Merge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux

Pull thermal management updates from Zhang Rui:
"Highlights:

- introduction of Dove thermal sensor driver.

- introduction of Kirkwood thermal sensor driver.

- introduction of intel_powerclamp thermal cooling device driver.

- add interrupt and DT support for rcar thermal driver.

- add thermal emulation support which allows platform thermal driver
to do software/hardware emulation for thermal issues."

* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux: (36 commits)
thermal: rcar: remove __devinitconst
thermal: return an error on failure to register thermal class
Thermal: rename thermal governor Kconfig option to avoid generic naming
thermal: exynos: Use the new thermal trend type for quick cooling action.
Thermal: exynos: Add support for temperature falling interrupt.
Thermal: Dove: Add Themal sensor support for Dove.
thermal: Add support for the thermal sensor on Kirkwood SoCs
thermal: rcar: add Device Tree support
thermal: rcar: remove machine_power_off() from rcar_thermal_notify()
thermal: rcar: add interrupt support
thermal: rcar: add read/write functions for common/priv data
thermal: rcar: multi channel support
thermal: rcar: use mutex lock instead of spin lock
thermal: rcar: enable CPCTL to use hardware TSC deciding
thermal: rcar: use parenthesis on macro
Thermal: fix a build warning when CONFIG_THERMAL_EMULATION cleared
Thermal: fix a wrong comment
thermal: sysfs: Add a new sysfs node emul_temp for thermal emulation
PM: intel_powerclamp: off by one in start_power_clamp()
thermal: exynos: Miscellaneous fixes to support falling threshold interrupt
...

Documentation/devicetree/bindings/thermal/dove-thermal.txt

···
+* Dove Thermal
+
+This driver is for Dove SoCs which contain a thermal sensor.
+
+Required properties:
+- compatible : "marvell,dove-thermal"
+- reg : Address range of the thermal registers
+
+The reg properties should contain two ranges. The first is for the
+three Thermal Manager registers, while the second range contains the
+Thermal Diode Control Registers.
+
+Example:
+
+	thermal@10078 {
+		compatible = "marvell,dove-thermal";
+		reg = <0xd001c 0x0c>, <0xd005c 0x08>;
+	};

Documentation/devicetree/bindings/thermal/kirkwood-thermal.txt

···
+* Kirkwood Thermal
+
+This version is for Kirkwood 88F8262 & 88F6283 SoCs. Other kirkwoods
+don't contain a thermal sensor.
+
+Required properties:
+- compatible : "marvell,kirkwood-thermal"
+- reg : Address range of the thermal registers
+
+Example:
+
+	thermal@10078 {
+		compatible = "marvell,kirkwood-thermal";
+		reg = <0x10078 0x4>;
+	};

Documentation/devicetree/bindings/thermal/rcar-thermal.txt

···
+* Renesas R-Car Thermal
+
+Required properties:
+- compatible		: "renesas,rcar-thermal"
+- reg			: Address range of the thermal registers.
+			  The 1st reg will be recognized as common register
+			  if it has "interrupts".
+
+Option properties:
+
+- interrupts		: use interrupt
+
+Example (non interrupt support):
+
+thermal@e61f0100 {
+	compatible = "renesas,rcar-thermal";
+	reg = <0xe61f0100 0x38>;
+};
+
+Example (interrupt support):
+
+thermal@e61f0000 {
+	compatible = "renesas,rcar-thermal";
+	reg = <0xe61f0000 0x14
+		0xe61f0100 0x38
+		0xe61f0200 0x38
+		0xe61f0300 0x38>;
+	interrupts = <0 69 4>;
+};

Documentation/thermal/exynos_thermal_emulation

···
+EXYNOS EMULATION MODE
+========================
+
+Copyright (C) 2012 Samsung Electronics
+
+Written by Jonghwa Lee <[e-mail]>
+
+Description
+-----------
+
+Exynos 4x12 (4212, 4412) and 5 series provide emulation mode for thermal management unit.
+Thermal emulation mode supports software debug for TMU's operation. User can set temperature
+manually with software code and TMU will read current temperature from user value not from
+sensor's value.
+
+Enabling CONFIG_EXYNOS_THERMAL_EMUL option will make this support in available.
+When it's enabled, sysfs node will be created under
+/sys/bus/platform/devices/'exynos device name'/ with name of 'emulation'.
+
+The sysfs node, 'emulation', will contain value 0 for the initial state. When you input any
+temperature you want to update to sysfs node, it automatically enable emulation mode and
+current temperature will be changed into it.
+(Exynos also supports user changable delay time which would be used to delay of
+ changing temperature. However, this node only uses same delay of real sensing time, 938us.)
+
+Exynos emulation mode requires synchronous of value changing and enabling. It means when you
+want to update the any value of delay or next temperature, then you have to enable emulation
+mode at the same time. (Or you have to keep the mode enabling.) If you don't, it fails to
+change the value to updated one and just use last succeessful value repeatedly. That's why
+this node gives users the right to change termerpature only. Just one interface makes it more
+simply to use.
+
+Disabling emulation mode only requires writing value 0 to sysfs node.
+
+
+TEMP	120 |
+	    |
+	100 |
+	    |
+	 80 |
+	    |		     	 	 +-----------
+	 60 |      		     	 |	    |
+	    |	           +-------------|          |
+	 40 |              |         	 |          |
+	    |		   |	     	 |          |
+	 20 |		   |	     	 |          +----------
+	    |	 	   |	     	 |          |          |
+	  0 |______________|_____________|__________|__________|_________
+		   A	    	 A	    A	   	       A     TIME
+		   |<----->|	 |<----->|  |<----->|	       |
+		   | 938us |  	 |	 |  |       |          |
+emulation    :  0  50	   |  	 70      |  20      |          0
+current temp :   sensor   50		 70         20	      sensor

Documentation/thermal/intel_powerclamp.txt

···
+			 =======================
+			 INTEL POWERCLAMP DRIVER
+			 =======================
+By: Arjan van de Ven <[e-mail]>
+    Jacob Pan <[e-mail]>
+
+Contents:
+	(*) Introduction
+	    - Goals and Objectives
+
+	(*) Theory of Operation
+	    - Idle Injection
+	    - Calibration
+
+	(*) Performance Analysis
+	    - Effectiveness and Limitations
+	    - Power vs Performance
+	    - Scalability
+	    - Calibration
+	    - Comparison with Alternative Techniques
+
+	(*) Usage and Interfaces
+	    - Generic Thermal Layer (sysfs)
+	    - Kernel APIs (TBD)
+
+============
+INTRODUCTION
+============
+
+Consider the situation where a system’s power consumption must be
+reduced at runtime, due to power budget, thermal constraint, or noise
+level, and where active cooling is not preferred. Software managed
+passive power reduction must be performed to prevent the hardware
+actions that are designed for catastrophic scenarios.
+
+Currently, P-states, T-states (clock modulation), and CPU offlining
+are used for CPU throttling.
+
+On Intel CPUs, C-states provide effective power reduction, but so far
+they’re only used opportunistically, based on workload. With the
+development of intel_powerclamp driver, the method of synchronizing
+idle injection across all online CPU threads was introduced. The goal
+is to achieve forced and controllable C-state residency.
+
+Test/Analysis has been made in the areas of power, performance,
+scalability, and user experience. In many cases, clear advantage is
+shown over taking the CPU offline or modulating the CPU clock.
+
+
+===================
+THEORY OF OPERATION
+===================
+
+Idle Injection
+--------------
+
+On modern Intel processors (Nehalem or later), package level C-state
+residency is available in MSRs, thus also available to the kernel.
+
+These MSRs are:
+      #define MSR_PKG_C2_RESIDENCY	0x60D
+      #define MSR_PKG_C3_RESIDENCY	0x3F8
+      #define MSR_PKG_C6_RESIDENCY	0x3F9
+      #define MSR_PKG_C7_RESIDENCY	0x3FA
+
+If the kernel can also inject idle time to the system, then a
+closed-loop control system can be established that manages package
+level C-state. The intel_powerclamp driver is conceived as such a
+control system, where the target set point is a user-selected idle
+ratio (based on power reduction), and the error is the difference
+between the actual package level C-state residency ratio and the target idle
+ratio.
+
+Injection is controlled by high priority kernel threads, spawned for
+each online CPU.
+
+These kernel threads, with SCHED_FIFO class, are created to perform
+clamping actions of controlled duty ratio and duration. Each per-CPU
+thread synchronizes its idle time and duration, based on the rounding
+of jiffies, so accumulated errors can be prevented to avoid a jittery
+effect. Threads are also bound to the CPU such that they cannot be
+migrated, unless the CPU is taken offline. In this case, threads
+belong to the offlined CPUs will be terminated immediately.
+
+Running as SCHED_FIFO and relatively high priority, also allows such
+scheme to work for both preemptable and non-preemptable kernels.
+Alignment of idle time around jiffies ensures scalability for HZ
+values. This effect can be better visualized using a Perf timechart.
+The following diagram shows the behavior of kernel thread
+kidle_inject/cpu. During idle injection, it runs monitor/mwait idle
+for a given "duration", then relinquishes the CPU to other tasks,
+until the next time interval.
+
+The NOHZ schedule tick is disabled during idle time, but interrupts
+are not masked. Tests show that the extra wakeups from scheduler tick
+have a dramatic impact on the effectiveness of the powerclamp driver
+on large scale systems (Westmere system with 80 processors).
+
+CPU0
+		  ____________          ____________
+kidle_inject/0   |   sleep    |  mwait |  sleep     |
+	_________|            |________|            |_______
+			       duration
+CPU1
+		  ____________          ____________
+kidle_inject/1   |   sleep    |  mwait |  sleep     |
+	_________|            |________|            |_______
+			      ^
+			      |
+			      |
+			      roundup(jiffies, interval)
+
+Only one CPU is allowed to collect statistics and update global
+control parameters. This CPU is referred to as the controlling CPU in
+this document. The controlling CPU is elected at runtime, with a
+policy that favors BSP, taking into account the possibility of a CPU
+hot-plug.
+
+In terms of dynamics of the idle control system, package level idle
+time is considered largely as a non-causal system where its behavior
+cannot be based on the past or current input. Therefore, the
+intel_powerclamp driver attempts to enforce the desired idle time
+instantly as given input (target idle ratio). After injection,
+powerclamp moniors the actual idle for a given time window and adjust
+the next injection accordingly to avoid over/under correction.
+
+When used in a causal control system, such as a temperature control,
+it is up to the user of this driver to implement algorithms where
+past samples and outputs are included in the feedback. For example, a
+PID-based thermal controller can use the powerclamp driver to
+maintain a desired target temperature, based on integral and
+derivative gains of the past samples.
+
+
+
+Calibration
+-----------
+During scalability testing, it is observed that synchronized actions
+among CPUs become challenging as the number of cores grows. This is
+also true for the ability of a system to enter package level C-states.
+
+To make sure the intel_powerclamp driver scales well, online
+calibration is implemented. The goals for doing such a calibration
+are:
+
+a) determine the effective range of idle injection ratio
+b) determine the amount of compensation needed at each target ratio
+
+Compensation to each target ratio consists of two parts:
+
+        a) steady state error compensation
+	This is to offset the error occurring when the system can
+	enter idle without extra wakeups (such as external interrupts).
+
+	b) dynamic error compensation
+	When an excessive amount of wakeups occurs during idle, an
+	additional idle ratio can be added to quiet interrupts, by
+	slowing down CPU activities.
+
+A debugfs file is provided for the user to examine compensation
+progress and results, such as on a Westmere system.
+[jacob@nex01 ~]$ cat
+/sys/kernel/debug/intel_powerclamp/powerclamp_calib
+controlling cpu: 0
+pct confidence steady dynamic (compensation)
+0	0	0	0
+1	1	0	0
+2	1	1	0
+3	3	1	0
+4	3	1	0
+5	3	1	0
+6	3	1	0
+7	3	1	0
+8	3	1	0
+...
+30	3	2	0
+31	3	2	0
+32	3	1	0
+33	3	2	0
+34	3	1	0
+35	3	2	0
+36	3	1	0
+37	3	2	0
+38	3	1	0
+39	3	2	0
+40	3	3	0
+41	3	1	0
+42	3	2	0
+43	3	1	0
+44	3	1	0
+45	3	2	0
+46	3	3	0
+47	3	0	0
+48	3	2	0
+49	3	3	0
+
+Calibration occurs during runtime. No offline method is available.
+Steady state compensation is used only when confidence levels of all
+adjacent ratios have reached satisfactory level. A confidence level
+is accumulated based on clean data collected at runtime. Data
+collected during a period without extra interrupts is considered
+clean.
+
+To compensate for excessive amounts of wakeup during idle, additional
+idle time is injected when such a condition is detected. Currently,
+we have a simple algorithm to double the injection ratio. A possible
+enhancement might be to throttle the offending IRQ, such as delaying
+EOI for level triggered interrupts. But it is a challenge to be
+non-intrusive to the scheduler or the IRQ core code.
+
+
+CPU Online/Offline
+------------------
+Per-CPU kernel threads are started/stopped upon receiving
+notifications of CPU hotplug activities. The intel_powerclamp driver
+keeps track of clamping kernel threads, even after they are migrated
+to other CPUs, after a CPU offline event.
+
+
+=====================
+Performance Analysis
+=====================
+This section describes the general performance data collected on
+multiple systems, including Westmere (80P) and Ivy Bridge (4P, 8P).
+
+Effectiveness and Limitations
+-----------------------------
+The maximum range that idle injection is allowed is capped at 50
+percent. As mentioned earlier, since interrupts are allowed during
+forced idle time, excessive interrupts could result in less
+effectiveness. The extreme case would be doing a ping -f to generated
+flooded network interrupts without much CPU acknowledgement. In this
+case, little can be done from the idle injection threads. In most
+normal cases, such as scp a large file, applications can be throttled
+by the powerclamp driver, since slowing down the CPU also slows down
+network protocol processing, which in turn reduces interrupts.
+
+When control parameters change at runtime by the controlling CPU, it
+may take an additional period for the rest of the CPUs to catch up
+with the changes. During this time, idle injection is out of sync,
+thus not able to enter package C- states at the expected ratio. But
+this effect is minor, in that in most cases change to the target
+ratio is updated much less frequently than the idle injection
+frequency.
+
+Scalability
+-----------
+Tests also show a minor, but measurable, difference between the 4P/8P
+Ivy Bridge system and the 80P Westmere server under 50% idle ratio.
+More compensation is needed on Westmere for the same amount of
+target idle ratio. The compensation also increases as the idle ratio
+gets larger. The above reason constitutes the need for the
+calibration code.
+
+On the IVB 8P system, compared to an offline CPU, powerclamp can
+achieve up to 40% better performance per watt. (measured by a spin
+counter summed over per CPU counting threads spawned for all running
+CPUs).
+
+====================
+Usage and Interfaces
+====================
+The powerclamp driver is registered to the generic thermal layer as a
+cooling device. Currently, it’s not bound to any thermal zones.
+
+jacob@chromoly:/sys/class/thermal/cooling_device14$ grep . *
+cur_state:0
+max_state:50
+type:intel_powerclamp
+
+Example usage:
+- To inject 25% idle time
+$ sudo sh -c "echo 25 > /sys/class/thermal/cooling_device80/cur_state
+"
+
+If the system is not busy and has more than 25% idle time already,
+then the powerclamp driver will not start idle injection. Using Top
+will not show idle injection kernel threads.
+
+If the system is busy (spin test below) and has less than 25% natural
+idle time, powerclamp kernel threads will do idle injection, which
+appear running to the scheduler. But the overall system idle is still
+reflected. In this example, 24.1% idle is shown. This helps the
+system admin or user determine the cause of slowdown, when a
+powerclamp driver is in action.
+
+
+Tasks: 197 total,   1 running, 196 sleeping,   0 stopped,   0 zombie
+Cpu(s): 71.2%us,  4.7%sy,  0.0%ni, 24.1%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
+Mem:   3943228k total,  1689632k used,  2253596k free,    74960k buffers
+Swap:  4087804k total,        0k used,  4087804k free,   945336k cached
+
+  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
+ 3352 jacob     20   0  262m  644  428 S  286  0.0   0:17.16 spin
+ 3341 root     -51   0     0    0    0 D   25  0.0   0:01.62 kidle_inject/0
+ 3344 root     -51   0     0    0    0 D   25  0.0   0:01.60 kidle_inject/3
+ 3342 root     -51   0     0    0    0 D   25  0.0   0:01.61 kidle_inject/1
+ 3343 root     -51   0     0    0    0 D   25  0.0   0:01.60 kidle_inject/2
+ 2935 jacob     20   0  696m 125m  35m S    5  3.3   0:31.11 firefox
+ 1546 root      20   0  158m  20m 6640 S    3  0.5   0:26.97 Xorg
+ 2100 jacob     20   0 1223m  88m  30m S    3  2.3   0:23.68 compiz
+
+Tests have shown that by using the powerclamp driver as a cooling
+device, a PID based userspace thermal controller can manage to
+control CPU temperature effectively, when no other thermal influence
+is added. For example, a UltraBook user can compile the kernel under
+certain temperature (below most active trip points).

