mine XDA I having total 32MB memory allows for moving memory tab in
start/settings/system/memory
between STORAGE and PROGRAM options so that it divides in-bulit memory but .... I am curious about one thing, I have just noticed that today - if I assign too small amount of memory for STORAGE (i.e. it automatically put the division tab sback in a way that memory is split in halfs (50:50) considering current utilization od PROGRAM part
the question is .... is it a limitations imposed from OS manufactirer ??
regards, monika
seems like similar issues have been covered here ... sorry for double posting
http://forum.xda-developers.com/viewtopic.php?t=5660
regards, monika
pflatlyne said:
Looking back at it,im not so sure anymore. I looked up some information on how the virtual memory in windows mobile works. It seems that whether a program is kept in memory depends on what kind of memory it is stored on. Programs that can be reloaded from flash will be unloaded if needed. Data structures that programs keep,will not be becuase there is no way to restore them.
It might well be possible to swap some of these things out to the transflash card. In fact,vista does something similar with readyboost. Ive actually used my phones card as the readyboost device without problems. What might be a problem is the speed of the card reader in the phone. Some people say the reader is very slow. I dont know if this is true or not,but as a general rule,the card itself is faster than a hard disk,so its not at all unreasonable. One problem of course would be wearing out the flash. Readyboost uses a special algorithm to spread the writes across the drive and reduce the amount of writes. If you just put the swap on flash,it would wear out rather fast.
Click to expand...
Click to collapse
Ive decided that this might not be so crazy as it sounds. This was originally discussed here. http://forum.xda-developers.com/showthread.php?t=350002&page=15
Consider this: http://blogs.msdn.com/ce_base/archive/2006/10/30/What-is-Virtual-Memory.aspx
Now direct your attention to this part from the above link
<QUOTE>
Windows CE will “demand commit” pages, meaning that it will usually delay committing them until the last possible moment. There are also some cases in which Windows CE will take committed pages out of memory again, returning them to “reserved” status:
* Since DLL and EXE code can easily be re-loaded from the file system, it is often decommitted.
* Memory-mapped file pages also have well defined files to read from, so those are decommitted as well.
* Thread stack can shrink; if the system is very low on memory we’ll scavenge the top pages from a stack if the thread is not currently using them.
* Heaps can shrink; if the system is very low on memory we’ll check whether there are heap pages without data in them that can be decommitted.
However that is where Windows CE stops. Other operating systems have a “page file” in which they will write other pages that don’t have corresponding files, notably:
* Stack
* Heap
* Allocations from VirtualAlloc()
* Memory-mapped files that don’t actually have files underneath them (CreateFileMapping with INVALID_HANDLE_VALUE)
* The writable data of executables (global variables)
Those operating systems have algorithms to decide that these pages have not been used in a while, and will write them to the page file and decommit the RAM. Windows CE does not have a page file. We’ll demand-commit to delay committing these pages as long as possible, but once they are committed, the operating system won’t page them out again.
So, as you see, virtual memory in its most minimal definition is just having a mapping between virtual addresses and physical addresses. To lay out allocations in the address space in an efficient manner and avoid wasting physical memory on unallocated address space. In more practical terms, we also use the virtual address space to implement paging, avoid wasting physical memory on allocated addresses that are not actively being used.
</QUOTE>
Here is another interesting article.
http://blogs.msdn.com/ce_base/archive/2008/01/19/Paging-and-the-Windows-CE-Paging-Pool.aspx
Consider:
<Quote>
I’d like to explain a little more about memory management in Windows CE. I already explained a bit about paging in Windows CE when I discussed virtual memory. In short, the OS will delay committing memory as long as possible by only allocating pages on first access (known as demand paging). And when memory is running low, the OS will page data out of memory if it corresponds to a file – a DLL or EXE, or a file-backed memory-mapped file – because the OS can always page the data from the file back into memory later. (Win32 allows you to create “memory-mapped files” which do or do not correspond to files on disk – I call these file-backed and RAM-backed memory-mapped files, respectively.) Windows CE does not use a page file, which means that non file-backed data such as heap and thread stacks is never paged out to disk. So for the discussion of paging in this blog post I’m really talking only about memory that is used by executables and by file-backed memory-mapped files.
