linux/drivers/cpufreq/cpufreq_governor.c

571 lines
16 KiB
C
Raw Normal View History

/*
* drivers/cpufreq/cpufreq_governor.c
*
* CPUFREQ governors common code
*
* Copyright (C) 2001 Russell King
* (C) 2003 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>.
* (C) 2003 Jun Nakajima <jun.nakajima@intel.com>
* (C) 2009 Alexander Clouter <alex@digriz.org.uk>
* (c) 2012 Viresh Kumar <viresh.kumar@linaro.org>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/export.h>
#include <linux/kernel_stat.h>
#include <linux/sched.h>
#include <linux/slab.h>
#include "cpufreq_governor.h"
static DEFINE_PER_CPU(struct cpu_dbs_info, cpu_dbs);
static DEFINE_MUTEX(gov_dbs_data_mutex);
/* Common sysfs tunables */
/**
* store_sampling_rate - update sampling rate effective immediately if needed.
*
* If new rate is smaller than the old, simply updating
* dbs.sampling_rate might not be appropriate. For example, if the
* original sampling_rate was 1 second and the requested new sampling rate is 10
* ms because the user needs immediate reaction from ondemand governor, but not
* sure if higher frequency will be required or not, then, the governor may
* change the sampling rate too late; up to 1 second later. Thus, if we are
* reducing the sampling rate, we need to make the new value effective
* immediately.
*
* This must be called with dbs_data->mutex held, otherwise traversing
* policy_dbs_list isn't safe.
*/
ssize_t store_sampling_rate(struct gov_attr_set *attr_set, const char *buf,
size_t count)
{
struct dbs_data *dbs_data = to_dbs_data(attr_set);
struct policy_dbs_info *policy_dbs;
unsigned int rate;
int ret;
ret = sscanf(buf, "%u", &rate);
if (ret != 1)
return -EINVAL;
dbs_data->sampling_rate = max(rate, dbs_data->min_sampling_rate);
/*
* We are operating under dbs_data->mutex and so the list and its
* entries can't be freed concurrently.
*/
list_for_each_entry(policy_dbs, &attr_set->policy_list, list) {
mutex_lock(&policy_dbs->timer_mutex);
/*
* On 32-bit architectures this may race with the
* sample_delay_ns read in dbs_update_util_handler(), but that
* really doesn't matter. If the read returns a value that's
* too big, the sample will be skipped, but the next invocation
* of dbs_update_util_handler() (when the update has been
* completed) will take a sample.
*
* If this runs in parallel with dbs_work_handler(), we may end
* up overwriting the sample_delay_ns value that it has just
* written, but it will be corrected next time a sample is
* taken, so it shouldn't be significant.
*/
gov_update_sample_delay(policy_dbs, 0);
mutex_unlock(&policy_dbs->timer_mutex);
}
return count;
}
EXPORT_SYMBOL_GPL(store_sampling_rate);
/**
* gov_update_cpu_data - Update CPU load data.
* @dbs_data: Top-level governor data pointer.
*
* Update CPU load data for all CPUs in the domain governed by @dbs_data
* (that may be a single policy or a bunch of them if governor tunables are
* system-wide).
*
* Call under the @dbs_data mutex.
*/
void gov_update_cpu_data(struct dbs_data *dbs_data)
{
struct policy_dbs_info *policy_dbs;
list_for_each_entry(policy_dbs, &dbs_data->attr_set.policy_list, list) {
unsigned int j;
for_each_cpu(j, policy_dbs->policy->cpus) {
struct cpu_dbs_info *j_cdbs = &per_cpu(cpu_dbs, j);
j_cdbs->prev_cpu_idle = get_cpu_idle_time(j, &j_cdbs->prev_update_time,
dbs_data->io_is_busy);
if (dbs_data->ignore_nice_load)
j_cdbs->prev_cpu_nice = kcpustat_cpu(j).cpustat[CPUTIME_NICE];
}
}
}
EXPORT_SYMBOL_GPL(gov_update_cpu_data);
unsigned int dbs_update(struct cpufreq_policy *policy)
{
struct policy_dbs_info *policy_dbs = policy->governor_data;
struct dbs_data *dbs_data = policy_dbs->dbs_data;
unsigned int ignore_nice = dbs_data->ignore_nice_load;
unsigned int max_load = 0;
unsigned int sampling_rate, io_busy, j;
/*
* Sometimes governors may use an additional multiplier to increase
* sample delays temporarily. Apply that multiplier to sampling_rate
* so as to keep the wake-up-from-idle detection logic a bit
* conservative.
*/
sampling_rate = dbs_data->sampling_rate * policy_dbs->rate_mult;
/*
* For the purpose of ondemand, waiting for disk IO is an indication
* that you're performance critical, and not that the system is actually
* idle, so do not add the iowait time to the CPU idle time then.
*/
io_busy = dbs_data->io_is_busy;
cpufreq: ondemand: Change the calculation of target frequency The ondemand governor calculates load in terms of frequency and increases it only if load_freq is greater than up_threshold multiplied by the current or average frequency. This appears to produce oscillations of frequency between min and max because, for example, a relatively small load can easily saturate minimum frequency and lead the CPU to the max. Then, it will decrease back to the min due to small load_freq. Change the calculation method of load and target frequency on the basis of the following two observations: - Load computation should not depend on the current or average measured frequency. For example, absolute load of 80% at 100MHz is not necessarily equivalent to 8% at 1000MHz in the next sampling interval. - It should be possible to increase the target frequency to any value present in the frequency table proportional to the absolute load, rather than to the max only, so that: Target frequency = C * load where we take C = policy->cpuinfo.max_freq / 100. Tested on Intel i7-3770 CPU @ 3.40GHz and on Quad core 1500MHz Krait. Phoronix benchmark of Linux Kernel Compilation 3.1 test shows an increase ~1.5% in performance. cpufreq_stats (time_in_state) shows that middle frequencies are used more, with this patch. Highest and lowest frequencies were used less by ~9%. [rjw: We have run multiple other tests on kernels with this change applied and in the vast majority of cases it turns out that the resulting performance improvement also leads to reduced consumption of energy. The change is additionally justified by the overall simplification of the code in question.] Signed-off-by: Stratos Karafotis <stratosk@semaphore.gr> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2013-06-05 16:01:25 +00:00
/* Get Absolute Load */
for_each_cpu(j, policy->cpus) {
struct cpu_dbs_info *j_cdbs = &per_cpu(cpu_dbs, j);
u64 update_time, cur_idle_time;
unsigned int idle_time, time_elapsed;
unsigned int load;
cur_idle_time = get_cpu_idle_time(j, &update_time, io_busy);
time_elapsed = update_time - j_cdbs->prev_update_time;
j_cdbs->prev_update_time = update_time;
idle_time = cur_idle_time - j_cdbs->prev_cpu_idle;
j_cdbs->prev_cpu_idle = cur_idle_time;
if (ignore_nice) {
u64 cur_nice = kcpustat_cpu(j).cpustat[CPUTIME_NICE];
idle_time += cputime_to_usecs(cur_nice - j_cdbs->prev_cpu_nice);
j_cdbs->prev_cpu_nice = cur_nice;
}
if (unlikely(!time_elapsed || time_elapsed < idle_time))
continue;
cpufreq: governor: Be friendly towards latency-sensitive bursty workloads Cpufreq governors like the ondemand governor calculate the load on the CPU periodically by employing deferrable timers. A deferrable timer won't fire if the CPU is completely idle (and there are no other timers to be run), in order to avoid unnecessary wakeups and thus save CPU power. However, the load calculation logic is agnostic to all this, and this can lead to the problem described below. Time (ms) CPU 1 100 Task-A running 110 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 110.5 Task-A running 120 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 125 Task-A went to sleep. With nothing else to do, CPU 1 went completely idle. 200 Task-A woke up and started running again. 200.5 Governor's deferred timer (which was originally programmed to fire at time 130) fires now. It calculates load for the time period 120 to 200.5, and finds the load is almost zero. Hence it decreases the CPU frequency to the minimum. 210 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. So, after the workload woke up and started running, the frequency was suddenly dropped to absolute minimum, and after that, there was an unnecessary delay of 10ms (sampling period) to increase the CPU frequency back to a reasonable value. And this pattern repeats for every wake-up-from-cpu-idle for that workload. This can be quite undesirable for latency- or response-time sensitive bursty workloads. So we need to fix the governor's logic to detect such wake-up-from- cpu-idle scenarios and start the workload at a reasonably high CPU frequency. One extreme solution would be to fake a load of 100% in such scenarios. But that might lead to undesirable side-effects such as frequency spikes (which might also need voltage changes) especially if the previous frequency happened to be very low. We just want to avoid the stupidity of dropping down the frequency to a minimum and then enduring a needless (and long) delay before ramping it up back again. So, let us simply carry forward the previous load - that is, let us just pretend that the 'load' for the current time-window is the same as the load for the previous window. That way, the frequency and voltage will continue to be set to whatever values they were set at previously. This means that bursty workloads will get a chance to influence the CPU frequency at which they wake up from cpu-idle, based on their past execution history. Thus, they might be able to avoid suffering from slow wakeups and long response-times. However, we should take care not to over-do this. For example, such a "copy previous load" logic will benefit cases like this: (where # represents busy and . represents idle) ##########.........#########.........###########...........##########........ but it will be detrimental in cases like the one shown below, because it will retain the high frequency (copied from the previous interval) even in a mostly idle system: ##########.........#.................#.....................#............... (i.e., the workload finished and the remaining tasks are such that their busy periods are smaller than the sampling interval, which causes the timer to always get deferred. So, this will make the copy-previous-load logic copy the initial high load to subsequent idle periods over and over again, thus keeping the frequency high unnecessarily). So, we modify this copy-previous-load logic such that it is used only once upon every wakeup-from-idle. Thus if we have 2 consecutive idle periods, the previous load won't get blindly copied over; cpufreq will freshly evaluate the load in the second idle interval, thus ensuring that the system comes back to its normal state. [ The right way to solve this whole problem is to teach the CPU frequency governors to also track load on a per-task basis, not just a per-CPU basis, and then use both the data sources intelligently to set the appropriate frequency on the CPUs. But that involves redesigning the cpufreq subsystem, so this patch should make the situation bearable until then. ] Experimental results: +-------------------+ I ran a modified version of ebizzy (called 'sleeping-ebizzy') that sleeps in between its execution such that its total utilization can be a user-defined value, say 10% or 20% (higher the utilization specified, lesser the amount of sleeps injected). This ebizzy was run with a single-thread, tied to CPU 8. Behavior observed with tracing (sample taken from 40% utilization runs): ------------------------------------------------------------------------ Without patch: ~~~~~~~~~~~~~~ kworker/8:2-12137 416.335742: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.335744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.345741: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.345744: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.345746: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.355738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> <...>-40753 416.402202: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 416.502130: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40753 416.505738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.505739: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.505741: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.515739: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.515742: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.515744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy Observation: Ebizzy went idle at 416.402202, and started running again at 416.502130. But cpufreq noticed the long idle period, and dropped the frequency at 416.505739, only to increase it back again at 416.515742, realizing that the workload is in-fact CPU bound. Thus ebizzy needlessly ran at the lowest frequency for almost 13 milliseconds (almost 1 full sample period), and this pattern repeats on every sleep-wakeup. This could hurt latency-sensitive workloads quite a lot. With patch: ~~~~~~~~~~~ kworker/8:2-29802 464.832535: cpu_frequency: state=2061000 cpu_id=8 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 464.962538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.972533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 464.972536: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-29802 464.972538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.982531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 465.022533: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.032531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.032532: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.035797: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 465.240178: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40738 465.242533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.242535: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.252531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 Observation: Ebizzy went idle at 465.035797, and started running again at 465.240178. Since ebizzy was the only real workload running on this CPU, cpufreq retained the frequency at 4.1Ghz throughout the run of ebizzy, no matter how many times ebizzy slept and woke-up in-between. Thus, ebizzy got the 10ms worth of 4.1 Ghz benefit during every sleep-wakeup (as compared to the run without the patch) and this boost gave a modest improvement in total throughput, as shown below. Sleeping-ebizzy records-per-second: ----------------------------------- Utilization Without patch With patch Difference (Absolute and % values) 10% 274767 277046 + 2279 (+0.829%) 20% 543429 553484 + 10055 (+1.850%) 40% 1090744 1107959 + 17215 (+1.578%) 60% 1634908 1662018 + 27110 (+1.658%) A rudimentary and somewhat approximately latency-sensitive workload such as sleeping-ebizzy itself showed a consistent, noticeable performance improvement with this patch. Hence, workloads that are truly latency-sensitive will benefit quite a bit from this change. Moreover, this is an overall win-win since this patch does not hurt power-savings at all (because, this patch does not reduce the idle time or idle residency; and the high frequency of the CPU when it goes to cpu-idle does not affect/hurt the power-savings of deep idle states). Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-06-07 20:41:43 +00:00
/*
* If the CPU had gone completely idle, and a task just woke up
* on this CPU now, it would be unfair to calculate 'load' the
* usual way for this elapsed time-window, because it will show
* near-zero load, irrespective of how CPU intensive that task
* actually is. This is undesirable for latency-sensitive bursty
* workloads.
