This is a long post, but I need assistance trying to get to the bottom of the random reboot and Black Screen of Death (BSOD) issue on the P93x series phones. Random chatter and dismissals are getting us nowhere, we need some structured approach.
I've submitted a CM bug report, and got Issue 6762 assigned. Feel free to star and follow there as well.
General Background Info
This seems to be a polarising issue: people who are in the know (technically) have devices without this problem, and people with this problem think the sky is falling and wondering why no fixes are coming. So before you jump all over me and tell me your P93x is fine and the problem does not exist, a little background is in order. I have two P930’s, one is humming along just nicely, thank-you-very-much, the other is experiencing the random reboot issue as well as the bsod, as validated by others in this forum.
Kernel OOPS on P930
Issues
First, I believe there are three issues at play (likely separate)
Random reboots especially when scrolling. Likely a kernel panic;
and, as per a good classification here
BSOD-A: Screen remains black, hardware buttons light up and touching buttons gives haptic feedback
BSOD-B: Screen and buttons remain black, no response to power button. Phone may become hot AND battery has taken a large hit when finally rebooted.
Scenarios for the above:
Using “heavy” apps like Google Maps, Waze, etc.
scrolling on lists or typing in some apps
device goes to sleep mode, can’t wake up the LCD. Keys light up, system up and running (background apps & loggers continue to operate) .. but can’t get to them.
devices reboots (often is is unseen since device is charging) and goes dead, can’t wake up. No lights, system down.
Both BSOD-A and BSOD-B require battery pull or long press of power button to restart device.
I am focusing on BSOD-B since ATM that's the bulk of empirical data I have at hand and it's a notoriously reoccurring problem.
If I manage to capture some good logs for BSOD-A (which are very rare for me), I will submit a separate bug report.
Troubleshooting BSOD-B
Here is a list of things I tried to narrow down where the problem may be:
wipes, full wipes, cache wipes, formats, etc.
nandroid clone of the stable P930 to the unstable P930
go through the unbrick process
flash back to Gingerbread
flash back to factory ICS
disable HW overlays in Developer Options
Here is what the effect these things had:
wipes stabilize the system for about a day or two before the problem reoccurs
nandroid clone had no effect. The unstable one still reboots/bsod
Unbrick process gives temporary relief. Problem still comes back after a day or so,
Gingerbread seems to be completely stable. No reboots or BSOD’s after a week of running it and incredible battery life.
ICS introduced random reboots, but no BSOD.
that seems to take care of the reboots while scrolling.
Observations
So first obvious observation: solution d) seems to fix all the problems. Buuuut … anyways moving on.
Second observation, solution f) seems to take care of vast majority of problem #1 under scenario 1b), so IMHO it’s a viable workaround, even though disabling HW overlays increases CPU usage and reduces battery life.
Third observation, BSOD-B most frequently occurs when under power, either AC or USB.
Fourth observation, based on analysis of kmsg for BSOD-B, the battery drain is due to a kernel fault, followed by a reboot, followed by CPU0 stall (after trying to initialize the LCD?). Here’s the kmsg related to this:
Code:
[ 2.021728] lm3530_probe: i2c probe start
[ 2.022369] mipi_lgit_lcd_probe: mipi lgit lcd probe start
[ 2.022521] setting pdata->panel_info.fb_num to 3. type: 8
[ 2.022979] mdp4_dsi_vsync_init: ndx=0
[ 2.029266] FrameBuffer[0] 720x1280 size=11304960 bytes is registered successfully!
