linux-acpi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [RESEND] X86 cpuidle cleanups
@ 2012-01-17 16:20 Thomas Renninger
  2012-01-17 16:20 ` [PATCH 1/3] Revert "intel_idle: disable auto_demotion for hotplugged CPUs" Thomas Renninger
                   ` (2 more replies)
  0 siblings, 3 replies; 9+ messages in thread
From: Thomas Renninger @ 2012-01-17 16:20 UTC (permalink / raw)
  To: lenb, linux-acpi; +Cc: deepthi

This is a rebase of some recently posted intel_idle processor_idle
related patches which needed refreshing, due to recent commits.

[PATCH 1/3] Revert "intel_idle: disable auto_demotion for hotplugged
-> because I think it's wrong/unneeded. I commented on the initial
posting.

I tested these patches before I had to rebase them.
Now they are compile tested, but adjusting the code wasn't that hard.

Would be great if these could get picked up for 3.3 inclusion.

Thanks,

   Thomas


^ permalink raw reply	[flat|nested] 9+ messages in thread

* [PATCH 1/3] Revert "intel_idle: disable auto_demotion for hotplugged CPUs"
  2012-01-17 16:20 [RESEND] X86 cpuidle cleanups Thomas Renninger
@ 2012-01-17 16:20 ` Thomas Renninger
  2012-01-17 16:20 ` [PATCH 2/3] ACPI processor: Remove unneeded variable passed by acpi_processor_hotadd_init Thomas Renninger
  2012-01-17 16:20 ` [PATCH 3/3] intel_idle: Split up and provide per CPU initialization func Thomas Renninger
  2 siblings, 0 replies; 9+ messages in thread
From: Thomas Renninger @ 2012-01-17 16:20 UTC (permalink / raw)
  To: lenb, linux-acpi; +Cc: deepthi, Thomas Renninger

This reverts commit e5f03b06dc99348df53118c6ef7b51a8611a7256.

Signed-off-by: Thomas Renninger <trenn@suse.de>
---
 drivers/idle/intel_idle.c |   53 +++++++++++++++++----------------------------
 1 files changed, 20 insertions(+), 33 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index e56ae0b..91aff18 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -296,42 +296,33 @@ static void __setup_broadcast_timer(void *arg)
 	clockevents_notify(reason, &cpu);
 }
 
-static void auto_demotion_disable(void *dummy)
-{
-	unsigned long long msr_bits;
-
-	rdmsrl(MSR_NHM_SNB_PKG_CST_CFG_CTL, msr_bits);
-	msr_bits &= ~auto_demotion_disable_flags;
-	wrmsrl(MSR_NHM_SNB_PKG_CST_CFG_CTL, msr_bits);
-}
-
-static void __intel_idle_notify_handler(void *arg)
-{
-	if (auto_demotion_disable_flags)
-		auto_demotion_disable(NULL);
-
-	if (lapic_timer_reliable_states != LAPIC_TIMER_ALWAYS_RELIABLE)
-		__setup_broadcast_timer((void *)true);
-}
-
-static int setup_intelidle_cpuhp_notify(struct notifier_block *n,
+static int setup_broadcast_cpuhp_notify(struct notifier_block *n,
 		unsigned long action, void *hcpu)
 {
 	int hotcpu = (unsigned long)hcpu;
 
 	switch (action & 0xf) {
 	case CPU_ONLINE:
-		smp_call_function_single(hotcpu, __intel_idle_notify_handler,
-			NULL, 1);
+		smp_call_function_single(hotcpu, __setup_broadcast_timer,
+			(void *)true, 1);
 		break;
 	}
 	return NOTIFY_OK;
 }
 
-static struct notifier_block setup_intelidle_notifier = {
-	.notifier_call = setup_intelidle_cpuhp_notify,
+static struct notifier_block setup_broadcast_notifier = {
+	.notifier_call = setup_broadcast_cpuhp_notify,
 };
 
