date Sept. 29, 2009, 2:07 p.m.
Message ID <20090929140718.GA23858@ioremap.net>
Download mbox | patch
Permalink /patch/34431/
State Not Applicable
Delegated to: David Miller
Headers show

Comments

Evgeniy Polyakov - Sept. 29, 2009, 2:07 p.m.
On Tue, Sep 29, 2009 at 03:47:21PM +0200, Christian Borntraeger (borntraeger@de.ibm.com) wrote:
> Ok,  can confirm that this patch fixes my problem, but I am not sure if the
> intended behaviour is still working as expected.

Your patch breaks assumption that task_session(current->group_leader) is
not equal to new session id, but to check task_session() we need either
rcu or task lock. Also setsid() return value is not zero or negative
error, but new session ID or negative error, so I believe attached patch
is a proper fix, although it looks rather ugly.

Also proc_sid_connector() uses GFP_KERNEL allocation which is way too
wrong to use under any locks.

Something like this (not tested :)
Christian Borntraeger - Sept. 29, 2009, 2:15 p.m.
Am Dienstag 29 September 2009 16:07:18 schrieb Evgeniy Polyakov:
> Your patch breaks assumption that task_session(current->group_leader) is
> not equal to new session id, but to check task_session() we need either
> rcu or task lock. Also setsid() return value is not zero or negative
> error, but new session ID or negative error,

Right.

> so I believe attached patch is a proper fix, although it looks rather ugly.
> 
> Also proc_sid_connector() uses GFP_KERNEL allocation which is way too
> wrong to use under any locks.
> 
> Something like this (not tested :)

Patch compiles and seems to work.

Christian
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Oleg Nesterov - Sept. 29, 2009, 2:25 p.m.
On 09/29, Evgeniy Polyakov wrote:
>
> On Tue, Sep 29, 2009 at 03:47:21PM +0200, Christian Borntraeger (borntraeger@de.ibm.com) wrote:
> > Ok,  can confirm that this patch fixes my problem, but I am not sure if the
> > intended behaviour is still working as expected.
>
> Your patch breaks assumption that task_session(current->group_leader) is
> not equal to new session id,

Afaics, no.

> --- a/kernel/sys.c
> +++ b/kernel/sys.c
> @@ -1090,6 +1090,7 @@ SYSCALL_DEFINE0(setsid)
>  	struct pid *sid = task_pid(group_leader);
>  	pid_t session = pid_vnr(sid);
>  	int err = -EPERM;
> +	int send_cn = 0;
>
>  	write_lock_irq(&tasklist_lock);
>  	/* Fail if I am already a session leader */
> @@ -1104,12 +1105,18 @@ SYSCALL_DEFINE0(setsid)
>
>  	group_leader->signal->leader = 1;
>  	__set_special_pids(sid);
> +	if (task_session(group_leader) != sid)
> +		send_cn = 1;

This is not right, task_session(group_leader) must be == sid after
__set_special_pids().

And I don't think "int send_cn" is needed. sys_setsid() must not
succeed if the caller lived in session == task_pid(group_leader).

Or I missed your point?

Oleg.

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Oleg Nesterov - Sept. 29, 2009, 2:45 p.m.
On 09/29, Oleg Nesterov wrote:
>
> On 09/29, Evgeniy Polyakov wrote:
> >
> > On Tue, Sep 29, 2009 at 03:47:21PM +0200, Christian Borntraeger (borntraeger@de.ibm.com) wrote:
> > > Ok,  can confirm that this patch fixes my problem, but I am not sure if the
> > > intended behaviour is still working as expected.
> >
> > Your patch breaks assumption that task_session(current->group_leader) is
> > not equal to new session id,
>
> Afaics, no.
>
> > --- a/kernel/sys.c
> > +++ b/kernel/sys.c
> > @@ -1090,6 +1090,7 @@ SYSCALL_DEFINE0(setsid)
> >  	struct pid *sid = task_pid(group_leader);
> >  	pid_t session = pid_vnr(sid);
> >  	int err = -EPERM;
> > +	int send_cn = 0;
> >
> >  	write_lock_irq(&tasklist_lock);
> >  	/* Fail if I am already a session leader */
> > @@ -1104,12 +1105,18 @@ SYSCALL_DEFINE0(setsid)
> >
> >  	group_leader->signal->leader = 1;
> >  	__set_special_pids(sid);
> > +	if (task_session(group_leader) != sid)
> > +		send_cn = 1;
>
> This is not right, task_session(group_leader) must be == sid after
> __set_special_pids().
>
> And I don't think "int send_cn" is needed. sys_setsid() must not
> succeed if the caller lived in session == task_pid(group_leader).

IOW, if sys_setsid() succeeds, we know it creates the new unique session,
we should report this change.

Note this check

	if (pid_task(sid, PIDTYPE_PGID))
		goto out;

before we actually change pids.

I think Christian's patch only needs the small fixup.

Oleg.