Documentation/thermal/sysfs-api.txt

···
 	.get_trip_type: get the type of certain trip point.
 	.get_trip_temp: get the temperature above which the certain trip point
 			will be fired.
+	.set_emul_temp: set the emulation temperature which helps in debugging
+			different threshold temperature points.
 
 1.1.2 void thermal_zone_device_unregister(struct thermal_zone_device *tz)
 
···
     |---trip_point_[0-*]_temp:	Trip point temperature
     |---trip_point_[0-*]_type:	Trip point type
     |---trip_point_[0-*]_hyst:	Hysteresis value for this trip point
+    |---emul_temp:		Emulated temperature set node
 
 Thermal cooling device sys I/F, created once it's registered:
 /sys/class/thermal/cooling_device[0-*]:
···
 	Valid values: 0 (disabled) or greater than 1000
 	RW, Optional
 
+emul_temp
+	Interface to set the emulated temperature method in thermal zone
+	(sensor). After setting this temperature, the thermal zone may pass
+	this temperature to platform emulation function if registered or
+	cache it locally. This is useful in debugging different temperature
+	threshold and its associated cooling action. This is write only node
+	and writing 0 on this node should disable emulation.
+	Unit: millidegree Celsius
+	WO, Optional
+
 *****************************
 * Cooling device attributes *
 *****************************
···
 netlink event. Netlink socket initialization is done during the _init_
 of the framework. Drivers which intend to use the notification mechanism
 just need to call thermal_generate_netlink_event() with two arguments viz
-(originator, event). Typically the originator will be an integer assigned
-to a thermal_zone_device when it registers itself with the framework. The
+(originator, event). The originator is a pointer to struct thermal_zone_device
+from where the event has been originated. An integer which represents the
+thermal zone device will be used in the message to identify the zone. The
 event will be one of:{THERMAL_AUX0, THERMAL_AUX1, THERMAL_CRITICAL,
 THERMAL_DEV_FAULT}. Notification can be sent when the current temperature
 crosses any of the configured thresholds.

arch/x86/kernel/nmi.c

···
 {
 	__this_cpu_write(last_nmi_rip, 0);
 }
+EXPORT_SYMBOL_GPL(local_touch_nmi);

drivers/thermal/Kconfig

···
 
 config THERMAL_DEFAULT_GOV_STEP_WISE
 	bool "step_wise"
-	select STEP_WISE
+	select THERMAL_GOV_STEP_WISE
 	help
 	  Use the step_wise governor as default. This throttles the
 	  devices one step at a time.
 
 config THERMAL_DEFAULT_GOV_FAIR_SHARE
 	bool "fair_share"
-	select FAIR_SHARE
+	select THERMAL_GOV_FAIR_SHARE
 	help
 	  Use the fair_share governor as default. This throttles the
 	  devices based on their 'contribution' to a zone. The
···
 
 config THERMAL_DEFAULT_GOV_USER_SPACE
 	bool "user_space"
-	select USER_SPACE
+	select THERMAL_GOV_USER_SPACE
 	help
 	  Select this if you want to let the user space manage the
 	  lpatform thermals.
 
 endchoice
 
-config FAIR_SHARE
+config THERMAL_GOV_FAIR_SHARE
 	bool "Fair-share thermal governor"
 	help
 	  Enable this to manage platform thermals using fair-share governor.
 
-config STEP_WISE
+config THERMAL_GOV_STEP_WISE
 	bool "Step_wise thermal governor"
 	help
 	  Enable this to manage platform thermals using a simple linear
 
-config USER_SPACE
+config THERMAL_GOV_USER_SPACE
 	bool "User_space thermal governor"
 	help
 	  Enable this to let the user space manage the platform thermals.
···
 	  and not the ACPI interface.
 	  If you want this support, you should say Y here.
 
+config THERMAL_EMULATION
+	bool "Thermal emulation mode support"
+	help
+	  Enable this option to make a emul_temp sysfs node in thermal zone
+	  directory to support temperature emulation. With emulation sysfs node,
+	  user can manually input temperature and test the different trip
+	  threshold behaviour for simulation purpose.
+
 config SPEAR_THERMAL
 	bool "SPEAr thermal sensor driver"
 	depends on PLAT_SPEAR
···
 	  Enable this to plug the R-Car thermal sensor driver into the Linux
 	  thermal framework
 
+config KIRKWOOD_THERMAL
+	tristate "Temperature sensor on Marvell Kirkwood SoCs"
+	depends on ARCH_KIRKWOOD
+	depends on OF
+	help
+	  Support for the Kirkwood thermal sensor driver into the Linux thermal
+	  framework. Only kirkwood 88F6282 and 88F6283 have this sensor.
+
 config EXYNOS_THERMAL
 	tristate "Temperature sensor on Samsung EXYNOS"
 	depends on (ARCH_EXYNOS4 || ARCH_EXYNOS5)
···
 	help
 	  If you say yes here you get support for TMU (Thermal Management
 	  Unit) on SAMSUNG EXYNOS series of SoC.
+
+config EXYNOS_THERMAL_EMUL
+	bool "EXYNOS TMU emulation mode support"
+	depends on EXYNOS_THERMAL
+	help
+	  Exynos 4412 and 4414 and 5 series has emulation mode on TMU.
+	  Enable this option will be make sysfs node in exynos thermal platform
+	  device directory to support emulation mode. With emulation mode sysfs
+	  node, you can manually input temperature to TMU for simulation purpose.
+
+config DOVE_THERMAL
+	tristate "Temperature sensor on Marvell Dove SoCs"
+	depends on ARCH_DOVE
+	depends on OF
+	help
+	  Support for the Dove thermal sensor driver in the Linux thermal
+	  framework.
 
 config DB8500_THERMAL
 	bool "DB8500 thermal management"
···
 	  bound to thermal zone trip points. When a trip point reached, the
 	  bound cpufreq cooling device turns active to set CPU frequency low to
 	  cool down the CPU.
+
+config INTEL_POWERCLAMP
+	tristate "Intel PowerClamp idle injection driver"
+	depends on THERMAL
+	depends on X86
+	depends on CPU_SUP_INTEL
+	help
+	  Enable this to enable Intel PowerClamp idle injection driver. This
+	  enforce idle time which results in more package C-state residency. The
+	  user interface is exposed via generic thermal framework.
 
 endif

drivers/thermal/Makefile

···
 obj-$(CONFIG_THERMAL)		+= thermal_sys.o
 
 # governors
-obj-$(CONFIG_FAIR_SHARE)	+= fair_share.o
-obj-$(CONFIG_STEP_WISE)		+= step_wise.o
-obj-$(CONFIG_USER_SPACE)	+= user_space.o
+obj-$(CONFIG_THERMAL_GOV_FAIR_SHARE)	+= fair_share.o
+obj-$(CONFIG_THERMAL_GOV_STEP_WISE)	+= step_wise.o
+obj-$(CONFIG_THERMAL_GOV_USER_SPACE)	+= user_space.o
 
 # cpufreq cooling
 obj-$(CONFIG_CPU_THERMAL)	+= cpu_cooling.o
···
 # platform thermal drivers
 obj-$(CONFIG_SPEAR_THERMAL)	+= spear_thermal.o
 obj-$(CONFIG_RCAR_THERMAL)	+= rcar_thermal.o
+obj-$(CONFIG_KIRKWOOD_THERMAL)  += kirkwood_thermal.o
 obj-$(CONFIG_EXYNOS_THERMAL)	+= exynos_thermal.o
+obj-$(CONFIG_DOVE_THERMAL)  	+= dove_thermal.o
 obj-$(CONFIG_DB8500_THERMAL)	+= db8500_thermal.o
 obj-$(CONFIG_DB8500_CPUFREQ_COOLING)	+= db8500_cpufreq_cooling.o
+obj-$(CONFIG_INTEL_POWERCLAMP)	+= intel_powerclamp.o
+

drivers/thermal/cpu_cooling.c

···
 /**
  * get_cpu_frequency - get the absolute value of frequency from level.
  * @cpu: cpu for which frequency is fetched.
- * @level: level of frequency of the CPU
- *	e.g level=1 --> 1st MAX FREQ, LEVEL=2 ---> 2nd MAX FREQ, .... etc
+ * @level: level of frequency, equals cooling state of cpu cooling device
+ *	e.g level=0 --> 1st MAX FREQ, level=1 ---> 2nd MAX FREQ, .... etc
  */
 static unsigned int get_cpu_frequency(unsigned int cpu, unsigned long level)
 {

drivers/thermal/db8500_cpufreq_cooling.c

···
 #include <linux/cpufreq.h>
 #include <linux/err.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>
 
···
 	{ .compatible = "stericsson,db8500-cpufreq-cooling" },
 	{},
 };
-#else
-#define db8500_cpufreq_cooling_match NULL
 #endif
 
 static struct platform_driver db8500_cpufreq_cooling_driver = {
 	.driver = {
 		.owner = THIS_MODULE,
 		.name = "db8500-cpufreq-cooling",
-		.of_match_table = db8500_cpufreq_cooling_match,
+		.of_match_table = of_match_ptr(db8500_cpufreq_cooling_match),
 	},
 	.probe = db8500_cpufreq_cooling_probe,
 	.suspend = db8500_cpufreq_cooling_suspend,

drivers/thermal/db8500_thermal.c

···
 	{ .compatible = "stericsson,db8500-thermal" },
 	{},
 };
-#else
-#define db8500_thermal_match NULL
 #endif
 
 static struct platform_driver db8500_thermal_driver = {
 	.driver = {
 		.owner = THIS_MODULE,
 		.name = "db8500-thermal",
-		.of_match_table = db8500_thermal_match,
+		.of_match_table = of_match_ptr(db8500_thermal_match),
 	},
 	.probe = db8500_thermal_probe,
 	.suspend = db8500_thermal_suspend,

