Merge branch 'idle-release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-idle-2.6

* 'idle-release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-idle-2.6:
  intel_idle: native hardware cpuidle driver for latest Intel processors
  ACPI: acpi_idle: touch TS_POLLING only in the non-MWAIT case
  acpi_pad: uses MONITOR/MWAIT, so it doesn't need to clear TS_POLLING
  sched: clarify commment for TS_POLLING
  ACPI: allow a native cpuidle driver to displace ACPI
  cpuidle: make cpuidle_curr_driver static
  cpuidle: add cpuidle_unregister_driver() error check
  cpuidle: fail to register if !CONFIG_CPU_IDLE

14 files changed, 538 insertions, 42 deletions

diff --git a/MAINTAINERS b/MAINTAINERS
index 33047a60543..13608bd2e79 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2887,6 +2887,13 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input.git
 S:	Maintained
 F:	drivers/input/
 
+INTEL IDLE DRIVER
+M:	Len Brown <lenb@kernel.org>
+L:	linux-pm@lists.linux-foundation.org
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-idle-2.6.git
+S:	Supported
+F:	drivers/idle/intel_idle.c
+
 INTEL FRAMEBUFFER DRIVER (excluding 810 and 815)
 M:	Maik Broemme <mbroemme@plusserver.de>
 L:	linux-fbdev@vger.kernel.org
diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h
index 62ba9400cc4..f0b6e5dbc5a 100644
--- a/arch/x86/include/asm/thread_info.h
+++ b/arch/x86/include/asm/thread_info.h
@@ -239,8 +239,8 @@ static inline struct thread_info *current_thread_info(void)
 #define TS_USEDFPU		0x0001	/* FPU was used by this task
 					   this quantum (SMP) */
 #define TS_COMPAT		0x0002	/* 32bit syscall active (64BIT)*/
-#define TS_POLLING		0x0004	/* true if in idle loop
-					   and not sleeping */
+#define TS_POLLING		0x0004	/* idle task polling need_resched,
+					   skip sending interrupt */
 #define TS_RESTORE_SIGMASK	0x0008	/* restore signal mask in do_signal() */
 
 #define tsk_is_polling(t) (task_thread_info(t)->status & TS_POLLING)
diff --git a/drivers/Makefile b/drivers/Makefile
index f42a03029b7..91874e04855 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -10,6 +10,7 @@ obj-$(CONFIG_PCI)		+= pci/
 obj-$(CONFIG_PARISC)		+= parisc/
 obj-$(CONFIG_RAPIDIO)		+= rapidio/
 obj-y				+= video/
+obj-y				+= idle/
 obj-$(CONFIG_ACPI)		+= acpi/
 obj-$(CONFIG_SFI)		+= sfi/
 # PnP must come after ACPI since it will eventually need to check if acpi
@@ -91,7 +92,6 @@ obj-$(CONFIG_EISA)		+= eisa/
 obj-y				+= lguest/
 obj-$(CONFIG_CPU_FREQ)		+= cpufreq/
 obj-$(CONFIG_CPU_IDLE)		+= cpuidle/
-obj-y				+= idle/
 obj-$(CONFIG_MMC)		+= mmc/
 obj-$(CONFIG_MEMSTICK)		+= memstick/
 obj-$(CONFIG_NEW_LEDS)		+= leds/
diff --git a/drivers/acpi/acpi_pad.c b/drivers/acpi/acpi_pad.c
index f169e516a1a..d269a8f3329 100644
--- a/drivers/acpi/acpi_pad.c
+++ b/drivers/acpi/acpi_pad.c
@@ -172,13 +172,6 @@ static int power_saving_thread(void *data)
 
 		do_sleep = 0;
 
-		current_thread_info()->status &= ~TS_POLLING;
-		/*
-		 * TS_POLLING-cleared state must be visible before we test
-		 * NEED_RESCHED:
-		 */
-		smp_mb();
-
 		expire_time = jiffies + HZ * (100 - idle_pct) / 100;
 
 		while (!need_resched()) {
@@ -209,8 +202,6 @@ static int power_saving_thread(void *data)
 			}
 		}
 
