[PATCH 10/16] cpuidle: pseries: Convert to hotplug state machine

Subsystems: cpu hotplug, cpu idle time management framework, the rest

STALE3633d

6 messages, 2 authors, 2016-08-24 · open the first message on its own page

[PATCH 10/16] cpuidle: pseries: Convert to hotplug state machine

From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: 2016-08-18 12:58:15

Install the callbacks via the state machine.

Cc: "Rafael J. Wysocki" <redacted>
Cc: Daniel Lezcano <redacted>
Cc: linux-pm@vger.kernel.org
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 drivers/cpuidle/cpuidle-pseries.c | 50 +++++++++++++++++----------------------
 include/linux/cpuhotplug.h        |  1 +
 2 files changed, 23 insertions(+), 28 deletions(-)
diff --git a/drivers/cpuidle/cpuidle-pseries.c b/drivers/cpuidle/cpuidle-pseries.c
index 07135e009d8b..10c4c51cd840 100644
--- a/drivers/cpuidle/cpuidle-pseries.c
+++ b/drivers/cpuidle/cpuidle-pseries.c
@@ -171,39 +171,29 @@ static struct cpuidle_state shared_states[] = {
 		.enter = &shared_cede_loop },
 };
 
-static int pseries_cpuidle_add_cpu_notifier(struct notifier_block *n,
-			unsigned long action, void *hcpu)
+static int pseries_cpuidle_cpu_online(unsigned int cpu)
 {
-	int hotcpu = (unsigned long)hcpu;
-	struct cpuidle_device *dev =
-				per_cpu(cpuidle_devices, hotcpu);
+	struct cpuidle_device *dev = per_cpu(cpuidle_devices, cpu);
 
 	if (dev && cpuidle_get_driver()) {
-		switch (action) {
-		case CPU_ONLINE:
-		case CPU_ONLINE_FROZEN:
-			cpuidle_pause_and_lock();
-			cpuidle_enable_device(dev);
-			cpuidle_resume_and_unlock();
-			break;
-
-		case CPU_DEAD:
-		case CPU_DEAD_FROZEN:
-			cpuidle_pause_and_lock();
-			cpuidle_disable_device(dev);
-			cpuidle_resume_and_unlock();
-			break;
-
-		default:
-			return NOTIFY_DONE;
-		}
+		cpuidle_pause_and_lock();
+		cpuidle_enable_device(dev);
+		cpuidle_resume_and_unlock();
 	}
-	return NOTIFY_OK;
+	return 0;
 }
 
-static struct notifier_block setup_hotplug_notifier = {
-	.notifier_call = pseries_cpuidle_add_cpu_notifier,
-};
+static int pseries_cpuidle_cpu_dead(unsigned int cpu)
+{
+	struct cpuidle_device *dev = per_cpu(cpuidle_devices, cpu);
+
+	if (dev && cpuidle_get_driver()) {
+		cpuidle_pause_and_lock();
+		cpuidle_disable_device(dev);
+		cpuidle_resume_and_unlock();
+	}
+	return 0;
+}
 
 /*
  * pseries_cpuidle_driver_init()
@@ -273,7 +263,11 @@ static int __init pseries_processor_idle_init(void)
 		return retval;
 	}
 
-	register_cpu_notifier(&setup_hotplug_notifier);
+	cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN, "CPUIDLE_PSERIES_ONLINE",
+				  pseries_cpuidle_cpu_online, NULL);
+	cpuhp_setup_state_nocalls(CPUHP_CPUIDLE_PSERIES_DEAD,
+				  "CPUIDLE_PSERIES_DEAD", NULL,
+				  pseries_cpuidle_cpu_dead);
 	printk(KERN_DEBUG "pseries_idle_driver registered\n");
 	return 0;
 }
diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
index 5811954809af..baecc4faf028 100644
--- a/include/linux/cpuhotplug.h
+++ b/include/linux/cpuhotplug.h
@@ -19,6 +19,7 @@ enum cpuhp_state {
 	CPUHP_MM_WRITEBACK_DEAD,
 	CPUHP_SOFTIRQ_DEAD,
 	CPUHP_NET_MVNETA_DEAD,
+	CPUHP_CPUIDLE_PSERIES_DEAD,
 	CPUHP_WORKQUEUE_PREP,
 	CPUHP_POWER_NUMA_PREPARE,
 	CPUHP_HRTIMERS_PREPARE,
-- 
2.9.3

Re: [PATCH 10/16] cpuidle: pseries: Convert to hotplug state machine

From: Daniel Lezcano <hidden>
Date: 2016-08-22 16:16:49

On 08/18/2016 02:57 PM, Sebastian Andrzej Siewior wrote:
quoted hunk
Install the callbacks via the state machine.

