Related
Hi Guys,
i thought it would be a good idea to put all the infos of the zzmoove governor around here together on one place
for better way to find it, to have a place to dump stuff for future versions and to give support for specific questions.
for now i just copied the allready existend posts here but will edit this further when i have more time
so lets start with the first
initial version:
ZZMoove Governor v0.1
(post from 18th December 2012, 10:27 PM)
Why that “zzmoove” governor and how it works:
I thought it were pretty cool to have one of my favorite governors back from the old SGS1-days on my actual device, so i decided to take the sources and give that a try on top of boeffla kernel. It worked well so this is now the result of that experiment.
More about the internals:
Basically this is the ported SGS1 version of the well known "smoove" governor from the good old midnight kernel from Michael Weingaertner (mialwe) with a modified CPU hotplug implementation of the ktoonservative governor from ktoonesz. The original implementation from ktoonesz worked well but I observed that on idle most of the time only one cpu was going to sleep. Well that was not enough for me so I made a modification to put the other cpu's also to sleep (except cpu0). That means that this governor uses more often only one cpu on idle and as a consequence of that it needs less energy. Depending on System load and governor settings all 4 cores will be instantly up again if it is needed.
In short:
So what you can expect now from this thingy is a battery-friendly behaving hotplug conservative governor which
uses a frequency lookup table for faster upscaling (so called "smooth scaling") So this is more a energy-safer than a performer.
Tuneables/Defaults:
Sampling Rate (default=2) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/sampling_rate
Sampling Down Factor (default=4) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/sampling_down_factor
Up Threshold (default=70) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold
Up Threshold Hotplug (default=68) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold_hotplug
Down Threshold (default=52) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold
Down Threshold Hotplug (default=55) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold_hotplug
Ignore Nice Load (default=0) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/ignore_nice_load
Freqency Step (default=5) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/freq_step
Smooth Up (default=75) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/smooth_up
Links:
Midnight kernel
KT747 Kernel
Common Infos about Governors,I/O Schedulers etc.
Credits to:
mialwe for his smoove governor
ktoonesz for original hotplug implementation
ZZMoove Governor v0.3
(Post from 25th February 2013, 05:06 PM - "more improvements")
there are now many new possibilities to adjust the governor more precisely via sysfs!
Following new tuneables were introduced in this new version:
The so called "sleep" values which were hardcoded in previous version are now changed
to sysfs-tuneables:
The Smooth Scaling for sleep (Screen off):
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/smooth_up_sleep
possible values from 1 to 100, default: 100
The up/down Threshold for sleep (Screen off):
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold_sleep
possible Values from above "down_threshold_sleep" to 100, default: 90
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold_sleep
possible Values from 11 to under "up_threshold_sleep", default: 44
The Sampling Rate for sleep (Screen off):
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/sampling_rate_sleep_multiplier
possible values 1 or 2, default: 2
The amound of cores which should run at "screen off":
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/hotplug_sleep
possible Values 0 = do not touch cores (the same behaving as in standard setting of zzmoove) 1, 2 or 3
the number of cores to run on screen off. btw. setting "4" doesn't exist as u can use "0" for that setting!
Beside of that you can now change the hotplug threshold per core independently (thx to gsw5700 for the inital idea!)
and turn cores off completely.
For that purpose following tuneables were introduced and are replacing
the old hotplug up/down threshold tuneables:
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold_hotplug1
hotplug up threshold for core 1 - 0 = turn off core 1, possible range from "down_threshold" to 100, default: 68
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold_hotplug2
hotplug up threshold for core 2 - 0 = turn off core 2, possible range from "down_threshold" to 100, default: 68
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold_hotplug3
hotplug up threshold for core 3 - 0 = turn off core 3, possible range from "down_threshold" to 100, default: 68
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold_hotplug1
hotplug down threshold for core 1 - possible range from 11 to under "up_threshold", default: 55
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold_hotplug2
hotplug down threshold for core 2 - possible range from 11 to under "up_threshold", default: 55
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold_hotplug3
hotplug down threshold for core 3 - possible range from 11 to unter "up_threshold", default: 55
Thanks to:
gsw5700 for the initial Idea "hotplug threshold per core".
brijmathew indirectly for the initial idea "just one core at screen off" (i think he did'nt meant exactly that in his post but hey he switched on the led in my head! *gg*)
ZZMoove Governor v0.4
(Post from 1st May 2013, 10:59 PM - "limits")
Changelog:
Frequency Limits:
First of all there is now a (by me so called) "soft" limit function for limiting frequencies at screen off but also at screen on if u wish. however i recommend to set the screen on limit always with the (if provided) max scaling functionality of the kernel as this is the better way of doing it and "works better" for following reasons: touchboost and wake up frequencies can go above that governor-soft-limit (mostly to 800/1000 mhz) because the governor has no control over these "events" and will be "bypassed" by cpufreq driver! nevertheless by setting this soft limit at screen off the use of frequencies higher than the given limit will be strongly reduced and therefore this will reduce power consumption at screen off. and that was the main intention - saving power if u do not use your phone!
For this function following new tuneables were indroduced to set in sysfs (/sys/devices/system/cpu/cpufreq/zzmoove/):
freq_limit_sleep: limit freqency at screen off (possible values 0 disable limit, 200000-1600000, default: 0)
freq_limit: limit freqency at screen on (possible values 0 disable limit, 200000-1600000, default: 0)
Fast Scaling:
As a second feature in this new version i added the so called (again by me *g*) "fast scaling" for faster up/down scaling! This should bring more performance but on the other hand this can be of course a little bit more power consumptive. try it, it makes things snappier and some people reported that with this setting touch boost can then also be disabled which wasn't really "possible" with previous versions of zzmoove governor.
For this function following new tuneables were indroduced to set in sysfs (/sys/devices/system/cpu/cpufreq/zzmoove/):
fast_scaling: fast scaling at screen on (possible values 0 disable or 1 enable, default: 0)
fast_scaling_sleep: fast scaling at screen off (possible values 0 disable or 1 enable, default: 0)
As last "feature" and to complete the "set" the tuneable "freq_step_sleep" was included to be able to change freq step only at screen off
possible settings are the same as the "freq_step" tuneable which are values from 0% (stops freq scaling) to 100% (switches frequencies from lowest to highest frequency and vice versa like ondmand governor) default is 1 in all settings (bat/opt/perf) at screen off.
ZZMoove Governor v0.5 aka "The Beast"
(performance and fixes)
Changelog:
- completely reworked fast scaling functionality. now using a "line jump" logic instead of fixed freq "colums".
fast scaling now in 4 steps and 2 modes possible (mode 1: only fast scaling up and mode2: fast scaling up/down)
- added support for "Dynamic Screen Frequency Scaling" (original implementation into zzmoove governor highly improved by Yank555)
originated by AndreiLux more info: http://forum.xda-developers.com/showpost.php?p=38499071&postcount=3
- re-enabled broken conservative sampling down factor functionality ("down skip" method).
originated by Stratosk - upstream kernel 3.10rc1: https://git.kernel.org/cgit/linux/k...id=refs/tags/v3.10-rc1&qt=author&q=Stratos+Ka
- changed down threshold check to act like it should.
originated by Stratosk - upstream kernel 3.10rc1: https://git.kernel.org/cgit/linux/k...id=refs/tags/v3.10-rc1&qt=author&q=Stratos+Ka
- implemented/ported "early demand" from ondemand governor.
originated by Stratosk - more info: http://www.semaphore.gr/80-latests/98-ondemand-early-demand
- implemented/ported "sampling down momentum" from ondemand governor.
originated by Stratosk - more info: http://www.semaphore.gr/80-latests/80-sampling-down-momentum
- modified some original conservative code parts regarding frequency scaling which should work better now.
originated by DerTeufel1980: https://github.com/DerTeufel/androi...mmit/6bab622344c548be853db19adf28c3917896f0a0
- added the possibility to use sampling down momentum or conservative "down skip" method.
- increased possible max sampling rate sleep multiplier to 4 and sampling down factor to 100000
accordingly to sampling down momentum implementation.
- added frequency search limit for more efficient frequency searching in scaling "table" and for improving
frequency "hard" and "soft" limit handling.
- added cpu idle exit time handling like it is in lulzactive
again work from ktoonsez : https://github.com/ktoonsez/KT747-JB/commit/a5931bee6ea9e69f386a340229745da6f2443b78
description in lulzactive governor: https://github.com/ktoonsez/KT747-J...f2443b78/drivers/cpufreq/cpufreq_lulzactive.c
- fixed a little scaling step mistake and added overclocking frequencies up to 1800 mhz in scaling frequency "tables".
- fixed possible freezes during start/stop/reload of governor and frequency limit change.
- fixed hotplugging logic at online core 0+3 or 0+2 situations and improved hotplugging in general by
removing mutex locks and skipping hotplugging when it is not needed.
- added possibility to disable hotplugging (that's a debugging relict but i thought maybe someone will find that usefull so i didn't remove it)
- try to fix lags when coming from suspend if hotplug limitation at sleep was active by enabling all offline cores during resume.
- code cleaning and documentation.
for this functions following new tuneables were indroduced:
Early Demand:
early_demand -> switch to enable/disable early demand functionality (possible values 0 disable or 1 enable, default: 0)
grad_up_threshold -> scale up frequency if the load goes up in one step of grad up value (possible range from 11 to 100, default 50)
little example for understanding: when the load rises up in one big 50% step then the
frequency will be scaled up immediately instead of wating till up_threshold is reached.
Fast Scaling (improved):
Fast scaling has now 8 levels which at the same time have 2 modes included. Values from 1-4 equals to scaling jumps in the frequency table and uses the Fast Scaling up but normal scaling down mode. Values from 5-8 equals to 1-4 scaling jumps but uses the fast scaling up and fast scaling down mode.
Hotplugging switch:
disable_hotplug -> switch to enable/disable hotplugging (possible values are any value above 0 to disable hotplugging and 0 to
enable it, default 0)
Sampling Down Factor and Sampling Down Momentum:
Description: From the original author of ondemand_sampling_factor David Niemi:
"This improves performance by reducing the overhead of load evaluation and helping the CPU stay
at its top speed when truly busy, rather than shifting back and forth in speed."
And that "Sampling Down Momentum" function from stratosk does this dynamicly now!
sampling_down_max_momentum -> max sampling down factor which should be set by momentum (0 disable momentum, possible range from sampling_down_factor up to MAX_SAMPLING_DOWN_FACTOR, default 0 disabled)
sampling_down_momentum_sensitivity -> how fast the sampling down factor should be switched (possible values from 1 to 500, default 50)
sampling_down_factor -> depending on which mode is active the factor for sampling rate multiplier which influences the whole
sampling rate or the value for stock "down skip" functionality which influences only the down scaling
mechanism (possible values are from 1 to MAX_SMPLING_DOWN_FACTOR, default 1 disabled)
Original conservative "down skip" or "stock" method can be enabled by setting the momentum tuneable to 0. so if momentum is inactive there will be a fallback to the stock method. as the name "down skip" says this method works "slightly" different from the ondemand stock sampling down method (on which momentum was based on). It just skips the scaling down code for the given samples. if u want to completely disable the sampling down functionality u can achieve this by setting sampling down factor to 1. so concluded: setting sampling_down_momentum = 0 and sampling_down_factor = 1 will disable sampling down completely (that is also the governor default setting)
Dynamic Screen Frequency Scaling:
Dynamicly switches the screen frequency to 40hz or 60hz depending on cpu scaling and hotplug settings. For compiling and enabling this functionality u have to do some more modification to the kernel sources, please take a look at AndreiLux Perseus repository and there at following commit: https://github.com/AndreiLux/Perseus-S3/commit/3476799587d93189a091ba1db26a36603ee43519 After adding this patch u can enable the feature by setting "CPU_FREQ_LCD_FREQ_DFS=y" in your kernel config and if u want to check if it is really working at runtime u can also enable the accounting which AndreiLux added by setting LCD_FREQ_SWITCH_ACCOUNTING=y in the kernel config. If all goes well and u have the DFS up and running u can use following tuneables to do some screen magic: (thx to Yank555 for highly extend and improving this!)
lcdfreq_enable -> to enable/disable LCDFreq scaling (possible values 0 disable or 1 enable, default: 0)
lcdfreq_kick_in_down_delay -> the amount of samples to wait below the threshold frequency before entering low display frequency mode (40hz)
lcdfreq_kick_in_up_delay -> the amount of samples to wait over the threshold frequency before entering high display frequency mode (60hz)
lcdfreq_kick_in_freq -> the frequency threshold - below this cpu frequency the low display frequency will be active
lcdfreq_kick_in_cores -> the number of cores which should be online before switching will be active. (also useable in combination
with kickin_freq)
So this version is a kind of "featured by" release as i took (again *g*) some ideas and work from other projects and even some of that work
comes directly from other devs so i wanna thank and give credits:
First of all to stratosk for his great work "sampling down momentum" and "early demand" and for all the code fixes which found their way into
the upstream kernel version of conservative governor! congrats and props on that stratos, happy to see such a nice and talented dev directly
contibuting to the upstream kernel, that is a real enrichment for all of us!
Second to Yank555 for coming up with the idea and improving/completeing (leaves nothing to be desired now *g*) my first
rudimentary implementation of Dynamic Screen Frequency Scaling from AndreiLux (credits for the idea/work also to him at this point!).
Third to DerTeufel1980 for his first implementation of stratosk's early demand functionality into version 0.3 of zzmoove governor
(even though i had to modify the original implementation a "little bit" to get it working properly ) and for some code optimizations/fixes regarding scaling.
Last but not least again to ktoonsez - I "cherry picked" again some code parts of his ktoonservative governor which should improve this governor
too!
ZZMoove Governor 0.5.1a
(bugfixes)
urgend bugfix release:
- fix governor switching issues (deadlocks) which oviously werend fixed in version v0.5
- optimised scaling function a lot (thx and credits to Yank555!)
ZZMoove Governor 0.5.1b (in cooperation with Yank555)
(bugfixes and more optimisations)
Changelog:
- now really fixed the governor switching issues! (gotcha *****! *g*)
- again some changes in scaling logic from Yank555 (thx and credits)
- simplified some tuneables by using already available stuff instead of using redundant code (thx Yank555)
- reduced/optimised hotplug logic and preperation for automatic detection of available cores
(maybe this fixes also the scaling/core stuck problems)
ZZMoove Governor 0.6 (in cooperation with Yank555)
(flexibility)
Changelog:
- removed fixed scaling lookup tables and use the system frequency table instead
changed scaling logic accordingly for this modification (thx and credits to Yank555)
- reduced new hotplug logic loop to a minimum
- again try to fix stuck issues by using seperate hotplug functions out of dbs_check_cpu (credits to ktoonesz)
- added support for 2 and 8 core systems and added automatic detection of cores were it is needed
(for setting the different core modes you can use the macro 'MAX_CORES'. possible values are: 2,4 or 8, default are 4 cores)
reduced core threshold defaults to only one up/down default and use an array to hold all threshold values
- fixed some mistakes in "frequency tuneables" (Yank555):
stop looping once the frequency has been found
return invalid error if new frequency is not found in the frequency table
ZZMoove Governor 0.6a (in cooperation with Yank555)
(scaling logic flexibility)
Changelog:
- added check if CPU freq. table is in ascending or descending order and scale accordingly
(compatibility for systems with 'inverted' frequency table like it is on OMAP4 platform)
thanks and credits to Yank555!
ZZMoove Governor 0.7 aka "Tamed Beast" (in cooperation with Yank555)
(slow down)
Changelog:
- reindroduced the 'old way' of hotplugging and scaling in form of the 'Legacy Mode' (macros for enabling/disabling this done by Yank555, thx!)
NOTE: this mode can only handle 4 cores and a scaling max frequency up to 1800mhz.
- added hotplug idle threshold for a balanced load at CPU idle to reduce possible higher idle temperatures when running on just one core.
(inspired by @JustArchi observations, thx!)
- added hotplug block cycles to reduce possible hotplugging overhead (credits to @ktoonsez)
- added possibility to disable hotplugging only at suspend (inspired by a request of @STAticKY, thx for the idea)
- introduced hotplug frequency thresholds (credits to Yank555)
- hotplug tuneables handling optimized (credits to Yank555)
- added version information tuneable (credits to Yank555)
for this functions following new tuneables were indroduced:
legacy_mode -> for switching to the 'old' method of scaling/hotplugging. possible values 0 to disable, any values above 0 to enable (default is 0)
NOTE: the legacy mode has to be enabled by uncommenting the macro ENABLE_LEGACY_MODE
hotplug_idle_threshold -> amount of load under which hotplugging should be disabled at idle times (respectively at scaling minimum). possible values 0 disable, from 1 to 100 (default is 0)
hotplug_block_cycles -> slow down hotplugging by waiting a given amount of cycles before plugging. possible values 0 disbale, any values above 0 (default is 0)
disable_hotplug_sleep -> same as disable_hotplug but will only take effect at suspend. possible values 0 disable, any values above 0 to enable (default is 0)
up_threshold_hotplug_freq1 -> hotplug up frequency threshold for core1. possible values 0 disable and range from over down_threshold_hotplug_freq1 to max scaling freqency (default is 0)
up_threshold_hotplug_freq2 -> hotplug up frequency threshold for core2. possible values 0 disable and range from over down_threshold_hotplug_freq2 to max scaling freqency (default is 0)
up_threshold_hotplug_freq3 -> hotplug up frequency threshold for core3. possible values 0 disable and range from over down_threshold_hotplug_freq3 to max scaling freqency (default is 0)
down_threshold_hotplug_freq1 -> hotplug down frequency threshold for core1. possible values 0 disable and range from min saling to under up_threshold_hotplug_freq1 freqency (default is 0)
down_threshold_hotplug_freq2 -> hotplug down frequency threshold for core2. possible values 0 disable and range from min saling to under up_threshold_hotplug_freq2 freqency (default is 0)
down_threshold_hotplug_freq3 -> hotplug down frequency threshold for core3. possible values 0 disable and range from min saling to under up_threshold_hotplug_freq3 freqency (default is 0)
version -> show the version of zzmoove governor
ZZMoove Governor 0.7a
(little fix)
Changelog:
- fixed a glitch in hotplug freq threshold tuneables which prevented setting of values in hotplug down freq thresholds when hotplug
up freq thresholds were set to 0. NOTE: enabling the legacy mode in the source is not part of that fix i just forgot to set it back to standard (disabled) before pushing, sry to lazy to fix that fix now
ZZMoove Governor 0.7b
(compatibility improved and forgotten things)
Changelog:
- fixed stuck at max scaling frequency when using stock kernel sources with unmodified cpufreq driver and without any oc capabilities.
