Power management updates for v4.12-rc1

- Rework the intel_pstate driver's sysfs interface to make it
    more straightforward and more intuitive (Rafael Wysocki).
 
  - Make intel_pstate support all processors which advertise HWP
    (hardware-managed P-states) to the kernel in all operation modes
    and make it use the load-based P-state selection algorithm on a
    wider range of systems in the active mode (Rafael Wysocki).
 
  - Add cpufreq driver for Tegra186 (Mikko Perttunen).
 
  - Add support for Gemini Lake SoCs to intel_pstate (David Box).
 
  - Add support for MT8176 and MT817x to the Mediatek cpufreq driver
    and clean up that driver a bit (Daniel Kurtz).
 
  - Clean up intel_pstate and optimize it slightly (Rafael Wysocki).
 
  - Update the schedutil cpufreq governor, mostly to fix a couple of
    issues with it related to specific workloads, and rework its sysfs
    tunable and initialization a bit (Rafael Wysocki, Viresh Kumar).
 
  - Fix minor issues in the imx6q, dbx500 and qoriq cpufreq drivers
    (Christophe Jaillet, Irina Tirdea, Leonard Crestez, Viresh Kumar,
    YuanTian Tang).
 
  - Add file patterns for cpufreq DT bindings to MAINTAINERS (Geert
    Uytterhoeven).
 
  - Add support for "always on" power domains to the genpd (generic
    power domains) framework and clean up that code somewhat (Ulf
    Hansson, Lina Iyer, Viresh Kumar).
 
  - Fix minor issues in the powernv cpuidle driver and clean it up
    (Anton Blanchard, Gautham Shenoy).
 
  - Move the AnalyzeSuspend utility under tools/power/pm-graph/ and
    add an analogous boot-profiling utility called AnalyzeBoot to it
    (Todd Brandt).
 
  - Add rk3328 support to the rockchip-io AVS (Adaptive Voltage
    Scaling) driver (David Wu).
 
  - Fix minor issues in the cpuidle core, the intel_pstate_tracer
    utility, the devfreq framework and the PM core documentation
    (Chanwoo Choi, Doug Smythies, Johan Hovold, Marcin Nowakowski).
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2
 
 iQIcBAABCAAGBQJZB5gGAAoJEILEb/54YlRx+78QAJRu+xAA9KtW2+loNyV8iBOB
 EFmQLrvz9jCDyYWsHE5huA1k6EVu5QE74HBfgDn4od9s1VqU1zWdEjqKYiaMwlCt
 EHxYCZ4YKeF31O3P3CtearBz9IXrckRx/XZ3F1jRsGGWooWv7o3U6PPN9iREmCzi
 9dB2j0UD4lCwrnpsDMrJ0GqLu4agn9pcIKDtu4VfszVwYtza0vOQvvlgg1fQS1jX
 BnNfaxN0lmpSlxDjtWfM//hfLzEWK8NlHiKWJFPnWFxJIAX1QL2QKnznF/Tqi5N5
 el9tQXCTRujlD7BLyl6FdsaowbiUUEcMqeh2k01Vz20WSmNDHIpfrV9MtzJ7biUK
 /DopyShUrpgJclKNF7BARJAJc19+PMkv3HMnOnsnhOsBNBpJjsL6FZPA95MMjI0o
 xmRQkixl31NWIMk60EIC6PaMLsxjhAWjiYi5D93bzkhnJTnQswvNb4ROPG+X2FCU
 6YgBogsYKkqp93TTJ49OFXIvu3o7NwOyEQQW8mnNY8ffaFdWuGzOX4HkOoKHCMTD
 rT0qT/2q+7LPK87YwTPIVtpGVltnCr/SVI/FtAGXPysyghu2Z+4GGP4eh2+pSAUj
 7Dqxdw3q7zs8ou6LOThTyNJrR+N/w1JPloprBleJR3TNcJjOy/SBjfp7yL0uopIK
 j5exr76ImVq7zzjyrYoa
 =6M3e
 -----END PGP SIGNATURE-----

Merge tag 'pm-4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

Pull power management updates from Rafael Wysocki:
 "This time the majority of changes go to the cpufreq subsystem (and to
  the intel_pstate driver in particular) and there are some updates in
  the generic power domains framework, cpuidle, tools and a couple of
  other places.

  One thing worth mentioning is that the intel_pstate's sysfs interface
  has been reworked to be more consistent with the general expectations
  of the cpufreq core and less confusing, hopefully for the better.
  Also, we have a new cpufreq driver for Tegra186 and new hardware
  support in intel_pstata and the Mediatek cpufreq driver.

  Apart from that, the AnalyzeSuspend utility for system suspend
  profiling gets a companion called AnalyzeBoot for the analogous
  profiling of system boot and they both go into one place under
  tools/power/pm-graph/.

  The rest is mostly fixes, cleanups and code reorganization.

  Specifics:

   - Rework the intel_pstate driver's sysfs interface to make it more
     straightforward and more intuitive (Rafael Wysocki).

   - Make intel_pstate support all processors which advertise HWP
     (hardware-managed P-states) to the kernel in all operation modes
     and make it use the load-based P-state selection algorithm on a
     wider range of systems in the active mode (Rafael Wysocki).

   - Add cpufreq driver for Tegra186 (Mikko Perttunen).

   - Add support for Gemini Lake SoCs to intel_pstate (David Box).

   - Add support for MT8176 and MT817x to the Mediatek cpufreq driver
     and clean up that driver a bit (Daniel Kurtz).

   - Clean up intel_pstate and optimize it slightly (Rafael Wysocki).

   - Update the schedutil cpufreq governor, mostly to fix a couple of
     issues with it related to specific workloads, and rework its sysfs
     tunable and initialization a bit (Rafael Wysocki, Viresh Kumar).

   - Fix minor issues in the imx6q, dbx500 and qoriq cpufreq drivers
     (Christophe Jaillet, Irina Tirdea, Leonard Crestez, Viresh Kumar,
     YuanTian Tang).

   - Add file patterns for cpufreq DT bindings to MAINTAINERS (Geert
     Uytterhoeven).

   - Add support for "always on" power domains to the genpd (generic
     power domains) framework and clean up that code somewhat (Ulf
     Hansson, Lina Iyer, Viresh Kumar).

   - Fix minor issues in the powernv cpuidle driver and clean it up
     (Anton Blanchard, Gautham Shenoy).

   - Move the AnalyzeSuspend utility under tools/power/pm-graph/ and add
     an analogous boot-profiling utility called AnalyzeBoot to it (Todd
     Brandt).

   - Add rk3328 support to the rockchip-io AVS (Adaptive Voltage
     Scaling) driver (David Wu).

   - Fix minor issues in the cpuidle core, the intel_pstate_tracer
     utility, the devfreq framework and the PM core documentation
     (Chanwoo Choi, Doug Smythies, Johan Hovold, Marcin Nowakowski)"

* tag 'pm-4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (56 commits)
  PM / runtime: Document autosuspend-helper side effects
  PM / runtime: Fix autosuspend documentation
  tools: power: pm-graph: Package makefile and man pages
  tools: power: pm-graph: AnalyzeBoot v2.0
  tools: power: pm-graph: AnalyzeSuspend v4.6
  cpufreq: Add Tegra186 cpufreq driver
  cpufreq: imx6q: Fix error handling code
  cpufreq: imx6q: Set max suspend_freq to avoid changes during suspend
  cpufreq: imx6q: Fix handling EPROBE_DEFER from regulator
  cpuidle: powernv: Avoid a branch in the core snooze_loop() loop
  cpuidle: powernv: Don't continually set thread priority in snooze_loop()
  cpuidle: powernv: Don't bounce between low and very low thread priority
  cpuidle: cpuidle-cps: remove unused variable
  tools/power/x86/intel_pstate_tracer: Adjust directory ownership
  cpufreq: schedutil: Use policy-dependent transition delays
  cpufreq: schedutil: Reduce frequencies slower
  PM / devfreq: Move struct devfreq_governor to devfreq directory
  PM / Domains: Ignore domain-idle-states that are not compatible
  cpufreq: intel_pstate: Add support for Gemini Lake
  powernv-cpuidle: Validate DT property array size
  ...

Documentation/devicetree/bindings/power/power_domain.txt

@@ -31,7 +31,9 @@ Optional properties:
 
 - domain-idle-states : A phandle of an idle-state that shall be soaked into a
                 generic domain power state. The idle state definitions are
-                compatible with domain-idle-state specified in [1].
+                compatible with domain-idle-state specified in [1]. phandles
+                that are not compatible with domain-idle-state will be
+                ignored.
   The domain-idle-state property reflects the idle state of this PM domain and
   not the idle states of the devices or sub-domains in the PM domain. Devices
   and sub-domains have their own idle-states independent of the parent

Documentation/devicetree/bindings/power/rockchip-io-domain.txt

@@ -33,6 +33,7 @@ Required properties:
 - compatible: should be one of:
   - "rockchip,rk3188-io-voltage-domain" for rk3188
   - "rockchip,rk3288-io-voltage-domain" for rk3288
+  - "rockchip,rk3328-io-voltage-domain" for rk3328
   - "rockchip,rk3368-io-voltage-domain" for rk3368
   - "rockchip,rk3368-pmu-io-voltage-domain" for rk3368 pmu-domains
   - "rockchip,rk3399-io-voltage-domain" for rk3399

Documentation/power/runtime_pm.txt

@@ -478,15 +478,23 @@ drivers/base/power/runtime.c and include/linux/pm_runtime.h:
     - set the power.last_busy field to the current time
 
   void pm_runtime_use_autosuspend(struct device *dev);
-    - set the power.use_autosuspend flag, enabling autosuspend delays
+    - set the power.use_autosuspend flag, enabling autosuspend delays; call
+      pm_runtime_get_sync if the flag was previously cleared and
+      power.autosuspend_delay is negative
 
   void pm_runtime_dont_use_autosuspend(struct device *dev);
-    - clear the power.use_autosuspend flag, disabling autosuspend delays
+    - clear the power.use_autosuspend flag, disabling autosuspend delays;
+      decrement the device's usage counter if the flag was previously set and
+      power.autosuspend_delay is negative; call pm_runtime_idle
 
   void pm_runtime_set_autosuspend_delay(struct device *dev, int delay);
     - set the power.autosuspend_delay value to 'delay' (expressed in
       milliseconds); if 'delay' is negative then runtime suspends are
-      prevented
+      prevented; if power.use_autosuspend is set, pm_runtime_get_sync may be
+      called or the device's usage counter may be decremented and
+      pm_runtime_idle called depending on if power.autosuspend_delay is
+      changed to or from a negative value; if power.use_autosuspend is clear,
+      pm_runtime_idle is called
 
   unsigned long pm_runtime_autosuspend_expiration(struct device *dev);
     - calculate the time when the current autosuspend delay period will expire,
@@ -836,9 +844,8 @@ of the non-autosuspend counterparts:
 	Instead of: pm_runtime_put_sync   use: pm_runtime_put_sync_autosuspend.
 
 Drivers may also continue to use the non-autosuspend helper functions; they
-will behave normally, not taking the autosuspend delay into account.
-Similarly, if the power.use_autosuspend field isn't set then the autosuspend
-helper functions will behave just like the non-autosuspend counterparts.
+will behave normally, which means sometimes taking the autosuspend delay into
+account (see pm_runtime_idle).
 
 Under some circumstances a driver or subsystem may want to prevent a device
 from autosuspending immediately, even though the usage counter is zero and the

MAINTAINERS

@@ -3471,6 +3471,7 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git
 T:	git git://git.linaro.org/people/vireshk/linux.git (For ARM Updates)
 B:	https://bugzilla.kernel.org
 F:	Documentation/cpu-freq/
+F:	Documentation/devicetree/bindings/cpufreq/
 F:	drivers/cpufreq/
 F:	include/linux/cpufreq.h
 F:	tools/testing/selftests/cpufreq/

arch/arm/boot/dts/ste-dbx5x0.dtsi

@@ -1189,11 +1189,6 @@
 			status = "disabled";
 		};
 
-		cpufreq-cooling {
-			compatible = "stericsson,db8500-cpufreq-cooling";
-			status = "disabled";
-		};
-
 		mcde@a0350000 {
 			compatible = "stericsson,mcde";
 			reg = <0xa0350000 0x1000>, /* MCDE */

drivers/base/power/domain.c

@@ -121,7 +121,9 @@ static const struct genpd_lock_ops genpd_spin_ops = {
 #define genpd_lock_interruptible(p)	p->lock_ops->lock_interruptible(p)
 #define genpd_unlock(p)			p->lock_ops->unlock(p)
 
+#define genpd_status_on(genpd)		(genpd->status == GPD_STATE_ACTIVE)
 #define genpd_is_irq_safe(genpd)	(genpd->flags & GENPD_FLAG_IRQ_SAFE)
+#define genpd_is_always_on(genpd)	(genpd->flags & GENPD_FLAG_ALWAYS_ON)
 
 static inline bool irq_safe_dev_in_no_sleep_domain(struct device *dev,
 		struct generic_pm_domain *genpd)
@@ -130,8 +132,12 @@ static inline bool irq_safe_dev_in_no_sleep_domain(struct device *dev,
 
 	ret = pm_runtime_is_irq_safe(dev) && !genpd_is_irq_safe(genpd);
 
-	/* Warn once if IRQ safe dev in no sleep domain */
-	if (ret)
+	/*
+	 * Warn once if an IRQ safe device is attached to a no sleep domain, as
+	 * to indicate a suboptimal configuration for PM. For an always on
+	 * domain this isn't case, thus don't warn.
+	 */
+	if (ret && !genpd_is_always_on(genpd))
 		dev_warn_once(dev, "PM domain %s will not be powered off\n",
 				genpd->name);
 
@@ -296,11 +302,15 @@ static int genpd_power_off(struct generic_pm_domain *genpd, bool one_dev_on,
 	 * (1) The domain is already in the "power off" state.
 	 * (2) System suspend is in progress.
 	 */
-	if (genpd->status == GPD_STATE_POWER_OFF
-	    || genpd->prepared_count > 0)
+	if (!genpd_status_on(genpd) || genpd->prepared_count > 0)
 		return 0;
 
