strict_strtoul() is obsolete. Use kstrtoul() instead.
Otherwise we run into following build error:
CC drivers/cpufreq/cpufreq_interactive.o
drivers/cpufreq/cpufreq_interactive.c: In function ‘store_hispeed_freq’:
drivers/cpufreq/cpufreq_interactive.c:784:2: error: implicit declaration of function ‘strict_strtoul’ [-Werror=implicit-function-declaration]
cc1: some warnings being treated as errors
make[2]: *** [drivers/cpufreq/cpufreq_interactive.o] Error 1
Change-Id: Ib91b9df3af5fe2a244861c2f598bd20ec8115e6c
Signed-off-by: Amit Pundir <amit.pundir@linaro.org>
Interactive governor doesn't rearm per-cpu timer if target_freq is
equal to policy->max. However, this does not have clear performance
benefits. Profiling doesn't show any difference in benchmarks, games
or other workloads, if timers are always rearmed.
At same time, there are a few issues caused by not rearming timer
at policy->max.
1) min_sample_time enforcement is inconsistent
For target frequency that is lower than policy->max, it will not
drop until min_sample_time has passed since last frequency evaluation
selected current frequency. However, for policy->max, it will
always drop immediately as long as CPU has been run for longer than
min_sample_time. This is because timer is not running and thus
floor_freq and floor_validate_time is not updated.
Example: assume min_sample_time is 59ms and timer_rate is 20ms.
Frequency X < Y. Let's say CPU would pick the following frequencies
before accounting for min_sample_time in each 20ms sampling window.
Y, Y, Y, Y, X, X, X, X, X
If Y is not policy->max, the final target_freq after considering
min_sample_time will be Y, Y, Y, Y, *Y, *Y, X, X, X
* marks the windows where frequency is prevented from dropping.
If Y is policy->max, the final target_freq will be
Y, Y, Y, Y, X, X, X, X, X
2) Rearm timer in IDLE_START does not work as intended
IDLE_START/END is sent in arch_cpu_idle_enter/exit(). However, next
wake up is decided in tick_nohz_idle_enter(), which traverses the
timer list before idle notification is sent out. Therefore, rearming
timer in idle notification won't take effect until CPU wakes up at
least once. In rare scenarios when a CPU goes to idle and sleeps for a
long time immediately after a heavy load stops, it may not wake up
to drop its frequency vote for a long time, defeating the purpose of
having a slack_timer.
3) Need to rearm timer for policy->max change
commit 535a553fc1c4b4c3627c73214ade6326615a7463
(cpufreq: interactive: restructure CPUFREQ_GOV_LIMITS) mentions the
problem of timer getting indefinitely pushed back due to frequency
changes in policy->min/max. However, it still cancels and rearms timer
if policy->max is increased, and same problem could still happen if
policy->max is frequently changing after the fix. The best solution is
to always rearm timer for each CPU even if it's running at
policy->max.
Rearming timers even if target_freq is policy->max solves these
problems cleanly. It also simplifies the design and code of interactive
governor.
Change-Id: I973853d2375ea6f697fa4cee04a89efe6b8bf735
Reviewed-by: Saravana Kannan <skannan@codeaurora.org>
Signed-off-by: Junjie Wu <junjiew@codeaurora.org>
Signed-off-by: Rohit Gupta <rohgup@codeaurora.org>
min_sample_time needs to be cluster-based to match
above_hispeed_delay. If each CPU keeps making local decisions, it's
possible min_sample_time is not correctly enforced at cluster level,
which results in undesired frequency drops.
Change-Id: Ia2ec2ad9b7a8d715d4408c924d6762b7e532e4b4
Reviewed-by: Saravana Kannan <skannan@codeaurora.org>
Signed-off-by: Junjie Wu <junjiew@codeaurora.org>
If a heavy task migrates between otherwise idle CPUs in a policy during
every sample window, the above hispeed delay window for the CPUs would get
restarted for every sample window. Due to the continuous restart of above
hispeed delay window, none of the CPUs would ever pick a target frequency
higher than hispeed frequency. This causes the policy's frequency to be
stuck at hispeed freq even if the load justifies a higher frequency.
