idle patches for 2.6.38-merge

idle patches for 2.6.38-merge

[Len Brown]

Here is my queue of idle patches for 2.6.38.
Please speak up if you see troubles with any of them.

thanks
Len Brown, Intel Open Source Technology Center

[PATCH 01/12] intel_idle: update Sandy Bridge core C-state residency targets

[Len Brown]

From: Len Brown <len.brown@intel.com>

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/intel_idle.c |    8 ++++----
 1 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index c131d58..94a6526 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -122,7 +122,7 @@ static struct cpuidle_state snb_cstates[MWAIT_MAX_NUM_CSTATES] = {
 		.driver_data = (void *) 0x00,
 		.flags = CPUIDLE_FLAG_TIME_VALID,
 		.exit_latency = 1,
-		.target_residency = 4,
+		.target_residency = 1,
 		.enter = &intel_idle },
 	{ /* MWAIT C2 */
 		.name = "SNB-C3",
@@ -130,7 +130,7 @@ static struct cpuidle_state snb_cstates[MWAIT_MAX_NUM_CSTATES] = {
 		.driver_data = (void *) 0x10,
 		.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
 		.exit_latency = 80,
-		.target_residency = 160,
+		.target_residency = 211,
 		.enter = &intel_idle },
 	{ /* MWAIT C3 */
 		.name = "SNB-C6",
@@ -138,7 +138,7 @@ static struct cpuidle_state snb_cstates[MWAIT_MAX_NUM_CSTATES] = {
 		.driver_data = (void *) 0x20,
 		.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
 		.exit_latency = 104,
-		.target_residency = 208,
+		.target_residency = 345,
 		.enter = &intel_idle },
 	{ /* MWAIT C4 */
 		.name = "SNB-C7",
@@ -146,7 +146,7 @@ static struct cpuidle_state snb_cstates[MWAIT_MAX_NUM_CSTATES] = {
 		.driver_data = (void *) 0x30,
 		.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
 		.exit_latency = 109,
-		.target_residency = 300,
+		.target_residency = 345,
 		.enter = &intel_idle },
 };
 
-- 
1.7.4.rc1.14.g7fad99

[PATCH 02/12] cpuidle: Make cpuidle_enable_device() call poll_idle_init()

[Len Brown]

From: Rafael J. Wysocki <rjw@sisk.pl>

The following scenario is possible with the current cpuidle code and
the ACPI cpuidle driver:
(1) acpi_processor_cst_has_changed() is called,
(2) cpuidle_disable_device() is called,
(3) cpuidle_remove_state_sysfs() is called to remove the (presumably
    outdated) states info from sysfs,
(3) acpi_processor_get_power_info() is called, the first entry in the
    pr->power.states[] table is filled with zeros,
(4) acpi_processor_setup_cpuidle() is called and it doesn't fill the
    first entry in pr->power.states[],
(5) cpuidle_enable_device() is called,
(6) __cpuidle_register_device() is _not_ called, since the device has
    already been registered,
(7) Consequently, poll_idle_init() is _not_ called either,
(8) cpuidle_add_state_sysfs() is called to create the sysfs attributes
    for the new states and it uses the bogus first table entry from
    acpi_processor_get_power_info() for creating state0.

This problem is avoided if cpuidle_enable_device()
unconditionally calls poll_idle_init().

Reported-by: Len Brown <len.brown@intel.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Len Brown <len.brown@intel.com>
cc: stable@kernel.org
---
 drivers/cpuidle/cpuidle.c |   82 ++++++++++++++++++++++----------------------
 1 files changed, 41 insertions(+), 41 deletions(-)

diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index a507108..97df791 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -154,6 +154,45 @@ void cpuidle_resume_and_unlock(void)
 
 EXPORT_SYMBOL_GPL(cpuidle_resume_and_unlock);
 
+#ifdef CONFIG_ARCH_HAS_CPU_RELAX
+static int poll_idle(struct cpuidle_device *dev, struct cpuidle_state *st)
+{
+	ktime_t	t1, t2;
+	s64 diff;
+	int ret;
+
+	t1 = ktime_get();
+	local_irq_enable();
+	while (!need_resched())
+		cpu_relax();
+
+	t2 = ktime_get();
+	diff = ktime_to_us(ktime_sub(t2, t1));
+	if (diff > INT_MAX)
+		diff = INT_MAX;
+
+	ret = (int) diff;
+	return ret;
+}
+
+static void poll_idle_init(struct cpuidle_device *dev)
+{
+	struct cpuidle_state *state = &dev->states[0];
+
+	cpuidle_set_statedata(state, NULL);
+
+	snprintf(state->name, CPUIDLE_NAME_LEN, "C0");
+	snprintf(state->desc, CPUIDLE_DESC_LEN, "CPUIDLE CORE POLL IDLE");
+	state->exit_latency = 0;
+	state->target_residency = 0;
+	state->power_usage = -1;
+	state->flags = CPUIDLE_FLAG_POLL;
+	state->enter = poll_idle;
+}
+#else
+static void poll_idle_init(struct cpuidle_device *dev) {}
+#endif /* CONFIG_ARCH_HAS_CPU_RELAX */
+
 /**
  * cpuidle_enable_device - enables idle PM for a CPU
  * @dev: the CPU
@@ -178,6 +217,8 @@ int cpuidle_enable_device(struct cpuidle_device *dev)
 			return ret;
 	}
 
