From 1c03a2d04d7ab6d27c1fef8614f08187d974bd21 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 2 Jun 2014 22:49:28 +0530 Subject: cpufreq: add support for intermediate (stable) frequencies Douglas Anderson, recently pointed out an interesting problem due to which udelay() was expiring earlier than it should. While transitioning between frequencies few platforms may temporarily switch to a stable frequency, waiting for the main PLL to stabilize. For example: When we transition between very low frequencies on exynos, like between 200MHz and 300MHz, we may temporarily switch to a PLL running at 800MHz. No CPUFREQ notification is sent for that. That means there's a period of time when we're running at 800MHz but loops_per_jiffy is calibrated at between 200MHz and 300MHz. And so udelay behaves badly. To get this fixed in a generic way, introduce another set of callbacks get_intermediate() and target_intermediate(), only for drivers with target_index() and CPUFREQ_ASYNC_NOTIFICATION unset. get_intermediate() should return a stable intermediate frequency platform wants to switch to, and target_intermediate() should set CPU to that frequency, before jumping to the frequency corresponding to 'index'. Core will take care of sending notifications and driver doesn't have to handle them in target_intermediate() or target_index(). NOTE: ->target_index() should restore to policy->restore_freq in case of failures as core would send notifications for that. Tested-by: Stephen Warren Signed-off-by: Viresh Kumar Reviewed-by: Doug Anderson Signed-off-by: Rafael J. Wysocki --- include/linux/cpufreq.h | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) (limited to 'include') diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 3f458896d45c..ec4112d257bc 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -75,6 +75,7 @@ struct cpufreq_policy { unsigned int max; /* in kHz */ unsigned int cur; /* in kHz, only needed if cpufreq * governors are used */ + unsigned int restore_freq; /* = policy->cur before transition */ unsigned int suspend_freq; /* freq to set during suspend */ unsigned int policy; /* see above */ @@ -221,11 +222,35 @@ struct cpufreq_driver { /* define one out of two */ int (*setpolicy) (struct cpufreq_policy *policy); + + /* + * On failure, should always restore frequency to policy->restore_freq + * (i.e. old freq). + */ int (*target) (struct cpufreq_policy *policy, /* Deprecated */ unsigned int target_freq, unsigned int relation); int (*target_index) (struct cpufreq_policy *policy, unsigned int index); + /* + * Only for drivers with target_index() and CPUFREQ_ASYNC_NOTIFICATION + * unset. + * + * get_intermediate should return a stable intermediate frequency + * platform wants to switch to and target_intermediate() should set CPU + * to to that frequency, before jumping to the frequency corresponding + * to 'index'. Core will take care of sending notifications and driver + * doesn't have to handle them in target_intermediate() or + * target_index(). + * + * Drivers can return '0' from get_intermediate() in case they don't + * wish to switch to intermediate frequency for some target frequency. + * In that case core will directly call ->target_index(). + */ + unsigned int (*get_intermediate)(struct cpufreq_policy *policy, + unsigned int index); + int (*target_intermediate)(struct cpufreq_policy *policy, + unsigned int index); /* should be defined, if possible */ unsigned int (*get) (unsigned int cpu); -- cgit v1.2.2 From bb3632c6101b2fad07e6246721466b984b1e0e9d Mon Sep 17 00:00:00 2001 From: Todd E Brandt Date: Fri, 6 Jun 2014 05:40:17 -0700 Subject: PM / sleep: trace events for suspend/resume Adds trace events that give finer resolution into suspend/resume. These events are graphed in the timelines generated by the analyze_suspend.py script. They represent large areas of time consumed that are typical to suspend and resume. The event is triggered by calling the function "trace_suspend_resume" with three arguments: a string (the name of the event to be displayed in the timeline), an integer (case specific number, such as the power state or cpu number), and a boolean (where true is used to denote the start of the timeline event, and false to denote the end). The suspend_resume trace event reproduces the data that the machine_suspend trace event did, so the latter has been removed. Signed-off-by: Todd Brandt Acked-by: Steven Rostedt Signed-off-by: Rafael J. Wysocki --- include/trace/events/power.h | 42 +++++++++++++++++++++++++----------------- 1 file changed, 25 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/include/trace/events/power.h b/include/trace/events/power.h index 9a7e08d61258..f88c8573e66c 100644 --- a/include/trace/events/power.h +++ b/include/trace/events/power.h @@ -7,6 +7,9 @@ #include #include #include +#include + +#define TPS(x) tracepoint_string(x) DECLARE_EVENT_CLASS(cpu, @@ -97,23 +100,6 @@ DEFINE_EVENT(cpu, cpu_frequency, TP_ARGS(frequency, cpu_id) ); -TRACE_EVENT(machine_suspend, - - TP_PROTO(unsigned int state), - - TP_ARGS(state), - - TP_STRUCT__entry( - __field( u32, state ) - ), - - TP_fast_assign( - __entry->state = state; - ), - - TP_printk("state=%lu", (unsigned long)__entry->state) -); - TRACE_EVENT(device_pm_report_time, TP_PROTO(struct device *dev, const char *pm_ops, s64 ops_time, @@ -151,6 +137,28 @@ TRACE_EVENT(device_pm_report_time, __entry->ops_time, __entry->error) ); +TRACE_EVENT(suspend_resume, + + TP_PROTO(const char *action, int val, bool start), + + TP_ARGS(action, val, start), + + TP_STRUCT__entry( + __field(const char *, action) + __field(int, val) + __field(bool, start) + ), + + TP_fast_assign( + __entry->action = action; + __entry->val = val; + __entry->start = start; + ), + + TP_printk("%s[%u] %s", __entry->action, (unsigned int)__entry->val, + (__entry->start)?"begin":"end") +); + DECLARE_EVENT_CLASS(wakeup_source, TP_PROTO(const char *name, unsigned int state), -- cgit v1.2.2 From e8bca479c3f269ebb3a3acea5ef63314bb677060 Mon Sep 17 00:00:00 2001 From: Todd E Brandt Date: Tue, 10 Jun 2014 07:31:22 -0700 Subject: PM / sleep: trace events for device PM callbacks Adds two trace events which supply the same info that initcall_debug provides, but via ftrace instead of dmesg. The existing initcall_debug calls require the pm_print_times_enabled var to be set (either via sysfs or via the kernel cmd line). The new trace events provide all the same info as the initcall_debug prints but with less overhead, and also with coverage of device prepare and complete device callbacks. These events replace the device_pm_report_time event (which has been removed). device_pm_callback_start is called first and provides the device and callback info. device_pm_callback_end is called after with the device name and error info. The time and pid are gathered from the trace data headers. Signed-off-by: Todd Brandt Signed-off-by: Rafael J. Wysocki --- include/trace/events/power.h | 60 +++++++++++++++++++++++++++++++------------- 1 file changed, 42 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/include/trace/events/power.h b/include/trace/events/power.h index f88c8573e66c..d19840b0cac8 100644 --- a/include/trace/events/power.h +++ b/include/trace/events/power.h @@ -93,6 +93,17 @@ TRACE_EVENT(pstate_sample, #define PWR_EVENT_EXIT -1 #endif +#define pm_verb_symbolic(event) \ + __print_symbolic(event, \ + { PM_EVENT_SUSPEND, "suspend" }, \ + { PM_EVENT_RESUME, "resume" }, \ + { PM_EVENT_FREEZE, "freeze" }, \ + { PM_EVENT_QUIESCE, "quiesce" }, \ + { PM_EVENT_HIBERNATE, "hibernate" }, \ + { PM_EVENT_THAW, "thaw" }, \ + { PM_EVENT_RESTORE, "restore" }, \ + { PM_EVENT_RECOVER, "recover" }) + DEFINE_EVENT(cpu, cpu_frequency, TP_PROTO(unsigned int frequency, unsigned int cpu_id), @@ -100,41 +111,54 @@ DEFINE_EVENT(cpu, cpu_frequency, TP_ARGS(frequency, cpu_id) ); -TRACE_EVENT(device_pm_report_time, +TRACE_EVENT(device_pm_callback_start, - TP_PROTO(struct device *dev, const char *pm_ops, s64 ops_time, - char *pm_event_str, int error), + TP_PROTO(struct device *dev, const char *pm_ops, int event), - TP_ARGS(dev, pm_ops, ops_time, pm_event_str, error), + TP_ARGS(dev, pm_ops, event), TP_STRUCT__entry( __string(device, dev_name(dev)) __string(driver, dev_driver_string(dev)) __string(parent, dev->parent ? dev_name(dev->parent) : "none") __string(pm_ops, pm_ops ? pm_ops : "none ") - __string(pm_event_str, pm_event_str) - __field(s64, ops_time) - __field(int, error) + __field(int, event) ), TP_fast_assign( - const char *tmp = dev->parent ? dev_name(dev->parent) : "none"; - const char *tmp_i = pm_ops ? pm_ops : "none "; + __assign_str(device, dev_name(dev)); + __assign_str(driver, dev_driver_string(dev)); + __assign_str(parent, + dev->parent ? dev_name(dev->parent) : "none"); + __assign_str(pm_ops, pm_ops ? pm_ops : "none "); + __entry->event = event; + ), + + TP_printk("%s %s, parent: %s, %s[%s]", __get_str(driver), + __get_str(device), __get_str(parent), __get_str(pm_ops), + pm_verb_symbolic(__entry->event)) +); + +TRACE_EVENT(device_pm_callback_end, + + TP_PROTO(struct device *dev, int error), + TP_ARGS(dev, error), + + TP_STRUCT__entry( + __string(device, dev_name(dev)) + __string(driver, dev_driver_string(dev)) + __field(int, error) + ), + + TP_fast_assign( __assign_str(device, dev_name(dev)); __assign_str(driver, dev_driver_string(dev)); - __assign_str(parent, tmp); - __assign_str(pm_ops, tmp_i); - __assign_str(pm_event_str, pm_event_str); - __entry->ops_time = ops_time; __entry->error = error; ), - /* ops_str has an extra space at the end */ - TP_printk("%s %s parent=%s state=%s ops=%snsecs=%lld err=%d", - __get_str(driver), __get_str(device), __get_str(parent), - __get_str(pm_event_str), __get_str(pm_ops), - __entry->ops_time, __entry->error) + TP_printk("%s %s, err=%d", + __get_str(driver), __get_str(device), __entry->error) ); TRACE_EVENT(suspend_resume, -- cgit v1.2.2