</QUOTE>
In short,people are comparing transflash speed to ram speed,realizing there is a vast difference and deciding it wont work. The fact is,the OS is already using virtual memory. It is just not using the transflash and it is not paging certain structures. It still may not be possible,but I think its going to be more of a matter of getting the OS to handle it rather than a speed thing. The quick easy and vastly oversimplified requirements would be to some software on the phone that would create a paging file,manage that file so as not to wear out the flash to fast,and do all the back end work necessary to swap them in and out. What is not clear is how much we can use machinery the OS already has.
Im now going to go into the realm of unsupported speculation. Could we perhaps have a program that manages the paging file(we will call this the page file manager). Its job would be to create a virtual file system inside the page file. Then we could set the pages that are ordinarily not pageable (through some magic algorithm that knows what we REALLY want to not page,and what we can get away with paging to the file) as pageable. I imagine we set them as "file backed" and allocate a virtual file inside the paging file.(our page file manager would manage this for us). The machinery of the OS might then handle all the paging for us. Of course the fact that these structures are writeable would be an issue. Writes to a page would cause an exception and the modified page would be recorded in the virtual file system. Like I said however,this is very early speculation,not even to the level of a plan. In short the idea is to take data structures,make a file for them somewhere,and then make the OS treat them like any other file backed memory page. It sounds almost plausible,doesn't it.
Still more speculatory rambling. I imagine the paging file manager would manage the list of free pages in the page file that are allocated to files and the free ones that can be allocated. It would also manage the order in which they are used,to spread the usage around as much as possible to extend the life of the flash card.
It would go something like this. A data structure is to be paged. A virtual file is created in the paging file,allocated to the size of the data structure to be paged. The structure would be copied to the file and the pages uncommitted. A read to the virtual memory pages would generate a exception and ultimatley cause structure to be reloaded. A write would generate an exception and cause the file to be updated. Page_file_manager would handle this. It would allocate the next available block to the page. It would then update the file allocation table to reflect the replacement of that block in the file and release the original. Of course this all really depends on the internal structure of windows mobile.
I am having trouble in understanding if and how best to use Compcache and/or Swap.
I see that the default config of UD not is CompCache AND Swap on. But I see no benefit. The memory never seems to be used (also it feels slower).
Even if I use CompCache 64Mb with swapiness at 80, if I check stats there is never more uses than about 32mb. But Android does go ahead and close programs in the mean time. Stuff does not keep running if I go to another program, it has to be restarted if I switch back.
So why does the system not use all 'RAM' available to it? Should swapiness be increased even more?
I wonder if your minfree settings may be keeping your compcache swap from being fully utilized. What does the output of 'cat /proc/swaps' in a terminal look like? What are your minfree settings?
I don't know if I understand the compcache thing fully, and I'm no Android expert....
From what I read from the net, both compcache and swap memory only swaps kernal memory, so it does not mean every bit of compcache will be used, as kernal only uses a limit amount of memory. If a certain program uses a lot of memory outside kernal, then other programs will still be closed to make room for it. In short, compcache is just an alternative solution, it's never a real solution for lack of ram.
For me, I actually experimented with setting 192MB compcache and swapiness of 100. It seems to work well for me, and the most I've seen it used is only about 30MB.
Thanks guys, I'll look at both things - minfree and setting swapiness to 100 - and report back!
As to minfree, I saw other threads that seemed to point in this direction as well, but maybe this is something different.
The other threads were talking about that the system/kernel would not get notified of the "additional free RAM" offered by the CompCache, so that it would still kill programs is there would still be CompCache (but not much real free RAM) available...?
[EDIT] this may help us to some answers, or part of, reading now myself...
Hello, last two days I thought about I/O performance of great Asus Transformer Prime.
It is actually PDA so it means that device is not in the real power off state so much.
It can be used for ANY kernel and ANY rom!
Works with ANY ASUS TRANSFORMER MODEL!!!!!!!!!!!!!!!!!!!!!