+	poll_idle_init(dev);
+
 	if ((ret = cpuidle_add_state_sysfs(dev)))
 		return ret;
 
@@ -232,45 +273,6 @@ void cpuidle_disable_device(struct cpuidle_device *dev)
 
 EXPORT_SYMBOL_GPL(cpuidle_disable_device);
 
-#ifdef CONFIG_ARCH_HAS_CPU_RELAX
-static int poll_idle(struct cpuidle_device *dev, struct cpuidle_state *st)
-{
-	ktime_t	t1, t2;
-	s64 diff;
-	int ret;
-
-	t1 = ktime_get();
-	local_irq_enable();
-	while (!need_resched())
-		cpu_relax();
-
-	t2 = ktime_get();
-	diff = ktime_to_us(ktime_sub(t2, t1));
-	if (diff > INT_MAX)
-		diff = INT_MAX;
-
-	ret = (int) diff;
-	return ret;
-}
-
-static void poll_idle_init(struct cpuidle_device *dev)
-{
-	struct cpuidle_state *state = &dev->states[0];
-
-	cpuidle_set_statedata(state, NULL);
-
-	snprintf(state->name, CPUIDLE_NAME_LEN, "C0");
-	snprintf(state->desc, CPUIDLE_DESC_LEN, "CPUIDLE CORE POLL IDLE");
-	state->exit_latency = 0;
-	state->target_residency = 0;
-	state->power_usage = -1;
-	state->flags = CPUIDLE_FLAG_POLL;
-	state->enter = poll_idle;
-}
-#else
-static void poll_idle_init(struct cpuidle_device *dev) {}
-#endif /* CONFIG_ARCH_HAS_CPU_RELAX */
-
 /**
  * __cpuidle_register_device - internal register function called before register
  * and enable routines
@@ -291,8 +293,6 @@ static int __cpuidle_register_device(struct cpuidle_device *dev)
 
 	init_completion(&dev->kobj_unregister);
 
-	poll_idle_init(dev);
-
 	/*
 	 * cpuidle driver should set the dev->power_specified bit
 	 * before registering the device if the driver provides
-- 
1.7.4.rc1.14.g7fad99

[PATCH 03/12] ACPI, intel_idle: Cleanup idle= internal variables

[Len Brown]

From: Thomas Renninger <trenn@suse.de>

Having four variables for the same thing:
  idle_halt, idle_nomwait, force_mwait and boot_option_idle_overrides
is rather confusing and unnecessary complex.

if idle= boot param is passed, only set up one variable:
boot_option_idle_overrides

Introduces following functional changes/fixes:
  - intel_idle driver does not register if any idle=xy
    boot param is passed.
  - processor_idle.c will also not register a cpuidle driver
    and get active if idle=halt is passed.
    Before a cpuidle driver with one (C1, halt) state got registered
    Now the default_idle function will be used which finally uses
    the same idle call to enter sleep state (safe_halt()), but
    without registering a whole cpuidle driver.

That means idle= param will always avoid cpuidle drivers to register
with one exception (same behavior as before):
idle=nomwait
may still register acpi_idle cpuidle driver, but C1 will not use
mwait, but hlt. This can be a workaround for IO based deeper sleep
states where C1 mwait causes problems.

Signed-off-by: Thomas Renninger <trenn@suse.de>
cc: x86@kernel.org
Signed-off-by: Len Brown <len.brown@intel.com>
---
 arch/ia64/include/asm/processor.h |    5 +++--
 arch/ia64/kernel/process.c        |    6 +-----
 arch/x86/include/asm/processor.h  |    5 +++--
 arch/x86/kernel/process.c         |   24 ++++++++----------------
 drivers/acpi/processor_core.c     |    4 ++--
 drivers/acpi/processor_idle.c     |   24 +++++++++++-------------
 drivers/idle/intel_idle.c         |    4 ++++
 7 files changed, 32 insertions(+), 40 deletions(-)

diff --git a/arch/ia64/include/asm/processor.h b/arch/ia64/include/asm/processor.h
index 348e44d..03afe79 100644
--- a/arch/ia64/include/asm/processor.h
+++ b/arch/ia64/include/asm/processor.h
@@ -717,8 +717,9 @@ prefetchw (const void *x)
 #define spin_lock_prefetch(x)	prefetchw(x)
 
 extern unsigned long boot_option_idle_override;
-extern unsigned long idle_halt;
-extern unsigned long idle_nomwait;
+
+enum idle_boot_override {IDLE_NO_OVERRIDE=0, IDLE_HALT, IDLE_FORCE_MWAIT,
+			 IDLE_NOMWAIT, IDLE_POLL};
 
