Re: [RFC][PATCH 1/4] acpi: Use CPUIDLE_FLAG_TIMER_STOP

[Rafael J. Wysocki]

On Tuesday, September 22, 2020 5:26:51 AM CEST Guenter Roeck wrote:
> On Tue, Sep 15, 2020 at 12:31:58PM +0200, Peter Zijlstra wrote:
> > Make acpi_processor_idle use the common broadcast code, there's no
> > reason not to. This also removes some RCU usage after
> > rcu_idle_enter().
> > 
> > Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
> > Acked-by: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
> > Reported-by: Borislav Petkov <bp@xxxxxxx>
> > Tested-by: Borislav Petkov <bp@xxxxxxx>
> > ---
> >  drivers/acpi/processor_idle.c |   49 +++++++++++++-----------------------------
> >  1 file changed, 16 insertions(+), 33 deletions(-)
> > 
> > --- a/drivers/acpi/processor_idle.c
> > +++ b/drivers/acpi/processor_idle.c
> > @@ -161,18 +161,10 @@ static void lapic_timer_propagate_broadc
> >  }
> >  
> >  /* Power(C) State timer broadcast control */
> > -static void lapic_timer_state_broadcast(struct acpi_processor *pr,
> > -				       struct acpi_processor_cx *cx,
> > -				       int broadcast)
> > -{
> > -	int state = cx - pr->power.states;
> > -
> > -	if (state >= pr->power.timer_broadcast_on_state) {
> > -		if (broadcast)
> > -			tick_broadcast_enter();
> > -		else
> > -			tick_broadcast_exit();
> > -	}
> > +static bool lapic_timer_needs_broadcast(struct acpi_processor *pr,
> > +					struct acpi_processor_cx *cx)
> > +{
> > +	return cx - pr->power.states >= pr->power.timer_broadcast_on_state;
> >  }
> >  
> >  #else
> > @@ -180,9 +172,9 @@ static void lapic_timer_state_broadcast(
> >  static void lapic_timer_check_state(int state, struct acpi_processor *pr,
> >  				   struct acpi_processor_cx *cstate) { }
> >  static void lapic_timer_propagate_broadcast(struct acpi_processor *pr) { }
> > -static void lapic_timer_state_broadcast(struct acpi_processor *pr,
> > -				       struct acpi_processor_cx *cx,
> > -				       int broadcast)
> > +
> > +static bool lapic_timer_needs_broadcast(struct acpi_processor *pr,
> > +					struct acpi_processor_cx *cx)
> >  {
> >  }
> 
> drivers/acpi/processor_idle.c: In function 'lapic_timer_needs_broadcast':
> drivers/acpi/processor_idle.c:179:1: warning: no return statement in function returning non-void [-Wreturn-type]
> 
> Should this return true or false ?

false - if the lapic timer doesn't stop, it doesn't need broadcast.

FWIW, patch appended.

Cheers!

---
From: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
Subject: [PATCH] ACPI: processor: Fix build for ARCH_APICTIMER_STOPS_ON_C3 unset

Fix the lapic_timer_needs_broadcast() stub for
ARCH_APICTIMER_STOPS_ON_C3 unset to actually return
a value.

Fixes: aa6b43d57f99 ("ACPI: processor: Use CPUIDLE_FLAG_TIMER_STOP")
Reported-by: Guenter Roeck <linux@xxxxxxxxxxxx>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
---
 drivers/acpi/processor_idle.c |    1 +
 1 file changed, 1 insertion(+)

Index: linux-pm/drivers/acpi/processor_idle.c
===================================================================
--- linux-pm.orig/drivers/acpi/processor_idle.c
+++ linux-pm/drivers/acpi/processor_idle.c
@@ -176,6 +176,7 @@ static void lapic_timer_propagate_broadc
 static bool lapic_timer_needs_broadcast(struct acpi_processor *pr,
 					struct acpi_processor_cx *cx)
 {
+	return false;
 }
 
 #endif