To fix this, the above high speed delay window is restarted only when the
policy frequency changes. This ensures that tasks migrating between CPUs in
a policy are handled correctly.
Also, the hispeed load/frequency heuristic is only necessary when the
information is insufficient to determine if the load on the CPU needs at
least hispeed frequency. When the policy frequency is already at or above
hispeed frequency, if the CPU load% based on policy frequency is not above
hispeed load, then the information is clearly sufficient to determine that
the load on the CPU does not need hispeed frequency.
Therefore, compute CPU load% (which is used only to compare against hispeed
load) based on policy frequency instead of CPU target frequency.
Change-Id: I8b5dfe6c50bee567a6719f0980e3f7757876ce4b
Signed-off-by: Saravana Kannan <skannan@codeaurora.org>
Signed-off-by: Junjie Wu <junjiew@codeaurora.org>
Timers are scheduled in unit of jiffies. Round up timer_rate so that
it matches the actual sampling period.
Change-Id: I88386a5a448e40333f9a9b9f0cf72af58cb54656
Signed-off-by: Junjie Wu <junjiew@codeaurora.org>
Frequency selection algorithm guarantees its chosen frequency
is not lower than hispeed_freq as long as boost is enabled.
Setting floor_freq and floor_validate_time during boost could block
CPU frequency from going below hispeed_freq even after
boostpulse_duration expires, if min_sample_time is higher than
boostpulse_duration. This conflicts with the intention of commit
de091367ead15b6e95dd1d0743a18f0da5a07ee5
(cpufreq: interactive: specify duration of CPU speed boost pulse)
to allow CPU to ramp down immediately after boost expires. It also
makes boost behavior inconsistent since it depends on min_sample_time.
Avoid setting floor_freq and floor_validate_time when boost starts.
Change-Id: I12852998af46cfbfaf8661eb5e8d5301b6f631e7
Signed-off-by: Junjie Wu <junjiew@codeaurora.org>
It is not correct to boost all the cpus when tunable boost
parameters are changed. It also does not need to boost the
cpus which is already boosted.
Signed-off-by: Lianwei Wang <a22439@motorola.com>
When __cpufreq_driver_target() in speedchange_task failed for some reason, the
policy->cur could be lower than the target_freq. The governor misses to change
the target_freq if the target_freq is equal to the next_freq at the next sample
time.
Added a check to prevent the CPU to stay at the speed that is lower than the
target_freq for long duration.
Change-Id: Ibfdcd193b8280390b8f8374a63218aa31267f310
Signed-off-by: Minsung Kim <ms925.kim@samsung.com>
When the load was below go_hispeed_load, there is a possibility that
choose_freq() would return a frequency which would be higher than the
hispeed_freq. According to the policy we should first jump to the
hispeed_freq, stay there for above_hispeed_delay and then be allowed to
raise higher than that.
Added a check to prevent the frequency to be directly raised to
something higher than the hispeed_freq.
Change-Id: Icda5d848dd9beadcc18835082ddf269732c75bd0
Signed-off-by: Ruchi Kandoi <kandoiruchi@google.com>
The cpufreq_interactive_timer gets cancelled and rescheduled
whenever the cpufreq_policy is changed. When the cpufreq policy is
changed at a rate faster than the sampling_rate of the interactive
governor, then the governor misses to change the target frequency
for long duration. The patch removes the need of cancelling the
timers when policy->min is changed.
Signed-off-by: Badhri Jagan Sridharan <Badhri@google.com>
Change-Id: Ibd98d151e1c73b8bd969484583ff98ee9f1135ef
2361be2366 changed cpufreq to add the
global cpufreq kobject to sysfs on demand.
To ensure this happens, cpufreq_interactive must hold a reference on
this object on devices where it intends to use it (i.e., devices where
have_governor_per_policy() returns false). Otherwise a parentless
kobject will be passed to sysfs_create_group() which will subsequently
BUG().
Change-Id: I7dd03956e1d3c6c3c0cc17c799882c235804ae09
Signed-off-by: Greg Hackmann <ghackmann@google.com>
Now that a generic version of get_cpu_idle_time() is available, use that for the interactive governor.
