[Kernel] - 3.1.10 - JB - 1.4GHZ - Galaxy Tab 10.1 (11/19/12)

[ipnekgs]

can i flash this kernel if Im using the latest TASK650 rom?

i been reading but cant seem to find the answer thanks

No, this is only for CM10.

Sent from my SGH-T999 using Tapatalk 2

 

[jazzespresso]

Where we used to have to set some things with additional tools/apps or in init scripts some thing are configurable in settings under performance. Governor settings are interactive, performance, conservative, ondemand, and userspace. We used to mostly use ondemand per pershoots recommendation... but that is no longer the case. As already stated above Interactive is the way to go now due to development focused on it by pershoot mainly. The I/O scheduler options are noop, deadline, cfq (default), and bfq:

noop

Inserts all the incoming I/O requests to a First In First Out queue and implements request merging. Best used with storage devices that does not depend on mechanical movement to access data (yes, like our flash drives). Advantage here is that flash drives does not require reordering of multiple I/O requests unlike in normal hard drives.

Advantages:
Serves I/O requests with least number of cpu cycles. (Battery friendly?)
Best for flash drives since there is no seeking penalty.
Good throughput on db systems.
Disadvantages:
Reduction in number of cpu cycles used is proportional to drop in performance.

Deadline

Goal is to minimize I/O latency or starvation of a request. The same is achieved by round robin policy to be fair among multiple I/O requests. Five queues are aggressively used to reorder incoming requests.

Advantages:
Nearly a real time scheduler.
Excels in reducing latency of any given single I/O.
Best scheduler for database access and queries.
Bandwidth requirement of a process - what percentage of CPU it needs, is easily calculated.
Like noop, a good scheduler for solid state/flash drives.
Disadvantages:
When system is overloaded, set of processes that may miss deadline is largely unpredictable.
Reduction in number of cpu cycles used is proportional to drop in performance.

CFQ

Completely Fair Queuing scheduler maintains a scalable per-process I/O queue and attempts to distribute the available I/O bandwidth equally among all I/O requests. Each per-process queue contains synchronous requests from processes. Time slice allocated for each queue depends on the priority of the 'parent' process. V2 of CFQ has some fixes which solves process' i/o starvation and some small backward seeks in the hope of improving responsiveness.

Advantages:
Considered to deliver a balanced i/o performance.
Easiest to tune.
Excels on multiprocessor systems.
Best database system performance after deadline.
Disadvantages:
Some users report media scanning takes longest to complete using CFQ. This could be because of the property that since the bandwidth is equally distributed to all i/o operations during boot-up, media scanning is not given any special priority.
Jitter (worst-case-delay) exhibited can sometimes be high, because of the number of tasks competing for the disk.

BFQ

Instead of time slices allocation by CFQ, BFQ assigns budgets. Disk is granted to an active process until it's budget (number of sectors) expires. BFQ assigns high budgets to non-read tasks. Budget assigned to a process varies over time as a function of it's behavior.

Advantages:
Believed to be very good for usb data transfer rate.
Believed to be the best scheduler for HD video recording and video streaming. (because of less jitter as compared to CFQ and others)
Considered an accurate i/o scheduler.
Achieves about 30% more throughput than CFQ on most workloads.
Disadvantages:
Not the best scheduler for benchmarking.
Higher budget assigned to a process can affect interactivity and increased latency.

So, as you can see, setting may help depending on workload. (Like playing or streaming video)

Note on governors: Interactive (default) and best to use at this point for everyday use except benchmarking where performance would be in order:
Can be considered a faster ondemand. So more snappier, less battery. Interactive is designed for latency-sensitive, interactive workloads. Instead of sampling at every interval like ondemand, it determines how to scale up when CPU comes out of idle. The governor has the following advantages: 1) More consistent ramping, because existing governors do their CPU load sampling in a workqueue context, but interactive governor does this in a timer context, which gives more consistent CPU load sampling. 2) Higher priority for CPU frequency increase, thus giving the remaining tasks the CPU performance benefit, unlike existing governors which schedule ramp-up work to occur after your performance starved tasks have completed. Interactive It's an intelligent Ondemand because of stability optimizations. Why??
Sampling the CPU load every X ms (like Ondemand) can lead to under-powering the CPU for X ms, leading to dropped frames, stuttering UI, etc. Instead of sampling the CPU at a specified rate, the interactive governor will check whether to scale the CPU frequency up soon after coming out of idle. When the CPU comes out of idle, a timer is configured to fire within 1-2 ticks. If the CPU is very busy between exiting idle and when the timer fires, then we assume the CPU is underpowered and ramp to max frequency. On top of this pershoot recommends this and has focused development on interactive governor... so just use it unless there's a reason not to and you know what you're doing...i.e. don't set it to performance and then complain in forums about battery life. :^}