*
* To avoid this, we reuse the 'load' from the previous
* time-window and give this task a chance to start with a
* reasonably high CPU frequency. (However, we shouldn't over-do
* this copy, lest we get stuck at a high load (high frequency)
* for too long, even when the current system load has actually
* dropped down. So we perform the copy only once, upon the
* first wake-up from idle.)
*
* Detecting this situation is easy: the governor's utilization
* update handler would not have run during CPU-idle periods.
* Hence, an unusually large 'time_elapsed' (as compared to the
* sampling rate) indicates this scenario.
2014-06-09 08:51:24 +00:00
*
* prev_load can be zero in two cases and we must recalculate it
* for both cases:
* - during long idle intervals
* - explicitly set to zero
cpufreq: governor: Be friendly towards latency-sensitive bursty workloads Cpufreq governors like the ondemand governor calculate the load on the CPU periodically by employing deferrable timers. A deferrable timer won't fire if the CPU is completely idle (and there are no other timers to be run), in order to avoid unnecessary wakeups and thus save CPU power. However, the load calculation logic is agnostic to all this, and this can lead to the problem described below. Time (ms) CPU 1 100 Task-A running 110 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 110.5 Task-A running 120 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 125 Task-A went to sleep. With nothing else to do, CPU 1 went completely idle. 200 Task-A woke up and started running again. 200.5 Governor's deferred timer (which was originally programmed to fire at time 130) fires now. It calculates load for the time period 120 to 200.5, and finds the load is almost zero. Hence it decreases the CPU frequency to the minimum. 210 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. So, after the workload woke up and started running, the frequency was suddenly dropped to absolute minimum, and after that, there was an unnecessary delay of 10ms (sampling period) to increase the CPU frequency back to a reasonable value. And this pattern repeats for every wake-up-from-cpu-idle for that workload. This can be quite undesirable for latency- or response-time sensitive bursty workloads. So we need to fix the governor's logic to detect such wake-up-from- cpu-idle scenarios and start the workload at a reasonably high CPU frequency. One extreme solution would be to fake a load of 100% in such scenarios. But that might lead to undesirable side-effects such as frequency spikes (which might also need voltage changes) especially if the previous frequency happened to be very low. We just want to avoid the stupidity of dropping down the frequency to a minimum and then enduring a needless (and long) delay before ramping it up back again. So, let us simply carry forward the previous load - that is, let us just pretend that the 'load' for the current time-window is the same as the load for the previous window. That way, the frequency and voltage will continue to be set to whatever values they were set at previously. This means that bursty workloads will get a chance to influence the CPU frequency at which they wake up from cpu-idle, based on their past execution history. Thus, they might be able to avoid suffering from slow wakeups and long response-times. However, we should take care not to over-do this. For example, such a "copy previous load" logic will benefit cases like this: (where # represents busy and . represents idle) ##########.........#########.........###########...........##########........ but it will be detrimental in cases like the one shown below, because it will retain the high frequency (copied from the previous interval) even in a mostly idle system: ##########.........#.................#.....................#............... (i.e., the workload finished and the remaining tasks are such that their busy periods are smaller than the sampling interval, which causes the timer to always get deferred. So, this will make the copy-previous-load logic copy the initial high load to subsequent idle periods over and over again, thus keeping the frequency high unnecessarily). So, we modify this copy-previous-load logic such that it is used only once upon every wakeup-from-idle. Thus if we have 2 consecutive idle periods, the previous load won't get blindly copied over; cpufreq will freshly evaluate the load in the second idle interval, thus ensuring that the system comes back to its normal state. [ The right way to solve this whole problem is to teach the CPU frequency governors to also track load on a per-task basis, not just a per-CPU basis, and then use both the data sources intelligently to set the appropriate frequency on the CPUs. But that involves redesigning the cpufreq subsystem, so this patch should make the situation bearable until then. ] Experimental results: +-------------------+ I ran a modified version of ebizzy (called 'sleeping-ebizzy') that sleeps in between its execution such that its total utilization can be a user-defined value, say 10% or 20% (higher the utilization specified, lesser the amount of sleeps injected). This ebizzy was run with a single-thread, tied to CPU 8. Behavior observed with tracing (sample taken from 40% utilization runs): ------------------------------------------------------------------------ Without patch: ~~~~~~~~~~~~~~ kworker/8:2-12137 416.335742: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.335744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.345741: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.345744: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.345746: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.355738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> <...>-40753 416.402202: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 416.502130: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40753 416.505738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.505739: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.505741: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.515739: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.515742: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.515744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy Observation: Ebizzy went idle at 416.402202, and started running again at 416.502130. But cpufreq noticed the long idle period, and dropped the frequency at 416.505739, only to increase it back again at 416.515742, realizing that the workload is in-fact CPU bound. Thus ebizzy needlessly ran at the lowest frequency for almost 13 milliseconds (almost 1 full sample period), and this pattern repeats on every sleep-wakeup. This could hurt latency-sensitive workloads quite a lot. With patch: ~~~~~~~~~~~ kworker/8:2-29802 464.832535: cpu_frequency: state=2061000 cpu_id=8 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 464.962538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.972533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 464.972536: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-29802 464.972538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.982531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 465.022533: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.032531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.032532: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.035797: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 465.240178: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40738 465.242533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.242535: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.252531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 Observation: Ebizzy went idle at 465.035797, and started running again at 465.240178. Since ebizzy was the only real workload running on this CPU, cpufreq retained the frequency at 4.1Ghz throughout the run of ebizzy, no matter how many times ebizzy slept and woke-up in-between. Thus, ebizzy got the 10ms worth of 4.1 Ghz benefit during every sleep-wakeup (as compared to the run without the patch) and this boost gave a modest improvement in total throughput, as shown below. Sleeping-ebizzy records-per-second: ----------------------------------- Utilization Without patch With patch Difference (Absolute and % values) 10% 274767 277046 + 2279 (+0.829%) 20% 543429 553484 + 10055 (+1.850%) 40% 1090744 1107959 + 17215 (+1.578%) 60% 1634908 1662018 + 27110 (+1.658%) A rudimentary and somewhat approximately latency-sensitive workload such as sleeping-ebizzy itself showed a consistent, noticeable performance improvement with this patch. Hence, workloads that are truly latency-sensitive will benefit quite a bit from this change. Moreover, this is an overall win-win since this patch does not hurt power-savings at all (because, this patch does not reduce the idle time or idle residency; and the high frequency of the CPU when it goes to cpu-idle does not affect/hurt the power-savings of deep idle states). Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-06-07 20:41:43 +00:00
*/
if (unlikely(time_elapsed > 2 * sampling_rate &&
2014-06-09 08:51:24 +00:00
j_cdbs->prev_load)) {
cpufreq: governor: Be friendly towards latency-sensitive bursty workloads Cpufreq governors like the ondemand governor calculate the load on the CPU periodically by employing deferrable timers. A deferrable timer won't fire if the CPU is completely idle (and there are no other timers to be run), in order to avoid unnecessary wakeups and thus save CPU power. However, the load calculation logic is agnostic to all this, and this can lead to the problem described below. Time (ms) CPU 1 100 Task-A running 110 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 110.5 Task-A running 120 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 125 Task-A went to sleep. With nothing else to do, CPU 1 went completely idle. 200 Task-A woke up and started running again. 200.5 Governor's deferred timer (which was originally programmed to fire at time 130) fires now. It calculates load for the time period 120 to 200.5, and finds the load is almost zero. Hence it decreases the CPU frequency to the minimum. 210 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. So, after the workload woke up and started running, the frequency was suddenly dropped to absolute minimum, and after that, there was an unnecessary delay of 10ms (sampling period) to increase the CPU frequency back to a reasonable value. And this pattern repeats for every wake-up-from-cpu-idle for that workload. This can be quite undesirable for latency- or response-time sensitive bursty workloads. So we need to fix the governor's logic to detect such wake-up-from- cpu-idle scenarios and start the workload at a reasonably high CPU frequency. One extreme solution would be to fake a load of 100% in such scenarios. But that might lead to undesirable side-effects such as frequency spikes (which might also need voltage changes) especially if the previous frequency happened to be very low. We just want to avoid the stupidity of dropping down the frequency to a minimum and then enduring a needless (and long) delay before ramping it up back again. So, let us simply carry forward the previous load - that is, let us just pretend that the 'load' for the current time-window is the same as the load for the previous window. That way, the frequency and voltage will continue to be set to whatever values they were set at previously. This means that bursty workloads will get a chance to influence the CPU frequency at which they wake up from cpu-idle, based on their past execution history. Thus, they might be able to avoid suffering from slow wakeups and long response-times. However, we should take care not to over-do this. For example, such a "copy previous load" logic will benefit cases like this: (where # represents busy and . represents idle) ##########.........#########.........###########...........##########........ but it will be detrimental in cases like the one shown below, because it will retain the high frequency (copied from the previous interval) even in a mostly idle system: ##########.........#.................#.....................