- readded forgotten frequency search optimisation in scaling logic (only effective when using governor soft frequency limit)
- readded forgotten minor optimisation in dbs_check_cpu function.
- as forgotten to switch in last version Legacy Mode now again disabled by default
- minor code format and comment fixes
ZZMoove Governor 0.7c
(again compatibility and optimisations)
Changelog:
- frequency search optimisation now fully compatible with ascending ordered system frequency tables (thx to @psndna88 for testing!)
- again minor optimisations at multiple points in dbs_check_cpu function
- code cleaning - removed some unnecessary things and whitespaces nuked - sry for the bigger diff but from now on it will be clean!
- corrected changelog for previous version regarding limits
ZZMoove Governor 0.7d (bugfix!)
(broken things)
Changelog:
- fixed hotplug up threshold tuneables to be able again to disable cores manually via sysfs by setting them to 0
- fixed the problem caused by a "wrong" tuneable apply order of non sticking values in hotplug down threshold tuneables when
hotplug up values are lower than down values during apply.
NOTE: due to this change right after start of the governor the full validation of given values to these tuneables is disabled till
all the tuneables were set for the first time. so if you set them for example with an init.d script or let them set automatically
with any tuning app be aware that there are illogical value combinations possible then which might not work properly!
simply be sure that all up values are higher than the down values and vice versa. after first set full validation checks are enabled
again and setting of values manually will be checked again.
- fixed a typo in hotplug threshold tuneable macros (would have been only a issue in 8-core mode)
- fixed unwanted disabling of cores when setting hotplug threshold tuneables to lowest or highest possible value
which would be a load of 100%/11% in up/down_hotplug_threshold and/or scaling frequency min/max in up/down_hotplug_threshold_freq
CPU Governor ZZMoove Changelog
ZZMoove Governor 0.8
(cool down baby!)
Changelog:
- indroduced scaling block cycles (in normal and legacy mode) to reduce unwanted jumps to higher frequencies (how high depends on settings) when a load comes up just for a short peroid of time or is hitting the scaling up threshold more often because it is within some higher to mid load range. reducing these jumps lowers CPU temperature in general and may also save some juice. so with this function u can influence the little bit odd scaling behaving when you are running apps like for example games which are constantly 'holding' system load in some range and the freq is scaled up too high for that range. in fact it just looks like so, monitoring apps are mostly too slow to catch load in realtime so you are watching almost only 'the past'. so actually it's not really odd it's more like: an app is stressing the system and holding it on a higher load level, due to this load level scaling up threshold is more often reached (even if monitoring shows a lower load than up_threshold!) the governor scales up very fast (usual zzmoove behaving) and almost never scales down again (even if monitoring shows a lower load than down_threshold!). now in patricular these scaling block cycles are throttling up scaling by just skipping it for the amount of cycles that you have set up and after that are making a forced scale down in addition for the same amount of cycles. this should bring us down to a 'appropriate' frequency when load is hanging around near up threshold.
- indroduced (D)ynamic (S)ampling (R)ate - thx to hellsgod for having the same idea at the same time and pointing me to an example. even though at the end i did 'my way' DSR switches between two sampling rates depending on the given load threshold from an 'idle' to a 'normal' one.
- added read only tuneable 'sampling_rate_current' for DSR to show current SR and internally use this sampling rate instead of automatically changing 'sampling_rate' tuneable in DSR. this keeps things more compatible and avoids problems when governor tuneables are set with tuning apps which are saving actual shown values.
- changed setting of sampling rate in governor from 'sampling_rate' to 'sampling_rate_current' and the value at suspend to 'sampling_rate_idle' value instead of using current active sampling rate to avoid accidentally setting of 'normal operation' sampling rate in sampling rate idle mode which has usually a much lower value.
- indroduced build-in profiles in seperate header file (credits to Yank555 for idea and prototype header file) you can switch between multible build in profiles by just piping a number into the tuneable 'profile_number'. all tuneable values of the set profile will be applied on-the-fly then. if a profile is active and you change any tuneable value from it to a custom one the profile will switch to '0' in 'profile_number' and 'custom' in 'profile' for information. with this profiles support developers can simplify their governor settings handling by adding all their desired and well proven governor settings into the governor itself instead of having to fiddle around with init.d scripts or tuning apps. NOTE: this is just an optional feature, you can still set all tuneables as usual just pay attention that the tuneable 'profile_number' is set to '0' then to avoid overwriting values by any build-in profile! for further details about profiles and adding own ones check provided 'cpufreq_zzmoove_profiles.h' file.
- added 'profiles_version' tuneable to be able to show seperate profiles header file version.
- added enabling of offline cores on governor exit to avoid cores 'stucking' in offline state when switching to a non-hotplug-able governor
and by the way reduced reduntant code by using an inline function for switching cores on and using the better 'sheduled_work_on-way' at all needed places in the code for that purpose.
- moved some code parts to legacy mode macro which has only relevance when including the legacy mode in the governor and in addition excluded it during runtime if legacy mode is disabled.
- improved freq limit handling in tuneables and in dbs_check_cpu function
- changed value restriction from '11' to '1' in hotplug down threshold and grad up tuneables as this restriction is only nessesary in
scaling down tuneable
- added missing fast scaling down/normal scaling up mode to fast scaling functionality (value range 9-12 and only available in non-legacy mode thx @OldBoy.Gr for pointing me to that missing mode!)
- added auto fast scaling aka 'insane' scaling mode to fast scaling functionality - lucky number 13 enables this mode in fast_scaling tuneable NOTE: a really funny mode, try it but keep in mind setting this in combination with a set freq limit (at sleep or awake)would not make much sense as there is not enough frequency range available to jump around then.
- back from the precautious 'mutex_try_lock' to normal 'mutex_lock' in governor 'LIMIT' case -> this should be save again, no deadlocks expected any more since hotplug logic has significantly changed in zzmoove version 0.6
- removed also the no longer required and precautious skipping of hotplugging and dbs_check_cpu on multiple places in code and removed the mutex locks at governor stop and early suspend/late resume
- added hotlug freq thresholds to legacy scaling mode (same usage as in normal scaling mode)
- seperated hotplug down and up block cycles to be more flexible. this replaces 'hotplug_block_cycles' with 'hotplug_block_up_cycles' tuneable and adds one new tunable 'hotplug_block_down_cycles'. functionality is the same as before but u can now differentiate the up and down value.
- added 'early demand sleep' combined with automatic fast scaling (fixed to scaling up mode 2) and if not set already automatic (depending on load) switching of sampling rate sleep multiplier to a fixed minimum possible multiplier of 2. this should avoid mostly audio or general device connection problems with 'resource hungrier' apps like some music players, skype, navigation apps etc. and/or in combination with using bluetooth equipment during screen is off. NOTE: this overwrites a possible fast 'scaling_sleep' setting so use either this or 'fast_scaling_sleep'
- added some missing governor tunebable default value definitions
- removed tuneable apply order exception and removed analog value checks in hotplug threshold and hotplug frequency tuneables to avoid tuneable values not changing issues. NOTE: keep in mind that all 'down' values should be lower then the analog 'up' values and vice versa!
- removed 200 mhz up hotplugging restriction, so up hotplugging starts at 200 mhz now
- removed some unnecessary macros in scaling logic
- added maximum fast scaling and frequency boost to late resume to react wakeup lags
- merged some improvements from ktoonservativeq governor version for the SGS4 (credits to ktoonsez)
changes from here: https://github.com/ktoonsez/KT-SGS4/commits/aosp4.4/drivers/cpufreq/cpufreq_ktoonservativeq.c
Use dedicated high-priority workqueues
Use NEW high-priority workqueue for hotplugging
Improved hotplugging in general by reducing calls to plugging functions if they are currently running,
by reducing calls to external function in up plugging and by changing the down plug loop to an optimized one
- added hotplug boost switch to early demand functionality and up hotplugging function
- added 'hotplug_idle_freq' tuneable to be able to adjust at which frequency idle should begin
- transfered code for frequency table order detection and limit optimisation into a inline function and use this function in START,LIMIT case and early suspend/late resume instead of using redundant code
- execute table order detection and freq limit optimization calculations at 'START' and 'LIMIT' case to avoid possible wrong setting of freq
max after governor start (currently set max frequency value was sometimes not applied) and a wrong soft limit optimization setting after undercutting the soft limit with a lower hard limit value
- minor optimisation in hotplug, hotplug block and in all freq search logic parts
- added debugging sysfs interface (can be included/excluded using #define ZZMOOVE_DEBUG) - credits to Yank555!
- added some missing annotation as a prepareation and mainly to avoid some errors when compiling zzmoove in combination with 3.4+ kernel sources
- fixed hotplugging issues when cpufreq limit was set under one or more hotplugging frequency thresholds NOTE: now hotplugging frequency thresholds will be completely disabled and a fall back to normal load thresholds will happen if the maximal possible frequency will undercut any frequency thresholds
- fixed stopping of up scaling at 100% load when up threshold tuneable is set to the max value of 100
- fixed smooth up not active at 100% load when smooth up tuneable is set to the max value of 100
- fixed many code style and dokumentation issues and made a massive code re-arrangement
for this functions following new tuneables were indroduced:
early_demand_sleep -> same function as early demand on awake but in addition combined with fast scaling and sampling rate switch and only active at sleep. (possible values 0 disable or 1 enable, default is 1)
grad_up_threshold_sleep -> 2 way functionality: early demand sleep grad up (load gradient) threshold and at the same time load threshold for switching internally (tuneables are staying at set values!) sampling_rate_sleep_multiplier to 2 and fast_scaling to 2 (possible values from 1 to 100, default is 35)
hotplug_block_up_cycles -> (replaces hotplug_block_cycles) slow down up hotplugging by waiting a given amount of cycles before plugging. possible values 0 disbale, any values above 0 (default is 0)
hotplug_block_down_cycles -> (replaces hotplug_block_cycles) slow down down hotplugging by waiting a given amount of cycles before plugging. possible values 0 disbale, any values above 0 (default is 0)
hotplug_idle_freq -> freq at which the idle should be active (possible values 0 disable and any possible scaling freq, default is 0)
sampling_rate_current -> read only and shows currently active sampling rate
sampling_rate_idle -> sampling rate which should be used at 'idle times' (possible values are any sampling rate > 'min_sampling_rate', 0 to disable whole function, default is 0)
sampling_rate_idle_delay -> delay in cycles for switching from idle to normal sampling rate and vice versa (possible values are any value and 0 to disable delay, default is 0)
sampling_rate_idle_threshold -> threshold under which idle sampling rate should be active (possible values 1 to 100, 0 to disable function, default is 0)
scaling_block_cycles -> amount of gradients which should be counted (if block threshold is set) and at the same time up scaling should be blocked and after that a forced down scaling should happen (possible values are any value, 0 to disable that function, default is 0)
scaling_bock_freg -> frequency at and above the blocking should be active (possible values are any possible scaling freq, 0 to enable blocking permanently at every frequency, default is 0)
scaling_block_threshold -> gradient (min value) of load in both directions (up/down) to count-up cycles (possible value are 1 to 100, 0 to disable gradient counting)
scaling_block_force_down -> multiplicator for the maximal amount of forced down scaling cycles (force down cycles = block_cycles * force_down) therefore the forced down scaling duration (possible value are 2 to any value, 0 to disable forced down scaling and use only scaling up blocks)
profile -> read only and shows name of currently active profile ('none' = no profile, 'custom' = a profile value has changed)
profile_number -> switches profile (possible value depends on amount of profiles in cpufreq_zzmoove_profiles.h file, please check this file for futher details!) 0 no profile set = tuneable mode, default 0)
version_profiles -> read only and shows version of profile header file
if ZZMOOVE_DEBUG is defined:
debug -> read only and shows various usefull debugging infos
ZZMoove Governor 1.0 Final
(EOL)
Changelog:
EOL commit: https://github.com/zanezam/cpufreq-governor-zzmoove/commit/b22eed13461fbb59b8e9e0121656582d6ef10696
Governor: https://github.com/zanezam/cpufreq-governor-zzmoove/blob/snapdragon/CHANGELOG.txt
Profiles: https://github.com/zanezam/cpufreq-governor-zzmoove/blob/snapdragon/CHANGELOG_PROFILES.txt
Current Test-Versions (obsolete but kept for reference):
Version 1.0 beta8
(outbreak)
Changelog:
https://github.com/zanezam/cpufreq-governor-zzmoove/commit/bef4355a652fdfc268b774e874b355bd269dbc07
Version 1.0 beta7a (bugfix)
(outbreak)
Changelog:
https://github.com/zanezam/cpufreq-governor-zzmoove/commit/1ce22579b0e6dfce0d9916b9f0644e7d707d8587
Version 1.0 beta7 (sync)
(outbreak)
Changelog:
https://github.com/zanezam/cpufreq-governor-zzmoove/commit/52ad61b169c24258ec1b0e6a630bbab5142faa67
Version 1.0 beta6a (Andip71 aka Lord Boeffla)
(outbreak)
Changelog (credits to Andip71):
https://github.com/zanezam/cpufreq-governor-zzmoove/commit/9b01250eb988e86c98d289dabe6773ba748f58dd
Version 1.0 beta6 (feature preview, for opo only atm.)
(outbreak)
Changelog:
https://github.com/zanezam/cpufreq-governor-zzmoove/commit/75c431aeff8354f224e4c688faa1559ac463ab8a
Version 1.0 beta5
(outbreak)
Changelog:
https://github.com/zanezam/cpufreq-governor-zzmoove/commit/1d2727cb9cefe0484573ace9e0dc9b55ba6c26f6
Version 1.0 beta4 (sync)
(outbreak)
Changelog:
- use again the conservative governor usual canceling of dbs work syncron instead of asyncron when exiting the governor as this change was
only needed in combination with older hotplug implementations. as also done in opo version removed again all previously merged kernel crash
fix attempts and precautions as they were not really needed
- bump version to beta4 to bring opo/i9300 versions in sync again
Version 1.0 beta3 (bugfix for opo-bacon)
(outbreak)
Changelog:
- changed back canceling of dbs work to syncron instead of asyncron in dbs_timer_exit function to avoid random kernel chrashes (again oops in smp.c)
when using this governor with the cpufreq implementation of kernel versions 3.10+. problem was initiated by governor restarts during hotplugging
- as an additional precaution check if a core is online before doing critical stuff in dbs_check_cpu main function (might be removeable at a later
time, more analyses/tests will show)
profile header file (Version 0.3 beta2 OPO)
- corrected/adjusted sampling rate values in settings where they were lower than the minimal possible value of 60000
Version 1.0 beta2
(outbreak)
Changelog:
- avoid kernel crash (usually a oops in smp.c) by checking if a core is online before putting work on it: this problem appeared on opo qualcomm
platform with proprietary mpdecision hotplugging service. assumption is that there is a delay between initiating hotplugging events from 'userland'
and gathering core state info in 'kernel land' so under some rare circumestances the governor doesn't 'know about' a changed core state and tries to
put work on a meanwhile offline core or that hotplug event happend during putting work on a core in the governor.