[ 62.040008] INFO: rcu_sched_state detected stalls on CPUs/tasks: { 0} (detected by 1, t=6002 jiffies)
[ 62.040344] Backtrace for cpu 1 (current):
[ 62.040435] [<c010cf9c>] (unwind_backtrace+0x0/0x12c) from [<c010b1f4>] (smp_send_all_cpu_backtrace+0x60/0xc4)
[ 62.040618] [<c010b1f4>] (smp_send_all_cpu_backtrace+0x60/0xc4) from [<c0214fe8>] (__rcu_pending+0x43c/0x5e0)
[ 62.040802] [<c0214fe8>] (__rcu_pending+0x43c/0x5e0) from [<c02152fc>] (rcu_check_callbacks+0x170/0x1e0)
[ 62.040954] [<c02152fc>] (rcu_check_callbacks+0x170/0x1e0) from [<c01cec1c>] (update_process_times+0x38/0x68)
[ 62.041107] [<c01cec1c>] (update_process_times+0x38/0x68) from [<c01f2798>] (tick_sched_timer+0x98/0x258)
[ 62.041290] [<c01f2798>] (tick_sched_timer+0x98/0x258) from [<c01e6930>] (__run_hrtimer+0x1c0/0x304)
[ 62.041381] [<c01e6930>] (__run_hrtimer+0x1c0/0x304) from [<c01e6da8>] (hrtimer_interrupt+0x168/0x2f8)
[ 62.041564] [<c01e6da8>] (hrtimer_interrupt+0x168/0x2f8) from [<c0117b54>] (msm_timer_interrupt+0x18/0x20)
[ 62.041717] [<c0117b54>] (msm_timer_interrupt+0x18/0x20) from [<c020f51c>] (handle_percpu_devid_irq+0xfc/0x1a8)
[ 62.041870] [<c020f51c>] (handle_percpu_devid_irq+0xfc/0x1a8) from [<c020bb74>] (generic_handle_irq+0x2c/0x40)
[ 62.042022] [<c020bb74>] (generic_handle_irq+0x2c/0x40) from [<c0106ea0>] (handle_IRQ+0x7c/0xc0)
[ 62.042205] [<c0106ea0>] (handle_IRQ+0x7c/0xc0) from [<c01003b4>] (gic_handle_irq+0x80/0xac)
[ 62.042297] [<c01003b4>] (gic_handle_irq+0x80/0xac) from [<c0853754>] (__irq_svc+0x54/0x80)
[ 62.042388] Exception stack(0xda65ff98 to 0xda65ffe0)
[ 62.042541] ff80: 00000020 c0c23fb8
[ 62.042633] ffa0: da65ffd8 00000000 da65e000 c0c2d5c8 c0cb4124 c086e55c 4020406a 510f02d2
[ 62.042785] ffc0: 0000001f 00000000 00000000 da65ffe0 c0106fdc c0107384 60000013 ffffffff
[ 62.042968] [<c0853754>] (__irq_svc+0x54/0x80) from [<c0107384>] (cpu_idle+0x84/0xe8)
[ 62.043121] [<c0107384>] (cpu_idle+0x84/0xe8) from [<40a4b754>] (0x40a4b754)
[ 62.043212]
[ 62.043212] sending IPI to all other CPUs:
Why battery discharge
Gist of the problem: pm8058-charger driver and related infrastructure - like USB - is not loaded yet, CPU0 is hung, CPU1 observes that its colleague is hung and faults, re-tries, and the party continues until the battery is dead.
Possible Workarounds
The only way to avoid to BSOD-B while charging is to shut down and put phone in power-off charge-mode. This is not completely viable since BSOD-B also occurs in the car when navigating or any time USB power is on.
Improving empirical data
Challenges and roadblocks
Hanging this way and so early during boot presents a problem when retrieving kmsg. The new kmsg with the CPU0 stall replaces last_kmsg on reboot thus wiping any evidence of the real kernel panic which caused this whole thing in the first place.
To make it even more fun, there is a non-zero probability that a kernel panic followed by a reboot does not hang CPU0 .. meaning you’ll never know if a kernel fault will cause CPU0 to hang until it hangs by which time it’s too late to recover the all important last_kmsg. Dead end.
For further kick in the nuts, when the CPU locks and drains the battery down to zero (or is powered off) the device turns off and boots to charge mode (since it’s under power), now wiping the CPU0 stall kmsg and stuffing last_kmsg with charge-mode messages, this time wiping evidence of the CPU0 lock described above. This can be avoided by catching the device early in the discharge cycle and unplugging the device before forcing a reboot .. but still that only gets us to read the CPU0 lock kmsg but not what caused this whole mess in the first place.
Next steps
Any ideas on how to backtrack and get better logs of the initial crash? Logcat appears useless here since it's a userspace process which gets yanked the moment a kernel oops occurs. Even piping kmsg to a file does not work due to the same problem (ie. userspace). Is there a way to tell the HW to shut down on kernel oops rather than reboot? Do I really need to disassemble these two phones in order to compare and determine differences?
Any constructive ideas, suggestions, or thoughts will be greatly appreciated.
EDIT: As of April my P930 which exhibited this behaviour is in the hands of tdmcode. This way he has something to test on and hopefully at some point he'll be able to figure out a workaround for this issue.
Finally!
Finally!
Thanks for officially posting the bug. Some of us didn't have the ability to technically describe the problem and now you have done it for us. Hopefully some fixes will come so that we can all start getting normal usage of our phones, which are powerful when working reliably.
Thanks again!
I'm glad you reported it. It's probably a combination of more than one bug, though, so hopefully it gets resolved. I'm mostly hoping CM10.1 somehow resolves the bugs in CM10 when Ricardo gets around to working on our phone.
Sent from my LG-P930 using Tapatalk 2
Very nice summary.