#............... (i.e., the workload finished and the remaining tasks are such that their busy periods are smaller than the sampling interval, which causes the timer to always get deferred. So, this will make the copy-previous-load logic copy the initial high load to subsequent idle periods over and over again, thus keeping the frequency high unnecessarily). So, we modify this copy-previous-load logic such that it is used only once upon every wakeup-from-idle. Thus if we have 2 consecutive idle periods, the previous load won't get blindly copied over; cpufreq will freshly evaluate the load in the second idle interval, thus ensuring that the system comes back to its normal state. [ The right way to solve this whole problem is to teach the CPU frequency governors to also track load on a per-task basis, not just a per-CPU basis, and then use both the data sources intelligently to set the appropriate frequency on the CPUs. But that involves redesigning the cpufreq subsystem, so this patch should make the situation bearable until then. ] Experimental results: +-------------------+ I ran a modified version of ebizzy (called 'sleeping-ebizzy') that sleeps in between its execution such that its total utilization can be a user-defined value, say 10% or 20% (higher the utilization specified, lesser the amount of sleeps injected). This ebizzy was run with a single-thread, tied to CPU 8. Behavior observed with tracing (sample taken from 40% utilization runs): ------------------------------------------------------------------------ Without patch: ~~~~~~~~~~~~~~ kworker/8:2-12137 416.335742: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.335744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.345741: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.345744: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.345746: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.355738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> <...>-40753 416.402202: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 416.502130: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40753 416.505738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.505739: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.505741: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.515739: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.515742: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.515744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy Observation: Ebizzy went idle at 416.402202, and started running again at 416.502130. But cpufreq noticed the long idle period, and dropped the frequency at 416.505739, only to increase it back again at 416.515742, realizing that the workload is in-fact CPU bound. Thus ebizzy needlessly ran at the lowest frequency for almost 13 milliseconds (almost 1 full sample period), and this pattern repeats on every sleep-wakeup. This could hurt latency-sensitive workloads quite a lot. With patch: ~~~~~~~~~~~ kworker/8:2-29802 464.832535: cpu_frequency: state=2061000 cpu_id=8 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 464.962538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.972533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 464.972536: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-29802 464.972538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.982531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 465.022533: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.032531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.032532: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.035797: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 465.240178: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40738 465.242533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.242535: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.252531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 Observation: Ebizzy went idle at 465.035797, and started running again at 465.240178. Since ebizzy was the only real workload running on this CPU, cpufreq retained the frequency at 4.1Ghz throughout the run of ebizzy, no matter how many times ebizzy slept and woke-up in-between. Thus, ebizzy got the 10ms worth of 4.1 Ghz benefit during every sleep-wakeup (as compared to the run without the patch) and this boost gave a modest improvement in total throughput, as shown below. Sleeping-ebizzy records-per-second: ----------------------------------- Utilization Without patch With patch Difference (Absolute and % values) 10% 274767 277046 + 2279 (+0.829%) 20% 543429 553484 + 10055 (+1.850%) 40% 1090744 1107959 + 17215 (+1.578%) 60% 1634908 1662018 + 27110 (+1.658%) A rudimentary and somewhat approximately latency-sensitive workload such as sleeping-ebizzy itself showed a consistent, noticeable performance improvement with this patch. Hence, workloads that are truly latency-sensitive will benefit quite a bit from this change. Moreover, this is an overall win-win since this patch does not hurt power-savings at all (because, this patch does not reduce the idle time or idle residency; and the high frequency of the CPU when it goes to cpu-idle does not affect/hurt the power-savings of deep idle states). Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-06-07 20:41:43 +00:00
load = j_cdbs->prev_load;
2014-06-09 08:51:24 +00:00
/*
* Perform a destructive copy, to ensure that we copy
* the previous load only once, upon the first wake-up
* from idle.
*/
j_cdbs->prev_load = 0;
cpufreq: governor: Be friendly towards latency-sensitive bursty workloads Cpufreq governors like the ondemand governor calculate the load on the CPU periodically by employing deferrable timers. A deferrable timer won't fire if the CPU is completely idle (and there are no other timers to be run), in order to avoid unnecessary wakeups and thus save CPU power. However, the load calculation logic is agnostic to all this, and this can lead to the problem described below. Time (ms) CPU 1 100 Task-A running 110 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 110.5 Task-A running 120 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 125 Task-A went to sleep. With nothing else to do, CPU 1 went completely idle. 200 Task-A woke up and started running again. 200.5 Governor's deferred timer (which was originally programmed to fire at time 130) fires now. It calculates load for the time period 120 to 200.5, and finds the load is almost zero. Hence it decreases the CPU frequency to the minimum. 210 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. So, after the workload woke up and started running, the frequency was suddenly dropped to absolute minimum, and after that, there was an unnecessary delay of 10ms (sampling period) to increase the CPU frequency back to a reasonable value. And this pattern repeats for every wake-up-from-cpu-idle for that workload. This can be quite undesirable for latency- or response-time sensitive bursty workloads. So we need to fix the governor's logic to detect such wake-up-from- cpu-idle scenarios and start the workload at a reasonably high CPU frequency. One extreme solution would be to fake a load of 100% in such scenarios. But that might lead to undesirable side-effects such as frequency spikes (which might also need voltage changes) especially if the previous frequency happened to be very low. We just want to avoid the stupidity of dropping down the frequency to a minimum and then enduring a needless (and long) delay before ramping it up back again. So, let us simply carry forward the previous load - that is, let us just pretend that the 'load' for the current time-window is the same as the load for the previous window. That way, the frequency and voltage will continue to be set to whatever values they were set at previously. This means that bursty workloads will get a chance to influence the CPU frequency at which they wake up from cpu-idle, based on their past execution history. Thus, they might be able to avoid suffering from slow wakeups and long response-times. However, we should take care not to over-do this. For example, such a "copy previous load" logic will benefit cases like this: (where # represents busy and . represents idle) ##########.........#########.........###########...........##########........ but it will be detrimental in cases like the one shown below, because it will retain the high frequency (copied from the previous interval) even in a mostly idle system: ##########.........#.................#.....................#............... (i.e., the workload finished and the remaining tasks are such that their busy periods are smaller than the sampling interval, which causes the timer to always get deferred. So, this will make the copy-previous-load logic copy the initial high load to subsequent idle periods over and over again, thus keeping the frequency high unnecessarily). So, we modify this copy-previous-load logic such that it is used only once upon every wakeup-from-idle. Thus if we have 2 consecutive idle periods, the previous load won't get blindly copied over; cpufreq will freshly evaluate the load in the second idle interval, thus ensuring that the system comes back to its normal state. [ The right way to solve this whole problem is to teach the CPU frequency governors to also track load on a per-task basis, not just a per-CPU basis, and then use both the data sources intelligently to set the appropriate frequency on the CPUs. But that involves redesigning the cpufreq subsystem, so this patch should make the situation bearable until then. ] Experimental results: +-------------------+ I ran a modified version of ebizzy (called 'sleeping-ebizzy') that sleeps in between its execution such that its total utilization can be a user-defined value, say 10% or 20% (higher the utilization specified, lesser the amount of sleeps injected). This ebizzy was run with a single-thread, tied to CPU 8. Behavior observed with tracing (sample taken from 40% utilization runs): ------------------------------------------------------------------------ Without patch: ~~~~~~~~~~~~~~ kworker/8:2-12137 416.335742: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.335744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.345741: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.345744: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.345746: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.355738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> <...>-40753 416.402202: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 416.502130: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40753 416.505738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.505739: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.505741: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.515739: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.515742: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.515744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy Observation: Ebizzy went idle at 416.402202, and started running again at 416.502130. But cpufreq noticed the long idle period, and dropped the frequency at 416.505739, only to increase it back again at 416.515742, realizing that the workload is in-fact CPU bound. Thus ebizzy needlessly ran at the lowest frequency for almost 13 milliseconds (almost 1 full sample period), and this pattern repeats on every sleep-wakeup. This could hurt latency-sensitive workloads quite a lot. With patch: ~~~~~~~~~~~ kworker/8:2-29802 464.832535: cpu_frequency: state=2061000 cpu_id=8 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 464.962538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.972533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 464.972536: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-29802 464.972538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.982531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 465.022533: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.032531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.032532: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.035797: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 465.240178: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40738 465.242533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.242535: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.252531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 Observation: Ebizzy went idle at 465.035797, and started running again at 465.240178. Since ebizzy was the only real workload running on this CPU, cpufreq retained the frequency at 4.1Ghz throughout the run of ebizzy, no matter how many times ebizzy slept and woke-up in-between. Thus, ebizzy got the 10ms worth of 4.1 Ghz benefit during every sleep-wakeup (as compared to the run without the patch) and this boost gave a modest improvement in total throughput, as shown below. Sleeping-ebizzy records-per-second: ----------------------------------- Utilization Without patch With patch Difference (Absolute and % values) 10% 274767 277046 + 2279 (+0.829%) 20% 543429 553484 + 10055 (+1.850%) 40% 1090744 1107959 + 17215 (+1.578%) 60% 1634908 1662018 + 27110 (+1.658%) A rudimentary and somewhat approximately latency-sensitive workload such as sleeping-ebizzy itself showed a consistent, noticeable performance improvement with this patch. Hence, workloads that are truly latency-sensitive will benefit quite a bit from this change. Moreover, this is an overall win-win since this patch does not hurt power-savings at all (because, this patch does not reduce the idle time or idle residency; and the high frequency of the CPU when it goes to cpu-idle does not affect/hurt the power-savings of deep idle states). Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-06-07 20:41:43 +00:00