Version 1.0 beta1
(outbreak)
Changelog:
- bump version to 1.0 beta1 because of brought forward plan 'outbreak'
- reworked scaling logic:
removed unessesary calls of external cpufreq function and use a static variable instead to hold the system freq table during runtime fixed frequency stuck on max hard and soft frequency limit (under some circumstances freq was out of scope for the main search loop) and added precautions to avoid problems when for what ever reason the freq table is 'messed' or even not available for the governor fixed not properly working scaling with descend ordered frequency table like it is for example on qualcomm platform added additional propotional scaling mode (mode '1' as usual decide and use the lowest freq, new mode '2' use only propotional frequencies like ondemand governor does - switchable as before in 'scaling_proportional' tuneable)
- use static frequency table variable in all frequency checks agains system frequency table in the governor
- fixed 'update_ts_time_stat idle accounting' (kernel patch for kernel version 3.0.x needed, example available in github zzmoove repository)
- fixed non setting of scaling down threshold tuneables under some circumstances (issue on kernel 3.4 when running multible zzmoove instances)
- changed some variable names in scaling range evaluation and debugging tuneable
- added compatibility for kernel version 3.4 (or higher, but only tested on 3.4.0 yet)
- added compatibility for cpufreq implementation used since kernel version 3.10 (NOTE: for backports u can use the macro CPU_IDLE_TIME_IN_CPUFREQ)
- added support for powersuspend (used on some platforms since kernel version 3.4)
- added support for opo specific 'backlight ext control' (kernel patch for opo bacon devices needed, example available in github zzmoove repository)
- added macros to exclude hotplugging functionality (default in this version is enabled=uncommented)
profile header file (Version 0.3 beta1)
- bump version to 0.3 beta1 because of brought forward plan 'outbreak'
- removed dynamic freq scaling tuneable leftovers
- added macros to switch code depending of used power management implementation
or used supend/resume backlight hook (opo specific)
- added macros to be able to disable hotplugging
Version 0.9 beta4
(slimline)
Changelog:
- removed 'freq_step' functionality as it never had any function in this governor. it was a left over from mialwes 'smoove' governor and also
had no function in his governor back then. so yeah all the 'feelings' about it's influence were placebo
- introduced 'proportional scaling' for more 'connectivity' to current load, this should give more 'balanced' frequencies
in general. when enabled all targeted frequencies in scaling logic will be compared with the ones from system table method and at the end
the lowest of them both will be used. so all used scaling frequencies will be 'tentential' lower in both directions
- added support for exynos4 CPU temperature reading (patches available in zzmoove repositories: https://github.com/zanezam)
this must be enabled via 'CONFIG_EXYNOS4_EXPORT_TEMP=y' in the config of a kernel which has exynos4 CPU temperature export
implementation
included. the default temp polling interval is 1000 ms and can be set with DEF_TMU_READ_DELAY. however the TMU driver has it's own polling
interval which is 10 seconds, so leaving it at the default value of 1 second is recommended temperature reading will only be enabled if the
tuneable 'scaling_block_temp' is set and will be disabled whenever early suspend is entered
- if exynos4 CPU temperature reading is enabled in the code use current CPU temperature in scaling block functionality to be able to 'hold' the cpu
temperatue to the given one in 'scaling_block_temp'. this function is used in combination with the already existent tuneable 'scaling_block_freq'
so u have to set both to enable it. the possible temperature range is 30°C to 80°C (lower temps are making no sense and higher temps would reach
into exynos4 TMU driver trottling range)
- if exynos4 CPU temperature reading is enabled added current CPU temperature to debug info tuneable
- added auto adjustment of all available frequency thresholds when scaling max limit has changed
- again some code style and comment changes/fixes
profile header file (Version 0.2 beta3):
- added scaling block temperature tuneable to all profiles if CONFIG_EXYNOS4_EXPORT_TEMP is defined
- use CPU temperature treshold of 65°C instead of 15 scaling block cycles in game profile if
CONFIG_EXYNOS4_EXPORT_TEMP is defined
- added proportional frequency tuneable to all profiles
- added auto adjust freq thresholds to all profiles (disabled by default)
- enabled scaling proportional in ybat, ybatext, zzbat, zzbatp, zzmod and zzgame profile
- enabled scaling fast down over 1200MHz and resposiveness over 400Mhz with up threshold of 20
in ybat, ybatext, zzbat, zzbatp, and zzmod profile
- added core macros to exclude not used code like it is in governor
- removed freq_step tuneable from all profiles
Version 0.9 beta3
(slimline)
Changelog:
merged some changes originated by ffolkes (all credits and thx to him)
(source https://github.com/ffolkes/android_...44e1e3190d5/drivers/cpufreq/cpufreq_zzmoove.c
- reordered sysfs attributes
description by ffolkes: some apps set tuneables by the order in which they are listed in the filesystem. this causes problems when one
tuneable needs another set first in order to correctly validate. (e.g. you cannot set down_threshold properly until you have first set up_threshold)
- added 'fast down' functionality (based on commits to pegasusq in perseus by andreilux and extended for this version by ZaneZam)
description by ffolkes: fastdown dynamically applies a (presumably) higher up_threshold and down_threshold after a frequency threshold has been reached. the goal is to encourage less time spent on the highest frequencies
- added 'hotplug engage' functionality
description by ffolkes: when set >0, will not bring any cores online until this frequency is met or exceeded
goal: reduce unnecessary cores online at low loads
- added 'scaling responsiveness' functioniality
description by ffolkes: similar to 'frequency for responsiveness' in other governors
defines a frequency below which we use a different up_threshold to help eliminate lag when starting tasks
- instead of failing when set too high in down_threshold tuneables set the value to the highest it can safely go
- increased possible sampling rate sleep multiplier value to a max of 8 (ZaneZam)
- added missing error handling to some tuneables (ZaneZam)
- fixed non setting of 'hotplug_sleep' tuneable when applying profiles (ZaneZam)
- added up/down threshold to debug tuneable (ZaneZam)
- some code style and comment changes/fixes (ZaneZam)
for this functions following new tuneables were introduced:
scaling_fastdown_freq-> will be enabled once this frequency has been met or exceeded (0 to disable, all possible system frequencies, default is 0)
scaling_fastdown_up_threshold-> once the above frequency threshold has been met, this will become the new up_threshold until we fall below the scaling_fastdown_freq again. (range from over fastdown_down_threshold to 100, default is 95)
scaling_fastdown_down_threshold-> once the above frequency threshold has been met, this will become the new down_threshold until we fall below the scaling_fastdown_freq again. (range from 11 to under fastdown_up_threshold, default is 90)
scaling_responsiveness_freq-> will be enabled once this frequency has been met or exceeded (0 to disable, all possible system frequencies, default is 0)
scaling_responsiveness_up_threshold-> the up_threshold that will take effect if scaling_responsiveness_freq is set (range from 11 to 100, default is 30)
hotplug_engage_freq -> will not bring any cores online until this frequency is met or exceeded (0 to disable, any possible system frequencies, default is 0)
profile header file (Version 0.2 beta2):
- added values for following new tuneables (credits to ffolkes):
hotplug_engage_freq (disabled by default in all profiles)
scaling_fastdown_freq (disabled by default in all profiles)
scaling_fastdown_up_threshold (default to 95 in all profiles)
scaling_fastdown_down_threshold (default to 90 in all profiles)
scaling_responsiveness_freq (disabled by default in all profiles)
scaling_responsiveness_up_threshold (default to 30 in all profiles)
- adjusted up/down thresholds for core 2 in moderate setting
- changed sampling rate sleep multiplier from 4 to 6 in all settings (except in default setting)
Version 0.9 beta2
(slimline)
Changelog:
- support for setting a default settings profile at governor start without the need of using the tuneable 'profile_number'
a default profile can be set with the already available macro 'DEF_PROFILE_NUMBER' check zzmoove_profiles.h for details about
which profile numbers are possible. this functionality was only half baken in previous versions, now any given profile will be really
applied when the governor starts. the value '0' (=profile name 'none') in 'DEF_PROFILE_NUMBER' disables this profile hardcoding and
that's also the default in the source. u still can (with or without enabling a default profile in the macro) as usual use the tuneable
'profile_number' to switch to a desired profile via sysfs at any later time after governor has started
- added 'blocking' of sysfs in all tuneables during apply of a settings profile to avoid a possible and unwanted overwriting/double
setting of tuneables mostly in combination with tuning apps where the tuneable apply order isn't influenceable
- added tuneable 'profile_list' for printing out a list of all profiles which are available in the profile header file
- fixed non setting of 'scaling_block_force_down' tuneable when applying profiles
- some documentation added and a little bit of source cleaning
Version 0.9 beta1
(slimline)
Changelog:
- bump version to beta for public
- added/corrected version informations and removed obsolete ones
profile header file (Version 0.2 beta1):
- bump version to beta for public
- corrected version informations
Version 0.9 alpha-2
Changelog:
- added auto fast scaling step tuneables:
afs_threshold1 for step one (range from 1 to 100)
afs_threshold2 for step two (range from 1 to 100)
afs_threshold3 for step three (range from 1 to 100)
afs_threshold4 for step four (range from 1 to 100)
profile header file (Version 0.2alpha-2):
- corrected documentation
- corrected version information
- added auto fast scaling step tuneables and values to all profiles
Version 0.9 alpha1 (Yank555.lu)
Changelog (credits to Yank555!):
- splitted fast_scaling into two separate tunables fast_scaling_up and fast_scaling_down so each can be set individually
to 0-4 (skip 0-4 frequency steps) or 5 to use autoscaling.
- splitted fast_scaling_sleep into two separate tunables fast_scaling_sleep_up and fast_scaling_sleep_down so each
can be set individually to 0-4 (skip 0-4 frequency steps) or 5 to use autoscaling.
- removed legacy mode (necessary to be able to split fast_scaling tunable)
- removed LCD frequency DFS
profile header file (Version 0.2):
- split fast_scaling and fast_scaling_sleep into fast_scaling_up/fast_scaling_down and
fast_scaling_sleep_up/fast_scaling_sleep_down
- adjusted values for profiles Yank Battery and Yank Battery Extreme
Currently in the Workshop:
working on Current versions of ZZMoove big.LITTLE Edition (bLE)
Sources:
https://github.com/zanezam/cpufreq-governor-zzmoove
General project status and repository changes:
NOTE: Changes in repository since 15.08.14
the test repository was deleted and all previous beta and stable commits are available now in the main repository.
all future commits will also land there! branches in this repo are "i9300" the new merged master branch and a new one
was added "desktop" for the "desktop edition" of zzmoove (which is btw. still WIP and far from stable, but it compiles).
Since 12.10.2014 there is a new branch called "opo-bacon" where u can find a special version for One Plus One devices and
some needed additional kernel patches for it.
NOTE: Changes in repository since 12.08.15:
renamed branch 'i9300' to 'exynos' and branch 'opo-bacon' to 'snapdragon'
added CHANGELOG_PROFILES file with the changelog of the profiles header file of this version
updated README with some quick infos about predefinded settings in this version
'desktop' branch synced with current version (same base for all now, just other settings)
new 'develop' branch based on 'snapdragon' version created for quick changes, pull requests, etc.
in short this will be the main develop branch in the future and will provide changes and fixes
earlier. but this branch also shall be considered as more experimental!
NOTE: Changes in repository since 26.05.20:
rebase: fixed all broken links to implementation examples linking to original boeffla kernels (dev has retired and removed all his work)
by using links to my still available forks. deleted 'develop' branch as it's obsolete now since old versions 'snapdragon' and 'exynos' and 'desktop'
are final and EOL now. changed default branch to 'bLE-develop-k49x' which is the newest development version.
Status:
I recently did a formal change by merging all the recent changes from the already
old 'develop' branch into a new final 1.0 version on variants 'exynos',
'snapdragon' and 'desktop' which all totally can be considered as end of life.
After 'some time' of beta state and after many successful implementations
over the years, at it's 'best times' running on tousands of devices in
different kernels, beeing stable as it currently is (used it for years 24/7
on all my meanwhile 10 devices, and still do!) and as it can be (well i have
to admit sometimes it was a beast!! *gg*) im going to make the almost 3 years
old 'develop' version finally the version 1.0 and consider it as stable.
By doing this i'm also closing this old development chapter by making the
'exynos,snapdragon and desktop' versions final and EOL. The 'develop' branch
is now obsolete and gets deleted.
Finally stable, phew what a journey!
Thx to all contributing devs and ppl who contributed with their ideas, tests
and nervs when it did 'mock around'! Last but not least thx to all of those
which let 'the beast' running on their devices! I hope u all enjoyed it like i did!
NOTE: Despite me beeing not as active as some years ago with development i
want to let u know that there meanwhile exists a 'big Little brother' (poor
one even has NO version number at all yet) of ZZMoove governor which is still
happily hopping around on 'newer' devices. This newer version exists already since
about 3 years and were already running in (meanwhile also EOL) boeffla kernels for
the OnePlus 3/5 and also can be found in some recent OnePlus 6 kernels (at least in the ones i
fiddled recently *g*). By the way i shamelessly take the opportunity: If u still
have a OnePlus 6/6T u really you should consider trying one of these builds, more
info can be found here: https://github.com/zanezam/ZZupreme-Builds
ZZMoove still rocks by significantly influencing speed and battery usage in a
positive way! Just saying.
Enjoy!
ZZMoove Governor Settings
Description of how u can switch build in Profiles
Beside of the well known ones like "Yank...","Battery","Performance" etc. settings
i did some new ones for ZZMoove v1.0 in addition which can be checked out
in particular in the new profile exynos header file provided HERE
So for those of u which have not the possibility to switch profiles with kernel dev provided tools/scripts/or what
ever but still want to switch to any of the build in settings you can do following:
use tools like Android Tuner ,SetCPU, Kernel Adiutor or similar tools which are supporting the change of multible tuneables on-the-fly or just do it directly in kernel sysfs via a terminal emulator and give the tuneable "profile_number" one of the following values:
for Default (set governor defaults)
for Yank Battery -> old untouched setting (a very good battery/performance balanced setting DEV-NOTE: highly recommended!)
for Yank Battery Extreme -> old untouched setting (like yank battery but focus on battery saving)
for ZaneZam Battery -> old untouched setting (a more 'harsh' setting strictly focused on battery saving DEV-NOTE: might give some lags!)
for ZaneZam Battery Plus -> NEW! reworked 'faster' battery setting (DEV-NOTE: recommended too! )
for ZaneZam Optimized -> old untouched setting (balanced setting with no focus in any direction DEV-NOTE: relict from back in the days, even though some people still like it!)
for ZaneZam Moderate -> NEW! setting based on 'zzopt' which has mainly (but not strictly only!) 2 cores online
for ZaneZam Performance -> old untouched setting (all you can get from zzmoove in terms of performance but still has the fast down scaling/hotplugging behaving)
for ZaneZam InZane -> NEW! based on performance with new auto fast scaling active. a new experience!
for ZaneZam Gaming -> NEW! based on performance with new scaling block enabled to avoid cpu overheating during gameplay
for ZaneZam Relax -> NEW! based on moderate (except hotplug settings) with relaxed sleep settings (to react audio/bluetooth/wakeup issues)
for asad007 lwk -> NEW! made by xda user asad007 with yet unknown direction
(since version 0.9 beta4: cpu temperature threshold of 65°C enabled if exynos4 cpu temperature reading support was compiled with the governor)
after a tunebable view-refresh in the tools/console u can see the name of the currently set profile in the "profile" tuneable,
but keep in mind not every tool supports returning chars in tunebables as some of them are internally handling them as numbers,
even though in fact they are chars in kernel sysfs so u might see a "-1" then (for example with SetCPU) not so with
Android Tuner which shows (again) what a great app this is.
Some last note for the game profile: plz do not complain about ingame performance issues or lags with this setting.
i tuned it to be more aggressive in terms of bringing freq down. reference game was angy birds star wars *g* calming
down zzmoove while playing this game was really a chellange, don't know why but it stresses the system a lot more
than others and u know how 'hasty' zzmoove can be another test game was temple run 2 and here u can see a significant
improvement with this setting. it's a bit difficult to give u THE universal setting for all games as they differ too much in terms
of how much the system is stressed by them. so we have to share our experiences to further optmize it till i made scaling
blocks more "intelligent" and therefore more automatically (idea already exists!!)
Have fun with all the settings and feel free to report back how they work for you, and/or post new ones! :highfive:
You're the Master of text!
Great and detailed
Gesendet von meinem GT-I9300 mit Tapatalk 2
romskii said:
You're the Master of text!
Great and detailed
Gesendet von meinem GT-I9300 mit Tapatalk 2
Click to expand...
Click to collapse
hehe thx i try my best to make it understandable
but i'm just about editing to shorten this novel
I do not know, but this makes the problem gov.mi / freeze and delay / I do not know where the problem is.?
I use boeffla kernel and I zzmove and still the problem
the gov.pegasusq so good
misacek said:
I do not know, but this makes the problem gov.mi / freeze and delay / I do not know where the problem is.?
I use boeffla kernel and I zzmove and still the problem
the gov.pegasusq so good
Click to expand...
Click to collapse
you are on actual beta boeffla kernel, right? do you have this on all provided settings up to "perf"? Do you uv? if so try to avoid uv for some time just to exclude this. and if you have zram enabled, can u try without this too? or maybe there is some app which is not dancing with the "beast" ? all guessed sorry can not give more advices, actual boeffla/zzmoove version runs without problems on my phone.
Great thanks I'll keep trying and testing
Finally we have a thread for zzmove governor!! :victory:
Great explanations Zane! I don't understand a few things though haha.
This is very useful, thank you very much. I'll try to learn and test tweaks for this governor, but will be probably asking you about some new features I don't get to understand
Thanks!! :good::good:
Enviado desde mi GT-I9300 usando Tapatalk 2
the best is disable touchboost or enable=?
I'm very happy for this thread! !
I will study it to understand better zzmove
un pò di tap qui e là...
Rom: Maya Rom v5.2
Kernel: Boeffla 2.12 beta 8
Modem: Buemc2
Operator: Tim
Recovery: Philz touch
@ZaneZam
I've been trying to setup lcdfreq to stay on 40hz regardless of frequency and cores.. No matter how I set it up I can't get it to sit on 40... I can get it to sit on 40 MOST of the time, but not ALL the time...
Any help appreciated..
Oh and I'm using temasek's ROM (RC5.4.1) and kernel (160613 wolfston with abb)..
Thanks
TP.
core720 said:
the best is disable touchboost or enable=?
Click to expand...
Click to collapse
since version 0.4 u can if u want, on earlier versions it lagged like hell.
it works quite well without touchboost on v0.5.x but for my taste i leave it on as i want my phone as touch responsive as possible.
but just try it maybe u like it without touchboost.
STAticKY said:
@ZaneZam
I've been trying to setup lcdfreq to stay on 40hz regardless of frequency and cores.. No matter how I set it up I can't get it to sit on 40... I can get it to sit on 40 MOST of the time, but ALL the time...
Any help appreciated..
Oh and I'm using temasek's ROM (RC5.4.1) and kernel (160613 wolfston with abb)..
Thanks
TP.
Click to expand...
Click to collapse
ok have u tried to enter 1300000? (assuming 1400000 is your max scaling setting) that should enable it all the time.
i dunno exaclty which lcd freq scaling implementation is used in temaseks kernel but i assume it's the original from
andreilux. in this implementation there is an additional touchboost switch which we removed in the boeffla
implementation. if i remember right this touchboost switch enabales hi frequency on touch. so your settings
might be overwritten by this.
ZaneZam said:
ok have u tried to enter 1300000? (assuming 1400000 is your max scaling setting) that should enable it all the time.
i dunno exaclty which lcd freq scaling implementation is used in temaseks kernel but i assume it's the original from
andreilux. in this implementation there is an additional touchboost switch which we removed in the boeffla
implementation. if i remember right this touchboost switch enabales hi frequency on touch. so your settings
might be overwritten by this.
Click to expand...
Click to collapse
Using max CPU freq at 1400 with lcdfreq freq threshold at 1300. Touch boost turned off.... Baffles me. I've also tryed setting max CPU freq to 1200 and lcdfreq freq threshold to 1300 because of something tema or yank said.. Can't remember what it was off hand.. I'll have a look back
TP.
core720 said:
the best is disable touchboost or enable=?
Click to expand...
Click to collapse
Zane
Enviado desde mi GT-I9300 usando Tapatalk 2
core720 said:
Zane
Enviado desde mi GT-I9300 usando Tapatalk 2
Click to expand...
Click to collapse
He answered you 3 posts up...
TP.
STAticKY said:
Using max CPU freq at 1400 with lcdfreq freq threshold at 1300. Touch boost turned off.... Baffles me. I've also tryed setting max CPU freq to 1200 and lcdfreq freq threshold to 1300 because of something tema or yank said.. Can't remember what it was off hand.. I'll have a look back
TP.
Click to expand...
Click to collapse
What Zane meant is that in Perseus, there was a touch boost for LCDfreq (I believe to remember sth like that at least), this has nothing to do with CPU touch boost. Whenever you touched the screen, LCDfreq would be bumped to 60Hz no matter what, with the idea to remove lags when actively using the device.
In case Temasek has that in his kernel, that would be something we do not have in either Boeffla nor mine ... LCDfreq scaling is purely based on the CPU freq (as per the tunables), touching the screen will not change anything in our implementation.
Might be worth checking...
JP.
Yank555 said:
What Zane meant is that in Perseus, there was a touch boost for LCDfreq (I believe to remember sth like that at least), this has nothing to do with CPU touch boost. Whenever you touched the screen, LCDfreq would be bumped to 60Hz no matter what, with the idea to remove lags when actively using the device.
In case Temasek has that in his kernel, that would be something we do not have in either Boeffla nor mine ... LCDfreq scaling is purely based on the CPU freq (as per the tunables), touching the screen will not change anything in our implementation.
Might be worth checking...
JP.
Click to expand...
Click to collapse
Yea I don't think he has that implemented. Upon touching the screen while benching, while its at 40hz it stays at 40hz
TP.
Q&A for [INFO] [v1.0 beta5] [07-01-2015] CPU Governor ZZMoove
Some developers prefer that questions remain separate from their main development thread to help keep things organized. Placing your question within this thread will increase its chances of being answered by a member of the community or by the developer.