However, given your own experience with two of them, and the "polarising" experiences, it seems to imply a hardware issue for at least one of the three situations, for example, a chip was revised at some point in production, or there was poor quality control on chip or component manufacture, etc.
I never had BSOD-B. The other two, well ... :crying:
Some other observations:
- P935, i.e., Telus, users seem to be more affected (or more proportionally more vocal) than Nitro HD users. Telus was/is offering affected users a new One X (or equivalent) as a straight swap. This is not typical carrier behaviour in my experience, especially after one has had a phone for over six months ...
- P936 users seem to be less affected.
Good luck on your quest, I hope it is successful!
I definitely agree with your hypothesis that it's hardware related. I suspected this when HONO first released an overclocked kernel for gingerbread. Some people could overclock to 1.8GHz without issue, and my phone would crash randomly and during benchmarks whilst underclocked massively. Same thing happened with Full Throttle for CM9.
Glad that someone actually submitted a bug report, thanks a lot!
Agree, this definitely presents itself like a QC or product rev issue. Based on how LG has handled it, it would seem it's a Qualcomm issue tho - otherwise I am sure LG would address it in their ROMs. This is why I don't hold out much hope ever seeing a fix/workaround for this.
I have no issues running the LG ICS releases (bell and att), except for an issue where the bluetooth driver becomes unresponsive. My head unit still reports being connected, but my phone will not turn off bluetooth or route audio through bluetooth OR speakers.
Sent from my LG-P930 using XDA Premium HD app
Blue tooth is your only issue? You sir are quite lucky.
Sent from my LG-P930 using xda app-developers app
Reboots and BSOD
Thanks for this post. I can add a couple of things, I have an AT&T LG Nitro, and as you say with GB work just fine, I can add that with v18f (ICS Leak) phone also works fine, no reboots or BSOD. I really think that this problem it's related with video, when I first installed stock v20c I had a very curious (and as far as I know common) problem, I have BSOD and after the phone reboots, a red halo appears on the screen, weird, it was solved after 3 or four reboots.
Another thing that I notice is that when I use Dolphin browser (it also happens with waze) and watch a site with flash content, and close dolphin, a strange drawing problem appears on screen, some kind of flicker, it can be solved reentering dolphin and close it.
I've been used liquid nitro for almost a week, after 6 hours of the installation I have the first reboot, then, I have 4 days without reboots or BSOD, incredible, but the fifth day the problem reappears, 2 BSOD.
Returning to what I think it's the root cause, maybe it could be related with video hardware acceleration, that could explain why v18f works, I think HW acceleration it's not complete implemented in this version, that's why it's not as smooth as v20c and works fine.
fguzmant said:
Thanks for this post. I can add a couple of things, I have an AT&T LG Nitro, and as you say with GB work just fine, I can add that with v18f (ICS Leak) phone also works fine, no reboots or BSOD. I really think that this problem it's related with video, when I first installed stock v20c I had a very curious (and as far as I know common) problem, I have BSOD and after the phone reboots, a red halo appears on the screen, weird, it was solved after 3 or four reboots.
Click to expand...
Click to collapse
Oh man, I've had the halo thing appear once as well and it would not go away .. that was a scary moment as I thought the LCD got fried for some reason. Very very weird effect.
fguzmant said:
Another thing that I notice is that when I use Dolphin browser (it also happens with waze) and watch a site with flash content, and close dolphin, a strange drawing problem appears on screen, some kind of flicker, it can be solved reentering dolphin and close it.
Click to expand...
Click to collapse
Did you try disabling "HW overlays" under Developer options? I have had weird drawing problems for graphically advanced apps, but disabling HW overlays seems to fix that up for me. Also, the overlay issue went away for me in the latest nightly CM10 builds .. no drawing artifacts, and no reboots I can link to video acceleration issues. Wish I could say the same thing about the BSODs ...
fguzmant said:
I've been used liquid nitro for almost a week, after 6 hours of the installation I have the first reboot, then, I have 4 days without reboots or BSOD, incredible, but the fifth day the problem reappears, 2 BSOD.
Click to expand...
Click to collapse
This is the other weird part about these BSODs .. they takes time to build up. When I do a full wipe, I will be OK for a few days, then BSOD hell begins.
Yes, I get the red halo effect too randomly after a BSOD, sometimes it has gone to the point of taking up the while screen with the red, and somewhat flickers as well. However it doesn't usually get that bad.
Sent from my LG-P930 using xda app-developers app
miji2 said:
Did you try disabling "HW overlays" under Developer options? I have had weird drawing problems for graphically advanced apps, but disabling HW overlays seems to fix that up for me. Also, the overlay issue went away for me in the latest nightly CM10 builds .. no drawing artifacts, and no reboots I can link to video acceleration issues. Wish I could say the same thing about the BSODs ...
