State of AOSP on Galaxy Nexus

[Jean-Baptiste Queru]

-AOSP Jelly Bean can be used on GSM Galaxy Nexus that were sold with a
yakju build (some devices in Europe) or takju build (Google Play).

-AOSP Jelly Bean can be used on VZW Galaxy Nexus. Those were all sold
with a mysid build.

-Jelly Bean can be used on those even if they're running the radio
firmware that came with 4.0.4, i.e. from the latest available factory
images. Of course, the 4.1.1 radio firmware also works.

-The binaries for android-4.1.1_r1 are JRO03C. They can also be used
on jb-dev and master. jb-dev should always use the latest Jelly Bean
binaries, master should always use the latest binaries.

-Graphics, cell radio, WiFi, Bluetooth, sensors all work fine.
Unchanged since Ice Cream Sandwich, the camera and GPS don't work. NFC
might not perform well for operations beyond reading tags.

JBQ

--
Jean-Baptiste M. "JBQ" Queru
Technical Lead, Android Open Source Project, Google.

Questions sent directly to me that have no reason for being private
will likely get ignored or forwarded to a public forum with no further
warning.

[Archon810]

Wait a second, so what are the differences that prompted you to say that mysid has issues compared to yakju/takju? What are those issues?

[Jean-Baptiste Queru]

There's a power management issue that only occurs on AOSP builds but
not on yakju/takju/mysid. From the logs, I assume that one of the
missing proprietary binaries causes a chip to trigger spurious
interrupts, which prevents the CPU from going to deep sleep, and that
shortens the battery life. In my tests, this seems more severe on toro
than on maguro. It's being investigated by kernel engineers.

JBQ

On Tue, Jul 10, 2012 at 8:15 PM, Archon810 <arch...@gmail.com> wrote:
> Wait a second, so what are the differences that prompted you to say that
> mysid has issues compared to yakju/takju? What are those issues?
>

> --
> You received this message because you are subscribed to the "Android
> Building" mailing list.
> To post to this group, send email to android-...@googlegroups.com
> To unsubscribe from this group, send email to
> android-buildi...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/android-building?hl=en

[jdkoreclipse]

JBQ:

The power management issue is caused by the Props for Google Now (everything in usr/srec/en-US for Google Now compat)

-jdkoreclipse

[Jean-Baptiste Queru]

Oh, interesting. Do you have any specific details? I'm wondering why
that'd only affect Galaxy Nexus.

JBQ

[Jean-Baptiste Queru]

I'll make those files immediately available if/when I get an
appropriate license for them. At the technical level, everything
works, I just can't distribute the files.

JBQ

On Wed, Jul 11, 2012 at 8:31 AM, iceandfire <arham...@gmail.com> wrote:
>> Fix for the camera and gps anytime soon?

>
> --
> You received this message because you are subscribed to the "Android
> Building" mailing list.
> To post to this group, send email to android-...@googlegroups.com
> To unsubscribe from this group, send email to
> android-buildi...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/android-building?hl=en

[bcbuc...@gmail.com]

What permissions should be set for including the camera and GPS files
with Jelly Bean, I included them in build but they are still failing
after I boot into the new images.

On Wed, Jul 11, 2012 at 11:36 AM, iceandfire <arham...@gmail.com> wrote:
> Thanks Shane. I'm currently pulling vendor stuff from the official build. If
> that doesn't work i'll do what you wrote.
>
> PS: Yeah I'm iceandfire :)
>
>
> On Wednesday, July 11, 2012 9:43:49 PM UTC+5:30, jdkoreclipse wrote:
>>
>> JBQ:
>> This is actually an odd issue, because not having props SHOULDNT break the
>> build as badly as it has (IE, the lack google now files causing deep sleep
>> issues). However, it makes sense that the deep sleep issue is just in AOSP
>> and not in retail builds, because as you know, Google Apps and Proprietaries
>> arent in AOSP, but they do make their way into retail builds.
>>
>> Also, for the person who asked about camera and gps, just take these files
>> out of ICS and put them in your build (i recognize your name from the
>> forums):
>> - /vendor/etc/sirfgps.conf
>> - /vendor/lib/hw/gps.omap4.so
>> - /vendor/firmware/ducati-m3..bin

[Jean-Baptiste Queru]

The makefiles that would be used in the packages that'd get
distributed on https://developers.google.com/android/nexus/drivers are
Open Source and are available in
device/samsung/maguro/self-extractors. Look specifically for the ones
under csr/staging and ti/staging.

JBQ

[Jean-Baptiste Queru]

Those are the files I'd use if I could distribute the matching
binaries. I didn't compare the exact file lists one by one.