Cc: "Rafael J. Wysocki" <redacted>
Cc: Daniel Lezcano <redacted>
Cc: linux-pm@vger.kernel.org
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
index 5811954809af..baecc4faf028 100644
--- a/include/linux/cpuhotplug.h
+++ b/include/linux/cpuhotplug.h
@@ -19,6 +19,7 @@ enum cpuhp_state {
 	CPUHP_MM_WRITEBACK_DEAD,
 	CPUHP_SOFTIRQ_DEAD,
 	CPUHP_NET_MVNETA_DEAD,
+	CPUHP_CPUIDLE_PSERIES_DEAD,
Can't we directly merge these into CPUHP_CPUIDLE_DEAD instead ? Or is it
planned to be done separately ?


-- 
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog

Re: [PATCH 10/16] cpuidle: pseries: Convert to hotplug state machine

From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: 2016-08-22 19:04:17

On 2016-08-22 18:09:47 [+0200], Daniel Lezcano wrote:
On 08/18/2016 02:57 PM, Sebastian Andrzej Siewior wrote:
quoted
Install the callbacks via the state machine.

Cc: "Rafael J. Wysocki" <redacted>
Cc: Daniel Lezcano <redacted>
Cc: linux-pm@vger.kernel.org
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
index 5811954809af..baecc4faf028 100644
--- a/include/linux/cpuhotplug.h
+++ b/include/linux/cpuhotplug.h
@@ -19,6 +19,7 @@ enum cpuhp_state {
 	CPUHP_MM_WRITEBACK_DEAD,
 	CPUHP_SOFTIRQ_DEAD,
 	CPUHP_NET_MVNETA_DEAD,
+	CPUHP_CPUIDLE_PSERIES_DEAD,
Can't we directly merge these into CPUHP_CPUIDLE_DEAD instead ? Or is it
planned to be done separately ?
You mean CPUHP_CPUIDLE_DEAD instead of _PSERIES_DEAD and _POWERNV_DEAD?
We could do that but you would have to ensure that only one CPUIDLE
driver registers itself at a time and for those powerpc drivers it looks
like you could have two registered (not sure about ARM's little/big (if
you could have two of those later at run-time)).
For the ONLINE state we have dynamic allocation of IDs. If it is
possible to rework the code to use only ONLINE & PRE_DOWN instead of
DEAD then we wouldn't have this. I can't say at this point if we do
dynamic allocation of the DEAD IDs.

Sebastian

Re: [PATCH 10/16] cpuidle: pseries: Convert to hotplug state machine

From: Daniel Lezcano <hidden>
Date: 2016-08-23 14:16:20

On 08/22/2016 09:04 PM, Sebastian Andrzej Siewior wrote:
On 2016-08-22 18:09:47 [+0200], Daniel Lezcano wrote:
quoted
On 08/18/2016 02:57 PM, Sebastian Andrzej Siewior wrote:
quoted
Install the callbacks via the state machine.

Cc: "Rafael J. Wysocki" <redacted>
Cc: Daniel Lezcano <redacted>
Cc: linux-pm@vger.kernel.org
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
index 5811954809af..baecc4faf028 100644
--- a/include/linux/cpuhotplug.h
+++ b/include/linux/cpuhotplug.h
@@ -19,6 +19,7 @@ enum cpuhp_state {
 	CPUHP_MM_WRITEBACK_DEAD,
 	CPUHP_SOFTIRQ_DEAD,
 	CPUHP_NET_MVNETA_DEAD,
+	CPUHP_CPUIDLE_PSERIES_DEAD,
Can't we directly merge these into CPUHP_CPUIDLE_DEAD instead ? Or is it
planned to be done separately ?
You mean CPUHP_CPUIDLE_DEAD instead of _PSERIES_DEAD and _POWERNV_DEAD?
Yes. If we can limit the number of duplicating enum for the same purpose
right now, it would be nice.
We could do that but you would have to ensure that only one CPUIDLE
driver registers itself at a time and for those powerpc drivers it looks
like you could have two registered (not sure about ARM's little/big (if
you could have two of those later at run-time)).
At the first glance, I don't think it is possible to register the cpu
hotplug callback twice because the cpuidle drivers are doing:

...
        retval = cpuidle_register(&pseries_idle_driver, NULL);
        if (retval) {
                printk(KERN_DEBUG "Registration of pseries driver
failed.\n");
                return retval;
        }

        register_cpu_notifier(&setup_hotplug_notifier);

So if a previous driver was already registered, cpuidle_register will
fail and register_cpu_notifier won't be hit.

There is the same scenario for intel_idle and processor_idle (acpi).
For the ONLINE state we have dynamic allocation of IDs. If it is
possible to rework the code to use only ONLINE & PRE_DOWN instead of
DEAD then we wouldn't have this. I can't say at this point if we do
dynamic allocation of the DEAD IDs.