+static void auto_demotion_disable(void *dummy)
+{
+	unsigned long long msr_bits;
+
+	rdmsrl(MSR_NHM_SNB_PKG_CST_CFG_CTL, msr_bits);
+	msr_bits &= ~auto_demotion_disable_flags;
+	wrmsrl(MSR_NHM_SNB_PKG_CST_CFG_CTL, msr_bits);
+}
+
 /*
  * intel_idle_probe()
  */
@@ -402,8 +393,10 @@ static int intel_idle_probe(void)
 
 	if (boot_cpu_has(X86_FEATURE_ARAT))	/* Always Reliable APIC Timer */
 		lapic_timer_reliable_states = LAPIC_TIMER_ALWAYS_RELIABLE;
-	else
+	else {
 		on_each_cpu(__setup_broadcast_timer, (void *)true, 1);
+		register_cpu_notifier(&setup_broadcast_notifier);
+	}
 
 	pr_debug(PREFIX "v" INTEL_IDLE_VERSION
 		" model 0x%X\n", boot_cpu_data.x86_model);
@@ -566,10 +559,6 @@ static int __init intel_idle_init(void)
 		return retval;
 	}
 
-	if (auto_demotion_disable_flags || lapic_timer_reliable_states !=
-	    LAPIC_TIMER_ALWAYS_RELIABLE)
-		register_cpu_notifier(&setup_intelidle_notifier);
-
 	return 0;
 }
 
@@ -578,12 +567,10 @@ static void __exit intel_idle_exit(void)
 	intel_idle_cpuidle_devices_uninit();
 	cpuidle_unregister_driver(&intel_idle_driver);
 
-	if (lapic_timer_reliable_states != LAPIC_TIMER_ALWAYS_RELIABLE)
+	if (lapic_timer_reliable_states != LAPIC_TIMER_ALWAYS_RELIABLE) {
 		on_each_cpu(__setup_broadcast_timer, (void *)false, 1);
-
-	if (auto_demotion_disable_flags || lapic_timer_reliable_states !=
-	    LAPIC_TIMER_ALWAYS_RELIABLE)
-		unregister_cpu_notifier(&setup_intelidle_notifier);
+		unregister_cpu_notifier(&setup_broadcast_notifier);
+	}
 
 	return;
 }
-- 
1.7.6.1


^ permalink raw reply related	[flat|nested] 9+ messages in thread

* [PATCH 2/3] ACPI processor: Remove unneeded variable passed by acpi_processor_hotadd_init
  2012-01-17 16:20 [RESEND] X86 cpuidle cleanups Thomas Renninger
  2012-01-17 16:20 ` [PATCH 1/3] Revert "intel_idle: disable auto_demotion for hotplugged CPUs" Thomas Renninger
@ 2012-01-17 16:20 ` Thomas Renninger
  2012-01-17 16:22   ` Thomas Renninger
  2012-01-17 16:29   ` Konrad Rzeszutek Wilk
  2012-01-17 16:20 ` [PATCH 3/3] intel_idle: Split up and provide per CPU initialization func Thomas Renninger
  2 siblings, 2 replies; 9+ messages in thread
From: Thomas Renninger @ 2012-01-17 16:20 UTC (permalink / raw)
  To: lenb, linux-acpi; +Cc: deepthi, Thomas Renninger, Bjorn Helgaas, Jiang, Yunhong

This is a very small part taken from patches which afaik
are coming from Yunhong Jiang for Xen.
Xen CPU hotplug things not existing in Linus kernel yet were
removed.

Cleanup only: no functional change.

Signed-off-by: Thomas Renninger <trenn@suse.de>
CC: Len Brown <lenb@kernel.org>
CC: linux-acpi@vger.kernel.org
CC: Bjorn Helgaas <bhelgaas@google.com>
CC: Jiang, Yunhong <yunhong.jiang@intel.com>
---
 drivers/acpi/processor_driver.c |   17 ++++++++---------
 1 files changed, 8 insertions(+), 9 deletions(-)

diff --git a/drivers/acpi/processor_driver.c b/drivers/acpi/processor_driver.c
index 90719d1..2291aa4 100644
--- a/drivers/acpi/processor_driver.c
+++ b/drivers/acpi/processor_driver.c
@@ -82,7 +82,7 @@ MODULE_LICENSE("GPL");
 static int acpi_processor_add(struct acpi_device *device);
 static int acpi_processor_remove(struct acpi_device *device, int type);
 static void acpi_processor_notify(struct acpi_device *device, u32 event);
-static acpi_status acpi_processor_hotadd_init(acpi_handle handle, int *p_cpu);
+static acpi_status acpi_processor_hotadd_init(struct acpi_processor *pr);
 static int acpi_processor_handle_eject(struct acpi_processor *pr);
 