Sooo let's tweak it for better performance!
We will reduce write times to SSD drive, no more hangups, ARNs etc.
I am using it on my own prime for one day and I have no delays at all.
For this test I need huge amount of volunteers to tune numbers in the script.
This change is only for actual session
If you want to save it after reboot use SManager for script autorun.
Here is script, run like. su
So download script on sdcard to root of sdcard, unzip and run:
Code:
su
cd /sdcard/
sh tune.sh
It is safe and it will not harm your system or break OTA update
The worst scenario is softresetting your device after
hang up (but it should be ok).
PLEASE I NEED RESPONSES FROM YOU!!!!
+ Faster system
+ less hangups (none on my Prime)
+ You don't have to flash new kernel
+ It is for actual session only, after reboot you have to run script again
+ Better battery life!
+ Longer life for internal SSD drive!
- If you reset your prime you can loose some of your datas
- If your system hangs you can loose some of your datas
- It is for actual session only, after reboot you have to run script again
EDIT: :::::::::::::::::: Ok version 0.2 is out. ::::::::::::::::::
With previous values I had problem with gmail.
With little tweaking I have no issues now and there is ALMOST no drop of permormance compared to version 0.1.
Changelog:
Code:
changed value for dirty_ratio from 90 to 70
changed value for dirty_background_ratio from 70 to 50
If you haven't gmail issues you don't have to install version 0.2
Nvm it seems that gmail issue had nothing to do with these changes
EDIT: 25.6.2012
Tune script is included here:
http://forum.xda-developers.com/showpost.php?p=27862975&postcount=149
:::::::::::::::::: EDIT 2: 26.6.2012 New version out v0.3 ::::::::::::::::::
+ tweak for battery (audio is suspended earlier when it is not used
+ some tweaks to cfs scheduler should be more responsive
Let me know your experiences with this version!!
::::::::::::::::::8.7. 2012 NEW VERSION OUT! v0.4 ::::::::::::::::::
Some of ideas taken from here ! http://forum.xda-developers.com/showthread.php?t=1205259
Optimalized for Prime
+ Added three types of scripts
+ 1st script = tune.v0.4.sh.zip = safe optimalization = only safe tweaks which will speed up IO operations a little bit
+ 2nd script = tune.v0.4.drive.sh.zip = drive optimalization = same like 1st script + optimalizations for internal ssd drive
+ 3rd script = tune.v0.4.max.sh.zip = heavy performance mode = 1st + 2nd script + "old" checked by users tweaks with background ratio, swapiness etc. + io mount tweaks + kernel tweaks + internet tweaks = max. perform. (for best turn sio TOO!!)
+ Longer life for internal SSD drive! Because noatime,noadatetime = less writes
- Somebody reported battery drain, so these relative functions were removed
Please for correct testing remove ATP application and run this script with SManager or manually (if you won't uninstall ATP it will change values...) If you don't want to remove ATP you have to wait for new developer of ATP when he will add these version of scripts to ATP !!!!!!!!!
If you have some problems after this script PLEASE try to remove ATP first, it seems that on some devices it has problems with running this script.
ok i m running it right now.
what are the default values?
Sent from my Transformer Prime TF201 using Tapatalk 2
Sorry I dont remember. Reboot your prime and you will see defaults.
batoo said:
Hello, last two days I thought about I/O performance of great Asus Transformer Prime.
It is actually PDA so it means that device is not in the real power off state so much.
It can be used for ANY kernel and ANY rom!
Sooo let's tweak it for better performance!
We will reduce write times to SSD drive, no more hangups, ARNs etc.
I am using it on my own prime for one day and I have no delays at all.
For this test I need huge amount of volunteers to tune numbers in the script.
This change is only for actual session
If you want to save it after reboot use SManager for script autorun.
Here is script, run like. su
So download script on sdcard to root of sdcard, unzip and run:
Code:
su
cd /sdcard/
sh ratio.sh
It is safe and it will not harm your system or break OTA update
The worst scenario is softresetting your device after
hang up (but it should be ok).
PLEASE I NEED RESPONSES FROM YOU!!!!