The whole list of hardware-related proprietary file is in
device/samsung/maguro/proprietary-blobs.txt

JBQ

On Thu, Jul 12, 2012 at 1:12 PM, iceandfire <arham...@gmail.com> wrote:
> So are the driver files updated to include those files stated by
> jdkoreclipse?

> --
> You received this message because you are subscribed to the "Android
> Building" mailing list.
> To post to this group, send email to android-...@googlegroups.com
> To unsubscribe from this group, send email to
> android-buildi...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/android-building?hl=en

[Chandra Gopalaiah]

Hi JBQ,

I just downloaded the binaries and add to my local source directory.

I tried to make a build, but unfortunately it is failing with the below error

Traceback (most recent call last):

  File "out/host/linux-x86/obj/EXECUTABLES/mksnapshot_intermediates/js2c.py", line 36, in <module>

    import bz2

ImportError: No module named bz2

Traceback (most recent call last):

  File "out/host/linux-x86/obj/EXECUTABLES/mksnapshot_intermediates/js2c.py", line 36, in <module>

    import bz2

ImportError: No module named bz2

Install: out/host/linux-x86/framework/commons-codec-1.4.jar

make: *** [out/host/linux-x86/obj/EXECUTABLES/mksnapshot_intermediates/libraries.cc] Error 1

make: *** Waiting for unfinished jobs....

make: *** [out/host/linux-x86/obj/EXECUTABLES/mksnapshot_intermediates/experimental-libraries.cc] Error 1

Can you help me on resolving this issue?

Highly appreciate your help.

Thanks,

Chandra

[Jeremy Schulte]

JBQ,

I think some of the AOSP problems we are seeing is due to a misplaced proprietary.  Maguro and Tor contain fRom from Samsung.

The included Android.mk file lists them to be installed to system/vendor/bin.  However, retail release has it in /system/bin. Proprietary-blobs.txt also lists system/bin as the location.

Installed in this location, the camera switches very slowly (front to back) as well as breaks the video camera with a nasty error.  Also, /data/misc/camera/R5_MVEN003_LD2_ND0_IR0_SH0_FL1_SVEN003_DCCID1044/calib.bin never gets created.  So to fix my problems, I changed Android.mk in the proprietary directory.

Before

include $(CLEAR_VARS)

LOCAL_MODULE := fRom

LOCAL_MODULE_OWNER := samsung

LOCAL_SRC_FILES := fRom

LOCAL_MODULE_TAGS := optional

LOCAL_MODULE_CLASS := EXECUTABLES

LOCAL_MODULE_PATH := $(TARGET_OUT_VENDOR)/bin

include $(BUILD_PREBUILT)

After

include $(CLEAR_VARS)

LOCAL_MODULE := fRom

LOCAL_MODULE_OWNER := samsung

LOCAL_SRC_FILES := fRom

LOCAL_MODULE_TAGS := optional

LOCAL_MODULE_CLASS := EXECUTABLES

LOCAL_MODULE_PATH := $(TARGET_OUT)/bin

include $(BUILD_PREBUILT)

This fixed my AOSP woes.

On Thu, Jul 12, 2012 at 1:19 PM, Jean-Baptiste Queru <j...@android.com> wrote:

[jdkoreclipse]

JBQ:

I've compared the lists one by one, and those seem to be the only missing ones that really matter. 

Iceandfire: Those files are in the maguro OTA. Just pull them from there, stick them in a vendor setup, and you know the rest.

-jdkoreclipse

[Jean-Baptiste Queru]

Are you sure that's related to the proprietary binaries? From here, it
seems to me that your version of python is missing a library and fails
at a point that has nothing to do with proprietary binaries or even
Galaxy Nexus (this is the part of the build system that compiles
WebKit). Are you using the version of python that's part of Ubuntu
LTS?

JBQ

[Jean-Baptiste Queru]

Ohhhh, very nice catch. That's definitely my bug. I'll get this fixed
at the first opportunity.

JBQ

[Chris Stratton]

On Jul 12, 4:19 pm, Jean-Baptiste Queru <j...@android.com> wrote:

> The whole list of hardware-related proprietary file is in
> device/samsung/maguro/proprietary-blobs.txt

Given that these don't seem to include any (odex'd) Dalvik classes,
would your earlier idea of using a recovery to flash an AOSP build
file-by-file "on top of" a restore image be workable in this case?

Obviously there are dependencies on non-public system libraries, but
unless those were changed from the versions they were built against it
would seem things should work?

[Jean-Baptiste Queru]

Yes, the approach of creating a custom recovery package that would
leave those files in place should still work.

JBQ

[Jean-Baptiste Queru]

-Did you do a flashall -w? (if not, you might have some leftovers from
a previous build).

-Did you wait a few minutes? (on a clean flash, the system then spends
a few minutes at first boot to create a new filesystem, and then
reboots).

