From: Youquan Song Date: Fri, 26 Oct 2012 10:26:50 +0000 (+0200) Subject: cpuidle: Quickly notice prediction failure in general case X-Git-Tag: next-20121205~51^2~7^2~10 X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=e11538d1f03914eb92af5a1a378375c05ae8520c;p=karo-tx-linux.git cpuidle: Quickly notice prediction failure in general case The prediction for future is difficult and when the cpuidle governor prediction fails and govenor possibly choose the shallower C-state than it should. How to quickly notice and find the failure becomes important for power saving. The patch extends to general case that prediction logic get a small predicted residency, so it choose a shallow C-state though the expected residency is large . Once the prediction will be fail, the CPU will keep staying at shallow C-state for a long time. Acutally, the CPU has change enter into deep C-state. So when the expected residency is long enough but governor choose a shallow C-state, an timer will be added in order to monitor if the prediction failure. When C-state is waken up prior to the adding timer, the timer will be cancelled initiatively. When the timer is triggered and menu governor will quickly notice prediction failure and re-evaluates deeper C-states possibility. Signed-off-by: Rik van Riel Signed-off-by: Youquan Song Signed-off-by: Rafael J. Wysocki --- diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c index 37c0ff6c805c..43a54fd6bfa2 100644 --- a/drivers/cpuidle/governors/menu.c +++ b/drivers/cpuidle/governors/menu.c @@ -34,7 +34,7 @@ static DEFINE_PER_CPU(struct hrtimer, menu_hrtimer); static DEFINE_PER_CPU(int, hrtimer_status); /* menu hrtimer mode */ -enum {MENU_HRTIMER_STOP, MENU_HRTIMER_REPEAT}; +enum {MENU_HRTIMER_STOP, MENU_HRTIMER_REPEAT, MENU_HRTIMER_GENERAL}; /* * Concepts and ideas behind the menu governor @@ -116,6 +116,13 @@ enum {MENU_HRTIMER_STOP, MENU_HRTIMER_REPEAT}; * */ +/* + * The C-state residency is so long that is is worthwhile to exit + * from the shallow C-state and re-enter into a deeper C-state. + */ +static unsigned int perfect_cstate_ms __read_mostly = 30; +module_param(perfect_cstate_ms, uint, 0000); + struct menu_device { int last_state_idx; int needs_update; @@ -216,6 +223,16 @@ EXPORT_SYMBOL_GPL(menu_hrtimer_cancel); static enum hrtimer_restart menu_hrtimer_notify(struct hrtimer *hrtimer) { int cpu = smp_processor_id(); + struct menu_device *data = &per_cpu(menu_devices, cpu); + + /* In general case, the expected residency is much larger than + * deepest C-state target residency, but prediction logic still + * predicts a small predicted residency, so the prediction + * history is totally broken if the timer is triggered. + * So reset the correction factor. + */ + if (per_cpu(hrtimer_status, cpu) == MENU_HRTIMER_GENERAL) + data->correction_factor[data->bucket] = RESOLUTION * DECAY; per_cpu(hrtimer_status, cpu) = MENU_HRTIMER_STOP; @@ -353,6 +370,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev) /* not deepest C-state chosen for low predicted residency */ if (low_predicted) { unsigned int timer_us = 0; + unsigned int perfect_us = 0; /* * Set a timer to detect whether this sleep is much @@ -363,12 +381,26 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev) */ timer_us = 2 * (data->predicted_us + MAX_DEVIATION); + perfect_us = perfect_cstate_ms * 1000; + if (repeat && (4 * timer_us < data->expected_us)) { hrtimer_start(hrtmr, ns_to_ktime(1000 * timer_us), HRTIMER_MODE_REL_PINNED); /* In repeat case, menu hrtimer is started */ per_cpu(hrtimer_status, cpu) = MENU_HRTIMER_REPEAT; + } else if (perfect_us < data->expected_us) { + /* + * The next timer is long. This could be because + * we did not make a useful prediction. + * In that case, it makes sense to re-enter + * into a deeper C-state after some time. + */ + hrtimer_start(hrtmr, ns_to_ktime(1000 * timer_us), + HRTIMER_MODE_REL_PINNED); + /* In general case, menu hrtimer is started */ + per_cpu(hrtimer_status, cpu) = MENU_HRTIMER_GENERAL; } + } return data->last_state_idx;