-	if (atomic_read(&genpd->sd_count) > 0)
+	/*
+	 * Abort power off for the PM domain in the following situations:
+	 * (1) The domain is configured as always on.
+	 * (2) When the domain has a subdomain being powered on.
+	 */
+	if (genpd_is_always_on(genpd) || atomic_read(&genpd->sd_count) > 0)
 		return -EBUSY;
 
 	list_for_each_entry(pdd, &genpd->dev_list, list_node) {
@@ -373,7 +383,7 @@ static int genpd_power_on(struct generic_pm_domain *genpd, unsigned int depth)
 	struct gpd_link *link;
 	int ret = 0;
 
-	if (genpd->status == GPD_STATE_ACTIVE)
+	if (genpd_status_on(genpd))
 		return 0;
 
 	/*
@@ -752,7 +762,7 @@ static void genpd_sync_power_off(struct generic_pm_domain *genpd, bool use_lock,
 {
 	struct gpd_link *link;
 
-	if (genpd->status == GPD_STATE_POWER_OFF)
+	if (!genpd_status_on(genpd) || genpd_is_always_on(genpd))
 		return;
 
 	if (genpd->suspended_count != genpd->device_count
@@ -761,7 +771,8 @@ static void genpd_sync_power_off(struct generic_pm_domain *genpd, bool use_lock,
 
 	/* Choose the deepest state when suspending */
 	genpd->state_idx = genpd->state_count - 1;
-	_genpd_power_off(genpd, false);
+	if (_genpd_power_off(genpd, false))
+		return;
 
 	genpd->status = GPD_STATE_POWER_OFF;
 
@@ -793,7 +804,7 @@ static void genpd_sync_power_on(struct generic_pm_domain *genpd, bool use_lock,
 {
 	struct gpd_link *link;
 
-	if (genpd->status == GPD_STATE_ACTIVE)
+	if (genpd_status_on(genpd))
 		return;
 
 	list_for_each_entry(link, &genpd->slave_links, slave_node) {
@@ -1329,8 +1340,7 @@ static int genpd_add_subdomain(struct generic_pm_domain *genpd,
 	genpd_lock(subdomain);
 	genpd_lock_nested(genpd, SINGLE_DEPTH_NESTING);
 
-	if (genpd->status == GPD_STATE_POWER_OFF
-	    &&  subdomain->status != GPD_STATE_POWER_OFF) {
+	if (!genpd_status_on(genpd) && genpd_status_on(subdomain)) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -1346,7 +1356,7 @@ static int genpd_add_subdomain(struct generic_pm_domain *genpd,
 	list_add_tail(&link->master_node, &genpd->master_links);
 	link->slave = subdomain;
 	list_add_tail(&link->slave_node, &subdomain->slave_links);
-	if (subdomain->status != GPD_STATE_POWER_OFF)
+	if (genpd_status_on(subdomain))
 		genpd_sd_counter_inc(genpd);
 
  out:
@@ -1406,7 +1416,7 @@ int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd,
 		list_del(&link->master_node);
 		list_del(&link->slave_node);
 		kfree(link);
-		if (subdomain->status != GPD_STATE_POWER_OFF)
+		if (genpd_status_on(subdomain))
 			genpd_sd_counter_dec(genpd);
 
 		ret = 0;
@@ -1492,6 +1502,10 @@ int pm_genpd_init(struct generic_pm_domain *genpd,
 		genpd->dev_ops.start = pm_clk_resume;
 	}
 
+	/* Always-on domains must be powered on at initialization. */
+	if (genpd_is_always_on(genpd) && !genpd_status_on(genpd))
+		return -EINVAL;
+
 	/* Use only one "off" state if there were no states declared */
 	if (genpd->state_count == 0) {
 		ret = genpd_set_default_power_state(genpd);
@@ -1700,12 +1714,12 @@ int of_genpd_add_provider_simple(struct device_node *np,
 
 	mutex_lock(&gpd_list_lock);
 
-	if (pm_genpd_present(genpd))
+	if (pm_genpd_present(genpd)) {
 		ret = genpd_add_provider(np, genpd_xlate_simple, genpd);
-
-	if (!ret) {
-		genpd->provider = &np->fwnode;
-		genpd->has_provider = true;
+		if (!ret) {
+			genpd->provider = &np->fwnode;
+			genpd->has_provider = true;
+		}
 	}
 
 	mutex_unlock(&gpd_list_lock);
@@ -2079,11 +2093,6 @@ static int genpd_parse_state(struct genpd_power_state *genpd_state,
 	int err;
 	u32 residency;
 	u32 entry_latency, exit_latency;
-	const struct of_device_id *match_id;
-
-	match_id = of_match_node(idle_state_match, state_node);
-	if (!match_id)
-		return -EINVAL;
 
 	err = of_property_read_u32(state_node, "entry-latency-us",
 						&entry_latency);
@@ -2132,6 +2141,7 @@ int of_genpd_parse_idle_states(struct device_node *dn,
 	int err, ret;
 	int count;
 	struct of_phandle_iterator it;
+	const struct of_device_id *match_id;
 
 	count = of_count_phandle_with_args(dn, "domain-idle-states", NULL);
 	if (count <= 0)
@@ -2144,6 +2154,9 @@ int of_genpd_parse_idle_states(struct device_node *dn,
 	/* Loop over the phandles until all the requested entry is found */
 	of_for_each_phandle(&it, err, dn, "domain-idle-states", NULL, 0) {
 		np = it.node;
+		match_id = of_match_node(idle_state_match, np);
+		if (!match_id)
+			continue;
 		ret = genpd_parse_state(&st[i++], np);
 		if (ret) {
 			pr_err
@@ -2155,8 +2168,11 @@ int of_genpd_parse_idle_states(struct device_node *dn,
 		}
 	}
 
-	*n = count;
-	*states = st;
+	*n = i;
+	if (!i)
+		kfree(st);
+	else
+		*states = st;
 
 	return 0;
 }
@@ -2221,7 +2237,7 @@ static int pm_genpd_summary_one(struct seq_file *s,
 
 	if (WARN_ON(genpd->status >= ARRAY_SIZE(status_lookup)))
 		goto exit;
-	if (genpd->status == GPD_STATE_POWER_OFF)
+	if (!genpd_status_on(genpd))
 		snprintf(state, sizeof(state), "%s-%u",
 			 status_lookup[genpd->status], genpd->state_idx);
 	else

drivers/cpufreq/Kconfig.arm

@@ -247,6 +247,12 @@ config ARM_TEGRA124_CPUFREQ
 	help
 	  This adds the CPUFreq driver support for Tegra124 SOCs.
 
+config ARM_TEGRA186_CPUFREQ
+	tristate "Tegra186 CPUFreq support"
+	depends on ARCH_TEGRA && TEGRA_BPMP
+	help
+	  This adds the CPUFreq driver support for Tegra186 SOCs.
+
 config ARM_TI_CPUFREQ
 	bool "Texas Instruments CPUFreq support"
 	depends on ARCH_OMAP2PLUS

drivers/cpufreq/Makefile

@@ -77,6 +77,7 @@ obj-$(CONFIG_ARM_SPEAR_CPUFREQ)		+= spear-cpufreq.o
 obj-$(CONFIG_ARM_STI_CPUFREQ)		+= sti-cpufreq.o
 obj-$(CONFIG_ARM_TEGRA20_CPUFREQ)	+= tegra20-cpufreq.o
 obj-$(CONFIG_ARM_TEGRA124_CPUFREQ)	+= tegra124-cpufreq.o
+obj-$(CONFIG_ARM_TEGRA186_CPUFREQ)	+= tegra186-cpufreq.o
 obj-$(CONFIG_ARM_TI_CPUFREQ)		+= ti-cpufreq.o
 obj-$(CONFIG_ARM_VEXPRESS_SPC_CPUFREQ)	+= vexpress-spc-cpufreq.o
 obj-$(CONFIG_ACPI_CPPC_CPUFREQ) += cppc_cpufreq.o

drivers/cpufreq/dbx500-cpufreq.c

@@ -11,6 +11,7 @@
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/cpufreq.h>
+#include <linux/cpu_cooling.h>
 #include <linux/delay.h>
 #include <linux/slab.h>
 #include <linux/platform_device.h>
@@ -18,6 +19,7 @@
 
 static struct cpufreq_frequency_table *freq_table;
 static struct clk *armss_clk;
+static struct thermal_cooling_device *cdev;
 
 static int dbx500_cpufreq_target(struct cpufreq_policy *policy,
 				unsigned int index)
@@ -32,6 +34,22 @@ static int dbx500_cpufreq_init(struct cpufreq_policy *policy)
 	return cpufreq_generic_init(policy, freq_table, 20 * 1000);
 }
 
+static int dbx500_cpufreq_exit(struct cpufreq_policy *policy)
+{
+	if (!IS_ERR(cdev))
+		cpufreq_cooling_unregister(cdev);
+	return 0;
+}
+
+static void dbx500_cpufreq_ready(struct cpufreq_policy *policy)
+{
+	cdev = cpufreq_cooling_register(policy->cpus);
+	if (IS_ERR(cdev))
+		pr_err("Failed to register cooling device %ld\n", PTR_ERR(cdev));
+	else
+		pr_info("Cooling device registered: %s\n", cdev->type);
+}
+
 static struct cpufreq_driver dbx500_cpufreq_driver = {
 	.flags  = CPUFREQ_STICKY | CPUFREQ_CONST_LOOPS |
 			CPUFREQ_NEED_INITIAL_FREQ_CHECK,
@@ -39,6 +57,8 @@ static struct cpufreq_driver dbx500_cpufreq_driver = {
 	.target_index = dbx500_cpufreq_target,
 	.get    = cpufreq_generic_get,
 	.init   = dbx500_cpufreq_init,
+	.exit  = dbx500_cpufreq_exit,
+	.ready  = dbx500_cpufreq_ready,
 	.name   = "DBX500",
 	.attr   = cpufreq_generic_attr,
 };

drivers/cpufreq/imx6q-cpufreq.c

@@ -161,8 +161,13 @@ static int imx6q_set_target(struct cpufreq_policy *policy, unsigned int index)
 
 static int imx6q_cpufreq_init(struct cpufreq_policy *policy)
 {
+	int ret;
+
 	policy->clk = arm_clk;
-	return cpufreq_generic_init(policy, freq_table, transition_latency);
+	ret = cpufreq_generic_init(policy, freq_table, transition_latency);
+	policy->suspend_freq = policy->max;
+
+	return ret;
 }
 
 static struct cpufreq_driver imx6q_cpufreq_driver = {
@@ -173,6 +178,7 @@ static struct cpufreq_driver imx6q_cpufreq_driver = {
 	.init = imx6q_cpufreq_init,
 	.name = "imx6q-cpufreq",
 	.attr = cpufreq_generic_attr,
+	.suspend = cpufreq_generic_suspend,
 };
 
 static int imx6q_cpufreq_probe(struct platform_device *pdev)
@@ -222,6 +228,13 @@ static int imx6q_cpufreq_probe(struct platform_device *pdev)
 	arm_reg = regulator_get(cpu_dev, "arm");
 	pu_reg = regulator_get_optional(cpu_dev, "pu");
 	soc_reg = regulator_get(cpu_dev, "soc");
+	if (PTR_ERR(arm_reg) == -EPROBE_DEFER ||
+			PTR_ERR(soc_reg) == -EPROBE_DEFER ||
+			PTR_ERR(pu_reg) == -EPROBE_DEFER) {
+		ret = -EPROBE_DEFER;
+		dev_dbg(cpu_dev, "regulators not ready, defer\n");
+		goto put_reg;
+	}
 	if (IS_ERR(arm_reg) || IS_ERR(soc_reg)) {
 		dev_err(cpu_dev, "failed to get regulators\n");
 		ret = -ENOENT;
@@ -255,7 +268,7 @@ static int imx6q_cpufreq_probe(struct platform_device *pdev)
 	ret = dev_pm_opp_init_cpufreq_table(cpu_dev, &freq_table);
 	if (ret) {
 		dev_err(cpu_dev, "failed to init cpufreq table: %d\n", ret);
-		goto put_reg;
+		goto out_free_opp;
 	}
 
 	/* Make imx6_soc_volt array's size same as arm opp number */

drivers/cpufreq/mt8173-cpufreq.c

@@ -573,14 +573,33 @@ static struct platform_driver mt8173_cpufreq_platdrv = {
 	.probe		= mt8173_cpufreq_probe,
 };
 
-static int mt8173_cpufreq_driver_init(void)
+/* List of machines supported by this driver */
+static const struct of_device_id mt8173_cpufreq_machines[] __initconst = {
+	{ .compatible = "mediatek,mt817x", },
+	{ .compatible = "mediatek,mt8173", },
+	{ .compatible = "mediatek,mt8176", },
+
+	{ }
+};
+
+static int __init mt8173_cpufreq_driver_init(void)
 {
+	struct device_node *np;
+	const struct of_device_id *match;
 	struct platform_device *pdev;
 	int err;
 
-	if (!of_machine_is_compatible("mediatek,mt8173"))
+	np = of_find_node_by_path("/");
+	if (!np)
 		return -ENODEV;
 
+	match = of_match_node(mt8173_cpufreq_machines, np);
+	of_node_put(np);
+	if (!match) {
+		pr_warn("Machine is not compatible with mt8173-cpufreq\n");
+		return -ENODEV;
+	}
+
 	err = platform_driver_register(&mt8173_cpufreq_platdrv);
 	if (err)
 		return err;

drivers/cpufreq/qoriq-cpufreq.c

@@ -52,17 +52,27 @@ static u32 get_bus_freq(void)
 {
 	struct device_node *soc;
 	u32 sysfreq;
+	struct clk *pltclk;
+	int ret;
 
+	/* get platform freq by searching bus-frequency property */
 	soc = of_find_node_by_type(NULL, "soc");
-	if (!soc)
-		return 0;
+	if (soc) {
+		ret = of_property_read_u32(soc, "bus-frequency", &sysfreq);
+		of_node_put(soc);
+		if (!ret)
+			return sysfreq;
+	}
 