} else {
load = 100 * (time_elapsed - idle_time) / time_elapsed;
cpufreq: governor: Be friendly towards latency-sensitive bursty workloads Cpufreq governors like the ondemand governor calculate the load on the CPU periodically by employing deferrable timers. A deferrable timer won't fire if the CPU is completely idle (and there are no other timers to be run), in order to avoid unnecessary wakeups and thus save CPU power. However, the load calculation logic is agnostic to all this, and this can lead to the problem described below. Time (ms) CPU 1 100 Task-A running 110 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 110.5 Task-A running 120 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 125 Task-A went to sleep. With nothing else to do, CPU 1 went completely idle. 200 Task-A woke up and started running again. 200.5 Governor's deferred timer (which was originally programmed to fire at time 130) fires now. It calculates load for the time period 120 to 200.5, and finds the load is almost zero. Hence it decreases the CPU frequency to the minimum. 210 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. So, after the workload woke up and started running, the frequency was suddenly dropped to absolute minimum, and after that, there was an unnecessary delay of 10ms (sampling period) to increase the CPU frequency back to a reasonable value. And this pattern repeats for every wake-up-from-cpu-idle for that workload. This can be quite undesirable for latency- or response-time sensitive bursty workloads. So we need to fix the governor's logic to detect such wake-up-from- cpu-idle scenarios and start the workload at a reasonably high CPU frequency. One extreme solution would be to fake a load of 100% in such scenarios. But that might lead to undesirable side-effects such as frequency spikes (which might also need voltage changes) especially if the previous frequency happened to be very low. We just want to avoid the stupidity of dropping down the frequency to a minimum and then enduring a needless (and long) delay before ramping it up back again. So, let us simply carry forward the previous load - that is, let us just pretend that the 'load' for the current time-window is the same as the load for the previous window. That way, the frequency and voltage will continue to be set to whatever values they were set at previously. This means that bursty workloads will get a chance to influence the CPU frequency at which they wake up from cpu-idle, based on their past execution history. Thus, they might be able to avoid suffering from slow wakeups and long response-times. However, we should take care not to over-do this. For example, such a "copy previous load" logic will benefit cases like this: (where # represents busy and . represents idle) ##########.........#########.........###########...........##########........ but it will be detrimental in cases like the one shown below, because it will retain the high frequency (copied from the previous interval) even in a mostly idle system: ##########.........#.................#.....................#............... (i.e., the workload finished and the remaining tasks are such that their busy periods are smaller than the sampling interval, which causes the timer to always get deferred. So, this will make the copy-previous-load logic copy the initial high load to subsequent idle periods over and over again, thus keeping the frequency high unnecessarily). So, we modify this copy-previous-load logic such that it is used only once upon every wakeup-from-idle. Thus if we have 2 consecutive idle periods, the previous load won't get blindly copied over; cpufreq will freshly evaluate the load in the second idle interval, thus ensuring that the system comes back to its normal state. [ The right way to solve this whole problem is to teach the CPU frequency governors to also track load on a per-task basis, not just a per-CPU basis, and then use both the data sources intelligently to set the appropriate frequency on the CPUs. But that involves redesigning the cpufreq subsystem, so this patch should make the situation bearable until then. ] Experimental results: +-------------------+ I ran a modified version of ebizzy (called 'sleeping-ebizzy') that sleeps in between its execution such that its total utilization can be a user-defined value, say 10% or 20% (higher the utilization specified, lesser the amount of sleeps injected). This ebizzy was run with a single-thread, tied to CPU 8. Behavior observed with tracing (sample taken from 40% utilization runs): ------------------------------------------------------------------------ Without patch: ~~~~~~~~~~~~~~ kworker/8:2-12137 416.335742: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.335744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.345741: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.345744: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.345746: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.355738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> <...>-40753 416.402202: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 416.502130: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40753 416.505738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.505739: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.505741: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.515739: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.515742: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.515744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy Observation: Ebizzy went idle at 416.402202, and started running again at 416.502130. But cpufreq noticed the long idle period, and dropped the frequency at 416.505739, only to increase it back again at 416.515742, realizing that the workload is in-fact CPU bound. Thus ebizzy needlessly ran at the lowest frequency for almost 13 milliseconds (almost 1 full sample period), and this pattern repeats on every sleep-wakeup. This could hurt latency-sensitive workloads quite a lot. With patch: ~~~~~~~~~~~ kworker/8:2-29802 464.832535: cpu_frequency: state=2061000 cpu_id=8 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 464.962538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.972533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 464.972536: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-29802 464.972538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.982531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 465.022533: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.032531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.032532: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.035797: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 465.240178: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40738 465.242533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.242535: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.252531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 Observation: Ebizzy went idle at 465.035797, and started running again at 465.240178. Since ebizzy was the only real workload running on this CPU, cpufreq retained the frequency at 4.1Ghz throughout the run of ebizzy, no matter how many times ebizzy slept and woke-up in-between. Thus, ebizzy got the 10ms worth of 4.1 Ghz benefit during every sleep-wakeup (as compared to the run without the patch) and this boost gave a modest improvement in total throughput, as shown below. Sleeping-ebizzy records-per-second: ----------------------------------- Utilization Without patch With patch Difference (Absolute and % values) 10% 274767 277046 + 2279 (+0.829%) 20% 543429 553484 + 10055 (+1.850%) 40% 1090744 1107959 + 17215 (+1.578%) 60% 1634908 1662018 + 27110 (+1.658%) A rudimentary and somewhat approximately latency-sensitive workload such as sleeping-ebizzy itself showed a consistent, noticeable performance improvement with this patch. Hence, workloads that are truly latency-sensitive will benefit quite a bit from this change. Moreover, this is an overall win-win since this patch does not hurt power-savings at all (because, this patch does not reduce the idle time or idle residency; and the high frequency of the CPU when it goes to cpu-idle does not affect/hurt the power-savings of deep idle states). Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-06-07 20:41:43 +00:00
j_cdbs->prev_load = load;
}
if (load > max_load)
max_load = load;
}
return max_load;
}
EXPORT_SYMBOL_GPL(dbs_update);
cpufreq: governor: replace per-CPU delayed work with timers cpufreq governors evaluate load at sampling rate and based on that they update frequency for a group of CPUs belonging to the same cpufreq policy. This is required to be done in a single thread for all policy->cpus, but because we don't want to wakeup idle CPUs to do just that, we use deferrable work for this. If we would have used a single delayed deferrable work for the entire policy, there were chances that the CPU required to run the handler can be in idle and we might end up not changing the frequency for the entire group with load variations. And so we were forced to keep per-cpu works, and only the one that expires first need to do the real work and others are rescheduled for next sampling time. We have been using the more complex solution until now, where we used a delayed deferrable work for this, which is a combination of a timer and a work. This could be made lightweight by keeping per-cpu deferred timers with a single work item, which is scheduled by the first timer that expires. This patch does just that and here are important changes: - The timer handler will run in irq context and so we need to use a spin_lock instead of the timer_mutex. And so a separate timer_lock is created. This also makes the use of the mutex and lock quite clear, as we know what exactly they are protecting. - A new field 'skip_work' is added to track when the timer handlers can queue a work. More comments present in code. Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Ashwin Chaugule <ashwin.chaugule@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-09 02:04:42 +00:00
static void dbs_work_handler(struct work_struct *work)
{
struct policy_dbs_info *policy_dbs;
struct cpufreq_policy *policy;
struct dbs_governor *gov;
policy_dbs = container_of(work, struct policy_dbs_info, work);
policy = policy_dbs->policy;
gov = dbs_governor_of(policy);
cpufreq: governor: replace per-CPU delayed work with timers cpufreq governors evaluate load at sampling rate and based on that they update frequency for a group of CPUs belonging to the same cpufreq policy. This is required to be done in a single thread for all policy->cpus, but because we don't want to wakeup idle CPUs to do just that, we use deferrable work for this. If we would have used a single delayed deferrable work for the entire policy, there were chances that the CPU required to run the handler can be in idle and we might end up not changing the frequency for the entire group with load variations. And so we were forced to keep per-cpu works, and only the one that expires first need to do the real work and others are rescheduled for next sampling time. We have been using the more complex solution until now, where we used a delayed deferrable work for this, which is a combination of a timer and a work. This could be made lightweight by keeping per-cpu deferred timers with a single work item, which is scheduled by the first timer that expires. This patch does just that and here are important changes: - The timer handler will run in irq context and so we need to use a spin_lock instead of the timer_mutex. And so a separate timer_lock is created. This also makes the use of the mutex and lock quite clear, as we know what exactly they are protecting. - A new field 'skip_work' is added to track when the timer handlers can queue a work. More comments present in code. Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Ashwin Chaugule <ashwin.chaugule@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-09 02:04:42 +00:00
/*
* Make sure cpufreq_governor_limits() isn't evaluating load or the
* ondemand governor isn't updating the sampling rate in parallel.