drivers/thermal/dove_thermal.c

···
+/*
+ * Dove thermal sensor driver
+ *
+ * Copyright (C) 2013 Andrew Lunn <[e-mail]>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that 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.
+ *
+ */
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/of.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/thermal.h>
+
+#define DOVE_THERMAL_TEMP_OFFSET	1
+#define DOVE_THERMAL_TEMP_MASK		0x1FF
+
+/* Dove Thermal Manager Control and Status Register */
+#define PMU_TM_DISABLE_OFFS		0
+#define PMU_TM_DISABLE_MASK		(0x1 << PMU_TM_DISABLE_OFFS)
+
+/* Dove Theraml Diode Control 0 Register */
+#define PMU_TDC0_SW_RST_MASK		(0x1 << 1)
+#define PMU_TDC0_SEL_VCAL_OFFS		5
+#define PMU_TDC0_SEL_VCAL_MASK		(0x3 << PMU_TDC0_SEL_VCAL_OFFS)
+#define PMU_TDC0_REF_CAL_CNT_OFFS	11
+#define PMU_TDC0_REF_CAL_CNT_MASK	(0x1FF << PMU_TDC0_REF_CAL_CNT_OFFS)
+#define PMU_TDC0_AVG_NUM_OFFS		25
+#define PMU_TDC0_AVG_NUM_MASK		(0x7 << PMU_TDC0_AVG_NUM_OFFS)
+
+/* Dove Thermal Diode Control 1 Register */
+#define PMU_TEMP_DIOD_CTRL1_REG		0x04
+#define PMU_TDC1_TEMP_VALID_MASK	(0x1 << 10)
+
+/* Dove Thermal Sensor Dev Structure */
+struct dove_thermal_priv {
+	void __iomem *sensor;
+	void __iomem *control;
+};
+
+static int dove_init_sensor(const struct dove_thermal_priv *priv)
+{
+	u32 reg;
+	u32 i;
+
+	/* Configure the Diode Control Register #0 */
+	reg = readl_relaxed(priv->control);
+
+	/* Use average of 2 */
+	reg &= ~PMU_TDC0_AVG_NUM_MASK;
+	reg |= (0x1 << PMU_TDC0_AVG_NUM_OFFS);
+
+	/* Reference calibration value */
+	reg &= ~PMU_TDC0_REF_CAL_CNT_MASK;
+	reg |= (0x0F1 << PMU_TDC0_REF_CAL_CNT_OFFS);
+
+	/* Set the high level reference for calibration */
+	reg &= ~PMU_TDC0_SEL_VCAL_MASK;
+	reg |= (0x2 << PMU_TDC0_SEL_VCAL_OFFS);
+	writel(reg, priv->control);
+
+	/* Reset the sensor */
+	reg = readl_relaxed(priv->control);
+	writel((reg | PMU_TDC0_SW_RST_MASK), priv->control);
+	writel(reg, priv->control);
+
+	/* Enable the sensor */
+	reg = readl_relaxed(priv->sensor);
+	reg &= ~PMU_TM_DISABLE_MASK;
+	writel(reg, priv->sensor);
+
+	/* Poll the sensor for the first reading */
+	for (i = 0; i < 1000000; i++) {
+		reg = readl_relaxed(priv->sensor);
+		if (reg & DOVE_THERMAL_TEMP_MASK)
+			break;
+	}
+
+	if (i == 1000000)
+		return -EIO;
+
+	return 0;
+}
+
+static int dove_get_temp(struct thermal_zone_device *thermal,
+			  unsigned long *temp)
+{
+	unsigned long reg;
+	struct dove_thermal_priv *priv = thermal->devdata;
+
+	/* Valid check */
+	reg = readl_relaxed(priv->control + PMU_TEMP_DIOD_CTRL1_REG);
+	if ((reg & PMU_TDC1_TEMP_VALID_MASK) == 0x0) {
+		dev_err(&thermal->device,
+			"Temperature sensor reading not valid\n");
+		return -EIO;
+	}
+
+	/*
+	 * Calculate temperature. See Section 8.10.1 of 88AP510,
+	 * Documentation/arm/Marvell/README
+	 */
+	reg = readl_relaxed(priv->sensor);
+	reg = (reg >> DOVE_THERMAL_TEMP_OFFSET) & DOVE_THERMAL_TEMP_MASK;
+	*temp = ((2281638UL - (7298*reg)) / 10);
+
+	return 0;
+}
+
+static struct thermal_zone_device_ops ops = {
+	.get_temp = dove_get_temp,
+};
+
+static const struct of_device_id dove_thermal_id_table[] = {
+	{ .compatible = "marvell,dove-thermal" },
+	{}
+};
+
+static int dove_thermal_probe(struct platform_device *pdev)
+{
+	struct thermal_zone_device *thermal = NULL;
+	struct dove_thermal_priv *priv;
+	struct resource *res;
+	int ret;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		dev_err(&pdev->dev, "Failed to get platform resource\n");
+		return -ENODEV;
+	}
+
+	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->sensor = devm_request_and_ioremap(&pdev->dev, res);
+	if (!priv->sensor) {
+		dev_err(&pdev->dev, "Failed to request_ioremap memory\n");
+		return -EADDRNOTAVAIL;
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	if (!res) {
+		dev_err(&pdev->dev, "Failed to get platform resource\n");
+		return -ENODEV;
+	}
+	priv->control = devm_request_and_ioremap(&pdev->dev, res);
+	if (!priv->control) {
+		dev_err(&pdev->dev, "Failed to request_ioremap memory\n");
+		return -EADDRNOTAVAIL;
+	}
+
+	ret = dove_init_sensor(priv);
+	if (ret) {
+		dev_err(&pdev->dev, "Failed to initialize sensor\n");
+		return ret;
+	}
+
+	thermal = thermal_zone_device_register("dove_thermal", 0, 0,
+					       priv, &ops, NULL, 0, 0);
+	if (IS_ERR(thermal)) {
+		dev_err(&pdev->dev,
+			"Failed to register thermal zone device\n");
+		return PTR_ERR(thermal);
+	}
+
+	platform_set_drvdata(pdev, thermal);
+
+	return 0;
+}
+
+static int dove_thermal_exit(struct platform_device *pdev)
+{
+	struct thermal_zone_device *dove_thermal =
+		platform_get_drvdata(pdev);
+
+	thermal_zone_device_unregister(dove_thermal);
+	platform_set_drvdata(pdev, NULL);
+
+	return 0;
+}
+
+MODULE_DEVICE_TABLE(of, dove_thermal_id_table);
+
+static struct platform_driver dove_thermal_driver = {
+	.probe = dove_thermal_probe,
+	.remove = dove_thermal_exit,
+	.driver = {
+		.name = "dove_thermal",
+		.owner = THIS_MODULE,
+		.of_match_table = of_match_ptr(dove_thermal_id_table),
+	},
+};
+
+module_platform_driver(dove_thermal_driver);
+
+MODULE_AUTHOR("Andrew Lunn <[e-mail]>");
+MODULE_DESCRIPTION("Dove thermal driver");
+MODULE_LICENSE("GPL");

drivers/thermal/exynos_thermal.c

···
 
 #define EXYNOS_TRIMINFO_RELOAD		0x1
 #define EXYNOS_TMU_CLEAR_RISE_INT	0x111
-#define EXYNOS_TMU_CLEAR_FALL_INT	(0x111 << 16)
+#define EXYNOS_TMU_CLEAR_FALL_INT	(0x111 << 12)
 #define EXYNOS_MUX_ADDR_VALUE		6
 #define EXYNOS_MUX_ADDR_SHIFT		20
 #define EXYNOS_TMU_TRIP_MODE_SHIFT	13
···
 #define SENSOR_NAME_LEN	16
 #define MAX_TRIP_COUNT	8
 #define MAX_COOLING_DEVICE 4
+#define MAX_THRESHOLD_LEVS 4
 
 #define ACTIVE_INTERVAL 500
 #define IDLE_INTERVAL 10000
 #define MCELSIUS	1000
+
+#ifdef CONFIG_EXYNOS_THERMAL_EMUL
+#define EXYNOS_EMUL_TIME	0x57F0
+#define EXYNOS_EMUL_TIME_SHIFT	16
+#define EXYNOS_EMUL_DATA_SHIFT	8
+#define EXYNOS_EMUL_DATA_MASK	0xFF
+#define EXYNOS_EMUL_ENABLE	0x1
+#endif /* CONFIG_EXYNOS_THERMAL_EMUL */
 
 /* CPU Zone information */
 #define PANIC_ZONE      4
···
 struct	thermal_trip_point_conf {
 	int trip_val[MAX_TRIP_COUNT];
 	int trip_count;
+	u8 trigger_falling;
 };
 
 struct	thermal_cooling_conf {
···
 
 	mutex_lock(&th_zone->therm_dev->lock);
 
-	if (mode == THERMAL_DEVICE_ENABLED)
+	if (mode == THERMAL_DEVICE_ENABLED &&
+		!th_zone->sensor_conf->trip_data.trigger_falling)
 		th_zone->therm_dev->polling_delay = IDLE_INTERVAL;
 	else
 		th_zone->therm_dev->polling_delay = 0;
···
 		case MONITOR_ZONE:
 		case WARN_ZONE:
 			if (thermal_zone_bind_cooling_device(thermal, i, cdev,
-								level, level)) {
+								level, 0)) {
 				pr_err("error binding cdev inst %d\n", i);
 				ret = -EINVAL;
 			}
···
 static int exynos_get_trend(struct thermal_zone_device *thermal,
 			int trip, enum thermal_trend *trend)
 {
-	if (thermal->temperature >= trip)
-		*trend = THERMAL_TREND_RAISING;
+	int ret;
+	unsigned long trip_temp;
+
+	ret = exynos_get_trip_temp(thermal, trip, &trip_temp);
+	if (ret < 0)
+		return ret;
+
+	if (thermal->temperature >= trip_temp)
+		*trend = THERMAL_TREND_RAISE_FULL;
 	else
-		*trend = THERMAL_TREND_DROPPING;
+		*trend = THERMAL_TREND_DROP_FULL;
 
 	return 0;
 }
···
 			break;
 	}
 
-	if (th_zone->mode == THERMAL_DEVICE_ENABLED) {
+	if (th_zone->mode == THERMAL_DEVICE_ENABLED &&
+		!th_zone->sensor_conf->trip_data.trigger_falling) {
 		if (i > 0)
 			th_zone->therm_dev->polling_delay = ACTIVE_INTERVAL;
 		else
···
 
 	th_zone->therm_dev = thermal_zone_device_register(sensor_conf->name,
 			EXYNOS_ZONE_COUNT, 0, NULL, &exynos_dev_ops, NULL, 0,
-			IDLE_INTERVAL);
+			sensor_conf->trip_data.trigger_falling ?
+			0 : IDLE_INTERVAL);
 
 	if (IS_ERR(th_zone->therm_dev)) {
 		pr_err("Failed to register thermal zone device\n");
···
 {
 	struct exynos_tmu_data *data = platform_get_drvdata(pdev);
 	struct exynos_tmu_platform_data *pdata = data->pdata;
-	unsigned int status, trim_info, rising_threshold;
-	int ret = 0, threshold_code;
+	unsigned int status, trim_info;
+	unsigned int rising_threshold = 0, falling_threshold = 0;
+	int ret = 0, threshold_code, i, trigger_levs = 0;
 
 	mutex_lock(&data->lock);
 	clk_enable(data->clk);
···
 			(data->temp_error2 != 0))
 		data->temp_error1 = pdata->efuse_value;
 
+	/* Count trigger levels to be enabled */
+	for (i = 0; i < MAX_THRESHOLD_LEVS; i++)
+		if (pdata->trigger_levels[i])
+			trigger_levs++;
+
 	if (data->soc == SOC_ARCH_EXYNOS4210) {
 		/* Write temperature code for threshold */
 		threshold_code = temp_to_code(data, pdata->threshold);
···
 		}
 		writeb(threshold_code,
 			data->base + EXYNOS4210_TMU_REG_THRESHOLD_TEMP);
-
-		writeb(pdata->trigger_levels[0],
-			data->base + EXYNOS4210_TMU_REG_TRIG_LEVEL0);
-		writeb(pdata->trigger_levels[1],
-			data->base + EXYNOS4210_TMU_REG_TRIG_LEVEL1);
-		writeb(pdata->trigger_levels[2],
-			data->base + EXYNOS4210_TMU_REG_TRIG_LEVEL2);
-		writeb(pdata->trigger_levels[3],
-			data->base + EXYNOS4210_TMU_REG_TRIG_LEVEL3);
+		for (i = 0; i < trigger_levs; i++)
+			writeb(pdata->trigger_levels[i],
+			data->base + EXYNOS4210_TMU_REG_TRIG_LEVEL0 + i * 4);
 
 		writel(EXYNOS4210_TMU_INTCLEAR_VAL,
 			data->base + EXYNOS_TMU_REG_INTCLEAR);
 	} else if (data->soc == SOC_ARCH_EXYNOS) {
-		/* Write temperature code for threshold */
-		threshold_code = temp_to_code(data, pdata->trigger_levels[0]);
-		if (threshold_code < 0) {
-			ret = threshold_code;
-			goto out;
+		/* Write temperature code for rising and falling threshold */
+		for (i = 0; i < trigger_levs; i++) {
+			threshold_code = temp_to_code(data,
+						pdata->trigger_levels[i]);
+			if (threshold_code < 0) {
+				ret = threshold_code;
+				goto out;
+			}
+			rising_threshold |= threshold_code << 8 * i;
+			if (pdata->threshold_falling) {
+				threshold_code = temp_to_code(data,
+						pdata->trigger_levels[i] -
+						pdata->threshold_falling);
+				if (threshold_code > 0)
+					falling_threshold |=
+						threshold_code << 8 * i;
+			}
 		}
-		rising_threshold = threshold_code;
-		threshold_code = temp_to_code(data, pdata->trigger_levels[1]);
-		if (threshold_code < 0) {
-			ret = threshold_code;
-			goto out;
-		}
-		rising_threshold |= (threshold_code << 8);
-		threshold_code = temp_to_code(data, pdata->trigger_levels[2]);
-		if (threshold_code < 0) {
-			ret = threshold_code;
-			goto out;
-		}
-		rising_threshold |= (threshold_code << 16);
 
 		writel(rising_threshold,
 				data->base + EXYNOS_THD_TEMP_RISE);
-		writel(0, data->base + EXYNOS_THD_TEMP_FALL);
+		writel(falling_threshold,
+				data->base + EXYNOS_THD_TEMP_FALL);
 
-		writel(EXYNOS_TMU_CLEAR_RISE_INT|EXYNOS_TMU_CLEAR_FALL_INT,
+		writel(EXYNOS_TMU_CLEAR_RISE_INT | EXYNOS_TMU_CLEAR_FALL_INT,
 				data->base + EXYNOS_TMU_REG_INTCLEAR);
 	}
 out:
···
 			pdata->trigger_level2_en << 8 |
 			pdata->trigger_level1_en << 4 |
 			pdata->trigger_level0_en;
+		if (pdata->threshold_falling)
+			interrupt_en |= interrupt_en << 16;
 	} else {
 		con |= EXYNOS_TMU_CORE_OFF;
 		interrupt_en = 0; /* Disable all interrupts */
···
 	struct exynos_tmu_data *data = container_of(work,
 			struct exynos_tmu_data, irq_work);
 
+	exynos_report_trigger();
 	mutex_lock(&data->lock);
 	clk_enable(data->clk);
-
-
 	if (data->soc == SOC_ARCH_EXYNOS)
-		writel(EXYNOS_TMU_CLEAR_RISE_INT,
+		writel(EXYNOS_TMU_CLEAR_RISE_INT |
+				EXYNOS_TMU_CLEAR_FALL_INT,
 				data->base + EXYNOS_TMU_REG_INTCLEAR);
 	else
 		writel(EXYNOS4210_TMU_INTCLEAR_VAL,
 				data->base + EXYNOS_TMU_REG_INTCLEAR);
-
 	clk_disable(data->clk);
 	mutex_unlock(&data->lock);
-	exynos_report_trigger();
+
 	enable_irq(data->irq);
 }
 