Click to expand...
Click to collapse
Sadly, liquid nitro or v18f does not have that option. My experience with CM10 was a BSOD nightmare.
I too have gotten all three types of hangs:
1) Reboot to CM logo (every couple days)
2) Can't turn screen back on but other buttons working and responsive (rare)
3) Reboot to LG logo followed by black screen and battery drain (every couple days)
I don't believe I ever had 2 and 3 happen on CM9 FT though, just CM10.
Thanks for submitting a report!
Sent from my LG-P930 using Tapatalk 2
On Official ICS, CM9, and CM10 I always have the random reboots. Doesn't happen in Gingerbread. I have tried every work around that I could find and the random reboots continue. I only started experiencing the wake locks in the newest CM10 nightlies. For now I have gone back to stock rooted gingerbread. At least until the reboot issue is resolved, if it is.
ahdoublexl said:
On Official ICS, CM9, and CM10 I always have the random reboots. Doesn't happen in Gingerbread. I have tried every work around that I could find and the random reboots continue. I only started experiencing the wake locks in the newest CM10 nightlies. For now I have gone back to stock rooted gingerbread. At least until the reboot issue is resolved, if it is.
Click to expand...
Click to collapse
Did you try ICS Leak v18f, I can confirm that there's no reboots or BSOD in this ROM, the problem with this it's battery life, but with setcpu and juice defender I can manage to survive a day.
OK, new theory to test: cache partition corruption causing BSOD
Background:
I observed several times that BSOD type B does not occur until the phone starts to reboot randomly (due to scrolling or otherwise). Same this time: after running the 20121201 nightly (+ reset) smoothly for the past 3 days, I got a reboot due to kernel oops, then my BSOD-B started.
Recovery
This time I was able to fix the Random reboot while charging leading to BSOD-B by just formatting the Cache partition from recovery. I am not sure if I tried that before, but it worked this time. But this kind-of explains why a factory reset would stabilize the phone for a bit ...
Testing
I need others to test so we can confirm this. To help, please try latest nightly, and watch the BSOD causing the phone to get hot. If you get it, format your cache and let me know if it made any difference.
Follow-up
Note, we still don't know what patterns of use cause the random kernel oops .. but at least we would not have to endure dead phones first thing in the morning.
miji2 said:
OK, new theory to test: cache partition corruption causing BSOD
Background:
I observed several times that BSOD type B does not occur until the phone starts to reboot randomly (due to scrolling or otherwise). Same this time: after running the 20121201 nightly (+ reset) smoothly for the past 3 days, I got a reboot due to kernel oops, then my BSOD-B started.
Recovery
This time I was able to fix the Random reboot while charging leading to BSOD-B by just formatting the Cache partition from recovery. I am not sure if I tried that before, but it worked this time. But this kind-of explains why a factory reset would stabilize the phone for a bit ...
Testing
I need others to test so we can confirm this. To help, please try latest nightly, and watch the BSOD causing the phone to get hot. If you get it, format your cache and let me know if it made any difference.
Follow-up
Note, we still don't know what patterns of use cause the random kernel oops .. but at least we would not have to endure dead phones first thing in the morning.
Click to expand...
Click to collapse
Tried the procedure above with todays nightly. So far only one reboot but that was due to the camera bug. No BSOD so far.
Will see if it happens overnight
EDIT: charged overnight and still no BSOD. So far so good. :thumbup:
Edit 2: just got BSOD, so I guess its back to the drawing board.
Sent from my LG-P930 using xda app-developers app
khanfed said:
Edit 2: just got BSOD, so I guess its back to the drawing board.
Click to expand...
Click to collapse
Same here
khanfed said:
Tried the procedure above with todays nightly. So far only one reboot but that was due to the camera bug. No BSOD so far.
Will see if it happens overnight
EDIT: charged overnight and still no BSOD. So far so good. :thumbup:
Edit 2: just got BSOD, so I guess its back to the drawing board.
Sent from my LG-P930 using xda app-developers app
Click to expand...
Click to collapse
Wasn't the whole point to clear cache after any sort of crash-reboot? The idea was a crash-reboot will corrupt the cache and then cause the BSOD.
Sent from my LG-P930 using Tapatalk 2
Yup the procedure above was done after a crash/reboot, but in the end it didn't resolve the issue, since I got the BSOD again.