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Evgeniy Polyakov - Sept. 29, 2009, 2:54 p.m.
On Tue, Sep 29, 2009 at 04:25:38PM +0200, Oleg Nesterov (oleg@redhat.com) wrote:
> > --- a/kernel/sys.c
> > +++ b/kernel/sys.c
> > @@ -1090,6 +1090,7 @@ SYSCALL_DEFINE0(setsid)
> >  	struct pid *sid = task_pid(group_leader);
> >  	pid_t session = pid_vnr(sid);
> >  	int err = -EPERM;
> > +	int send_cn = 0;
> >
> >  	write_lock_irq(&tasklist_lock);
> >  	/* Fail if I am already a session leader */
> > @@ -1104,12 +1105,18 @@ SYSCALL_DEFINE0(setsid)
> >
> >  	group_leader->signal->leader = 1;
> >  	__set_special_pids(sid);
> > +	if (task_session(group_leader) != sid)
> > +		send_cn = 1;
> 
> This is not right, task_session(group_leader) must be == sid after
> __set_special_pids().

Yeah, that check should be done before __set_special_pids().

> And I don't think "int send_cn" is needed. sys_setsid() must not
> succeed if the caller lived in session == task_pid(group_leader).

Doesn't it only check pgid while patch intention was to send
notification about sid? I.e. setsid() succeeds, although nothing
happens.
Oleg Nesterov - Sept. 29, 2009, 3:36 p.m.
On 09/29, Evgeniy Polyakov wrote:
>
> On Tue, Sep 29, 2009 at 04:25:38PM +0200, Oleg Nesterov (oleg@redhat.com) wrote:
> > > --- a/kernel/sys.c
> > > +++ b/kernel/sys.c
> > > @@ -1090,6 +1090,7 @@ SYSCALL_DEFINE0(setsid)
> > >  	struct pid *sid = task_pid(group_leader);
> > >  	pid_t session = pid_vnr(sid);
> > >  	int err = -EPERM;
> > > +	int send_cn = 0;
> > >
> > >  	write_lock_irq(&tasklist_lock);
> > >  	/* Fail if I am already a session leader */
> > > @@ -1104,12 +1105,18 @@ SYSCALL_DEFINE0(setsid)
> > >
> > >  	group_leader->signal->leader = 1;
> > >  	__set_special_pids(sid);
> > > +	if (task_session(group_leader) != sid)
> > > +		send_cn = 1;
> >
> > This is not right, task_session(group_leader) must be == sid after
> > __set_special_pids().
>
> Yeah, that check should be done before __set_special_pids().
>
> > And I don't think "int send_cn" is needed. sys_setsid() must not
> > succeed if the caller lived in session == task_pid(group_leader).
>
> Doesn't it only check pgid while patch intention was to send
> notification about sid?

If the proposed sid already was the session id, then prgp shouldn't
be empty.

but this doesn't really matter, we also check ->signal->leader
(not sure, but afaics this check is not strictly necessary because
 of PIDTYPE_PGID check)

> I.e. setsid() succeeds, although nothing
> happens.

This shouldn't happen, or sys_setsid() is buggy. Look, the new session
id is task_pid(current). If sys_setsid() succeeds but we don't change
the session, this means we were already the leader. In that case we
should return -EPERM.

Oleg.

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Evgeniy Polyakov - Sept. 29, 2009, 5:08 p.m.
On Tue, Sep 29, 2009 at 05:36:31PM +0200, Oleg Nesterov (oleg@redhat.com) wrote:
> > Doesn't it only check pgid while patch intention was to send
> > notification about sid?
> 
> If the proposed sid already was the session id, then prgp shouldn't
> be empty.
> 
> but this doesn't really matter, we also check ->signal->leader
> (not sure, but afaics this check is not strictly necessary because
>  of PIDTYPE_PGID check)

Ok, I see, thanks.

Patch

diff --git a/kernel/exit.c b/kernel/exit.c
index 5859f59..1565baf 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -359,10 +359,8 @@  void __set_special_pids(struct pid *pid)
 {
 	struct task_struct *curr = current->group_leader;
 
-	if (task_session(curr) != pid) {
+	if (task_session(curr) != pid)
 		change_pid(curr, PIDTYPE_SID, pid);
-		proc_sid_connector(curr);
-	}
 
 	if (task_pgrp(curr) != pid)
 		change_pid(curr, PIDTYPE_PGID, pid);
diff --git a/kernel/sys.c b/kernel/sys.c
index 255475d..b852a8b 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1090,6 +1090,7 @@  SYSCALL_DEFINE0(setsid)
 	struct pid *sid = task_pid(group_leader);
 	pid_t session = pid_vnr(sid);
 	int err = -EPERM;
+	int send_cn = 0;
 
 	write_lock_irq(&tasklist_lock);
 	/* Fail if I am already a session leader */
@@ -1104,12 +1105,18 @@  SYSCALL_DEFINE0(setsid)
 
 	group_leader->signal->leader = 1;
 	__set_special_pids(sid);
+	if (task_session(group_leader) != sid)
+		send_cn = 1;
 
 	proc_clear_tty(group_leader);
 
 	err = session;
 out:
 	write_unlock_irq(&tasklist_lock);
+
+	if (send_cn)
+		proc_sid_connector(group_leader);
+
 	return err;
 }