[ROM][MOD][4.4.2][EXPERIMENTAL] KitKat OS2SD

skinbark

Well-known member
Nov 22, 2013
289
0
0
Visit site
OK. It's probably something minor because the headset is detected and identified - it just never completes pairing and gives up after a while.
Correction: one of the p500 users reported a similar pairing/functioning issue so it might actually be something that hasn't yet been worked out upstream.
 

yaconsult

Well-known member
Jun 14, 2011
129
0
0
Visit site
Possible usb charging issue?

I left the phone with os2sd kitkat on my optimus v plugged into the computer when I went out for the day yesterday. It wasn't mounted or anything - just plugged into a usb port to stay charged.

This morning I noticed that the phone was looping on the LG logo. So I unplugged it, removed and reinserted the battery and tried to start it. The phone and battery were completely dead. Plugged in usb and got the LG logo boot. Also noticed that I couldn't get into recovery.

By swapping batteries with my other phone I determined that the problem was that the battery was completely dead. And it won't charge in that phone. Put the good battery in the os2sd kitkat phone and it boots right up. I put the dead battery in the non-kitkat phone to charge and it says to wait because the battery is so low it can't be charged. After a while, it starts charging the battery.

I'm a little confused about why the battery wouldn't charge on the phone even when plugged into a proper charger instead of the computer's usb port. When plugged into my good phone, it displayed the expected battery charge graphic when the phone was turned off and plugged into the charger. After charging for 10 minutes in the other phone, I was able to plug the battery back into the os2sd phone and boot it. I have it plugged into the charger now and will report on whether it charges as expected. It's showing the battery as "charging 1%".

Postscript: it seems to be charging normally now - it's up to 5% from the high current charger I have it plugged in to. I will switch it back to usb computer connection after it's fully charged and see if it stays charged or discharges again.
 
Feb 19, 2011
1,971
284
0
Visit site
Re: Possible usb charging issue?

I left the phone with os2sd kitkat on my optimus v plugged into the computer when I went out for the day yesterday. It wasn't mounted or anything - just plugged into a usb port to stay charged.

This morning I noticed that the phone was looping on the LG logo. So I unplugged it, removed and reinserted the battery and tried to start it. The phone and battery were completely dead. Plugged in usb and got the LG logo boot. Also noticed that I couldn't get into recovery.

By swapping batteries with my other phone I determined that the problem was that the battery was completely dead. And it won't charge in that phone. Put the good battery in the os2sd kitkat phone and it boots right up. I put the dead battery in the non-kitkat phone to charge and it says to wait because the battery is so low it can't be charged. After a while, it starts charging the battery.

I'm a little confused about why the battery wouldn't charge on the phone even when plugged into a proper charger instead of the computer's usb port. When plugged into my good phone, it displayed the expected battery charge graphic when the phone was turned off and plugged into the charger. After charging for 10 minutes in the other phone, I was able to plug the battery back into the os2sd phone and boot it. I have it plugged into the charger now and will report on whether it charges as expected. It's showing the battery as "charging 1%".
here and there in the jbc thread a few people complained about that happening to their phones with the 3.0 kernel. charging wouldn't work even with a different ROM installed. I can't account for it... and nothing like that has happened to any of the 4 OV's I've flashed to, zv5, zv5 and zv9.
 

brotherswing

Well-known member
Feb 24, 2011
555
167
0
Visit site
I encountered that issue. The variables that I isolated were the port (some USB ports on my PC do better than others) and the cable itself. I have a cable that will charge my Milestone X but not my OV regardless if it's on a wall charger or a USB port. The additional drain noted on many Roms that require a 3.0 kernel coupled with the inconsistencies of charging from a USB port could easily account for the symptoms you've described.

VM670 on 4.4.2? Yes, indeed.
 
Feb 19, 2011
1,971
284
0
Visit site
I encountered that issue. The variables that I isolated were the port (some USB ports on my PC do better than others) and the cable itself. I have a cable that will charge my Milestone X but not my OV regardless if it's on a wall charger or a USB port. The additional drain noted on many Roms that require a 3.0 kernel coupled with the inconsistencies of charging from a USB port could easily account for the symptoms you've described.