-	if (of_property_read_u32(soc, "bus-frequency", &sysfreq))
-		sysfreq = 0;
+	/* get platform freq by its clock name */
+	pltclk = clk_get(NULL, "cg-pll0-div1");
+	if (IS_ERR(pltclk)) {
+		pr_err("%s: can't get bus frequency %ld\n",
+		       __func__, PTR_ERR(pltclk));
+		return PTR_ERR(pltclk);
+	}
 
-	of_node_put(soc);
-
-	return sysfreq;
+	return clk_get_rate(pltclk);
 }
 
 static struct clk *cpu_to_clk(int cpu)

drivers/cpufreq/tegra186-cpufreq.c

@@ -0,0 +1,275 @@
+/*
+ * Copyright (c) 2017, NVIDIA CORPORATION. All rights reserved
+ *
+ * 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.
+ */
+
+#include <linux/cpufreq.h>
+#include <linux/dma-mapping.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+
+#include <soc/tegra/bpmp.h>
+#include <soc/tegra/bpmp-abi.h>
+
+#define EDVD_CORE_VOLT_FREQ(core)		(0x20 + (core) * 0x4)
+#define EDVD_CORE_VOLT_FREQ_F_SHIFT		0
+#define EDVD_CORE_VOLT_FREQ_V_SHIFT		16
+
+struct tegra186_cpufreq_cluster_info {
+	unsigned long offset;
+	int cpus[4];
+	unsigned int bpmp_cluster_id;
+};
+
+#define NO_CPU -1
+static const struct tegra186_cpufreq_cluster_info tegra186_clusters[] = {
+	/* Denver cluster */
+	{
+		.offset = SZ_64K * 7,
+		.cpus = { 1, 2, NO_CPU, NO_CPU },
+		.bpmp_cluster_id = 0,
+	},
+	/* A57 cluster */
+	{
+		.offset = SZ_64K * 6,
+		.cpus = { 0, 3, 4, 5 },
+		.bpmp_cluster_id = 1,
+	},
+};
+
+struct tegra186_cpufreq_cluster {
+	const struct tegra186_cpufreq_cluster_info *info;
+	struct cpufreq_frequency_table *table;
+};
+
+struct tegra186_cpufreq_data {
+	void __iomem *regs;
+
+	size_t num_clusters;
+	struct tegra186_cpufreq_cluster *clusters;
+};
+
+static int tegra186_cpufreq_init(struct cpufreq_policy *policy)
+{
+	struct tegra186_cpufreq_data *data = cpufreq_get_driver_data();
+	unsigned int i;
+
+	for (i = 0; i < data->num_clusters; i++) {
+		struct tegra186_cpufreq_cluster *cluster = &data->clusters[i];
+		const struct tegra186_cpufreq_cluster_info *info =
+			cluster->info;
+		int core;
+
+		for (core = 0; core < ARRAY_SIZE(info->cpus); core++) {
+			if (info->cpus[core] == policy->cpu)
+				break;
+		}
+		if (core == ARRAY_SIZE(info->cpus))
+			continue;
+
+		policy->driver_data =
+			data->regs + info->offset + EDVD_CORE_VOLT_FREQ(core);
+		cpufreq_table_validate_and_show(policy, cluster->table);
+	}
+
+	policy->cpuinfo.transition_latency = 300 * 1000;
+
+	return 0;
+}
+
+static int tegra186_cpufreq_set_target(struct cpufreq_policy *policy,
+				       unsigned int index)
+{
+	struct cpufreq_frequency_table *tbl = policy->freq_table + index;
+	void __iomem *edvd_reg = policy->driver_data;
+	u32 edvd_val = tbl->driver_data;
+
+	writel(edvd_val, edvd_reg);
+
+	return 0;
+}
+
+static struct cpufreq_driver tegra186_cpufreq_driver = {
+	.name = "tegra186",
+	.flags = CPUFREQ_STICKY | CPUFREQ_HAVE_GOVERNOR_PER_POLICY,
+	.verify = cpufreq_generic_frequency_table_verify,
+	.target_index = tegra186_cpufreq_set_target,
+	.init = tegra186_cpufreq_init,
+	.attr = cpufreq_generic_attr,
+};
+
+static struct cpufreq_frequency_table *init_vhint_table(
+	struct platform_device *pdev, struct tegra_bpmp *bpmp,
+	unsigned int cluster_id)
+{
+	struct cpufreq_frequency_table *table;
+	struct mrq_cpu_vhint_request req;
+	struct tegra_bpmp_message msg;
+	struct cpu_vhint_data *data;
+	int err, i, j, num_rates = 0;
+	dma_addr_t phys;
+	void *virt;
+
+	virt = dma_alloc_coherent(bpmp->dev, sizeof(*data), &phys,
+				  GFP_KERNEL | GFP_DMA32);
+	if (!virt)
+		return ERR_PTR(-ENOMEM);
+
+	data = (struct cpu_vhint_data *)virt;
+
+	memset(&req, 0, sizeof(req));
+	req.addr = phys;
+	req.cluster_id = cluster_id;
+
+	memset(&msg, 0, sizeof(msg));
+	msg.mrq = MRQ_CPU_VHINT;
+	msg.tx.data = &req;
+	msg.tx.size = sizeof(req);
+
+	err = tegra_bpmp_transfer(bpmp, &msg);
+	if (err) {
+		table = ERR_PTR(err);
+		goto free;
+	}
+
+	for (i = data->vfloor; i <= data->vceil; i++) {
+		u16 ndiv = data->ndiv[i];
+
+		if (ndiv < data->ndiv_min || ndiv > data->ndiv_max)
+			continue;
+
+		/* Only store lowest voltage index for each rate */
+		if (i > 0 && ndiv == data->ndiv[i - 1])
+			continue;
+
+		num_rates++;
+	}
+
+	table = devm_kcalloc(&pdev->dev, num_rates + 1, sizeof(*table),
+			     GFP_KERNEL);
+	if (!table) {
+		table = ERR_PTR(-ENOMEM);
+		goto free;
+	}
+
+	for (i = data->vfloor, j = 0; i <= data->vceil; i++) {
+		struct cpufreq_frequency_table *point;
+		u16 ndiv = data->ndiv[i];
+		u32 edvd_val = 0;
+
+		if (ndiv < data->ndiv_min || ndiv > data->ndiv_max)
+			continue;
+
+		/* Only store lowest voltage index for each rate */
+		if (i > 0 && ndiv == data->ndiv[i - 1])
+			continue;
+
+		edvd_val |= i << EDVD_CORE_VOLT_FREQ_V_SHIFT;
+		edvd_val |= ndiv << EDVD_CORE_VOLT_FREQ_F_SHIFT;
+
+		point = &table[j++];
+		point->driver_data = edvd_val;
+		point->frequency = data->ref_clk_hz * ndiv / data->pdiv /
+			data->mdiv / 1000;
+	}
+
+	table[j].frequency = CPUFREQ_TABLE_END;
+
+free:
+	dma_free_coherent(bpmp->dev, sizeof(*data), virt, phys);
+
+	return table;
+}
+
+static int tegra186_cpufreq_probe(struct platform_device *pdev)
+{
+	struct tegra186_cpufreq_data *data;
+	struct tegra_bpmp *bpmp;
+	struct resource *res;
+	unsigned int i = 0, err;
+
+	data = devm_kzalloc(&pdev->dev, sizeof(*data), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	data->clusters = devm_kcalloc(&pdev->dev, ARRAY_SIZE(tegra186_clusters),
+				      sizeof(*data->clusters), GFP_KERNEL);
+	if (!data->clusters)
+		return -ENOMEM;
+
+	data->num_clusters = ARRAY_SIZE(tegra186_clusters);
+
+	bpmp = tegra_bpmp_get(&pdev->dev);
+	if (IS_ERR(bpmp))
+		return PTR_ERR(bpmp);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	data->regs = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(data->regs)) {
+		err = PTR_ERR(data->regs);
+		goto put_bpmp;
+	}
+
+	for (i = 0; i < data->num_clusters; i++) {
+		struct tegra186_cpufreq_cluster *cluster = &data->clusters[i];
+
+		cluster->info = &tegra186_clusters[i];
+		cluster->table = init_vhint_table(
+			pdev, bpmp, cluster->info->bpmp_cluster_id);
+		if (IS_ERR(cluster->table)) {
+			err = PTR_ERR(cluster->table);
+			goto put_bpmp;
+		}
+	}
+
+	tegra_bpmp_put(bpmp);
+
+	tegra186_cpufreq_driver.driver_data = data;
+
+	err = cpufreq_register_driver(&tegra186_cpufreq_driver);
+	if (err)
+		return err;
+
+	return 0;
+
+put_bpmp:
+	tegra_bpmp_put(bpmp);
+
+	return err;
+}
+
+static int tegra186_cpufreq_remove(struct platform_device *pdev)
+{
+	cpufreq_unregister_driver(&tegra186_cpufreq_driver);
+
+	return 0;
+}
+
+static const struct of_device_id tegra186_cpufreq_of_match[] = {
+	{ .compatible = "nvidia,tegra186-ccplex-cluster", },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, tegra186_cpufreq_of_match);
+
+static struct platform_driver tegra186_cpufreq_platform_driver = {
+	.driver = {
+		.name = "tegra186-cpufreq",
+		.of_match_table = tegra186_cpufreq_of_match,
+	},
+	.probe = tegra186_cpufreq_probe,
+	.remove = tegra186_cpufreq_remove,
+};
+module_platform_driver(tegra186_cpufreq_platform_driver);
+
+MODULE_AUTHOR("Mikko Perttunen <mperttunen@nvidia.com>");
+MODULE_DESCRIPTION("NVIDIA Tegra186 cpufreq driver");
+MODULE_LICENSE("GPL v2");

drivers/cpuidle/cpuidle-cps.c

@@ -118,7 +118,7 @@ static void __init cps_cpuidle_unregister(void)
 
 static int __init cps_cpuidle_init(void)
 {
-	int err, cpu, core, i;
+	int err, cpu, i;
 	struct cpuidle_device *device;
 
 	/* Detect supported states */
@@ -160,7 +160,6 @@ static int __init cps_cpuidle_init(void)
 	}
 
 	for_each_possible_cpu(cpu) {
-		core = cpu_data[cpu].core;
 		device = &per_cpu(cpuidle_dev, cpu);
 		device->cpu = cpu;
 #ifdef CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED

drivers/cpuidle/cpuidle-powernv.c

@@ -56,10 +56,9 @@ static int snooze_loop(struct cpuidle_device *dev,
 
 	snooze_exit_time = get_tb() + snooze_timeout;
 	ppc64_runlatch_off();
+	HMT_very_low();
 	while (!need_resched()) {
-		HMT_low();
-		HMT_very_low();
-		if (snooze_timeout_en && get_tb() > snooze_exit_time)
+		if (likely(snooze_timeout_en) && get_tb() > snooze_exit_time)
 			break;
 	}
 
@@ -215,11 +214,25 @@ static inline void add_powernv_state(int index, const char *name,
 	stop_psscr_table[index].mask = psscr_mask;
 }
 
+/*
+ * Returns 0 if prop1_len == prop2_len. Else returns -1
+ */
+static inline int validate_dt_prop_sizes(const char *prop1, int prop1_len,
+					 const char *prop2, int prop2_len)
+{
+	if (prop1_len == prop2_len)
+		return 0;
+
+	pr_warn("cpuidle-powernv: array sizes don't match for %s and %s\n",
+		prop1, prop2);
+	return -1;
+}
+
 static int powernv_add_idle_states(void)
 {
 	struct device_node *power_mgt;
 	int nr_idle_states = 1; /* Snooze */
-	int dt_idle_states;
+	int dt_idle_states, count;
 	u32 latency_ns[CPUIDLE_STATE_MAX];
 	u32 residency_ns[CPUIDLE_STATE_MAX];
 	u32 flags[CPUIDLE_STATE_MAX];
@@ -244,6 +257,21 @@ static int powernv_add_idle_states(void)
 		goto out;
 	}
 
+	count = of_property_count_u32_elems(power_mgt,
+					    "ibm,cpu-idle-state-latencies-ns");
+
+	if (validate_dt_prop_sizes("ibm,cpu-idle-state-flags", dt_idle_states,
+				   "ibm,cpu-idle-state-latencies-ns",
+				   count) != 0)
+		goto out;
+
+	count = of_property_count_strings(power_mgt,
+					  "ibm,cpu-idle-state-names");
+	if (validate_dt_prop_sizes("ibm,cpu-idle-state-flags", dt_idle_states,
+				   "ibm,cpu-idle-state-names",
+				   count) != 0)
+		goto out;
+
 	/*
 	 * Since snooze is used as first idle state, max idle states allowed is
 	 * CPUIDLE_STATE_MAX -1
@@ -278,6 +306,22 @@ static int powernv_add_idle_states(void)
 	has_stop_states = (flags[0] &
 			   (OPAL_PM_STOP_INST_FAST | OPAL_PM_STOP_INST_DEEP));
 	if (has_stop_states) {
+		count = of_property_count_u64_elems(power_mgt,
+						    "ibm,cpu-idle-state-psscr");
+		if (validate_dt_prop_sizes("ibm,cpu-idle-state-flags",
+					   dt_idle_states,
+					   "ibm,cpu-idle-state-psscr",
+					   count) != 0)
+			goto out;
+
+		count = of_property_count_u64_elems(power_mgt,
+						    "ibm,cpu-idle-state-psscr-mask");
+		if (validate_dt_prop_sizes("ibm,cpu-idle-state-flags",
+					   dt_idle_states,
+					   "ibm,cpu-idle-state-psscr-mask",
+					   count) != 0)
+			goto out;
+
 		if (of_property_read_u64_array(power_mgt,
 		    "ibm,cpu-idle-state-psscr", psscr_val, dt_idle_states)) {
 			pr_warn("cpuidle-powernv: missing ibm,cpu-idle-state-psscr in DT\n");
@@ -292,8 +336,21 @@ static int powernv_add_idle_states(void)
 		}
 	}
 