 #endif /* !__ASSEMBLY__ */
 
diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index 16f1c7b..6d33c5c 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -53,12 +53,8 @@
 
 void (*ia64_mark_idle)(int);
 
-unsigned long boot_option_idle_override = 0;
+unsigned long boot_option_idle_override = IDLE_NO_OVERRIDE;
 EXPORT_SYMBOL(boot_option_idle_override);
-unsigned long idle_halt;
-EXPORT_SYMBOL(idle_halt);
-unsigned long idle_nomwait;
-EXPORT_SYMBOL(idle_nomwait);
 void (*pm_idle) (void);
 EXPORT_SYMBOL(pm_idle);
 void (*pm_power_off) (void);
diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index cae9c3c..b79bd98 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -762,10 +762,11 @@ extern void select_idle_routine(const struct cpuinfo_x86 *c);
 extern void init_c1e_mask(void);
 
 extern unsigned long		boot_option_idle_override;
-extern unsigned long		idle_halt;
-extern unsigned long		idle_nomwait;
 extern bool			c1e_detected;
 
+enum idle_boot_override {IDLE_NO_OVERRIDE=0, IDLE_HALT, IDLE_NOMWAIT,
+			 IDLE_POLL, IDLE_FORCE_MWAIT};
+
 extern void enable_sep_cpu(void);
 extern int sysenter_setup(void);
 
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 57d1868..b647215 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -22,11 +22,6 @@
 #include <asm/i387.h>
 #include <asm/debugreg.h>
 
-unsigned long idle_halt;
-EXPORT_SYMBOL(idle_halt);
-unsigned long idle_nomwait;
-EXPORT_SYMBOL(idle_nomwait);
-
 struct kmem_cache *task_xstate_cachep;
 EXPORT_SYMBOL_GPL(task_xstate_cachep);
 
@@ -328,7 +323,7 @@ long sys_execve(const char __user *name,
 /*
  * Idle related variables and functions
  */
-unsigned long boot_option_idle_override = 0;
+unsigned long boot_option_idle_override = IDLE_NO_OVERRIDE;
 EXPORT_SYMBOL(boot_option_idle_override);
 
 /*
@@ -499,7 +494,6 @@ static void poll_idle(void)
  *
  * idle=mwait overrides this decision and forces the usage of mwait.
  */
-static int __cpuinitdata force_mwait;
 
 #define MWAIT_INFO			0x05
 #define MWAIT_ECX_EXTENDED_INFO		0x01
@@ -509,7 +503,7 @@ static int __cpuinit mwait_usable(const struct cpuinfo_x86 *c)
 {
 	u32 eax, ebx, ecx, edx;
 
-	if (force_mwait)
+	if (boot_option_idle_override == IDLE_FORCE_MWAIT)
 		return 1;
 
 	if (c->cpuid_level < MWAIT_INFO)
@@ -629,9 +623,10 @@ static int __init idle_setup(char *str)
 	if (!strcmp(str, "poll")) {
 		printk("using polling idle threads.\n");
 		pm_idle = poll_idle;
-	} else if (!strcmp(str, "mwait"))
-		force_mwait = 1;
-	else if (!strcmp(str, "halt")) {
+		boot_option_idle_override = IDLE_POLL;
+	} else if (!strcmp(str, "mwait")) {
+		boot_option_idle_override = IDLE_FORCE_MWAIT;
+	} else if (!strcmp(str, "halt")) {
 		/*
 		 * When the boot option of idle=halt is added, halt is
 		 * forced to be used for CPU idle. In such case CPU C2/C3
@@ -640,8 +635,7 @@ static int __init idle_setup(char *str)
 		 * the boot_option_idle_override.
 		 */
 		pm_idle = default_idle;
-		idle_halt = 1;
-		return 0;
+		boot_option_idle_override = IDLE_HALT;
 	} else if (!strcmp(str, "nomwait")) {
 		/*
 		 * If the boot option of "idle=nomwait" is added,
@@ -649,12 +643,10 @@ static int __init idle_setup(char *str)
 		 * states. In such case it won't touch the variable
 		 * of boot_option_idle_override.
 		 */
-		idle_nomwait = 1;
-		return 0;
+		boot_option_idle_override = IDLE_NOMWAIT;
 	} else
 		return -1;
 