VM670 on 4.4.2? Yes, indeed.
I have one OV that refuses to connect to usb or charger with anything except one single cable. Every other cable that works with all kinds of different devices and 3 other OV phones, it's blind to.
 

brotherswing

Well-known member
Feb 24, 2011
555
167
0
Visit site
And that, my friend, is why I don't bother reporting most charging issues. USB charging is known to be flaky. I do think something needs to happen for the battery life before I could recommend OS2SD as a daily driver for the typical user through. My battery drops about eighty percent in six hours and that's probably more drain than most people would care to deal with. Of course, my SD card might be contributing to that. How fast are the batteries draining on other phones testing this build?

VM670 on 4.4.2? Yes, indeed.
 
Feb 19, 2011
1,971
284
0
Visit site
And that, my friend, is why I don't bother reporting most charging issues. USB charging is known to be flaky. I do think something needs to happen for the battery life before I could recommend OS2SD as a daily driver for the typical user through. My battery drops about eighty percent in six hours and that's probably more drain than most people would care to deal with. Of course, my SD card might be contributing to that. How fast are the batteries draining on other phones testing this build?

VM670 on 4.4.2? Yes, indeed.
mine seems comparable to pac (not the latest but the one before.) lasts over a day without taxing wifi or 3g but only a matter of hours if pulling data.
 

willmon22

Well-known member
Oct 2, 2011
109
3
0
Visit site
Seems like something that might be a little complicating to do. I have only been flashing roms and have never used os2sd. Is it possible someone can make a video showing step by step instructions on how to do this so I will feel more comfortable doing this to my phone? It may be a back up phone for me but I would just like to learn how to use os2sd properly so I can use it in the future if necessary.
 

brotherswing

Well-known member
Feb 24, 2011
555
167
0
Visit site
That is a cool idea, and I need to redo my installation (got the partition sizes wrong.) I'll do it up in a few days. Need to fix my laptop first...

VM670 on 4.4.2? Yes, indeed.
 
Feb 19, 2011
1,971
284
0
Visit site
Since this is the experimental thread, I'll mention aufs again.
It'd be nice to have a minimal install in NAND that would run without the card, and overlay partitions onto system, data, and cache (all three because of the dalvik and other bits that wouldn't enjoy only existing in one of two places of the card was removed,) so it'd be possible to pull the card while running but still have the ridiculous space option from the ext partitions.
settings>storage would probably need modified to handle the two partitions beyond what it expects for a clean unmount/remount.
 
Feb 19, 2011
1,971
284
0
Visit site
That would be handy! Just what would be involved in implementing this?

VM670 on 4.4.2? Yes, indeed.
insmod aufs module in init.*.rc
script to mount ext's over nand's

mostly. I couldn't get it to work on first try, but the OS2SD wasn't working yet either at that point.

While I'm posting,
of interest in kernel, this is in 3.10
https://github.com/bigsupersquid/codeaurora_kernel_msm-3.10/tree/msm-3.10/android/configs
The files in this directory are meant to be used as a base for an Android
kernel config. All devices should have the options in android-base.cfg enabled.
While not mandatory, the options in android-recommended.cfg enable advanced
Android features.

Assuming you already have a minimalist defconfig for your device, a possible
way to enable these options would be:

ARCH=<arch> scripts/kconfig/merge_config.sh <path_to>/<device>_defconfig android/configs/android-base.cfg android/configs/android-recommended.cfg

This will generate a .config that can then be used to save a new defconfig or
compile a new kernel with Android features enabled.
 

TechVendetta

Active member
May 16, 2011
31
0
0
Visit site
Well I installed the rom last night on my 1st gen vm670, I'm already starting to see what patches and fixes it needs so I'll start playing with it when I get off work tonight and post my fixes here. Great work so far, all the bugs my personal rom of jellybean had are fixed here.
 