Before posting, please use the forum search and read through the discussion thread for [INFO] [v1.0 beta5] [07-01-2015] CPU Governor ZZMoove. If you can't find an answer, post it here, being sure to give as much information as possible (firmware version, steps to reproduce, logcat if available) so that you can get help.
Thanks for understanding and for helping to keep XDA neat and tidy!
ZaneZam said:
Hi Guys,
i thought it would be a good idea to put all the infos of the zzmoove governor around here together on one place
for better way to find it, to have a place to dump stuff for future versions and to give support for specific questions.
for now i just copied the allready existend posts here but will edit this further when i have more time
so lets start with the first
initial version:
ZZMoove Governor v0.1
(post from 18th December 2012, 10:27 PM)
Why that “zzmoove” governor and how it works:
I thought it were pretty cool to have one of my favorite governors back from the old SGS1-days on my actual device, so i decided to take the sources and give that a try on top of boeffla kernel. It worked well so this is now the result of that experiment.
More about the internals:
Basically this is the ported SGS1 version of the well known "smoove" governor from the good old midnight kernel from Michael Weingaertner (mialwe) with a modified CPU hotplug implementation of the ktoonservative governor from ktoonesz. The original implementation from ktoonesz worked well but I observed that on idle most of the time only one cpu was going to sleep. Well that was not enough for me so I made a modification to put the other cpu's also to sleep (except cpu0). That means that this governor uses more often only one cpu on idle and as a consequence of that it needs less energy. Depending on System load and governor settings all 4 cores will be instantly up again if it is needed.
In short:
So what you can expect now from this thingy is a battery-friendly behaving hotplug conservative governor which
uses a frequency lookup table for faster upscaling (so called "smooth scaling") So this is more a energy-safer than a performer.
Tuneables/Defaults:
Sampling Rate (default=2) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/sampling_rate
Sampling Down Factor (default=4) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/sampling_down_factor
Up Threshold (default=70) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold
Up Threshold Hotplug (default=68) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold_hotplug
Down Threshold (default=52) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold
Down Threshold Hotplug (default=55) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold_hotplug
Ignore Nice Load (default=0) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/ignore_nice_load
Freqency Step (default=5) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/freq_step
Smooth Up (default=75) tuneable: /sys/devices/system/cpu/cpufreq/zzmoove/smooth_up
Links:
Midnight kernel
KT747 Kernel
Common Infos about Governors,I/O Schedulers etc.
Credits to:
mialwe for his smoove governor
ktoonesz for original hotplug implementation
ZZMoove Governor v0.3
(Post from 25th February 2013, 05:06 PM - "more improvements")
there are now many new possibilities to adjust the governor more precisely via sysfs!
Following new tuneables were introduced in this new version:
The so called "sleep" values which were hardcoded in previous version are now changed
to sysfs-tuneables:
The Smooth Scaling for sleep (Screen off):
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/smooth_up_sleep
possible values from 1 to 100, default: 100
The up/down Threshold for sleep (Screen off):
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold_sleep
possible Values from above "down_threshold_sleep" to 100, default: 90
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold_sleep
possible Values from 11 to under "up_threshold_sleep", default: 44
The Sampling Rate for sleep (Screen off):
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/sampling_rate_sleep_multiplier
possible values 1 or 2, default: 2
The amound of cores which should run at "screen off":
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/hotplug_sleep
possible Values 0 = do not touch cores (the same behaving as in standard setting of zzmoove) 1, 2 or 3
the number of cores to run on screen off. btw. setting "4" doesn't exist as u can use "0" for that setting!
Beside of that you can now change the hotplug threshold per core independently (thx to gsw5700 for the inital idea!)
and turn cores off completely.
For that purpose following tuneables were introduced and are replacing
the old hotplug up/down threshold tuneables:
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold_hotplug1
hotplug up threshold for core 1 - 0 = turn off core 1, possible range from "down_threshold" to 100, default: 68
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold_hotplug2
hotplug up threshold for core 2 - 0 = turn off core 2, possible range from "down_threshold" to 100, default: 68
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/up_threshold_hotplug3
hotplug up threshold for core 3 - 0 = turn off core 3, possible range from "down_threshold" to 100, default: 68
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold_hotplug1
hotplug down threshold for core 1 - possible range from 11 to under "up_threshold", default: 55
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold_hotplug2
hotplug down threshold for core 2 - possible range from 11 to under "up_threshold", default: 55
tuneable -> /sys/devices/system/cpu/cpufreq/zzmoove/down_threshold_hotplug3
hotplug down threshold for core 3 - possible range from 11 to unter "up_threshold", default: 55
Thanks to:
gsw5700 for the initial Idea "hotplug threshold per core".
brijmathew indirectly for the initial idea "just one core at screen off" (i think he did'nt meant exactly that in his post but hey he switched on the led in my head! *gg*)
ZZMoove Governor v0.4
(Post from 1st May 2013, 10:59 PM - "limits")
Changelog:
Frequency Limits:
First of all there is now a (by me so called) "soft" limit function for limiting frequencies at screen off but also at screen on if u wish. however i recommend to set the screen on limit always with the (if provided) max scaling functionality of the kernel as this is the better way of doing it and "works better" for following reasons: touchboost and wake up frequencies can go above that governor-soft-limit (mostly to 800/1000 mhz) because the governor has no control over these "events" and will be "bypassed" by cpufreq driver! nevertheless by setting this soft limit at screen off the use of frequencies higher than the given limit will be strongly reduced and therefore this will reduce power consumption at screen off. and that was the main intention - saving power if u do not use your phone!
For this function following new tuneables were indroduced to set in sysfs (/sys/devices/system/cpu/cpufreq/zzmoove/):
freq_limit_sleep: limit freqency at screen off (possible values 0 disable limit, 200000-1600000, default: 0)
freq_limit: limit freqency at screen on (possible values 0 disable limit, 200000-1600000, default: 0)
Fast Scaling:
As a second feature in this new version i added the so called (again by me *g*) "fast scaling" for faster up/down scaling! This should bring more performance but on the other hand this can be of course a little bit more power consumptive. try it, it makes things snappier and some people reported that with this setting touch boost can then also be disabled which wasn't really "possible" with previous versions of zzmoove governor.
For this function following new tuneables were indroduced to set in sysfs (/sys/devices/system/cpu/cpufreq/zzmoove/):
fast_scaling: fast scaling at screen on (possible values 0 disable or 1 enable, default: 0)
fast_scaling_sleep: fast scaling at screen off (possible values 0 disable or 1 enable, default: 0)
As last "feature" and to complete the "set" the tuneable "freq_step_sleep" was included to be able to change freq step only at screen off
possible settings are the same as the "freq_step" tuneable which are values from 0% (stops freq scaling) to 100% (switches frequencies from lowest to highest frequency and vice versa like ondmand governor) default is 1 in all settings (bat/opt/perf) at screen off.
ZZMoove Governor v0.5 aka "The Beast"
(performance and fixes)
Changelog:
- completely reworked fast scaling functionality. now using a "line jump" logic instead of fixed freq "colums".
fast scaling now in 4 steps and 2 modes possible (mode 1: only fast scaling up and mode2: fast scaling up/down)
- added support for "Dynamic Screen Frequency Scaling" (original implementation into zzmoove governor highly improved by Yank555)
originated by AndreiLux more info: http://forum.xda-developers.com/showpost.php?p=38499071&postcount=3
- re-enabled broken conservative sampling down factor functionality ("down skip" method).
originated by Stratosk - upstream kernel 3.10rc1: https://git.kernel.org/cgit/linux/k...id=refs/tags/v3.10-rc1&qt=author&q=Stratos+Ka
- changed down threshold check to act like it should.
originated by Stratosk - upstream kernel 3.10rc1: https://git.kernel.org/cgit/linux/k...id=refs/tags/v3.10-rc1&qt=author&q=Stratos+Ka
- implemented/ported "early demand" from ondemand governor.
originated by Stratosk - more info: http://www.semaphore.gr/80-latests/98-ondemand-early-demand
- implemented/ported "sampling down momentum" from ondemand governor.
originated by Stratosk - more info: http://www.semaphore.gr/80-latests/80-sampling-down-momentum
- modified some original conservative code parts regarding frequency scaling which should work better now.
originated by DerTeufel1980: https://github.com/DerTeufel/androi...mmit/6bab622344c548be853db19adf28c3917896f0a0
- added the possibility to use sampling down momentum or conservative "down skip" method.
- increased possible max sampling rate sleep multiplier to 4 and sampling down factor to 100000
accordingly to sampling down momentum implementation.
- added frequency search limit for more efficient frequency searching in scaling "table" and for improving
frequency "hard" and "soft" limit handling.
- added cpu idle exit time handling like it is in lulzactive
again work from ktoonsez : https://github.com/ktoonsez/KT747-JB/commit/a5931bee6ea9e69f386a340229745da6f2443b78
description in lulzactive governor: https://github.com/ktoonsez/KT747-J...f2443b78/drivers/cpufreq/cpufreq_lulzactive.c
- fixed a little scaling step mistake and added overclocking frequencies up to 1800 mhz in scaling frequency "tables".
- fixed possible freezes during start/stop/reload of governor and frequency limit change.
- fixed hotplugging logic at online core 0+3 or 0+2 situations and improved hotplugging in general by
removing mutex locks and skipping hotplugging when it is not needed.
- added possibility to disable hotplugging (that's a debugging relict but i thought maybe someone will find that usefull so i didn't remove it)
- try to fix lags when coming from suspend if hotplug limitation at sleep was active by enabling all offline cores during resume.
- code cleaning and documentation.
for this functions following new tuneables were indroduced:
Early Demand:
early_demand -> switch to enable/disable early demand functionality (possible values 0 disable or 1 enable, default: 0)
grad_up_threshold -> scale up frequency if the load goes up in one step of grad up value (possible range from 11 to 100, default 50)
little example for understanding: when the load rises up in one big 50% step then the
frequency will be scaled up immediately instead of wating till up_threshold is reached.
Fast Scaling (improved):
Fast scaling has now 8 levels which at the same time have 2 modes included. Values from 1-4 equals to scaling jumps in the frequency table and uses the Fast Scaling up but normal scaling down mode. Values from 5-8 equals to 1-4 scaling jumps but uses the fast scaling up and fast scaling down mode.
Hotplugging switch:
disable_hotplug -> switch to enable/disable hotplugging (possible values are any value above 0 to disable hotplugging and 0 to
enable it, default 0)
Sampling Down Factor and Sampling Down Momentum:
Description: From the original author of ondemand_sampling_factor David Niemi:
"This improves performance by reducing the overhead of load evaluation and helping the CPU stay
at its top speed when truly busy, rather than shifting back and forth in speed."
And that "Sampling Down Momentum" function from stratosk does this dynamicly now!
sampling_down_max_momentum -> max sampling down factor which should be set by momentum (0 disable momentum, possible range from sampling_down_factor up to MAX_SAMPLING_DOWN_FACTOR, default 0 disabled)
sampling_down_momentum_sensitivity -> how fast the sampling down factor should be switched (possible values from 1 to 500, default 50)
sampling_down_factor -> depending on which mode is active the factor for sampling rate multiplier which influences the whole
sampling rate or the value for stock "down skip" functionality which influences only the down scaling
mechanism (possible values are from 1 to MAX_SMPLING_DOWN_FACTOR, default 1 disabled)
Original conservative "down skip" or "stock" method can be enabled by setting the momentum tuneable to 0. so if momentum is inactive there will be a fallback to the stock method. as the name "down skip" says this method works "slightly" different from the ondemand stock sampling down method (on which momentum was based on). It just skips the scaling down code for the given samples. if u want to completely disable the sampling down functionality u can achieve this by setting sampling down factor to 1. so concluded: setting sampling_down_momentum = 0 and sampling_down_factor = 1 will disable sampling down completely (that is also the governor default setting)
Dynamic Screen Frequency Scaling:
Dynamicly switches the screen frequency to 40hz or 60hz depending on cpu scaling and hotplug settings. For compiling and enabling this functionality u have to do some more modification to the kernel sources, please take a look at AndreiLux Perseus repository and there at following commit: https://github.com/AndreiLux/Perseus-S3/commit/3476799587d93189a091ba1db26a36603ee43519 After adding this patch u can enable the feature by setting "CPU_FREQ_LCD_FREQ_DFS=y" in your kernel config and if u want to check if it is really working at runtime u can also enable the accounting which AndreiLux added by setting LCD_FREQ_SWITCH_ACCOUNTING=y in the kernel config. If all goes well and u have the DFS up and running u can use following tuneables to do some screen magic: (thx to Yank555 for highly extend and improving this!)
lcdfreq_enable -> to enable/disable LCDFreq scaling (possible values 0 disable or 1 enable, default: 0)
lcdfreq_kick_in_down_delay -> the amount of samples to wait below the threshold frequency before entering low display frequency mode (40hz)
lcdfreq_kick_in_up_delay -> the amount of samples to wait over the threshold frequency before entering high display frequency mode (60hz)
lcdfreq_kick_in_freq -> the frequency threshold - below this cpu frequency the low display frequency will be active
lcdfreq_kick_in_cores -> the number of cores which should be online before switching will be active. (also useable in combination
with kickin_freq)
So this version is a kind of "featured by" release as i took (again *g*) some ideas and work from other projects and even some of that work
comes directly from other devs so i wanna thank and give credits:
First of all to stratosk for his great work "sampling down momentum" and "early demand" and for all the code fixes which found their way into
the upstream kernel version of conservative governor! congrats and props on that stratos, happy to see such a nice and talented dev directly
contibuting to the upstream kernel, that is a real enrichment for all of us!
Second to Yank555 for coming up with the idea and improving/completeing (leaves nothing to be desired now *g*) my first
rudimentary implementation of Dynamic Screen Frequency Scaling from AndreiLux (credits for the idea/work also to him at this point!).
Third to DerTeufel1980 for his first implementation of stratosk's early demand functionality into version 0.3 of zzmoove governor
(even though i had to modify the original implementation a "little bit" to get it working properly ) and for some code optimizations/fixes regarding scaling.
Last but not least again to ktoonsez - I "cherry picked" again some code parts of his ktoonservative governor which should improve this governor
too!
ZZMoove Governor 0.5.1a
(bugfixes)
urgend bugfix release:
- fix governor switching issues (deadlocks) which oviously werend fixed in version v0.5
- optimised scaling function a lot (thx and credits to Yank555!)
ZZMoove Governor 0.5.1b (in cooperation with Yank555)
(bugfixes and more optimisations)
Changelog:
- now really fixed the governor switching issues! (gotcha *****! *g*)
- again some changes in scaling logic from Yank555 (thx and credits)
- simplified some tuneables by using already available stuff instead of using redundant code (thx Yank555)
- reduced/optimised hotplug logic and preperation for automatic detection of available cores
(maybe this fixes also the scaling/core stuck problems)
ZZMoove Governor 0.6 (in cooperation with Yank555)
(flexibility)
Changelog:
- removed fixed scaling lookup tables and use the system frequency table instead
changed scaling logic accordingly for this modification (thx and credits to Yank555)
- reduced new hotplug logic loop to a minimum
- again try to fix stuck issues by using seperate hotplug functions out of dbs_check_cpu (credits to ktoonesz)
- added support for 2 and 8 core systems and added automatic detection of cores were it is needed
(for setting the different core modes you can use the macro 'MAX_CORES'. possible values are: 2,4 or 8, default are 4 cores)
reduced core threshold defaults to only one up/down default and use an array to hold all threshold values
- fixed some mistakes in "frequency tuneables" (Yank555):
stop looping once the frequency has been found
return invalid error if new frequency is not found in the frequency table
ZZMoove Governor 0.6a (in cooperation with Yank555)
(scaling logic flexibility)
Changelog:
- added check if CPU freq. table is in ascending or descending order and scale accordingly
(compatibility for systems with 'inverted' frequency table like it is on OMAP4 platform)
thanks and credits to Yank555!
ZZMoove Governor 0.7 aka "Tamed Beast" (in cooperation with Yank555)
(slow down)
Changelog:
- reindroduced the 'old way' of hotplugging and scaling in form of the 'Legacy Mode' (macros for enabling/disabling this done by Yank555, thx!)
NOTE: this mode can only handle 4 cores and a scaling max frequency up to 1800mhz.
- added hotplug idle threshold for a balanced load at CPU idle to reduce possible higher idle temperatures when running on just one core.
(inspired by @JustArchi observations, thx!)
- added hotplug block cycles to reduce possible hotplugging overhead (credits to @ktoonsez)
- added possibility to disable hotplugging only at suspend (inspired by a request of @STAticKY, thx for the idea)
- introduced hotplug frequency thresholds (credits to Yank555)
- hotplug tuneables handling optimized (credits to Yank555)
- added version information tuneable (credits to Yank555)
for this functions following new tuneables were indroduced:
legacy_mode -> for switching to the 'old' method of scaling/hotplugging. possible values 0 to disable, any values above 0 to enable (default is 0)
NOTE: the legacy mode has to be enabled by uncommenting the macro ENABLE_LEGACY_MODE
hotplug_idle_threshold -> amount of load under which hotplugging should be disabled at idle times (respectively at scaling minimum). possible values 0 disable, from 1 to 100 (default is 0)
hotplug_block_cycles -> slow down hotplugging by waiting a given amount of cycles before plugging. possible values 0 disbale, any values above 0 (default is 0)
disable_hotplug_sleep -> same as disable_hotplug but will only take effect at suspend. possible values 0 disable, any values above 0 to enable (default is 0)
up_threshold_hotplug_freq1 -> hotplug up frequency threshold for core1. possible values 0 disable and range from over down_threshold_hotplug_freq1 to max scaling freqency (default is 0)
up_threshold_hotplug_freq2 -> hotplug up frequency threshold for core2. possible values 0 disable and range from over down_threshold_hotplug_freq2 to max scaling freqency (default is 0)
up_threshold_hotplug_freq3 -> hotplug up frequency threshold for core3. possible values 0 disable and range from over down_threshold_hotplug_freq3 to max scaling freqency (default is 0)
down_threshold_hotplug_freq1 -> hotplug down frequency threshold for core1. possible values 0 disable and range from min saling to under up_threshold_hotplug_freq1 freqency (default is 0)
down_threshold_hotplug_freq2 -> hotplug down frequency threshold for core2. possible values 0 disable and range from min saling to under up_threshold_hotplug_freq2 freqency (default is 0)
down_threshold_hotplug_freq3 -> hotplug down frequency threshold for core3. possible values 0 disable and range from min saling to under up_threshold_hotplug_freq3 freqency (default is 0)
version -> show the version of zzmoove governor
ZZMoove Governor 0.7a
(little fix)
Changelog:
- fixed a glitch in hotplug freq threshold tuneables which prevented setting of values in hotplug down freq thresholds when hotplug
up freq thresholds were set to 0. NOTE: enabling the legacy mode in the source is not part of that fix i just forgot to set it back to standard (disabled) before pushing, sry to lazy to fix that fix now
ZZMoove Governor 0.7b
(compatibility improved and forgotten things)
Changelog:
- fixed stuck at max scaling frequency when using stock kernel sources with unmodified cpufreq driver and without any oc capabilities.