+ Faster system
+ less hangups (none on my Prime)
+ You don't have to flash new kernel
+ It is for actual session only, after reboot you have to run script again
+ Better battery life!
- If you reset your prime you can loose some of your datas
- If your system hangs you can loose some of your datas
- It is for actual session only, after reboot you have to run script again
Click to expand...
Click to collapse
Tempie007 said:
ok i m running it right now.
what are the default values?
Sent from my Transformer Prime TF201 using Tapatalk 2
Click to expand...
Click to collapse
Hi, it will be interesting to see how this works. swappiness=0 will cause less swapping to disk (it will only happen when it needs to). Default for swappiness=60 in our Android and mainline Linux kernels. These settings here are aggressive in trying to reduce swap disk i/o, so it will be great to see if it succeeds and whether or not OOM errors occur when RAM is running short when using RAM hungry applications or doing extreme multi-tasking.
I have also been testing something at the vm level in the kernel that I found in the HTC OneX source. It is something the HTC engineers added that would actually set the default to swappiness=100, the complete other direction from what you are doing here. This theory is to clear out the stale RAM and open up the RAM for something more useful. While this may be counter-intuitive to I/O performance since swapping equates to more disk i/o, the code change also implements some changes to the block, fs, and mm portions of the kernel. Essentially, from what I can tell, it flags and handles disk operations (aka "bios") differently if they are involved in swap operations (going to or coming from swap). A "bio" is essentially a manifest of an ongoing I/O block device operation, what sectors and memory locations it contains etc. Once the RAM is freed up, I have also been increasing the Dalvik vm settings in the build.prop to allow for more RAM consumption. So far, this combination (at least for me so far) has really given me good UI performance and I haven't had to kill the browser often like I did before. I have yet to play with vm.dirty_ratio and vm.dirty_background_ratio, but I am wanting to do this as well to see what combination can work the best for us.
Anyhow, thought I would share. I think you are looking at some important things here with the tuning of virtual memory to see how they can affect disk performance and more importantly application fluidity.
Cheers
Hi
I'm using it for about two hours and I had one reboot when I opened File explorer (before this never happend to me on .28) Im using also IO Sio from Your other topic (it seems to be alright...everything seems to work smoother). On tune.sh I have better write to sd speed in antutu. Now I have 8 mb...before 2 mb. Thx for that.
_motley said:
Hi, it will be interesting to see how this works. swappiness=0 will cause less swapping to disk (it will only happen when it needs to). Default for swappiness=60 in our Android and mainline Linux kernels. These settings here are aggressive in trying to reduce swap disk i/o, so it will be great to see if it succeeds and whether or not OOM errors occur when RAM is running short when using RAM hungry applications or doing extreme multi-tasking.
I have also been testing something at the vm level in the kernel that I found in the HTC OneX source. It is something the HTC engineers added that would actually set the default to swappiness=100, the complete other direction from what you are doing here. This theory is to clear out the stale RAM and open up the RAM for something more useful. While this may be counter-intuitive to I/O performance since swapping equates to more disk i/o, the code change also implements some changes to the block, fs, and mm portions of the kernel. Essentially, from what I can tell, it flags and handles disk operations (aka "bios") differently if they are involved in swap operations (going to or coming from swap). A "bio" is essentially a manifest of an ongoing I/O block device operation, what sectors and memory locations it contains etc. Once the RAM is freed up, I have also been increasing the Dalvik vm settings in the build.prop to allow for more RAM consumption. So far, this combination (at least for me so far) has really given me good UI performance and I haven't had to kill the browser often like I did before. I have yet to play with vm.dirty_ratio and vm.dirty_background_ratio, but I am wanting to do this as well to see what combination can work the best for us.
Anyhow, thought I would share. I think you are looking at some important things here with the tuning of virtual memory to see how they can affect disk performance and more importantly application fluidity.
Cheers
Click to expand...
Click to collapse
Ice increased my dalvik vm also through build.prop. Like the heapsize, growth limit, startsize etc...how much can you increase those vm to where its still safe? On stock kernel.