[toddpoynor@google.com: commit text changes]
Change-Id: Ia38b57085aac99ec3d415fe44471d5dfde519c2c
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Jon Medhurst <tixy@linaro.org>
Signed-off-by: John Stultz <john.stultz@linaro.org>
sysfs ops for target_loads and above_hispeed_delay can be called before
initializing tunables at CPUFREQ_GOV_POLICY_INIT. Create sysfs entries after
initialization.
Change-Id: I50356198d7629731c0d32a3066d61fe8354e0001
Signed-off-by: Minsung Kim <ms925.kim@samsung.com>
The gcc warns like:
cpufreq_interactive.c:745:6: warning: operation on 'ret' may be undefined [-Wsequence-point]
It was introduced by commit cf0fad49d17cb8273ce555dd5b7afab67d7923bf.
Since sprintf(...) just return 1 (one character) in this case, ret should not changed.
Just discarding the result of sprintf(...) leads to the result that
the committer of cf0fad49d17cb8273ce555dd5b7afab67d7923bf wants.
Change-Id: Ifed1cef6d6a31c3ed23dad03a567b3b9eddf3a57
Signed-off-by: Chih-Wei Huang <cwhuang@android-x86.org>
Make sure that timers cpu_timer and cpu_slack_timer
deactivated before addition of new.
Change-Id: If31c4049606871df6f00efdc24b1d713c86a6f69
Signed-off-by: Shridhar Rasal <srasal@nvidia.com>
Signed-off-by: Bharat Nihalani <bnihalani@nvidia.com>
If we have a multi-package system, where we have multiple instances of struct
policy (per package), currently we can't have multiple instances of same
governor. i.e. We can't have multiple instances of Interactive governor for
multiple packages.
This is a bottleneck for multicluster system, where we want different packages
to use Interactive governor, but with different tunables.
This patch uses the infrastructure provided by earlier patches pushed in
Mainline in v3.10-rc1/rc2 and implements per policy instances of Interactive
governor.
Change-Id: I70436d4a5a45c6cb6edf37f3e46d0b9fbc930982
[toddpoynor@google.com: merge with later code, minor changes]
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
This moves definition of cpufreq_gov_interactive towards the bottom of file, so
that we don't have to add prototype of cpufreq_governor_interactive() in the
beginning of file.
Change-Id: I04bd1004954eb36502c5cd7e35d3d7274cddaf95
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Cpufreq no longer calls governor callback for offlined cpus. i.e. All
policy->cpus are guaranteed to be online. Hence we don't need explicit check to
see if cpu is online or not.
Change-Id: I9ad85ea4addd5b4a40952e59ed730dd15e328690
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Remove a trailing whitespace from target_loads and above_hispeed_delay. Problem
happens when user-space program tried to restore parameters that saved before
changing parameters. In this case was returned error(EINVAL).
Change-Id: I5a74e3824602cd6f2b74651adda5ec1b627e61e9
Signed-off-by: Minsung Kim <ms925.kim@samsung.com>
When the policy max freq is raised, and before the timer is
rescheduled in idle callback, the cpu freq may stuck at a
lower freq.
The target_freq shall be updated too, else on a high load
situation, the new_freq is always equal to target_freq and
which will cause freq stuck at a lower freq too.
Reschedule the timer on gov limits callback.
Change-Id: I6c187001ab43e859731429b64f75a74eebc37a24
Signed-off-by: Lianwei Wang <a22439@motorola.com>
The cpufreq TRANSTION notifier callback does not check the
governor_enabled state on affected CPUS, which will case
kernel panic in update_load because the policy object maybe
NULL or invalid when governor_enabled is false.
Change-Id: Ie0f1718124f61e2f9b5da57abc6981ada5b83908
Signed-off-by: Lianwei Wang <a22439@motorola.com>
Check for idle time delta less than elapsed time delta, avoid
underflow computing active time.
Change-Id: I3e4c6ef1ad794eec49ed379c0c50fa727fd6ad28
Signed-off-by: Minsung Kim <ms925.kim@samsung.com>
Reschedule load sampling timer after timestamp of sample start taken,
hold spinlock across entire sequence to avoid preemption. Avoid the
WARN for zero time delta in the load sampling timer function.