-	rc = of_property_read_u32_array(power_mgt,
-		"ibm,cpu-idle-state-residency-ns", residency_ns, dt_idle_states);
+	count = of_property_count_u32_elems(power_mgt,
+					    "ibm,cpu-idle-state-residency-ns");
+
+	if (count < 0) {
+		rc = count;
+	} else if (validate_dt_prop_sizes("ibm,cpu-idle-state-flags",
+					  dt_idle_states,
+					  "ibm,cpu-idle-state-residency-ns",
+					  count) != 0) {
+		goto out;
+	} else {
+		rc = of_property_read_u32_array(power_mgt,
+						"ibm,cpu-idle-state-residency-ns",
+						residency_ns, dt_idle_states);
+	}
 
 	for (i = 0; i < dt_idle_states; i++) {
 		unsigned int exit_latency, target_residency;

drivers/devfreq/governor.h

@@ -25,6 +25,35 @@
 #define DEVFREQ_GOV_SUSPEND			0x4
 #define DEVFREQ_GOV_RESUME			0x5
 
+/**
+ * struct devfreq_governor - Devfreq policy governor
+ * @node:		list node - contains registered devfreq governors
+ * @name:		Governor's name
+ * @immutable:		Immutable flag for governor. If the value is 1,
+ *			this govenror is never changeable to other governor.
+ * @get_target_freq:	Returns desired operating frequency for the device.
+ *			Basically, get_target_freq will run
+ *			devfreq_dev_profile.get_dev_status() to get the
+ *			status of the device (load = busy_time / total_time).
+ *			If no_central_polling is set, this callback is called
+ *			only with update_devfreq() notified by OPP.
+ * @event_handler:      Callback for devfreq core framework to notify events
+ *                      to governors. Events include per device governor
+ *                      init and exit, opp changes out of devfreq, suspend
+ *                      and resume of per device devfreq during device idle.
+ *
+ * Note that the callbacks are called with devfreq->lock locked by devfreq.
+ */
+struct devfreq_governor {
+	struct list_head node;
+
+	const char name[DEVFREQ_NAME_LEN];
+	const unsigned int immutable;
+	int (*get_target_freq)(struct devfreq *this, unsigned long *freq);
+	int (*event_handler)(struct devfreq *devfreq,
+				unsigned int event, void *data);
+};
+
 /* Caution: devfreq->lock must be locked before calling update_devfreq */
 extern int update_devfreq(struct devfreq *devfreq);
 

drivers/power/avs/rockchip-io-domain.c

@@ -43,6 +43,10 @@
 #define RK3288_SOC_CON2_FLASH0		BIT(7)
 #define RK3288_SOC_FLASH_SUPPLY_NUM	2
 
+#define RK3328_SOC_CON4			0x410
+#define RK3328_SOC_CON4_VCCIO2		BIT(7)
+#define RK3328_SOC_VCCIO2_SUPPLY_NUM	1
+
 #define RK3368_SOC_CON15		0x43c
 #define RK3368_SOC_CON15_FLASH0		BIT(14)
 #define RK3368_SOC_FLASH_SUPPLY_NUM	2
@@ -166,6 +170,25 @@ static void rk3288_iodomain_init(struct rockchip_iodomain *iod)
 		dev_warn(iod->dev, "couldn't update flash0 ctrl\n");
 }
 
+static void rk3328_iodomain_init(struct rockchip_iodomain *iod)
+{
+	int ret;
+	u32 val;
+
+	/* if no vccio2 supply we should leave things alone */
+	if (!iod->supplies[RK3328_SOC_VCCIO2_SUPPLY_NUM].reg)
+		return;
+
+	/*
+	 * set vccio2 iodomain to also use this framework
+	 * instead of a special gpio.
+	 */
+	val = RK3328_SOC_CON4_VCCIO2 | (RK3328_SOC_CON4_VCCIO2 << 16);
+	ret = regmap_write(iod->grf, RK3328_SOC_CON4, val);
+	if (ret < 0)
+		dev_warn(iod->dev, "couldn't update vccio2 vsel ctrl\n");
+}
+
 static void rk3368_iodomain_init(struct rockchip_iodomain *iod)
 {
 	int ret;
@@ -247,6 +270,20 @@ static const struct rockchip_iodomain_soc_data soc_data_rk3288 = {
 	.init = rk3288_iodomain_init,
 };
 
+static const struct rockchip_iodomain_soc_data soc_data_rk3328 = {
+	.grf_offset = 0x410,
+	.supply_names = {
+		"vccio1",
+		"vccio2",
+		"vccio3",
+		"vccio4",
+		"vccio5",
+		"vccio6",
+		"pmuio",
+	},
+	.init = rk3328_iodomain_init,
+};
+
 static const struct rockchip_iodomain_soc_data soc_data_rk3368 = {
 	.grf_offset = 0x900,
 	.supply_names = {
@@ -311,6 +348,10 @@ static const struct of_device_id rockchip_iodomain_match[] = {
 		.compatible = "rockchip,rk3288-io-voltage-domain",
 		.data = (void *)&soc_data_rk3288
 	},
+	{
+		.compatible = "rockchip,rk3328-io-voltage-domain",
+		.data = (void *)&soc_data_rk3328
+	},
 	{
 		.compatible = "rockchip,rk3368-io-voltage-domain",
 		.data = (void *)&soc_data_rk3368

drivers/thermal/Kconfig

@@ -291,18 +291,6 @@ config ARMADA_THERMAL
 	  Enable this option if you want to have support for thermal management
 	  controller present in Armada 370 and Armada XP SoC.
 
-config DB8500_CPUFREQ_COOLING
-	tristate "DB8500 cpufreq cooling"
-	depends on ARCH_U8500 || COMPILE_TEST
-	depends on HAS_IOMEM
-	depends on CPU_THERMAL
-	default y
-	help
-	  Adds DB8500 cpufreq cooling devices, and these cooling devices can be
-	  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

drivers/thermal/Makefile

@@ -41,7 +41,6 @@ obj-$(CONFIG_TANGO_THERMAL)	+= tango_thermal.o
 obj-$(CONFIG_IMX_THERMAL)	+= imx_thermal.o
 obj-$(CONFIG_MAX77620_THERMAL)	+= max77620_thermal.o
 obj-$(CONFIG_QORIQ_THERMAL)	+= qoriq_thermal.o
-obj-$(CONFIG_DB8500_CPUFREQ_COOLING)	+= db8500_cpufreq_cooling.o
 obj-$(CONFIG_INTEL_POWERCLAMP)	+= intel_powerclamp.o
 obj-$(CONFIG_X86_PKG_TEMP_THERMAL)	+= x86_pkg_temp_thermal.o
 obj-$(CONFIG_INTEL_SOC_DTS_IOSF_CORE)	+= intel_soc_dts_iosf.o

drivers/thermal/db8500_cpufreq_cooling.c

@@ -1,105 +0,0 @@
-/*
- * db8500_cpufreq_cooling.c - DB8500 cpufreq works as cooling device.
- *
- * Copyright (C) 2012 ST-Ericsson
- * Copyright (C) 2012 Linaro Ltd.
- *
- * Author: Hongbo Zhang <hongbo.zhang@linaro.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * 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/cpu_cooling.h>
-#include <linux/err.h>
-#include <linux/module.h>
-#include <linux/of.h>
-#include <linux/platform_device.h>
-#include <linux/slab.h>
-
-static int db8500_cpufreq_cooling_probe(struct platform_device *pdev)
-{
-	struct thermal_cooling_device *cdev;
-
-	cdev = cpufreq_cooling_register(cpu_present_mask);
-	if (IS_ERR(cdev)) {
-		int ret = PTR_ERR(cdev);
-
-		if (ret != -EPROBE_DEFER)
-			dev_err(&pdev->dev,
-				"Failed to register cooling device %d\n",
-				ret);
-				
-		return ret;
-	}
-
-	platform_set_drvdata(pdev, cdev);
-
-	dev_info(&pdev->dev, "Cooling device registered: %s\n",	cdev->type);
-
-	return 0;
-}
-
-static int db8500_cpufreq_cooling_remove(struct platform_device *pdev)
-{
-	struct thermal_cooling_device *cdev = platform_get_drvdata(pdev);
-
-	cpufreq_cooling_unregister(cdev);
-
-	return 0;
-}
-
-static int db8500_cpufreq_cooling_suspend(struct platform_device *pdev,
-		pm_message_t state)
-{
-	return -ENOSYS;
-}
-
-static int db8500_cpufreq_cooling_resume(struct platform_device *pdev)
-{
-	return -ENOSYS;
-}
-
-#ifdef CONFIG_OF
-static const struct of_device_id db8500_cpufreq_cooling_match[] = {
-	{ .compatible = "stericsson,db8500-cpufreq-cooling" },
-	{},
-};
-MODULE_DEVICE_TABLE(of, db8500_cpufreq_cooling_match);
-#endif
-
-static struct platform_driver db8500_cpufreq_cooling_driver = {
-	.driver = {
-		.name = "db8500-cpufreq-cooling",
-		.of_match_table = of_match_ptr(db8500_cpufreq_cooling_match),
-	},
-	.probe = db8500_cpufreq_cooling_probe,
-	.suspend = db8500_cpufreq_cooling_suspend,
-	.resume = db8500_cpufreq_cooling_resume,
-	.remove = db8500_cpufreq_cooling_remove,
-};
-
-static int __init db8500_cpufreq_cooling_init(void)
-{
-	return platform_driver_register(&db8500_cpufreq_cooling_driver);
-}
-
-static void __exit db8500_cpufreq_cooling_exit(void)
-{
-	platform_driver_unregister(&db8500_cpufreq_cooling_driver);
-}
-
-/* Should be later than db8500_cpufreq_register */
-late_initcall(db8500_cpufreq_cooling_init);
-module_exit(db8500_cpufreq_cooling_exit);
-
-MODULE_AUTHOR("Hongbo Zhang <hongbo.zhang@stericsson.com>");
-MODULE_DESCRIPTION("DB8500 cpufreq cooling driver");
-MODULE_LICENSE("GPL");

include/linux/cpufreq.h

@@ -120,6 +120,13 @@ struct cpufreq_policy {
 	bool			fast_switch_possible;
 	bool			fast_switch_enabled;
 
+	/*
+	 * Preferred average time interval between consecutive invocations of
+	 * the driver to set the frequency for this policy.  To be set by the
+	 * scaling driver (0, which is the default, means no preference).
+	 */
+	unsigned int		transition_delay_us;
+
 	 /* Cached frequency lookup from cpufreq_driver_resolve_freq. */
 	unsigned int cached_target_freq;
 	int cached_resolved_idx;

include/linux/devfreq.h

@@ -27,6 +27,7 @@
 #define DEVFREQ_POSTCHANGE		(1)
 
 struct devfreq;
+struct devfreq_governor;
 
 /**
  * struct devfreq_dev_status - Data given from devfreq user device to
@@ -100,35 +101,6 @@ struct devfreq_dev_profile {
 	unsigned int max_state;
 };
 
-/**
- * struct devfreq_governor - Devfreq policy governor
- * @node:		list node - contains registered devfreq governors
- * @name:		Governor's name
- * @immutable:		Immutable flag for governor. If the value is 1,
- *			this govenror is never changeable to other governor.
- * @get_target_freq:	Returns desired operating frequency for the device.
- *			Basically, get_target_freq will run
- *			devfreq_dev_profile.get_dev_status() to get the
- *			status of the device (load = busy_time / total_time).
- *			If no_central_polling is set, this callback is called
- *			only with update_devfreq() notified by OPP.
- * @event_handler:      Callback for devfreq core framework to notify events
- *                      to governors. Events include per device governor
- *                      init and exit, opp changes out of devfreq, suspend
- *                      and resume of per device devfreq during device idle.
- *
- * Note that the callbacks are called with devfreq->lock locked by devfreq.
- */
-struct devfreq_governor {
-	struct list_head node;
-
-	const char name[DEVFREQ_NAME_LEN];
-	const unsigned int immutable;
-	int (*get_target_freq)(struct devfreq *this, unsigned long *freq);
-	int (*event_handler)(struct devfreq *devfreq,
-				unsigned int event, void *data);
-};
-
 /**
  * struct devfreq - Device devfreq structure
  * @node:	list node - contains the devices with devfreq that have been

include/linux/pm_domain.h

@@ -20,6 +20,7 @@
 /* Defines used for the flags field in the struct generic_pm_domain */
 #define GENPD_FLAG_PM_CLK	(1U << 0) /* PM domain uses PM clk */
 #define GENPD_FLAG_IRQ_SAFE	(1U << 1) /* PM domain operates in atomic */
+#define GENPD_FLAG_ALWAYS_ON	(1U << 2) /* PM domain is always powered on */
 
 enum gpd_status {
 	GPD_STATE_ACTIVE = 0,	/* PM domain is active */

include/linux/tick.h

@@ -117,6 +117,7 @@ extern void tick_nohz_idle_enter(void);
 extern void tick_nohz_idle_exit(void);
 extern void tick_nohz_irq_exit(void);
 extern ktime_t tick_nohz_get_sleep_length(void);
+extern unsigned long tick_nohz_get_idle_calls(void);
 extern u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time);
 extern u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time);
 #else /* !CONFIG_NO_HZ_COMMON */

kernel/sched/cpufreq_schedutil.c

@@ -61,6 +61,11 @@ struct sugov_cpu {
 	unsigned long util;
 	unsigned long max;
 	unsigned int flags;
+
+	/* The field below is for single-CPU policies only. */
+#ifdef CONFIG_NO_HZ_COMMON
+	unsigned long saved_idle_calls;
+#endif
 };
 
 static DEFINE_PER_CPU(struct sugov_cpu, sugov_cpu);
@@ -93,22 +98,23 @@ static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time,
 {
 	struct cpufreq_policy *policy = sg_policy->policy;
 
