Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1310537 > unrolled thread

[PATCH 1/2] sched / idle: Drop default_idle_call() fallback from call_cpuidle()

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [PATCH 1/2] sched / idle: Drop default_idle_call() fallback from call_cpuidle() "Rafael J. Wysocki" <rjw@rjwysocki.net> - 2016-01-16 01:00 +0100
    Re: [PATCH 1/2] sched / idle: Drop default_idle_call() fallback from  call_cpuidle() Peter Zijlstra <peterz@infradead.org> - 2016-01-18 15:40 +0100

#1310537 — [PATCH 1/2] sched / idle: Drop default_idle_call() fallback from call_cpuidle()

From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

After commit 9c4b2867ed7c (cpuidle: menu: Fix menu_select() for
CPUIDLE_DRIVER_STATE_START == 0) it is clear that menu_select()
cannot return negative values.  Moreover, ladder_select_state()
will never return a negative value too, so make find_deepest_state()
return non-negative values too and drop the default_idle_call()
fallback from call_cpuidle().

This eliminates one branch from the idle loop and makes the governors
and find_deepest_state() handle the case when all states have been
disabled from sysfs consistently.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---

Note: Commit 9c4b2867ed7c (cpuidle: menu: Fix menu_select() for
CPUIDLE_DRIVER_STATE_START == 0) is in linux-next only ATM.

---
 drivers/cpuidle/cpuidle.c |    6 +++---
 kernel/sched/idle.c       |    6 ------
 2 files changed, 3 insertions(+), 9 deletions(-)

Index: linux-pm/drivers/cpuidle/cpuidle.c
===================================================================
--- linux-pm.orig/drivers/cpuidle/cpuidle.c
+++ linux-pm/drivers/cpuidle/cpuidle.c
@@ -79,9 +79,9 @@ static int find_deepest_state(struct cpu
 			      bool freeze)
 {
 	unsigned int latency_req = 0;
-	int i, ret = -ENXIO;
+	int i, ret = 0;
 
-	for (i = 0; i < drv->state_count; i++) {
+	for (i = 1; i < drv->state_count; i++) {
 		struct cpuidle_state *s = &drv->states[i];
 		struct cpuidle_state_usage *su = &dev->states_usage[i];
 
@@ -243,7 +243,7 @@ int cpuidle_enter_state(struct cpuidle_d
  * @drv: the cpuidle driver
  * @dev: the cpuidle device
  *
- * Returns the index of the idle state.
+ * Returns the index of the idle state.  The return value must not be negative.
  */
 int cpuidle_select(struct cpuidle_driver *drv, struct cpuidle_device *dev)
 {
Index: linux-pm/kernel/sched/idle.c
===================================================================
--- linux-pm.orig/kernel/sched/idle.c
+++ linux-pm/kernel/sched/idle.c
@@ -97,12 +97,6 @@ void default_idle_call(void)
 static int call_cpuidle(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 		      int next_state)
 {
-	/* Fall back to the default arch idle method on errors. */
-	if (next_state < 0) {
-		default_idle_call();
-		return next_state;
-	}
-
 	/*
 	 * The idle task must be scheduled, it is pointless to go to idle, just
 	 * update no idle residency and return.

[toc] | [next] | [standalone]

#1311582 — Re: [PATCH 1/2] sched / idle: Drop default_idle_call() fallback from call_cpuidle()

On Sat, Jan 16, 2016 at 12:54:53AM +0100, Rafael J. Wysocki wrote:
> This eliminates one branch from the idle loop and makes the governors
> and find_deepest_state() handle the case when all states have been
> disabled from sysfs consistently.

> Index: linux-pm/kernel/sched/idle.c
> ===================================================================
> --- linux-pm.orig/kernel/sched/idle.c
> +++ linux-pm/kernel/sched/idle.c
> @@ -97,12 +97,6 @@ void default_idle_call(void)
>  static int call_cpuidle(struct cpuidle_driver *drv, struct cpuidle_device *dev,
>  		      int next_state)
>  {
> -	/* Fall back to the default arch idle method on errors. */
> -	if (next_state < 0) {
> -		default_idle_call();
> -		return next_state;
> -	}
> -
>  	/*
>  	 * The idle task must be scheduled, it is pointless to go to idle, just
>  	 * update no idle residency and return.


Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web