 
@@ -324,10 +324,8 @@ static int acpi_processor_get_info(struct acpi_device *device)
 	 *  they are physically not present.
 	 */
 	if (pr->id == -1) {
-		if (ACPI_FAILURE
-		    (acpi_processor_hotadd_init(pr->handle, &pr->id))) {
+		if (ACPI_FAILURE (acpi_processor_hotadd_init(pr->handle)))
 			return -ENODEV;
-		}
 	}
 	/*
 	 * On some boxes several processors use the same processor bus id.
@@ -721,18 +719,19 @@ processor_walk_namespace_cb(acpi_handle handle,
 	return (AE_OK);
 }
 
-static acpi_status acpi_processor_hotadd_init(acpi_handle handle, int *p_cpu)
+static acpi_status acpi_processor_hotadd_init(struct acpi_processor *pr)
 {
+	acpi_handle handle = pr->handle;
 
 	if (!is_processor_present(handle)) {
 		return AE_ERROR;
 	}
 
-	if (acpi_map_lsapic(handle, p_cpu))
+	if (acpi_map_lsapic(handle, &pr-id))
 		return AE_ERROR;
 
-	if (arch_register_cpu(*p_cpu)) {
-		acpi_unmap_lsapic(*p_cpu);
+	if (arch_register_cpu(pr->id)) {
+		acpi_unmap_lsapic(pr->id);
 		return AE_ERROR;
 	}
 
@@ -749,7 +748,7 @@ static int acpi_processor_handle_eject(struct acpi_processor *pr)
 	return (0);
 }
 #else
-static acpi_status acpi_processor_hotadd_init(acpi_handle handle, int *p_cpu)
+static acpi_status acpi_processor_hotadd_init(struct acpi_processor *pr)
 {
 	return AE_ERROR;
 }
-- 
1.7.6.1


^ permalink raw reply related	[flat|nested] 9+ messages in thread

* [PATCH 3/3] intel_idle: Split up and provide per CPU initialization func
  2012-01-17 16:20 [RESEND] X86 cpuidle cleanups Thomas Renninger
  2012-01-17 16:20 ` [PATCH 1/3] Revert "intel_idle: disable auto_demotion for hotplugged CPUs" Thomas Renninger
  2012-01-17 16:20 ` [PATCH 2/3] ACPI processor: Remove unneeded variable passed by acpi_processor_hotadd_init Thomas Renninger
@ 2012-01-17 16:20 ` Thomas Renninger
  2 siblings, 0 replies; 9+ messages in thread
From: Thomas Renninger @ 2012-01-17 16:20 UTC (permalink / raw)
  To: lenb, linux-acpi; +Cc: deepthi, Thomas Renninger, Shaohua Li, Andrew Morton

Function split up, should have no functional change.

Provides entry point for physically hotplugged CPUs
to initialize and activate cpuidle.

Signed-off-by: Thomas Renninger <trenn@suse.de>
CC: Len Brown <lenb@kernel.org>
CC: linux-acpi@vger.kernel.org
CC: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
CC: Shaohua Li <shaohua.li@intel.com>
CC: Andrew Morton <akpm@linux-foundation.org>
---
 drivers/idle/intel_idle.c |   84 +++++++++++++++++++++++----------------------
 include/linux/cpuidle.h   |    7 ++++
 2 files changed, 50 insertions(+), 41 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 91aff18..be96715 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -478,64 +478,60 @@ static int intel_idle_cpuidle_driver_init(void)
 