+	if (sg_policy->next_freq == next_freq)
+		return;
+
+	if (sg_policy->next_freq > next_freq)
+		next_freq = (sg_policy->next_freq + next_freq) >> 1;
+
+	sg_policy->next_freq = next_freq;
 	sg_policy->last_freq_update_time = time;
 
 	if (policy->fast_switch_enabled) {
-		if (sg_policy->next_freq == next_freq) {
-			trace_cpu_frequency(policy->cur, smp_processor_id());
-			return;
-		}
-		sg_policy->next_freq = next_freq;
 		next_freq = cpufreq_driver_fast_switch(policy, next_freq);
 		if (next_freq == CPUFREQ_ENTRY_INVALID)
 			return;
 
 		policy->cur = next_freq;
 		trace_cpu_frequency(next_freq, smp_processor_id());
-	} else if (sg_policy->next_freq != next_freq) {
-		sg_policy->next_freq = next_freq;
+	} else {
 		sg_policy->work_in_progress = true;
 		irq_work_queue(&sg_policy->irq_work);
 	}
@@ -192,6 +198,19 @@ static void sugov_iowait_boost(struct sugov_cpu *sg_cpu, unsigned long *util,
 	sg_cpu->iowait_boost >>= 1;
 }
 
+#ifdef CONFIG_NO_HZ_COMMON
+static bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu)
+{
+	unsigned long idle_calls = tick_nohz_get_idle_calls();
+	bool ret = idle_calls == sg_cpu->saved_idle_calls;
+
+	sg_cpu->saved_idle_calls = idle_calls;
+	return ret;
+}
+#else
+static inline bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu) { return false; }
+#endif /* CONFIG_NO_HZ_COMMON */
+
 static void sugov_update_single(struct update_util_data *hook, u64 time,
 				unsigned int flags)
 {
@@ -200,6 +219,7 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
 	struct cpufreq_policy *policy = sg_policy->policy;
 	unsigned long util, max;
 	unsigned int next_f;
+	bool busy;
 
 	sugov_set_iowait_boost(sg_cpu, time, flags);
 	sg_cpu->last_update = time;
@@ -207,40 +227,37 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
 	if (!sugov_should_update_freq(sg_policy, time))
 		return;
 
+	busy = sugov_cpu_is_busy(sg_cpu);
+
 	if (flags & SCHED_CPUFREQ_RT_DL) {
 		next_f = policy->cpuinfo.max_freq;
 	} else {
 		sugov_get_util(&util, &max);
 		sugov_iowait_boost(sg_cpu, &util, &max);
 		next_f = get_next_freq(sg_policy, util, max);
+		/*
+		 * Do not reduce the frequency if the CPU has not been idle
+		 * recently, as the reduction is likely to be premature then.
+		 */
+		if (busy && next_f < sg_policy->next_freq)
+			next_f = sg_policy->next_freq;
 	}
 	sugov_update_commit(sg_policy, time, next_f);
 }
 
-static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu,
-					   unsigned long util, unsigned long max,
-					   unsigned int flags)
+static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu)
 {
 	struct sugov_policy *sg_policy = sg_cpu->sg_policy;
 	struct cpufreq_policy *policy = sg_policy->policy;
-	unsigned int max_f = policy->cpuinfo.max_freq;
 	u64 last_freq_update_time = sg_policy->last_freq_update_time;
+	unsigned long util = 0, max = 1;
 	unsigned int j;
 
-	if (flags & SCHED_CPUFREQ_RT_DL)
-		return max_f;
-
-	sugov_iowait_boost(sg_cpu, &util, &max);
-
 	for_each_cpu(j, policy->cpus) {
-		struct sugov_cpu *j_sg_cpu;
+		struct sugov_cpu *j_sg_cpu = &per_cpu(sugov_cpu, j);
 		unsigned long j_util, j_max;
 		s64 delta_ns;
 
-		if (j == smp_processor_id())
-			continue;
-
-		j_sg_cpu = &per_cpu(sugov_cpu, j);
 		/*
 		 * If the CPU utilization was last updated before the previous
 		 * frequency update and the time elapsed between the last update
@@ -254,7 +271,7 @@ static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu,
 			continue;
 		}
 		if (j_sg_cpu->flags & SCHED_CPUFREQ_RT_DL)
-			return max_f;
+			return policy->cpuinfo.max_freq;
 
 		j_util = j_sg_cpu->util;
 		j_max = j_sg_cpu->max;
@@ -289,7 +306,11 @@ static void sugov_update_shared(struct update_util_data *hook, u64 time,
 	sg_cpu->last_update = time;
 
 	if (sugov_should_update_freq(sg_policy, time)) {
-		next_f = sugov_next_freq_shared(sg_cpu, util, max, flags);
+		if (flags & SCHED_CPUFREQ_RT_DL)
+			next_f = sg_policy->policy->cpuinfo.max_freq;
+		else
+			next_f = sugov_next_freq_shared(sg_cpu);
+
 		sugov_update_commit(sg_policy, time, next_f);
 	}
 
@@ -473,7 +494,6 @@ static int sugov_init(struct cpufreq_policy *policy)
 {
 	struct sugov_policy *sg_policy;
 	struct sugov_tunables *tunables;
-	unsigned int lat;
 	int ret = 0;
 
 	/* State should be equivalent to EXIT */
@@ -512,10 +532,16 @@ static int sugov_init(struct cpufreq_policy *policy)
 		goto stop_kthread;
 	}
 
-	tunables->rate_limit_us = LATENCY_MULTIPLIER;
-	lat = policy->cpuinfo.transition_latency / NSEC_PER_USEC;
-	if (lat)
-		tunables->rate_limit_us *= lat;
+	if (policy->transition_delay_us) {
+		tunables->rate_limit_us = policy->transition_delay_us;
+	} else {
+		unsigned int lat;
+
+		tunables->rate_limit_us = LATENCY_MULTIPLIER;
+		lat = policy->cpuinfo.transition_latency / NSEC_PER_USEC;
+		if (lat)
+			tunables->rate_limit_us *= lat;
+	}
 
 	policy->governor_data = sg_policy;
 	sg_policy->tunables = tunables;

kernel/time/tick-sched.c

@@ -993,6 +993,18 @@ ktime_t tick_nohz_get_sleep_length(void)
 	return ts->sleep_length;
 }
 
+/**
+ * tick_nohz_get_idle_calls - return the current idle calls counter value
+ *
+ * Called from the schedutil frequency scaling governor in scheduler context.
+ */
+unsigned long tick_nohz_get_idle_calls(void)
+{
+	struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched);
+
+	return ts->idle_calls;
+}
+
 static void tick_nohz_account_idle_ticks(struct tick_sched *ts)
 {
 #ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE

tools/power/pm-graph/Makefile

@@ -0,0 +1,28 @@
+PREFIX		?= /usr
+DESTDIR		?=
+
+all:
+	@echo "Nothing to build"
+
+install :
+	install -d  $(DESTDIR)$(PREFIX)/lib/pm-graph
+	install analyze_suspend.py $(DESTDIR)$(PREFIX)/lib/pm-graph
+	install analyze_boot.py $(DESTDIR)$(PREFIX)/lib/pm-graph
+
+	ln -s $(DESTDIR)$(PREFIX)/lib/pm-graph/analyze_boot.py $(DESTDIR)$(PREFIX)/bin/bootgraph
+	ln -s $(DESTDIR)$(PREFIX)/lib/pm-graph/analyze_suspend.py $(DESTDIR)$(PREFIX)/bin/sleepgraph
+
+	install -d  $(DESTDIR)$(PREFIX)/share/man/man8
+	install bootgraph.8 $(DESTDIR)$(PREFIX)/share/man/man8
+	install sleepgraph.8 $(DESTDIR)$(PREFIX)/share/man/man8
+
+uninstall :
+	rm $(DESTDIR)$(PREFIX)/share/man/man8/bootgraph.8
+	rm $(DESTDIR)$(PREFIX)/share/man/man8/sleepgraph.8
+
+	rm $(DESTDIR)$(PREFIX)/bin/bootgraph
+	rm $(DESTDIR)$(PREFIX)/bin/sleepgraph
+
+	rm $(DESTDIR)$(PREFIX)/lib/pm-graph/analyze_boot.py
+	rm $(DESTDIR)$(PREFIX)/lib/pm-graph/analyze_suspend.py
+	rmdir $(DESTDIR)$(PREFIX)/lib/pm-graph