-	boot_option_idle_override = 1;
 	return 0;
 }
 early_param("idle", idle_setup);
diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index bec561c..3c1a2fe 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -23,7 +23,7 @@ static int set_no_mwait(const struct dmi_system_id *id)
 {
 	printk(KERN_NOTICE PREFIX "%s detected - "
 		"disabling mwait for CPU C-states\n", id->ident);
-	idle_nomwait = 1;
+	boot_option_idle_override = IDLE_NOMWAIT;
 	return 0;
 }
 
@@ -283,7 +283,7 @@ acpi_processor_eval_pdc(acpi_handle handle, struct acpi_object_list *pdc_in)
 {
 	acpi_status status = AE_OK;
 
-	if (idle_nomwait) {
+	if (boot_option_idle_override == IDLE_NOMWAIT) {
 		/*
 		 * If mwait is disabled for CPU C-states, the C2C3_FFH access
 		 * mode will be disabled in the parameter of _PDC object.
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index dcb38f8..eefd4aa 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -79,6 +79,13 @@ module_param(bm_check_disable, uint, 0000);
 static unsigned int latency_factor __read_mostly = 2;
 module_param(latency_factor, uint, 0644);
 
+static int disabled_by_idle_boot_param(void)
+{
+	return boot_option_idle_override == IDLE_POLL ||
+		boot_option_idle_override == IDLE_FORCE_MWAIT ||
+		boot_option_idle_override == IDLE_HALT;
+}
+
 /*
  * IBM ThinkPad R40e crashes mysteriously when going into C2 or C3.
  * For now disable this. Probably a bug somewhere else.
@@ -455,7 +462,7 @@ static int acpi_processor_get_power_info_cst(struct acpi_processor *pr)
 				continue;
 			}
 			if (cx.type == ACPI_STATE_C1 &&
-					(idle_halt || idle_nomwait)) {
+			    (boot_option_idle_override == IDLE_NOMWAIT)) {
 				/*
 				 * In most cases the C1 space_id obtained from
 				 * _CST object is FIXED_HARDWARE access mode.
@@ -1058,7 +1065,7 @@ int acpi_processor_cst_has_changed(struct acpi_processor *pr)
 {
 	int ret = 0;
 
-	if (boot_option_idle_override)
+	if (disabled_by_idle_boot_param())
 		return 0;
 
 	if (!pr)
@@ -1089,19 +1096,10 @@ int __cpuinit acpi_processor_power_init(struct acpi_processor *pr,
 	acpi_status status = 0;
 	static int first_run;
 
-	if (boot_option_idle_override)
+	if (disabled_by_idle_boot_param())
 		return 0;
 
 	if (!first_run) {
-		if (idle_halt) {
-			/*
-			 * When the boot option of "idle=halt" is added, halt
-			 * is used for CPU IDLE.
-			 * In such case C2/C3 is meaningless. So the max_cstate
-			 * is set to one.
-			 */
-			max_cstate = 1;
-		}
 		dmi_check_system(processor_power_dmi_table);
 		max_cstate = acpi_processor_cstate_check(max_cstate);
 		if (max_cstate < ACPI_C_STATES_MAX)
@@ -1142,7 +1140,7 @@ int __cpuinit acpi_processor_power_init(struct acpi_processor *pr,
 int acpi_processor_power_exit(struct acpi_processor *pr,
 			      struct acpi_device *device)
 {
-	if (boot_option_idle_override)
+	if (disabled_by_idle_boot_param())
 		return 0;
 
 	cpuidle_unregister_device(&pr->power.dev);
diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 94a6526..21d3871 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -404,6 +404,10 @@ static int __init intel_idle_init(void)
 {
 	int retval;
 
+	/* Do not load intel_idle at all for now if idle= is passed */
+	if (boot_option_idle_override != IDLE_NO_OVERRIDE)
+		return -ENODEV;
+
 	retval = intel_idle_probe();
 	if (retval)
 		return retval;
-- 
1.7.4.rc1.14.g7fad99

[PATCH 04/12] cpuidle: Rename X86 specific idle poll state[0] from C0 to POLL

[Len Brown]

From: Thomas Renninger <trenn@suse.de>

C0 means and is well know as "not idle".
All documentation out there uses this term as "running"/"not idle"
state. Also Linux userspace tools (e.g. cpufreq-aperf and turbostat)
show C0 residency which there is correct, but means something totally
else than cpuidle "POLL" state.

Signed-off-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/cpuidle/cpuidle.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index 97df791..37e4460 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -181,7 +181,7 @@ static void poll_idle_init(struct cpuidle_device *dev)
 
 	cpuidle_set_statedata(state, NULL);
 
-	snprintf(state->name, CPUIDLE_NAME_LEN, "C0");
+	snprintf(state->name, CPUIDLE_NAME_LEN, "POLL");
 	snprintf(state->desc, CPUIDLE_DESC_LEN, "CPUIDLE CORE POLL IDLE");
 	state->exit_latency = 0;
 	state->target_residency = 0;
-- 
1.7.4.rc1.14.g7fad99

[PATCH 05/12] ACPI: processor_idle: delete use of NOP CPUIDLE_FLAGs

[Len Brown]

From: Len Brown <len.brown@intel.com>

CPUIDLE_FLAG_SHALLOW
CPUIDLE_FLAG_BALANCED
CPUIDLE_FLAG_DEEP
CPUIDLE_FLAG_CHECK_BM

were set by acpi_processor_setup_cpuidle(),
but never used by cpuidle or by acpi_idle.
So stop setting them.