···
 
 #if defined(CONFIG_SOC_EXYNOS5250) || defined(CONFIG_SOC_EXYNOS4412)
 static struct exynos_tmu_platform_data const exynos_default_tmu_data = {
+	.threshold_falling = 10,
 	.trigger_levels[0] = 85,
 	.trigger_levels[1] = 103,
 	.trigger_levels[2] = 110,
···
 	{},
 };
 MODULE_DEVICE_TABLE(of, exynos_tmu_match);
-#else
-#define  exynos_tmu_match NULL
 #endif
 
 static struct platform_device_id exynos_tmu_driver_ids[] = {
···
 	return (struct exynos_tmu_platform_data *)
 			platform_get_device_id(pdev)->driver_data;
 }
+
+#ifdef CONFIG_EXYNOS_THERMAL_EMUL
+static ssize_t exynos_tmu_emulation_show(struct device *dev,
+					 struct device_attribute *attr,
+					 char *buf)
+{
+	struct platform_device *pdev = container_of(dev,
+					struct platform_device, dev);
+	struct exynos_tmu_data *data = platform_get_drvdata(pdev);
+	unsigned int reg;
+	u8 temp_code;
+	int temp = 0;
+
+	if (data->soc == SOC_ARCH_EXYNOS4210)
+		goto out;
+
+	mutex_lock(&data->lock);
+	clk_enable(data->clk);
+	reg = readl(data->base + EXYNOS_EMUL_CON);
+	clk_disable(data->clk);
+	mutex_unlock(&data->lock);
+
+	if (reg & EXYNOS_EMUL_ENABLE) {
+		reg >>= EXYNOS_EMUL_DATA_SHIFT;
+		temp_code = reg & EXYNOS_EMUL_DATA_MASK;
+		temp = code_to_temp(data, temp_code);
+	}
+out:
+	return sprintf(buf, "%d\n", temp * MCELSIUS);
+}
+
+static ssize_t exynos_tmu_emulation_store(struct device *dev,
+					struct device_attribute *attr,
+					const char *buf, size_t count)
+{
+	struct platform_device *pdev = container_of(dev,
+					struct platform_device, dev);
+	struct exynos_tmu_data *data = platform_get_drvdata(pdev);
+	unsigned int reg;
+	int temp;
+
+	if (data->soc == SOC_ARCH_EXYNOS4210)
+		goto out;
+
+	if (!sscanf(buf, "%d\n", &temp) || temp < 0)
+		return -EINVAL;
+
+	mutex_lock(&data->lock);
+	clk_enable(data->clk);
+
+	reg = readl(data->base + EXYNOS_EMUL_CON);
+
+	if (temp) {
+		/* Both CELSIUS and MCELSIUS type are available for input */
+		if (temp > MCELSIUS)
+			temp /= MCELSIUS;
+
+		reg = (EXYNOS_EMUL_TIME << EXYNOS_EMUL_TIME_SHIFT) |
+			(temp_to_code(data, (temp / MCELSIUS))
+			 << EXYNOS_EMUL_DATA_SHIFT) | EXYNOS_EMUL_ENABLE;
+	} else {
+		reg &= ~EXYNOS_EMUL_ENABLE;
+	}
+
+	writel(reg, data->base + EXYNOS_EMUL_CON);
+
+	clk_disable(data->clk);
+	mutex_unlock(&data->lock);
+
+out:
+	return count;
+}
+
+static DEVICE_ATTR(emulation, 0644, exynos_tmu_emulation_show,
+					exynos_tmu_emulation_store);
+static int create_emulation_sysfs(struct device *dev)
+{
+	return device_create_file(dev, &dev_attr_emulation);
+}
+static void remove_emulation_sysfs(struct device *dev)
+{
+	device_remove_file(dev, &dev_attr_emulation);
+}
+#else
+static inline int create_emulation_sysfs(struct device *dev) { return 0; }
+static inline void remove_emulation_sysfs(struct device *dev) {}
+#endif
+
 static int exynos_tmu_probe(struct platform_device *pdev)
 {
 	struct exynos_tmu_data *data;
···
 		exynos_sensor_conf.trip_data.trip_val[i] =
 			pdata->threshold + pdata->trigger_levels[i];
 
+	exynos_sensor_conf.trip_data.trigger_falling = pdata->threshold_falling;
+
 	exynos_sensor_conf.cooling_data.freq_clip_count =
 						pdata->freq_tab_count;
 	for (i = 0; i < pdata->freq_tab_count; i++) {
···
 		dev_err(&pdev->dev, "Failed to register thermal interface\n");
 		goto err_clk;
 	}
+
+	ret = create_emulation_sysfs(&pdev->dev);
+	if (ret)
+		dev_err(&pdev->dev, "Failed to create emulation mode sysfs node\n");
+
 	return 0;
 err_clk:
 	platform_set_drvdata(pdev, NULL);
···
 static int exynos_tmu_remove(struct platform_device *pdev)
 {
 	struct exynos_tmu_data *data = platform_get_drvdata(pdev);
+
+	remove_emulation_sysfs(&pdev->dev);
 
 	exynos_tmu_control(pdev, false);
 
···
 		.name   = "exynos-tmu",
 		.owner  = THIS_MODULE,
 		.pm     = EXYNOS_TMU_PM,
-		.of_match_table = exynos_tmu_match,
+		.of_match_table = of_match_ptr(exynos_tmu_match),
 	},
 	.probe = exynos_tmu_probe,
 	.remove	= exynos_tmu_remove,