cpufreq: governor: replace per-CPU delayed work with timers cpufreq governors evaluate load at sampling rate and based on that they update frequency for a group of CPUs belonging to the same cpufreq policy. This is required to be done in a single thread for all policy->cpus, but because we don't want to wakeup idle CPUs to do just that, we use deferrable work for this. If we would have used a single delayed deferrable work for the entire policy, there were chances that the CPU required to run the handler can be in idle and we might end up not changing the frequency for the entire group with load variations. And so we were forced to keep per-cpu works, and only the one that expires first need to do the real work and others are rescheduled for next sampling time. We have been using the more complex solution until now, where we used a delayed deferrable work for this, which is a combination of a timer and a work. This could be made lightweight by keeping per-cpu deferred timers with a single work item, which is scheduled by the first timer that expires. This patch does just that and here are important changes: - The timer handler will run in irq context and so we need to use a spin_lock instead of the timer_mutex. And so a separate timer_lock is created. This also makes the use of the mutex and lock quite clear, as we know what exactly they are protecting. - A new field 'skip_work' is added to track when the timer handlers can queue a work. More comments present in code. Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Ashwin Chaugule <ashwin.chaugule@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-09 02:04:42 +00:00
*/
mutex_lock(&policy_dbs->timer_mutex);
gov_update_sample_delay(policy_dbs, gov->gov_dbs_timer(policy));
mutex_unlock(&policy_dbs->timer_mutex);
cpufreq: governor: replace per-CPU delayed work with timers cpufreq governors evaluate load at sampling rate and based on that they update frequency for a group of CPUs belonging to the same cpufreq policy. This is required to be done in a single thread for all policy->cpus, but because we don't want to wakeup idle CPUs to do just that, we use deferrable work for this. If we would have used a single delayed deferrable work for the entire policy, there were chances that the CPU required to run the handler can be in idle and we might end up not changing the frequency for the entire group with load variations. And so we were forced to keep per-cpu works, and only the one that expires first need to do the real work and others are rescheduled for next sampling time. We have been using the more complex solution until now, where we used a delayed deferrable work for this, which is a combination of a timer and a work. This could be made lightweight by keeping per-cpu deferred timers with a single work item, which is scheduled by the first timer that expires. This patch does just that and here are important changes: - The timer handler will run in irq context and so we need to use a spin_lock instead of the timer_mutex. And so a separate timer_lock is created. This also makes the use of the mutex and lock quite clear, as we know what exactly they are protecting. - A new field 'skip_work' is added to track when the timer handlers can queue a work. More comments present in code. Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Ashwin Chaugule <ashwin.chaugule@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-09 02:04:42 +00:00
/* Allow the utilization update handler to queue up more work. */
atomic_set(&policy_dbs->work_count, 0);
/*
* If the update below is reordered with respect to the sample delay
* modification, the utilization update handler may end up using a stale
* sample delay value.
*/
smp_wmb();
policy_dbs->work_in_progress = false;
}
static void dbs_irq_work(struct irq_work *irq_work)
{
struct policy_dbs_info *policy_dbs;
cpufreq: governor: replace per-CPU delayed work with timers cpufreq governors evaluate load at sampling rate and based on that they update frequency for a group of CPUs belonging to the same cpufreq policy. This is required to be done in a single thread for all policy->cpus, but because we don't want to wakeup idle CPUs to do just that, we use deferrable work for this. If we would have used a single delayed deferrable work for the entire policy, there were chances that the CPU required to run the handler can be in idle and we might end up not changing the frequency for the entire group with load variations. And so we were forced to keep per-cpu works, and only the one that expires first need to do the real work and others are rescheduled for next sampling time. We have been using the more complex solution until now, where we used a delayed deferrable work for this, which is a combination of a timer and a work. This could be made lightweight by keeping per-cpu deferred timers with a single work item, which is scheduled by the first timer that expires. This patch does just that and here are important changes: - The timer handler will run in irq context and so we need to use a spin_lock instead of the timer_mutex. And so a separate timer_lock is created. This also makes the use of the mutex and lock quite clear, as we know what exactly they are protecting. - A new field 'skip_work' is added to track when the timer handlers can queue a work. More comments present in code. Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Ashwin Chaugule <ashwin.chaugule@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-09 02:04:42 +00:00
policy_dbs = container_of(irq_work, struct policy_dbs_info, irq_work);
schedule_work_on(smp_processor_id(), &policy_dbs->work);
cpufreq: governor: replace per-CPU delayed work with timers cpufreq governors evaluate load at sampling rate and based on that they update frequency for a group of CPUs belonging to the same cpufreq policy. This is required to be done in a single thread for all policy->cpus, but because we don't want to wakeup idle CPUs to do just that, we use deferrable work for this. If we would have used a single delayed deferrable work for the entire policy, there were chances that the CPU required to run the handler can be in idle and we might end up not changing the frequency for the entire group with load variations. And so we were forced to keep per-cpu works, and only the one that expires first need to do the real work and others are rescheduled for next sampling time. We have been using the more complex solution until now, where we used a delayed deferrable work for this, which is a combination of a timer and a work. This could be made lightweight by keeping per-cpu deferred timers with a single work item, which is scheduled by the first timer that expires. This patch does just that and here are important changes: - The timer handler will run in irq context and so we need to use a spin_lock instead of the timer_mutex. And so a separate timer_lock is created. This also makes the use of the mutex and lock quite clear, as we know what exactly they are protecting. - A new field 'skip_work' is added to track when the timer handlers can queue a work. More comments present in code. Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Ashwin Chaugule <ashwin.chaugule@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-09 02:04:42 +00:00
}
static void dbs_update_util_handler(struct update_util_data *data, u64 time,
unsigned long util, unsigned long max)
{
struct cpu_dbs_info *cdbs = container_of(data, struct cpu_dbs_info, update_util);
struct policy_dbs_info *policy_dbs = cdbs->policy_dbs;
u64 delta_ns, lst;
cpufreq: governor: replace per-CPU delayed work with timers cpufreq governors evaluate load at sampling rate and based on that they update frequency for a group of CPUs belonging to the same cpufreq policy. This is required to be done in a single thread for all policy->cpus, but because we don't want to wakeup idle CPUs to do just that, we use deferrable work for this. If we would have used a single delayed deferrable work for the entire policy, there were chances that the CPU required to run the handler can be in idle and we might end up not changing the frequency for the entire group with load variations. And so we were forced to keep per-cpu works, and only the one that expires first need to do the real work and others are rescheduled for next sampling time. We have been using the more complex solution until now, where we used a delayed deferrable work for this, which is a combination of a timer and a work. This could be made lightweight by keeping per-cpu deferred timers with a single work item, which is scheduled by the first timer that expires. This patch does just that and here are important changes: - The timer handler will run in irq context and so we need to use a spin_lock instead of the timer_mutex. And so a separate timer_lock is created. This also makes the use of the mutex and lock quite clear, as we know what exactly they are protecting. - A new field 'skip_work' is added to track when the timer handlers can queue a work. More comments present in code. Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Ashwin Chaugule <ashwin.chaugule@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-09 02:04:42 +00:00
/*
* The work may not be allowed to be queued up right now.
* Possible reasons:
* - Work has already been queued up or is in progress.
* - It is too early (too little time from the previous sample).
cpufreq: governor: replace per-CPU delayed work with timers cpufreq governors evaluate load at sampling rate and based on that they update frequency for a group of CPUs belonging to the same cpufreq policy. This is required to be done in a single thread for all policy->cpus, but because we don't want to wakeup idle CPUs to do just that, we use deferrable work for this. If we would have used a single delayed deferrable work for the entire policy, there were chances that the CPU required to run the handler can be in idle and we might end up not changing the frequency for the entire group with load variations. And so we were forced to keep per-cpu works, and only the one that expires first need to do the real work and others are rescheduled for next sampling time. We have been using the more complex solution until now, where we used a delayed deferrable work for this, which is a combination of a timer and a work. This could be made lightweight by keeping per-cpu deferred timers with a single work item, which is scheduled by the first timer that expires. This patch does just that and here are important changes: - The timer handler will run in irq context and so we need to use a spin_lock instead of the timer_mutex. And so a separate timer_lock is created. This also makes the use of the mutex and lock quite clear, as we know what exactly they are protecting. - A new field 'skip_work' is added to track when the timer handlers can queue a work. More comments present in code. Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Ashwin Chaugule <ashwin.chaugule@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-12-09 02:04:42 +00:00
*/
if (policy_dbs->work_in_progress)
return;
/*
* If the reads below are reordered before the check above, the value
* of sample_delay_ns used in the computation may be stale.
*/
smp_rmb();
lst = READ_ONCE(policy_dbs->last_sample_time);
delta_ns = time - lst;
if ((s64)delta_ns < policy_dbs->sample_delay_ns)
return;
/*
* If the policy is not shared, the irq_work may be queued up right away
* at this point. Otherwise, we need to ensure that only one of the
* CPUs sharing the policy will do that.
*/
if (policy_dbs->is_shared) {
if (!atomic_add_unless(&policy_dbs->work_count, 1, 1))
return;
/*
* If another CPU updated last_sample_time in the meantime, we
* shouldn't be here, so clear the work counter and bail out.