 
 /*
- * intel_idle_cpuidle_devices_init()
+ * intel_idle_cpu_init()
  * allocate, initialize, register cpuidle_devices
+ * @cpu: cpu/core to initialize
  */
-static int intel_idle_cpuidle_devices_init(void)
+int intel_idle_cpu_init(int cpu)
 {
-	int i, cstate;
+	int cstate;
 	struct cpuidle_device *dev;
 
-	intel_idle_cpuidle_devices = alloc_percpu(struct cpuidle_device);
-	if (intel_idle_cpuidle_devices == NULL)
-		return -ENOMEM;
-
-	for_each_online_cpu(i) {
-		dev = per_cpu_ptr(intel_idle_cpuidle_devices, i);
-
-		dev->state_count = 1;
+	dev = per_cpu_ptr(intel_idle_cpuidle_devices, cpu);
 
-		for (cstate = 1; cstate < MWAIT_MAX_NUM_CSTATES; ++cstate) {
-			int num_substates;
+	dev->state_count = 1;
 
-			if (cstate > max_cstate) {
-				printk(PREFIX "max_cstate %d reached\n",
-					max_cstate);
-				break;
-			}
+	for (cstate = 1; cstate < MWAIT_MAX_NUM_CSTATES; ++cstate) {
+		int num_substates;
 
-			/* does the state exist in CPUID.MWAIT? */
-			num_substates = (mwait_substates >> ((cstate) * 4))
-						& MWAIT_SUBSTATE_MASK;
-			if (num_substates == 0)
-				continue;
-			/* is the state not enabled? */
-			if (cpuidle_state_table[cstate].enter == NULL) {
-				continue;
-			}
+		if (cstate > max_cstate) {
+			printk(PREFIX "max_cstate %d reached\n",
+			       max_cstate);
+			break;
+		}
 
-			dev->states_usage[dev->state_count].driver_data =
-				(void *)get_driver_data(cstate);
+		/* does the state exist in CPUID.MWAIT? */
+		num_substates = (mwait_substates >> ((cstate) * 4))
+			& MWAIT_SUBSTATE_MASK;
+		if (num_substates == 0)
+			continue;
+		/* is the state not enabled? */
+		if (cpuidle_state_table[cstate].enter == NULL) {
+			continue;
+		}
+		dev->states_usage[dev->state_count].driver_data =
+			(void *)get_driver_data(cstate);
 
 			dev->state_count += 1;
 		}
+	dev->cpu = cpu;
 
-		dev->cpu = i;
-		if (cpuidle_register_device(dev)) {
-			pr_debug(PREFIX "cpuidle_register_device %d failed!\n",
-				 i);
-			intel_idle_cpuidle_devices_uninit();
-			return -EIO;
-		}
+	if (cpuidle_register_device(dev)) {
+		pr_debug(PREFIX "cpuidle_register_device %d failed!\n", cpu);
+		intel_idle_cpuidle_devices_uninit();
+		return -EIO;
 	}
 
+ 	if (auto_demotion_disable_flags)
+		smp_call_function_single(cpu, auto_demotion_disable, NULL, 1);
+
 	return 0;
 }
 
 
 static int __init intel_idle_init(void)
 {
-	int retval;
+	int retval, i;
 
 	/* Do not load intel_idle at all for now if idle= is passed */
 	if (boot_option_idle_override != IDLE_NO_OVERRIDE)
@@ -553,10 +549,16 @@ static int __init intel_idle_init(void)
 		return retval;
 	}
 
-	retval = intel_idle_cpuidle_devices_init();
-	if (retval) {
-		cpuidle_unregister_driver(&intel_idle_driver);
-		return retval;
+	intel_idle_cpuidle_devices = alloc_percpu(struct cpuidle_device);
+	if (intel_idle_cpuidle_devices == NULL)
+		return -ENOMEM;
+
+	for_each_online_cpu(i) {
+		retval = intel_idle_cpu_init(i);
+		if (retval) {
+			cpuidle_unregister_driver(&intel_idle_driver);
+			return retval;
+		}
 	}
 
 	return 0;
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index 7408af8..93df66e 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -188,7 +188,14 @@ struct cpuidle_governor {
 extern int cpuidle_register_governor(struct cpuidle_governor *gov);
 extern void cpuidle_unregister_governor(struct cpuidle_governor *gov);
 