drivers/thermal/intel_powerclamp.c

···
+/*
+ * intel_powerclamp.c - package c-state idle injection
+ *
+ * Copyright (c) 2012, Intel Corporation.
+ *
+ * Authors:
+ *     Arjan van de Ven <[e-mail]>
+ *     Jacob Pan <[e-mail]>
+ *
+ * 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.
+ *
+ *
+ *	TODO:
+ *           1. better handle wakeup from external interrupts, currently a fixed
+ *              compensation is added to clamping duration when excessive amount
+ *              of wakeups are observed during idle time. the reason is that in
+ *              case of external interrupts without need for ack, clamping down
+ *              cpu in non-irq context does not reduce irq. for majority of the
+ *              cases, clamping down cpu does help reduce irq as well, we should
+ *              be able to differenciate the two cases and give a quantitative
+ *              solution for the irqs that we can control. perhaps based on
+ *              get_cpu_iowait_time_us()
+ *
+ *	     2. synchronization with other hw blocks
+ *
+ *
+ */
+
+#define pr_fmt(fmt)	KBUILD_MODNAME ": " fmt
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/kthread.h>
+#include <linux/freezer.h>
+#include <linux/cpu.h>
+#include <linux/thermal.h>
+#include <linux/slab.h>
+#include <linux/tick.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+
+#include <asm/nmi.h>
+#include <asm/msr.h>
+#include <asm/mwait.h>
+#include <asm/cpu_device_id.h>
+#include <asm/idle.h>
+#include <asm/hardirq.h>
+
+#define MAX_TARGET_RATIO (50U)
+/* For each undisturbed clamping period (no extra wake ups during idle time),
+ * we increment the confidence counter for the given target ratio.
+ * CONFIDENCE_OK defines the level where runtime calibration results are
+ * valid.
+ */
+#define CONFIDENCE_OK (3)
+/* Default idle injection duration, driver adjust sleep time to meet target
+ * idle ratio. Similar to frequency modulation.
+ */
+#define DEFAULT_DURATION_JIFFIES (6)
+
+static unsigned int target_mwait;
+static struct dentry *debug_dir;
+
+/* user selected target */
+static unsigned int set_target_ratio;
+static unsigned int current_ratio;
+static bool should_skip;
+static bool reduce_irq;
+static atomic_t idle_wakeup_counter;
+static unsigned int control_cpu; /* The cpu assigned to collect stat and update
+				  * control parameters. default to BSP but BSP
+				  * can be offlined.
+				  */
+static bool clamping;
+
+
+static struct task_struct * __percpu *powerclamp_thread;
+static struct thermal_cooling_device *cooling_dev;
+static unsigned long *cpu_clamping_mask;  /* bit map for tracking per cpu
+					   * clamping thread
+					   */
+
+static unsigned int duration;
+static unsigned int pkg_cstate_ratio_cur;
+static unsigned int window_size;
+
+static int duration_set(const char *arg, const struct kernel_param *kp)
+{
+	int ret = 0;
+	unsigned long new_duration;
+
+	ret = kstrtoul(arg, 10, &new_duration);
+	if (ret)
+		goto exit;
+	if (new_duration > 25 || new_duration < 6) {
+		pr_err("Out of recommended range %lu, between 6-25ms\n",
+			new_duration);
+		ret = -EINVAL;
+	}
+
+	duration = clamp(new_duration, 6ul, 25ul);
+	smp_mb();
+
+exit:
+
+	return ret;
+}
+
+static struct kernel_param_ops duration_ops = {
+	.set = duration_set,
+	.get = param_get_int,
+};
+
+
+module_param_cb(duration, &duration_ops, &duration, 0644);
+MODULE_PARM_DESC(duration, "forced idle time for each attempt in msec.");
+
+struct powerclamp_calibration_data {
+	unsigned long confidence;  /* used for calibration, basically a counter
+				    * gets incremented each time a clamping
+				    * period is completed without extra wakeups
+				    * once that counter is reached given level,
+				    * compensation is deemed usable.
+				    */
+	unsigned long steady_comp; /* steady state compensation used when
+				    * no extra wakeups occurred.
+				    */
+	unsigned long dynamic_comp; /* compensate excessive wakeup from idle
+				     * mostly from external interrupts.
+				     */
+};
+
+static struct powerclamp_calibration_data cal_data[MAX_TARGET_RATIO];
+
+static int window_size_set(const char *arg, const struct kernel_param *kp)
+{
+	int ret = 0;
+	unsigned long new_window_size;
+
+	ret = kstrtoul(arg, 10, &new_window_size);
+	if (ret)
+		goto exit_win;
+	if (new_window_size > 10 || new_window_size < 2) {
+		pr_err("Out of recommended window size %lu, between 2-10\n",
+			new_window_size);
+		ret = -EINVAL;
+	}
+
+	window_size = clamp(new_window_size, 2ul, 10ul);
+	smp_mb();
+
+exit_win:
+
+	return ret;
+}
+
+static struct kernel_param_ops window_size_ops = {
+	.set = window_size_set,
+	.get = param_get_int,
+};
+
+module_param_cb(window_size, &window_size_ops, &window_size, 0644);
+MODULE_PARM_DESC(window_size, "sliding window in number of clamping cycles\n"
+	"\tpowerclamp controls idle ratio within this window. larger\n"
+	"\twindow size results in slower response time but more smooth\n"
+	"\tclamping results. default to 2.");
+
+static void find_target_mwait(void)
+{
+	unsigned int eax, ebx, ecx, edx;
+	unsigned int highest_cstate = 0;
+	unsigned int highest_subcstate = 0;
+	int i;
+
+	if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF)
+		return;
+
+	cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &edx);
+
+	if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) ||
+	    !(ecx & CPUID5_ECX_INTERRUPT_BREAK))
+		return;
+
+	edx >>= MWAIT_SUBSTATE_SIZE;
+	for (i = 0; i < 7 && edx; i++, edx >>= MWAIT_SUBSTATE_SIZE) {
+		if (edx & MWAIT_SUBSTATE_MASK) {
+			highest_cstate = i;
+			highest_subcstate = edx & MWAIT_SUBSTATE_MASK;
+		}
+	}
+	target_mwait = (highest_cstate << MWAIT_SUBSTATE_SIZE) |
+		(highest_subcstate - 1);
+
+}
+
+static u64 pkg_state_counter(void)
+{
+	u64 val;
+	u64 count = 0;
+
+	static bool skip_c2;
+	static bool skip_c3;
+	static bool skip_c6;
+	static bool skip_c7;
+
+	if (!skip_c2) {
+		if (!rdmsrl_safe(MSR_PKG_C2_RESIDENCY, &val))
+			count += val;
+		else
+			skip_c2 = true;
+	}
+
+	if (!skip_c3) {
+		if (!rdmsrl_safe(MSR_PKG_C3_RESIDENCY, &val))
+			count += val;
+		else
+			skip_c3 = true;
+	}
+
+	if (!skip_c6) {
+		if (!rdmsrl_safe(MSR_PKG_C6_RESIDENCY, &val))
+			count += val;
+		else
+			skip_c6 = true;
+	}
+
+	if (!skip_c7) {
+		if (!rdmsrl_safe(MSR_PKG_C7_RESIDENCY, &val))
+			count += val;
+		else
+			skip_c7 = true;
+	}
+
+	return count;
+}
+
+static void noop_timer(unsigned long foo)
+{
+	/* empty... just the fact that we get the interrupt wakes us up */
+}
+
+static unsigned int get_compensation(int ratio)
+{
+	unsigned int comp = 0;
+
+	/* we only use compensation if all adjacent ones are good */
+	if (ratio == 1 &&
+		cal_data[ratio].confidence >= CONFIDENCE_OK &&
+		cal_data[ratio + 1].confidence >= CONFIDENCE_OK &&
+		cal_data[ratio + 2].confidence >= CONFIDENCE_OK) {
+		comp = (cal_data[ratio].steady_comp +
+			cal_data[ratio + 1].steady_comp +
+			cal_data[ratio + 2].steady_comp) / 3;
+	} else if (ratio == MAX_TARGET_RATIO - 1 &&
+		cal_data[ratio].confidence >= CONFIDENCE_OK &&
+		cal_data[ratio - 1].confidence >= CONFIDENCE_OK &&
+		cal_data[ratio - 2].confidence >= CONFIDENCE_OK) {
+		comp = (cal_data[ratio].steady_comp +
+			cal_data[ratio - 1].steady_comp +
+			cal_data[ratio - 2].steady_comp) / 3;
+	} else if (cal_data[ratio].confidence >= CONFIDENCE_OK &&
+		cal_data[ratio - 1].confidence >= CONFIDENCE_OK &&
+		cal_data[ratio + 1].confidence >= CONFIDENCE_OK) {
+		comp = (cal_data[ratio].steady_comp +
+			cal_data[ratio - 1].steady_comp +
+			cal_data[ratio + 1].steady_comp) / 3;
+	}
+
+	/* REVISIT: simple penalty of double idle injection */
+	if (reduce_irq)
+		comp = ratio;
+	/* do not exceed limit */
+	if (comp + ratio >= MAX_TARGET_RATIO)
+		comp = MAX_TARGET_RATIO - ratio - 1;
+
+	return comp;
+}
+
+static void adjust_compensation(int target_ratio, unsigned int win)
+{
+	int delta;
+	struct powerclamp_calibration_data *d = &cal_data[target_ratio];
+
+	/*
+	 * adjust compensations if confidence level has not been reached or
+	 * there are too many wakeups during the last idle injection period, we
+	 * cannot trust the data for compensation.
+	 */
+	if (d->confidence >= CONFIDENCE_OK ||
+		atomic_read(&idle_wakeup_counter) >
+		win * num_online_cpus())
+		return;
+
+	delta = set_target_ratio - current_ratio;
+	/* filter out bad data */
+	if (delta >= 0 && delta <= (1+target_ratio/10)) {
+		if (d->steady_comp)
+			d->steady_comp =
+				roundup(delta+d->steady_comp, 2)/2;
+		else
+			d->steady_comp = delta;
+		d->confidence++;
+	}
+}
+
+static bool powerclamp_adjust_controls(unsigned int target_ratio,
+				unsigned int guard, unsigned int win)
+{
+	static u64 msr_last, tsc_last;
+	u64 msr_now, tsc_now;
+	u64 val64;
+
+	/* check result for the last window */
+	msr_now = pkg_state_counter();
+	rdtscll(tsc_now);
+
+	/* calculate pkg cstate vs tsc ratio */
+	if (!msr_last || !tsc_last)
+		current_ratio = 1;
+	else if (tsc_now-tsc_last) {
+		val64 = 100*(msr_now-msr_last);
+		do_div(val64, (tsc_now-tsc_last));
+		current_ratio = val64;
+	}
+
+	/* update record */
+	msr_last = msr_now;
+	tsc_last = tsc_now;
+
+	adjust_compensation(target_ratio, win);
+	/*
+	 * too many external interrupts, set flag such
+	 * that we can take measure later.
+	 */
+	reduce_irq = atomic_read(&idle_wakeup_counter) >=
+		2 * win * num_online_cpus();
+
+	atomic_set(&idle_wakeup_counter, 0);
+	/* if we are above target+guard, skip */
+	return set_target_ratio + guard <= current_ratio;
+}
+
+static int clamp_thread(void *arg)
+{
+	int cpunr = (unsigned long)arg;
+	DEFINE_TIMER(wakeup_timer, noop_timer, 0, 0);
+	static const struct sched_param param = {
+		.sched_priority = MAX_USER_RT_PRIO/2,
+	};
+	unsigned int count = 0;
+	unsigned int target_ratio;
+
+	set_bit(cpunr, cpu_clamping_mask);
+	set_freezable();
+	init_timer_on_stack(&wakeup_timer);
+	sched_setscheduler(current, SCHED_FIFO, &param);
+
+	while (true == clamping && !kthread_should_stop() &&
+		cpu_online(cpunr)) {
+		int sleeptime;
+		unsigned long target_jiffies;
+		unsigned int guard;
+		unsigned int compensation = 0;
+		int interval; /* jiffies to sleep for each attempt */
+		unsigned int duration_jiffies = msecs_to_jiffies(duration);
+		unsigned int window_size_now;
+
+		try_to_freeze();
+		/*
+		 * make sure user selected ratio does not take effect until
+		 * the next round. adjust target_ratio if user has changed
+		 * target such that we can converge quickly.
+		 */
+		target_ratio = set_target_ratio;
+		guard = 1 + target_ratio/20;
+		window_size_now = window_size;
+		count++;
+
+		/*
+		 * systems may have different ability to enter package level
+		 * c-states, thus we need to compensate the injected idle ratio
+		 * to achieve the actual target reported by the HW.
+		 */
+		compensation = get_compensation(target_ratio);
+		interval = duration_jiffies*100/(target_ratio+compensation);
+
+		/* align idle time */
+		target_jiffies = roundup(jiffies, interval);
+		sleeptime = target_jiffies - jiffies;
+		if (sleeptime <= 0)
+			sleeptime = 1;
+		schedule_timeout_interruptible(sleeptime);
+		/*
+		 * only elected controlling cpu can collect stats and update
+		 * control parameters.
+		 */
+		if (cpunr == control_cpu && !(count%window_size_now)) {
+			should_skip =
+				powerclamp_adjust_controls(target_ratio,
+							guard, window_size_now);
+			smp_mb();
+		}
+
+		if (should_skip)
+			continue;
+
+		target_jiffies = jiffies + duration_jiffies;
+		mod_timer(&wakeup_timer, target_jiffies);
+		if (unlikely(local_softirq_pending()))
+			continue;
+		/*
+		 * stop tick sched during idle time, interrupts are still
+		 * allowed. thus jiffies are updated properly.
+		 */
+		preempt_disable();
+		tick_nohz_idle_enter();
+		/* mwait until target jiffies is reached */
+		while (time_before(jiffies, target_jiffies)) {
+			unsigned long ecx = 1;
+			unsigned long eax = target_mwait;
+
+			/*
+			 * REVISIT: may call enter_idle() to notify drivers who
+			 * can save power during cpu idle. same for exit_idle()
+			 */
+			local_touch_nmi();
+			stop_critical_timings();
+			__monitor((void *)&current_thread_info()->flags, 0, 0);
+			cpu_relax(); /* allow HT sibling to run */
+			__mwait(eax, ecx);
+			start_critical_timings();
+			atomic_inc(&idle_wakeup_counter);
+		}
+		tick_nohz_idle_exit();
+		preempt_enable_no_resched();
+	}
+	del_timer_sync(&wakeup_timer);
+	clear_bit(cpunr, cpu_clamping_mask);
+
+	return 0;
+}
+
+/*
+ * 1 HZ polling while clamping is active, useful for userspace
+ * to monitor actual idle ratio.
+ */
+static void poll_pkg_cstate(struct work_struct *dummy);
+static DECLARE_DELAYED_WORK(poll_pkg_cstate_work, poll_pkg_cstate);
+static void poll_pkg_cstate(struct work_struct *dummy)
+{
+	static u64 msr_last;
+	static u64 tsc_last;
+	static unsigned long jiffies_last;
+
+	u64 msr_now;
+	unsigned long jiffies_now;
+	u64 tsc_now;
+	u64 val64;
+
+	msr_now = pkg_state_counter();
+	rdtscll(tsc_now);
+	jiffies_now = jiffies;
+
+	/* calculate pkg cstate vs tsc ratio */
+	if (!msr_last || !tsc_last)
+		pkg_cstate_ratio_cur = 1;
+	else {
+		if (tsc_now - tsc_last) {
+			val64 = 100 * (msr_now - msr_last);
+			do_div(val64, (tsc_now - tsc_last));
+			pkg_cstate_ratio_cur = val64;
+		}
+	}
+
+	/* update record */
+	msr_last = msr_now;
+	jiffies_last = jiffies_now;
+	tsc_last = tsc_now;
+
+	if (true == clamping)
+		schedule_delayed_work(&poll_pkg_cstate_work, HZ);
+}
+
+static int start_power_clamp(void)
+{
+	unsigned long cpu;
+	struct task_struct *thread;
+
+	/* check if pkg cstate counter is completely 0, abort in this case */
+	if (!pkg_state_counter()) {
+		pr_err("pkg cstate counter not functional, abort\n");
+		return -EINVAL;
+	}
+
+	set_target_ratio = clamp(set_target_ratio, 0U, MAX_TARGET_RATIO - 1);
+	/* prevent cpu hotplug */
+	get_online_cpus();
+
+	/* prefer BSP */
+	control_cpu = 0;
+	if (!cpu_online(control_cpu))
+		control_cpu = smp_processor_id();
+
+	clamping = true;
+	schedule_delayed_work(&poll_pkg_cstate_work, 0);
+
+	/* start one thread per online cpu */
+	for_each_online_cpu(cpu) {
+		struct task_struct **p =
+			per_cpu_ptr(powerclamp_thread, cpu);
+
+		thread = kthread_create_on_node(clamp_thread,
+						(void *) cpu,
+						cpu_to_node(cpu),
+						"kidle_inject/%ld", cpu);
+		/* bind to cpu here */
+		if (likely(!IS_ERR(thread))) {
+			kthread_bind(thread, cpu);
+			wake_up_process(thread);
+			*p = thread;
+		}
+
+	}
+	put_online_cpus();
+
+	return 0;
+}
+
+static void end_power_clamp(void)
+{
+	int i;
+	struct task_struct *thread;
+
+	clamping = false;
+	/*
+	 * make clamping visible to other cpus and give per cpu clamping threads
+	 * sometime to exit, or gets killed later.
+	 */
+	smp_mb();
+	msleep(20);
+	if (bitmap_weight(cpu_clamping_mask, num_possible_cpus())) {
+		for_each_set_bit(i, cpu_clamping_mask, num_possible_cpus()) {
+			pr_debug("clamping thread for cpu %d alive, kill\n", i);
+			thread = *per_cpu_ptr(powerclamp_thread, i);
+			kthread_stop(thread);
+		}
+	}
+}
+
+static int powerclamp_cpu_callback(struct notifier_block *nfb,
+				unsigned long action, void *hcpu)
+{
+	unsigned long cpu = (unsigned long)hcpu;
+	struct task_struct *thread;
+	struct task_struct **percpu_thread =
+		per_cpu_ptr(powerclamp_thread, cpu);
+
+	if (false == clamping)
+		goto exit_ok;
+
+	switch (action) {
+	case CPU_ONLINE:
+		thread = kthread_create_on_node(clamp_thread,
+						(void *) cpu,
+						cpu_to_node(cpu),
+						"kidle_inject/%lu", cpu);
+		if (likely(!IS_ERR(thread))) {
+			kthread_bind(thread, cpu);
+			wake_up_process(thread);
+			*percpu_thread = thread;
+		}
+		/* prefer BSP as controlling CPU */
+		if (cpu == 0) {
+			control_cpu = 0;
+			smp_mb();
+		}
+		break;
+	case CPU_DEAD:
+		if (test_bit(cpu, cpu_clamping_mask)) {
+			pr_err("cpu %lu dead but powerclamping thread is not\n",
+				cpu);
+			kthread_stop(*percpu_thread);
+		}
+		if (cpu == control_cpu) {
+			control_cpu = smp_processor_id();
+			smp_mb();
+		}
+	}
+
+exit_ok:
+	return NOTIFY_OK;
+}
+
+static struct notifier_block powerclamp_cpu_notifier = {
+	.notifier_call = powerclamp_cpu_callback,
+};
+
+static int powerclamp_get_max_state(struct thermal_cooling_device *cdev,
+				 unsigned long *state)
+{
+	*state = MAX_TARGET_RATIO;
+
+	return 0;
+}
+
+static int powerclamp_get_cur_state(struct thermal_cooling_device *cdev,
+				 unsigned long *state)
+{
+	if (true == clamping)
+		*state = pkg_cstate_ratio_cur;
+	else
+		/* to save power, do not poll idle ratio while not clamping */
+		*state = -1; /* indicates invalid state */
+
+	return 0;
+}
+
+static int powerclamp_set_cur_state(struct thermal_cooling_device *cdev,
+				 unsigned long new_target_ratio)
+{
+	int ret = 0;
+
+	new_target_ratio = clamp(new_target_ratio, 0UL,
+				(unsigned long) (MAX_TARGET_RATIO-1));
+	if (set_target_ratio == 0 && new_target_ratio > 0) {
+		pr_info("Start idle injection to reduce power\n");
+		set_target_ratio = new_target_ratio;
+		ret = start_power_clamp();
+		goto exit_set;
+	} else	if (set_target_ratio > 0 && new_target_ratio == 0) {
+		pr_info("Stop forced idle injection\n");
+		set_target_ratio = 0;
+		end_power_clamp();
+	} else	/* adjust currently running */ {
+		set_target_ratio = new_target_ratio;
+		/* make new set_target_ratio visible to other cpus */
+		smp_mb();
+	}
+
+exit_set:
+	return ret;
+}
+
+/* bind to generic thermal layer as cooling device*/
+static struct thermal_cooling_device_ops powerclamp_cooling_ops = {
+	.get_max_state = powerclamp_get_max_state,
+	.get_cur_state = powerclamp_get_cur_state,
+	.set_cur_state = powerclamp_set_cur_state,
+};
+
+/* runs on Nehalem and later */
+static const struct x86_cpu_id intel_powerclamp_ids[] = {
+	{ X86_VENDOR_INTEL, 6, 0x1a},
+	{ X86_VENDOR_INTEL, 6, 0x1c},
+	{ X86_VENDOR_INTEL, 6, 0x1e},
+	{ X86_VENDOR_INTEL, 6, 0x1f},
+	{ X86_VENDOR_INTEL, 6, 0x25},
+	{ X86_VENDOR_INTEL, 6, 0x26},
+	{ X86_VENDOR_INTEL, 6, 0x2a},
+	{ X86_VENDOR_INTEL, 6, 0x2c},
+	{ X86_VENDOR_INTEL, 6, 0x2d},
+	{ X86_VENDOR_INTEL, 6, 0x2e},
+	{ X86_VENDOR_INTEL, 6, 0x2f},
+	{ X86_VENDOR_INTEL, 6, 0x3a},
+	{}
+};
+MODULE_DEVICE_TABLE(x86cpu, intel_powerclamp_ids);
+
+static int powerclamp_probe(void)
+{
+	if (!x86_match_cpu(intel_powerclamp_ids)) {
+		pr_err("Intel powerclamp does not run on family %d model %d\n",
+				boot_cpu_data.x86, boot_cpu_data.x86_model);
+		return -ENODEV;
+	}
+	if (!boot_cpu_has(X86_FEATURE_NONSTOP_TSC) ||
+		!boot_cpu_has(X86_FEATURE_CONSTANT_TSC) ||
+		!boot_cpu_has(X86_FEATURE_MWAIT) ||
+		!boot_cpu_has(X86_FEATURE_ARAT))
+		return -ENODEV;
+
+	/* find the deepest mwait value */
+	find_target_mwait();
+
+	return 0;
+}
+
+static int powerclamp_debug_show(struct seq_file *m, void *unused)
+{
+	int i = 0;
+
+	seq_printf(m, "controlling cpu: %d\n", control_cpu);
+	seq_printf(m, "pct confidence steady dynamic (compensation)\n");
+	for (i = 0; i < MAX_TARGET_RATIO; i++) {
+		seq_printf(m, "%d\t%lu\t%lu\t%lu\n",
+			i,
+			cal_data[i].confidence,
+			cal_data[i].steady_comp,
+			cal_data[i].dynamic_comp);
+	}
+
+	return 0;
+}
+
+static int powerclamp_debug_open(struct inode *inode,
+			struct file *file)
+{
+	return single_open(file, powerclamp_debug_show, inode->i_private);
+}
+
+static const struct file_operations powerclamp_debug_fops = {
+	.open		= powerclamp_debug_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.owner		= THIS_MODULE,
+};
+
+static inline void powerclamp_create_debug_files(void)
+{
+	debug_dir = debugfs_create_dir("intel_powerclamp", NULL);
+	if (!debug_dir)
+		return;
+
+	if (!debugfs_create_file("powerclamp_calib", S_IRUGO, debug_dir,
+					cal_data, &powerclamp_debug_fops))
+		goto file_error;
+
+	return;
+
+file_error:
+	debugfs_remove_recursive(debug_dir);
+}
+
+static int powerclamp_init(void)
+{
+	int retval;
+	int bitmap_size;
+
+	bitmap_size = BITS_TO_LONGS(num_possible_cpus()) * sizeof(long);
+	cpu_clamping_mask = kzalloc(bitmap_size, GFP_KERNEL);
+	if (!cpu_clamping_mask)
+		return -ENOMEM;
+
+	/* probe cpu features and ids here */
+	retval = powerclamp_probe();
+	if (retval)
+		return retval;
+	/* set default limit, maybe adjusted during runtime based on feedback */
+	window_size = 2;
+	register_hotcpu_notifier(&powerclamp_cpu_notifier);
+	powerclamp_thread = alloc_percpu(struct task_struct *);
+	cooling_dev = thermal_cooling_device_register("intel_powerclamp", NULL,
+						&powerclamp_cooling_ops);
+	if (IS_ERR(cooling_dev))
+		return -ENODEV;
+
+	if (!duration)
+		duration = jiffies_to_msecs(DEFAULT_DURATION_JIFFIES);
+	powerclamp_create_debug_files();
+
+	return 0;
+}
+module_init(powerclamp_init);
+
+static void powerclamp_exit(void)
+{
+	unregister_hotcpu_notifier(&powerclamp_cpu_notifier);
+	end_power_clamp();
+	free_percpu(powerclamp_thread);
+	thermal_cooling_device_unregister(cooling_dev);
+	kfree(cpu_clamping_mask);
+
+	cancel_delayed_work_sync(&poll_pkg_cstate_work);
+	debugfs_remove_recursive(debug_dir);
+}
+module_exit(powerclamp_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Arjan van de Ven <[e-mail]>");
+MODULE_AUTHOR("Jacob Pan <[e-mail]>");
+MODULE_DESCRIPTION("Package Level C-state Idle Injection for Intel CPUs");