-		current_thread_info()->status |= TS_POLLING;
-
 		/*
 		 * current sched_rt has threshold for rt task running time.
 		 * When a rt task uses 95% CPU time, the rt thread will be
diff --git a/drivers/acpi/processor_driver.c b/drivers/acpi/processor_driver.c
index 5675d9747e8..b1034a9ada4 100644
--- a/drivers/acpi/processor_driver.c
+++ b/drivers/acpi/processor_driver.c
@@ -616,7 +616,8 @@ static int __cpuinit acpi_processor_add(struct acpi_device *device)
 	acpi_processor_get_limit_info(pr);
 
 
-	acpi_processor_power_init(pr, device);
+	if (cpuidle_get_driver() == &acpi_idle_driver)
+		acpi_processor_power_init(pr, device);
 
 	pr->cdev = thermal_cooling_device_register("Processor", device,
 						&processor_cooling_ops);
@@ -920,9 +921,14 @@ static int __init acpi_processor_init(void)
 	if (!acpi_processor_dir)
 		return -ENOMEM;
 #endif
-	result = cpuidle_register_driver(&acpi_idle_driver);
-	if (result < 0)
-		goto out_proc;
+
+	if (!cpuidle_register_driver(&acpi_idle_driver)) {
+		printk(KERN_DEBUG "ACPI: %s registered with cpuidle\n",
+			acpi_idle_driver.name);
+	} else {
+		printk(KERN_DEBUG "ACPI: acpi_idle yielding to %s",
+			cpuidle_get_driver()->name);
+	}
 
 	result = acpi_bus_register_driver(&acpi_processor_driver);
 	if (result < 0)
@@ -941,7 +947,6 @@ static int __init acpi_processor_init(void)
 out_cpuidle:
 	cpuidle_unregister_driver(&acpi_idle_driver);
 
-out_proc:
 #ifdef CONFIG_ACPI_PROCFS
 	remove_proc_entry(ACPI_PROCESSOR_CLASS, acpi_root_dir);
 #endif
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 13c6cb703f1..2e8c27d48f2 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -872,6 +872,7 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev,
 		return(acpi_idle_enter_c1(dev, state));
 
 	local_irq_disable();
+
 	if (cx->entry_method != ACPI_CSTATE_FFH) {
 		current_thread_info()->status &= ~TS_POLLING;
 		/*
@@ -879,12 +880,12 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev,
 		 * NEED_RESCHED:
 		 */
 		smp_mb();
-	}
 
-	if (unlikely(need_resched())) {
-		current_thread_info()->status |= TS_POLLING;
-		local_irq_enable();
-		return 0;
+		if (unlikely(need_resched())) {
+			current_thread_info()->status |= TS_POLLING;
+			local_irq_enable();
+			return 0;
+		}
 	}
 
 	/*
@@ -911,7 +912,8 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev,
 	sched_clock_idle_wakeup_event(idle_time_ns);
 
 	local_irq_enable();
-	current_thread_info()->status |= TS_POLLING;
+	if (cx->entry_method != ACPI_CSTATE_FFH)
+		current_thread_info()->status |= TS_POLLING;
 
 	cx->usage++;
 
@@ -962,6 +964,7 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev,
 	}
 
 	local_irq_disable();
+
 	if (cx->entry_method != ACPI_CSTATE_FFH) {
 		current_thread_info()->status &= ~TS_POLLING;
 		/*
@@ -969,12 +972,12 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev,
 		 * NEED_RESCHED:
 		 */
 		smp_mb();
-	}
 