+#ifdef CONFIG_INTEL_IDLE
+extern int intel_idle_cpu_init(int cpu);
 #else
+static inline int intel_idle_cpu_init(int cpu) { return -1; }
+#endif
+
+#else
+static inline int intel_idle_cpu_init(int cpu) { return -1; }
 
 static inline int cpuidle_register_governor(struct cpuidle_governor *gov)
 {return 0;}
-- 
1.7.6.1


^ permalink raw reply related	[flat|nested] 9+ messages in thread

* Re: [PATCH 2/3] ACPI processor: Remove unneeded variable passed by acpi_processor_hotadd_init
  2012-01-17 16:20 ` [PATCH 2/3] ACPI processor: Remove unneeded variable passed by acpi_processor_hotadd_init Thomas Renninger
@ 2012-01-17 16:22   ` Thomas Renninger
  2012-01-17 16:29   ` Konrad Rzeszutek Wilk
  1 sibling, 0 replies; 9+ messages in thread
From: Thomas Renninger @ 2012-01-17 16:22 UTC (permalink / raw)
  To: lenb; +Cc: linux-acpi, deepthi, Bjorn Helgaas, Jiang, Yunhong

On Tuesday, January 17, 2012 05:20:29 PM Thomas Renninger wrote:
...
> +	if (acpi_map_lsapic(handle, &pr-id))
oops, blindly used guilt import.
I resend.
Sorry,

   Thomas

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: [PATCH 2/3] ACPI processor: Remove unneeded variable passed by acpi_processor_hotadd_init
  2012-01-17 16:20 ` [PATCH 2/3] ACPI processor: Remove unneeded variable passed by acpi_processor_hotadd_init Thomas Renninger
  2012-01-17 16:22   ` Thomas Renninger
@ 2012-01-17 16:29   ` Konrad Rzeszutek Wilk
  2012-01-17 16:47     ` Thomas Renninger
  1 sibling, 1 reply; 9+ messages in thread
From: Konrad Rzeszutek Wilk @ 2012-01-17 16:29 UTC (permalink / raw)
  To: Thomas Renninger; +Cc: lenb, linux-acpi, deepthi, Bjorn Helgaas, Jiang, Yunhong

On Tue, Jan 17, 2012 at 05:20:29PM +0100, Thomas Renninger wrote:
> This is a very small part taken from patches which afaik
> are coming from Yunhong Jiang for Xen.
> Xen CPU hotplug things not existing in Linus kernel yet were
> removed.

Could you point out the git commit please?
I did this:
 git log --grep=Yunhong drivers/acpi/processor_driver.c