tools/power/pm-graph/analyze_boot.py

@@ -0,0 +1,824 @@
+#!/usr/bin/python
+#
+# Tool for analyzing boot timing
+# Copyright (c) 2013, Intel Corporation.
+#
+# 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.
+#
+# Authors:
+#	 Todd Brandt <todd.e.brandt@linux.intel.com>
+#
+# Description:
+#	 This tool is designed to assist kernel and OS developers in optimizing
+#	 their linux stack's boot time. It creates an html representation of
+#	 the kernel boot timeline up to the start of the init process.
+#
+
+# ----------------- LIBRARIES --------------------
+
+import sys
+import time
+import os
+import string
+import re
+import platform
+import shutil
+from datetime import datetime, timedelta
+from subprocess import call, Popen, PIPE
+import analyze_suspend as aslib
+
+# ----------------- CLASSES --------------------
+
+# Class: SystemValues
+# Description:
+#	 A global, single-instance container used to
+#	 store system values and test parameters
+class SystemValues(aslib.SystemValues):
+	title = 'BootGraph'
+	version = 2.0
+	hostname = 'localhost'
+	testtime = ''
+	kernel = ''
+	dmesgfile = ''
+	ftracefile = ''
+	htmlfile = 'bootgraph.html'
+	outfile = ''
+	phoronix = False
+	addlogs = False
+	useftrace = False
+	usedevsrc = True
+	suspendmode = 'boot'
+	max_graph_depth = 2
+	graph_filter = 'do_one_initcall'
+	reboot = False
+	manual = False
+	iscronjob = False
+	timeformat = '%.6f'
+	def __init__(self):
+		if('LOG_FILE' in os.environ and 'TEST_RESULTS_IDENTIFIER' in os.environ):
+			self.phoronix = True
+			self.addlogs = True
+			self.outfile = os.environ['LOG_FILE']
+			self.htmlfile = os.environ['LOG_FILE']
+		self.hostname = platform.node()
+		self.testtime = datetime.now().strftime('%Y-%m-%d_%H:%M:%S')
+		if os.path.exists('/proc/version'):
+			fp = open('/proc/version', 'r')
+			val = fp.read().strip()
+			fp.close()
+			self.kernel = self.kernelVersion(val)
+		else:
+			self.kernel = 'unknown'
+	def kernelVersion(self, msg):
+		return msg.split()[2]
+	def kernelParams(self):
+		cmdline = 'initcall_debug log_buf_len=32M'
+		if self.useftrace:
+			cmdline += ' trace_buf_size=128M trace_clock=global '\
+			'trace_options=nooverwrite,funcgraph-abstime,funcgraph-cpu,'\
+			'funcgraph-duration,funcgraph-proc,funcgraph-tail,'\
+			'nofuncgraph-overhead,context-info,graph-time '\
+			'ftrace=function_graph '\
+			'ftrace_graph_max_depth=%d '\
+			'ftrace_graph_filter=%s' % \
+				(self.max_graph_depth, self.graph_filter)
+		return cmdline
+	def setGraphFilter(self, val):
+		fp = open(self.tpath+'available_filter_functions')
+		master = fp.read().split('\n')
+		fp.close()
+		for i in val.split(','):
+			func = i.strip()
+			if func not in master:
+				doError('function "%s" not available for ftrace' % func)
+		self.graph_filter = val
+	def cronjobCmdString(self):
+		cmdline = '%s -cronjob' % os.path.abspath(sys.argv[0])
+		args = iter(sys.argv[1:])
+		for arg in args:
+			if arg in ['-h', '-v', '-cronjob', '-reboot']:
+				continue
+			elif arg in ['-o', '-dmesg', '-ftrace', '-filter']:
+				args.next()
+				continue
+			cmdline += ' '+arg
+		if self.graph_filter != 'do_one_initcall':
+			cmdline += ' -filter "%s"' % self.graph_filter
+		cmdline += ' -o "%s"' % os.path.abspath(self.htmlfile)
+		return cmdline
+	def manualRebootRequired(self):
+		cmdline = self.kernelParams()
+		print 'To generate a new timeline manually, follow these steps:\n'
+		print '1. Add the CMDLINE string to your kernel command line.'
+		print '2. Reboot the system.'
+		print '3. After reboot, re-run this tool with the same arguments but no command (w/o -reboot or -manual).\n'
+		print 'CMDLINE="%s"' % cmdline
+		sys.exit()
+
+sysvals = SystemValues()
+
+# Class: Data
+# Description:
+#	 The primary container for test data.
+class Data(aslib.Data):
+	dmesg = {}  # root data structure
+	start = 0.0 # test start
+	end = 0.0   # test end
+	dmesgtext = []   # dmesg text file in memory
+	testnumber = 0
+	idstr = ''
+	html_device_id = 0
+	valid = False
+	initstart = 0.0
+	boottime = ''
+	phases = ['boot']
+	do_one_initcall = False
+	def __init__(self, num):
+		self.testnumber = num
+		self.idstr = 'a'
+		self.dmesgtext = []
+		self.dmesg = {
+			'boot': {'list': dict(), 'start': -1.0, 'end': -1.0, 'row': 0, 'color': '#dddddd'}
+		}
+	def deviceTopology(self):
+		return ''
+	def newAction(self, phase, name, start, end, ret, ulen):
+		# new device callback for a specific phase
+		self.html_device_id += 1
+		devid = '%s%d' % (self.idstr, self.html_device_id)
+		list = self.dmesg[phase]['list']
+		length = -1.0
+		if(start >= 0 and end >= 0):
+			length = end - start
+		i = 2
+		origname = name
+		while(name in list):
+			name = '%s[%d]' % (origname, i)
+			i += 1
+		list[name] = {'name': name, 'start': start, 'end': end,
+			'pid': 0, 'length': length, 'row': 0, 'id': devid,
+			'ret': ret, 'ulen': ulen }
+		return name
+	def deviceMatch(self, cg):
+		if cg.end - cg.start == 0:
+			return True
+		list = self.dmesg['boot']['list']
+		for devname in list:
+			dev = list[devname]
+			if cg.name == 'do_one_initcall':
+				if(cg.start <= dev['start'] and cg.end >= dev['end'] and dev['length'] > 0):
+					dev['ftrace'] = cg
+					self.do_one_initcall = True
+					return True
+			else:
+				if(cg.start > dev['start'] and cg.end < dev['end']):
+					if 'ftraces' not in dev:
+						dev['ftraces'] = []
+					dev['ftraces'].append(cg)
+					return True
+		return False
+
+# ----------------- FUNCTIONS --------------------
+
+# Function: loadKernelLog
+# Description:
+#	 Load a raw kernel log from dmesg
+def loadKernelLog():
+	data = Data(0)
+	data.dmesg['boot']['start'] = data.start = ktime = 0.0
+	sysvals.stamp = {
+		'time': datetime.now().strftime('%B %d %Y, %I:%M:%S %p'),
+		'host': sysvals.hostname,
+		'mode': 'boot', 'kernel': ''}
+
+	devtemp = dict()
+	if(sysvals.dmesgfile):
+		lf = open(sysvals.dmesgfile, 'r')
+	else:
+		lf = Popen('dmesg', stdout=PIPE).stdout
+	for line in lf:
+		line = line.replace('\r\n', '')
+		idx = line.find('[')
+		if idx > 1:
+			line = line[idx:]
+		m = re.match('[ \t]*(\[ *)(?P<ktime>[0-9\.]*)(\]) (?P<msg>.*)', line)
+		if(not m):
+			continue
+		ktime = float(m.group('ktime'))
+		if(ktime > 120):
+			break
+		msg = m.group('msg')
+		data.end = data.initstart = ktime
+		data.dmesgtext.append(line)
+		if(ktime == 0.0 and re.match('^Linux version .*', msg)):
+			if(not sysvals.stamp['kernel']):
+				sysvals.stamp['kernel'] = sysvals.kernelVersion(msg)
+			continue
+		m = re.match('.* setting system clock to (?P<t>.*) UTC.*', msg)
+		if(m):
+			bt = datetime.strptime(m.group('t'), '%Y-%m-%d %H:%M:%S')
+			bt = bt - timedelta(seconds=int(ktime))
+			data.boottime = bt.strftime('%Y-%m-%d_%H:%M:%S')
+			sysvals.stamp['time'] = bt.strftime('%B %d %Y, %I:%M:%S %p')
+			continue
+		m = re.match('^calling *(?P<f>.*)\+.*', msg)
+		if(m):
+			devtemp[m.group('f')] = ktime
+			continue
+		m = re.match('^initcall *(?P<f>.*)\+.* returned (?P<r>.*) after (?P<t>.*) usecs', msg)
+		if(m):
+			data.valid = True
+			f, r, t = m.group('f', 'r', 't')
+			if(f in devtemp):
+				data.newAction('boot', f, devtemp[f], ktime, int(r), int(t))
+				data.end = ktime
+				del devtemp[f]
+			continue
+		if(re.match('^Freeing unused kernel memory.*', msg)):
+			break
+
+	data.dmesg['boot']['end'] = data.end
+	lf.close()
+	return data
+
+# Function: loadTraceLog
+# Description:
+#	 Check if trace is available and copy to a temp file
+def loadTraceLog(data):
+	# load the data to a temp file if none given
+	if not sysvals.ftracefile:
+		lib = aslib.sysvals
+		aslib.rootCheck(True)
+		if not lib.verifyFtrace():
+			doError('ftrace not available')
+		if lib.fgetVal('current_tracer').strip() != 'function_graph':
+			doError('ftrace not configured for a boot callgraph')
+		sysvals.ftracefile = '/tmp/boot_ftrace.%s.txt' % os.getpid()
+		call('cat '+lib.tpath+'trace > '+sysvals.ftracefile, shell=True)
+	if not sysvals.ftracefile:
+		doError('No trace data available')
+
+	# parse the trace log
+	ftemp = dict()
+	tp = aslib.TestProps()
+	tp.setTracerType('function_graph')
+	tf = open(sysvals.ftracefile, 'r')
+	for line in tf:
+		if line[0] == '#':
+			continue
+		m = re.match(tp.ftrace_line_fmt, line.strip())
+		if(not m):
+			continue
+		m_time, m_proc, m_pid, m_msg, m_dur = \
+			m.group('time', 'proc', 'pid', 'msg', 'dur')
+		if float(m_time) > data.end:
+			break
+		if(m_time and m_pid and m_msg):
+			t = aslib.FTraceLine(m_time, m_msg, m_dur)
+			pid = int(m_pid)
+		else:
+			continue
+		if t.fevent or t.fkprobe:
+			continue
+		key = (m_proc, pid)
+		if(key not in ftemp):
+			ftemp[key] = []
+			ftemp[key].append(aslib.FTraceCallGraph(pid))
+		cg = ftemp[key][-1]
+		if(cg.addLine(t)):
+			ftemp[key].append(aslib.FTraceCallGraph(pid))
+	tf.close()
+
+	# add the callgraph data to the device hierarchy
+	for key in ftemp:
+		proc, pid = key
+		for cg in ftemp[key]:
+			if len(cg.list) < 1 or cg.invalid:
+				continue
+			if(not cg.postProcess()):
+				print('Sanity check failed for %s-%d' % (proc, pid))
+				continue
+			# match cg data to devices
+			if not data.deviceMatch(cg):
+				print ' BAD: %s %s-%d [%f - %f]' % (cg.name, proc, pid, cg.start, cg.end)
+
+# Function: colorForName
+# Description:
+#	 Generate a repeatable color from a list for a given name
+def colorForName(name):
+	list = [
+		('c1', '#ec9999'),
+		('c2', '#ffc1a6'),
+		('c3', '#fff0a6'),
+		('c4', '#adf199'),
+		('c5', '#9fadea'),
+		('c6', '#a699c1'),
+		('c7', '#ad99b4'),
+		('c8', '#eaffea'),
+		('c9', '#dcecfb'),
+		('c10', '#ffffea')
+	]
+	i = 0
+	total = 0
+	count = len(list)
+	while i < len(name):
+		total += ord(name[i])
+		i += 1
+	return list[total % count]
+
+def cgOverview(cg, minlen):
+	stats = dict()
+	large = []
+	for l in cg.list:
+		if l.fcall and l.depth == 1:
+			if l.length >= minlen:
+				large.append(l)
+			if l.name not in stats:
+				stats[l.name] = [0, 0.0]
+			stats[l.name][0] += (l.length * 1000.0)
+			stats[l.name][1] += 1
+	return (large, stats)
+
+# Function: createBootGraph
+# Description:
+#	 Create the output html file from the resident test data
+# Arguments:
+#	 testruns: array of Data objects from parseKernelLog or parseTraceLog
+# Output:
+#	 True if the html file was created, false if it failed
+def createBootGraph(data, embedded):
+	# html function templates
+	html_srccall = '<div id={6} title="{5}" class="srccall" style="left:{1}%;top:{2}px;height:{3}px;width:{4}%;line-height:{3}px;">{0}</div>\n'
+	html_timetotal = '<table class="time1">\n<tr>'\
+		'<td class="blue">Time from Kernel Boot to start of User Mode: <b>{0} ms</b></td>'\
+		'</tr>\n</table>\n'
+
+	# device timeline
+	devtl = aslib.Timeline(100, 20)
+
+	# write the test title and general info header
+	devtl.createHeader(sysvals, 'noftrace')
+
+	# Generate the header for this timeline
+	t0 = data.start
+	tMax = data.end
+	tTotal = tMax - t0
+	if(tTotal == 0):
+		print('ERROR: No timeline data')
+		return False
+	boot_time = '%.0f'%(tTotal*1000)
+	devtl.html += html_timetotal.format(boot_time)
+
+	# determine the maximum number of rows we need to draw
+	phase = 'boot'
+	list = data.dmesg[phase]['list']
+	devlist = []
+	for devname in list:
+		d = aslib.DevItem(0, phase, list[devname])
+		devlist.append(d)
+	devtl.getPhaseRows(devlist)
+	devtl.calcTotalRows()
+
+	# draw the timeline background
+	devtl.createZoomBox()
+	boot = data.dmesg[phase]
+	length = boot['end']-boot['start']
+	left = '%.3f' % (((boot['start']-t0)*100.0)/tTotal)
+	width = '%.3f' % ((length*100.0)/tTotal)
+	devtl.html += devtl.html_tblock.format(phase, left, width, devtl.scaleH)
+	devtl.html += devtl.html_phase.format('0', '100', \
+		'%.3f'%devtl.scaleH, '%.3f'%devtl.bodyH, \
+		'white', '')
+
+	# draw the device timeline
+	num = 0
+	devstats = dict()
+	for devname in sorted(list):
+		cls, color = colorForName(devname)
+		dev = list[devname]
+		info = '@|%.3f|%.3f|%.3f|%d' % (dev['start']*1000.0, dev['end']*1000.0,
+			dev['ulen']/1000.0, dev['ret'])
+		devstats[dev['id']] = {'info':info}
+		dev['color'] = color
+		height = devtl.phaseRowHeight(0, phase, dev['row'])
+		top = '%.6f' % ((dev['row']*height) + devtl.scaleH)
+		left = '%.6f' % (((dev['start']-t0)*100)/tTotal)
+		width = '%.6f' % (((dev['end']-dev['start'])*100)/tTotal)
+		length = ' (%0.3f ms) ' % ((dev['end']-dev['start'])*1000)
+		devtl.html += devtl.html_device.format(dev['id'],
+			devname+length+'kernel_mode', left, top, '%.3f'%height,
+			width, devname, ' '+cls, '')
+		rowtop = devtl.phaseRowTop(0, phase, dev['row'])
+		height = '%.6f' % (devtl.rowH / 2)
+		top = '%.6f' % (rowtop + devtl.scaleH + (devtl.rowH / 2))
+		if data.do_one_initcall:
+			if('ftrace' not in dev):
+				continue
+			cg = dev['ftrace']
+			large, stats = cgOverview(cg, 0.001)
+			devstats[dev['id']]['fstat'] = stats
+			for l in large:
+				left = '%f' % (((l.time-t0)*100)/tTotal)
+				width = '%f' % (l.length*100/tTotal)
+				title = '%s (%0.3fms)' % (l.name, l.length * 1000.0)