Sent from my LG-P930 using xda app-developers app
Related
Have a VERY weird issue going on. I've been using Burnt Droid 1.4.2 by Netarchy since it came out. I decided to flash his new 4.3.4 CFS-NoHavs,noSBC, with static undervolt (i wanted to try and see if it was stable since I had success using undervolted kernels last year). I used proper procedure and cleared cache/dalvik-cache before flash. I downloaded the new Angry Birds from amazon and started playing it for a few minutes phone started lagging very bad, then had to restart. I said forget this; undervolt is probably causing this, and reverted back to the original 4.3.1-cfs-nohavs-nosbc which I never had a problem with(used proper procedure above to flash back kernel as well).
Now a day later at night, I decide to run my phone down because I'm trying to recalibrate my battery(I'm at less than 10% at this point). So I load up Wifi, turn on Pandora, and browse the internet. Not even 5 minutes into it, here's whats occuring in order:
1) Bad lag.. Pandora starts stuttering, so I hit the home button
2) HTC Sense restarts, It doesn't even force close, or at least it's not telling me
3) a Hot restart occurs a few moments later... (no white boot screen, just starts bootin from ROM screen)
4) At this point, sometimes its still laggy, sometimes it's fine. If it's laggy I must do a battery pull because It will keep looping to this point.
I can rinse and repeat this running pandora + browsing the web, and the same 4 steps above will happen within moments of eachother. I've been using this ROM since January with no major problems like this
A few side notes:
-I have Evo hardware 0002 (launch day evo)
-Rooted with Toast's original root method, eng .76 boot loader
-Newest radio/PRI/NV/Wimax
-My phone does not randomly reboot on me normally. Reboots only occur when following the above steps.
-I do not use Set CPU
Here is where I am very superstitious. I've been closely monitoring the "Boot looping + no Recovery Access" forum, and similar forums to this. Within the past few months I have tried out a few different CM7 nightlies. I have never flashed RC2 but I have flashed some after and before and tried those out for a while. I just hope my phone isn't starting a dying phase in relation to the problems people are having now.
Is there anything else recommended before I end up doing a full wipe and reflash the rom?
bigblueshock said:
Have a VERY weird issue going on. I've been using Burnt Droid 1.4.2 by Netarchy since it came out. I decided to flash his new 4.3.4 CFS-NoHavs,noSBC, with static undervolt (i wanted to try and see if it was stable since I had success using undervolted kernels last year). I used proper procedure and cleared cache/dalvik-cache before flash. I downloaded the new Angry Birds from amazon and started playing it for a few minutes phone started lagging very bad, then had to restart. I said forget this; undervolt is probably causing this, and reverted back to the original 4.3.1-cfs-nohavs-nosbc which I never had a problem with(used proper procedure above to flash back kernel as well).
Now a day later at night, I decide to run my phone down because I'm trying to recalibrate my battery(I'm at less than 10% at this point). So I load up Wifi, turn on Pandora, and browse the internet. Not even 5 minutes into it, here's whats occuring in order:
1) Bad lag.. Pandora starts stuttering, so I hit the home button
2) HTC Sense restarts, It doesn't even force close, or at least it's not telling me
3) a Hot restart occurs a few moments later... (no white boot screen, just starts bootin from ROM screen)
4) At this point, sometimes its still laggy, sometimes it's fine. If it's laggy I must do a battery pull because It will keep looping to this point.
I can rinse and repeat this running pandora + browsing the web, and the same 4 steps above will happen within moments of eachother. I've been using this ROM since January with no major problems like this
A few side notes:
-I have Evo hardware 0002 (launch day evo)
-Rooted with Toast's original root method, eng .76 boot loader
-Newest radio/PRI/NV/Wimax
-My phone does not randomly reboot on me normally. Reboots only occur when following the above steps.
-I do not use Set CPU
Here is where I am very superstitious. I've been closely monitoring the "Boot looping + no Recovery Access" forum, and similar forums to this. Within the past few months I have tried out a few different CM7 nightlies. I have never flashed RC2 but I have flashed some after and before and tried those out for a while. I just hope my phone isn't starting a dying phase in relation to the problems people are having now.
Is there anything else recommended before I end up doing a full wipe and reflash the rom?
Click to expand...
Click to collapse
I'm not really sure what's causing your problem, and If i was in your situation, I'd most likely do a complete wipe, including calkulin's format all, and either restore a good nandroid, or reflash the rom from scratch.
That said, I don't believe CM to be 'dangerous' to run or anything like that. Yes, it is odd that an increasing number of people are reporting somewhat major issues, with boot looping the splash screen, and being permanently stuck there. However, that doesn't mean CM is the cause. It does seem to be a common denominator in the people that this has happened to, but there are soooo many other variables that come into play with something like this. There is no proof that CM caused it, but it does seem odd that everyone it happened to was running CM. Also, I'm sure there are thousands of CM users, and a select 8-10 people have reported these issues on here. That's a very low percentage, overall. With all that said, I myself won't be running CM in the near future, either until it's confirmed or denied that it has anything to do with these issues. I'll just be playing it safe, just in case.