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/processor_idle.c |    4 ----
 1 files changed, 0 insertions(+), 4 deletions(-)

diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index eefd4aa..70599d2 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -1023,7 +1023,6 @@ static int acpi_processor_setup_cpuidle(struct acpi_processor *pr)
 		state->flags = 0;
 		switch (cx->type) {
 			case ACPI_STATE_C1:
-			state->flags |= CPUIDLE_FLAG_SHALLOW;
 			if (cx->entry_method == ACPI_CSTATE_FFH)
 				state->flags |= CPUIDLE_FLAG_TIME_VALID;
 
@@ -1032,16 +1031,13 @@ static int acpi_processor_setup_cpuidle(struct acpi_processor *pr)
 			break;
 
 			case ACPI_STATE_C2:
-			state->flags |= CPUIDLE_FLAG_BALANCED;
 			state->flags |= CPUIDLE_FLAG_TIME_VALID;
 			state->enter = acpi_idle_enter_simple;
 			dev->safe_state = state;
 			break;
 
 			case ACPI_STATE_C3:
-			state->flags |= CPUIDLE_FLAG_DEEP;
 			state->flags |= CPUIDLE_FLAG_TIME_VALID;
-			state->flags |= CPUIDLE_FLAG_CHECK_BM;
 			state->enter = pr->flags.bm_check ?
 					acpi_idle_enter_bm :
 					acpi_idle_enter_simple;
-- 
1.7.4.rc1.14.g7fad99

[PATCH 06/12] cpuidle: delete NOP CPUIDLE_FLAG_POLL

[Len Brown]

From: Len Brown <len.brown@intel.com>

it serves no purpose

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/cpuidle/cpuidle.c |    2 +-
 include/linux/cpuidle.h   |    1 -
 2 files changed, 1 insertions(+), 2 deletions(-)

diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index 37e4460..dd5f1ea 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -186,7 +186,7 @@ static void poll_idle_init(struct cpuidle_device *dev)
 	state->exit_latency = 0;
 	state->target_residency = 0;
 	state->power_usage = -1;
-	state->flags = CPUIDLE_FLAG_POLL;
+	state->flags = 0;
 	state->enter = poll_idle;
 }
 #else
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index 1be416b..dee3285 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -48,7 +48,6 @@ struct cpuidle_state {
 /* Idle State Flags */
 #define CPUIDLE_FLAG_TIME_VALID	(0x01) /* is residency time measurable? */
 #define CPUIDLE_FLAG_CHECK_BM	(0x02) /* BM activity will exit state */
-#define CPUIDLE_FLAG_POLL	(0x10) /* no latency, no savings */
 #define CPUIDLE_FLAG_SHALLOW	(0x20) /* low latency, minimal savings */
 #define CPUIDLE_FLAG_BALANCED	(0x40) /* medium latency, moderate savings */
 #define CPUIDLE_FLAG_DEEP	(0x80) /* high latency, large savings */
-- 
1.7.4.rc1.14.g7fad99

[PATCH 07/12] SH, cpuidle: delete use of NOP CPUIDLE_FLAGS_SHALLOW

[Len Brown]

From: Len Brown <len.brown@intel.com>

set but not checked.

Signed-off-by: Len Brown <len.brown@intel.com>
---
 arch/sh/kernel/cpu/shmobile/cpuidle.c |    1 -
 1 files changed, 0 insertions(+), 1 deletions(-)

diff --git a/arch/sh/kernel/cpu/shmobile/cpuidle.c b/arch/sh/kernel/cpu/shmobile/cpuidle.c
index 83972aa..c19e2a9 100644
--- a/arch/sh/kernel/cpu/shmobile/cpuidle.c
+++ b/arch/sh/kernel/cpu/shmobile/cpuidle.c
@@ -81,7 +81,6 @@ void sh_mobile_setup_cpuidle(void)
 	state->target_residency = 1 * 2;
 	state->power_usage = 3;
 	state->flags = 0;
-	state->flags |= CPUIDLE_FLAG_SHALLOW;
 	state->flags |= CPUIDLE_FLAG_TIME_VALID;
 	state->enter = cpuidle_sleep_enter;
 
-- 
1.7.4.rc1.14.g7fad99

[PATCH 08/12] cpuidle: delete unused CPUIDLE_FLAG_SHALLOW, BALANCED, DEEP definitions

[Len Brown]

