[Kernel][android_kernel_lge_msm7x27-3.0.x][bigsuperkernel-3.08]

_alexkta

Well-known member
Nov 3, 2012
289
0
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

Finally got to flash test 50 with build 6 of JBC. It's great! Accelerometer is definitely working, and charging status shows correctly as well, after wake of course. I ran into a similar situation as did at least one other person.

I flashed in this order:
  1. ROM
  2. Kernel
  3. Virgin Fix 3 (- /system/usr/)
  4. Gapps (KonstaKang)

I then rebooted, signed into my Google account, waited until contacts synced, and changed a few UI settings. Everything was running really smooth, so I rebooted into recovery.

Once in recovery, I made a nandroid, and then flashed Crossbreeder. I rebooted and got the welcome screen again with no radio service. I figured Crossbreeder had screwed something up, so I rebooted into recovery and restored the nandroid.

After restoring, I booted back into Android, and of course, everything was fine. I tested cell and wifi, and both were working fine. I rebooted, and then after a few minutes rebooted again into recovery with the intent of flashing an int2ext script. I flashed CronMod int2ext+, rebooted, and got the welcome screen again, yet again with no cell service. Wifi was still working, so I figured why not try to sign in again and see what happens. After signing in, I still had no cell service, so I rebooted.

After this reboot, everything was/has been fine. After a few more reboots, I booted into recovery and flashed Crossbreeder. Everything is running fine now. Weird. I don't have a clue as to what in either Crossbreeder or the int2ext script may have caused that to happen. But that's what happened. Hopefully that's of some use to you. If not, at least know that running your kernel with JBC is extremely satisfying! Thanks again for all your work!

I was also having this problem and it was pretty annoying it would happen everyday... Anyways after using the int2extv2+ & a different gapps then the Konsta ones I haven't had a problem in 3 days

Sent from my LG-VM670 using Tapatalk 4 Beta
 
Feb 19, 2011
1,971
284
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

test 60
cleans up charger notification a little... breaks into two separate functions for charger and battery rpc calls.
//edit: my testing says battery shows full regardless of voltage. binary pulled until repaired.
still no instant notification. sleep/wake only. looking at da9030_battery.c (as suggested by something in the Documentation/power files in the kernel source) for a timed refresh instead of a callback format to at least get closer to realtime.
 
Last edited:

zemm356

Member
May 16, 2013
13
0
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

Hi to all!!
I am just user and beginner, so I am have LG Vortex Verizon Mobile. All good if it is a clean rom 2.2.1 or 2.2.2, I am tried many custom ( all which found for Vortex, all for Optimus V, Optimus S). All are good but some bugs, main is no autorotation.
Your kernel is perfect, but some strange, if I use CM9 quatrimmus beta2 for VS660 + init pac or init cai + your kernel 45 - 60, Rotation works well, all fast and clear, but WiFi cannot On((( and camera says need to install SD card((( tested on Maceks Mod 10.0, 4.1.2 - no rotation, no wifi
If I use b6-cm-10.1-20130715-thekraven-LS670, 4.2.2 for Optimus V/S, all works good, hard keys no light, rotation like mirror ( meaning up is down and down is up) for all positions((( wifi and BT work well
can you do some amendments in kernel for VS 660 (LG Vortex)? IF need to Donate, indicate amount.
Regrards, Igor
 
Last edited:
Feb 19, 2011
1,971
284
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

Hi to all!!
I am just user and beginner, so I am have LG Vortex Verizon Mobile. All good if it is a clean rom 2.2.1 or 2.2.2, I am tried many custom ( all which found for Vortex, all for Optimus V, Optimus S). All are good but some bugs, main is no autorotation.
Your kernel is perfect, but some strange, if I use CM9 quatrimmus beta2 for VS660 + init pac or init cai + your kernel 45 - 60, Rotation works well, all fast and clear, but WiFi cannot On((( and camera says need to install SD card((( tested on Maceks Mod 10.0, 4.1.2 - no rotation, no wifi
If I use b6-cm-10.1-20130715-thekraven-LS670, 4.2.2 for Optimus V/S, all works good, hard keys no light, rotation like mirror ( meaning up is down and down is up) for all positions((( wifi and BT work well
can you do some amendments in kernel for VS 660 (LG Vortex)? IF need to Donate, indicate amount.
Regrards, Igor
I didn't realize VM670 kernel would work at all on VS660!
I would need a VS660 to test with to do any work on that model. Not sure how long it would take.
 

zemm356

Member
May 16, 2013
13
0
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

I didn't realize VM670 kernel would work at all on VS660!
I would need a VS660 to test with to do any work on that model. Not sure how long it would take.