Change-Id: Idc10a756f09141decb6df92669521a1ebf0dbc10
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Add checks for error return from cpufreq_frequency_table_target, and be
less noisy on the existing call with an error check. CPU hotplug and
system shutdown may cause this call to return -EINVAL.
Bug: 8613560
Change-Id: Id78d8829920462c0db1c7e14e717d91740d6cb44
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Time to wait should be based on the intended target speed, not the
actual speed (which may be held high by another CPU).
Change-Id: Ifc5bb55d06adddb9a02af90af05398a78f282272
Reported-by: Arve Hjønnevåg <arve@android.com>
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Previously the idle time returned from get_cpu_idle_time_us included the
iowait time. So the iowait time was always calculated as idle time.
But now the idle time returned from get_cpu_idle_time_us does not include
the iowait time anymore because of below commit which cause the iowait time
always calculated as busy time:
6beea0c nohz: Fix update_ts_time_stat idle accounting
Add the io_is_busy interface, as does the ondemand governor, and let the user
configure the iowait time as busy or idle through the io_is_busy sysfs
interface.
By default, io_is_busy is disabled.
[toddpoynor@google.com: minor updates]
Change-Id: If7d70ff864c43bc9c8d7fd7cfc66f930d339f9b4
Signed-off-by: Lianwei Wang <lian-wei.wang@motorola.com>
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Accept a string of delays and speeds at which to apply the delay before
raising each step above hispeed. For example, "80000 1300000:200000
1500000:40000" means that the delay at or above 1GHz, until 1.3GHz is 80 msecs,
the delay until 1.5GHz is 200 msecs and the delay at or above 1.5GHz is 40
msecs when hispeed_freq is 1GHz.
[toddpoynor@google.com: add documentation]
Change-Id: Ifeebede8b1acbdd0a53e5c6916bccbf764dc854f
Signed-off-by: Minsung Kim <ms925.kim@samsung.com>
There is race condition when both two cpu do CPUFREQ_GOV_STOP and one cpu
do CPUFREQ_GOV_START soon. The sysfs_remove_group is not done yet on one
cpu, but sysfs_create_group is called on another cpu, which cause governor
start failed and then kernel panic in timer callback because the policy and
cpu mask are all kfree in cpufreq driver.
Replace atomic with mutex to lock the whole START/STOP sequence.
Change-Id: I3762b3d44315ae021b8275aca84f5ea9147cc540
Signed-off-by: Lianwei Wang <a22439@motorola.com>
Need to use irqsave/restore spinlock calls to avoid a deadlock in calls
from the timer.
Change-Id: I15b6b590045ba1447e34ca7b5ff342723e53a605
Signed-off-by: Todd Poynor <toddpoynor@google.com>
If multiple governors are in use then avoid processing frequency transition
notifications for CPUs on which the interactive governor is not enabled.
Change-Id: Ibd75255b921d887501a64774a8c4f62302f2d4e4
Reported-by: Francisco Franco <francisco.franco@cloudcar.com>
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Update default go_hispeed_load from 85% to 99%. Recent changes to the
governor now use a default target_load of 90%. go_hispeed_load should
not be lower than the target load for hispeed_freq, which could lead
to oscillating speed decisions. Other recent changes reduce the need
to dampen speed jumps on load spikes, while input event boosts from
userspace are the preferred method for anticipating load spikes with
UI impacts.
General update to the documentation to reflect recent changes.
Change-Id: I1b92f3091f42c04b10503cd1169a943b5dfd6faf
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Starting the governor, or restarting on a hotplugged-in CPU, can race
with the timer start in idle, triggering a BUG on timer already pending.
Start the timer before setting the enable flag, and use enable_sem to
protect the sequence (and ensure correct order of the update to the
enable flag). Delete any existing timer for safety.
Change-Id: Ife77cf9fe099e8fd8543224cbf148c6722c2ffb0
Reported-by: Francisco Franco <francisco.franco@cloudcar.com>
Signed-off-by: Todd Poynor <toddpoynor@google.com>
When stopping the governor, del_timer_sync() can race against an
invocation of the idle notifier callback, which has the potential
to reactivate the timer.