-	if (unlikely(need_resched())) {
-		current_thread_info()->status |= TS_POLLING;
-		local_irq_enable();
-		return 0;
+		if (unlikely(need_resched())) {
+			current_thread_info()->status |= TS_POLLING;
+			local_irq_enable();
+			return 0;
+		}
 	}
 
 	acpi_unlazy_tlb(smp_processor_id());
@@ -1028,7 +1031,8 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev,
 	sched_clock_idle_wakeup_event(idle_time_ns);
 
 	local_irq_enable();
-	current_thread_info()->status |= TS_POLLING;
+	if (cx->entry_method != ACPI_CSTATE_FFH)
+		current_thread_info()->status |= TS_POLLING;
 
 	cx->usage++;
 
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index 12fdd3987a3..199488576a0 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -156,7 +156,7 @@ int cpuidle_enable_device(struct cpuidle_device *dev)
 
 	if (dev->enabled)
 		return 0;
-	if (!cpuidle_curr_driver || !cpuidle_curr_governor)
+	if (!cpuidle_get_driver() || !cpuidle_curr_governor)
 		return -EIO;
 	if (!dev->state_count)
 		return -EINVAL;
@@ -207,7 +207,7 @@ void cpuidle_disable_device(struct cpuidle_device *dev)
 {
 	if (!dev->enabled)
 		return;
-	if (!cpuidle_curr_driver || !cpuidle_curr_governor)
+	if (!cpuidle_get_driver() || !cpuidle_curr_governor)
 		return;
 
 	dev->enabled = 0;
@@ -271,10 +271,11 @@ static int __cpuidle_register_device(struct cpuidle_device *dev)
 {
 	int ret;
 	struct sys_device *sys_dev = get_cpu_sysdev((unsigned long)dev->cpu);
+	struct cpuidle_driver *cpuidle_driver = cpuidle_get_driver();
 
 	if (!sys_dev)
 		return -EINVAL;
-	if (!try_module_get(cpuidle_curr_driver->owner))
+	if (!try_module_get(cpuidle_driver->owner))
 		return -EINVAL;
 
 	init_completion(&dev->kobj_unregister);
@@ -284,7 +285,7 @@ static int __cpuidle_register_device(struct cpuidle_device *dev)
 	per_cpu(cpuidle_devices, dev->cpu) = dev;
 	list_add(&dev->device_list, &cpuidle_detected_devices);
 	if ((ret = cpuidle_add_sysfs(sys_dev))) {
-		module_put(cpuidle_curr_driver->owner);
+		module_put(cpuidle_driver->owner);
 		return ret;
 	}
 
@@ -325,6 +326,7 @@ EXPORT_SYMBOL_GPL(cpuidle_register_device);
 void cpuidle_unregister_device(struct cpuidle_device *dev)
 {
 	struct sys_device *sys_dev = get_cpu_sysdev((unsigned long)dev->cpu);
+	struct cpuidle_driver *cpuidle_driver = cpuidle_get_driver();
 
 	if (dev->registered == 0)
 		return;
@@ -340,7 +342,7 @@ void cpuidle_unregister_device(struct cpuidle_device *dev)
 
 	cpuidle_resume_and_unlock();
 
-	module_put(cpuidle_curr_driver->owner);
+	module_put(cpuidle_driver->owner);
 }
 
 EXPORT_SYMBOL_GPL(cpuidle_unregister_device);
diff --git a/drivers/cpuidle/cpuidle.h b/drivers/cpuidle/cpuidle.h
index 9476ba33ee2..33e50d556f1 100644
--- a/drivers/cpuidle/cpuidle.h
+++ b/drivers/cpuidle/cpuidle.h
@@ -9,7 +9,6 @@
 
 /* For internal use only */
 extern struct cpuidle_governor *cpuidle_curr_governor;
-extern struct cpuidle_driver *cpuidle_curr_driver;
 extern struct list_head cpuidle_governors;
 extern struct list_head cpuidle_detected_devices;
 extern struct mutex cpuidle_lock;
diff --git a/drivers/cpuidle/driver.c b/drivers/cpuidle/driver.c
index 2257004fe33..fd1601e3d12 100644
--- a/drivers/cpuidle/driver.c
+++ b/drivers/cpuidle/driver.c
@@ -14,7 +14,7 @@
 