drivers/thermal/kirkwood_thermal.c

···
+/*
+ * Kirkwood thermal sensor driver
+ *
+ * Copyright (C) 2012 Nobuhiro Iwamatsu <[e-mail]>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that 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.
+ *
+ */
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/of.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/thermal.h>
+
+#define KIRKWOOD_THERMAL_VALID_OFFSET	9
+#define KIRKWOOD_THERMAL_VALID_MASK	0x1
+#define KIRKWOOD_THERMAL_TEMP_OFFSET	10
+#define KIRKWOOD_THERMAL_TEMP_MASK	0x1FF
+
+/* Kirkwood Thermal Sensor Dev Structure */
+struct kirkwood_thermal_priv {
+	void __iomem *sensor;
+};
+
+static int kirkwood_get_temp(struct thermal_zone_device *thermal,
+			  unsigned long *temp)
+{
+	unsigned long reg;
+	struct kirkwood_thermal_priv *priv = thermal->devdata;
+
+	reg = readl_relaxed(priv->sensor);
+
+	/* Valid check */
+	if (!(reg >> KIRKWOOD_THERMAL_VALID_OFFSET) &
+	    KIRKWOOD_THERMAL_VALID_MASK) {
+		dev_err(&thermal->device,
+			"Temperature sensor reading not valid\n");
+		return -EIO;
+	}
+
+	/*
+	 * Calculate temperature. See Section 8.10.1 of the 88AP510,
+	 * datasheet, which has the same sensor.
+	 * Documentation/arm/Marvell/README
+	 */
+	reg = (reg >> KIRKWOOD_THERMAL_TEMP_OFFSET) &
+		KIRKWOOD_THERMAL_TEMP_MASK;
+	*temp = ((2281638UL - (7298*reg)) / 10);
+
+	return 0;
+}
+
+static struct thermal_zone_device_ops ops = {
+	.get_temp = kirkwood_get_temp,
+};
+
+static const struct of_device_id kirkwood_thermal_id_table[] = {
+	{ .compatible = "marvell,kirkwood-thermal" },
+	{}
+};
+
+static int kirkwood_thermal_probe(struct platform_device *pdev)
+{
+	struct thermal_zone_device *thermal = NULL;
+	struct kirkwood_thermal_priv *priv;
+	struct resource *res;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		dev_err(&pdev->dev, "Failed to get platform resource\n");
+		return -ENODEV;
+	}
+
+	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->sensor = devm_request_and_ioremap(&pdev->dev, res);
+	if (!priv->sensor) {
+		dev_err(&pdev->dev, "Failed to request_ioremap memory\n");
+		return -EADDRNOTAVAIL;
+	}
+
+	thermal = thermal_zone_device_register("kirkwood_thermal", 0, 0,
+					       priv, &ops, NULL, 0, 0);
+	if (IS_ERR(thermal)) {
+		dev_err(&pdev->dev,
+			"Failed to register thermal zone device\n");
+		return PTR_ERR(thermal);
+	}
+
+	platform_set_drvdata(pdev, thermal);
+
+	return 0;
+}
+
+static int kirkwood_thermal_exit(struct platform_device *pdev)
+{
+	struct thermal_zone_device *kirkwood_thermal =
+		platform_get_drvdata(pdev);
+
+	thermal_zone_device_unregister(kirkwood_thermal);
+	platform_set_drvdata(pdev, NULL);
+
+	return 0;
+}
+
+MODULE_DEVICE_TABLE(of, kirkwood_thermal_id_table);
+
+static struct platform_driver kirkwood_thermal_driver = {
+	.probe = kirkwood_thermal_probe,
+	.remove = kirkwood_thermal_exit,
+	.driver = {
+		.name = "kirkwood_thermal",
+		.owner = THIS_MODULE,
+		.of_match_table = of_match_ptr(kirkwood_thermal_id_table),
+	},
+};
+
+module_platform_driver(kirkwood_thermal_driver);
+
+MODULE_AUTHOR("Nobuhiro Iwamatsu <[e-mail]>");
+MODULE_DESCRIPTION("kirkwood thermal driver");
+MODULE_LICENSE("GPL");

drivers/thermal/rcar_thermal.c

···
  */
 #include <linux/delay.h>
 #include <linux/err.h>
+#include <linux/irq.h>
+#include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/module.h>
 #include <linux/platform_device.h>
+#include <linux/reboot.h>
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 #include <linux/thermal.h>
 
-#define THSCR	0x2c
-#define THSSR	0x30
+#define IDLE_INTERVAL	5000
+
+#define COMMON_STR	0x00
+#define COMMON_ENR	0x04
+#define COMMON_INTMSK	0x0c
+
+#define REG_POSNEG	0x20
+#define REG_FILONOFF	0x28
+#define REG_THSCR	0x2c
+#define REG_THSSR	0x30
+#define REG_INTCTRL	0x34
 
 /* THSCR */
-#define CPTAP	0xf
+#define CPCTL	(1 << 12)
 
 /* THSSR */
 #define CTEMP	0x3f
 
+struct rcar_thermal_common {
+	void __iomem *base;
+	struct device *dev;
+	struct list_head head;
+	spinlock_t lock;
+};
 
 struct rcar_thermal_priv {
 	void __iomem *base;
-	struct device *dev;
-	spinlock_t lock;
-	u32 comp;
+	struct rcar_thermal_common *common;
+	struct thermal_zone_device *zone;
+	struct delayed_work work;
+	struct mutex lock;
+	struct list_head list;
+	int id;
+	int ctemp;
 };
 
+#define rcar_thermal_for_each_priv(pos, common)	\
+	list_for_each_entry(pos, &common->head, list)
+
 #define MCELSIUS(temp)			((temp) * 1000)
-#define rcar_zone_to_priv(zone)		(zone->devdata)
+#define rcar_zone_to_priv(zone)		((zone)->devdata)
+#define rcar_priv_to_dev(priv)		((priv)->common->dev)
+#define rcar_has_irq_support(priv)	((priv)->common->base)
+#define rcar_id_to_shift(priv)		((priv)->id * 8)
+
+#ifdef DEBUG
+# define rcar_force_update_temp(priv)	1
+#else
+# define rcar_force_update_temp(priv)	0
+#endif
 
 /*
  *		basic functions
  */
-static u32 rcar_thermal_read(struct rcar_thermal_priv *priv, u32 reg)
+#define rcar_thermal_common_read(c, r) \
+	_rcar_thermal_common_read(c, COMMON_ ##r)
+static u32 _rcar_thermal_common_read(struct rcar_thermal_common *common,
+				     u32 reg)
 {
-	unsigned long flags;
-	u32 ret;
-
-	spin_lock_irqsave(&priv->lock, flags);
-
-	ret = ioread32(priv->base + reg);
-
-	spin_unlock_irqrestore(&priv->lock, flags);
-
-	return ret;
+	return ioread32(common->base + reg);
 }
 
-#if 0 /* no user at this point */
-static void rcar_thermal_write(struct rcar_thermal_priv *priv,
-			       u32 reg, u32 data)
+#define rcar_thermal_common_write(c, r, d) \
+	_rcar_thermal_common_write(c, COMMON_ ##r, d)
+static void _rcar_thermal_common_write(struct rcar_thermal_common *common,
+				       u32 reg, u32 data)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&priv->lock, flags);
-
-	iowrite32(data, priv->base + reg);
-
-	spin_unlock_irqrestore(&priv->lock, flags);
+	iowrite32(data, common->base + reg);
 }
-#endif
 
-static void rcar_thermal_bset(struct rcar_thermal_priv *priv, u32 reg,
-			      u32 mask, u32 data)
+#define rcar_thermal_common_bset(c, r, m, d) \
+	_rcar_thermal_common_bset(c, COMMON_ ##r, m, d)
+static void _rcar_thermal_common_bset(struct rcar_thermal_common *common,
+				      u32 reg, u32 mask, u32 data)
 {
-	unsigned long flags;
 	u32 val;
 
-	spin_lock_irqsave(&priv->lock, flags);
+	val = ioread32(common->base + reg);
+	val &= ~mask;
+	val |= (data & mask);
+	iowrite32(val, common->base + reg);
+}
+
+#define rcar_thermal_read(p, r) _rcar_thermal_read(p, REG_ ##r)
+static u32 _rcar_thermal_read(struct rcar_thermal_priv *priv, u32 reg)
+{
+	return ioread32(priv->base + reg);
+}
+
+#define rcar_thermal_write(p, r, d) _rcar_thermal_write(p, REG_ ##r, d)
+static void _rcar_thermal_write(struct rcar_thermal_priv *priv,
+				u32 reg, u32 data)
+{
+	iowrite32(data, priv->base + reg);
+}
+
+#define rcar_thermal_bset(p, r, m, d) _rcar_thermal_bset(p, REG_ ##r, m, d)
+static void _rcar_thermal_bset(struct rcar_thermal_priv *priv, u32 reg,
+			       u32 mask, u32 data)
+{
+	u32 val;
 
 	val = ioread32(priv->base + reg);
 	val &= ~mask;
 	val |= (data & mask);
 	iowrite32(val, priv->base + reg);
-
-	spin_unlock_irqrestore(&priv->lock, flags);
 }
 
 /*
  *		zone device functions
  */
-static int rcar_thermal_get_temp(struct thermal_zone_device *zone,
-			   unsigned long *temp)
+static int rcar_thermal_update_temp(struct rcar_thermal_priv *priv)
 {
-	struct rcar_thermal_priv *priv = rcar_zone_to_priv(zone);
-	int val, min, max, tmp;
+	struct device *dev = rcar_priv_to_dev(priv);
+	int i;
+	int ctemp, old, new;
 
-	tmp = -200; /* default */
-	while (1) {
-		if (priv->comp < 1 || priv->comp > 12) {
-			dev_err(priv->dev,
-				"THSSR invalid data (%d)\n", priv->comp);
-			priv->comp = 4; /* for next thermal */
-			return -EINVAL;
-		}
+	mutex_lock(&priv->lock);
 
-		/*
-		 * THS comparator offset and the reference temperature
-		 *
-		 * Comparator	| reference	| Temperature field
-		 * offset	| temperature	| measurement
-		 *		| (degrees C)	| (degrees C)
-		 * -------------+---------------+-------------------
-		 *  1		|  -45		|  -45 to  -30
-		 *  2		|  -30		|  -30 to  -15
-		 *  3		|  -15		|  -15 to    0
-		 *  4		|    0		|    0 to  +15
-		 *  5		|  +15		|  +15 to  +30
-		 *  6		|  +30		|  +30 to  +45
-		 *  7		|  +45		|  +45 to  +60
-		 *  8		|  +60		|  +60 to  +75
-		 *  9		|  +75		|  +75 to  +90
-		 * 10		|  +90		|  +90 to +105
-		 * 11		| +105		| +105 to +120
-		 * 12		| +120		| +120 to +135
-		 */
+	/*
+	 * TSC decides a value of CPTAP automatically,
+	 * and this is the conditions which validate interrupt.
+	 */
+	rcar_thermal_bset(priv, THSCR, CPCTL, CPCTL);
 