Here no so big choice of ROMs for VS660 at all. I started to test all similar ROMs for LG phones, ROM for Optimus One refused by phone)) so I begin to see for CDMA phones. Optimus V and S are similar in spec, so if use S, hard keys need to remap, V ROM better, but here 2 types, one do not work at all, second works +/- well.
So in fact I have LG Verizon VS660 with b6-cm-10.1-20130715-thekraven-LS670 ROM + Virgin.fix4.3.0kernel.jbc-CM10.1 + 3.0.8.test_60
this is most stable working, only 100% batt all time and accelerometer working +180 from real position.
in case if used 3.0.8.test_55 - can work 1-2 hours, 10 hours after just switch off, need to remove batt and put back
3.0.8.test_50 testing now
main problem, if phone connected to charger or to PC and to reload it, it freezing at booting(((( need to re-instal kernel than and all ok)))
 
Last edited:
Feb 19, 2011
1,971
284
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

there's still issues getting the charger data right in the source code. the 3.0.8 stock msm_battery.c doesn't read most of the battery data, and doesn't read charger data right until after boot is finished. So it is still freezing up boot on some ROMS because the offline charging is locking up trying to read the charger info. You can remove /sbin/chargerlogo from the ramdisk to stop it or disable it in init.thunderc.rc in the ramdisk as well to allow booting until I get the battery fixed. I can maybe figure out what the difference between test 40 and on up is, because those pre-accelerometer versions boot ok with the offline charging enabled (they just don't see the charger at all.) I pulled that test 60 build because of the 100% battery issue. trying to get an alternate up soon.
the inverted accelerometer could be worked out in the driver itself. I just need hardware to test on, because of all the boot/check/rebuild cycles needed to figure out glitches.
 

zemm356

Member
May 16, 2013
13
0
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

the inverted accelerometer could be worked out in the driver itself. I just need hardware to test on, because of all the boot/check/rebuild cycles needed to figure out glitches.
that's very strange, as on 4.0.4 + your kernel 45-55 accelerometer working perfectly. but wifi disabl by this new kernel.
with 4.2 wifi working well, but accelerometer inverted, and this specially can see on photo, in any position it rotate +180 from current. so up became down)))
 
Feb 19, 2011
1,971
284
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

compare the /thunderc.init.rc files for wifi differences between 4.0.4 and 4.2. match the 4.2 settings into the 4.0.4 file and maybe your wifi will inititialize correctly.
 
Feb 19, 2011
1,971
284
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

test 61
'fixes' battery readout from test 60. bounces around like a jumping bean, though. manual calculation, probably slows down the kernel too. need to retrieve the value from battery rpc like before, no more time this afternoon though.
 
Last edited:

zemm356

Member
May 16, 2013
13
0
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

compare the /thunderc.init.rc files for wifi differences between 4.0.4 and 4.2. match the 4.2 settings into the 4.0.4 file and maybe your wifi will inititialize correctly.

compare, size differences first of all and what in it
so working composition is next step by step: quattrimus-VS660-cm9-beta2 + 3.0.8.test_61.zip + init.thunderc.rc.cai.zip + OS-keyswap.zip
WIFI +
GPS +
Accelerometer + ( all working perfectly)
Camera and SD card + ( disappeared bug with SD, this only with happens using init.thunderc.rc.pac.zip)

and In fact it is VS660, 4.0.4, kernel 3.0.8.test_61

BUGS
cannot install GAPPS
if for ICS 4.0 - home button correct, but Google play closed in time of loadings after authorization
if for 4.2 - home hard key works like PG UP button... Google same story..
and known BUG indicated in 1st post
init.thunderc.rc.cai.zip for CAI to stop bootloop. no adb in CAI with this... in fact, no usb functionality at all. no charging indicator

night is long... try to see the turn for installing...
* note if I am installing GAPPS before kernel, apps no want to work, if after, same((( ** if use Virgin.fix4.3.0kernel.jbc-CM10.1 - hard key +, but WIFI -
If no installing GAPPS, manually install from SD all I need, all works perfectly for me:p:p:p:p

EDIT:
F**K that was amazing
so step by step
1. WIPE ALL + DALVIK + SYSTEM
2.quattrimus-VS660-cm9-beta2
3. gapps ics
4. kernel 61
5. init cai
6. os-kayswap
7. reboot

working WiFi, Accelerometer all rotations as need, Bt, GPS, root explorer ( put in apps), Navitel, Kate Mobile, Hard Keys as need.
installed new Google Play Store 4,2,9 working 2-3 sec after - Unfortunately. Google Play Store has stopped. If clear apps data, start, again accept conditions, it show 3 apdates for my apps, and 3-5 sec - stopped
so all apps need to discover in the internet and install from SD\PC (((
same no charge indicator and usb((
continue to testing)))
// LAST EDIT
no want to talk a lot
done next way
1. Wipe
2. ROM CM9
3. Kernel 061
4. Keyswap
5. Init ( took original 4.0.4 init.thunderc.rc, deleted boot logo line, replaced all wifi lines from init.thunderc.rc.cai)
6. flash gapps-ics-20120429-small
7. Reboot