what can I say, this is probably one of the most detail answer ever, very informative indeed.

 

[notfaded1]

pershoots new mainline 3.1.10 kernel rocks!

what can I say, this is probably one of the most detail answer ever, very informative indeed.

Thanks jazzespresso... it's definitely an area of interest to me. Imagine what pershoot deals with in some of these low level kernel level coding decisions... tiny changes can have drastic affect on entire function... much to his dismay I'm sure sometimes... :^}

Semper fidelis

 

[jazzespresso]

Thanks jazzespresso... it's definitely an area of interest to me. Imagine what pershoot deals with in some of these low level kernel level coding decisions... tiny changes can have drastic affect on entire function... much to his dismay I'm sure sometimes... :^}

Semper fidelis

I always trying to make my galaxy tablet better. Considering this tablet is over 18 month old, and with 2.5 year old hardware specs (yes, quite outdated), what we are doing here is pushing the limits, and surely it is good enough for daily drive, and can take some small work, but it is still not up for heavy work if you ask me because of hardware. Yes it has Nvidia Tegra 2 dual core processor, but it was the first release of this kind processor, since then Nvidia did a lot of improvement with next releases. Developers, like pershoot, bringing JB ROM to our tablet, and it is so far the best JB ROM you can get our 18 month old galaxy tab 10.1. However, still it suffers when it comes to performance, that's probably what Samsung found out and decided to not release JB for galaxy tab 10.1. I remember when this first come out, it was the one of the fastest Android tablet on the market. Like every other technology, it eventually becomes absolute. And tablets are not like laptop or desktop to open it and throw extra RAM or better processors, so we are limited with software improvements, and see how much developer can improve it.

So far with Nightly CM10 and latest kernel from pershoot, our tablet is as good as it can get; with considering it is over 18 years old, and with 2.5 year old hardware specs. Needless to say, there are some budget tablets (range $150 to $200) out there that has better specs than our galaxy tablet, and performance is much better of course. Again, as pershoot outlined in one of his post, it all comes down to 2.5 year old hardware specs.

 

[kbluhm]

However, still it suffers when it comes to performance, that's probably what Samsung found out and decided to not release JB for galaxy tab 10.1.

lol, 18 years, it sure feels like it, right?

Despite your accurate analysis of the processor's poor performance, which has been addressed countless times prior, I doubt that had much to do with their decision. Samsung is famously known for their lack of long term upgrade support. They typically release new devices with newer software in lieu of updating existing devices.

Were it not for pershoot's hard work I would undoubtedly own a Nexus 10 today. Don't get me wrong, I will own one eventually, but his work has kept me infinitely more patient and will save me $$$ in the long run... some of which will end up in his pocket.

Sent from my GT-P7510 using Tapatalk 2

 

[ole12]

lol, 18 years, it sure feels like it, right?

Despite your accurate analysis of the processor's poor performance, which has been addressed countless times prior, I doubt that had much to do with their decision. Samsung is famously known for their lack of long term upgrade support. They typically release new devices with newer software in lieu of updating existing devices.

Were it not for pershoot's hard work I would undoubtedly own a Nexus 10 today. Don't get me wrong, I will own one eventually, but his work has kept me infinitely more patient and will save me $$$ in the long run... some of which will end up in his pocket.

Sent from my GT-P7510 using Tapatalk 2

You are so god damn right.

With task previously up-to-date roms, and pershot fast kernels, there isn't any need for an device upgrade.

The difference between an stock tab 10.1 and CM10 or any rom with pershot kernel for that matter, is so huge.