- readded forgotten frequency search optimisation in scaling logic (only effective when using governor soft frequency limit)
- readded forgotten minor optimisation in dbs_check_cpu function.
- as forgotten to switch in last version Legacy Mode now again disabled by default
- minor code format and comment fixes
ZZMoove Governor 0.7c
(again compatibility and optimisations)
Changelog:
- frequency search optimisation now fully compatible with ascending ordered system frequency tables (thx to @psndna88 for testing!)
- again minor optimisations at multiple points in dbs_check_cpu function
- code cleaning - removed some unnecessary things and whitespaces nuked - sry for the bigger diff but from now on it will be clean!
- corrected changelog for previous version regarding limits
ZZMoove Governor 0.7d (bugfix!)
(broken things)
Changelog:
- fixed hotplug up threshold tuneables to be able again to disable cores manually via sysfs by setting them to 0
- fixed the problem caused by a "wrong" tuneable apply order of non sticking values in hotplug down threshold tuneables when
hotplug up values are lower than down values during apply.
NOTE: due to this change right after start of the governor the full validation of given values to these tuneables is disabled till
all the tuneables were set for the first time. so if you set them for example with an init.d script or let them set automatically
with any tuning app be aware that there are illogical value combinations possible then which might not work properly!
simply be sure that all up values are higher than the down values and vice versa. after first set full validation checks are enabled
again and setting of values manually will be checked again.
- fixed a typo in hotplug threshold tuneable macros (would have been only a issue in 8-core mode)
- fixed unwanted disabling of cores when setting hotplug threshold tuneables to lowest or highest possible value
which would be a load of 100%/11% in up/down_hotplug_threshold and/or scaling frequency min/max in up/down_hotplug_threshold_freq
Click to expand...
Click to collapse
Thanks for great inputs
Hi!
Where I can read is responsible for what in one way or another option?
Now I'm interested in the inputboost options.
What's the difference between inputboost_cycles and inputboost_punch_cycles?
What give these options: inputboost_punch_epenmove and inputboost_punch_fingermove?
Forgive me if I ask stupid questions))
Thanks!
ilfat12 said:
Hi!
Where I can read is responsible for what in one way or another option?
Now I'm interested in the inputboost options.
What's the difference between inputboost_cycles and inputboost_punch_cycles?
What give these options: inputboost_punch_epenmove and inputboost_punch_fingermove?
Forgive me if I ask stupid questions))
Thanks!
Click to expand...
Click to collapse
hi ilfat12,
you can get more infos about the new features here : https://github.com/zanezam/cpufreq-governor-zzmoove/commit/75c431aeff8354f224e4c688faa1559ac463ab8a
and especially for the info u are searching i want to quote ffolkes description from here http://www.plasmakernel.com/?p=214:
Added native input booster to ZZMoove (based on faux123′s intelliactive)
I’ve added my own port of the native input booster from intelliactive into ZZMoove. Input events such as buttons, touchkeys, and the touchscreen will temporarily modify different ZZMoove settings. Using this allows for extremely conservative governor settings to be used without affecting performance. The boost is made up of two stages; the first is meant to be a very short high speed burst, just a few hundred milliseconds. The second stage is a reduced up threshold that is applied for the duration of the entire boost.
Click to expand...
Click to collapse
in short this means: inputboost_cycles = main switch = duration of whole input boost
inputboost_punch_cycles = duration of the freq "punch" therefore the freq boost duration
inputboost_punch_epenmove and inputboost_punch_fingermove are addons to this functionality
and are influencing the boost of the freq in following way: if one of these tuneables are set then
either on epen move (for devices with epen) or on finger move the punch will be reinitialized constantly
without expiring so actually its behaving like the regular system touch booster which has every device in some
way implemented. so that mode is kind of redundant if u use the system touch booster in addition, exception would be
if u want to overwrite the system booster with the governor boost. if i use it i usually like to use the native booster with
fingerdown enabled thats a good compromise i think.
hope it's somehow more understandable now for u
and hey never be sorry to ask, there are no stupid questions only stupid people who don't ask
Deep Sleep
How do I enable deep sleep mode at zzmove? Because mine is not working
Henriquefeira said:
How do I enable deep sleep mode at zzmove? Because mine is not working
Click to expand...
Click to collapse
you mean the governor "sleep" settings? it depends on the used kernel and how it has implemented zzmoove. they can be switched off by intention in the source code or are disabled automatically because no power management implementation or display state detection is used in the kernel.
but if u mean that u have wake locks and your device doesn't deep sleep then you have another problem which is not related to kernel/governors
in that case u can check what service/app is keeping your device awake with better battery stats for example to find the root cause.
Default profile Set.
Is this possible to make make yang battery extreme (3) profile set by default.....? Every time i set this but changed automatically zzrelax(11).....After unlock my phone....!
Hi ZaneZam. I use your Governor with Boeffla Kernel Beta on my S5.
I use the optimized profile with following additional tunables:
Fast Scaling Sleep Up: 5
Fast Scaling Sleep Down: 5
Fast Scaling Up: 5
Fast Scaling Down: 5
Sampling Rate Sleep Multiplier: 2
I use apps like Facebook or Chrome and than the phone freezes and reboots. This happened 2 times every day.
I checked the Insane profile and that's very similar to the optimized with my tunables. Why it doesn't work?
tuhinxp04 said:
Is this possible to make make yang battery extreme (3) profile set by default.....? Every time i set this but changed automatically zzrelax(11).....After unlock my phone....!
Click to expand...
Click to collapse
Yes it is possible either via a default value in the governor itself or maybe via the tool u use to switch it which u didn't reveal to us
i just can say that there is nothing included from my side which changes a profile "at sleep" so it must be something which is dealing with the tuneables automatically. I can imagine some kind of "at sleep changing settings" mode.
Solvin said:
Hi ZaneZam. I use your Governor with Boeffla Kernel Beta on my S5.
I use the optimized profile with following additional tunables:
Fast Scaling Sleep Up: 5
Fast Scaling Sleep Down: 5
Fast Scaling Up: 5
Fast Scaling Down: 5
Sampling Rate Sleep Multiplier: 2
I use apps like Facebook or Chrome and than the phone freezes and reboots. This happened 2 times every day.
I checked the Insane profile and that's very similar to the optimized with my tunables. Why it doesn't work?
Click to expand...
Click to collapse
hm I don't have to ask if u are using uv/oc, don't I? maybe any chance of a crash log? the S5 seems still disliking zzmoove in some situations.
ZaneZam said:
Yes it is possible either via a default value in the governor itself or maybe via the tool u use to switch it which u didn't reveal to us
i just can say that there is nothing included from my side which changes a profile "at sleep" so it must be something which is dealing with the tuneables automatically. I can imagine some kind of "at sleep changing settings" mode.
hm I don't have to ask if u are using uv/oc, don't I? maybe any chance of a crash log? the S5 seems still disliking zzmoove in some situations.
Click to expand...
Click to collapse
I don't use UV, only OC but without OC it's the same. How can I create a crashlog?
Solvin said:
I don't use UV, only OC but without OC it's the same. How can I create a crashlog?
Click to expand...
Click to collapse
good at least no uv in case of boeffla kernel its quite easy to create a crash log: just let the crash happen with kernel logger enabled (don't forget to enable it beforehand in config app!) then after system is up again right after the crash-reboot. go to action menu->create debug file. this file will be created in boeffla-data folder on your sdcard and is called "debug-date-time.txt" (the coming popup shows it more exactly) send this file to [email protected] then i can see what happend - hopefully, as the S5 crash logs are almost never much meaningful. but maybe u can give me the crucial hint with this log, lets see. the "good thing" for me is that u can reproduce it! thx in advance...
OK I will do that and send it to you. Thank you!
Currently no reboots or freezes. I think that was the problem:
http://forum.cyanogenmod.org/topic/114602-random-reboots/#entry534858
Today I got a Greenify update from playstore with that changes.
I hope that solved my problem.
And it crashed again (reboot). I send you the crashlog via mail
Solvin said:
And it crashed again (reboot). I send you the crashlog via mail
Click to expand...
Click to collapse
thx for the log, it rebooted yes but it didn't crash (at least the kernel did not!)
so in short: the log is clean, no kernel problem in my opinion but one thing: you have UV set to "custom" in boeffla config which means "the setting for custom values"
and which could still contain values u have changed previously (or have changed by a settings preset)! if u really want stock settings then choose "no undervolting"!
Lord boeffla recently changed the naming from "none" to "custom" upon the advice of me because i had one user which had a similar problem like u (i ask about uv because
of a reason, it's "smelling like that" *g*) and this user had previously tested values still active because he thought "none" is "off" so try to choose "no undervolting" then
u really have no uv values set.
hope this stabilizes things for u. otherwise feel free to get back with another log
:fingers-crossed:
ZaneZam said:
thx for the log, it rebooted yes but it didn't crash (at least the kernel did not!)
so in short: the log is clean, no kernel problem in my opinion but one thing: you have UV set to "custom" in boeffla config which means "the setting for custom values"
and which could still contain values u have changed previously (or have changed by a settings preset)! if u really want stock settings then choose "no undervolting"!
Lord boeffla recently changed the naming from "none" to "custom" upon the advice of me because i had one user which had a similar problem like u (i ask about uv because
of a reason, it's "smelling like that" *g*) and this user had previously tested values still active because he thought "none" is "off" so try to choose "no undervolting" then
u really have no uv values set.
hope this stabilizes things for u. otherwise feel free to get back with another log
:fingers-crossed:
Click to expand...
Click to collapse
Damn! I thought it was off, but you are right, my device was UV. But that was a stock setting from the kernel. I hope that was the Problem. Thank you very much ZaneZam.
Samsung Galaxy S5 - G900F
ROM: CM12.1 Nightly
Kernel: Boeffla 2.2-beta2
CP/BL: G900FXXU1BOH4
Recovery: TWRP 2.8.7.0
Can I get some help with my LG G3 zzmove profile 2 /5 settings? It runs good with the default settings, but I'd like to get some more performance and speed and get rid of the lag I have. SlimSaber 5.1.1 Nebula 12.0 using Kernel Adiutor
Someone explain about profile? i read it many time but dont understand it.
thaibinh262 said:
Someone explain about profile? i read it many time but dont understand it.
Click to expand...
Click to collapse
profiles are just predefined settings but included in the governor itself and switchable via a tuneable.
nytebird said:
Can I get some help with my LG G3 zzmove profile 2 /5 settings? It runs good with the default settings, but I'd like to get some more performance and speed and get rid of the lag I have. SlimSaber 5.1.1 Nebula 12.0 using Kernel Adiutor
Click to expand...
Click to collapse
still relevant? . sry completely missed that I would suggest to try all other settings then and see... if none of them suits u u anyway have to adjust the tuneables in particular to. make things better.. nut this is a huge topic then.
music_state
What does music_state tuneable do?
Hey guys, I'm back with a new KERNEL for both Variants of the Tab S 10.5 (T800 and T805)!
Some guys probably know me from the IronRom . I decided myselfe to create a custom kernel for our really great Tab s 10.5, I'm getting better and better at this stuff, I don't thought that
It is basically the normal kernel with some modifications for better performance and hopefully also batterylife. It is for the stock kitkat (4.4.2) touchwiz and not for cyanogenmod or something else.
I excuse all devs here visiting my github page, it is such a mess (with the commits)! I know it, but I'm doing this the first time, so hopefully you will forgive me.
The kernel is pretty stable, I just call it a beta version, because I can't test the T800 version, so if with thisone also all works great -> stable
IF YOU FOLLOW MY STEPS BELOW, YOU WILL MAY LOSE YOUR WARRANTY, KNOX WILL DISPLAY 0x1! I'M NOT RESPONSIBLE FOR ANY DAMAGED DEVICE!
You can try to use the kernel adiutor app, or just the preinstalled sTweaks, but not both at the same time! This will cause problems.
Notice, V2.0 and onwards is only for TW lollipop and not for kitkat anymore!!
Features of my Kernel::- Built with latest 5.2 Toolchain compiled by myself!!
- Latest Kernel version 3.4.109, includes all updates from linux mainstream, patched it form 3.4.39 up to 3.4.109 (was a lot of work)
- Choose between different CPU governors: Interactive (default), Powersave, Performance, Userspace, Conservative, Intellidemand, Intelliactive, Ondemand, Adaptive, Abyssplug, AbyssplugV2, Badass, DanceDance, ZZmove, Nightmare, Wheatley, Lionheart, Darkness, PegasusQ and Intellimm
- Built with latest ramdisk sources from samsung
- Kernelsource from T805XXU1BOG2
- Underclock to 200MHz and Overclock to 2.0GHz
- GPU works from 100MHz to 733MHz (default)
- I/O schedulers: ROW (Default), cfq, No-op, Deadline, Test, BFQ, FIOPS, SIO, VR, ZEN, FIFO and SIOplus
- Readahead can be set
- UKSM (Ultar Kernel Samepage Merging)
- Gentlefairsleeper and ARCH power
- Android Logger
- data and cache f2fs support!
- Init.d Support
- Busybox support
- Full STweaks support
- Charging Control
- Cpufreq voltcontrol
- ZRam and Swap
- Allow ADB-Insecure
- Low Memory Kill
- TCP (Network) control: Cubic (default), Reno, Bic, Westwood, Highspeed, Hybla, HTCP, Vegas, Veno, Scalable, LP, Yeah and Illinois
- SeLinux is set to permissive
- Compiled as small as it could be (just around 6MB)
Download:In the second post
Googy Max STweaks
Bugs/Problems:-sTweaks can't enable the right GPU over and underclock freqency
-Some other stweaks stuff, you will see
-Didn't tried the voltage table
Instructions:
If you want to install the Kernel, follow this:
1. Install a custom recovery for your tablet, like this one here: TWRP Recovery
2. Follow the instructions on the page above, until you get a working recovery
3. Download the Kernel from below and copy it to your external SD Card
4. Reboot to your recovery by pressing volume up, home button and power button at the same time.
5. Install zip, and select the kernel
6. Wipe cache and dalvik cache (recommand)
7. Reboot
Support:If you like my work, please hit a thanks down on my posts. A thanks is enough!:highfive: If you really really really really really like my work, you can donate something to me, but it is not necessary. I created a paypal account, just in case, someone would give me a small donation. :good:
As I said, you don't have to give me something, but this keeps me motivated to built better roms and keep updating everything. It's your choice, and I'm very thankfull for every donation! No matter how big it is! Thank you so much for supporting me, cheers and have a nice day :fingers-crossed:
Donators for the Kernel:- @Hookmt Thank you very much for your support giving to me! I really love that and it also wasn't the first time you donate something to me! Thank you so much, I really really really appreciat that mate. Transaction number: 42P214019W495221S
Credits/Thanks:- Samsung for sources
- @Christopher83 for the compiler
- @UpInTheAir for the work he already did in his own kernel (could use some of his commits on github (opensource) and see what he did when I didn't know what I did wrong). He also inspired me to work on my own stuff and kernel, thank you very much!
- @googy_anas, without him, I would not have a working kernel here, he did so much for me and also for his own kernel! He
let me use everything he already did, I got so much stuff from his page and included it in my kernel. I'm so thankfull for all the support he gave to me! I know a thank you isn't enough, but I wanted to write it here.
- @googy_anas (again this great man!) and @kryten2k35 thank you so much for let me using your stweaks app! Great work you have done on thatone!
- @faux123 for all the great stuff he did for the kernels
- @Yank555
If you want to take my work and need it somewhere, or do other things with it, please ask me first for the permission. Otherwise you are not allowed to take it! Thank you !
XDA:DevDB Information
Stock Based Kernel for Tab S 10.5, Kernel for the Samsung Galaxy Tab S
Contributors
Tkkg1994, @googy_anas
Source Code: https://github.com/Tkkg1994/IronKernel
Kernel Special Features:
Version Information
Status: Stable
Current Stable Version: V2.5
Stable Release Date: 2015-10-13
Current Beta Version: 1.0
Beta Release Date: 2015-02-19
Created 2015-02-20
Last Updated 2015-10-12
Changelog:
Kitkat
IronKernel Beta V1.0 on 20.02.2015:
Initial release!
Ironkernel V1.1 26.02.2015:
-Temporarly removed Intelli-Plug
-Added voltcontroll for CPU
-Added Stweaks support and Stweaks app (all credits and stuff goes to @googy_anas and @Kryten2k35
-Hell lot of improvements
-Init.d support and busybox support
-GPU overclock to 733MHz
-CPU overclock to 2.1GHz (sorry, I couldn't get it stable )
-for more what I had done, visit here: Commits IronKernel
-Added fast charge support
IronKernel V1.2 01.03.2015:
- Built with latest toolchain (2015.02) by christopher83
- Use frandom from now
- Enable dynamic page writeback with earlysuspend
- Better battery charging control (kernel and stweaks)
- Auto install the right sTweaks version
- Reduce overestimating rq->avg_idle
- Optimize find_busiest_queue()
- Some CPUfreq optimizations
- Dynamic sync control with earlysuspend support
- Lowmemorykiller: implement task's adj rbtree
- Check free memory when tasks switch to background
- Dynamic FSync
- SOO Much more but I don't remember all
- After flashing the kernel, it will be successfull, but then show an error (becaus of mounting partition) don't worry, just reboot. Just ignore it
Changelog V1.3.5 08.03.15:
- Prerelease of V1.3 was on the ironrom
- Script auto-removes all knox containors/apps
- cpufreq: Retain only online cpus in managed_policy->cpus
- cpufreq: make the "scaling_cur_freq" sysfs entry pollable
- cpufreq: Make the "scaling_governor" sysfs node pollable
- cpufreq: Save and restore min and max frequencies
- fix some missing stuff with default governor
- cpufreq: Notify governors when cpus are hot-[un]plugged
- Updated nightmare and zzmoove governors
- net: ipv6: Add a sysctl to make optimistic addresses useful candidates
- fs/proc/task_mmu.c: add user-space support for resetting mm
- net: ipv6: allow choosing optimistic addresses with use_optimistic
- sched/idle: Avoid spurious wakeup IPIs
- cpufreq: Return directly in __cpufreq_get if policy is NULL
- new relation between governors
- ARM: 8226/1: cacheflush: get rid of restarting block
Changelog V1.4 23.03.15:
- It is simply to much to write everything here.. what I did
- Wolfson Control for the sound on our Tab S
- Added voltage control (doesen't work 100%)
- sched: Add an rq migration call-back to sched_class
- sched: Account for blocked load waking back up …
- sched: Normalize tg load contributions against runnable time
- sched: Refactor update_shares_cpu() -> update_blocked_avgs()
- sched/fair: Set se->vruntime directly in place_entity()
- sched: provide per cpu-cgroup option to notify on migrations
- sched: Make sure to not re-read variables after validation
- sched: Add WAKEUP_PREEMPTION feature flag, on by default
And this goes on for like 2 or 3 pages lol So the changelog is tooooooo long.