Well, all working well +/-. no usb, no hardkey light, no battery indications
maybe can find some solution soon)))
wifi, accelerometr, finally works Play Stoere, Google play services, Gmail, etc ( all Google)
Installed apps
Facebook + Messenger
Foursquare
Skype
Kate Mobile
Navitel navigation full
Smart Keyboard Pro
MXplayer +codec
Unnistaller for root

Many thanks to bigsupersquid
 

Attachments

  • Screenshot_2013-07-31-13-09-43.png
    Screenshot_2013-07-31-13-09-43.png
    72 KB · Views: 175
  • Screenshot_2013-07-31-13-09-58.png
    Screenshot_2013-07-31-13-09-58.png
    80.4 KB · Views: 177
Last edited:
Feb 19, 2011
1,971
284
0
Visit site
Re: [SOURCE][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

test 65
should make battery readout more accurate and less jumpy. Reads 100% until after sleeping first time, though.
patch for Novatek driver dmesg error.
usb host mode turned off since I can't get it working yet.
has unnecessary file system support in it that will be removed later (ISO, etc.)
 
Last edited:

zemm356

Member
May 16, 2013
13
0
0
Visit site
Re: [SOURCE][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

test 65
should make battery readout more accurate and less jumpy. Reads 100% until after sleeping first time, though.
patch for Novatek driver dmesg error.
usb host mode turned off since I can't get it working yet.
has unnecessary file system support in it that will be removed later (ISO, etc.)
HI
with this new kernel got same problem like with 60 ( batt 100% all time)
but this is on my VS660 with 4.0.4
 
Feb 19, 2011
1,971
284
0
Visit site
Re: [SOURCE][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

test 70: remove extra filesystems, tweak at battery some more. should read battery level correctly on boot again. (nope, reads 100% till it dies. use at own risk.)
still wake/sleep deal for charger notify and battery level update.
 
Last edited:

zemm356

Member
May 16, 2013
13
0
0
Visit site
Re: [SOURCE][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

test 70: remove extra filesystems, tweak at battery some more. should read battery level correctly on boot again. (nope, reads 100% till it dies. use at own risk.)
still wake/sleep deal for charger notify and battery level update.

got same problem like with 60 ( batt 100% all time)
but this is on my VS660 with 4.0.4
so i am used 50test
 
Last edited:
Feb 19, 2011
1,971
284
0
Visit site
Re: [SOURCE][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

I am getting really aggravated with the battery code.
every time I try to incorporate callbacks for battery info, the system reboots without warning or anything in the kernel log.
I tried to implement a 10-sec update function instead of handling callbacks, but it caused a straight-up kernel panic with the work thread instead. Same crash as trying to use the 2.6 msm_battery_thunderc.c code.
I cannot find diddly squat with google about using RPC callbacks. Barely any info about RPC calls either.
Anybody got any pointers to RPC/callback programming examples?
 

flapjack.fiasco

Well-known member
Sep 13, 2011
776
102
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

Way over my head at this point, but I'll do a little digging later tonight. Maybe I'll find something relevant.

Sent from my LG-VM670 using Tapatalk 4
 

flapjack.fiasco

Well-known member
Sep 13, 2011
776
102
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

Could you point me to which files I should be looking at? I've done a little C++ in school, and I assume that the battery code is in C, right? I can probably follow along, at least a little. Maybe that will better inform my searches as well. I'll take a look at your github, but if you could point me straight to the relevant files, that would be awesome.
 
Feb 19, 2011
1,971
284
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

drivers/power/msm_battery.c is stock and has the callback code.
you'll need the commit from before I started messing with that file. look under commits and you'll find it.
drivers/power/msm_battery_thunderc.c is the modified version.
thekraven's 2.6 kernel has the original of those two files to compare. the link is in the OP.
there's bits in drivers/power/power_supply_sysfs.c in the 2.6 kernel, some in arch/arm/mach-msm/lge/LG_rapi_client.c too. probably more I haven't found.
lots missing in the 3.0 kernel I didn't bother to patch in cause the callbacks crash.
 

dingermtb

Well-known member
Aug 16, 2011
203
26
0
Visit site
Re: [Kernel][ALPHA][android_kernel_lge_msm7x27-3.0.x (3.0.8)]

So far, the best version for me, and most accurate battery reading is test 55. This is the version I keep going back too :D

Sent from my ZombieBean-VM670 using Xparent Blue Tapatalk 2