Re: [PATCH] rcu: Is it safe to enter an RCU read-side critical section?

[Frederic Weisbecker <fweisbec@gmail.com>]

On Mon, Sep 09, 2013 at 08:39:26AM -0400, Steven Rostedt wrote:
> On Mon, 9 Sep 2013 14:13:31 +0200
> Frederic Weisbecker <fweisbec@gmail.com> wrote:
> 
> 
> > > In any case the preempt_disable/enable pair there is just plain wrong as
> > > Eric pointed out.
> > 
> > Check this:
> > 
> > 34240697d619c439c55f21989680024dcb604aab "rcu: Disable preemption in rcu_is_cpu_idle()"
> 
> 
> Ug, and that patch does nothing to fix the bug that it reported!
> 
>     1.  Task A on CPU 1 enters rcu_is_cpu_idle() and picks up the
>     pointer to CPU 1's per-CPU variables.
>     
>     2.  Task B preempts Task A and starts running on CPU 1.
> 
> Let's say that B preempts Task A here:
> 
>        preempt_disable();
>        ret = (atomic_read(&__get_cpu_var(rcu_dynticks).dynticks) &  0x1) == 0;
>        preempt_enable();
>  <preempt>
>        return ret;
> 
>     
>     3.  Task A migrates to CPU 2.
>     
>     4.  Task B blocks, leaving CPU 1 idle.
>     
>     5.  Task A continues execution on CPU 2, accessing CPU 1's
>     dyntick-idle information using the pointer fetched in step 1 above,
>     and finds that CPU 1 is idle.
> 
> Yeah, and Task A is using the "ret" from CPU 1!
>     
>     6.  Task A therefore incorrectly concludes that it is executing in
>         an extended quiescent state, possibly issuing a spurious splat.
>     
>     Therefore, this commit disables preemption within the
>     rcu_is_cpu_idle() function.
> 
> Where this commit is totally bogus. Sorry, but it is.
> 
> This just proves that the caller of rcu_is_cpu_idle() must disable
> preemption itself for the entire time that it needs to use the result
> of rcu_is_cpu_idle().

Sorry, I don't understand your point here. What's wrong with checking the
ret from another CPU?