JBQ

On Fri, Jul 13, 2012 at 3:05 PM, Erik Rainey <erik....@gmail.com> wrote:
> JBQ,
> I'm building off "android-4.1.1_r1" branch for full_maguro for a
> GT-I9250M with "otapackage". After the "flashall", it gets trapped in the
> Android boot logo with this constantly cycling in the logcat.:
>
> ... lots of unknown permissions.... then
> W/PackageManager( 6573): Unknown permission android.permission.BLUETOOTH in
> package com.android.musicfx
> D/dalvikvm( 6573): GC_CONCURRENT freed 473K, 24% free 11826K/15367K, paused
> 11ms+2ms, total 30ms
> D/dalvikvm( 6573): WAIT_FOR_CONCURRENT_GC blocked 0ms
> D/dalvikvm( 6573): GC_EXPLICIT freed 99K, 24% free 11749K/15367K, paused
> 2ms+4ms, total 33ms
> I/SystemServer( 6573): Account Manager
> D/PackageManager( 6573):
> generateServicesMap(android.accounts.AccountAuthenticator): 2 services
> unchanged
> I/SystemServer( 6573): Content Manager
> D/PackageManager( 6573): generateServicesMap(android.content.SyncAdapter): 8
> services unchanged
> I/SystemServer( 6573): System Content Providers
> E/ActivityThread( 6573): Failed to find provider info for settings
> E/System ( 6573): ******************************************
> E/System ( 6573): ************ Failure starting core service
> E/System ( 6573): java.lang.NullPointerException
> E/System ( 6573): at
> android.provider.Settings$NameValueCache.getString(Settings.java:737)
> E/System ( 6573): at
> android.provider.Settings$Secure.getString(Settings.java:2343)
> E/System ( 6573): at
> android.provider.Settings$Secure.getInt(Settings.java:2411)
> E/System ( 6573): at
> com.android.server.am.CoreSettingsObserver.populateCoreSettings(CoreSettingsObserver.java:93)
> E/System ( 6573): at
> com.android.server.am.CoreSettingsObserver.sendCoreSettings(CoreSettingsObserver.java:70)
> E/System ( 6573): at
> com.android.server.am.CoreSettingsObserver.<init>(CoreSettingsObserver.java:55)
> E/System ( 6573): at
> com.android.server.am.ActivityManagerService.installSystemProviders(ActivityManagerService.java:6709)
> E/System ( 6573): at
> com.android.server.ServerThread.run(SystemServer.java:204)
> I/SystemServer( 6573): Input Method Service
> W/InputMethodManagerService( 6573): Couldn't create dir.:
> /data/system/inputmethod
> E/ActivityThread( 6573): Failed to find provider info for settings
> F/SystemServer( 6573): BOOT FAILURE starting Input Manager Service
> F/SystemServer( 6573): java.lang.NullPointerException
> F/SystemServer( 6573): at
> android.provider.Settings$NameValueCache.getString(Settings.java:737)
> F/SystemServer( 6573): at
> android.provider.Settings$Secure.getString(Settings.java:2343)
> F/SystemServer( 6573): at
> com.android.server.InputMethodManagerService.<init>(InputMethodManagerService.java:618)
> F/SystemServer( 6573): at
> com.android.server.ServerThread.run(SystemServer.java:287)
> F/SystemServer( 6573): BOOT FAILURE making display ready
> F/SystemServer( 6573): java.lang.NullPointerException
> F/SystemServer( 6573): at
> com.android.server.ServerThread.run(SystemServer.java:303)
> W/SystemServer( 6573): ***********************************************
> I/SystemServer( 6573): Accessibility Manager
> W/SystemServer( 6573): ***********************************************
> W/SystemServer( 6573): ***********************************************
> F/SystemServer( 6573): BOOT FAILURE performing boot dexopt
> F/SystemServer( 6573): java.lang.NullPointerException
> F/SystemServer( 6573): at
> com.android.server.am.ActivityManagerService.showBootMessage(ActivityManagerService.java:4187)
> F/SystemServer( 6573): at
> com.android.server.pm.PackageManagerService.performBootDexOpt(PackageManagerService.java:3328)
> F/SystemServer( 6573): at
> com.android.server.ServerThread.run(SystemServer.java:309)
> W/dalvikvm( 6573): threadid=12: thread exiting with uncaught exception
> (group=0x40ce0300)
> I/Process ( 6573): Sending signal. PID: 6573 SIG: 9
> E/AndroidRuntime( 6573): *** FATAL EXCEPTION IN SYSTEM PROCESS:
> android.server.ServerThread
> E/AndroidRuntime( 6573): java.lang.NullPointerException
> E/AndroidRuntime( 6573): at
> com.android.server.am.ActivityManagerService.showBootMessage(ActivityManagerService.java:4187)
> E/AndroidRuntime( 6573): at
> com.android.server.ServerThread.run(SystemServer.java:315)
> I/ServiceManager( 116): service 'accessibility' died
> I/ServiceManager( 116): service 'content' died
> I/ServiceManager( 116): service 'account' died
> I/ServiceManager( 116): service 'telephony.registry' died
> I/ServiceManager( 116): service 'batteryinfo' died
> I/ServiceManager( 116): service 'gfxinfo' died
> I/ServiceManager( 116): service 'scheduling_policy' died
> I/ServiceManager( 116): service 'usagestats' died
> I/ServiceManager( 116): service 'dbinfo' died
> I/ServiceManager( 116): service 'sensorservice' died
> I/ServiceManager( 116): service 'cpuinfo' died
> I/ServiceManager( 116): service 'package' died
> I/ServiceManager( 116): service 'activity' died
> I/ServiceManager( 116): service 'meminfo' died
> I/ServiceManager( 116): service 'entropy' died
> I/ServiceManager( 116): service 'power' died
> I/ServiceManager( 116): service 'permission' died
> W/AudioFlinger( 6445): power manager service died !!!
> W/AudioFlinger( 6445): power manager service died !!!
> E/installd( 129): eof
> E/installd( 129): failed to read size
> I/installd( 129): closing connection
> I/Zygote ( 6447): Exit zygote because system server (6573) has terminated
> I/ServiceManager( 116): service 'media.audio_flinger' died
> I/ServiceManager( 116): service 'media.player' died
> I/ServiceManager( 116): service 'media.camera' died
> I/ServiceManager( 116): service 'media.audio_policy' died
>
> Is this the permissions issue you were mentioning above which causes
> crashes? /data/system/inputmethod is present, so I think that's a herring.