From: Len Brown <len.brown@intel.com>

Signed-off-by: Len Brown <len.brown@intel.com>
---
 include/linux/cpuidle.h |    3 ---
 1 files changed, 0 insertions(+), 3 deletions(-)

diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index dee3285..c252953 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -48,9 +48,6 @@ struct cpuidle_state {
 /* Idle State Flags */
 #define CPUIDLE_FLAG_TIME_VALID	(0x01) /* is residency time measurable? */
 #define CPUIDLE_FLAG_CHECK_BM	(0x02) /* BM activity will exit state */
-#define CPUIDLE_FLAG_SHALLOW	(0x20) /* low latency, minimal savings */
-#define CPUIDLE_FLAG_BALANCED	(0x40) /* medium latency, moderate savings */
-#define CPUIDLE_FLAG_DEEP	(0x80) /* high latency, large savings */
 #define CPUIDLE_FLAG_IGNORE	(0x100) /* ignore during this idle period */
 #define CPUIDLE_FLAG_TLB_FLUSHED (0x200) /* tlb will be flushed */
 
-- 
1.7.4.rc1.14.g7fad99

[PATCH 09/12] cpuidle: CPUIDLE_FLAG_TLB_FLUSHED is specific to intel_idle

[Len Brown]

From: Len Brown <len.brown@intel.com>

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/intel_idle.c |    8 ++++++++
 include/linux/cpuidle.h   |    1 -
 2 files changed, 8 insertions(+), 1 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 21d3871..8256309 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -82,6 +82,14 @@ static int intel_idle(struct cpuidle_device *dev, struct cpuidle_state *state);
 static struct cpuidle_state *cpuidle_state_table;
 
 /*
+ * Set this flag for states where the HW flushes the TLB for us
+ * and so we don't need cross-calls to keep it consistent.
+ * If this flag is set, SW flushes the TLB, so even if the
+ * HW doesn't do the flushing, this flag is safe to use.
+ */
+#define CPUIDLE_FLAG_TLB_FLUSHED	0x10000
+
+/*
  * States are indexed by the cstate number,
  * which is also the index into the MWAIT hint array.
  * Thus C0 is a dummy.
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index c252953..6be722c 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -49,7 +49,6 @@ struct cpuidle_state {
 #define CPUIDLE_FLAG_TIME_VALID	(0x01) /* is residency time measurable? */
 #define CPUIDLE_FLAG_CHECK_BM	(0x02) /* BM activity will exit state */
 #define CPUIDLE_FLAG_IGNORE	(0x100) /* ignore during this idle period */
-#define CPUIDLE_FLAG_TLB_FLUSHED (0x200) /* tlb will be flushed */
 
 #define CPUIDLE_DRIVER_FLAGS_MASK (0xFFFF0000)
 
-- 
1.7.4.rc1.14.g7fad99

[PATCH 10/12] cpuidle: CPUIDLE_FLAG_CHECK_BM is omap3_idle specific

[Len Brown]

From: Len Brown <len.brown@intel.com>

Signed-off-by: Len Brown <len.brown@intel.com>
---
 arch/arm/mach-omap2/cpuidle34xx.c |    2 ++
 include/linux/cpuidle.h           |    1 -
 2 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/arch/arm/mach-omap2/cpuidle34xx.c b/arch/arm/mach-omap2/cpuidle34xx.c
index 0d50b45..f9be4e3 100644
--- a/arch/arm/mach-omap2/cpuidle34xx.c
+++ b/arch/arm/mach-omap2/cpuidle34xx.c
@@ -47,6 +47,8 @@
 
 #define OMAP3_STATE_MAX OMAP3_STATE_C7
 
+#define CPUIDLE_FLAG_CHECK_BM	0x10000	/* use omap3_enter_idle_bm() */
+
 struct omap3_processor_cx {
 	u8 valid;
 	u8 type;
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index 6be722c..36719ea 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -47,7 +47,6 @@ struct cpuidle_state {
 
 /* Idle State Flags */
 #define CPUIDLE_FLAG_TIME_VALID	(0x01) /* is residency time measurable? */
-#define CPUIDLE_FLAG_CHECK_BM	(0x02) /* BM activity will exit state */
 #define CPUIDLE_FLAG_IGNORE	(0x100) /* ignore during this idle period */
 
 #define CPUIDLE_DRIVER_FLAGS_MASK (0xFFFF0000)
-- 
1.7.4.rc1.14.g7fad99

[PATCH 11/12] intel_idle: open broadcast clock event

[Len Brown]

From: Shaohua Li <shaohua.li@intel.com>

Intel_idle driver uses CLOCK_EVT_NOTIFY_BROADCAST_ENTER
CLOCK_EVT_NOTIFY_BROADCAST_EXIT
for broadcast clock events. The _ENTER/_EXIT doesn't really open broadcast clock
events, please see processor_idle.c for an example. In some situation, this will
cause boot hang, because some CPUs enters idle but local APIC timer stalls.