Lollipop
Changelog V2.0 10.04.15:
- Full lollipop compatible! (not with kitkat anymore!)
- Support CIFS
- Overclock to 2.0 GHz (other will be back soon)
- Include all features of all previous releases! Such as stweask, overclocked gpu etc.
- Based on latest samsung opensource T800XXU1BOCC
Was a lot of work to port all to Lollipop
If you got problems with sTweaks showing "loaded", but nothing happen, go to a root explorer, navigate to system/xbin, copy Busybox and past it in /sbin direction!
Changelog V2.1 16.05.15:
- Fix stweaks problems
- Voltage is still NOT working
- add tripndroid scheduler
- add row scheduler
- add and enable UKSM (ultra kernel samepage merging)
- update ramdisk to latest versions
- f2fs
- update linux to 3.4.107
- more stuff I may forgot
Changelog V2.2 08.07.15:
- Update to latest linux mainstream (3.4.108)
- Updated ramdisk source to OE3
- New source patches from official kernel opensource center (samsung)
- this does only work on the newer bases, as BOE3, the old ones will get a bootloop!
Changelog V2.2.5 16.07.15:
- Fixed "slow" charging (Thanks to AndreiLux and UpInTheAir!)
- Incrased sound so it will be a louder by default
- Some other small ramdisk fixes
Changelog V2.3 08.09.15:
- Kernel Rebased on latest OG2 kernel source
- Fix some heating issues that where reported
- Ramdisk update to OG2
- Fully support f2fs in /data and /cache
- Build with latest 4.9.4 toolchain
Changelog V2.4 05.10.15:
- Update to 5.2 toolchain compiled by myself!
- updated to 3.4.109 linux
- ftrace: Make all inline tags also include notrace
- compiler-gcc4.h: correct verion check for __compiletime_error
- compiler.h: add __visible
- compiler{,-gcc4}.h, bug.h: Remove duplicate macros
- some more optimisations concerning compiler
- msm: cpufreq: Only apply driver limits for scaling_min/max_freq writes
- drivers: cpufreq: Send a uevent when governor changes
- cpufreq: Save user policy min/max instead of policy min/max during hotplug
- cpufreq: Fix broken uevents for cpufreq governor and cpu devices
- cpufreq: Always allow update of user policy
- drivers: cpufreq: Upstream optimizations
- cpufreq: Export user_policy min/max
- cpufreq: Add policy notifiers
- cpufreq: Simplify cpufreq_add_dev()
- some more cpufreq things that I made
- cpufreq_stats: do not remove sysfs files if frequency table is not present
- sched/numa: Rewrite the CONFIG_NUMA sched domain support
- sched/numa: Fix the new NUMA topology bits
- sched/numa: Don't scale the imbalance
- sched/debug: Fix printing large integers on 32-bit platforms
- sched: Remove stale power aware scheduling remnants and dysfunctional knobs
- f2fs: update to latest version
- tima debug log disabled
- uksm: disabled by default
Changelog V2.5 13.10.15:
- enabled tima debug again
- fixed some Random reboots people had
- added pegasusq cpugovernor
- arm/crypto: Add optimized AES and SHA1 routines
- added cifs, nfs, exportfs, cdrom, all ramdisk support (joystick too)
- ARM: 7626/1: arm/crypto: Make asm SHA-1 and AES code Thumb-2 compatible
- ARM: 7723/1: crypto: sha1-armv4-large.S: fix SP handling
- ARM: 8119/1: crypto: sha1: add ARM NEON implementation
- ARM: 8120/1: crypto: sha512: add ARM NEON implementation
- a lot of other crypto optimisations (like 10 patches)
- cpufreq: Move get_cpu_idle_time() to cpufreq.c
- workqueue: set delayed_work->timer function on initialization
- workqueue: don't use WQ_HIGHPRI for unbound workqueues
- workqueue: factor out worker_pool from global_cwq
- workqueue: use @pool instead of @gcwq or @Cpu where applicable
- workqueue: separate out worker_pool flags
- workqueue: introduce NR_WORKER_POOLS and for_each_worker_pool()
- workqueue: reimplement WQ_HIGHPRI using a separate worker_pool
- hashtable: introduce a small and naive hashtable
- workqueue: use new hashtable implementation
- workqueue: drop @bind from create_worker()
- much more workqueue updates, to see them all, please visit here: Github Kernel Updates
Governors and I/O Scheduler:
Original Thread: Governor Explained, all credits go to @stempox
1: OnDemand
2: OndemandX
3: Performance
4: Powersave
5: Conservative
6: Userspace
7: Min Max
8: Interactive
9: InteractiveX
10: Smartass
11: SmartassV2
12: Scary
13: Lagfree
14: Smoothass
15: Brazilianwax
16: SavagedZen
17: Lazy
18: Lionheart
19: LionheartX
20: Intellidemand
21: Hotplug
22: BadAss
23: Wheatley
24: Lulzactive
25: Pegasusq/Pegasusd
26: hotplugx
27: AbissPlug
28: MSM DCVS
29: IntelliActive
30: Adaptive
31: Nightmare
32: ZZmove
1: OnDemand Governor:
This governor has a hair trigger for boosting clockspeed to the maximum speed set by the user. If the CPU load placed by the user abates, the OnDemand governor will slowly step back down through the kernel's frequency steppings until it settles at the lowest possible frequency, or the user executes another task to demand a ramp.
OnDemand has excellent interface fluidity because of its high-frequency bias, but it can also have a relatively negative effect on battery life versus other governors. OnDemand is commonly chosen by smartphone manufacturers because it is well-tested, reliable, and virtually guarantees the smoothest possible performance for the phone. This is so because users are vastly more likely to ***** about performance than they are the few hours of extra battery life another governor could have granted them.
This final fact is important to know before you read about the Interactive governor: OnDemand scales its clockspeed in a work queue context. In other words, once the task that triggered the clockspeed ramp is finished, OnDemand will attempt to move the clockspeed back to minimum. If the user executes another task that triggers OnDemand's ramp, the clockspeed will bounce from minimum to maximum. This can happen especially frequently if the user is multi-tasking. This, too, has negative implications for battery life.
2: OndemandX:
Basically an ondemand with suspend/wake profiles. This governor is supposed to be a battery friendly ondemand. When screen is off, max frequency is capped at 500 mhz. Even though ondemand is the default governor in many kernel and is considered safe/stable, the support for ondemand/ondemandX depends on CPU capability to do fast frequency switching which are very low latency frequency transitions. I have read somewhere that the performance of ondemand/ondemandx were significantly varying for different i/o schedulers. This is not true for most of the other governors. I personally feel ondemand/ondemandx goes best with SIO I/O scheduler.
3: Performance Governor:
This locks the phone's CPU at maximum frequency. While this may sound like an ugly idea, there is growing evidence to suggest that running a phone at its maximum frequency at all times will allow a faster race-to-idle. Race-to-idle is the process by which a phone completes a given task, such as syncing email, and returns the CPU to the extremely efficient low-power state. This still requires extensive testing, and a kernel that properly implements a given CPU's C-states (low power states).
4: Powersave Governor:
The opposite of the Performance governor, the Powersave governor locks the CPU frequency at the lowest frequency set by the user.
5:Conservative Governor:
This biases the phone to prefer the lowest possible clockspeed as often as possible. In other words, a larger and more persistent load must be placed on the CPU before the conservative governor will be prompted to raise the CPU clockspeed. Depending on how the developer has implemented this governor, and the minimum clockspeed chosen by the user, the conservative governor can introduce choppy performance. On the other hand, it can be good for battery life.
The Conservative Governor is also frequently described as a "slow OnDemand," if that helps to give you a more complete picture of its functionality.
6: Userspace Governor:
This governor, exceptionally rare for the world of mobile devices, allows any program executed by the user to set the CPU's operating frequency. This governor is more common amongst servers or desktop PCs where an application (like a power profile app) needs privileges to set the CPU clockspeed.
7: Min Max
well this governor makes use of only min & maximum frequency based on workload... no intermediate frequencies are used.
8: Interactive Governor:
Much like the OnDemand governor, the Interactive governor dynamically scales CPU clockspeed in response to the workload placed on the CPU by the user. This is where the similarities end. Interactive is significantly more responsive than OnDemand, because it's faster at scaling to maximum frequency.
Unlike OnDemand, which you'll recall scales clockspeed in the context of a work queue, Interactive scales the clockspeed over the course of a timer set arbitrarily by the kernel developer. In other words, if an application demands a ramp to maximum clockspeed (by placing 100% load on the CPU), a user can execute another task before the governor starts reducing CPU frequency. This can eliminate the frequency bouncing discussed in the OnDemand section. Because of this timer, Interactive is also better prepared to utilize intermediate clockspeeds that fall between the minimum and maximum CPU frequencies. This is another pro-battery life benefit of Interactive.
However, because Interactive is permitted to spend more time at maximum frequency than OnDemand (for device performance reasons), the battery-saving benefits discussed above are effectively negated. Long story short, Interactive offers better performance than OnDemand (some say the best performance of any governor) and negligibly different battery life.
Interactive also makes the assumption that a user turning the screen on will shortly be followed by the user interacting with some application on their device. Because of this, screen on triggers a ramp to maximum clockspeed, followed by the timer behavior described above.
9: InteractiveX Governor:
Created by kernel developer "Imoseyon," the InteractiveX governor is based heavily on the Interactive governor, enhanced with tuned timer parameters to better balance battery vs. performance. The InteractiveX governor's defining feature, however, is that it locks the CPU frequency to the user's lowest defined speed when the screen is off.
10: Smartass
Is based on the concept of the interactive governor.
I have always agreed that in theory the way interactive works – by taking over the idle loop – is very attractive. I have never managed to tweak it so it would behave decently in real life. Smartass is a complete rewrite of the code plus more. I think its a success. Performance is on par with the “old” minmax and I think smartass is a bit more responsive. Battery life is hard to quantify precisely but it does spend much more time at the lower frequencies.
Smartass will also cap the max frequency when sleeping to 352Mhz (or if your min frequency is higher than 352 – why?! – it will cap it to your min frequency). Lets take for example the 528/176 kernel, it will sleep at 352/176. No need for sleep profiles any more!"
11: SmartassV2:
Version 2 of the original smartass governor from Erasmux. Another favorite for many a people. The governor aim for an "ideal frequency", and ramp up more aggressively towards this freq and less aggressive after. It uses different ideal frequencies for screen on and screen off, namely awake_ideal_freq and sleep_ideal_freq. This governor scales down CPU very fast (to hit sleep_ideal_freq soon) while screen is off and scales up rapidly to awake_ideal_freq (500 mhz for GS2 by default) when screen is on. There's no upper limit for frequency while screen is off (unlike Smartass). So the entire frequency range is available for the governor to use during screen-on and screen-off state. The motto of this governor is a balance between performance and battery.
12: Scary
A new governor wrote based on conservative with some smartass features, it scales accordingly to conservatives laws. So it will start from the bottom, take a load sample, if it's above the upthreshold, ramp up only one speed at a time, and ramp down one at a time. It will automatically cap the off screen speeds to 245Mhz, and if your min freq is higher than 245mhz, it will reset the min to 120mhz while screen is off and restore it upon screen awakening, and still scale accordingly to conservatives laws. So it spends most of its time at lower frequencies. The goal of this is to get the best battery life with decent performance. It will give the same performance as conservative right now, it will get tweaked over time.
13: Lagfree:
Lagfree is similar to ondemand. Main difference is it's optimization to become more battery friendly. Frequency is gracefully decreased and increased, unlike ondemand which jumps to 100% too often. Lagfree does not skip any frequency step while scaling up or down. Remember that if there's a requirement for sudden burst of power, lagfree can not satisfy that since it has to raise cpu through each higher frequency step from current. Some users report that video playback using lagfree stutters a little.
14: Smoothass:
The same as the Smartass “governor” But MUCH more aggressive & across the board this one has a better battery life that is about a third better than stock KERNEL
15: Brazilianwax:
Similar to smartassV2. More aggressive ramping, so more performance, less battery
16: SavagedZen:
Another smartassV2 based governor. Achieves good balance between performance & battery as compared to brazilianwax.
17: Lazy:
This governor from Ezekeel is basically an ondemand with an additional parameter min_time_state to specify the minimum time CPU stays on a frequency before scaling up/down. The Idea here is to eliminate any instabilities caused by fast frequency switching by ondemand. Lazy governor polls more often than ondemand, but changes frequency only after completing min_time_state on a step overriding sampling interval. Lazy also has a screenoff_maxfreq parameter which when enabled will cause the governor to always select the maximum frequency while the screen is off.
18: Lionheart:
Lionheart is a conservative-based governor which is based on samsung's update3 source.
The tunables (such as the thresholds and sampling rate) were changed so the governor behaves more like the performance one, at the cost of battery as the scaling is very aggressive.
19: LionheartX
LionheartX is based on Lionheart but has a few changes on the tunables and features a suspend profile based on Smartass governor.
20: Intellidemand:
Intellidemand aka Intelligent Ondemand from Faux is yet another governor that's based on ondemand. Unlike what some users believe, this governor is not the replacement for OC Daemon (Having different governors for sleep and awake). The original intellidemand behaves differently according to GPU usage. When GPU is really busy (gaming, maps, benchmarking, etc) intellidemand behaves like ondemand. When GPU is 'idling' (or moderately busy), intellidemand limits max frequency to a step depending on frequencies available in your device/kernel for saving battery. This is called browsing mode. We can see some 'traces' of interactive governor here. Frequency scale-up decision is made based on idling time of CPU. Lower idling time (<20%) causes CPU to scale-up from current frequency. Frequency scale-down happens at steps=5% of max frequency. (This parameter is tunable only in conservative, among the popular governors)
To sum up, this is an intelligent ondemand that enters browsing mode to limit max frequency when GPU is idling, and (exits browsing mode) behaves like ondemand when GPU is busy; to deliver performance for gaming and such. Intellidemand does not jump to highest frequency when screen is off.
21: Hotplug Governor:
The Hotplug governor performs very similarly to the OnDemand governor, with the added benefit of being more precise about how it steps down through the kernel's frequency table as the governor measures the user's CPU load. However, the Hotplug governor's defining feature is its ability to turn unused CPU cores off during periods of low CPU utilization. This is known as "hotplugging."
22: BadAss Governor:
Badass removes all of this "fast peaking" to the max frequency. On a typical system the cpu won't go above 918Mhz and therefore stay cool and will use less power. To trigger a frequency increase, the system must run a bit @ 918Mhz with high load, then the frequency is bumped to 1188Mhz. If that is still not enough the governor gives you full throttle. (this transition should not take longer than 1-2 seconds, depending on the load your system is experiencing)
Badass will also take the gpu load into consideration. If the gpu is moderately busy it will bypass the above check and clock the cpu with 1188Mhz. If the gpu is crushed under load, badass will lift the restrictions to the cpu.
23: Wheatley:
Building on the classic 'ondemand' governor is implemented Wheatley governor. The governor has two additional parameters:
target_residency - The minimum average residency in µs which is considered acceptable for a proper efficient usage of the C4 state. Default is 10000 = 10ms.
allowed_misses - The number sampling intervals in a row the average residency is allowed to be lower than target_residency before the governor reduces the frequency. This ensures that the governor is not too aggressive in scaling down the frequency and reduces it just because some background process was temporarily causing a larger number of wakeups. The default is 5.
Wheatley works as planned and does not hinder the proper C4 usage for task where the C4 can be used properly .
For internet browsing the time spend in C4 has increased by 10% points and the average residency has increased by about 1ms. I guess these differences are mostly due to the different browsing behaviour (I spend the last time more multi-tabbing). But at least we can say that Wheatley does not interfere with the proper use of the C4 state during 'light' tasks. For music playback with screen off the time spend in C4 is practically unchanged, however the average residency is reduced from around 30ms to around 18ms, but this is still more than acceptable.
So the results show that Wheatley works as intended and ensures that the C4 state is used whenever the task allows a proper efficient usage of the C4 state. For more demanding tasks which cause a large number of wakeups and prevent the efficient usage of the C4 state, the governor resorts to the next best power saving mechanism and scales down the frequency. So with the new highly-flexible Wheatley governor one can have the best of both worlds.
Obviously, this governor is only available on multi-core devices.
24: Lulzactive:
Lulzactive:
This new find from Tegrak is based on Interactive & Smartass governors and is one of the favorites.
Old Version: When workload is greater than or equal to 60%, the governor scales up CPU to next higher step. When workload is less than 60%, governor scales down CPU to next lower step. When screen is off, frequency is locked to global scaling minimum frequency.
New Version: Three more user configurable parameters: inc_cpu_load, pump_up_step, pump_down_step. Unlike older version, this one gives more control for the user. We can set the threshold at which governor decides to scale up/down. We can also set number of frequency steps to be skipped while polling up and down.
When workload greater than or equal to inc_cpu_load, governor scales CPU pump_up_step steps up. When workload is less than inc_cpu_load, governor scales CPU down pump_down_step steps down.
Example:
Consider
inc_cpu_load=70
pump_up_step=2
pump_down_step=1
If current frequency=200, Every up_sampling_time Us if cpu load >= 70%, cpu is scaled up 2 steps - to 800.
If current frequency =1200, Every down_sampling_time Us if cpu load < 70%, cpu is scaled down 1 step - to 1000.
25: Pegasusq/Pegasusd
The Pegasus-q / d is a multi-core based on the Ondemand governor and governor with integrated hot-plugging.
Ongoing processes in the queue, we know that multiple processes can run simultaneously on. These processes are active in an array, which is a field called "Run Queue" queue that is ongoing, with their priority values arranged (priority will be used by the task scheduler, which then decides which process to run next).
To ensure that each process has its fair share of resources, each running for a certain period and will eventually stop and then again placed in the queue until it is your turn again. If a program is terminated, so that others can run the program with the highest priority in the current queue is executed.
26: hotplugx
It 'a Hotplug modified and optimized for the suspension in off-screen
27: AbissPlug
It 'a Governor derived hotplug, it works the same way, but with the changes in savings for a better battery.
