[SCRIPT] Slaid480's LagFree V3 script -fixed- -optimized- - Galaxy Ace S5830 General

Update 03-04-12:
This project moved to:
http://forum.xda-developers.com/showthread.php?t=1529746
---
Update 02-18-12:
Just to clarify: DONT FLASH THIS ZIP-FILE !!! It's NOT a flashable zip.
On sidenote, maybe the bootloop problem is caused by traing to flash this zip ? Maybe the .txt file in there confuses android ? I don't know but DONT flash this! Use manual installation.
---
Would any OP be willing to move this thread to the Ace Development Subforum ? Thanks a lot !
---
Update 02-09-12 (number 2):
The touchscreen sensitivity tweak was removed from the script as suggested by the posters here. It did not work anyway as the path to the files was non-existant.
Attached the updated script with short installation instructions.
---
Update 02-09-12:
I went through the LagFreeV3 script and cleaned it up / optimized it so there are no more useless repetitions of the same commands. Succesful startup of the script was tested using DDMS.
Attached the cleaned up script with short installation instructions.
---
All,
longtime lurker - first time poster here. Since i'm not allowed to post into the Ace Development subforum i have to post right here.
Looks like there are some bugs in Slaid480's latest LagFreeV3 script. On my ace running DDM during boot i can see it erroring out right here:
Code:
02-08 14:26:50.239: I/run-parts(100): /system/etc/init.d/S95LagfreeV3: 351: Syntax error: word unexpected
02-08 14:26:50.239: I/run-parts(100): run-parts: /system/etc/init.d/S95LagfreeV3 exited with code 2
so it's not even running to its EOF due to this error. Looking at he script the lines around line 351 look like this:
Code:
echo ${READ_AHEAD_KB} > /sys/block/mmcblk1/queue/read_ahead_kb
for i in \ `find /data -iname "*.db"` do \ sqlite3
$i 'VACUUM;'
done
echo "6144 87380 524288" > /proc/sys/net/ipv4/tcp_wmem
with " $i 'VACUUM;' " being on line 351. So: i'm not a BASH pro but maybe someone knows what is the syntax error in here and a way to fix it ?
Also by looking at the script there seem to be a lot of lines / functions doing the same things over and over, like setting minfree or readahead_kb - maybe in a community effort it would be possible to consolidate that script into something smaller and more streamlined ?
Just a thought

---

Yes I think it might have to do with that word.....try deleting the line .
Herpderp Defy.

---

maybe slaid can give this a clue?

---

g_BonE said:
...
Also by looking at the script there seem to be a lot of lines / functions doing the same things over and over, like setting minfree or readahead_kb - maybe in a community effort it would be possible to consolidate that script into something smaller and more streamlined ?
Click to expand...
Click to collapse
slaid did exactly that to
One Script To Roll Them ALL
Click to expand...
Click to collapse
I guess while cutting, pasting and reformatting all of known scripts to streamline them, two more newlines were accidentally added here?
Try to edit the above code snippet to
Code:
for i in \ `find /data -iname "*.db"` do \ sqlite[B][COLOR=Red]3[/COLOR][/B] [COLOR=Blue][B]$[/B][/COLOR]i 'VACUUM;'
and see, if it runs till the end. ...
^^

---

i will give this a try and come back here to post about the results. meanwhile i was also thinking about shuffling through all the parts in the script that essentially do the same thing, dont know if i'd be allowed to clean this up and repost the cleaned version here ?

---

I'd PM Slaid and ask him. The android spirit is one of openness, as is the spirit of almost all the work on modifying our favourite android ROMs. LagfreeV3 doesn't specify a software licence so it is not clear what slaid's intentions are in terms of modifications to his code. Ideally code should be licenced (and forkable via github) to make these things clearer. I think the broken bits are supposed to be formatted as in the following (the equivalent function in engengis, written more robustly):
Code:
for i in \
`busybox find /data -iname "*.db"`;
do \
/system/xbin/sqlite3 $i 'VACUUM;';
/system/xbin/sqlite3 $i 'REINDEX;';
done;
There are other sorts of errors, like the broken hashbang on line 387 (which is just a stock turboboost copy'n'pasted in) -- and yes, loads of places in which lagfreeV3 does something which is then done again by the inline turboboost. As some ROMs include turboboost and lagfree, when they upgrade to V3, things will often be repeated 3 times!!!
Looking at the code a bit is another good reason to try engengis, which combines many tweaks in a flexible system without all the duplication.

