Re: [PATCH v2 3/9] rcu/sync: Remove custom check for reader-section

[Paul E. McKenney]

On Sun, Jul 14, 2019 at 02:38:20PM -0400, Joel Fernandes wrote:
> On Sun, Jul 14, 2019 at 02:10:53PM -0400, Joel Fernandes wrote:
> > On Sat, Jul 13, 2019 at 02:28:12PM -0700, Paul E. McKenney wrote:
> [snip]
> > > > > > > > > > 
> > > > > > > > > > Cc: Oleg Nesterov <oleg@xxxxxxxxxx>
> > > > > > > > > > Signed-off-by: Joel Fernandes (Google) <joel@xxxxxxxxxxxxxxxxx>
> > > > > > > > > > ---
> > > > > > > > > >  include/linux/rcu_sync.h | 4 +---
> > > > > > > > > >  1 file changed, 1 insertion(+), 3 deletions(-)
> > > > > > > > > > 
> > > > > > > > > > diff --git a/include/linux/rcu_sync.h b/include/linux/rcu_sync.h
> > > > > > > > > > index 9b83865d24f9..0027d4c8087c 100644
> > > > > > > > > > --- a/include/linux/rcu_sync.h
> > > > > > > > > > +++ b/include/linux/rcu_sync.h
> > > > > > > > > > @@ -31,9 +31,7 @@ struct rcu_sync {
> > > > > > > > > >   */
> > > > > > > > > >  static inline bool rcu_sync_is_idle(struct rcu_sync *rsp)
> > > > > > > > > >  {
> > > > > > > > > > -	RCU_LOCKDEP_WARN(!rcu_read_lock_held() &&
> > > > > > > > > > -			 !rcu_read_lock_bh_held() &&
> > > > > > > > > > -			 !rcu_read_lock_sched_held(),
> > > > > > > > > > +	RCU_LOCKDEP_WARN(!rcu_read_lock_any_held(),
> > > > > > > > > 
> > > > > > > > > I believe that replacing rcu_read_lock_sched_held() with preemptible()
> > > > > > > > > in a CONFIG_PREEMPT=n kernel will give you false-positive splats here.
> > > > > > > > > If you have not already done so, could you please give it a try?
> > > > > > > > 
> > > > > > > > Hi Paul,
> > > > > > > > I don't think it will cause splats for !CONFIG_PREEMPT.
> > > > > > > > 
> > > > > > > > Currently, rcu_read_lock_any_held() introduced in this patch returns true if
> > > > > > > > !preemptible(). This means that:
> > > > > > > > 
> > > > > > > > The following expression above:
> > > > > > > > RCU_LOCKDEP_WARN(!rcu_read_lock_any_held(),...)
> > > > > > > > 
> > > > > > > > Becomes:
> > > > > > > > RCU_LOCKDEP_WARN(preemptible(), ...)
> > > > > > > > 
> > > > > > > > For, CONFIG_PREEMPT=n kernels, this means:
> > > > > > > > RCU_LOCKDEP_WARN(0, ...)
> > > > > > > > 
> > > > > > > > Which would mean no splats. Or, did I miss the point?
> > > > > > > 
> > > > > > > I suggest trying it out on a CONFIG_PREEMPT=n kernel.
> > > > > > 
> > > > > > Sure, will do, sorry did not try it out yet because was busy with weekend
> > > > > > chores but will do soon, thanks!
> > > > > 
> > > > > I am not faulting you for taking the weekend off, actually.  ;-)
> > > > 
> > > > ;-) 
> > > > 
> > > > I tried doing RCU_LOCKDEP_WARN(preemptible(), ...) in this code path and I
> > > > don't get any splats. I also disassembled the code and it seems to me
> > > > RCU_LOCKDEP_WARN() becomes a NOOP which also the above reasoning confirms.
> > > 
> > > OK, very good.  Could you do the same thing for the RCU_LOCKDEP_WARN()
> > > in synchronize_rcu()?  Why or why not?
> > > 
> > 
> > Hi Paul,
> > 
> > Yes synchronize_rcu() can also make use of this technique since it is
> > strictly illegal to call synchronize_rcu() within a reader section.
> > 
> > I will add this to the set of my patches as well and send them all out next
> > week, along with the rcu-sync and bh clean ups we discussed.
> 
> After sending this email, it occurs to me it wont work in synchronize_rcu()
> for !CONFIG_PREEMPT kernels. This is because in a !CONFIG_PREEMPT kernel,
> executing in kernel mode itself looks like being in an RCU reader. So we
> should leave that as is. However it will work fine for rcu_sync_is_idle (for
> CONFIG_PREEMPT=n kernels) as I mentioned earlier.
> 
> Were trying to throw me a Quick-Quiz ? ;-) In that case, hope I passed!

You did pass.  This time.  ;-)

							Thanx, Paul