Any chance you pulled any of the /proc/last_kmsg from the reboots?
-
After 15+ years, we've made a big change: Android Forums is now Early Bird Club. Learn more here.
Root Cyanogenmod-Tweaked Kernel
- Thread starter Yoinx
- Start date
-
- Tags
- lg spectrum
Any chance you pulled any of the /proc/last_kmsg from the reboots?
Dear God that file is huge! I started the cmd let it run for like at least 5 minutes and terminal never really finished. I had to do a ctrl c to stop the running command. Should I have kept it running until the command ran it's course? Logcat is an active running process correct? If so it was just journalling in process then.
If all good with the above, I will probably send it to my dropbox and provide a link
Yeah, sorry about that.
Yeah you only need to let it run for a couple seconds. If you want it to be smaller... it's easy enough to just do
logcat -c
logcat
do wahtever you're trying to get logged.
ctrl +c
hahaha so now what I need to do is get in the car start start the logcat process, then let it run then when it messes up, catch it in logcat then ctrl c
yeah, it's kinda hard sometimes.
You could also use an app like catlog, I think it'll record your logcat to sdcard when you tell it to record and stop.
Right now, I'm trying to just pull in changes/fixes etc from the main linux kernel source.
But, I'm having a hell of a time adding the kernel source as a remote.
You could also use an app like catlog, I think it'll record your logcat to sdcard when you tell it to record and stop.
Right now, I'm trying to just pull in changes/fixes etc from the main linux kernel source.
But, I'm having a hell of a time adding the kernel source as a remote.
Sounds like good times! Thanks for at least trying to get it working dood
Here's the real question...
Can you get it to fail pretty reliably?
I'm manually inserting the bt fixes from kernel.org right now. There's a couple for HCI, which should be how you're connecting to your head unit. If anyone can get it to fail reliably, they'd be the best to test it out. I'll have a new build up here before long.
Can you get it to fail pretty reliably?
I'm manually inserting the bt fixes from kernel.org right now. There's a couple for HCI, which should be how you're connecting to your head unit. If anyone can get it to fail reliably, they'd be the best to test it out. I'll have a new build up here before long.
Yeah... Nooooooo but here is what I have as the latest log. I got it to hold steady by connecting from the phone to the deck. But the issue is on reconnect. The deck reaches out to bluetooth devices and then reports no connection do I want to retry. In the retry it fails as well.
Hope this helps
http://db.tt/CSNFReO9
Yeah, I uploaded v36. Link's in the OP.
See if maybe it fixes it at all for you?
I'm still looking through commits and such to see if there's anything out there that helps. I do notice a few
06-07 16:42:32.982 E/BTLD ( 1214): ######################################################################
06-07 16:42:32.982 E/BTLD ( 1214): #
06-07 16:42:32.982 E/BTLD ( 1214): # WARNING : BTU HCI(id=0) command timeout. opcode=0x409
06-07 16:42:32.982 E/BTLD ( 1214): #
06-07 16:42:32.982 E/BTLD ( 1214): ######################################################################
In there. Which isn't especially great >.>
Looks like I missed http://git.kernel.org/cgit/linux/ke...y&id=6ffb80e739e282d7b9ffa43c2ec2a9766c8099a1 on v36.
Also missed: http://git.kernel.org/cgit/linux/ke...y&id=073c4aec557b87d0c5b6f8b6b1910b356088eaf3
Which actually sounds like it might be what's happening. I'll post up a test build with those two included. (instead of bumping build numbers)
Try this one, it's got the two previously mentioned missed fixes.
Dev-Host - Cyanogenmod-Tweaked.36-test.zip - The Ultimate Free File Hosting / File Sharing Service
Dev-Host - Cyanogenmod-Tweaked.36-test.zip - The Ultimate Free File Hosting / File Sharing Service
Got it downloaded bout to flash
Seems smooth. Do you want a logcat of just running? I am not in the car now so I can't test that
Nah. I'm just curious if any of the fixes from upstream fixed anything.
I'm building another copy with the changes TDM merged into his quatrimus kernel.
Kool. Release it and I will run it
Don't know if it would do any good. Now I'm getting some errors that look like they point back to the RoM... or maybe something's not handshaking right. I had to change some settings on Ubuntu to make it try to use it for audio though, could be there... gonna have to try under windows I guess..