And could not find it?
> 
> Cleanup only: no functional change.
> 
> Signed-off-by: Thomas Renninger <trenn@suse.de>
> CC: Len Brown <lenb@kernel.org>
> CC: linux-acpi@vger.kernel.org
> CC: Bjorn Helgaas <bhelgaas@google.com>
> CC: Jiang, Yunhong <yunhong.jiang@intel.com>
> ---
>  drivers/acpi/processor_driver.c |   17 ++++++++---------
>  1 files changed, 8 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/acpi/processor_driver.c b/drivers/acpi/processor_driver.c
> index 90719d1..2291aa4 100644
> --- a/drivers/acpi/processor_driver.c
> +++ b/drivers/acpi/processor_driver.c
> @@ -82,7 +82,7 @@ MODULE_LICENSE("GPL");
>  static int acpi_processor_add(struct acpi_device *device);
>  static int acpi_processor_remove(struct acpi_device *device, int type);
>  static void acpi_processor_notify(struct acpi_device *device, u32 event);
> -static acpi_status acpi_processor_hotadd_init(acpi_handle handle, int *p_cpu);
> +static acpi_status acpi_processor_hotadd_init(struct acpi_processor *pr);
>  static int acpi_processor_handle_eject(struct acpi_processor *pr);
>  
>  
> @@ -324,10 +324,8 @@ static int acpi_processor_get_info(struct acpi_device *device)
>  	 *  they are physically not present.
>  	 */
>  	if (pr->id == -1) {
> -		if (ACPI_FAILURE
> -		    (acpi_processor_hotadd_init(pr->handle, &pr->id))) {
> +		if (ACPI_FAILURE (acpi_processor_hotadd_init(pr->handle)))
>  			return -ENODEV;
> -		}
>  	}
>  	/*
>  	 * On some boxes several processors use the same processor bus id.
> @@ -721,18 +719,19 @@ processor_walk_namespace_cb(acpi_handle handle,
>  	return (AE_OK);
>  }
>  
> -static acpi_status acpi_processor_hotadd_init(acpi_handle handle, int *p_cpu)
> +static acpi_status acpi_processor_hotadd_init(struct acpi_processor *pr)
>  {
> +	acpi_handle handle = pr->handle;
>  
>  	if (!is_processor_present(handle)) {
>  		return AE_ERROR;
>  	}
>  
> -	if (acpi_map_lsapic(handle, p_cpu))
> +	if (acpi_map_lsapic(handle, &pr-id))
>  		return AE_ERROR;
>  
> -	if (arch_register_cpu(*p_cpu)) {
> -		acpi_unmap_lsapic(*p_cpu);
> +	if (arch_register_cpu(pr->id)) {
> +		acpi_unmap_lsapic(pr->id);
>  		return AE_ERROR;
>  	}
>  
> @@ -749,7 +748,7 @@ static int acpi_processor_handle_eject(struct acpi_processor *pr)
>  	return (0);
>  }
>  #else
> -static acpi_status acpi_processor_hotadd_init(acpi_handle handle, int *p_cpu)
> +static acpi_status acpi_processor_hotadd_init(struct acpi_processor *pr)
>  {
>  	return AE_ERROR;
>  }
> -- 
> 1.7.6.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [PATCH 3/3] intel_idle: Split up and provide per CPU initialization func
  2012-01-17 16:31 [Resend] X86 cpuidle cleanups V2 Thomas Renninger
@ 2012-01-17 16:31 ` Thomas Renninger
  0 siblings, 0 replies; 9+ messages in thread
From: Thomas Renninger @ 2012-01-17 16:31 UTC (permalink / raw)
  To: linux-acpi, lenb; +Cc: deepthi, shaohua.li, Thomas Renninger, Andrew Morton

Function split up, should have no functional change.

Provides entry point for physically hotplugged CPUs
to initialize and activate cpuidle.

Signed-off-by: Thomas Renninger <trenn@suse.de>
CC: Len Brown <lenb@kernel.org>
CC: linux-acpi@vger.kernel.org
CC: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
CC: Shaohua Li <shaohua.li@intel.com>
CC: Andrew Morton <akpm@linux-foundation.org>
---
 drivers/idle/intel_idle.c |   84 +++++++++++++++++++++++----------------------
 include/linux/cpuidle.h   |    7 ++++
 2 files changed, 50 insertions(+), 41 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 91aff18..be96715 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -478,64 +478,60 @@ static int intel_idle_cpuidle_driver_init(void)
 
 
 /*
- * intel_idle_cpuidle_devices_init()
+ * intel_idle_cpu_init()
  * allocate, initialize, register cpuidle_devices
+ * @cpu: cpu/core to initialize
  */
-static int intel_idle_cpuidle_devices_init(void)
+int intel_idle_cpu_init(int cpu)
 {
-	int i, cstate;
+	int cstate;
 	struct cpuidle_device *dev;
 
-	intel_idle_cpuidle_devices = alloc_percpu(struct cpuidle_device);
-	if (intel_idle_cpuidle_devices == NULL)
-		return -ENOMEM;
-
-	for_each_online_cpu(i) {
-		dev = per_cpu_ptr(intel_idle_cpuidle_devices, i);
-
-		dev->state_count = 1;
+	dev = per_cpu_ptr(intel_idle_cpuidle_devices, cpu);
 