+				devtl.html += html_srccall.format(l.name, left,
+					top, height, width, title, 'x%d'%num)
+				num += 1
+			continue
+		if('ftraces' not in dev):
+			continue
+		for cg in dev['ftraces']:
+			left = '%f' % (((cg.start-t0)*100)/tTotal)
+			width = '%f' % ((cg.end-cg.start)*100/tTotal)
+			cglen = (cg.end - cg.start) * 1000.0
+			title = '%s (%0.3fms)' % (cg.name, cglen)
+			cg.id = 'x%d' % num
+			devtl.html += html_srccall.format(cg.name, left,
+				top, height, width, title, dev['id']+cg.id)
+			num += 1
+
+	# draw the time scale, try to make the number of labels readable
+	devtl.createTimeScale(t0, tMax, tTotal, phase)
+	devtl.html += '</div>\n'
+
+	# timeline is finished
+	devtl.html += '</div>\n</div>\n'
+
+	if(sysvals.outfile == sysvals.htmlfile):
+		hf = open(sysvals.htmlfile, 'a')
+	else:
+		hf = open(sysvals.htmlfile, 'w')
+
+	# add the css if this is not an embedded run
+	extra = '\
+		.c1 {background:rgba(209,0,0,0.4);}\n\
+		.c2 {background:rgba(255,102,34,0.4);}\n\
+		.c3 {background:rgba(255,218,33,0.4);}\n\
+		.c4 {background:rgba(51,221,0,0.4);}\n\
+		.c5 {background:rgba(17,51,204,0.4);}\n\
+		.c6 {background:rgba(34,0,102,0.4);}\n\
+		.c7 {background:rgba(51,0,68,0.4);}\n\
+		.c8 {background:rgba(204,255,204,0.4);}\n\
+		.c9 {background:rgba(169,208,245,0.4);}\n\
+		.c10 {background:rgba(255,255,204,0.4);}\n\
+		.vt {transform:rotate(-60deg);transform-origin:0 0;}\n\
+		table.fstat {table-layout:fixed;padding:150px 15px 0 0;font-size:10px;column-width:30px;}\n\
+		.fstat th {width:55px;}\n\
+		.fstat td {text-align:left;width:35px;}\n\
+		.srccall {position:absolute;font-size:10px;z-index:7;overflow:hidden;color:black;text-align:center;white-space:nowrap;border-radius:5px;border:1px solid black;background:linear-gradient(to bottom right,#CCC,#969696);}\n\
+		.srccall:hover {color:white;font-weight:bold;border:1px solid white;}\n'
+	if(not embedded):
+		aslib.addCSS(hf, sysvals, 1, False, extra)
+
+	# write the device timeline
+	hf.write(devtl.html)
+
+	# add boot specific html
+	statinfo = 'var devstats = {\n'
+	for n in sorted(devstats):
+		statinfo += '\t"%s": [\n\t\t"%s",\n' % (n, devstats[n]['info'])
+		if 'fstat' in devstats[n]:
+			funcs = devstats[n]['fstat']
+			for f in sorted(funcs, key=funcs.get, reverse=True):
+				if funcs[f][0] < 0.01 and len(funcs) > 10:
+					break
+				statinfo += '\t\t"%f|%s|%d",\n' % (funcs[f][0], f, funcs[f][1])
+		statinfo += '\t],\n'
+	statinfo += '};\n'
+	html = \
+		'<div id="devicedetailtitle"></div>\n'\
+		'<div id="devicedetail" style="display:none;">\n'\
+		'<div id="devicedetail0">\n'\
+		'<div id="kernel_mode" class="phaselet" style="left:0%;width:100%;background:#DDDDDD"></div>\n'\
+		'</div>\n</div>\n'\
+		'<script type="text/javascript">\n'+statinfo+\
+		'</script>\n'
+	hf.write(html)
+
+	# add the callgraph html
+	if(sysvals.usecallgraph):
+		aslib.addCallgraphs(sysvals, hf, data)
+
+	# add the dmesg log as a hidden div
+	if sysvals.addlogs:
+		hf.write('<div id="dmesglog" style="display:none;">\n')
+		for line in data.dmesgtext:
+			line = line.replace('<', '&lt').replace('>', '&gt')
+			hf.write(line)
+		hf.write('</div>\n')
+
+	if(not embedded):
+		# write the footer and close
+		aslib.addScriptCode(hf, [data])
+		hf.write('</body>\n</html>\n')
+	else:
+		# embedded out will be loaded in a page, skip the js
+		hf.write('<div id=bounds style=display:none>%f,%f</div>' % \
+			(data.start*1000, data.initstart*1000))
+	hf.close()
+	return True
+
+# Function: updateCron
+# Description:
+#    (restore=False) Set the tool to run automatically on reboot
+#    (restore=True) Restore the original crontab
+def updateCron(restore=False):
+	if not restore:
+		sysvals.rootUser(True)
+	crondir = '/var/spool/cron/crontabs/'
+	cronfile = crondir+'root'
+	backfile = crondir+'root-analyze_boot-backup'
+	if not os.path.exists(crondir):
+		doError('%s not found' % crondir)
+	out = Popen(['which', 'crontab'], stdout=PIPE).stdout.read()
+	if not out:
+		doError('crontab not found')
+	# on restore: move the backup cron back into place
+	if restore:
+		if os.path.exists(backfile):
+			shutil.move(backfile, cronfile)
+		return
+	# backup current cron and install new one with reboot
+	if os.path.exists(cronfile):
+		shutil.move(cronfile, backfile)
+	else:
+		fp = open(backfile, 'w')
+		fp.close()
+	res = -1
+	try:
+		fp = open(backfile, 'r')
+		op = open(cronfile, 'w')
+		for line in fp:
+			if '@reboot' not in line:
+				op.write(line)
+				continue
+		fp.close()
+		op.write('@reboot python %s\n' % sysvals.cronjobCmdString())
+		op.close()
+		res = call('crontab %s' % cronfile, shell=True)
+	except Exception, e:
+		print 'Exception: %s' % str(e)
+		shutil.move(backfile, cronfile)
+		res = -1
+	if res != 0:
+		doError('crontab failed')
+
+# Function: updateGrub
+# Description:
+#	 update grub.cfg for all kernels with our parameters
+def updateGrub(restore=False):
+	# call update-grub on restore
+	if restore:
+		try:
+			call(['update-grub'], stderr=PIPE, stdout=PIPE,
+				env={'PATH': '.:/sbin:/usr/sbin:/usr/bin:/sbin:/bin'})
+		except Exception, e:
+			print 'Exception: %s\n' % str(e)
+		return
+	# verify we can do this
+	sysvals.rootUser(True)
+	grubfile = '/etc/default/grub'
+	if not os.path.exists(grubfile):
+		print 'ERROR: Unable to set the kernel parameters via grub.\n'
+		sysvals.manualRebootRequired()
+	out = Popen(['which', 'update-grub'], stdout=PIPE).stdout.read()
+	if not out:
+		print 'ERROR: Unable to set the kernel parameters via grub.\n'
+		sysvals.manualRebootRequired()
+
+	# extract the option and create a grub config without it
+	tgtopt = 'GRUB_CMDLINE_LINUX_DEFAULT'
+	cmdline = ''
+	tempfile = '/etc/default/grub.analyze_boot'
+	shutil.move(grubfile, tempfile)
+	res = -1
+	try:
+		fp = open(tempfile, 'r')
+		op = open(grubfile, 'w')
+		cont = False
+		for line in fp:
+			line = line.strip()
+			if len(line) == 0 or line[0] == '#':
+				continue
+			opt = line.split('=')[0].strip()
+			if opt == tgtopt:
+				cmdline = line.split('=', 1)[1].strip('\\')
+				if line[-1] == '\\':
+					cont = True
+			elif cont:
+				cmdline += line.strip('\\')
+				if line[-1] != '\\':
+					cont = False
+			else:
+				op.write('%s\n' % line)
+		fp.close()
+		# if the target option value is in quotes, strip them
+		sp = '"'
+		val = cmdline.strip()
+		if val[0] == '\'' or val[0] == '"':
+			sp = val[0]
+			val = val.strip(sp)
+		cmdline = val
+		# append our cmd line options
+		if len(cmdline) > 0:
+			cmdline += ' '
+		cmdline += sysvals.kernelParams()
+		# write out the updated target option
+		op.write('\n%s=%s%s%s\n' % (tgtopt, sp, cmdline, sp))
+		op.close()
+		res = call('update-grub')
+		os.remove(grubfile)
+	except Exception, e:
+		print 'Exception: %s' % str(e)
+		res = -1
+	# cleanup
+	shutil.move(tempfile, grubfile)
+	if res != 0:
+		doError('update-grub failed')
+
+# Function: doError
+# Description:
+#	 generic error function for catastrphic failures
+# Arguments:
+#	 msg: the error message to print
+#	 help: True if printHelp should be called after, False otherwise
+def doError(msg, help=False):
+	if help == True:
+		printHelp()
+	print 'ERROR: %s\n' % msg
+	sys.exit()
+
+# Function: printHelp
+# Description:
+#	 print out the help text
+def printHelp():
+	print('')
+	print('%s v%.1f' % (sysvals.title, sysvals.version))
+	print('Usage: bootgraph <options> <command>')
+	print('')
+	print('Description:')
+	print('  This tool reads in a dmesg log of linux kernel boot and')
+	print('  creates an html representation of the boot timeline up to')
+	print('  the start of the init process.')
+	print('')
+	print('  If no specific command is given the tool reads the current dmesg')
+	print('  and/or ftrace log and outputs bootgraph.html')
+	print('')
+	print('Options:')
+	print('  -h            Print this help text')
+	print('  -v            Print the current tool version')
+	print('  -addlogs      Add the dmesg log to the html output')
+	print('  -o file       Html timeline name (default: bootgraph.html)')
+	print(' [advanced]')
+	print('  -f            Use ftrace to add function detail (default: disabled)')
+	print('  -callgraph    Add callgraph detail, can be very large (default: disabled)')
+	print('  -maxdepth N   limit the callgraph data to N call levels (default: 2)')
+	print('  -mincg ms     Discard all callgraphs shorter than ms milliseconds (e.g. 0.001 for us)')
+	print('  -timeprec N   Number of significant digits in timestamps (0:S, 3:ms, [6:us])')
+	print('  -expandcg     pre-expand the callgraph data in the html output (default: disabled)')
+	print('  -filter list  Limit ftrace to comma-delimited list of functions (default: do_one_initcall)')
+	print(' [commands]')
+	print('  -reboot       Reboot the machine automatically and generate a new timeline')
+	print('  -manual       Show the requirements to generate a new timeline manually')
+	print('  -dmesg file   Load a stored dmesg file (used with -ftrace)')
+	print('  -ftrace file  Load a stored ftrace file (used with -dmesg)')
+	print('  -flistall     Print all functions capable of being captured in ftrace')
+	print('')
+	return True
+
+# ----------------- MAIN --------------------
+# exec start (skipped if script is loaded as library)
+if __name__ == '__main__':
+	# loop through the command line arguments
+	cmd = ''
+	simplecmds = ['-updategrub', '-flistall']
+	args = iter(sys.argv[1:])
+	for arg in args:
+		if(arg == '-h'):
+			printHelp()
+			sys.exit()
+		elif(arg == '-v'):
+			print("Version %.1f" % sysvals.version)
+			sys.exit()
+		elif(arg in simplecmds):
+			cmd = arg[1:]
+		elif(arg == '-f'):
+			sysvals.useftrace = True
+		elif(arg == '-callgraph'):
+			sysvals.useftrace = True
+			sysvals.usecallgraph = True
+		elif(arg == '-mincg'):
+			sysvals.mincglen = aslib.getArgFloat('-mincg', args, 0.0, 10000.0)
+		elif(arg == '-timeprec'):
+			sysvals.setPrecision(aslib.getArgInt('-timeprec', args, 0, 6))
+		elif(arg == '-maxdepth'):
+			sysvals.max_graph_depth = aslib.getArgInt('-maxdepth', args, 0, 1000)
+		elif(arg == '-filter'):
+			try:
+				val = args.next()
+			except:
+				doError('No filter functions supplied', True)
+			aslib.rootCheck(True)
+			sysvals.setGraphFilter(val)
+		elif(arg == '-ftrace'):
+			try:
+				val = args.next()
+			except:
+				doError('No ftrace file supplied', True)
+			if(os.path.exists(val) == False):
+				doError('%s does not exist' % val)
+			sysvals.ftracefile = val
+		elif(arg == '-addlogs'):
+			sysvals.addlogs = True
+		elif(arg == '-expandcg'):
+			sysvals.cgexp = True
+		elif(arg == '-dmesg'):
+			try:
+				val = args.next()
+			except:
+				doError('No dmesg file supplied', True)
+			if(os.path.exists(val) == False):
+				doError('%s does not exist' % val)
+			if(sysvals.htmlfile == val or sysvals.outfile == val):
+				doError('Output filename collision')
+			sysvals.dmesgfile = val
+		elif(arg == '-o'):
+			try:
+				val = args.next()
+			except:
+				doError('No HTML filename supplied', True)
+			if(sysvals.dmesgfile == val or sysvals.ftracefile == val):
+				doError('Output filename collision')
+			sysvals.htmlfile = val
+		elif(arg == '-reboot'):
+			if sysvals.iscronjob:
+				doError('-reboot and -cronjob are incompatible')
+			sysvals.reboot = True
+		elif(arg == '-manual'):
+			sysvals.reboot = True
+			sysvals.manual = True
+		# remaining options are only for cron job use
+		elif(arg == '-cronjob'):
+			sysvals.iscronjob = True
+			if sysvals.reboot:
+				doError('-reboot and -cronjob are incompatible')
+		else:
+			doError('Invalid argument: '+arg, True)
+
+	if cmd != '':
+		if cmd == 'updategrub':
+			updateGrub()
+		elif cmd == 'flistall':
+			sysvals.getFtraceFilterFunctions(False)
+		sys.exit()
+
+	# update grub, setup a cronjob, and reboot
+	if sysvals.reboot:
+		if not sysvals.manual:
+			updateGrub()
+			updateCron()
+			call('reboot')
+		else:
+			sysvals.manualRebootRequired()
+		sys.exit()
+
+	# disable the cronjob
+	if sysvals.iscronjob:
+		updateCron(True)
+		updateGrub(True)
+
+	data = loadKernelLog()
+	if sysvals.useftrace:
+		loadTraceLog(data)
+		if sysvals.iscronjob:
+			try:
+				sysvals.fsetVal('0', 'tracing_on')
+			except:
+				pass
+
+	if(sysvals.outfile and sysvals.phoronix):
+		fp = open(sysvals.outfile, 'w')
+		fp.write('pass %s initstart %.3f end %.3f boot %s\n' %
+			(data.valid, data.initstart*1000, data.end*1000, data.boottime))
+		fp.close()
+	if(not data.valid):
+		if sysvals.dmesgfile:
+			doError('No initcall data found in %s' % sysvals.dmesgfile)
+		else:
+			doError('No initcall data found, is initcall_debug enabled?')
+
+	print('          Host: %s' % sysvals.hostname)
+	print('     Test time: %s' % sysvals.testtime)
+	print('     Boot time: %s' % data.boottime)
+	print('Kernel Version: %s' % sysvals.kernel)
+	print('  Kernel start: %.3f' % (data.start * 1000))
+	print('    init start: %.3f' % (data.initstart * 1000))
+
+	createBootGraph(data, sysvals.phoronix)