[HIGH]
D/BluetoothAdapterService(1089588768)(23977): getState(): mAdapterProperties: com.android.bluetooth.btservice.AdapterProperties@40f1d1d0
D/dalvikvm( 943): GC_FOR_ALLOC freed 2779K, 20% free 22604K/28224K, paused 171ms, total 171ms
D/BluetoothAdapterService(1089588768)(23977): getState(): mAdapterProperties: com.android.bluetooth.btservice.AdapterProperties@40f1d1d0
I/HeadsetService(10837): Audio session removed: 61
I/HeadsetService(10837): Selected configuration: bluetooth
D/dalvikvm( 4277): GC_EXPLICIT freed 1830K, 72% free 3187K/11060K, paused 2ms+5ms, total 32ms
D/dalvikvm( 5734): GC_FOR_ALLOC freed 1413K, 73% free 3150K/11420K, paused 21ms, total 21ms
I/ActivityManager( 943): Start proc com.andrew.apollo for broadcast com.andrew.apollo/.MediaButtonIntentReceiver: pid=24664 uid=10000 gids={50000, 1015, 1023, 3003, 1028}
I/audio_a2dp_hw( 219): out_set_parameters: state 3
I/str_params( 219): key: 'exiting' value: '1'W/AudioService( 943): Audio Settings Permission Denial: setBluetoothA2dpOn() from pid=11692, uid=10073
I/audio_a2dp_hw( 219): out_set_parameters: state 3
I/str_params( 219): key: 'closing' value: 'true'
I/audio_a2dp_hw( 219): out_set_parameters: state 2
I/str_params( 219): key: 'exiting' value: '1'
D/A2dpStateMachine(23977): Connection state BC:77:37:49:CD:A6: 2->0
D/BluetoothAdapterService(1089588768)(23977): getState(): mAdapterProperties: com.android.bluetooth.btservice.AdapterProperties@40f1d1d0
D/BluetoothAdapterProperties(23977): CONNECTION_STATE_CHANGE: BC:77:37:49:CD:A6: 2 -> 0
D/BluetoothAdapterService(1089588768)(23977): getState(): mAdapterProperties: com.android.bluetooth.btservice.AdapterProperties@40f1d1d0
D/BluetoothAdapterService(1089588768)(23977): Get Bonded Devices being called
I/audio_a2dp_hw( 219): adev_close_output_stream: closing output (state 2)
I/audio_a2dp_hw( 219): stop_audio_datapath: state 2
I/audio_a2dp_hw( 219): skt_disconnect: fd -1
I/audio_a2dp_hw( 219): skt_disconnect: fd 13
W/AudioService( 943): Audio Settings Permission Denial: setBluetoothA2dpOn() from pid=11692, uid=10073
[/HIGH]
Tried setting everything in /etc/bluetooth/ to 755 to see if that might have started causing the permission denial... no dice.
The kernel? I don't know that there's a different source for it anymore.
I thought the 3.0.x kernels ran on ICS.
Maybe try LG's stock source? https://wws.lg.com/global/support/opensource/opensourceList?types=ALL&search=vs920
If you'd rather it on github, I could push it to a repo there. This has the stock kernel and android sources. No mods though... I imagine it should be easy enough to clone/cherry pick commits though. Enjoy.
There is a permission in ubuntu or Linux that is a wide open permission setting that might be good for testing. I wanna say is 999. But don't hold me to that. Ubuntu Linux or Linux Mint is all I use. None of that windows crap
You're thinking of 777.
That gives world read/write/execute.
It's horrible to use, but yeah... it removed that issue for testing.
No, was waiting for a phone call so just dirty flashed AOKP real quick... I also realized I hadn't adjusted freq or governor, which my phone's never liked over clocking so that may of been it, I'll try it again tomorrow.
Yeah. That's very likely it. I never went back and reset it to 1.5 ghz as default. And right now the frequencies above 1'5 are iffy with the voltage table.. I'll get those a less conservative tackle and it might fix that. I mean...if you're using anything over 1.5 you don't probably care about battery anyway
Personally for the kmsg I have an init.d script pulling them to the cache with the date and time as the filename on reboot. Just on case there are multiple reboots or if it happens while you're out. Though it does it on every reboot whether you crashed or not.
Sent from my VS920 4G using Tapatalk 2
V38 up, still no improvement as far as I can tell in Bluetooth. I baked in a couple things from V6SuperCharger. Should help make things a little snappier hopefully.
Yes, I missed V37 because I bumped it, then changed a few more things. I decided to leave it bumped.
@Nokoko, I changed the default max freq back to 1.5ghz.
If people want it higher they can turn it up themselves. Should help with some stability. I also turned the voltages for anything over 1.5ghz back up a little. Should also help with stability.
Yes, I missed V37 because I bumped it, then changed a few more things. I decided to leave it bumped.
@Nokoko, I changed the default max freq back to 1.5ghz.
If people want it higher they can turn it up themselves. Should help with some stability. I also turned the voltages for anything over 1.5ghz back up a little. Should also help with stability.
Hm. I wonder what I did to make it mad. :s
I'll work on undoing whatever to get it back to where it was.
What was the last one you had used that wasn't like that?