 #include "cpuidle.h"
 
-struct cpuidle_driver *cpuidle_curr_driver;
+static struct cpuidle_driver *cpuidle_curr_driver;
 DEFINE_SPINLOCK(cpuidle_driver_lock);
 
 /**
@@ -40,13 +40,25 @@ int cpuidle_register_driver(struct cpuidle_driver *drv)
 EXPORT_SYMBOL_GPL(cpuidle_register_driver);
 
 /**
+ * cpuidle_get_driver - return the current driver
+ */
+struct cpuidle_driver *cpuidle_get_driver(void)
+{
+	return cpuidle_curr_driver;
+}
+EXPORT_SYMBOL_GPL(cpuidle_get_driver);
+
+/**
  * cpuidle_unregister_driver - unregisters a driver
  * @drv: the driver
  */
 void cpuidle_unregister_driver(struct cpuidle_driver *drv)
 {
-	if (!drv)
+	if (drv != cpuidle_curr_driver) {
+		WARN(1, "invalid cpuidle_unregister_driver(%s)\n",
+			drv->name);
 		return;
+	}
 
 	spin_lock(&cpuidle_driver_lock);
 	cpuidle_curr_driver = NULL;
diff --git a/drivers/cpuidle/sysfs.c b/drivers/cpuidle/sysfs.c
index 0ba9c8b8ee7..0310ffaec9d 100644
--- a/drivers/cpuidle/sysfs.c
+++ b/drivers/cpuidle/sysfs.c
@@ -47,10 +47,11 @@ static ssize_t show_current_driver(struct sysdev_class *class,
 				   char *buf)
 {
 	ssize_t ret;
+	struct cpuidle_driver *cpuidle_driver = cpuidle_get_driver();
 
 	spin_lock(&cpuidle_driver_lock);
-	if (cpuidle_curr_driver)
-		ret = sprintf(buf, "%s\n", cpuidle_curr_driver->name);
+	if (cpuidle_driver)
+		ret = sprintf(buf, "%s\n", cpuidle_driver->name);
 	else
 		ret = sprintf(buf, "none\n");
 	spin_unlock(&cpuidle_driver_lock);
diff --git a/drivers/idle/Kconfig b/drivers/idle/Kconfig
index f15e90a453d..fb5c5186d4a 100644
--- a/drivers/idle/Kconfig
+++ b/drivers/idle/Kconfig
@@ -1,3 +1,14 @@
+config INTEL_IDLE
+	tristate "Cpuidle Driver for Intel Processors"
+	depends on CPU_IDLE
+	depends on X86
+	depends on CPU_SUP_INTEL
+	depends on EXPERIMENTAL
+	help
+	  Enable intel_idle, a cpuidle driver that includes knowledge of
+	  native Intel hardware idle features.  The acpi_idle driver
+	  can be configured at the same time, in order to handle
+	  processors intel_idle does not support.
 
 menu "Memory power savings"
 depends on X86_64
diff --git a/drivers/idle/Makefile b/drivers/idle/Makefile
index 5f68fc377e2..23d295cf10f 100644
--- a/drivers/idle/Makefile
+++ b/drivers/idle/Makefile
@@ -1,2 +1,3 @@
 obj-$(CONFIG_I7300_IDLE)			+= i7300_idle.o
+obj-$(CONFIG_INTEL_IDLE)			+= intel_idle.o
 
diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
new file mode 100755
index 00000000000..54f0fb4cd5d
--- /dev/null
+++ b/drivers/idle/intel_idle.c
@@ -0,0 +1,461 @@
+/*
+ * intel_idle.c - native hardware idle loop for modern Intel processors
+ *
+ * Copyright (c) 2010, Intel Corporation.
+ * Len Brown <len.brown@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+/*
+ * intel_idle is a cpuidle driver that loads on specific Intel processors
+ * in lieu of the legacy ACPI processor_idle driver.  The intent is to
+ * make Linux more efficient on these processors, as intel_idle knows
+ * more than ACPI, as well as make Linux more immune to ACPI BIOS bugs.
+ */
+
+/*
+ * Design Assumptions
+ *
+ * All CPUs have same idle states as boot CPU
+ *
+ * Chipset BM_STS (bus master status) bit is a NOP
+ *	for preventing entry into deep C-stats
+ */
+
+/*
+ * Known limitations
+ *
+ * The driver currently initializes for_each_online_cpu() upon modprobe.
+ * It it unaware of subsequent processors hot-added to the system.
+ * This means that if you boot with maxcpus=n and later online
+ * processors above n, those processors will use C1 only.
+ *
+ * ACPI has a .suspend hack to turn off deep c-statees during suspend
+ * to avoid complications with the lapic timer workaround.
+ * Have not seen issues with suspend, but may need same workaround here.
+ *
+ * There is currently no kernel-based automatic probing/loading mechanism
+ * if the driver is built as a module.
+ */
+
+/* un-comment DEBUG to enable pr_debug() statements */
+#define DEBUG
+
+#include <linux/kernel.h>
+#include <linux/cpuidle.h>
+#include <linux/clockchips.h>
+#include <linux/hrtimer.h>	/* ktime_get_real() */
+#include <trace/events/power.h>
+#include <linux/sched.h>
+
+#define INTEL_IDLE_VERSION "0.4"
+#define PREFIX "intel_idle: "
+
+#define MWAIT_SUBSTATE_MASK	(0xf)
+#define MWAIT_CSTATE_MASK	(0xf)
+#define MWAIT_SUBSTATE_SIZE	(4)
+#define MWAIT_MAX_NUM_CSTATES	8
+#define CPUID_MWAIT_LEAF (5)
+#define CPUID5_ECX_EXTENSIONS_SUPPORTED (0x1)
+#define CPUID5_ECX_INTERRUPT_BREAK	(0x2)
+
+static struct cpuidle_driver intel_idle_driver = {
+	.name = "intel_idle",
+	.owner = THIS_MODULE,
+};
+/* intel_idle.max_cstate=0 disables driver */
+static int max_cstate = MWAIT_MAX_NUM_CSTATES - 1;
+static int power_policy = 7; /* 0 = max perf; 15 = max powersave */
+
+static unsigned int substates;
+static int (*choose_substate)(int);
+
+/* Reliable LAPIC Timer States, bit 1 for C1 etc.  */
+static unsigned int lapic_timer_reliable_states;
+
+static struct cpuidle_device *intel_idle_cpuidle_devices;
+static int intel_idle(struct cpuidle_device *dev, struct cpuidle_state *state);
+
+static struct cpuidle_state *cpuidle_state_table;
+
+/*
+ * States are indexed by the cstate number,
+ * which is also the index into the MWAIT hint array.
+ * Thus C0 is a dummy.
+ */
+static struct cpuidle_state nehalem_cstates[MWAIT_MAX_NUM_CSTATES] = {
+	{ /* MWAIT C0 */ },
+	{ /* MWAIT C1 */
+		.name = "NHM-C1",
+		.desc = "MWAIT 0x00",
+		.driver_data = (void *) 0x00,
+		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.exit_latency = 3,
+		.power_usage = 1000,
+		.target_residency = 6,
+		.enter = &intel_idle },
+	{ /* MWAIT C2 */
+		.name = "NHM-C3",
+		.desc = "MWAIT 0x10",
+		.driver_data = (void *) 0x10,
+		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.exit_latency = 20,
+		.power_usage = 500,
+		.target_residency = 80,
+		.enter = &intel_idle },
+	{ /* MWAIT C3 */
+		.name = "NHM-C6",
+		.desc = "MWAIT 0x20",
+		.driver_data = (void *) 0x20,
+		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.exit_latency = 200,
+		.power_usage = 350,
+		.target_residency = 800,
+		.enter = &intel_idle },
+};
+
+static struct cpuidle_state atom_cstates[MWAIT_MAX_NUM_CSTATES] = {
+	{ /* MWAIT C0 */ },
+	{ /* MWAIT C1 */
+		.name = "ATM-C1",
+		.desc = "MWAIT 0x00",
+		.driver_data = (void *) 0x00,
+		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.exit_latency = 1,