I'm still wondering how the hell sammy would release an device with an UI so laggy out of the box.

It is useless without any modifications.

Sent from my GT-I9300 using Tapatalk 2

 

[jazzespresso]

lol, 18 years, it sure feels like it, right?

Despite your accurate analysis of the processor's poor performance, which has been addressed countless times prior, I doubt that had much to do with their decision. Samsung is famously known for their lack of long term upgrade support. They typically release new devices with newer software in lieu of updating existing devices.

Were it not for pershoot's hard work I would undoubtedly own a Nexus 10 today. Don't get me wrong, I will own one eventually, but his work has kept me infinitely more patient and will save me $$$ in the long run... some of which will end up in his pocket.

Sent from my GT-P7510 using Tapatalk 2

heh heh heh...you guys know that I meant 18 moths:silly:

---------- Post added at 04:29 PM ---------- Previous post was at 04:14 PM ----------

The difference between an stock tab 10.1 and CM10 or any rom with pershot kernel for that matter, is so huge.

I agree 100%, but again it still comes down to hardware specs, processors, RAM,.etc. Keep in mind that; now 4.2 up, and many apps, games are start utilizing for 4.2, and for faster processors. For example, most of gameloft games runs much better and smooth on my Nexus 7 then my galaxy tab 10.1. I have the same games and apps on my both Nexus 7 and Galaxy Tab 10.1, and you know there's big differences when it comes to performance. I know it is quite stupid to compare two tablets, but my point here is the app/games performances and they run better on newest devices.

Nexus 10 will be my next 10' tablet for sure, it is just I need 32GB, which is quite pricey for me ($499 + shipping + tax), I will wait and hope Google will cut the price (like they did for Nexus 7), and I don't think price cut for Nexus 10 going to happen anytime soon, so I am sticking with my Galaxy Tab 10.1 for now.

 

[daniel_fm]

I am not having as much deep sleep with this kernel + pershoot rom image. The tablet spends at 250mhz almost as much as deep sleeping and I have it to disabe wifi and all when it is not being used, I have no sim card in it either so no data usage, battery stats shows after the screen the inactive device and inactive tablet are the ones more consuming. Battery last hardly two days with little use while with jellybeanrom by toldo and a1 kernel it could last over 5,6 days even more with no use and wuth this is ckeary not the case.

As much as Im loving the performance here and persoots work, there`s always the will to improve so:

Is it a rom issue, a kernel issue, any tips, clues? Thanks in advance

Sent from my GT-P7500 using xda app-developers app

 

[zerong]

I am not having as much deep sleep with this kernel + pershoot rom image. The tablet spends at 250mhz almost as much as deep sleeping and I have it to disabe wifi and all when it is not being used, I have no sim card in it either so no data usage, battery stats shows after the screen the inactive device and inactive tablet are the ones more consuming. Battery last hardly two days with little use while with jellybeanrom by toldo and a1 kernel it could last over 5,6 days even more with no use and wuth this is ckeary not the case.

As much as Im loving the performance here and persoots work, there`s always the will to improve so:

Is it a rom issue, a kernel issue, any tips, clues? Thanks in advance

Sent from my GT-P7500 using xda app-developers app

parameters, battery, wifi, advanced wifi options : actif wifi in sleep : "NEVER" for me !
So disable all google synchronisation service ! now no more drain battery for me !

 

[mcmb03]

... considering it is over 18 years old, and with 2.5 year old hardware specs...

Damn, I didn't know my tablet was almost as old as I am!

Sent from my GT-P7510 using XDA Premium HD app

 

[Leogp]

I am not having as much deep sleep with this kernel + pershoot rom image. The tablet spends at 250mhz almost as much as deep sleeping and I have it to disabe wifi and all when it is not being used, I have no sim card in it either so no data usage, battery stats shows after the screen the inactive device and inactive tablet are the ones more consuming. Battery last hardly two days with little use while with jellybeanrom by toldo and a1 kernel it could last over 5,6 days even more with no use and wuth this is ckeary not the case.

