Re: [PATCH v2] do_exit(): Make sure we run with get_fs() == USER_DS.

[KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>]

> > > diff --git a/kernel/exit.c b/kernel/exit.c
> > > index 21aa7b3..68899b3 100644
> > > --- a/kernel/exit.c
> > > +++ b/kernel/exit.c
> > > @@ -914,6 +914,14 @@ NORET_TYPE void do_exit(long code)
> > >  	if (unlikely(!tsk->pid))
> > >  		panic("Attempted to kill the idle task!");
> > >  
> > > +	/*
> > > +	 * If do_exit is called because this processes oopsed, it's possible
> > > +	 * that get_fs() was left as KERNEL_DS, so reset it to USER_DS before
> > > +	 * continuing. This is relevant at least for clearing clear_child_tid in
> > > +	 * mm_release.
> > > +	 */
> > > +	set_fs(USER_DS);
> > 
> > "This is relevant" is no good explanation ;)
> > Please recognize this is tricky code and Please consider to write more 
> > careful and looooong comments.
> 
> I've seen worse comments.  And occasionally none at all :)
> 
> Is this better?
> 
> --- a/kernel/exit.c~do_exit-make-sure-we-run-with-get_fs-==-user_ds-fix
> +++ a/kernel/exit.c
> @@ -917,8 +917,9 @@ NORET_TYPE void do_exit(long code)
>  	/*
>  	 * If do_exit is called because this processes oopsed, it's possible
>  	 * that get_fs() was left as KERNEL_DS, so reset it to USER_DS before
> -	 * continuing. This is relevant at least for clearing clear_child_tid in
> -	 * mm_release.
> +	 * continuing. Amongst other possible reasons, this is to prevent
> +	 * mm_release()->clear_child_tid() from writing to a user-controlled
> +	 * kernel address.
>  	 */

Great!