28: MSM DCVS
a very efficient and wide range of Dynamic Clock and
Voltage Scaling (DCVS) which addresses usage models from
active standby to mid and high level processing requirements.
A Krait CPU can smoothly scale from low power, low
leakage mode to blazingly fast performance.
Believe it's a governor that is mfg'd by qualcomm to utilize new on chip features.
MSM is the prefix for the SOC (MSM8960) and DCVS is Dynamic Clock and Voltage Scaling. Makes sense, MSM-DCVS
29: IntelliActive
Based off Google's Interactive governor with the following enhancements:
1. self-boost capability from input drivers (no need for PowerHAL assist)
2. two phase scheduling (idle/busy phases to prevent from jumping directly to max freq
3. Checks for offline cpus and short circuits some unnecessary checks to improve code execution paths
30: Adaptive
This driver adds a dynamic cpufreq policy governor
designed for latency-sensitive workloads and also for demanding
performance.
This governor attempts to reduce the latency of clock
increases so that the system is more responsive to
interactive workloads in loweset steady-state but to
to reduce power consumption in middle operation level level up
will be done in step by step to prohibit system from going to
max operation level.
31: Nightmare
A PegasusQ modified, less aggressive and more stable. A good compromise between performance and battery.
In addition to the SoD is a prevention because it usually does not hotplug.
32: ZZmove
ZZmove Governor optimized for low power consumption with the screen off, with particular attention to the limitation of consumption applications in the background with the screen off, such as listening to music. It has three settings: battery saver, balanced and performance. In addition to a performance boost, there is also the governor zzmove optimized.
I/O Scheduler: Thread:I/O Scheduler explained
The Scheduler is an algorithm that, given a set of requests for access to a resource, establishing a temporal order for the execution of such requests, favoring those that meet certain criteria in order to optimize access to that resource.
The difference between the various scheduler is the focus on certain criteria rather than on others.
The choice of a given scheduler does not produce visible changes so as to the choice of the governor, but still provides some improvements.
As usual schedulers are personally tested to find one that best suits your needs.
Deadline
It aims to provide a deadline, a deadline for all requests in order to avoid undesirable phenomena such as the "starvation" or the eternal waiting for some requests that occurs when one or more background processes are left indefinitely in the queue the ready, because there is always at least one of the highest priority ready process.
V (r)
The next request is performed according to the distance from the last request. In the network running good opinions about this scheduler.
No-op
Push all requests in a single queue simply by their arrival order, grouping together those contiguous.
SIO
E 'the scheduler simpler, does not make any type of sort, is intended only for the purpose of obtaining a low-latency, ie to reduce the amount of time that elapses between the instant at which the request is generated and that in which the request is satisfied.
CFQ
Order requests of different processes in queues for each queue type and assigns a specific interval of time whose duration depends on the priorities assigned to processes. Can be considered the Ondemand the scheduler, the scheduler is in fact more balanced, doing its job in an honest manner.
BFQ
It 's based on CFQ but, instead of the intervals of time, assigns a part of the bandwidth of the disc to each process running in a proportional manner.
Anticipatory
Order requests based on criteria predictive, that puts the demands paused for a short period of time in anticipation that more of this to come to aggregate them.
ADAPTIVE ANTICIPATORY SCHEDULER
For the anticipatory scheduler, we scale up the anticipation timeout (antic expire) using the latency scaling factor over time. When the virtual disk latencies are low a small scaling of the timeout is sucient to prevent deceptive idleness, whereas when the latencies are high a larger scaling of the timeout value may be required to achieve the same. Note that such dynamic setting of the timeout value ensures that we attain a good trade-o between throughput (lost due to idling) and deceptive idleness mitigation. Setting a high value for the scaling factor (increasing idling time) only happens when the disk service latencies themselves are higher. This may not necessarily cause a signicant loss in throughput, because submitting a request from another process instead of idling is not going to improve throughput if the virtual disk itself does not get any faster than it is at the current period. A higher anticipation timeout might also be capable of absorbing process scheduling eects inside the VM. The results for the adaptive anticipatory scheduler are shown in Figure 2. The read time with our modied implementation (third bar in the dierent scheduler combinations) shows that it is possible to mitigate the eects of deceptive idleness by adapting the timeout. An interesting related observation is that the level to which the improve- ment is possible varies for dierent Domain-0 schedulers; noop - 39%, anticipatory - 67% and cfq - 36%. This again points to the fact that the I/O scheduler used in Domain-0 is important for the VM's ability in enforcing I/O scheduling guarantees. Dierent Domain-0 I/O schedulers likely have a dierent service latency footprint inside the VMs, contributing to dierent levels of improvement.
ROW
Row stands for READ Over WRITE which is the main requests dispatch policy of this algorithm. The ROW IO scheduler was developed with the mobile devices needs in mind. In mobile devices we favor user experience upon everything else, thus we want to give READ IO requests as much priority as possible. In mobile devices we won't have as much parallel threads as on desktops. Usually it's a single thread or at most 2 simultaneous working threads for read & write. Favoring READ requests over WRITEs decreases the READ latency greatly.
The main idea of the ROW scheduling policy is: If there are READ requests in pipe - dispatch them but don't starve the WRITE requests too much. Bellow you'll find a small comparison of ROW to existing schedulers. The test that was run for these measurements is parallel read and write.
FIOS
Flash-based solid-state drives (SSDs) have the potential to eliminate the I/O bottlenecks in data-intensive applications However the large performance discrepancy between Flash reads and writes introduces challenges for fair resource usage. Further, existing fair queuing and quanta-based I/O schedulers poorly manage the I/O anticipation for Flash I/O fairness and efficiency. Some also suppress the I/O parallelism which causes substantial performance degradation on Flash. This paper develops FIOS, a new Flash I/O scheduler that attains fairness and high efficiency at the same time. FIOS employs a fair I/O time-slice management with mechanisms for read preference, parallelism, and fairness-oriented I/O anticipation. Evaluation demonstrates that FIOS achieves substantially better fairness and efficiency compared to the Linux CFQ scheduler, the SFQ(D) fair queuing scheduler, and the Argon quanta-based scheduler on several Flash-based storage devices (including a CompactFlash card in a low-power wimpy node). In particular, FIOS reduces the worst-case slowdown bya factor of 2.3 or more when the read-only SPECweb workload runs together with the write-intensive TPC
Sweet! In terms of battery life, did it improve for you? You made one of the best TW roms and now this great job!
DUHAsianSKILLZ said:
Sweet! In terms of battery life, did it improve for you? You made one of the best TW roms and now this great job!
Click to expand...
Click to collapse
Thank you! I think we have less devs here in Tab S section, so I have to do something here
I couldn't test the batterylife till now, I was testing whole 2 weeks to get the kernel working (got holidays) So I will report you back!
Intelli-Plug speeks for better batterylife, it shuts down 3 cores if they are not needed (screen off = only 1 core working with 200Mhz or something)
Tkkg1994 said:
Thank you! I think we have less devs here in Tab S section, so I have to do something here
I couldn't test the batterylife till now, I was testing whole 2 weeks to get the kernel working (got holidays) So I will report you back!
Intelli-Plug speeks for better batterylife, it shuts down 3 cores if they are not needed (screen off = only 1 core working with 200Mhz or something)
Click to expand...
Click to collapse
Thanks! Ill downgrade tommrow or so back to kitkat and try xnote rom. I can try your kernel when I do that.
DUHAsianSKILLZ said:
Thanks! Ill downgrade tommrow or so back to kitkat and try xnote rom. I can try your kernel when I do that.
Click to expand...
Click to collapse
Always a pleasure to have you as a user!
Not the ironrom
Wrong thread
Mokum020 said:
Thank you for your great work (again)!
Your kernel is running perfect here on T805 with X-Note.
would love to be able to try 2.1/2.2Ghz, 2.1Ghz is no problem for my Tab S with SkyHigh.
View attachment 3174370View attachment 3174371
Click to expand...
Click to collapse
I always make kind of a "stresstest" for the CPU freq. I set min freq to 2.1GHz and max freq to 2.1GHz. Than see what happen. I couldn't get it stable enough for daily use, but if I can, I will enable 2.1GHz for sure
Confirmed working on the T800 with xnote rom.
DUHAsianSKILLZ said:
Confirmed working on the T800 with xnote rom.
Click to expand...
Click to collapse
Thank you for testing
I added a description of governors and I/O schedulers above
To install this with Ironrom, should I do anything with Synapse first to either return settings to default or uninstall? Or should I reflash IronRom, leaving out Synapse in Aroma and then flash the new kernel? I plan on trying Kernel Auditor.
Also, will you be adding this kernel to the aroma options soon (or are you waiting for more testing to come in)?
Hookmt said:
To install this with Ironrom, should I do anything with Synapse first to either return settings to default or uninstall? Or should I reflash IronRom, leaving out Synapse in Aroma and then flash the new kernel? I plan on trying Kernel Auditor.
Also, will you be adding this kernel to the aroma options soon (or are you waiting for more testing to come in)?
Click to expand...
Click to collapse
The kernel doesn' support synapse. You can just uninstall it if you like to and than install your kernel auditor.
Yes it will come to aroma soon, I'm waiting for samsung to release a new base for T800
And yes, testing is always welcome
Can you include some tips for let's say performance settings, battery settings, both battery and performance settings etc. I'm still testing a few things too, but can't figure out the best battery settings in faux.
DUHAsianSKILLZ said:
Can you include some tips for let's say performance settings, battery settings, both battery and performance settings etc. I'm still testing a few things too, but can't figure out the best battery settings in faux.
Click to expand...
Click to collapse
Intelli-plug, KSM enabled, Governor... I would say intellimm is good for battery but a bit slow. Interactive with intelliplug is nice
With Stweaks Support does the Kernel support BLN (blinking Soft Keys)? God, this would be awesome!!
If not, could you create a Custom Kernel with this Feature?
haselchen said:
With Stweaks Support does the Kernel support BLN (blinking Soft Keys)? God, this would be awesome!!
If not, could you create a Custom Kernel with this Feature?
Click to expand...
Click to collapse
If you would have read correctly, you will see that I'm currently adding sTweaks support. And it don't supports it now, but seems like a good idea!
Please realize this idea
Tkkg1994, this is what I call a very promising start.
Detailed walkthrough, FAQ style, and a very interesting feature rich kernel... I'm on CM12 right now, but already eager to try this in the coming days with IronRom, of course.
A Big Thanks from the other side of the world... Porto, Portugal.
Tkkg1994 said:
Intelli-plug, KSM enabled, Governor... I would say intellimm is good for battery but a bit slow. Interactive with intelliplug is nice
Click to expand...
Click to collapse
IntellIplug is nice but sometimes you cant turn the tablet on when it's enabled. I even set the screen off frequency to something above 200mhz. I had to force reboot a couple times. Did you managed to get it work? I even tryed all the profiles including balanced etc and also changing the governer. For now I'll keep it off but battery seems good so far!
i have tried many kernel...from them...dragon xia and xceed works the best...
the thing is i am a moderate use guy...just anime episodes ...songs...web surfing and lots of calls...
i main aim is to get better battery life....performance doesnt matter to me much...as long as its stable...
from my experience xceed gives better battery life then dragon xia...but since i am new to these kind of stuff....
i would really love to have knowledge from you guys...
i would love to see if someone can provide me with the adiutor settings they have used...
currently i am on resurrection remix rom 5.8.4 (mi5 standard)
i really want a better battery life from my device
Like how much battery life you are getting? SOT as well as overall usage + SOT.. try my setup on dragonxia.. I don't use it now though..
*CPU 3 &4
Min - 307
Max - 1401
Gov - interactive
*CPU 1 & 2
Min - 307
Max - 1036
Gov - interactive
*Input boost freq core 1 - 556
*Input boost freq core 2 - disabled
*Input boost freq core 3 - 729
*Input boost freq core 4 - disabled
*Touch boost - disabled
*Set GPU at max
*Turn on backlight dimmer
*I/O sched - bfq
*Swap - 100
With daily usage on this, I can say that there's no performance decrease..it actually performs well and battery life is pretty awesome..
Tip: you can personally tweak the kernel depending on your usage..rule of thumb, CPU 3 & 4 (cores with lower clock speed) is the main driver over all..means CPU interaction triggers from here most of the time..and CPU 1 & 2 (higher clock speed) will work if you're doing extensive task like gaming..
viking_kong16 said:
Like how much battery life you are getting? SOT as well as overall usage + SOT.. try my setup on dragonxia.. I don't use it now though..
*CPU 3 &4
Min - 307
Max - 1401
Gov - interactive
*CPU 1 & 2
Min - 307
Max - 1036
Gov - interactive
*Input boost freq core 1 - 556
*Input boost freq core 2 - disabled
*Input boost freq core 3 - 729
*Input boost freq core 4 - disabled
*Touch boost - disabled
*Set GPU at max
*Turn on backlight dimmer
*I/O sched - bfq
*Swap - 100
With daily usage on this, I can say that there's no performance decrease..it actually performs well and battery life is pretty awesome..
Tip: you can personally tweak the kernel depending on your usage..rule of thumb, CPU 3 & 4 (cores with lower clock speed) is the main driver over all..means CPU interaction triggers from here most of the time..and CPU 1 & 2 (higher clock speed) will work if you're doing extensive task like gaming..
Click to expand...
Click to collapse
What does swappiness do actually?
Sent from my MI 5 using Tapatalk
cheaterfruity said:
What does swappiness do actually?
Click to expand...
Click to collapse
Speed up read and write speed of the zram which is enabled by default in our RAM.. which means snappy switching of apps..
viking_kong16 said:
Speed up read and write speed of the zram which is enabled by default in our RAM.. which means snappy switching of apps..
Click to expand...
Click to collapse
Consumes more battery?
Sent from my MI 5 using Tapatalk
cheaterfruity said:
Consumes more battery?
Click to expand...
Click to collapse
Nope.. our phone is using zram as swap..not internal storage.. zram is actually a compressed RAM that sits within RAM itself.. try to google it..
Hello all.
After about a month of researching and testing with the Galaxy S5, I'm finally happy with my SmartKernel profile, with the interactive governor carefully tuned, using known resources and countless trials and errors, as well as other various tweaks, like VM and I/O scheduler, and decided to publish on it's own thread.
The main resources I've used for the Interactive governor tuning includes the well known:
Android Modders Guide;
[GUIDE] Advanced Interactive Governor Tweaks; Buttery smooth and insane battery life! for Nexus 5X; and it's twin
[GUIDE] Advanced Interactive Governor Tweaks; Buttery smooth and insane battery life! for HTC Evo 4G.
First of all, this tweaks should be a little sensible to the ROM, kernel, apps, and other tweaks your using. Like, I just found out that Havoc pie style quicktile settings use way more juice then if I turn it off and go back to Oreo default. Bellow you will see the apps I mainly crafted this profile in mind.
For reference: I have a klte with latest Oreo Havoc installed, nano OpenGapps, Magisk and the SmartPack kernel. For apps I use Facebook lite, cause the normal app is just a big hog, whatsapp and instagram social apps. Chrome. I don't use the Google App or Greenify(uninstall/delete velvet). And play lots of games like Clash Royale, Star Wars Force Arena and Arena of Valor. BetterBatteryStats.
And a lot of random apps that normally don't stay on the background.
DESCRIPTION
On the SmartPack manager profile:
. HIghly Efficient Interactive Governor Tunables (most important part);
. No Touchboost or any other boost, only the governor dictates to CPU in which clock it should to be;
. Overclock disabled, but can be enabled at you will;
. No underclock, I do undervolt my CPU but this you need to find your specific device numbers, mine won't cut;
. LazyPlug Hotplug with all 4 cores on all the time (better performance while using and battery savings while at idle);
. I/O Schedulers: ZEN (the L-Speed profile complement this part, with it's scheduler tunables);
. READ-AHEAD internal 1024kb (for 16GB or more) and external 512 kb (for my 8GB SDCard, adjust accordingly to yours SD Card size conform described here
. Adreno Idler disabled: it doesn't make any effect;
. Speaker Driver Leakage disabled and Boeffla Sound enabled with 0 gain as it does make a difference, at least with ViperFX magisk module installed;
. Screen minimum RGB set to 1 (0 won't stick), for a darker dark on our AMOLED, plus some tweaks;
. Led blinking fade enable;
. VM tweaks: dirty_ratio 30 and dirty_background_ratio 15; for minor battery improvement, with a perceptible lower termperature/cpu usage and almost imperceptible performance hit;
. VM tweaks: page-cluster 1; for better multitasking/memory management
. VM tweaks: oom_dump_tasks 0; disable depuration of dumping tasks, less cpu needed.
. LMK values: 32 48 64 128 176 208 (MBs)
L-Speed Profile
. Logging and I/O stats disabled;
. Animations speed set to 0.25x;
. System battery save trigger at 20%;
If you need to provide or read logs, enable logging and i/o stats back on l speed; i/o stats and oom_dump_tasks 1 on smartpack manager
INSTALLATION
Unzip the attached file and import with SmartPack Manager:
The attached profile should be imported, applied and marked as to run "On Boot" to make effect. It will only work with SmartPack Manager and Kernels for both Nougat and Oreo, maybe even Pie. Just try it, and report back. If you wanna fine tune it. You need to use an app or enable the "show cpu clocks" option if your rom supports it (like Havoc, RR and many more), and monitor at which frequencies the lags happens, while doing the jobs you want the CPU to be efficient at. And mainly tweak the target_load according, maybe above_high_speed delays of 1,7GHz clock and above. You need to read the guides more in-dept too see exactly how to do it, but I'll paste here the most important parts on how to tweak this settings more to your Galaxy S5, with your particularly apps and ROM:
soniCron said:
Optimize Idle Frequency
Now that you've got the base configuration, we need to tweak it so that the CPU stays at your efficient idle frequency (384Mhz in this case) without spontaneously jumping when your phone is actually idle. To do this, open a CPU monitor that displays the current core frequencies (I like CoolTool, but you can use what you like as long as it doesn't significantly impact the CPU use--you're best off using a passive monitor and checking the results after 30-60 seconds of no activity), watch the frequencies and see how often they go above your efficient idle frequency when you're not doing anything at all, and adjust the following:
timer_rate - If your idle frequency is not being exceeded much, adjust this downward in increments of 5000 until it is, then increase it by 5000. If your idle frequency is being exceeded often, adjust this upward in increments of 5000 until your CPU primarily stays at or below your desired idle frequency.
above_highspeed_delay - Only if your timer_rate has matched or exceeded 50000 and still won't stay at or below your desired idle frequency most of the time, set timer_rate to 50000 and adjust the "20000" portion of the value upwards in increments of 5000 until the idle frequency has stabilized.