+		.power_usage = 1000,
+		.target_residency = 4,
+		.enter = &intel_idle },
+	{ /* MWAIT C2 */
+		.name = "ATM-C2",
+		.desc = "MWAIT 0x10",
+		.driver_data = (void *) 0x10,
+		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.exit_latency = 20,
+		.power_usage = 500,
+		.target_residency = 80,
+		.enter = &intel_idle },
+	{ /* MWAIT C3 */ },
+	{ /* MWAIT C4 */
+		.name = "ATM-C4",
+		.desc = "MWAIT 0x30",
+		.driver_data = (void *) 0x30,
+		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.exit_latency = 100,
+		.power_usage = 250,
+		.target_residency = 400,
+		.enter = &intel_idle },
+	{ /* MWAIT C5 */ },
+	{ /* MWAIT C6 */
+		.name = "ATM-C6",
+		.desc = "MWAIT 0x40",
+		.driver_data = (void *) 0x40,
+		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.exit_latency = 200,
+		.power_usage = 150,
+		.target_residency = 800,
+		.enter = NULL },	/* disabled */
+};
+
+/*
+ * choose_tunable_substate()
+ *
+ * Run-time decision on which C-state substate to invoke
+ * If power_policy = 0, choose shallowest substate (0)
+ * If power_policy = 15, choose deepest substate
+ * If power_policy = middle, choose middle substate etc.
+ */
+static int choose_tunable_substate(int cstate)
+{
+	unsigned int num_substates;
+	unsigned int substate_choice;
+
+	power_policy &= 0xF;	/* valid range: 0-15 */
+	cstate &= 7;	/* valid range: 0-7 */
+
+	num_substates = (substates >> ((cstate) * 4)) & MWAIT_SUBSTATE_MASK;
+
+	if (num_substates <= 1)
+		return 0;
+
+	substate_choice = ((power_policy + (power_policy + 1) *
+				(num_substates - 1)) / 16);
+
+	return substate_choice;
+}
+
+/*
+ * choose_zero_substate()
+ */
+static int choose_zero_substate(int cstate)
+{
+	return 0;
+}
+
+/**
+ * intel_idle
+ * @dev: cpuidle_device
+ * @state: cpuidle state
+ *
+ */
+static int intel_idle(struct cpuidle_device *dev, struct cpuidle_state *state)
+{
+	unsigned long ecx = 1; /* break on interrupt flag */
+	unsigned long eax = (unsigned long)cpuidle_get_statedata(state);
+	unsigned int cstate;
+	ktime_t kt_before, kt_after;
+	s64 usec_delta;
+	int cpu = smp_processor_id();
+
+	cstate = (((eax) >> MWAIT_SUBSTATE_SIZE) & MWAIT_CSTATE_MASK) + 1;
+
+	eax = eax + (choose_substate)(cstate);
+
+	local_irq_disable();
+
+	if (!(lapic_timer_reliable_states & (1 << (cstate))))
+		clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, &cpu);
+
+	kt_before = ktime_get_real();
+
+	stop_critical_timings();
+#ifndef MODULE
+	trace_power_start(POWER_CSTATE, (eax >> 4) + 1);
+#endif
+	if (!need_resched()) {
+
+		__monitor((void *)&current_thread_info()->flags, 0, 0);
+		smp_mb();
+		if (!need_resched())
+			__mwait(eax, ecx);
+	}
+
+	start_critical_timings();
+
+	kt_after = ktime_get_real();
+	usec_delta = ktime_to_us(ktime_sub(kt_after, kt_before));
+
+	local_irq_enable();
+
+	if (!(lapic_timer_reliable_states & (1 << (cstate))))
+		clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_EXIT, &cpu);
+
+	return usec_delta;
+}
+
+/*
+ * intel_idle_probe()
+ */
+static int intel_idle_probe(void)
+{
+	unsigned int eax, ebx, ecx, edx;
+
+	if (max_cstate == 0) {
+		pr_debug(PREFIX "disabled\n");
+		return -EPERM;
+	}
+
+	if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL)
+		return -ENODEV;
+
+	if (!boot_cpu_has(X86_FEATURE_MWAIT))
+		return -ENODEV;
+
+	if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF)
+		return -ENODEV;
+
+	cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &edx);
+
+	if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) ||
+		!(ecx & CPUID5_ECX_INTERRUPT_BREAK))
+			return -ENODEV;
+#ifdef DEBUG
+	if (substates == 0)	/* can over-ride via modparam */
+#endif
+		substates = edx;
+
+	pr_debug(PREFIX "MWAIT substates: 0x%x\n", substates);
+
+	if (boot_cpu_has(X86_FEATURE_ARAT))	/* Always Reliable APIC Timer */
+		lapic_timer_reliable_states = 0xFFFFFFFF;
+
+	if (boot_cpu_data.x86 != 6)	/* family 6 */
+		return -ENODEV;
+
+	switch (boot_cpu_data.x86_model) {
