Re: [PATCH 1/4] nohz: Split extended quiescent state handling from nohz switch

[Frederic Weisbecker <fweisbec@gmail.com>]

On Sun, Sep 04, 2011 at 02:01:30PM -0700, Paul E. McKenney wrote:
> On Sat, Aug 20, 2011 at 07:30:49PM +0200, Frederic Weisbecker wrote:
> > It is assumed that rcu won't be used once we switch to tickless
> > mode and until we restart the tick. However this is not always
> > true, as in x86-64 where we dereference the idle notifiers after
> > the tick is stopped.
> > 
> > To prepare for fixing this, split the tickless mode switching and
> > RCU extended quiescent state logics.
> > Make tick_nohz_stop/restart_sched_tick() RCU agnostic but provide
> > a new pair of APIs tick_nohz_enter/exit_idle() that keep the
> > old behaviour by handling both the nohz mode and RCU extended
> > quiescent states, then convert every archs to use these.
> > 
> > Archs that want to switch to RCU extended QS to some custom points
> > can do it later by changing the parameter in tick_nohz_enter,exit_idle()
> > to false and call rcu_enter,exit() separately.
> 
> Hello, Frederic,
> 
> Some of my testing indicates that it is necessary to push
> tick_nohz_enter_idle() and tick_nohz_exit_idle() further down in the
> powerpc case due to tracing that happens in functions called from
> ppc_md.power_save().  My current guess is that for the pSeries flavor
> of powerpc, these must be called from __trace_hcall_entry() and
> __trace_hcall_exit(), probably controlled by a powerpc-specific per-CPU
> variable.
> 
> The following functions likely need help.
> 
> 	cbe_power_save()
> 	cpm_idle()
> 	e500_idle()
> 	idle_doze()
> 	idle_spin()
> 	ppc44x_idle()
> 	ppc6xx_idle()
> 	ps3_power_save()
> 	pseries_dedicated_idle_sleep()
> 	pseries_shared_idle_sleep()
> 
> I am continuing testing.  In the meantime, other thoughts?
> 
> 							Thanx, Paul

Yeah sure, I still have that in my todo list. I wanted to ensure
first that at least you are fine with the new interface before I go
fixing more detected bad callsites.

Thanks.