-		/* calculate thermal limitation */
-		min = (priv->comp * 15) - 60;
-		max = min + 15;
-
+	ctemp = 0;
+	old = ~0;
+	for (i = 0; i < 128; i++) {
 		/*
 		 * we need to wait 300us after changing comparator offset
 		 * to get stable temperature.
 		 * see "Usage Notes" on datasheet
 		 */
-		rcar_thermal_bset(priv, THSCR, CPTAP, priv->comp);
 		udelay(300);
 
-		/* calculate current temperature */
-		val = rcar_thermal_read(priv, THSSR) & CTEMP;
-		val = (val * 5) - 65;
-
-		dev_dbg(priv->dev, "comp/min/max/val = %d/%d/%d/%d\n",
-			priv->comp, min, max, val);
-
-		/*
-		 * If val is same as min/max, then,
-		 * it should try again on next comparator.
-		 * But the val might be correct temperature.
-		 * Keep it on "tmp" and compare with next val.
-		 */
-		if (tmp == val)
-			break;
-
-		if (val <= min) {
-			tmp = min;
-			priv->comp--; /* try again */
-		} else if (val >= max) {
-			tmp = max;
-			priv->comp++; /* try again */
-		} else {
-			tmp = val;
+		new = rcar_thermal_read(priv, THSSR) & CTEMP;
+		if (new == old) {
+			ctemp = new;
 			break;
 		}
+		old = new;
 	}
 
-	*temp = MCELSIUS(tmp);
+	if (!ctemp) {
+		dev_err(dev, "thermal sensor was broken\n");
+		return -EINVAL;
+	}
+
+	/*
+	 * enable IRQ
+	 */
+	if (rcar_has_irq_support(priv)) {
+		rcar_thermal_write(priv, FILONOFF, 0);
+
+		/* enable Rising/Falling edge interrupt */
+		rcar_thermal_write(priv, POSNEG,  0x1);
+		rcar_thermal_write(priv, INTCTRL, (((ctemp - 0) << 8) |
+						   ((ctemp - 1) << 0)));
+	}
+
+	dev_dbg(dev, "thermal%d  %d -> %d\n", priv->id, priv->ctemp, ctemp);
+
+	priv->ctemp = ctemp;
+
+	mutex_unlock(&priv->lock);
+
+	return 0;
+}
+
+static int rcar_thermal_get_temp(struct thermal_zone_device *zone,
+				 unsigned long *temp)
+{
+	struct rcar_thermal_priv *priv = rcar_zone_to_priv(zone);
+
+	if (!rcar_has_irq_support(priv) || rcar_force_update_temp(priv))
+		rcar_thermal_update_temp(priv);
+
+	mutex_lock(&priv->lock);
+	*temp =  MCELSIUS((priv->ctemp * 5) - 65);
+	mutex_unlock(&priv->lock);
+
+	return 0;
+}
+
+static int rcar_thermal_get_trip_type(struct thermal_zone_device *zone,
+				      int trip, enum thermal_trip_type *type)
+{
+	struct rcar_thermal_priv *priv = rcar_zone_to_priv(zone);
+	struct device *dev = rcar_priv_to_dev(priv);
+
+	/* see rcar_thermal_get_temp() */
+	switch (trip) {
+	case 0: /* +90 <= temp */
+		*type = THERMAL_TRIP_CRITICAL;
+		break;
+	default:
+		dev_err(dev, "rcar driver trip error\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int rcar_thermal_get_trip_temp(struct thermal_zone_device *zone,
+				      int trip, unsigned long *temp)
+{
+	struct rcar_thermal_priv *priv = rcar_zone_to_priv(zone);
+	struct device *dev = rcar_priv_to_dev(priv);
+
+	/* see rcar_thermal_get_temp() */
+	switch (trip) {
+	case 0: /* +90 <= temp */
+		*temp = MCELSIUS(90);
+		break;
+	default:
+		dev_err(dev, "rcar driver trip error\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int rcar_thermal_notify(struct thermal_zone_device *zone,
+			       int trip, enum thermal_trip_type type)
+{
+	struct rcar_thermal_priv *priv = rcar_zone_to_priv(zone);
+	struct device *dev = rcar_priv_to_dev(priv);
+
+	switch (type) {
+	case THERMAL_TRIP_CRITICAL:
+		/* FIXME */
+		dev_warn(dev, "Thermal reached to critical temperature\n");
+		break;
+	default:
+		break;
+	}
+
 	return 0;
 }
 
 static struct thermal_zone_device_ops rcar_thermal_zone_ops = {
-	.get_temp = rcar_thermal_get_temp,
+	.get_temp	= rcar_thermal_get_temp,
+	.get_trip_type	= rcar_thermal_get_trip_type,
+	.get_trip_temp	= rcar_thermal_get_trip_temp,
+	.notify		= rcar_thermal_notify,
 };
+
+/*
+ *		interrupt
+ */
+#define rcar_thermal_irq_enable(p)	_rcar_thermal_irq_ctrl(p, 1)
+#define rcar_thermal_irq_disable(p)	_rcar_thermal_irq_ctrl(p, 0)
+static void _rcar_thermal_irq_ctrl(struct rcar_thermal_priv *priv, int enable)
+{
+	struct rcar_thermal_common *common = priv->common;
+	unsigned long flags;
+	u32 mask = 0x3 << rcar_id_to_shift(priv); /* enable Rising/Falling */
+
+	spin_lock_irqsave(&common->lock, flags);
+
+	rcar_thermal_common_bset(common, INTMSK, mask, enable ? 0 : mask);
+
+	spin_unlock_irqrestore(&common->lock, flags);
+}
+
+static void rcar_thermal_work(struct work_struct *work)
+{
+	struct rcar_thermal_priv *priv;
+
+	priv = container_of(work, struct rcar_thermal_priv, work.work);
+
+	rcar_thermal_update_temp(priv);
+	rcar_thermal_irq_enable(priv);
+	thermal_zone_device_update(priv->zone);
+}
+
+static u32 rcar_thermal_had_changed(struct rcar_thermal_priv *priv, u32 status)
+{
+	struct device *dev = rcar_priv_to_dev(priv);
+
+	status = (status >> rcar_id_to_shift(priv)) & 0x3;
+
+	if (status & 0x3) {
+		dev_dbg(dev, "thermal%d %s%s\n",
+			priv->id,
+			(status & 0x2) ? "Rising " : "",
+			(status & 0x1) ? "Falling" : "");
+	}
+
+	return status;
+}
+
+static irqreturn_t rcar_thermal_irq(int irq, void *data)
+{
+	struct rcar_thermal_common *common = data;
+	struct rcar_thermal_priv *priv;
+	unsigned long flags;
+	u32 status, mask;
+
+	spin_lock_irqsave(&common->lock, flags);
+
+	mask	= rcar_thermal_common_read(common, INTMSK);
+	status	= rcar_thermal_common_read(common, STR);
+	rcar_thermal_common_write(common, STR, 0x000F0F0F & mask);
+
+	spin_unlock_irqrestore(&common->lock, flags);
+
+	status = status & ~mask;
+
+	/*
+	 * check the status
+	 */
+	rcar_thermal_for_each_priv(priv, common) {
+		if (rcar_thermal_had_changed(priv, status)) {
+			rcar_thermal_irq_disable(priv);
+			schedule_delayed_work(&priv->work,
+					      msecs_to_jiffies(300));
+		}
+	}
+
+	return IRQ_HANDLED;
+}
 
 /*
  *		platform functions
  */
 static int rcar_thermal_probe(struct platform_device *pdev)
 {
-	struct thermal_zone_device *zone;
+	struct rcar_thermal_common *common;
 	struct rcar_thermal_priv *priv;
-	struct resource *res;
+	struct device *dev = &pdev->dev;
+	struct resource *res, *irq;
+	int mres = 0;
+	int i;
+	int idle = IDLE_INTERVAL;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		dev_err(&pdev->dev, "Could not get platform resource\n");
-		return -ENODEV;
-	}
-
-	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
-	if (!priv) {
-		dev_err(&pdev->dev, "Could not allocate priv\n");
+	common = devm_kzalloc(dev, sizeof(*common), GFP_KERNEL);
+	if (!common) {
+		dev_err(dev, "Could not allocate common\n");
 		return -ENOMEM;
 	}
 
-	priv->comp = 4; /* basic setup */
-	priv->dev = &pdev->dev;
-	spin_lock_init(&priv->lock);
-	priv->base = devm_ioremap_nocache(&pdev->dev,
-					  res->start, resource_size(res));
-	if (!priv->base) {
-		dev_err(&pdev->dev, "Unable to ioremap thermal register\n");
-		return -ENOMEM;
+	INIT_LIST_HEAD(&common->head);
+	spin_lock_init(&common->lock);
+	common->dev = dev;
+
+	irq = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (irq) {
+		int ret;
+
+		/*
+		 * platform has IRQ support.
+		 * Then, drier use common register
+		 */
+		res = platform_get_resource(pdev, IORESOURCE_MEM, mres++);
+		if (!res) {
+			dev_err(dev, "Could not get platform resource\n");
+			return -ENODEV;
+		}
+
+		ret = devm_request_irq(dev, irq->start, rcar_thermal_irq, 0,
+				       dev_name(dev), common);
+		if (ret) {
+			dev_err(dev, "irq request failed\n ");
+			return ret;
+		}
+
+		/*
+		 * rcar_has_irq_support() will be enabled
+		 */
+		common->base = devm_request_and_ioremap(dev, res);
+		if (!common->base) {
+			dev_err(dev, "Unable to ioremap thermal register\n");
+			return -ENOMEM;
+		}
+
+		/* enable temperature comparation */
+		rcar_thermal_common_write(common, ENR, 0x00030303);
+
+		idle = 0; /* polling delaye is not needed */
 	}
 
-	zone = thermal_zone_device_register("rcar_thermal", 0, 0, priv,
-				    &rcar_thermal_zone_ops, NULL, 0, 0);
-	if (IS_ERR(zone)) {
-		dev_err(&pdev->dev, "thermal zone device is NULL\n");
-		return PTR_ERR(zone);
+	for (i = 0;; i++) {
+		res = platform_get_resource(pdev, IORESOURCE_MEM, mres++);
+		if (!res)
+			break;
+
+		priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
+		if (!priv) {
+			dev_err(dev, "Could not allocate priv\n");
+			return -ENOMEM;
+		}
+
+		priv->base = devm_request_and_ioremap(dev, res);
+		if (!priv->base) {
+			dev_err(dev, "Unable to ioremap priv register\n");
+			return -ENOMEM;
+		}
+
+		priv->common = common;
+		priv->id = i;
+		mutex_init(&priv->lock);
+		INIT_LIST_HEAD(&priv->list);
+		INIT_DELAYED_WORK(&priv->work, rcar_thermal_work);
+		rcar_thermal_update_temp(priv);
+
+		priv->zone = thermal_zone_device_register("rcar_thermal",
+						1, 0, priv,
+						&rcar_thermal_zone_ops, NULL, 0,
+						idle);
+		if (IS_ERR(priv->zone)) {
+			dev_err(dev, "can't register thermal zone\n");
+			goto error_unregister;
+		}
+
+		list_move_tail(&priv->list, &common->head);
+
+		if (rcar_has_irq_support(priv))
+			rcar_thermal_irq_enable(priv);
 	}
 
-	platform_set_drvdata(pdev, zone);
+	platform_set_drvdata(pdev, common);
 
-	dev_info(&pdev->dev, "proved\n");
+	dev_info(dev, "%d sensor proved\n", i);
 
 	return 0;
+
+error_unregister:
+	rcar_thermal_for_each_priv(priv, common)
+		thermal_zone_device_unregister(priv->zone);
+
+	return -ENODEV;
 }
 
 static int rcar_thermal_remove(struct platform_device *pdev)
 {
-	struct thermal_zone_device *zone = platform_get_drvdata(pdev);
+	struct rcar_thermal_common *common = platform_get_drvdata(pdev);
+	struct rcar_thermal_priv *priv;
 
-	thermal_zone_device_unregister(zone);
+	rcar_thermal_for_each_priv(priv, common)
+		thermal_zone_device_unregister(priv->zone);
+
 	platform_set_drvdata(pdev, NULL);
 
 	return 0;
 }
 
+static const struct of_device_id rcar_thermal_dt_ids[] = {
+	{ .compatible = "renesas,rcar-thermal", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, rcar_thermal_dt_ids);
+
 static struct platform_driver rcar_thermal_driver = {
 	.driver	= {
 		.name	= "rcar_thermal",
+		.of_match_table = rcar_thermal_dt_ids,
 	},
 	.probe		= rcar_thermal_probe,
 	.remove		= rcar_thermal_remove,

drivers/thermal/spear_thermal.c

···
 		return -ENOMEM;
 	}
 
-	stdev->clk = clk_get(&pdev->dev, NULL);
+	stdev->clk = devm_clk_get(&pdev->dev, NULL);
 	if (IS_ERR(stdev->clk)) {
 		dev_err(&pdev->dev, "Can't get clock\n");
 		return PTR_ERR(stdev->clk);
···
 	ret = clk_enable(stdev->clk);
 	if (ret) {
 		dev_err(&pdev->dev, "Can't enable clock\n");
-		goto put_clk;
+		return ret;
 	}
 
 	stdev->flags = val;
···
 
 disable_clk:
 	clk_disable(stdev->clk);
-put_clk:
-	clk_put(stdev->clk);
 
 	return ret;
 }
···
 	writel_relaxed(actual_mask & ~stdev->flags, stdev->thermal_base);
 
 	clk_disable(stdev->clk);
-	clk_put(stdev->clk);
 
 	return 0;
 }

drivers/thermal/step_wise.c

···
  *       state for this trip point
  *    b. if the trend is THERMAL_TREND_DROPPING, use lower cooling
  *       state for this trip point
+ *    c. if the trend is THERMAL_TREND_RAISE_FULL, use upper limit
+ *       for this trip point
+ *    d. if the trend is THERMAL_TREND_DROP_FULL, use lower limit
+ *       for this trip point
+ * If the temperature is lower than a trip point,
+ *    a. if the trend is THERMAL_TREND_RAISING, do nothing
+ *    b. if the trend is THERMAL_TREND_DROPPING, use lower cooling
+ *       state for this trip point, if the cooling state already
+ *       equals lower limit, deactivate the thermal instance
+ *    c. if the trend is THERMAL_TREND_RAISE_FULL, do nothing
+ *    d. if the trend is THERMAL_TREND_DROP_FULL, use lower limit,
+ *       if the cooling state already equals lower limit,
+ *       deactive the thermal instance
  */
 static unsigned long get_target_state(struct thermal_instance *instance,
-					enum thermal_trend trend)
+				enum thermal_trend trend, bool throttle)
 {
 	struct thermal_cooling_device *cdev = instance->cdev;
 	unsigned long cur_state;
 
 	cdev->ops->get_cur_state(cdev, &cur_state);
 