+
+	case 0x1A:	/* Core i7, Xeon 5500 series */
+	case 0x1E:	/* Core i7 and i5 Processor - Lynnfield Jasper Forest */
+	case 0x1F:	/* Core i7 and i5 Processor - Nehalem */
+	case 0x2E:	/* Nehalem-EX Xeon */
+		lapic_timer_reliable_states = (1 << 1);	 /* C1 */
+
+	case 0x25:	/* Westmere */
+	case 0x2C:	/* Westmere */
+		cpuidle_state_table = nehalem_cstates;
+		choose_substate = choose_tunable_substate;
+		break;
+
+	case 0x1C:	/* 28 - Atom Processor */
+		lapic_timer_reliable_states = (1 << 2) | (1 << 1); /* C2, C1 */
+		cpuidle_state_table = atom_cstates;
+		choose_substate = choose_zero_substate;
+		break;
+#ifdef FUTURE_USE
+	case 0x17:	/* 23 - Core 2 Duo */
+		lapic_timer_reliable_states = (1 << 2) | (1 << 1); /* C2, C1 */
+#endif
+
+	default:
+		pr_debug(PREFIX "does not run on family %d model %d\n",
+			boot_cpu_data.x86, boot_cpu_data.x86_model);
+		return -ENODEV;
+	}
+
+	pr_debug(PREFIX "v" INTEL_IDLE_VERSION
+		" model 0x%X\n", boot_cpu_data.x86_model);
+
+	pr_debug(PREFIX "lapic_timer_reliable_states 0x%x\n",
+		lapic_timer_reliable_states);
+	return 0;
+}
+
+/*
+ * intel_idle_cpuidle_devices_uninit()
+ * unregister, free cpuidle_devices
+ */
+static void intel_idle_cpuidle_devices_uninit(void)
+{
+	int i;
+	struct cpuidle_device *dev;
+
+	for_each_online_cpu(i) {
+		dev = per_cpu_ptr(intel_idle_cpuidle_devices, i);
+		cpuidle_unregister_device(dev);
+	}
+
+	free_percpu(intel_idle_cpuidle_devices);
+	return;
+}
+/*
+ * intel_idle_cpuidle_devices_init()
+ * allocate, initialize, register cpuidle_devices
+ */
+static int intel_idle_cpuidle_devices_init(void)
+{
+	int i, 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;
+
+		for (cstate = 1; cstate < MWAIT_MAX_NUM_CSTATES; ++cstate) {
+			int num_substates;
+
+			if (cstate > max_cstate) {
+				printk(PREFIX "max_cstate %d reached\n",
+					max_cstate);
+				break;
+			}
+
+			/* does the state exist in CPUID.MWAIT? */
+			num_substates = (substates >> ((cstate) * 4))
+						& MWAIT_SUBSTATE_MASK;
+			if (num_substates == 0)
+				continue;
+			/* is the state not enabled? */
+			if (cpuidle_state_table[cstate].enter == NULL) {
+				/* does the driver not know about the state? */
+				if (*cpuidle_state_table[cstate].name == '\0')
+					pr_debug(PREFIX "unaware of model 0x%x"
+						" MWAIT %d please"
+						" contact lenb@kernel.org",
+					boot_cpu_data.x86_model, cstate);
+				continue;
+			}
+
+			if ((cstate > 2) &&
+				!boot_cpu_has(X86_FEATURE_NONSTOP_TSC))
+				mark_tsc_unstable("TSC halts in idle"
+					" states deeper than C2");
+
+			dev->states[dev->state_count] =	/* structure copy */
+				cpuidle_state_table[cstate];
+
+			dev->state_count += 1;
+		}
+
+		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;
+		}
+	}
+
+	return 0;
+}
+
+
+static int __init intel_idle_init(void)
+{
+	int retval;
+
+	retval = intel_idle_probe();
+	if (retval)
+		return retval;
+
+	retval = cpuidle_register_driver(&intel_idle_driver);
+	if (retval) {
+		printk(KERN_DEBUG PREFIX "intel_idle yielding to %s",
+			cpuidle_get_driver()->name);
+		return retval;
+	}
+
+	retval = intel_idle_cpuidle_devices_init();
+	if (retval) {
+		cpuidle_unregister_driver(&intel_idle_driver);
+		return retval;
+	}
+
+	return 0;
+}
+
+static void __exit intel_idle_exit(void)
+{
+	intel_idle_cpuidle_devices_uninit();
+	cpuidle_unregister_driver(&intel_idle_driver);
+
+	return;
+}
+
+module_init(intel_idle_init);
+module_exit(intel_idle_exit);
+
+module_param(power_policy, int, 0644);
+module_param(max_cstate, int, 0444);
+#ifdef DEBUG
+module_param(substates, int, 0444);
+#endif
+
+MODULE_AUTHOR("Len Brown <len.brown@intel.com>");
+MODULE_DESCRIPTION("Cpuidle driver for Intel Hardware v" INTEL_IDLE_VERSION);
+MODULE_LICENSE("GPL");
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index dcf77fa826b..55215cce500 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -125,6 +125,7 @@ struct cpuidle_driver {
 #ifdef CONFIG_CPU_IDLE
 