*/
if (unlikely(lst != READ_ONCE(policy_dbs->last_sample_time))) {
atomic_set(&policy_dbs->work_count, 0);
return;
}
}
policy_dbs->last_sample_time = time;
policy_dbs->work_in_progress = true;
irq_work_queue(&policy_dbs->irq_work);
}
static void gov_set_update_util(struct policy_dbs_info *policy_dbs,
unsigned int delay_us)
{
struct cpufreq_policy *policy = policy_dbs->policy;
int cpu;
gov_update_sample_delay(policy_dbs, delay_us);
policy_dbs->last_sample_time = 0;
for_each_cpu(cpu, policy->cpus) {
struct cpu_dbs_info *cdbs = &per_cpu(cpu_dbs, cpu);
cpufreq_add_update_util_hook(cpu, &cdbs->update_util,
dbs_update_util_handler);
}
}
static inline void gov_clear_update_util(struct cpufreq_policy *policy)
{
int i;
for_each_cpu(i, policy->cpus)
cpufreq_remove_update_util_hook(i);
synchronize_sched();
}
static void gov_cancel_work(struct cpufreq_policy *policy)
{
struct policy_dbs_info *policy_dbs = policy->governor_data;
gov_clear_update_util(policy_dbs->policy);
irq_work_sync(&policy_dbs->irq_work);
cancel_work_sync(&policy_dbs->work);
atomic_set(&policy_dbs->work_count, 0);
policy_dbs->work_in_progress = false;
}
static struct policy_dbs_info *alloc_policy_dbs_info(struct cpufreq_policy *policy,
struct dbs_governor *gov)
{
struct policy_dbs_info *policy_dbs;
int j;
/* Allocate memory for per-policy governor data. */
policy_dbs = gov->alloc();
if (!policy_dbs)
return NULL;
policy_dbs->policy = policy;
mutex_init(&policy_dbs->timer_mutex);
atomic_set(&policy_dbs->work_count, 0);
init_irq_work(&policy_dbs->irq_work, dbs_irq_work);
INIT_WORK(&policy_dbs->work, dbs_work_handler);
/* Set policy_dbs for all CPUs, online+offline */
for_each_cpu(j, policy->related_cpus) {
struct cpu_dbs_info *j_cdbs = &per_cpu(cpu_dbs, j);
j_cdbs->policy_dbs = policy_dbs;
}
return policy_dbs;
}
static void free_policy_dbs_info(struct policy_dbs_info *policy_dbs,
struct dbs_governor *gov)
{
int j;
mutex_destroy(&policy_dbs->timer_mutex);
for_each_cpu(j, policy_dbs->policy->related_cpus) {
struct cpu_dbs_info *j_cdbs = &per_cpu(cpu_dbs, j);
j_cdbs->policy_dbs = NULL;
j_cdbs->update_util.func = NULL;
}
gov->free(policy_dbs);
}
static int cpufreq_governor_init(struct cpufreq_policy *policy)
{
struct dbs_governor *gov = dbs_governor_of(policy);
struct dbs_data *dbs_data;
struct policy_dbs_info *policy_dbs;
unsigned int latency;
int ret = 0;
/* State should be equivalent to EXIT */
if (policy->governor_data)
return -EBUSY;
policy_dbs = alloc_policy_dbs_info(policy, gov);
if (!policy_dbs)
return -ENOMEM;
/* Protect gov->gdbs_data against concurrent updates. */
mutex_lock(&gov_dbs_data_mutex);
dbs_data = gov->gdbs_data;
if (dbs_data) {
if (WARN_ON(have_governor_per_policy())) {
ret = -EINVAL;
goto free_policy_dbs_info;
}
policy_dbs->dbs_data = dbs_data;
policy->governor_data = policy_dbs;
gov_attr_set_get(&dbs_data->attr_set, &policy_dbs->list);
goto out;
}
dbs_data = kzalloc(sizeof(*dbs_data), GFP_KERNEL);
if (!dbs_data) {
ret = -ENOMEM;
goto free_policy_dbs_info;
}
gov_attr_set_init(&dbs_data->attr_set, &policy_dbs->list);
ret = gov->init(dbs_data, !policy->governor->initialized);
if (ret)
goto free_policy_dbs_info;
/* policy latency is in ns. Convert it to us first */
latency = policy->cpuinfo.transition_latency / 1000;
if (latency == 0)
latency = 1;
/* Bring kernel and HW constraints together */
dbs_data->min_sampling_rate = max(dbs_data->min_sampling_rate,
MIN_LATENCY_MULTIPLIER * latency);
dbs_data->sampling_rate = max(dbs_data->min_sampling_rate,
LATENCY_MULTIPLIER * latency);
if (!have_governor_per_policy())
gov->gdbs_data = dbs_data;
policy_dbs->dbs_data = dbs_data;
policy->governor_data = policy_dbs;
cpufreq: governor: New sysfs show/store callbacks for governor tunables The ondemand and conservative governors use the global-attr or freq-attr structures to represent sysfs attributes corresponding to their tunables (which of them is actually used depends on whether or not different policy objects can use the same governor with different tunables at the same time and, consequently, on where those attributes are located in sysfs). Unfortunately, in the freq-attr case, the standard cpufreq show/store sysfs attribute callbacks are applied to the governor tunable attributes and they always acquire the policy->rwsem lock before carrying out the operation. That may lead to an ABBA deadlock if governor tunable attributes are removed under policy->rwsem while one of them is being accessed concurrently (if sysfs attributes removal wins the race, it will wait for the access to complete with policy->rwsem held while the attribute callback will block on policy->rwsem indefinitely). We attempted to address this issue by dropping policy->rwsem around governor tunable attributes removal (that is, around invocations of the ->governor callback with the event arg equal to CPUFREQ_GOV_POLICY_EXIT) in cpufreq_set_policy(), but that opened up race conditions that had not been possible with policy->rwsem held all the time. Therefore policy->rwsem cannot be dropped in cpufreq_set_policy() at any point, but the deadlock situation described above must be avoided too. To that end, use the observation that in principle governor tunables may be represented by the same data type regardless of whether the governor is system-wide or per-policy and introduce a new structure, struct governor_attr, for representing them and new corresponding macros for creating show/store sysfs callbacks for them. Also make their parent kobject use a new kobject type whose default show/store callbacks are not related to the standard core cpufreq ones in any way (and they don't acquire policy->rwsem in particular). Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Tested-by: Juri Lelli <juri.lelli@arm.com> Tested-by: Shilpasri G Bhat <shilpa.bhat@linux.vnet.ibm.com> [ rjw: Subject & changelog + rebase ] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-02-09 03:31:33 +00:00
gov->kobj_type.sysfs_ops = &governor_sysfs_ops;
ret = kobject_init_and_add(&dbs_data->attr_set.kobj, &gov->kobj_type,
cpufreq: governor: New sysfs show/store callbacks for governor tunables The ondemand and conservative governors use the global-attr or freq-attr structures to represent sysfs attributes corresponding to their tunables (which of them is actually used depends on whether or not different policy objects can use the same governor with different tunables at the same time and, consequently, on where those attributes are located in sysfs). Unfortunately, in the freq-attr case, the standard cpufreq show/store sysfs attribute callbacks are applied to the governor tunable attributes and they always acquire the policy->rwsem lock before carrying out the operation. That may lead to an ABBA deadlock if governor tunable attributes are removed under policy->rwsem while one of them is being accessed concurrently (if sysfs attributes removal wins the race, it will wait for the access to complete with policy->rwsem held while the attribute callback will block on policy->rwsem indefinitely). We attempted to address this issue by dropping policy->rwsem around governor tunable attributes removal (that is, around invocations of the ->governor callback with the event arg equal to CPUFREQ_GOV_POLICY_EXIT) in cpufreq_set_policy(), but that opened up race conditions that had not been possible with policy->rwsem held all the time. Therefore policy->rwsem cannot be dropped in cpufreq_set_policy() at any point, but the deadlock situation described above must be avoided too. To that end, use the observation that in principle governor tunables may be represented by the same data type regardless of whether the governor is system-wide or per-policy and introduce a new structure, struct governor_attr, for representing them and new corresponding macros for creating show/store sysfs callbacks for them. Also make their parent kobject use a new kobject type whose default show/store callbacks are not related to the standard core cpufreq ones in any way (and they don't acquire policy->rwsem in particular). Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Tested-by: Juri Lelli <juri.lelli@arm.com> Tested-by: Shilpasri G Bhat <shilpa.bhat@linux.vnet.ibm.com> [ rjw: Subject & changelog + rebase ] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-02-09 03:31:33 +00:00
get_governor_parent_kobj(policy),
"%s", gov->gov.name);
if (!ret)
goto out;
/* Failure, so roll back. */
cpufreq: governor: New sysfs show/store callbacks for governor tunables The ondemand and conservative governors use the global-attr or freq-attr structures to represent sysfs attributes corresponding to their tunables (which of them is actually used depends on whether or not different policy objects can use the same governor with different tunables at the same time and, consequently, on where those attributes are located in sysfs). Unfortunately, in the freq-attr case, the standard cpufreq show/store sysfs attribute callbacks are applied to the governor tunable attributes and they always acquire the policy->rwsem lock before carrying out the operation. That may lead to an ABBA deadlock if governor tunable attributes are removed under policy->rwsem while one of them is being accessed concurrently (if sysfs attributes removal wins the race, it will wait for the access to complete with policy->rwsem held while the attribute callback will block on policy->rwsem indefinitely). We attempted to address this issue by dropping policy->rwsem around governor tunable attributes removal (that is, around invocations of the ->governor callback with the event arg equal to CPUFREQ_GOV_POLICY_EXIT) in cpufreq_set_policy(), but that opened up race conditions that had not been possible with policy->rwsem held all the time. Therefore policy->rwsem cannot be dropped in cpufreq_set_policy() at any point, but the deadlock situation described above must be avoided too. To that end, use the observation that in principle governor tunables may be represented by the same data type regardless of whether the governor is system-wide or per-policy and introduce a new structure, struct governor_attr, for representing them and new corresponding macros for creating show/store sysfs callbacks for them. Also make their parent kobject use a new kobject type whose default show/store callbacks are not related to the standard core cpufreq ones in any way (and they don't acquire policy->rwsem in particular). Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Tested-by: Juri Lelli <juri.lelli@arm.com> Tested-by: Shilpasri G Bhat <shilpa.bhat@linux.vnet.ibm.com> [ rjw: Subject & changelog + rebase ] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-02-09 03:31:33 +00:00
pr_err("cpufreq: Governor initialization failed (dbs_data kobject init error %d)\n", ret);