What values are you running at? IM also using that dalvik jit mod.
demandarin said:
Ice increased my dalvik vm also through build.prop. Like the heapsize, growth limit, startsize etc...how much can you increase those vm to where its still safe? On stock kernel.
What values are you running at? IM also using that dalvik jit mod.
Click to expand...
Click to collapse
Not sure how much room there is since I have only tweaked on the prime recently. I am running 8m, 64m, and 256m now. One of my other tabs used higher dalvik settings like this on the HC stock ROM (GT 8.9 or Iconia A500 can't remember). There are so many parameters to take into account for any given test etc., as they say "you never step into the same river twice" LOL.
I have posted my latest alpha kernel with the aforementioned changes. It will be interesting to see if folks like it or not. It is working great for me thus far.
Guys pls no offtopic and chat.
Ill try this new tweak once I get off work.
my first problem. cmclient has stopped.
first time I have seen this in weeks and it comes after 45 min of this tweak.
However it may be related to the fact that I have not used the default browser in weeks.
whycali said:
my first problem. cmclient has stopped.
first time I have seen this in weeks and it comes after 45 min of this tweak.
However it may be related to the fact that I have not used the default browser in weeks.
Click to expand...
Click to collapse
a reboot should clear it. did you enter in commands exactly like listed in op? it tells me file not found. when i unzipped the file, it gives me a tune.sh txt file or whatever. in the commands, is it supposed to be "tune.sh" instead of "ratio.sh"? i see i can just click the file and execute the file like that, bypassing the need to enter in commands. same like with overclock files. i could just execute them manually instead of using an overclock app.
have you noticed any other issues? what benefits have you noticed so far? you have me hesistant to execute the file now, since those commands give me error.
demandarin said:
a reboot should clear it. did you enter in commands exactly like listed in op? it tells me file not found. when i unzipped the file, it gives me a tune.sh txt file or whatever. in the commands, is it supposed to be "tune.sh" instead of "ratio.sh"? i see i can just click the file and execute the file like that, bypassing the need to enter in commands. same like with overclock files. i could just execute them manually instead of using an overclock app.
have you noticed any other issues? what benefits have you noticed so far? you have me hesistant to execute the file now, since those commands give me error.
Click to expand...
Click to collapse
PM'd you. it was just an alert. no need for reboot. hard to see or report anything yet.
whycali said:
PM'd you. it was just an alert. no need for reboot. hard to see or report anything yet.
Click to expand...
Click to collapse
Did you guys get this to work, I keep getting the file not found message
whycali said:
PM'd you. it was just an alert. no need for reboot. hard to see or report anything yet.
Click to expand...
Click to collapse
MRCANNADY said:
Did you guys get this to work, I keep getting the file not found message
Click to expand...
Click to collapse
If you want to try the OP's settings, you can easily change these and other virtual memory related settings graphically using the free app System Tuner under "SysCtl". This may be easier for those that don't like messing around with scripts. You can also optionally have the settings to be activated on boot if you test them thoroughly and want them to persist.
I ran
sh tune.sh
as the third command since that was the file name. nothing to report beyond the first one I posted here. I have not sodded yet and have folded my prime closed for a while since implementation.
im getting great results so far. no lag during app store apps updating and installing. plus the apps download and install so fast now. I think this mod along with SIO scheduler, build.prop tweaks, and overclock makes the prime scream speed. no negative drawbacks running this so far. i havent tested playing games yet.
whycali is correct. the filename in the command needs to be changed to what he posted, tune.sh
I am so sorry I changed typo in firat post to tune.sh, my test version had diff. Name.
i am going to try this based on the awesome feedback i have read. i will be testing alongside Chainfire driver and CFQ using ATP Tweak. i will report back. i do have some dalvik tweaks that may yield extreme lag free by forcing zygote to recognize 4 cores along with JIT.
Sent from my Transformer Prime TF201 using xda premium
I don't think that tweaking over heap size and experimenting with JIT will help so much.
This is the cleariest way how to speed up system.
swapiness is useful mainly with swap location but prime doesn't use swap at all.
It's true that these values are agressively focused on smooth UI but it's the point.