 extern int cpuidle_register_driver(struct cpuidle_driver *drv);
+struct cpuidle_driver *cpuidle_get_driver(void);
 extern void cpuidle_unregister_driver(struct cpuidle_driver *drv);
 extern int cpuidle_register_device(struct cpuidle_device *dev);
 extern void cpuidle_unregister_device(struct cpuidle_device *dev);
@@ -137,16 +138,17 @@ extern void cpuidle_disable_device(struct cpuidle_device *dev);
 #else
 
 static inline int cpuidle_register_driver(struct cpuidle_driver *drv)
-{return 0;}
+{return -ENODEV; }
+static inline struct cpuidle_driver *cpuidle_get_driver(void) {return NULL; }
 static inline void cpuidle_unregister_driver(struct cpuidle_driver *drv) { }
 static inline int cpuidle_register_device(struct cpuidle_device *dev)
-{return 0;}
+{return -ENODEV; }
 static inline void cpuidle_unregister_device(struct cpuidle_device *dev) { }
 
 static inline void cpuidle_pause_and_lock(void) { }
 static inline void cpuidle_resume_and_unlock(void) { }
 static inline int cpuidle_enable_device(struct cpuidle_device *dev)
-{return 0;}
+{return -ENODEV; }
 static inline void cpuidle_disable_device(struct cpuidle_device *dev) { }
 
 #endif