Sebastian

-- 
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog

Re: [PATCH 10/16] cpuidle: pseries: Convert to hotplug state machine

From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: 2016-08-23 16:32:16

On 2016-08-23 16:16:12 [+0200], Daniel Lezcano wrote:
Yes. If we can limit the number of duplicating enum for the same purpose
right now, it would be nice.
Okay.
quoted
We could do that but you would have to ensure that only one CPUIDLE
driver registers itself at a time and for those powerpc drivers it looks
like you could have two registered (not sure about ARM's little/big (if
you could have two of those later at run-time)).
At the first glance, I don't think it is possible to register the cpu
hotplug callback twice because the cpuidle drivers are doing:
So if a previous driver was already registered, cpuidle_register will
fail and register_cpu_notifier won't be hit.

There is the same scenario for intel_idle and processor_idle (acpi).
I tried to preserve everything as-is during the conversation. However if
you are explicitly asking for this and you are sure that it will work
then you can get it. No problem :)

Sebastian

[PATCH 10/16 v2] cpuidle: pseries: Convert to hotplug state machine

From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: 2016-08-24 09:09:41

Install the callbacks via the state machine.

v1…v2: Use only CPUHP_CPUIDLE_DEAD (requested by Daniel Lezcano)

Cc: "Rafael J. Wysocki" <redacted>
Cc: Daniel Lezcano <redacted>
Cc: linux-pm@vger.kernel.org
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 drivers/cpuidle/cpuidle-pseries.c | 53 ++++++++++++++++++---------------------
 include/linux/cpuhotplug.h        |  1 +
 2 files changed, 26 insertions(+), 28 deletions(-)
diff --git a/drivers/cpuidle/cpuidle-pseries.c b/drivers/cpuidle/cpuidle-pseries.c
index 07135e009d8b..abbce8f96d93 100644
--- a/drivers/cpuidle/cpuidle-pseries.c
+++ b/drivers/cpuidle/cpuidle-pseries.c
@@ -171,39 +171,29 @@ static struct cpuidle_state shared_states[] = {
 		.enter = &shared_cede_loop },
 };
 
-static int pseries_cpuidle_add_cpu_notifier(struct notifier_block *n,
-			unsigned long action, void *hcpu)
+static int pseries_cpuidle_cpu_online(unsigned int cpu)
 {
-	int hotcpu = (unsigned long)hcpu;
-	struct cpuidle_device *dev =
-				per_cpu(cpuidle_devices, hotcpu);
+	struct cpuidle_device *dev = per_cpu(cpuidle_devices, cpu);
 
 	if (dev && cpuidle_get_driver()) {
-		switch (action) {
-		case CPU_ONLINE:
-		case CPU_ONLINE_FROZEN:
-			cpuidle_pause_and_lock();
-			cpuidle_enable_device(dev);
-			cpuidle_resume_and_unlock();
-			break;
-
-		case CPU_DEAD:
-		case CPU_DEAD_FROZEN:
-			cpuidle_pause_and_lock();
-			cpuidle_disable_device(dev);
-			cpuidle_resume_and_unlock();
-			break;
-
-		default:
-			return NOTIFY_DONE;
-		}
+		cpuidle_pause_and_lock();
+		cpuidle_enable_device(dev);
+		cpuidle_resume_and_unlock();
 	}
-	return NOTIFY_OK;
+	return 0;
 }
 
-static struct notifier_block setup_hotplug_notifier = {
-	.notifier_call = pseries_cpuidle_add_cpu_notifier,
-};
+static int pseries_cpuidle_cpu_dead(unsigned int cpu)
+{
+	struct cpuidle_device *dev = per_cpu(cpuidle_devices, cpu);
+
+	if (dev && cpuidle_get_driver()) {
+		cpuidle_pause_and_lock();
+		cpuidle_disable_device(dev);
+		cpuidle_resume_and_unlock();
+	}
+	return 0;
+}
 
 /*
  * pseries_cpuidle_driver_init()
@@ -273,7 +263,14 @@ static int __init pseries_processor_idle_init(void)
 		return retval;
 	}
 
-	register_cpu_notifier(&setup_hotplug_notifier);
+	retval = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
+					   "CPUIDLE_PSERIES_ONLINE",
+					   pseries_cpuidle_cpu_online, NULL);
+	WARN_ON(retval < 0);
+	retval = cpuhp_setup_state_nocalls(CPUHP_CPUIDLE_DEAD,
+					   "CPUIDLE_PSERIES_DEAD", NULL,
+					   pseries_cpuidle_cpu_dead);
+	WARN_ON(retval < 0);
 	printk(KERN_DEBUG "pseries_idle_driver registered\n");
 	return 0;
 }
diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
index 5811954809af..14b58a9d1450 100644
--- a/include/linux/cpuhotplug.h
+++ b/include/linux/cpuhotplug.h
@@ -19,6 +19,7 @@ enum cpuhp_state {
 	CPUHP_MM_WRITEBACK_DEAD,
 	CPUHP_SOFTIRQ_DEAD,
 	CPUHP_NET_MVNETA_DEAD,
+	CPUHP_CPUIDLE_DEAD,
 	CPUHP_WORKQUEUE_PREP,
 	CPUHP_POWER_NUMA_PREPARE,
 	CPUHP_HRTIMERS_PREPARE,
-- 
2.9.3
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help