Reported-and-tested-by: Yan Zheng <zheng.z.yan@intel.com>
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
cc: stable@kernel.org
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/intel_idle.c |   47 ++++++++++++++++++++++++++++++++++++++++++++-
 1 files changed, 46 insertions(+), 1 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 8256309..fc39358 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -59,6 +59,8 @@
 #include <linux/hrtimer.h>	/* ktime_get_real() */
 #include <trace/events/power.h>
 #include <linux/sched.h>
+#include <linux/notifier.h>
+#include <linux/cpu.h>
 #include <asm/mwait.h>
 
 #define INTEL_IDLE_VERSION "0.4"
@@ -73,6 +75,7 @@ static int max_cstate = MWAIT_MAX_NUM_CSTATES - 1;
 
 static unsigned int mwait_substates;
 
+#define LAPIC_TIMER_ALWAYS_RELIABLE 0xFFFFFFFF
 /* Reliable LAPIC Timer States, bit 1 for C1 etc.  */
 static unsigned int lapic_timer_reliable_states = (1 << 1);	 /* Default to only C1 */
 
@@ -252,6 +255,39 @@ static int intel_idle(struct cpuidle_device *dev, struct cpuidle_state *state)
 	return usec_delta;
 }
 
+static void __setup_broadcast_timer(void *arg)
+{
+	unsigned long reason = (unsigned long)arg;
+	int cpu = smp_processor_id();
+
+	reason = reason ?
+		CLOCK_EVT_NOTIFY_BROADCAST_ON : CLOCK_EVT_NOTIFY_BROADCAST_OFF;
+
+	clockevents_notify(reason, &cpu);
+}
+
+static int __cpuinit 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, __setup_broadcast_timer,
+			(void *)true, 1);
+		break;
+	case CPU_DOWN_PREPARE:
+		smp_call_function_single(hotcpu, __setup_broadcast_timer,
+			(void *)false, 1);
+		break;
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block __cpuinitdata setup_broadcast_notifier = {
+	.notifier_call = setup_broadcast_cpuhp_notify,
+};
+
 /*
  * intel_idle_probe()
  */
@@ -314,7 +350,11 @@ static int intel_idle_probe(void)
 	}
 
 	if (boot_cpu_has(X86_FEATURE_ARAT))	/* Always Reliable APIC Timer */
-		lapic_timer_reliable_states = 0xFFFFFFFF;
+		lapic_timer_reliable_states = LAPIC_TIMER_ALWAYS_RELIABLE;
+	else {
+		smp_call_function(__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);
@@ -441,6 +481,11 @@ 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) {
+		smp_call_function(__setup_broadcast_timer, (void *)false, 1);
+		unregister_cpu_notifier(&setup_broadcast_notifier);
+	}
+
 	return;
 }
 
-- 
1.7.4.rc1.14.g7fad99

[PATCH 12/12] cpuidle/x86/perf: fix power:cpu_idle double end events and throw cpu_idle events from the cpuidle layer

[Len Brown]

From: Thomas Renninger <trenn@suse.de>

Currently intel_idle and acpi_idle driver show double cpu_idle "exit idle"
events -> this patch fixes it and makes cpu_idle events throwing less complex.

It also introduces cpu_idle events for all architectures which use
the cpuidle subsystem, namely:
  - arch/arm/mach-at91/cpuidle.c
  - arch/arm/mach-davinci/cpuidle.c
  - arch/arm/mach-kirkwood/cpuidle.c
  - arch/arm/mach-omap2/cpuidle34xx.c
  - arch/drivers/acpi/processor_idle.c (for all cases, not only mwait)
  - arch/x86/kernel/process.c (did throw events before, but was a mess)
  - drivers/idle/intel_idle.c (did throw events before)

Convention should be:
Fire cpu_idle events inside the current pm_idle function (not somewhere
down the the callee tree) to keep things easy.

Current possible pm_idle functions in X86:
c1e_idle, poll_idle, cpuidle_idle_call, mwait_idle, default_idle
-> this is really easy is now.

This affects userspace:
The type field of the cpu_idle power event can now direclty get
mapped to:
/sys/devices/system/cpu/cpuX/cpuidle/stateX/{name,desc,usage,time,...}
instead of throwing very CPU/mwait specific values.
This change is not visible for the intel_idle driver.
For the acpi_idle driver it should only be visible if the vendor
misses out C-states in his BIOS.
Another (perf timechart) patch reads out cpuidle info of cpu_idle
events from:
/sys/.../cpuidle/stateX/*, then the cpuidle events are mapped
to the correct C-/cpuidle state again, even if e.g. vendors miss
out C-states in their BIOS and for example only export C1 and C3.
-> everything is fine.