I'm stock everything, no root or mods and it restarts at the 4g screen skips everything else...
Sent from my PC36100 using XDA Premium App
I think I will do the format all and completetly wipe it and reflash the rom from scratch. Thank you!
Logcat #1 - http://pastebin.com/QwiBtnWg
Logcat #2 - http://pastebin.com/GnZe6S4s
Hiya,
DHD, running Cyanogenmod (running the 16th November 2011 nightly), with LordMod 8.5 CFS kernel. The nightly seems to me to be better than the stable release; the battery drain has been fixed and there don't seem to be any noticable bugs, so I don't think this is a CM7 nightly issue.
----
I recently downloaded "Traffic Panic Attack 3d" from the android market - and it's a great game There's over a million downloads so should all be good. However when you cause a large crash sometimes a car explodes, but this explosion seems to crash my phone - about 2/3 of the time a car explodes, the game freezes and eventually the phone completely locks up and dies.
The first time it happened I assume my phone had had a mishap or something and ignored it, but then I realised it was the same game causing at at this certain point. I attached it to ADB and tried to logcat it, but it seems to stop logcatting after a while so the second logcat is started after I lost responsiveness of the phone.
The screen freezes and the soft touch buttons still respond, but then they stop working and it completely freezes. If you quickly do "adb reboot" it'll reboot the phone but another time I just left the phone to see if it could recover itself, and then not even adb reboot would work, it just went to a new line, as if it were trying to connect, but didn't ever get through. I pulled the battery (it then said no status, so was clearly trying to reboot the phone) and turned it on, and all is fine again.
Any idea if this is an issue with the app, my phone or what? I can only assume the explosions are particularly GPU/CPU heavy and it causes my phone to fall over.
First logcat (http://pastebin.com/QwiBtnWg) stops after the first crash (phone rebooted itself so logcat stopped) the second (http://pastebin.com/GnZe6S4s) is started during a lockup I think, and I had to pull the battery to get it do anything.
Are you using ADW launcher by any chance?
Yes, I am - the free version that comes with CM7
Why?
No such game on Market :-?
https://market.android.com/details?id=com.neonplay.trafficpanic2
And just seen the reviews (I saw different ones on my phone that talk about any problems) - it seems I'm not alone with this issue, and it's not just the DHD either
Sorry for the delay in replying.
I asked about ADW Launcher because I was reading this as that was describing similar symptoms & error messages. This was caused by an OpenGL issue & ADW Launcher, as far as I could tell.
Have you downloaded Chainfire3D from the market & installed the OpenGL driver. If you have, try uninstalling the driver & try again. And if you haven't install the driver & try again.
Thanks for the info.
I installed the driver, but no luck. But then again I think it must be the app rather than anything else as so many others are also having problems.
I can't post in the Dev section so I'm posting here in the hopes that some one from the actual thread : http://forum.xda-developers.com/showthread.php?t=1714187 Will see this and respond and maybe even quote this in that thread.
I notice that my phone is very slow when it comes to preforming the basic function of well being a phone. When I try to initiate calls it takes it's sweet time. Sometimes even rebooting while attempting to connect a call.
I've also noticed some sluggishness with the Rom in general, usually after it's been running from a reboot or cold boot for a few hours. I've disabled the Virtuous OC Since it seemed to be cause issues and once I did that my phone's responsiveness went up considerably but I'm still noticing some of these slow down issues.
I'm considering giving the EliteMod a shot as most report supper quick responsiveness from that Rom.
A few things:
T-Mobile G2 / Desire-Z
I don't have Swap enabled,
I did a full wipe and format prior to installing the Rom
I use Apex as my Launcher
Ondemand Governor 368 --> 1209
I noticed that, when someone calls me, the phone starts ringing first with the screen still black, and only after a few seconds it shows the phone app with the option to answer, etc.
The phone app slowdown (when I need to call someone) has just started to happen some 7 days ago. It took ages to dial a number, but it went back to (almost) normal speed after reboot.
I did try to clear call logs, maybe that helped.
I'm using touch pal dialer since I installed mimicry - but the slowdown only appeared recently, with the reboot fixing the thing.
A bit of a nit pick, but it's mimicry, not MimiCry.
Mimicry is a real word yanno..
Anyway, your symptoms are likely the result of a bad app abusing memory. The phone app in general is slow on our devices, due to our hardware limitations, but hot reboots should not be happening.
Debug, figure out the bad app(s) and remove them.