As much as Im loving the performance here and persoots work, there`s always the will to improve so:

Is it a rom issue, a kernel issue, any tips, clues? Thanks in advance

Sent from my GT-P7500 using xda app-developers app

Im having the very same problem with my tab. Battery stats showed that my tab went from 99% to 14% in 2 days, while the screen was on for less than 40min....
I have a P7500, but never used a SIM card. I'm on the latest nightly so far, 26/11

As mentioned above, love the performance and really thanks Pershoot for the work, otherwise would prob still be on Honeycomb. Just dont understand why battery drains so quick, without even use it.

 

[mcmb03]

Correct me if I'm wrong, but could this simply be because our device wasn't made for jb?

Sent from my GT-P7510 using XDA Premium HD app

 

[damngood98]

Correct me if I'm wrong, but could this simply be because our device wasn't made for jb?

Sent from my GT-P7510 using XDA Premium HD app

haven't touched my tab in 4 days since getting a nexus 10....tab is still at 54%. not seeing issues like this at all. likely you have an app causing problems. try using cpu spy or doing a full wipe.

 

[Leogp]

haven't touched my tab in 4 days since getting a nexus 10....tab is still at 54%. not seeing issues like this at all. likely you have an app causing problems. try using cpu spy or doing a full wipe.

Did a full wipe and a factory reset couple for times already, still the same. What does cpu spy do, that will help me with this issue?

Thanks
Leo

 

[muzzy996]

Did a full wipe and a factory reset couple for times already, still the same. What does cpu spy do, that will help me with this issue?

Thanks
Leo

I haven't used CPU Spy but if it's anything like Better Battery Stats the intent is to rank processes and applications in order of usage so that you can determine what's keeping the tablet awake.

One thing's for certain; if you completely format system and data and then install CM10 clean without installing any other applications the tab will deep sleep reliably. Its when you start loading up applications that the wakelocks start.

For me I keep some things disabled using Titanium Backup shortcuts on my desktop to freeze/defrost them. Things I disable include Google+, Google Talk, Google Maps. I do not have Facebook or any instant messaging apps but if those were things I used then they'd be on the list as well. Finally I keep location services turned off unless I intend to use it.

 

[jlokos]

I haven't used CPU Spy but if it's anything like Better Battery Stats the intent is to rank processes and applications in order of usage so that you can determine what's keeping the tablet awake.

One thing's for certain; if you completely format system and data and then install CM10 clean without installing any other applications the tab will deep sleep reliably. Its when you start loading up applications that the wakelocks start.

For me I keep some things disabled using Titanium Backup shortcuts on my desktop to freeze/defrost them. Things I disable include Google+, Google Talk, Google Maps. I do not have Facebook or any instant messaging apps but if those were things I used then they'd be on the list as well. Finally I keep location services turned off unless I intend to use it.

CPU Spy provides data on your cpu state, not what apps or processes are running. It's helpful to show how your kernel is operating.

 

[muzzy996]

CPU Spy provides data on your cpu state, not what apps or processes are running. It's helpful to show how your kernel is operating.

In that case perhaps Better Battery Stats is more useful in determining wakelock issues, though I don't know for sure. It gives you actual cpu times for applications/processes and wakelocks as well as wakeup calls per application from given events like since unplugged or since screen off etc. CPU state is included as a reporting option within BBS as well.

 

[kbluhm]

haven't touched my tab in 4 days since getting a nexus 10....tab is still at 54%. not seeing issues like this at all. likely you have an app causing problems. try using cpu spy or doing a full wipe.

Unplugged mine about 9.5 hours ago after crawling out of bed. Just picked it up after it's been sitting all day doing nothing. Battery's strong at 98%

Sent from my GT-P7510 using Tapatalk 2

 

[notfaded1]

better battery stats xda edition...

Did a full wipe and a factory reset couple for times already, still the same. What does cpu spy do, that will help me with this issue?

Thanks
Leo

Try this to see what's going on with you're 7500... Better battery stats-
http://forum.xda-developers.com/showthread.php?t=1179809

 

[Leogp]

Try this to see what's going on with you're 7500... Better battery stats-
http://forum.xda-developers.com/showthread.php?t=1179809

I'm actually using battery stats for a while now, and that's what shows me (attached). Any suggestions?