Re: [PATCH 5/5] Use copy_wait_opts_to_user() in wait_task_continued()

[Oleg Nesterov <oleg@redhat.com>]

On 06/15, Oleg Nesterov wrote:
>
> On 05/11, Vitaly Mayatskikh wrote:
> >
> > --- a/kernel/exit.c
> > +++ b/kernel/exit.c
> > @@ -1416,19 +1416,16 @@ static int wait_task_continued(struct wait_opts *wo, struct task_struct *p)
> >  	get_task_struct(p);
> >  	read_unlock(&tasklist_lock);
> >
> > -	if (!wo->wo_info) {
> > -		retval = wo->wo_rusage
> > -			? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0;
> > -		put_task_struct(p);
> > -		if (!retval && wo->wo_stat)
> > -			retval = put_user(0xffff, wo->wo_stat);
> > -		if (!retval)
> > -			retval = pid;
> > -	} else {
> > -		retval = wait_noreap_copyout(wo, p, pid, uid,
> > -					     CLD_CONTINUED, SIGCONT);
> > -		BUG_ON(retval == 0);
> > -	}
> > +	retval = copy_wait_opts_to_user(wo, p, pid, uid,
> > +					CLD_CONTINUED, SIGCONT, SIGCHLD);
> > +	put_task_struct(p);
> > +
> > +	if (!retval && wo->wo_stat)
> > +		retval = put_user(0xffff, wo->wo_stat);
>
> This adds another user-visible change. Before the patch, we ignore
> ->wo_stat if ->wo_info != NULL.

Ugh! sorry I was wrong.

->wo_info != NULL means waitid() which sets ->wo_stat = NULL, so this
change is correct.

> I could send the fixes for 3 and 4, but unfortunately Vitaly has
> the vacation now. I think it is better to drop
>
> 	wait_task_-cleanups-split-wait_noreap_copyout.patch
> 	wait_task_-cleanups-use-copy_wait_opts_to_user-in-wait_task_stopped.patch
> 	wait_task_-cleanups-use-copy_wait_opts_to_user-in-do_wait.patch
> 	wait_task_-cleanups-use-copy_wait_opts_to_user-in-wait_task_zombie.patch
> 	wait_task_-cleanups-use-copy_wait_opts_to_user-in-wait_task_continued.patch
>
> patches, and wait for Vitaly to redo/resend.

Still I think this would be right :/


Unless Vitaly can see my emails and reply.

Oleg.