-		for (cstate = 1; cstate < MWAIT_MAX_NUM_CSTATES; ++cstate) {
-			int num_substates;
+	dev->state_count = 1;
 
-			if (cstate > max_cstate) {
-				printk(PREFIX "max_cstate %d reached\n",
-					max_cstate);
-				break;
-			}
+	for (cstate = 1; cstate < MWAIT_MAX_NUM_CSTATES; ++cstate) {
+		int num_substates;
 
-			/* does the state exist in CPUID.MWAIT? */
-			num_substates = (mwait_substates >> ((cstate) * 4))
-						& MWAIT_SUBSTATE_MASK;
-			if (num_substates == 0)
-				continue;
-			/* is the state not enabled? */
-			if (cpuidle_state_table[cstate].enter == NULL) {
-				continue;
-			}
+		if (cstate > max_cstate) {
+			printk(PREFIX "max_cstate %d reached\n",
+			       max_cstate);
+			break;
+		}
 
-			dev->states_usage[dev->state_count].driver_data =
-				(void *)get_driver_data(cstate);
+		/* does the state exist in CPUID.MWAIT? */
+		num_substates = (mwait_substates >> ((cstate) * 4))
+			& MWAIT_SUBSTATE_MASK;
+		if (num_substates == 0)
+			continue;
+		/* is the state not enabled? */
+		if (cpuidle_state_table[cstate].enter == NULL) {
+			continue;
+		}
+		dev->states_usage[dev->state_count].driver_data =
+			(void *)get_driver_data(cstate);
 
 			dev->state_count += 1;
 		}
+	dev->cpu = cpu;
 
-		dev->cpu = i;
-		if (cpuidle_register_device(dev)) {
-			pr_debug(PREFIX "cpuidle_register_device %d failed!\n",
-				 i);
-			intel_idle_cpuidle_devices_uninit();
-			return -EIO;
-		}
+	if (cpuidle_register_device(dev)) {
+		pr_debug(PREFIX "cpuidle_register_device %d failed!\n", cpu);
+		intel_idle_cpuidle_devices_uninit();
+		return -EIO;
 	}
 
+ 	if (auto_demotion_disable_flags)
+		smp_call_function_single(cpu, auto_demotion_disable, NULL, 1);
+
 	return 0;
 }
 
 
 static int __init intel_idle_init(void)
 {
-	int retval;
+	int retval, i;
 
 	/* Do not load intel_idle at all for now if idle= is passed */
 	if (boot_option_idle_override != IDLE_NO_OVERRIDE)
@@ -553,10 +549,16 @@ static int __init intel_idle_init(void)
 		return retval;
 	}
 
-	retval = intel_idle_cpuidle_devices_init();
-	if (retval) {
-		cpuidle_unregister_driver(&intel_idle_driver);
-		return retval;
+	intel_idle_cpuidle_devices = alloc_percpu(struct cpuidle_device);
+	if (intel_idle_cpuidle_devices == NULL)
+		return -ENOMEM;
+
+	for_each_online_cpu(i) {
+		retval = intel_idle_cpu_init(i);
+		if (retval) {
+			cpuidle_unregister_driver(&intel_idle_driver);
+			return retval;
+		}
 	}
 
 	return 0;
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index 7408af8..93df66e 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -188,7 +188,14 @@ struct cpuidle_governor {
 extern int cpuidle_register_governor(struct cpuidle_governor *gov);
 extern void cpuidle_unregister_governor(struct cpuidle_governor *gov);
 
+#ifdef CONFIG_INTEL_IDLE
+extern int intel_idle_cpu_init(int cpu);
 #else
+static inline int intel_idle_cpu_init(int cpu) { return -1; }
+#endif
+
+#else
+static inline int intel_idle_cpu_init(int cpu) { return -1; }
 
 static inline int cpuidle_register_governor(struct cpuidle_governor *gov)
 {return 0;}
-- 
1.7.6.1