tools/power/pm-graph/bootgraph.8

@@ -0,0 +1,132 @@
+.TH BOOTGRAPH 8
+.SH NAME
+bootgraph \- Kernel boot timing analysis
+.SH SYNOPSIS
+.ft B
+.B bootgraph
+.RB [ OPTIONS ]
+.RB [ COMMAND ]
+.SH DESCRIPTION
+\fBbootgraph \fP reads the dmesg log from kernel boot and
+creates an html representation of the initcall timeline up to the start
+of the init process.
+.PP
+If no specific command is given, the tool reads the current dmesg log and
+outputs bootgraph.html.
+.PP
+The tool can also augment the timeline with ftrace data on custom target
+functions as well as full trace callgraphs.
+.SH OPTIONS
+.TP
+\fB-h\fR
+Print this help text
+.TP
+\fB-v\fR
+Print the current tool version
+.TP
+\fB-addlogs\fR
+Add the dmesg log to the html output. It will be viewable by
+clicking a button in the timeline.
+.TP
+\fB-o \fIfile\fR
+Override the HTML output filename (default: bootgraph.html)
+.SS "Ftrace Debug"
+.TP
+\fB-f\fR
+Use ftrace to add function detail (default: disabled)
+.TP
+\fB-callgraph\fR
+Use ftrace to create initcall callgraphs (default: disabled). If -filter
+is not used there will be one callgraph per initcall. This can produce
+very large outputs, i.e. 10MB - 100MB.
+.TP
+\fB-maxdepth \fIlevel\fR
+limit the callgraph trace depth to \fIlevel\fR (default: 2). This is
+the best way to limit the output size when using -callgraph.
+.TP
+\fB-mincg \fIt\fR
+Discard all callgraphs shorter than \fIt\fR milliseconds (default: 0=all).
+This reduces the html file size as there can be many tiny callgraphs
+which are barely visible in the timeline.
+The value is a float: e.g. 0.001 represents 1 us.
+.TP
+\fB-timeprec \fIn\fR
+Number of significant digits in timestamps (0:S, 3:ms, [6:us])
+.TP
+\fB-expandcg\fR
+pre-expand the callgraph data in the html output (default: disabled)
+.TP
+\fB-filter \fI"func1,func2,..."\fR
+Instead of tracing each initcall, trace a custom list of functions (default: do_one_initcall)
+
+.SH COMMANDS
+.TP
+\fB-reboot\fR
+Reboot the machine and generate a new timeline automatically. Works in 4 steps.
+  1. updates grub with the required kernel parameters
+  2. installs a cron job which re-runs the tool after reboot
+  3. reboots the system
+  4. after startup, extracts the data and generates the timeline
+.TP
+\fB-manual\fR
+Show the requirements to generate a new timeline manually. Requires 3 steps.
+  1. append the string to the kernel command line via your native boot manager.
+  2. reboot the system
+  3. after startup, re-run the tool with the same arguments and no command
+.TP
+\fB-dmesg \fIfile\fR
+Create HTML output from an existing dmesg file.
+.TP
+\fB-ftrace \fIfile\fR
+Create HTML output from an existing ftrace file (used with -dmesg).
+.TP
+\fB-flistall\fR
+Print all ftrace functions capable of being captured. These are all the
+possible values you can add to trace via the -filter argument.
+
+.SH EXAMPLES
+Create a timeline using the current dmesg log.
+.IP
+\f(CW$ bootgraph\fR
+.PP
+Create a timeline using the current dmesg and ftrace log.
+.IP
+\f(CW$ bootgraph -callgraph\fR
+.PP
+Create a timeline using the current dmesg, add the log to the html and change the name.
+.IP
+\f(CW$ bootgraph -addlogs -o myboot.html\fR
+.PP
+Capture a new boot timeline by automatically rebooting the machine.
+.IP
+\f(CW$ sudo bootgraph -reboot -addlogs -o latestboot.html\fR
+.PP
+Capture a new boot timeline with function trace data.
+.IP
+\f(CW$ sudo bootgraph -reboot -f\fR
+.PP
+Capture a new boot timeline with trace & callgraph data. Skip callgraphs smaller than 5ms.
+.IP
+\f(CW$ sudo bootgraph -reboot -callgraph -mincg 5\fR
+.PP
+Capture a new boot timeline with callgraph data over custom functions.
+.IP
+\f(CW$ sudo bootgraph -reboot -callgraph -filter "acpi_ps_parse_aml,msleep"\fR
+.PP
+Capture a brand new boot timeline with manual reboot.
+.IP
+\f(CW$ sudo bootgraph -callgraph -manual\fR
+.IP
+\f(CW$ vi /etc/default/grub      # add the CMDLINE string to your kernel params\fR
+.IP
+\f(CW$ sudo reboot               # reboot the machine\fR
+.IP
+\f(CW$ sudo bootgraph -callgraph # re-run the tool after restart\fR
+.PP
+
+.SH "SEE ALSO"
+dmesg(1), update-grub(8), crontab(1), reboot(8)
+.PP
+.SH AUTHOR
+.nf
+Written by Todd Brandt <todd.e.brandt@linux.intel.com>

tools/power/pm-graph/sleepgraph.8

@@ -0,0 +1,243 @@
+.TH SLEEPGRAPH 8
+.SH NAME
+sleepgraph \- Suspend/Resume timing analysis
+.SH SYNOPSIS
+.ft B
+.B sleepgraph
+.RB [ OPTIONS ]
+.RB [ COMMAND ]
+.SH DESCRIPTION
+\fBsleepgraph \fP is designed to assist kernel and OS developers
+in optimizing their linux stack's suspend/resume time. Using a kernel
+image built with a few extra options enabled, the tool will execute a
+suspend and capture dmesg and ftrace data until resume is complete.
+This data is transformed into a device timeline and an optional
+callgraph to give a detailed view of which devices/subsystems are
+taking the most time in suspend/resume.
+.PP
+If no specific command is given, the default behavior is to initiate
+a suspend/resume.
+.PP
+Generates output files in subdirectory: suspend-yymmdd-HHMMSS
+   html timeline   :     <hostname>_<mode>.html
+   raw dmesg file  :     <hostname>_<mode>_dmesg.txt
+   raw ftrace file :     <hostname>_<mode>_ftrace.txt
+.SH OPTIONS
+.TP
+\fB-h\fR
+Print the help text.
+.TP
+\fB-v\fR
+Print the current tool version.
+.TP
+\fB-verbose\fR
+Print extra information during execution and analysis.
+.TP
+\fB-config \fIfile\fR
+Pull arguments and config options from a file.
+.TP
+\fB-m \fImode\fR
+Mode to initiate for suspend e.g. standby, freeze, mem (default: mem).
+.TP
+\fB-o \fIsubdir\fR
+Override the output subdirectory. Use {date}, {time}, {hostname} for current values.
+.sp
+e.g. suspend-{hostname}-{date}-{time}
+.TP
+\fB-rtcwake \fIt\fR | off
+Use rtcwake to autoresume after \fIt\fR seconds (default: 15). Set t to "off" to
+disable rtcwake and require a user keypress to resume.
+.TP
+\fB-addlogs\fR
+Add the dmesg and ftrace logs to the html output. They will be viewable by
+clicking buttons in the timeline.
+
+.SS "Advanced"
+.TP
+\fB-cmd \fIstr\fR
+Run the timeline over a custom suspend command, e.g. pm-suspend. By default
+the tool forces suspend via /sys/power/state so this allows testing over
+an OS's official suspend method. The output file will change to
+hostname_command.html and will autodetect which suspend mode was triggered.
+.TP
+\fB-filter \fI"d1,d2,..."\fR
+Filter out all but these device callbacks. These strings can be device names
+or module names. e.g. 0000:00:02.0, ata5, i915, usb, etc.
+.TP
+\fB-mindev \fIt\fR
+Discard all device callbacks shorter than \fIt\fR milliseconds (default: 0.0).
+This reduces the html file size as there can be many tiny callbacks which are barely
+visible. The value is a float: e.g. 0.001 represents 1 us.
+.TP
+\fB-proc\fR
+Add usermode process info into the timeline (default: disabled).
+.TP
+\fB-dev\fR
+Add kernel source calls and threads to the timeline (default: disabled).
+.TP
+\fB-x2\fR
+Run two suspend/resumes back to back (default: disabled).
+.TP
+\fB-x2delay \fIt\fR
+Include \fIt\fR ms delay between multiple test runs (default: 0 ms).
+.TP
+\fB-predelay \fIt\fR
+Include \fIt\fR ms delay before 1st suspend (default: 0 ms).
+.TP
+\fB-postdelay \fIt\fR
+Include \fIt\fR ms delay after last resume (default: 0 ms).
+.TP
+\fB-multi \fIn d\fR
+Execute \fIn\fR consecutive tests at \fId\fR seconds intervals. The outputs will
+be created in a new subdirectory with a summary page: suspend-xN-{date}-{time}.
+
+.SS "Ftrace Debug"
+.TP
+\fB-f\fR
+Use ftrace to create device callgraphs (default: disabled). This can produce
+very large outputs, i.e. 10MB - 100MB.
+.TP
+\fB-maxdepth \fIlevel\fR
+limit the callgraph trace depth to \fIlevel\fR (default: 0=all). This is
+the best way to limit the output size when using callgraphs via -f.
+.TP
+\fB-expandcg\fR
+pre-expand the callgraph data in the html output (default: disabled)
+.TP
+\fB-fadd \fIfile\fR
+Add functions to be graphed in the timeline from a list in a text file
+.TP
+\fB-mincg \fIt\fR
+Discard all callgraphs shorter than \fIt\fR milliseconds (default: 0.0).
+This reduces the html file size as there can be many tiny callgraphs
+which are barely visible in the timeline.
+The value is a float: e.g. 0.001 represents 1 us.
+.TP
+\fB-cgphase \fIp\fR
+Only show callgraph data for phase \fIp\fR (e.g. suspend_late).
+.TP
+\fB-cgtest \fIn\fR
+In an x2 run, only show callgraph data for test \fIn\fR (e.g. 0 or 1).
+.TP
+\fB-timeprec \fIn\fR
+Number of significant digits in timestamps (0:S, [3:ms], 6:us).
+
+.SH COMMANDS
+.TP
+\fB-ftrace \fIfile\fR
+Create HTML output from an existing ftrace file.
+.TP
+\fB-dmesg \fIfile\fR
+Create HTML output from an existing dmesg file.
+.TP
+\fB-summary \fIindir\fR
+Create a summary page of all tests in \fIindir\fR. Creates summary.html
+in the current folder. The output page is a table of tests with
+suspend and resume values sorted by suspend mode, host, and kernel.
+Includes test averages by mode and links to the test html files.
+.TP
+\fB-modes\fR
+List available suspend modes.
+.TP
+\fB-status\fR
+Test to see if the system is able to run this tool. Use this along
+with any options you intend to use to see if they will work.
+.TP
+\fB-fpdt\fR
+Print out the contents of the ACPI Firmware Performance Data Table.
+.TP
+\fB-usbtopo\fR
+Print out the current USB topology with power info.
+.TP
+\fB-usbauto\fR
+Enable autosuspend for all connected USB devices.
+.TP
+\fB-flist\fR
+Print the list of ftrace functions currently being captured. Functions
+that are not available as symbols in the current kernel are shown in red.
+By default, the tool traces a list of important suspend/resume functions
+in order to better fill out the timeline. If the user has added their own
+with -fadd they will also be checked.
+.TP
+\fB-flistall\fR
+Print all ftrace functions capable of being captured. These are all the
+possible values you can add to trace via the -fadd argument.
+
+.SH EXAMPLES
+.SS "Simple Commands"
+Check which suspend modes are currently supported.
+.IP
+\f(CW$ sleepgraph -modes\fR
+.PP
+Read the Firmware Performance Data Table (FPDT)
+.IP
+\f(CW$ sudo sleepgraph -fpdt\fR
+.PP
+Print out the current USB power topology
+.IP
+\f(CW$ sleepgraph -usbtopo
+.PP
+Verify that you can run a command with a set of arguments
+.IP
+\f(CW$ sudo sleepgraph -f -rtcwake 30 -status
+.PP
+Generate a summary of all timelines in a particular folder.
+.IP
+\f(CW$ sleepgraph -summary ~/workspace/myresults/\fR
+.PP
+Re-generate the html output from a previous run's dmesg and ftrace log.
+.IP
+\f(CW$ sleepgraph -dmesg myhost_mem_dmesg.txt -ftrace myhost_mem_ftrace.txt\fR
+.PP
+
+.SS "Capturing Simple Timelines"
+Execute a mem suspend with a 15 second wakeup. Include the logs in the html.
+.IP
+\f(CW$ sudo sleepgraph -rtcwake 15 -addlogs\fR
+.PP
+Execute a standby with a 15 second wakeup. Change the output folder name.
+.IP
+\f(CW$ sudo sleepgraph -m standby -rtcwake 15 -o "standby-{hostname}-{date}-{time}"\fR
+.PP
+Execute a freeze with no wakeup (require keypress). Change output folder name.
+.IP
+\f(CW$ sudo sleepgraph -m freeze -rtcwake off -o "freeze-{hostname}-{date}-{time}"\fR
+.PP
+
+.SS "Capturing Advanced Timelines"
+Execute a suspend & include dev mode source calls, limit callbacks to 5ms or larger.
+.IP
+\f(CW$ sudo sleepgraph -m mem -rtcwake 15 -dev -mindev 5\fR
+.PP
+Run two suspends back to back, include a 500ms delay before, after, and in between runs.
+.IP
+\f(CW$ sudo sleepgraph -m mem -rtcwake 15 -x2 -predelay 500 -x2delay 500 -postdelay 500\fR
+.PP
+Do a batch run of 10 freezes with 30 seconds delay between runs.
+.IP
+\f(CW$ sudo sleepgraph -m freeze -rtcwake 15 -multi 10 30\fR
+.PP
+Execute a suspend using a custom command.
+.IP
+\f(CW$ sudo sleepgraph -cmd "echo mem > /sys/power/state" -rtcwake 15\fR
+.PP
+
+
+.SS "Capturing Timelines with Callgraph Data"
+Add device callgraphs. Limit the trace depth and only show callgraphs 10ms or larger.
+.IP
+\f(CW$ sudo sleepgraph -m mem -rtcwake 15 -f -maxdepth 5 -mincg 10\fR
+.PP
+Capture a full callgraph across all suspend, then filter the html by a single phase.
+.IP
+\f(CW$ sudo sleepgraph -m mem -rtcwake 15 -f\fR
+.IP
+\f(CW$ sleepgraph -dmesg host_mem_dmesg.txt -ftrace host_mem_ftrace.txt -f -cgphase resume
+.PP
+
+.SH "SEE ALSO"
+dmesg(1)
+.PP
+.SH AUTHOR
+.nf
+Written by Todd Brandt <todd.e.brandt@linux.intel.com>

tools/power/x86/intel_pstate_tracer/intel_pstate_tracer.py

@@ -353,6 +353,14 @@ def split_csv():
                 os.system('grep -m 1 common_cpu cpu.csv > cpu{:0>3}.csv'.format(index))
                 os.system('grep CPU_{:0>3} cpu.csv >> cpu{:0>3}.csv'.format(index, index))
 
+def fix_ownership(path):
+    """Change the owner of the file to SUDO_UID, if required"""
+
+    uid = os.environ.get('SUDO_UID')
+    gid = os.environ.get('SUDO_GID')
+    if uid is not None:
+        os.chown(path, int(uid), int(gid))
+
 def cleanup_data_files():
     """ clean up existing data files """
 
@@ -518,12 +526,16 @@ else:
 
 if not os.path.exists('results'):
     os.mkdir('results')
+    # The regular user needs to own the directory, not root.
+    fix_ownership('results')
 
 os.chdir('results')
 if os.path.exists(testname):
     print('The test name directory already exists. Please provide a unique test name. Test re-run not supported, yet.')
     sys.exit()
 os.mkdir(testname)
+# The regular user needs to own the directory, not root.
+fix_ownership(testname)
 os.chdir(testname)
 
 # Temporary (or perhaps not)
@@ -566,4 +578,9 @@ plot_scaled_cpu()
 plot_boost_cpu()
 plot_ghz_cpu()
 
+# It is preferrable, but not necessary, that the regular user owns the files, not root.
+for root, dirs, files in os.walk('.'):
+    for f in files:
+        fix_ownership(f)
+
 os.chdir('../../')