Sent from my SAMSUNG-SGH-I747 using xda premium
I got a freezing problem with my note2. It's quite a strange behavior, got worse pretty fast since yesterday (tried to reflect the situation exactly below).
The freeze behaves like: system freezes, apps freeze, buttons stop working, sometimes the notification bar pull-down still works for 1-2 seconds, BUT: animated icons in the notification bar are sometimes (not always) still animating up to 2-3 minutes. Can also happen with display off, can't turn on anymore.
Since I'm using a ROM that's safe from the SDS problem, it cant be that. Still I mention that I do have that insane chip (Type: VTU00M, FwRev: 0xf1).
My guess is its a hardware related problem, maybe the internal flash memory(?).
Still, it will be hard to restore back to stock for warranty with the usb not working to use Odin.
Original Baseband: XXALJ2
History of this phone (maybe not 100% accurate, it's hard to recall everything exactly):
Bought on release
rooted some weeks later
flashed CM ~spring, regularly updated
July: stopped showing modem signal strength
July: switched to Asylum CM10.x, updated on new version releases
September: USB stopped working correctly when connecting to computer (charging but no data, different cables, ports, computers)
October: started freezing randomly, approx every second day, getting slighly more often
Since Yesterday:
-did a full wipe/factory reset, reflashed rom. more and faster freezes
-as i was messing around, decided to do a modem update: from XXALJ2 to XXDMF1: radio signal strength working again. Freezing increased.
-reverted to backup, also restored old modem: Crazy fast freezings, withing 30sek after every boot.
-again some full wipes/clean install with different but still recent asylum roms: freezings a little better than with a backup, but still crazy fast.
-full wipe with format internal storage: freezes got better, occur after minutes instead of seconds. They still occur faster when using a backup instead of a new install. Same with CM 10.1.3.
No custom kernels used. Default governors. No TB restores or something that I could think of messing with stuff.
Any help or ideas are very appreciated
I could manage to restore stock via Odin (1 in 100 times it detects the usb connection on my laptop), with 4.1.1 Had a freeze, made the OTR update to 4.1.2. Had another freeze. After that 2 more updates. It also changed the Baseband every time, its now DMF1 (same as i upgraded). No freezes anymore. I mean, what the heck? This makes no sense for me at all. If someone can lighten me up, please, go ahead.
It all sounds like SDS. Don't forget that SDS happens if there is a memory block corrupted. If you were in an insane rom before (and you said you bought it in release so almost sure you did) the memory blocks got corrupted. that happened "a lot" of people in the SDS thread(between commas taking in account that here in xda there is not the whole group of people who bought it and percentatge is low).
I recommend you to use emmc brickbug check app and check the memory blocks(if you are rooted)
then if it is SDS the only thing that we can do is pray or wait to samsung not to give free repairs only to China.
hoping that you'll pass the blocks test and it was only a normal problem..
Sent from my GT-N7100 using xda premium
Thanks for your reply.
SDS was the first thing I thought could be the issue. But it run for months without a single freeze on a SDS-safe rom before it started.
The last thing I have done before reverting to stock was a memory check with emmc brickbug app, and it passed. I forgot to mention it, even it's very interesting: while the memory check was running I had the longest runtime without a freeze.
I'm thinking now of returning the phone for warranty because of the USB problem. It's probably too risky to get back to stock once again. If I even could manage to root it again. But tell them too I had freezes? hmm, it's really not happening now anymore
I have the same problem... no idea what should I do
well... Yesterday I took a try and swtiched the CWM. I had TWRP and I've switch to Phliz Touch, so far... 16 hours without freeze
Amit BL said:
well... Yesterday I took a try and swtiched the CWM. I had TWRP and I've switch to Phliz Touch, so far... 16 hours without freeze
Click to expand...
Click to collapse
hmm, I also switched back to CWM recovery (from TWRP) for the last boot (to restore the nandroid stock backup I did, but could manage to flash with Odin before I had to do it that way), that was the point where I was surprised that the emmc memory check did go through without a freeze. I didn't saw any relation to the problem here, but who knows?
So far 34 hours without a freeze
Sent from my Galaxy Note II using Tapatalk
dedors said:
I got a freezing problem with my note2. It's quite a strange behavior, got worse pretty fast since yesterday (tried to reflect the situation exactly below).
The freeze behaves like: system freezes, apps freeze, buttons stop working, sometimes the notification bar pull-down still works for 1-2 seconds, BUT: animated icons in the notification bar are sometimes (not always) still animating up to 2-3 minutes. Can also happen with display off, can't turn on anymore.
Click to expand...