---

iandol said:
...
LagfreeV3 doesn't specify a software licence so it is not clear what slaid's intentions are in terms of modifications to his code.
...
Click to expand...
Click to collapse
There ain't any, because he also only copied it from several sources. You can find there names and "Copyrights" verbatim in there.
iandol said:
...
Looking at the code a bit is another good reason to try engengis, which combines many tweaks in a flexible system without all the duplication.
Click to expand...
Click to collapse
Good find! I'll take a closer look. Like it already, that the options are well explained and documented.
^^

---

iandol said:
Looking at the code a bit is another good reason to try engengis, which combines many tweaks in a flexible system without all the duplication.
Click to expand...
Click to collapse
Maybe a little out of topic, but what if we want to switch to engengis from lagfix? I have both lagfix and turboboost installed on my stock rom and i'd love to switch to this new tweak, but how? Is removing S95lagfix in /system/etc/init.d good enough?

---

As i mention in his thread, if u looks closedly u will find more lines actually not working...its maybe taken from other device tweaks
regards,
•ƒγяύs ργяσ™•

---

Delete S95Lagfree to get rid of lagfreeV3.
87calibration, S95zipalign S98CFSK and S98system_tweak remove turboboost.
Engengis contains the cache dropping (S98CFSK), zipaligning (S95zipalign) and deep tweaks (S98system_tweak) of turboboost, and I delete 87calibration as all it does is delete batterystats the first time it is run which I'd rather not do (and engengis can do this via the commandline control when I choose to).
Engengis also contains the same sqlite3 version as turboboost to optimise the sqlite databases, and performs both a a vacuum and a reindex, which is more thorough optimisation than the lagfree/turboboost combination seems to do...

---

Alright, after some ~6 hours of fiddling with the lagfree V3 script i managed to get it semi-cleaned up. I'll post the zip-file in my first post of this thread.
Regarding Engengis: I'll have a look at this soon and give it a shot

---

Touchscreen tweak will not working coz there are no path name like that exist...
regards,
•ƒγяύs ργяσ™•

---

will try it later on! So still thiscompatible with turboboost? Delete s78touchscreen part from turboboost, right?
Thanks...6 hours is a lot of work dude!

---

thanks g_BonE -- you still have the touchscreen tweaks in at line250
Reading through the code this is much better than lagfreeV3 original, and should combine better with turboboost...

---

Thanks for the nudge to the non-working touchscreen tweak. I removed the tweak alltogether and reuploaded the fixed script. See first post
I'll have a go at the turboboost v8.5 script: maybe i can integrate all / most of it into the cleaned LFv3 script. Maybe also the "journaling disabler" script can be integrated as it is just a matter of mount options.

---

g_BonE said:
Thanks for the nudge to the non-working touchscreen tweak. I removed the tweak alltogether and reuploaded the fixed script. See first post
I'll have a go at the turboboost v8.5 script: maybe i can integrate all / most of it into the cleaned LFv3 script. Maybe also the "journaling disabler" script can be integrated as it is just a matter of mount options.
Click to expand...
Click to collapse
I am really not an expert, not even close, but doing a script with so many lines will probably not have the desired result?
As member pyronia suggested it's better to do small scripts for a specific task than having one for all. Maybe the one for all script would suit devices with more power to process the script?
Anyway, already thanked, glad someone cleaned that script, watching you closely for the updates!

---

I was used this fix script. But when I check in terminal manager 'a2sd check' the end value min RAM free change to 14xxx, when used lagfixv3 the value is 23xxx.
So actually it's works or not? Thanks
Sent from my GT-S5830 using Tapatalk

---

Hello g_BonE!
I'm having an issue with your last version (I got stucked in Android Screen). Is there anything I did wrong? I just entered in /system/init.d/deleted s95lagfix and simply copied you s99 with all permissions (with root explorer) and then I got stucked.
What can I do?
Thank you very much for your script!

---

I knew it there was something fishy with these scripts.LOL Thank you g_BonE for your effort! I'll be hitting the thanks button buddy. Oh and another thing, should I delete V8.5 TurboBoost too?
---------- Post added at 08:38 AM ---------- Previous post was at 08:26 AM ----------
For the people who don't read the instructions:
You should delete S95Lagfree first, located at etc/init.d. Then paste S99_LFv3gB on the same folder (etc/init.d), then set permissions to rwe-rwe-rwe.
I used to think that it should be at rw-r-r sir g_BonE.

---

itasoulas said:
I am really not an expert, not even close, but doing a script with so many lines will probably not have the desired result?
As member pyronia suggested it's better to do small scripts for a specific task than having one for all. Maybe the one for all script would suit devices with more power to process the script?
Anyway, already thanked, glad someone cleaned that script, watching you closely for the updates!
Click to expand...
Click to collapse
Yeah bro, i test so many script but not all working well coz of too much lines..the devices itself maybe could not read the script properly when too much lines in it, thats why i make a single line script that actually u can see the different...
regards,
•ƒγяύs ργяσ™•