Signed-off-by: Thomas Renninger <trenn@suse.de>
CC: Robert Schoene <robert.schoene@tu-dresden.de>
CC: Jean Pihet <j-pihet@ti.com>
CC: Arjan van de Ven <arjan@linux.intel.com>
CC: Ingo Molnar <mingo@elte.hu>
CC: Frederic Weisbecker <fweisbec@gmail.com>
CC: linux-pm@lists.linux-foundation.org
CC: linux-acpi@vger.kernel.org
CC: linux-kernel@vger.kernel.org
CC: linux-perf-users@vger.kernel.org
CC: linux-omap@vger.kernel.org
Signed-off-by: Len Brown <len.brown@intel.com>
---
 arch/x86/kernel/process.c    |    6 ++++--
 arch/x86/kernel/process_32.c |    4 ----
 arch/x86/kernel/process_64.c |    6 ------
 drivers/cpuidle/cpuidle.c    |   10 ++++++++--
 drivers/idle/intel_idle.c    |    2 --
 5 files changed, 12 insertions(+), 16 deletions(-)

diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 09c08a1..67e96e6 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -386,6 +386,8 @@ void default_idle(void)
 		else
 			local_irq_enable();
 		current_thread_info()->status |= TS_POLLING;
+		trace_power_end(smp_processor_id());
+		trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id());
 	} else {
 		local_irq_enable();
 		/* loop is done by the caller */
@@ -443,8 +445,6 @@ EXPORT_SYMBOL_GPL(cpu_idle_wait);
  */
 void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
 {
-	trace_power_start(POWER_CSTATE, (ax>>4)+1, smp_processor_id());
-	trace_cpu_idle((ax>>4)+1, smp_processor_id());
 	if (!need_resched()) {
 		if (cpu_has(__this_cpu_ptr(&cpu_info), X86_FEATURE_CLFLUSH_MONITOR))
 			clflush((void *)&current_thread_info()->flags);
@@ -471,6 +471,8 @@ static void mwait_idle(void)
 			__sti_mwait(0, 0);
 		else
 			local_irq_enable();
+		trace_power_end(smp_processor_id());
+		trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id());
 	} else
 		local_irq_enable();
 }
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 4b9befa..8d12878 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -57,8 +57,6 @@
 #include <asm/syscalls.h>
 #include <asm/debugreg.h>
 
-#include <trace/events/power.h>
-
 asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
 
 /*
@@ -113,8 +111,6 @@ void cpu_idle(void)
 			stop_critical_timings();
 			pm_idle();
 			start_critical_timings();
-			trace_power_end(smp_processor_id());
-			trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id());
 		}
 		tick_nohz_restart_sched_tick();
 		preempt_enable_no_resched();
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 4c818a7..bd387e8 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -51,8 +51,6 @@
 #include <asm/syscalls.h>
 #include <asm/debugreg.h>
 
-#include <trace/events/power.h>
-
 asmlinkage extern void ret_from_fork(void);
 
 DEFINE_PER_CPU(unsigned long, old_rsp);
@@ -141,10 +139,6 @@ void cpu_idle(void)
 			pm_idle();
 			start_critical_timings();
 
-			trace_power_end(smp_processor_id());
-			trace_cpu_idle(PWR_EVENT_EXIT,
-				       smp_processor_id());
-
 			/* In many cases the interrupt that ended idle
 			   has already called exit_idle. But some idle
 			   loops can be woken up without interrupt. */
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index 386888f..e4855c3 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -96,7 +96,15 @@ static void cpuidle_idle_call(void)
 
 	/* enter the state and update stats */
 	dev->last_state = target_state;
+
+	trace_power_start(POWER_CSTATE, next_state, dev->cpu);
+	trace_cpu_idle(next_state, dev->cpu);
+
 	dev->last_residency = target_state->enter(dev, target_state);
+
+	trace_power_end(dev->cpu);
+	trace_cpu_idle(PWR_EVENT_EXIT, dev->cpu);
+
 	if (dev->last_state)
 		target_state = dev->last_state;
 
@@ -106,8 +114,6 @@ static void cpuidle_idle_call(void)
 	/* give the governor an opportunity to reflect on the outcome */
 	if (cpuidle_curr_governor->reflect)
 		cpuidle_curr_governor->reflect(dev);
-	trace_power_end(smp_processor_id());
-	trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id());
 }
 
 /**
diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 56ac09d..60fa6ec 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -220,8 +220,6 @@ static int intel_idle(struct cpuidle_device *dev, struct cpuidle_state *state)
 	kt_before = ktime_get_real();
 
 	stop_critical_timings();
-	trace_power_start(POWER_CSTATE, (eax >> 4) + 1, cpu);
-	trace_cpu_idle((eax >> 4) + 1, cpu);
 	if (!need_resched()) {
 
 		__monitor((void *)&current_thread_info()->flags, 0, 0);
-- 
1.7.4.rc1.14.g7fad99