To fix this issue, a read-write semaphore is used. Multiple readers are
allowed as long as pcpu->governor_enabled is true. However it can be
moved to false only after taking a write semaphore which would wait for
any on-going asynchronous activities to complete and prevent any more of
those activities to be initiated.
[toddpoynor@google.com: cosmetic and commit text changes]
Change-Id: Ib51165a735d73dcf964a06754c48bdc1913e13d0
Signed-off-by: Nicolas Pitre <nicolas.pitre@linaro.org>
35a84de cpufreq: interactive: apply above_hispeed_delay to each step above hispeed
caused the speed choice logic to osciallate between boosting and not boosting.
Add back code to ensure speed does not drop below boost frequency while
boosting.
Change-Id: Id420068480fcc7f5c4989ff523e2a8d22e2f4db2
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Always use deferrable timer for load sampling.
Set a non-deferrable timer to an additional slack time to allow prior to
waking up from idle to drop speed when not at minimum speed. Slack value
-1 avoids wakeups to drop speed. Default is 80ms.
Remove the governidle module param and its timer management in idle. For
platforms on which holding speed above mimum in idle costs power, use the
new timer slack to select how long to wait before waking up to drop speed.
Change-Id: I270b3980667e2c70a68e5bff534124b4411dbad5
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Sysfs attribute boostpulse_duration specifies the duration of boosting CPU
speed in response to bootpulse events. Duration is specified in usecs,
default 80ms.
Change-Id: Ifd41625574891a44f1787a4e85d1e7b4f2afb52b
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Add notifier for speed transitions. Keep a count of CPU active
microseconds times current frequency, converted to a percentage relative
to the current frequency when load is evaluated.
Change-Id: I5c27adb11081c50490219784ca57cc46e97fc28c
Signed-off-by: Todd Poynor <toddpoynor@google.com>
The longer-term load since last speed change isn't terribly useful,
may delay recognition of dropping load, and would need forthcoming
changes to adjust load for changing CPU speeds. Drop it.
Change-Id: Ic3cbb0542cc3484617031787e03ed9bdd632dec1
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Accept a string of target loads and speeds at which to apply the
target loads, per the documentation update in this patch. For example,
"85 1000000:90 1700000:99" targets CPU load 85% below speed 1GHz, 90%
at or above 1GHz, until 1.7GHz and above, at which load 99% is targeted.
Attempt to avoid oscillations by evaluating the current speed
weighted by current load against each new choice of speed, choosing a
higher speed if the current load requires a higher speed.
Change-Id: Ie3300206047c84eca5a26b0b63ea512e5207550e
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Apply above_hispeed_delay whenever increasing speed to a new speed above
hispeed (not just the first step above hispeed).
Change-Id: Ibb7add7db47f2a4306a9458c4e1ebabb60698636
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Add a target_load attribute that specifies how aggressively the governor is
to adjust speed to meet the observed load. New target speed is calculated
as the current actual speed (may be higher than target speed on SMP) times
the CPU load (as a fraction) divided by target load (fraction).
cpufreq_frequency_table_target() call use CPUFREQ_RELATION_L to set
the next higher speed rather than next lower speed.
Change-Id: If432451da82f5fed12e15c9421d7d27792376150
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Tracing adds actual speed since this is expected to be key to the
choice of target speed.
Change-Id: Iec936102d0010c4e9dfa143c38a9fd0d551189c3
Signed-off-by: Todd Poynor <toddpoynor@google.com>
The deferrable timer list isn't checked on all idle exits, such as when
hi-res timers expire or ISRs schedule workers. If the idle loop is
exited and it's past time to run the governor load polling timer,
run it immediately. This ensures we handle load spikes caused by actvity
that does not run the normal timer list.
Rename the field that timestamps the "time_in_idle" value to be more
accurate.
Change-Id: Ied590ecbefc83c9a9ec5eb9e31903557f6fa1614
Signed-off-by: Todd Poynor <toddpoynor@google.com>
