Re: 3.0-git15 Atomic scheduling in pidmap_init

[Josh Boyer <jwboyer@redhat.com>]

On Wed, Aug 10, 2011 at 04:53:41PM +0200, Frederic Weisbecker wrote:
> On Wed, Aug 10, 2011 at 08:45:29AM -0400, Josh Boyer wrote:
> > On Tue, Aug 09, 2011 at 01:35:18PM +0200, Frederic Weisbecker wrote:
> > > > But setting rdp->qs_pending to 1 in rcu_init_percpu_data() has no effect
> > > > until a grace period starts.  So, if grace periods are prevented from
> > 
> > Er... really?  Because it gets set and __rcu_pending looks at it
> > unconditionally in the case that is calling set_need_resched.  It
> > doesn't check if there is anything about a grace period going on or not.
> 
> You mean this?
> 
> 	if (rdp->qs_pending && !rdp->passed_quiesc) {
> 
> 		/*
> 		 * If force_quiescent_state() coming soon and this CPU
> 		 * needs a quiescent state, and this is either RCU-sched
> 		 * or RCU-bh, force a local reschedule.
> 		 */
> 		rdp->n_rp_qs_pending++;
> 		if (!rdp->preemptible &&
> 		    ULONG_CMP_LT(ACCESS_ONCE(rsp->jiffies_force_qs) - 1,
> 				 jiffies))
> 			set_need_resched();
> 	}
> 

Yes.

> On boot, if we don't start a grace period, we don't schedule a grace period forcing,
> so rsp->jiffies_force_qs is 0.

Ok.  I missed that cpu_needs_another_gp() would somehow prevent
rsp->jiffies_force_qs from getting set.

> With ULONG_CMP_LT taking care of (-1 < jiffies) to be valid even with
> ulong, then we are fine I guess.

I'd be happy to test a patch to make sure :)

josh