Click to collapse
Same problem, been suffering with freezes for the last 3 weeks and the last 2 days have been 400% worse, found out about this emmc bug but cannot find any information as to how to fix it, some mention kernals, vodafone will not replace the phone unless their software update does not fix it, Running 3.99 RC2 Paranoid android. Anybody got any ideas?
Seems my phone is one of the effect chips with the emmc issue. Funny thing is i just flashed back to stock in order to take my phone back to the shop and its working fine.... not a single issue all day.
Is there something custom roms are doing that party causes this issue?
Sent from my GT-N7100 using xda app-developers app
Madnessx9 said:
Seems my phone is one of the effect chips with the emmc issue. Funny thing is i just flashed back to stock in order to take my phone back to the shop and its working fine.... not a single issue all day.
Is there something custom roms are doing that party causes this issue?
Sent from my GT-N7100 using xda app-developers app
Click to expand...
Click to collapse
I was having different issues including freezes and WiFi/BT loss when kernels with undervolt installed. Usually from custom ROMs. The effect is prolonged sometimes till hardware reset done.
The only working solution (for me) is to use a kernel with controlled voltages and set all the voltages about 1000mV for frequencies below 800MHz. 100% stable,
First off I'd like to say that I've already got the power switch replaced on this device. So I've ruled that out to some extent.
That being said, my device pretty much will freeze up at random times then either turn off or reboot. If it reboots, it will load the OS (or sometimes not even finish the boot), then either freeze up then reboot again, turn off, or freeze up and stay frozen for eternity. I've tried reverting to stock then reflashing all my **** and redownloading all my apps and stuff. While that helped for maybe, 3 days, it did not fix the issue. Also notable that this usually happens several times a day, and that sometimes leaving it on the charger helps it make a stable boot. I've also purchased a new battery with no luck. Not sure if this is related but figured it might be relevant that the device wakes randomly sometimes. This is not related to my specific ROM either, this happens on several pieces of firmware. Only cause I can think of is age (this model is almost 4 years old) or a faulty power switch (I've already had the switch replaced).
This has started to affect my schedule in that my phone loves to turn off when I'm sleeping so I miss my phone's alarm, which I highly depend on.
Completely unrelated issue but I'm on Sultan's Lollipop ROM and I can't download themes, apps, etc. The device gives me an error saying that the app/theme "failed to install". All of my current apps come from restoring /data from my 4.4.4 daily driver backup. Help with this would be very nice as well.
Thanks a ton in advance for the help.
Fixed! My overclock was too high. I can only assume that it was overheating, causing the freezing and rebooting. Min frequency is at 192 MHz and max is 1512. Stable for the time being.
As far as your reboots go, if it happens after flashing stock 4.1.2 with Odin, then it is hardware. Either the button still, or something else disrupting power momentarily.
Not sure about issue 2 though.
?
LiquidPlatinum said:
First off I'd like to say that I've already got the power switch replaced on this device. So I've ruled that out to some extent.
That being said, my device pretty much will freeze up at random times then either turn off or reboot. If it reboots, it will load the OS (or sometimes not even finish the boot), then either freeze up then reboot again, turn off, or freeze up and stay frozen for eternity. I've tried reverting to stock then reflashing all my **** and redownloading all my apps and stuff. While that helped for maybe, 3 days, it did not fix the issue. Also notable that this usually happens several times a day, and that sometimes leaving it on the charger helps it make a stable boot. I've also purchased a new battery with no luck. Not sure if this is related but figured it might be relevant that the device wakes randomly sometimes. This is not related to my specific ROM either, this happens on several pieces of firmware. Only cause I can think of is age (this model is almost 4 years old) or a faulty power switch (I've already had the switch replaced).
This has started to affect my schedule in that my phone loves to turn off when I'm sleeping so I miss my phone's alarm, which I highly depend on.
Completely unrelated issue but I'm on Sultan's Lollipop ROM and I can't download themes, apps, etc. The device gives me an error saying that the app/theme "failed to install". All of my current apps come from restoring /data from my 4.4.4 daily driver backup. Help with this would be very nice as well.
Thanks a ton in advance for the help.
Click to expand...
Click to collapse
Same happened with me... Since I've problem with soft navigation keys.... Refer here for more... http://forum.xda-developers.com/showthread.php?t=2999362....I think it's the same in ur case too
pawanzZ said:
Same happened with me... Since I've problem with soft navigation keys.... Refer here for more... http://forum.xda-developers.com/showthread.php?t=2999362....I think it's the same in ur case too
Click to expand...
Click to collapse
i don't think that's the same issue. even so the thread you linked me to doesn't help anything..
LiquidPlatinum said:
i don't think that's the same issue. even so the thread you linked me to doesn't help anything..
Click to expand...
Click to collapse
Hmm...hope u'll get fix soon..