-	if (trend == THERMAL_TREND_RAISING) {
-		cur_state = cur_state < instance->upper ?
-			    (cur_state + 1) : instance->upper;
-	} else if (trend == THERMAL_TREND_DROPPING) {
-		cur_state = cur_state > instance->lower ?
-			    (cur_state - 1) : instance->lower;
+	switch (trend) {
+	case THERMAL_TREND_RAISING:
+		if (throttle)
+			cur_state = cur_state < instance->upper ?
+				    (cur_state + 1) : instance->upper;
+		break;
+	case THERMAL_TREND_RAISE_FULL:
+		if (throttle)
+			cur_state = instance->upper;
+		break;
+	case THERMAL_TREND_DROPPING:
+		if (cur_state == instance->lower) {
+			if (!throttle)
+				cur_state = -1;
+		} else
+			cur_state -= 1;
+		break;
+	case THERMAL_TREND_DROP_FULL:
+		if (cur_state == instance->lower) {
+			if (!throttle)
+				cur_state = -1;
+		} else
+			cur_state = instance->lower;
+		break;
+	default:
+		break;
 	}
 
 	return cur_state;
···
 		tz->passive += value;
 }
 
-static void update_instance_for_throttle(struct thermal_zone_device *tz,
-				int trip, enum thermal_trip_type trip_type,
-				enum thermal_trend trend)
-{
-	struct thermal_instance *instance;
-
-	list_for_each_entry(instance, &tz->thermal_instances, tz_node) {
-		if (instance->trip != trip)
-			continue;
-
-		instance->target = get_target_state(instance, trend);
-
-		/* Activate a passive thermal instance */
-		if (instance->target == THERMAL_NO_TARGET)
-			update_passive_instance(tz, trip_type, 1);
-
-		instance->cdev->updated = false; /* cdev needs update */
-	}
-}
-
-static void update_instance_for_dethrottle(struct thermal_zone_device *tz,
-				int trip, enum thermal_trip_type trip_type)
-{
-	struct thermal_instance *instance;
-	struct thermal_cooling_device *cdev;
-	unsigned long cur_state;
-
-	list_for_each_entry(instance, &tz->thermal_instances, tz_node) {
-		if (instance->trip != trip ||
-			instance->target == THERMAL_NO_TARGET)
-			continue;
-
-		cdev = instance->cdev;
-		cdev->ops->get_cur_state(cdev, &cur_state);
-
-		instance->target = cur_state > instance->lower ?
-			    (cur_state - 1) : THERMAL_NO_TARGET;
-
-		/* Deactivate a passive thermal instance */
-		if (instance->target == THERMAL_NO_TARGET)
-			update_passive_instance(tz, trip_type, -1);
-
-		cdev->updated = false; /* cdev needs update */
-	}
-}
-
 static void thermal_zone_trip_update(struct thermal_zone_device *tz, int trip)
 {
 	long trip_temp;
 	enum thermal_trip_type trip_type;
 	enum thermal_trend trend;
+	struct thermal_instance *instance;
+	bool throttle = false;
+	int old_target;
 
 	if (trip == THERMAL_TRIPS_NONE) {
 		trip_temp = tz->forced_passive;
···
 
 	trend = get_tz_trend(tz, trip);
 
+	if (tz->temperature >= trip_temp)
+		throttle = true;
+
 	mutex_lock(&tz->lock);
 
-	if (tz->temperature >= trip_temp)
-		update_instance_for_throttle(tz, trip, trip_type, trend);
-	else
-		update_instance_for_dethrottle(tz, trip, trip_type);
+	list_for_each_entry(instance, &tz->thermal_instances, tz_node) {
+		if (instance->trip != trip)
+			continue;
+
+		old_target = instance->target;
+		instance->target = get_target_state(instance, trend, throttle);
+
+		/* Activate a passive thermal instance */
+		if (old_target == THERMAL_NO_TARGET &&
+			instance->target != THERMAL_NO_TARGET)
+			update_passive_instance(tz, trip_type, 1);
+		/* Deactivate a passive thermal instance */
+		else if (old_target != THERMAL_NO_TARGET &&
+			instance->target == THERMAL_NO_TARGET)
+			update_passive_instance(tz, trip_type, -1);
+
+
+		instance->cdev->updated = false; /* cdev needs update */
+	}
 
 	mutex_unlock(&tz->lock);
 }

drivers/thermal/thermal_sys.c

···
 #include <linux/kdev_t.h>
 #include <linux/idr.h>
 #include <linux/thermal.h>
-#include <linux/spinlock.h>
 #include <linux/reboot.h>
 #include <net/netlink.h>
 #include <net/genetlink.h>
···
 		tz->ops->notify(tz, trip, trip_type);
 
 	if (trip_type == THERMAL_TRIP_CRITICAL) {
-		pr_emerg("Critical temperature reached(%d C),shutting down\n",
-			 tz->temperature / 1000);
+		dev_emerg(&tz->device,
+			  "critical temperature reached(%d C),shutting down\n",
+			  tz->temperature / 1000);
 		orderly_poweroff(true);
 	}
 }
···
 	monitor_thermal_zone(tz);
 }
 
+static int thermal_zone_get_temp(struct thermal_zone_device *tz,
+				unsigned long *temp)
+{
+	int ret = 0;
+#ifdef CONFIG_THERMAL_EMULATION
+	int count;
+	unsigned long crit_temp = -1UL;
+	enum thermal_trip_type type;
+#endif
+
+	mutex_lock(&tz->lock);
+
+	ret = tz->ops->get_temp(tz, temp);
+#ifdef CONFIG_THERMAL_EMULATION
+	if (!tz->emul_temperature)
+		goto skip_emul;
+
+	for (count = 0; count < tz->trips; count++) {
+		ret = tz->ops->get_trip_type(tz, count, &type);
+		if (!ret && type == THERMAL_TRIP_CRITICAL) {
+			ret = tz->ops->get_trip_temp(tz, count, &crit_temp);
+			break;
+		}
+	}
+
+	if (ret)
+		goto skip_emul;
+
+	if (*temp < crit_temp)
+		*temp = tz->emul_temperature;
+skip_emul:
+#endif
+	mutex_unlock(&tz->lock);
+	return ret;
+}
+
 static void update_temperature(struct thermal_zone_device *tz)
 {
 	long temp;
 	int ret;
 
-	mutex_lock(&tz->lock);
-
-	ret = tz->ops->get_temp(tz, &temp);
+	ret = thermal_zone_get_temp(tz, &temp);
 	if (ret) {
-		pr_warn("failed to read out thermal zone %d\n", tz->id);
-		goto exit;
+		dev_warn(&tz->device, "failed to read out thermal zone %d\n",
+			 tz->id);
+		return;
 	}
 
+	mutex_lock(&tz->lock);
 	tz->last_temperature = tz->temperature;
 	tz->temperature = temp;
-
-exit:
 	mutex_unlock(&tz->lock);
 }
 
···
 	long temperature;
 	int ret;
 
-	if (!tz->ops->get_temp)
-		return -EPERM;
-
-	ret = tz->ops->get_temp(tz, &temperature);
+	ret = thermal_zone_get_temp(tz, &temperature);
 
 	if (ret)
 		return ret;
···
 	return sprintf(buf, "%s\n", tz->governor->name);
 }
 
+#ifdef CONFIG_THERMAL_EMULATION
+static ssize_t
+emul_temp_store(struct device *dev, struct device_attribute *attr,
+		     const char *buf, size_t count)
+{
+	struct thermal_zone_device *tz = to_thermal_zone(dev);
+	int ret = 0;
+	unsigned long temperature;
+
+	if (kstrtoul(buf, 10, &temperature))
+		return -EINVAL;
+
+	if (!tz->ops->set_emul_temp) {
+		mutex_lock(&tz->lock);
+		tz->emul_temperature = temperature;
+		mutex_unlock(&tz->lock);
+	} else {
+		ret = tz->ops->set_emul_temp(tz, temperature);
+	}
+
+	return ret ? ret : count;
+}
+static DEVICE_ATTR(emul_temp, S_IWUSR, NULL, emul_temp_store);
+#endif/*CONFIG_THERMAL_EMULATION*/
+
 static DEVICE_ATTR(type, 0444, type_show, NULL);
 static DEVICE_ATTR(temp, 0444, temp_show, NULL);
 static DEVICE_ATTR(mode, 0644, mode_show, mode_store);
···
 				       temp_input);
 	struct thermal_zone_device *tz = temp->tz;
 
-	ret = tz->ops->get_temp(tz, &temperature);
+	ret = thermal_zone_get_temp(tz, &temperature);
 
 	if (ret)
 		return ret;
···
 	if (!ops || !ops->get_temp)
 		return ERR_PTR(-EINVAL);
 
+	if (trips > 0 && !ops->get_trip_type)
+		return ERR_PTR(-EINVAL);
+
 	tz = kzalloc(sizeof(struct thermal_zone_device), GFP_KERNEL);
 	if (!tz)
 		return ERR_PTR(-ENOMEM);
···
 			goto unregister;
 	}
 
+#ifdef CONFIG_THERMAL_EMULATION
+	result = device_create_file(&tz->device, &dev_attr_emul_temp);
+	if (result)
+		goto unregister;
+#endif
 	/* Create policy attribute */
 	result = device_create_file(&tz->device, &dev_attr_policy);
 	if (result)
···
 	.name = THERMAL_GENL_MCAST_GROUP_NAME,
 };
 
-int thermal_generate_netlink_event(u32 orig, enum events event)
+int thermal_generate_netlink_event(struct thermal_zone_device *tz,
+					enum events event)
 {
 	struct sk_buff *skb;
 	struct nlattr *attr;
···
 	int size;
 	int result;
 	static unsigned int thermal_event_seqnum;
+
+	if (!tz)
+		return -EINVAL;
 
 	/* allocate memory */
 	size = nla_total_size(sizeof(struct thermal_genl_event)) +
···
 
 	memset(thermal_event, 0, sizeof(struct thermal_genl_event));
 
-	thermal_event->orig = orig;
+	thermal_event->orig = tz->id;
 	thermal_event->event = event;
 
 	/* send multicast genetlink message */
···
 
 	result = genlmsg_multicast(skb, 0, thermal_event_mcgrp.id, GFP_ATOMIC);
 	if (result)
-		pr_info("failed to send netlink event:%d\n", result);
+		dev_err(&tz->device, "Failed to send netlink event:%d", result);
 
 	return result;
 }
···
 		idr_destroy(&thermal_cdev_idr);
 		mutex_destroy(&thermal_idr_lock);
 		mutex_destroy(&thermal_list_lock);
+		return result;
 	}
 	result = genetlink_init();
 	return result;

include/linux/platform_data/exynos_thermal.h

···
  * struct exynos_tmu_platform_data
  * @threshold: basic temperature for generating interrupt
  *	       25 <= threshold <= 125 [unit: degree Celsius]
+ * @threshold_falling: differntial value for setting threshold
+ *		       of temperature falling interrupt.
  * @trigger_levels: array for each interrupt levels
  *	[unit: degree Celsius]
  *	0: temperature for trigger_level0 interrupt
···
  */
 struct exynos_tmu_platform_data {
 	u8 threshold;
+	u8 threshold_falling;
 	u8 trigger_levels[4];
 	bool trigger_level0_en;
 	bool trigger_level1_en;

include/linux/thermal.h

···
 	THERMAL_TREND_STABLE, /* temperature is stable */
 	THERMAL_TREND_RAISING, /* temperature is raising */
 	THERMAL_TREND_DROPPING, /* temperature is dropping */
+	THERMAL_TREND_RAISE_FULL, /* apply highest cooling action */
+	THERMAL_TREND_DROP_FULL, /* apply lowest cooling action */
 };
 
 /* Events supported by Thermal Netlink */
···
 	int (*set_trip_hyst) (struct thermal_zone_device *, int,
 			      unsigned long);
 	int (*get_crit_temp) (struct thermal_zone_device *, unsigned long *);
+	int (*set_emul_temp) (struct thermal_zone_device *, unsigned long);
 	int (*get_trend) (struct thermal_zone_device *, int,
 			  enum thermal_trend *);
 	int (*notify) (struct thermal_zone_device *, int,
···
 	int polling_delay;
 	int temperature;
 	int last_temperature;
+	int emul_temperature;
 	int passive;
 	unsigned int forced_passive;
 	const struct thermal_zone_device_ops *ops;
···
 void thermal_unregister_governor(struct thermal_governor *);
 
 #ifdef CONFIG_NET
-extern int thermal_generate_netlink_event(u32 orig, enum events event);
+extern int thermal_generate_netlink_event(struct thermal_zone_device *tz,
+						enum events event);
 #else
-static inline int thermal_generate_netlink_event(u32 orig, enum events event)
+static int thermal_generate_netlink_event(struct thermal_zone_device *tz,
+						enum events event)
 {
 	return 0;
 }

kernel/time/tick-sched.c

···
 
 	local_irq_enable();
 }
+EXPORT_SYMBOL_GPL(tick_nohz_idle_enter);
 
 /**
  * tick_nohz_irq_exit - update next tick event from interrupt exit
···
 
 	local_irq_enable();
 }
+EXPORT_SYMBOL_GPL(tick_nohz_idle_exit);
 
 static int tick_nohz_reprogram(struct tick_sched *ts, ktime_t now)
 {