cpufreq: Fix NULL reference crash while accessing policy->governor_data There is a race discovered by Juri, where we are able to: - create and read a sysfs file before policy->governor_data is being set to a non NULL value. OR - set policy->governor_data to NULL, and reading a file before being destroyed. And so such a crash is reported: Unable to handle kernel NULL pointer dereference at virtual address 0000000c pgd = edfc8000 [0000000c] *pgd=bfc8c835 Internal error: Oops: 17 [#1] SMP ARM Modules linked in: CPU: 4 PID: 1730 Comm: cat Not tainted 4.5.0-rc1+ #463 Hardware name: ARM-Versatile Express task: ee8e8480 ti: ee930000 task.ti: ee930000 PC is at show_ignore_nice_load_gov_pol+0x24/0x34 LR is at show+0x4c/0x60 pc : [<c058f1bc>] lr : [<c058ae88>] psr: a0070013 sp : ee931dd0 ip : ee931de0 fp : ee931ddc r10: ee4bc290 r9 : 00001000 r8 : ef2cb000 r7 : ee4bc200 r6 : ef2cb000 r5 : c0af57b0 r4 : ee4bc2e0 r3 : 00000000 r2 : 00000000 r1 : c0928df4 r0 : ef2cb000 Flags: NzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: adfc806a DAC: 00000051 Process cat (pid: 1730, stack limit = 0xee930210) Stack: (0xee931dd0 to 0xee932000) 1dc0: ee931dfc ee931de0 c058ae88 c058f1a4 1de0: edce3bc0 c07bfca4 edce3ac0 00001000 ee931e24 ee931e00 c01fcb90 c058ae48 1e00: 00000001 edce3bc0 00000000 00000001 ee931e50 ee8ff480 ee931e34 ee931e28 1e20: c01fb33c c01fcb0c ee931e8c ee931e38 c01a5210 c01fb314 ee931e9c ee931e48 1e40: 00000000 edce3bf0 befe4a00 ee931f78 00000000 00000000 000001e4 00000000 1e60: c00545a8 edce3ac0 00001000 00001000 befe4a00 ee931f78 00000000 00001000 1e80: ee931ed4 ee931e90 c01fbed8 c01a5038 ed085a58 00020000 00000000 00000000 1ea0: c0ad72e4 ee931f78 ee8ff488 ee8ff480 c077f3fc 00001000 befe4a00 ee931f78 1ec0: 00000000 00001000 ee931f44 ee931ed8 c017c328 c01fbdc4 00001000 00000000 1ee0: ee8ff480 00001000 ee931f44 ee931ef8 c017c65c c03deb10 ee931fac ee931f08 1f00: c0009270 c001f290 c0a8d968 ef2cb000 ef2cb000 ee8ff480 00000020 ee8ff480 1f20: ee8ff480 befe4a00 00001000 ee931f78 00000000 00000000 ee931f74 ee931f48 1f40: c017d1ec c017c2f8 c019c724 c019c684 ee8ff480 ee8ff480 00001000 befe4a00 1f60: 00000000 00000000 ee931fa4 ee931f78 c017d2a8 c017d160 00000000 00000000 1f80: 000a9f20 00001000 befe4a00 00000003 c000ffe4 ee930000 00000000 ee931fa8 1fa0: c000fe40 c017d264 000a9f20 00001000 00000003 befe4a00 00001000 00000000 Unable to handle kernel NULL pointer dereference at virtual address 0000000c 1fc0: 000a9f20 00001000 befe4a00 00000003 00000000 00000000 00000003 00000001 pgd = edfc4000 [0000000c] *pgd=bfcac835 1fe0: 00000000 befe49dc 000197f8 b6e35dfc 60070010 00000003 3065b49d 134ac2c9 [<c058f1bc>] (show_ignore_nice_load_gov_pol) from [<c058ae88>] (show+0x4c/0x60) [<c058ae88>] (show) from [<c01fcb90>] (sysfs_kf_seq_show+0x90/0xfc) [<c01fcb90>] (sysfs_kf_seq_show) from [<c01fb33c>] (kernfs_seq_show+0x34/0x38) [<c01fb33c>] (kernfs_seq_show) from [<c01a5210>] (seq_read+0x1e4/0x4e4) [<c01a5210>] (seq_read) from [<c01fbed8>] (kernfs_fop_read+0x120/0x1a0) [<c01fbed8>] (kernfs_fop_read) from [<c017c328>] (__vfs_read+0x3c/0xe0) [<c017c328>] (__vfs_read) from [<c017d1ec>] (vfs_read+0x98/0x104) [<c017d1ec>] (vfs_read) from [<c017d2a8>] (SyS_read+0x50/0x90) [<c017d2a8>] (SyS_read) from [<c000fe40>] (ret_fast_syscall+0x0/0x1c) Code: e5903044 e1a00001 e3081df4 e34c1092 (e593300c) ---[ end trace 5994b9a5111f35ee ]--- Fix that by making sure, policy->governor_data is updated at the right places only. Cc: 4.2+ <stable@vger.kernel.org> # 4.2+ Reported-and-tested-by: Juri Lelli <juri.lelli@arm.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-01-25 17:03:46 +00:00
policy->governor_data = NULL;
if (!have_governor_per_policy())
gov->gdbs_data = NULL;
gov->exit(dbs_data, !policy->governor->initialized);
kfree(dbs_data);
free_policy_dbs_info:
free_policy_dbs_info(policy_dbs, gov);
out:
mutex_unlock(&gov_dbs_data_mutex);
return ret;
}
static int cpufreq_governor_exit(struct cpufreq_policy *policy)
{
struct dbs_governor *gov = dbs_governor_of(policy);
struct policy_dbs_info *policy_dbs = policy->governor_data;
struct dbs_data *dbs_data = policy_dbs->dbs_data;
unsigned int count;
/* Protect gov->gdbs_data against concurrent updates. */
mutex_lock(&gov_dbs_data_mutex);
count = gov_attr_set_put(&dbs_data->attr_set, &policy_dbs->list);
policy->governor_data = NULL;
cpufreq: Fix NULL reference crash while accessing policy->governor_data There is a race discovered by Juri, where we are able to: - create and read a sysfs file before policy->governor_data is being set to a non NULL value. OR - set policy->governor_data to NULL, and reading a file before being destroyed. And so such a crash is reported: Unable to handle kernel NULL pointer dereference at virtual address 0000000c pgd = edfc8000 [0000000c] *pgd=bfc8c835 Internal error: Oops: 17 [#1] SMP ARM Modules linked in: CPU: 4 PID: 1730 Comm: cat Not tainted 4.5.0-rc1+ #463 Hardware name: ARM-Versatile Express task: ee8e8480 ti: ee930000 task.ti: ee930000 PC is at show_ignore_nice_load_gov_pol+0x24/0x34 LR is at show+0x4c/0x60 pc : [<c058f1bc>] lr : [<c058ae88>] psr: a0070013 sp : ee931dd0 ip : ee931de0 fp : ee931ddc r10: ee4bc290 r9 : 00001000 r8 : ef2cb000 r7 : ee4bc200 r6 : ef2cb000 r5 : c0af57b0 r4 : ee4bc2e0 r3 : 00000000 r2 : 00000000 r1 : c0928df4 r0 : ef2cb000 Flags: NzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: adfc806a DAC: 00000051 Process cat (pid: 1730, stack limit = 0xee930210) Stack: (0xee931dd0 to 0xee932000) 1dc0: ee931dfc ee931de0 c058ae88 c058f1a4 1de0: edce3bc0 c07bfca4 edce3ac0 00001000 ee931e24 ee931e00 c01fcb90 c058ae48 1e00: 00000001 edce3bc0 00000000 00000001 ee931e50 ee8ff480 ee931e34 ee931e28 1e20: c01fb33c c01fcb0c ee931e8c ee931e38 c01a5210 c01fb314 ee931e9c ee931e48 1e40: 00000000 edce3bf0 befe4a00 ee931f78 00000000 00000000 000001e4 00000000 1e60: c00545a8 edce3ac0 00001000 00001000 befe4a00 ee931f78 00000000 00001000 1e80: ee931ed4 ee931e90 c01fbed8 c01a5038 ed085a58 00020000 00000000 00000000 1ea0: c0ad72e4 ee931f78 ee8ff488 ee8ff480 c077f3fc 00001000 befe4a00 ee931f78 1ec0: 00000000 00001000 ee931f44 ee931ed8 c017c328 c01fbdc4 00001000 00000000 1ee0: ee8ff480 00001000 ee931f44 ee931ef8 c017c65c c03deb10 ee931fac ee931f08 1f00: c0009270 c001f290 c0a8d968 ef2cb000 ef2cb000 ee8ff480 00000020 ee8ff480 1f20: ee8ff480 befe4a00 00001000 ee931f78 00000000 00000000 ee931f74 ee931f48 1f40: c017d1ec c017c2f8 c019c724 c019c684 ee8ff480 ee8ff480 00001000 befe4a00 1f60: 00000000 00000000 ee931fa4 ee931f78 c017d2a8 c017d160 00000000 00000000 1f80: 000a9f20 00001000 befe4a00 00000003 c000ffe4 ee930000 00000000 ee931fa8 1fa0: c000fe40 c017d264 000a9f20 00001000 00000003 befe4a00 00001000 00000000 Unable to handle kernel NULL pointer dereference at virtual address 0000000c 1fc0: 000a9f20 00001000 befe4a00 00000003 00000000 00000000 00000003 00000001 pgd = edfc4000 [0000000c] *pgd=bfcac835 1fe0: 00000000 befe49dc 000197f8 b6e35dfc 60070010 00000003 3065b49d 134ac2c9 [<c058f1bc>] (show_ignore_nice_load_gov_pol) from [<c058ae88>] (show+0x4c/0x60) [<c058ae88>] (show) from [<c01fcb90>] (sysfs_kf_seq_show+0x90/0xfc) [<c01fcb90>] (sysfs_kf_seq_show) from [<c01fb33c>] (kernfs_seq_show+0x34/0x38) [<c01fb33c>] (kernfs_seq_show) from [<c01a5210>] (seq_read+0x1e4/0x4e4) [<c01a5210>] (seq_read) from [<c01fbed8>] (kernfs_fop_read+0x120/0x1a0) [<c01fbed8>] (kernfs_fop_read) from [<c017c328>] (__vfs_read+0x3c/0xe0) [<c017c328>] (__vfs_read) from [<c017d1ec>] (vfs_read+0x98/0x104) [<c017d1ec>] (vfs_read) from [<c017d2a8>] (SyS_read+0x50/0x90) [<c017d2a8>] (SyS_read) from [<c000fe40>] (ret_fast_syscall+0x0/0x1c) Code: e5903044 e1a00001 e3081df4 e34c1092 (e593300c) ---[ end trace 5994b9a5111f35ee ]--- Fix that by making sure, policy->governor_data is updated at the right places only. Cc: 4.2+ <stable@vger.kernel.org> # 4.2+ Reported-and-tested-by: Juri Lelli <juri.lelli@arm.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-01-25 17:03:46 +00:00
if (!count) {
if (!have_governor_per_policy())
gov->gdbs_data = NULL;
gov->exit(dbs_data, policy->governor->initialized == 1);
kfree(dbs_data);
}
free_policy_dbs_info(policy_dbs, gov);
mutex_unlock(&gov_dbs_data_mutex);
return 0;
}
static int cpufreq_governor_start(struct cpufreq_policy *policy)
{
struct dbs_governor *gov = dbs_governor_of(policy);
struct policy_dbs_info *policy_dbs = policy->governor_data;
struct dbs_data *dbs_data = policy_dbs->dbs_data;
unsigned int sampling_rate, ignore_nice, j;
unsigned int io_busy;
if (!policy->cur)
return -EINVAL;
policy_dbs->is_shared = policy_is_shared(policy);
policy_dbs->rate_mult = 1;
sampling_rate = dbs_data->sampling_rate;
ignore_nice = dbs_data->ignore_nice_load;
io_busy = dbs_data->io_is_busy;
for_each_cpu(j, policy->cpus) {
struct cpu_dbs_info *j_cdbs = &per_cpu(cpu_dbs, j);
j_cdbs->prev_cpu_idle = get_cpu_idle_time(j, &j_cdbs->prev_update_time, io_busy);
cpufreq: governor: Fix prev_load initialization in cpufreq_governor_start() The way cpufreq_governor_start() initializes j_cdbs->prev_load is questionable. First off, j_cdbs->prev_cpu_wall used as a denominator in the computation may be zero. The case this happens is when get_cpu_idle_time_us() returns -1 and get_cpu_idle_time_jiffy() used to return that number is called exactly at the jiffies_64 wrap time. It is rather hard to trigger that error, but it is not impossible and it will just crash the kernel then. Second, j_cdbs->prev_load is computed as the average load during the entire time since the system started and it may not reflect the load in the previous sampling period (as it is expected to). That doesn't play well with the way dbs_update() uses that value. Namely, if the update time delta (wall_time) happens do be greater than twice the sampling rate on the first invocation of it, the initial value of j_cdbs->prev_load (which may be completely off) will be returned to the caller as the current load (unless it is equal to zero and unless another CPU sharing the same policy object has a greater load value). For this reason, notice that the prev_load field of struct cpu_dbs_info is only used by dbs_update() and only in that one place, so if cpufreq_governor_start() is modified to always initialize it to 0, it will make dbs_update() always compute the actual load first time it checks the update time delta against the doubled sampling rate (after initialization) and there won't be any side effects of it. Consequently, modify cpufreq_governor_start() as described. Fixes: 18b46abd0009 (cpufreq: governor: Be friendly towards latency-sensitive bursty workloads) Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
2016-04-25 14:21:34 +00:00
/*
* Make the first invocation of dbs_update() compute the load.