TechVendetta

Active member
May 16, 2011
31
0
0
Visit site
The issues I'm currently looking into is improper mount/detection of the sdcard vfat partition (it mounts just fine in /mnt/media_rw but never seems to be mountable, or accessible from /sdcard), The time either lags or freezes after going into deep sleep (notice this most while charging for the night and waking up to see a 2 hour old time, solved on reboot), Maps works but has the rendering glitch, usually caused by gsf. Also the ramdisk needs to be cleaned and organized. There's unnecessary hardware loads in some of the files and duplicate loads, this makes the bootup logcat harder to debug. It seems that for sdcard mounting we need to not use the external sdcard method but instead duplicate the mount method used by phones with only internal storage. As for the mtdblocks, we can probably either have a slim stable build there which the boot can load after checking for /dev/mmcblk0's presence or use it as swapspace.

Can someone link me to the current repo and device config for the release? I think there's something framework level blocking some of my fix attempts from the ramdisk.
 
Feb 19, 2011
1,971
284
0
Visit site
The issues I'm currently looking into is improper mount/detection of the sdcard vfat partition (it mounts just fine in /mnt/media_rw but never seems to be mountable, or accessible from /sdcard), The time either lags or freezes after going into deep sleep (notice this most while charging for the night and waking up to see a 2 hour old time, solved on reboot), Maps works but has the rendering glitch, usually caused by gsf. Also the ramdisk needs to be cleaned and organized. There's unnecessary hardware loads in some of the files and duplicate loads, this makes the bootup logcat harder to debug. It seems that for sdcard mounting we need to not use the external sdcard method but instead duplicate the mount method used by phones with only internal storage. As for the mtdblocks, we can probably either have a slim stable build there which the boot can load after checking for /dev/mmcblk0's presence or use it as swapspace.

Can someone link me to the current repo and device config for the release? I think there's something framework level blocking some of my fix attempts from the ramdisk.

this'll get you going.
init a repo from github.com/androidarmv6/android, branch cm-11.0
drop this xml in .repo/local_manifests, modify as desired.
View attachment roomservice-full.xml.zip
substitute skinbark for bigsupersquid in the kernel line if you want optimus S keylayout and other changes.
the basic files for boot from sd are in this flashable zip, also modify as desired.
kitkat-sd.zip
post your improvements!
 

skinbark

Well-known member
Nov 22, 2013
289
0
0
Visit site
Since this is the experimental thread, I'll mention aufs again.
It'd be nice to have a minimal install in NAND that would run without the card, and overlay partitions onto system, data, and cache (all three because of the dalvik and other bits that wouldn't enjoy only existing in one of two places of the card was removed,) so it'd be possible to pull the card while running but still have the ridiculous space option from the ext partitions.
settings>storage would probably need modified to handle the two partitions beyond what it expects for a clean unmount/remount.
Maybe a modified vold that can activate/deactivate the aufs along with the vfat partition?
 
Feb 19, 2011
1,971
284
0
Visit site
Maybe a modified vold that can activate/deactivate the aufs along with the vfat partition?
you could try... I have yet to edit any Android source except for assembly. Kernel, yes, and patching in Android fixes, init.rc mods, and other files, yes, but not actual new source. Java, my experience is limited to J2ME until I dig into android. I have reasons to but haven't yet.
here, look at this. apparently I do things the long way around.
m.ksy said:
Hi! My device (Motorola Droid X) supports only fakeflash recovery. For CWM Recovery to build it I use "make -j# recoveryzip". Does TWRP support fakeflash building?
from the xda twrp-building thread
 

TechVendetta

Active member
May 16, 2011
31
0
0
Visit site
I still think dualboot via recovery would be best, eventually ill get around to making the recovery scripts for it. I just recently started setting my build servers back up so I'm hoping to pull down the source over the weekend and merge some nexus code to make the os2sd mount properly. Also I wonder how much battery is used by having the button leds on constantly. And did anyone on virgin mobile notice this didnt require the apn patch? Data just works. I know VM announced network upgrades, maybe its 100% sprint compatible now.
 

Trending Posts

Forum statistics

Threads
942,377
Messages
6,913,745
Members
3,158,380
Latest member
knowledge is power