I am working on another tweaks.
These things can be of course written to startup scripts to the system, but I don't want to unlock my Prime and when I don't have CWM I won't risk some mistyping resulting to bootloop.
This is the most effective way how to safely reach good performance.
Don't blame Asus that they didn't setup these values, they need to keep some "safe" standards.
Anyway I will try to explore more effective ways.
Guys let me knows your experiences!!
I already tested some heavy applications like Nova 3 and it's working without problems...
Results are in
With out Tune.sh script and SIO at balanced setting:
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Same but at Performance Setting:
Now with Tune.sh and SIO enabled at balanced setting:
Same with Performance Setting:
You be the judge. Running .28 ASUS Stock and LOCKED bootloader
Sent from my Transformer Prime TF201 Biotchh
Hello,
I am working on an Acer Iconia B1-A71 at the moment.
I managed to unlock the BootLoader, thanks to @FireDiamond ) & I also managed to root it, thanks to @entonjackson ).:highfive:
Now I'ld like to increase the memory dedicated to applications.
I know that I'm able to swap internal & external storage, but what I'm willing to do is increase what in the settings is called "applications memory".
Talking with a friend we thought that repartitioning internal & application memory, & then swapping modified internal with external memory, would grant lots of application memory & lots of internal memory.
Is this possible ? what should I use or do to achieve this ??
Thanks to all the comunity!!!
P.S.: More specifically:
App 1GB, Int 5GB, Ext 30GB =Partition=> App 5GB, Int 1GB, Ext 30GB =Swap=> App 5GB, Int 30GB, Ext 1GB.
My useless opinion:
So, what will you do with that increased memory? I thought only about games with large cache like Dead Trigger (about 0.7 Gb) or apps that require a heavy amount of memory like cSploit+MetaSploit Framework+Ruby (about 3 Gb).
Did you know we need another resources to get apps working, like RAM and CPU? That's why I bolded RAM - we need to increase it first!
I did a little research earlier:
(tablet status) — (free RAM in standby mode, total is 480 Mb)
1. Full stock RV04RC04 — ~90 Mb
2. Stock, DVM heap 30 Mb, Smart Launcher v2 — 130 Mb
3. Same as 2, but with removed bloatware and stock launcher — 150 Mb.
Now let's check minimal and optimal requirements for some apps:
Minecraft PE — 64/128 Mb (passes)
NfS: MW — 100/250 Mb (passes on minimal graphics)
GTA: SA — 300/1024 Mb (passes veeery laggy on minimal graphics then crashes)
DJ Studio 5 — 500/1024 Mb (btw it passes, so strange!)
Firefox (even Beta) — ?/? (crashes on second opened tab)
Also, when playing some heavy games you can see that ugly LOW MEMORY sign in edge of screen.
What we need to make:
1. Custom kernel with another compression and enabled 2nd CPU kernel (it's disabled forever with stock one).
2. ROM with a high RAM economy, so we can use free RAM in heavy apps.
Unfortunately, I don't think we are able to do it, because there are no many developers who ready to work with Iconias; anyway we have devs, but in the most cases we can't co-work because of "language barrier". I know some guys from 4PDA, they can look on the B1-A71 if I ask them, but we need more than only 2 russian devs to make all that stuff...
Now about your problem with memory swapping
Use google: android create vold fstab
@FireDiamond
I know that RAM is needed & I noticed it had quite a low amount.
The owner ( a friend of mine ) wants to have a bit more app memory because by installing the few apps she uses, it fills up pretty fast.
She uses it to read some pdf, browsing & play low level games as Fruit Ninja or any Saga game.
I'm searching now on the keywords, thanks for them.
Probably it would be enough to set 3GB for App & 3GB for Int storage, though if possible, it could be more usefull if I also did the memory swap, in which case the "Ext" storage ( = Int memory left out of App storage ) would be useless since there would be an "Int" 30GB.
That is why I thought of giving everything to App memory.
EDIT: I searched a bit about thevold.fstab & I recognized that it is usefull for swapping, which I will do following an other topic. What about resizing interanl memory partitions ? Does anyone know what I could use ?