^ permalink raw reply related	[flat|nested] 9+ messages in thread

* Re: [PATCH 2/3] ACPI processor: Remove unneeded variable passed by acpi_processor_hotadd_init
  2012-01-17 16:29   ` Konrad Rzeszutek Wilk
@ 2012-01-17 16:47     ` Thomas Renninger
  2012-01-17 16:52       ` Konrad Rzeszutek Wilk
  0 siblings, 1 reply; 9+ messages in thread
From: Thomas Renninger @ 2012-01-17 16:47 UTC (permalink / raw)
  To: Konrad Rzeszutek Wilk
  Cc: lenb, linux-acpi, deepthi, Bjorn Helgaas, Jiang, Yunhong

On Tuesday, January 17, 2012 05:29:51 PM Konrad Rzeszutek Wilk wrote:
> On Tue, Jan 17, 2012 at 05:20:29PM +0100, Thomas Renninger wrote:
> > This is a very small part taken from patches which afaik
> > are coming from Yunhong Jiang for Xen.
> > Xen CPU hotplug things not existing in Linus kernel yet were
> > removed.
> 
> Could you point out the git commit please?
> I did this:
>  git log --grep=Yunhong drivers/acpi/processor_driver.c
> 
> And could not find it?
Jiang may know in which Xen Linux kernel repo this went in.
I grepped the commit from our distri where Jiang was
mentioned as author.

Anyway, it's an improvement (one param less)
and makes things easier when dealing with ACPI CPU hotplug.

   Thomas


^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: [PATCH 2/3] ACPI processor: Remove unneeded variable passed by acpi_processor_hotadd_init
  2012-01-17 16:47     ` Thomas Renninger
@ 2012-01-17 16:52       ` Konrad Rzeszutek Wilk
  0 siblings, 0 replies; 9+ messages in thread
From: Konrad Rzeszutek Wilk @ 2012-01-17 16:52 UTC (permalink / raw)
  To: Thomas Renninger; +Cc: lenb, linux-acpi, deepthi, Bjorn Helgaas, Jiang, Yunhong

On Tue, Jan 17, 2012 at 05:47:53PM +0100, Thomas Renninger wrote:
> On Tuesday, January 17, 2012 05:29:51 PM Konrad Rzeszutek Wilk wrote:
> > On Tue, Jan 17, 2012 at 05:20:29PM +0100, Thomas Renninger wrote:
> > > This is a very small part taken from patches which afaik
> > > are coming from Yunhong Jiang for Xen.
> > > Xen CPU hotplug things not existing in Linus kernel yet were
> > > removed.
> > 
> > Could you point out the git commit please?
> > I did this:
> >  git log --grep=Yunhong drivers/acpi/processor_driver.c
> > 
> > And could not find it?
> Jiang may know in which Xen Linux kernel repo this went in.
> I grepped the commit from our distri where Jiang was
> mentioned as author.
> 
> Anyway, it's an improvement (one param less)
> and makes things easier when dealing with ACPI CPU hotplug.

Then please provide that description in the git commit.

How does it make it "easier" ?

^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~2012-01-17 16:54 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-01-17 16:20 [RESEND] X86 cpuidle cleanups Thomas Renninger
2012-01-17 16:20 ` [PATCH 1/3] Revert "intel_idle: disable auto_demotion for hotplugged CPUs" Thomas Renninger
2012-01-17 16:20 ` [PATCH 2/3] ACPI processor: Remove unneeded variable passed by acpi_processor_hotadd_init Thomas Renninger
2012-01-17 16:22   ` Thomas Renninger
2012-01-17 16:29   ` Konrad Rzeszutek Wilk
2012-01-17 16:47     ` Thomas Renninger
2012-01-17 16:52       ` Konrad Rzeszutek Wilk
2012-01-17 16:20 ` [PATCH 3/3] intel_idle: Split up and provide per CPU initialization func Thomas Renninger
  -- strict thread matches above, loose matches on Subject: below --
2012-01-17 16:31 [Resend] X86 cpuidle cleanups V2 Thomas Renninger
2012-01-17 16:31 ` [PATCH 3/3] intel_idle: Split up and provide per CPU initialization func Thomas Renninger

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).