*/
j_cdbs->prev_load = 0;
cpufreq: governor: Be friendly towards latency-sensitive bursty workloads Cpufreq governors like the ondemand governor calculate the load on the CPU periodically by employing deferrable timers. A deferrable timer won't fire if the CPU is completely idle (and there are no other timers to be run), in order to avoid unnecessary wakeups and thus save CPU power. However, the load calculation logic is agnostic to all this, and this can lead to the problem described below. Time (ms) CPU 1 100 Task-A running 110 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 110.5 Task-A running 120 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. 125 Task-A went to sleep. With nothing else to do, CPU 1 went completely idle. 200 Task-A woke up and started running again. 200.5 Governor's deferred timer (which was originally programmed to fire at time 130) fires now. It calculates load for the time period 120 to 200.5, and finds the load is almost zero. Hence it decreases the CPU frequency to the minimum. 210 Governor's timer fires, finds load as 100% in the last 10ms interval and increases the CPU frequency. So, after the workload woke up and started running, the frequency was suddenly dropped to absolute minimum, and after that, there was an unnecessary delay of 10ms (sampling period) to increase the CPU frequency back to a reasonable value. And this pattern repeats for every wake-up-from-cpu-idle for that workload. This can be quite undesirable for latency- or response-time sensitive bursty workloads. So we need to fix the governor's logic to detect such wake-up-from- cpu-idle scenarios and start the workload at a reasonably high CPU frequency. One extreme solution would be to fake a load of 100% in such scenarios. But that might lead to undesirable side-effects such as frequency spikes (which might also need voltage changes) especially if the previous frequency happened to be very low. We just want to avoid the stupidity of dropping down the frequency to a minimum and then enduring a needless (and long) delay before ramping it up back again. So, let us simply carry forward the previous load - that is, let us just pretend that the 'load' for the current time-window is the same as the load for the previous window. That way, the frequency and voltage will continue to be set to whatever values they were set at previously. This means that bursty workloads will get a chance to influence the CPU frequency at which they wake up from cpu-idle, based on their past execution history. Thus, they might be able to avoid suffering from slow wakeups and long response-times. However, we should take care not to over-do this. For example, such a "copy previous load" logic will benefit cases like this: (where # represents busy and . represents idle) ##########.........#########.........###########...........##########........ but it will be detrimental in cases like the one shown below, because it will retain the high frequency (copied from the previous interval) even in a mostly idle system: ##########.........#.................#.....................#............... (i.e., the workload finished and the remaining tasks are such that their busy periods are smaller than the sampling interval, which causes the timer to always get deferred. So, this will make the copy-previous-load logic copy the initial high load to subsequent idle periods over and over again, thus keeping the frequency high unnecessarily). So, we modify this copy-previous-load logic such that it is used only once upon every wakeup-from-idle. Thus if we have 2 consecutive idle periods, the previous load won't get blindly copied over; cpufreq will freshly evaluate the load in the second idle interval, thus ensuring that the system comes back to its normal state. [ The right way to solve this whole problem is to teach the CPU frequency governors to also track load on a per-task basis, not just a per-CPU basis, and then use both the data sources intelligently to set the appropriate frequency on the CPUs. But that involves redesigning the cpufreq subsystem, so this patch should make the situation bearable until then. ] Experimental results: +-------------------+ I ran a modified version of ebizzy (called 'sleeping-ebizzy') that sleeps in between its execution such that its total utilization can be a user-defined value, say 10% or 20% (higher the utilization specified, lesser the amount of sleeps injected). This ebizzy was run with a single-thread, tied to CPU 8. Behavior observed with tracing (sample taken from 40% utilization runs): ------------------------------------------------------------------------ Without patch: ~~~~~~~~~~~~~~ kworker/8:2-12137 416.335742: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.335744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.345741: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.345744: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.345746: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.355738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> <...>-40753 416.402202: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 416.502130: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40753 416.505738: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.505739: cpu_frequency: state=2061000 cpu_id=8 kworker/8:2-12137 416.505741: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40753 416.515739: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-12137 416.515742: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-12137 416.515744: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy Observation: Ebizzy went idle at 416.402202, and started running again at 416.502130. But cpufreq noticed the long idle period, and dropped the frequency at 416.505739, only to increase it back again at 416.515742, realizing that the workload is in-fact CPU bound. Thus ebizzy needlessly ran at the lowest frequency for almost 13 milliseconds (almost 1 full sample period), and this pattern repeats on every sleep-wakeup. This could hurt latency-sensitive workloads quite a lot. With patch: ~~~~~~~~~~~ kworker/8:2-29802 464.832535: cpu_frequency: state=2061000 cpu_id=8 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 464.962538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.972533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 464.972536: cpu_frequency: state=4123000 cpu_id=8 kworker/8:2-29802 464.972538: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 464.982531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 <snip> --------------------------------------------------------------------- <snip> kworker/8:2-29802 465.022533: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.032531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.032532: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.035797: sched_switch: prev_comm=ebizzy ==> next_comm=swapper/8 <idle>-0 465.240178: sched_switch: prev_comm=swapper/8 ==> next_comm=ebizzy <...>-40738 465.242533: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 kworker/8:2-29802 465.242535: sched_switch: prev_comm=kworker/8:2 ==> next_comm=ebizzy <...>-40738 465.252531: sched_switch: prev_comm=ebizzy ==> next_comm=kworker/8:2 Observation: Ebizzy went idle at 465.035797, and started running again at 465.240178. Since ebizzy was the only real workload running on this CPU, cpufreq retained the frequency at 4.1Ghz throughout the run of ebizzy, no matter how many times ebizzy slept and woke-up in-between. Thus, ebizzy got the 10ms worth of 4.1 Ghz benefit during every sleep-wakeup (as compared to the run without the patch) and this boost gave a modest improvement in total throughput, as shown below. Sleeping-ebizzy records-per-second: ----------------------------------- Utilization Without patch With patch Difference (Absolute and % values) 10% 274767 277046 + 2279 (+0.829%) 20% 543429 553484 + 10055 (+1.850%) 40% 1090744 1107959 + 17215 (+1.578%) 60% 1634908 1662018 + 27110 (+1.658%) A rudimentary and somewhat approximately latency-sensitive workload such as sleeping-ebizzy itself showed a consistent, noticeable performance improvement with this patch. Hence, workloads that are truly latency-sensitive will benefit quite a bit from this change. Moreover, this is an overall win-win since this patch does not hurt power-savings at all (because, this patch does not reduce the idle time or idle residency; and the high frequency of the CPU when it goes to cpu-idle does not affect/hurt the power-savings of deep idle states). Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-06-07 20:41:43 +00:00
if (ignore_nice)
j_cdbs->prev_cpu_nice = kcpustat_cpu(j).cpustat[CPUTIME_NICE];
}
gov->start(policy);
gov_set_update_util(policy_dbs, sampling_rate);
return 0;
}
static int cpufreq_governor_stop(struct cpufreq_policy *policy)
{
gov_cancel_work(policy);
return 0;
}
static int cpufreq_governor_limits(struct cpufreq_policy *policy)
{
struct policy_dbs_info *policy_dbs = policy->governor_data;
mutex_lock(&policy_dbs->timer_mutex);
if (policy->max < policy->cur)
__cpufreq_driver_target(policy, policy->max, CPUFREQ_RELATION_H);
else if (policy->min > policy->cur)
__cpufreq_driver_target(policy, policy->min, CPUFREQ_RELATION_L);
gov_update_sample_delay(policy_dbs, 0);
mutex_unlock(&policy_dbs->timer_mutex);
return 0;
}
int cpufreq_governor_dbs(struct cpufreq_policy *policy, unsigned int event)
{
if (event == CPUFREQ_GOV_POLICY_INIT) {
return cpufreq_governor_init(policy);
} else if (policy->governor_data) {
switch (event) {
case CPUFREQ_GOV_POLICY_EXIT:
return cpufreq_governor_exit(policy);
case CPUFREQ_GOV_START:
return cpufreq_governor_start(policy);
case CPUFREQ_GOV_STOP:
return cpufreq_governor_stop(policy);
case CPUFREQ_GOV_LIMITS:
return cpufreq_governor_limits(policy);
}
}
return -EINVAL;
}
EXPORT_SYMBOL_GPL(cpufreq_governor_dbs);