The lower these two values are, the more snappy/lag free your system will be. So try to get them as low as possible without the idle frequency being exceeded too much, as this inversely affects the snappiness and efficiency of your phone when you're not doing anything. Lower = snappier but uses more CPU when you're not doing anything (such as reading a webpage); higher = less snappy but stays in a power saving state more often reducing CPU use when you're not interacting with the device. These are the most critical in determining your idle power savings, so keep that in mind if you want the most battery life!
Enhance Task Responsiveness
Now use the efficiency and nominal clock rate correlations you made for your master clock rate list in the section above and adjust your frequencies to suit your usage patterns. For example, I had web page scrolling as my 710Mhz/864Mhz rates, so I will open a web page and scroll and see how everything feels. If it feels sluggish, I will increase all the references to "710000" in both above_highspeed_delay and target_loads upwards to the next available clock rate until that task is smooth. What you are looking for is constant poor/sluggish performance when the task you're testing for is using its highest CPU use. If the task becomes sluggish/stuttery as it winds down (such as a scrolling webpage slowing to a stop), we will address that next, so do not take that behavior into consideration as you adjust these values! If the task is smooth until (or after) it slows down, then you have reached your optimal clock rate and can move on.
If you need to exceed your nominal clock rate for a particular task, first measure it again just to be sure you had it correct. If you did indeed have it correct, leave it at your nominal clock rate and adjust the value after the colon next to the task frequency you're tuning downward in increments of 5. For example, if my setting of "864000:80" is still not sufficient, I will adjust it first to "864000:75", then "864000:70", and so on until the task is smooth. However, it almost certainly won't come to this, but if you reach ":50" and the task still isn't performing how you want, set it back to ":80" and increase the clock step once more, then decrease the ":80" until it is smooth.
Do the same for each other frequency in your master clock rate list until you are satisfied. If you have chosen to use more than 2 primary clock rates, add them and use ":##" values between the two surrounding frequency values.
Fix Stuttering
Now that you have adjusted your frequencies for optimal high CPU use in each given task, you may notice some stuttering as the task winds down. (Such as a scrolling webpage slowing to a stop.) If this bothers you, you can tweak this at the expense of some (minor) battery life by adjusting min_sample_time up in increments of 5000 until you are satisfied.
If you have exceeded a value of 100000 for the min_sample_time setting and still are not satisfied, change it back to 40000 and increase (and re-optimize) your idle frequency by one step. This will impact battery life more, but less than if you were to keep increasing the value of min_sample_time.
Adjust High Load Clock Rates
You're almost done! Now you can leave everything as is and be satisfied with your amazing, buttery smooth, snappy experience, or you can optionally tweak things further to either increase the responsiveness of high load tasks (such as loading image previews in Gallery) or increase battery life somewhat.
Adjust the final delay value in above_highspeed_delay to suit your needs. The default ("150000") means that the CPU load at the highest set frequency (default "1026000") will have to be sustained for 150ms before it allows the load to go above that frequency. Increasing this value will prevent the CPU from reaching higher frequencies (which may be unnecessary) as often, saving battery life. This will come at the expense of burst-type high CPU load tasks. Reducing it will allow the CPU to reach higher frequencies more often, at the expense of battery life. However, adjusting this is probably unnecessary, as it will most likely not yield any perceptible difference in performance. It is recommended to leave this value at its default.
Click to expand...
Click to collapse
Besides CPU Voltage and Battery, all tabs on the manager are modified and tuned to achieve best performance, while having best efficiency possible. Is not a battery or a performance, but a efficiency profile.
Refer to this thread if you wanna undervolt your device with a well know secure margin for the CPU Snapdragon 801 2.5ghz MSM8974AC, which our Galaxy S5 contains:
[GUIDE] Snapdragon 805/801/800/600 Clock & Voltage (PVS bin) guide by HD2Owner I've managed to achieve much lower voltages then PSV15+ devices (refer to the sheets).
I also attached the excel spreadsheet I've made with all this thread information, both governor guide equations on target loads, undervolting guide findings, and made my own base calculations and settings. Feel free to use, modify, and discuss it with me. You will see that I based the most efficient clocks in an original thought about which ones are the most efficient, instead of plotting the differentials between voltages of each clocks, I did plotted the difference of the clock divided by voltage, which on itself should be how much voltage 1 mhz uses, on each clock rate. So, the higher the number, more speed each clock rate give us by voltage used. It's kinda complicated and idk if I explained it the right way, and even if it really makes sense under scrutiny, but I couldn't think why not myself, so, any inputs are welcome.
I own my thanks to all the following XDA fellows, without them, I could not have achieved this:
@sunilpaulmathew for the SmartPack Kernel which is the only kernel for the S5 that can turn that damned MPDecision off and SmartPack Manager;
@soniCron for both of the governos Guides;
@Saber for the Android Modders Guide which is immensely helpful.
CHANGELOGS
L-Speed Profile (download the app on PlayStore):
011118 lspeed profile
- first release
031118 lspeed profile
- Removed most tweaks, only left minor stuff, refer to the OP.
L Speed profile is not really needed, SmartPack will do 99% of the job.
SmartPack Manger Profile (download the kernel and the app here):
301018
- first release.
011118 smartpack profile:
- A few Interactive governor tweaks;
- Removed Virtual Memory and LMK tweaks, let it on default or use L-Speed to optimize, as it does a much better job then me.
031118 smartpack profile:
- Governor tunning: better high load management;
- Included back only 3 sane VM configurations, no more freezing, better cooling (less cpu needed, while performance barely took a hit)
- Sane LMK configurations, kills apps not being used faster, retain some multitasking while not let it slow down the device
081118 smartpack profile:
- target_load (no changes up to 1497600) ...1728000:89 1958400:91 2265600:95 -> ...1728000:88 1958400:90 2265600:95
- above_hispeed 20000 1190400:60000 1497600:64000 1728000:77000 1958400:84000 2265600:130000 -> 20000 1190400:60000 1728000:68000 1958400:79000 2265600:110000
- external storage read-ahead from 512 -> 2048 (because I've gone from a 8GB to a 32 GB SDCard, ADJUST YOURS ACCORDINGLY TO https://androidmodguide.blogspot.com/p/io-schedulers.html)
- cleaned unused and already default values from profile
101118 smartpack profile:
- Turned Alucard off, accidentally activated it with Lazyplug also enabled, not good!
- Managed to go 1 point higher on freq 1497 MHz, the 2 hotplugs enabled were messing with me trying to test this change before, also 1 point lower on the idle freq 268 MHz for smoother scrolling while still staying at freq 268 while idle. And some more high load optimizations now that I only got 1 hotplug enabled as it should always be.
- target_loads from 268800:29 ... 1497600:86 1574400:5 1728000:88 1958400:90 2265600:95 to -> 268800:28 ... 1497600:87 1574400:5 1728000:89 1958400:91 2265600:94
- above_hispeed 20000 1190400:60000 1728000:68000 1958400:79000 2265600:110000 -> 20000 1190400:60000 1728000:74000 1958400:82000 2265600:120000
- dirty_background_ratio 15 -> 10
221118 smartpack profile:
. Reverted new SmartPack Kernel v14r4 changes to Virtual Memory back to original default configurations, if you've have had reboots this should fix it, please report back here and/or the kernel's thread;
. More changes to Interactive governor aiming to optimize high load scenarios according to the profile philosophy:
. above_hispeed_delay 20000 1190400:60000 1728000:74000 1958400:82000 2265600:120000 -> 20000 1190400:60000 1728000:74000 1958400:80000 2265600:105000;
. Enabled fast charge configurations, set at 1200 mhA as I found it's a good charging speed without heating the phone too much on my hot city, nothing you can't change at your will.
241218 smartpack profile:
. Restored missing min_sample_time tunable since 081018 profile
. dirty_ratio 30 -> 25
. General cleanup
. Tested on Pie
@justjr
Nice work friend. Great to see that your finally open a place to share your findings. In my opinion, your profile should work on any klte device with minimum kernel support. I haven't seen much SmartPack specific stuff in your profile except some hotplug related things. So, if you make it as a shell script instead of KA/SP-Kernel Manager profile, it shall be beneficial for everyone. Anyway, as usual, I'll kang your changes to my kernel default profile
sunilpaulmathew said:
@justjr
Nice work friend. Great to see that your finally open a place to share your findings. In my opinion, your profile should work on any klte device with minimum kernel support. I haven't seen much SmartPack specific stuff in your profile except some hotplug related things. So, if you make it as a shell script instead of KA/SP-Kernel Manager profile, it shall be beneficial for everyone. Anyway, as usual, I'll kang your changes to my kernel default profile
Click to expand...
Click to collapse
I think this profile should work on original Kernel Adiutor, or any fork of it, shouldn't it?
It should work on any other kernel if the changes really stick, and uses the same paths, but MPDecision will mess with frequencies all the time. It would still follow the governor tunables anyway, but it will interfere with it and in the end will not gain too much efficiency out of it.
Actually I only state it is for SmartPack specifically because of the fact that is the only one I can disable MPDecision on our device, and because I included all the tweaks other then just governor tweaks.
Actually I'm kinda lazy right now, but I could do a shell script if any demand for it shows up.
justjr said:
I think this profile should work on original Kernel Adiutor, or any fork of it, shouldn't it?
It should work on any other kernel if the changes really stick, and uses the same paths, but MPDecision will mess with frequencies all the time. It would still follow the governor tunables anyway, but it will interfere with it and in the end will not gain too much efficiency out of it.
Actually I only state it is for SmartPack specifically because of the fact that is the only one I can disable MPDecision on our device, and because I included all the tweaks other then just governor tweaks.
Actually I'm kinda lazy right now, but I could do a shell script if any demand for it shows up.
Click to expand...
Click to collapse
Well, official KA (free version) doesn't allow to import profiles (paid feature), but all other mods does.
and yes, it is supposed to work on every klte device as long as the sysfs paths exist. Means it should work on any custom Kernel with lazyplug support (most of the other stuff are actually included in the stock kernel itself). Of course, the default settings provided by the kernel devs might conflict. e.g., as you said, MPDecision, although the line "stop mpdecison" in your profile will disable it. By the way, I'm not the only one who disabled mpdecision and relay on other hotplugs in this klte community
sunilpaulmathew said:
Well, official KA (free version) doesn't allow to import profiles (paid feature), but all other mods does.
and yes, it is supposed to work on every klte device as long as the sysfs paths exist. Means it should work on any custom Kernel with lazyplug support (most of the other stuff are actually included in the stock kernel itself). Of course, the default settings provided by the kernel devs might conflict. e.g., as you said, MPDecision, although the line "stop mpdecison" in your profile will disable it. By the way, I'm not the only one who disabled mpdecision and relay on other hotplugs in this klte community
Click to expand...
Click to collapse
Oh, really? Which one? I must had missed it. I've tested all kernels I could find. At least all the remotely up-to-date, like venom, tuned and boeffla kernels. I didn't see any option to change hotplugs on any. There were hotplug profiles, to keep cores online and stuff, but everyone of them keep changing min and max frequency at MPDecision will.
justjr said:
Oh, really? Which one? I must had missed it. I've tested all kernels I could find. At least all the remotely up-to-date, like venom, tuned and boeffla kernels. I didn't see any option to change hotplugs on any. There were hotplug profiles, to keep cores online and stuff, but everyone of them keep changing min and max frequency at MPDecision will.
Click to expand...
Click to collapse
Boeffla and Venom largely depends on MPDecision. However, as I remember correctly (on the basis of the code review, not from my experience, I never used it by myself), the Tuned kernel by @fbs disabled MPDecision upon booting to work well with its own Tuned hotplug.
sunilpaulmathew said:
Boeffla and Venom largely depends on MPDecision. However, as I remember correctly (on the basis of the code review, not from my experience, I never used it by myself), the Tuned kernel by @fbs disabled MPDecision upon booting to work well with its own Tuned hotplug.
Click to expand...
Click to collapse
I tested it too. And although he claims he uses hes own hotplug, it behave the same as boeffla and venom, it has the same profiles, and it does changes min and max freq out of my control.
justjr said:
I tested it too. And although he claims he uses hes own hotplug, it behave the same as boeffla and venom, it has the same profiles, and it does changes min and max freq out of my control.
Click to expand...
Click to collapse
no it doesn't change any freqs
it works by disabling or enabling cores, just that.
if any cpu reaches the maximum frequency, it enables one more core (as the other ones are already giving their best)
if any cpu reaches the minimum frequency too many times, it disables it (as it doesn't seem to be needed)
so in any moment you can have all 4 cores enabled or only 1. even with display on or off, it doesn't matter
mpdecision will NEVER let you use just 1 core, and it doesn't react as fast: battery hog
fbs said:
no it doesn't change any freqs
it works by disabling or enabling cores, just that.
if any cpu reaches the maximum frequency, it enables one more core (as the other ones are already giving their best)
if any cpu reaches the minimum frequency too many times, it disables it (as it doesn't seem to be needed)
so in any moment you can have all 4 cores enabled or only 1. even with display on or off, it doesn't matter
mpdecision will NEVER let you use just 1 core, and it doesn't react as fast: battery hog
Click to expand...
Click to collapse
Alright, sorry then, it seems my memories got clouded or something, as I've tested it about a month ago. I might go back any day just to test that. Thanks for giving us one more kernel option! :good:
UPDATE OP WITH
Description
Changelogs
New profile 011118, changelog:
. Few governor tweaks
. Removed Virtual Memory and LMK tweaks, let it on default or use L-Speed to optimize, it does a much better job then me
Also uploading the L-Speed profile I use so those who want to use it like I do, but you can choose any VM and LMK profile that fits your needs on the app. Just don't use the governor tuner because it will mess with my tunings, and l-speed governor tuning is a generic one for all devices, VM and LMK is OK to use generic tweaks, but not on governor.
@sunilpaulmathew I took a look at l-speed virtual memory and lmk profiles and they make incredible sense, take a look yourself, they may be what you need to put o that spectrum profiles, because above all they are device independent and do make a noticeable difference.
Is it valide for stock rom (6.0)?
lollazzo said:
Is it valide for stock rom (6.0)?
Click to expand...
Click to collapse
What kernel? It should work if the kernel have lazyplug or alucard hotplug, if is the late you just have to enable it.
Updates
SmartPack Manager Profile 031118:
. Governor tunning: better high load management;
. Included back only 3 sane VM configurations, no more freezing, better cooling (less cpu needed, while performance barely took a hit)
. Sane LMK configurations, kills apps not being used faster, retain some multitasking while not let it slow down the device
LSpeed Profile 031118:
. Removed most tweaks, only left minor stuff, refer to the OP.
L Speed profile is not really needed, SmartPack will do 99% of the job.
OP: descriptions for both profiles updated.
New profile.
I returned to Nougat, RR 5.8.5, same configs works awesomely and the device is cooler/faster then with Oreo. But still will works the same with both N/O and even Pie, not tested.
I also reinstalled Hearthstone as a high load app so I could tune the governor better for it, and up to 1490 MHz nothing is changed, and changed a bit target_loads and above_hispeed of the clocks above it so Hearthstone (and any other high load apps, or, using split screen with youtube) runs smoother/without lags and tasks like opening an app will finish faster, and also go back to a lower clock faster because of that. So, in the end it stays most of the time at lower clocks anyway, only difference is that it will jump faster when needed for less waiting time/lag.
Just to clarify, this is not suppose to waste battery, or drain it faster. As an efficiency profile the goal is to do the job you the want faster the possible, ramping up to the clocks that the jobs demands, without lags (or minimal lags) and go back to idle/lower clocks as soon as high clocks aren't needed anymore, so it don't overstay at a higher clocks then it's needed, very simple.
So, going to a high clock doesn't mean less battery life, finishing a job fast and going back to idle is the key to achieve more battery life, specially during deep sleep, when you really want your device go back to deep sleep fast, but also at any other time. Watching youtube, browsing and using low demand apps still uses the same clocks.
Also, on top of that you will spend more time USING the device instead of WAITING for it to finish a job. Battery life is very subjective, and SoT doesn't mean nothing IRL, I mean, are you spend that SoT waiting for a job to finish or to actually use the device?
081118 smartpack profile:
- target_load (no changes up to 1497600) ...1728000:89 1958400:91 2265600:95 -> ...1728000:88 1958400:90 2265600:95
- above_hispeed 20000 1190400:60000 1497600:64000 1728000:77000 1958400:84000 2265600:130000 -> 20000 1190400:60000 1728000:68000 1958400:79000 2265600:110000
- external storage read-ahead from 512 -> 2048 (because I've gone from a 8GB to a 32 GB SDCard, ADJUST YOURS ACCORDINGLY TO https://androidmodguide.blogspot.com/p/io-schedulers.html)
- cleaned unused and already default values from profile
File attached on OP.
I don't use SD card so what do I do?
razor17 said:
I don't use SD card so what do I do?
Click to expand...
Click to collapse
In that case nothing is needed, the configurations related to the absent sd card will not be applied.
Ok guys. I was wondering why my device was heating a lot more these last 2 days. Turns out both Alucard and Lazyplug were accidentally activated on 081119 profile. Turn one of them off and everything will be a lot better. Sorry for that. I will upload a new profile very soon.
edit:
101118 smartpack profile:
- Turned Alucard off, accidentally activated it with Lazyplug also enabled, not good!
- Managed to go 1 point higher on freq 1497 MHz, the 2 hotplugs enabled were messing with me trying to test this change before, also 1 point lower on the idle freq 268 MHz for smoother scrolling while still staying at freq 268 while idle. And some more high load optimizations now that I only got 1 hotplug enabled as it should always be.
- target_loads from 268800:29 ... 1497600:86 1574400:5 1728000:88 1958400:90 2265600:95 to -> 268800:28 ... 1497600:87 1574400:5 1728000:89 1958400:91 2265600:94
- above_hispeed 20000 1190400:60000 1728000:68000 1958400:79000 2265600:110000 -> 20000 1190400:60000 1728000:74000 1958400:82000 2265600:120000
- dirty_background_ratio 15 -> 10
I will give this a try. Hope it works well...
Yeah.
You know, try it and report back. I don't see any reports, so I assume is working well for people.
Any reports are welcome.
lentm said:
I will give this a try. Hope it works well...
Click to expand...
Click to collapse
Enviado de meu SM-G900M usando o Tapatalk
justjr said:
Yeah.
You know, try it and report back. I don't see any reports, so I assume is working well for people.
Any reports are welcome.
Enviado de meu SM-G900M usando o Tapatalk
Click to expand...
Click to collapse
No problems so far...greats for daily use..scrolling smoother than default..but pubg still laggy on lower res...may i know which rom are u using?