[Jean-Baptiste Queru]

I'll distribute the GPS libraries for Galaxy Nexus if/when their owner
allows Google to do so.

JBQ

On Mon, Jul 16, 2012 at 1:56 AM, Nexus Man <cm6...@gmail.com> wrote:
> Hi Jean,
>
> Will you update GPS driver soon ? Basically, with ICS & current JB , gps do
> not work fine on my GNex GSM . (awful accuracy)
>
> Modifying gps.conf / ntp servers ans other settings do not resolve this
> issue (known issue)
>
> I ve also a Nexus S GSM which is perfect when driving.
>
> Regards

> --
> You received this message because you are subscribed to the "Android
> Building" mailing list.
> To post to this group, send email to android-...@googlegroups.com
> To unsubscribe from this group, send email to
> android-buildi...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/android-building?hl=en

[Jean-Baptiste Queru]

If you're doing a mix-and-match of one version of the source code with
binaries from another version, all bets are off, so nothing would
surprise me, including pauses at boot.

JBQ

On Sun, Jul 15, 2012 at 4:47 PM, yarly <lanni...@gmail.com> wrote:
> Is it possible that the binary incompatibilities for the Galaxy Nexus Toro
> could be related to the random pauses users experience randomly on booting
> up the device? Just for reference, maybe 1 out of 10 times, I will have it
> pause for maybe a minute (give or take) at the GOOGLE boot screen before
> going into the boot animation.

>
> --
> You received this message because you are subscribed to the "Android
> Building" mailing list.
> To post to this group, send email to android-...@googlegroups.com
> To unsubscribe from this group, send email to
> android-buildi...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/android-building?hl=en

[Jean-Baptiste Queru]

Unfortunately, they have not.

JBQ

On Mon, Dec 17, 2012 at 12:22 AM, Roger Chen <cxr514...@gmail.com> wrote:
> Haven't the owner allow to publish the lib yet?
>
> 在 2012年7月11日星期三UTC+8上午10时37分43秒，Jean-Baptiste Queru写道：

> --
> You received this message because you are subscribed to the "Android
> Building" mailing list.
> To post to this group, send email to android-...@googlegroups.com
> To unsubscribe from this group, send email to
> android-buildi...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/android-building?hl=en

[Jean-Baptiste Queru]

I might be seeing the same crash. Does this look like it?

D/ConnectivityService( 8638): ConnectivityChange for mobile: CONNECTED/CONNECTED
I/NetworkPolicy( 8638): no policy for active mobile network;
generating default policy
W/dalvikvm( 8638): threadid=36: thread exiting with uncaught exception
(group=0x41c4b990)
E/AndroidRuntime( 8638): *** FATAL EXCEPTION IN SYSTEM PROCESS: NetworkPolicy
E/AndroidRuntime( 8638): java.lang.RuntimeException: Error receiving
broadcast Intent { act=android.net.conn.CONNECTIVITY_CHANGE_IMMEDIATE
flg=0x8000010 (has extras) } in
com.android.server.net.NetworkPolicyManagerService$12@4257ae98
E/AndroidRuntime( 8638): at
android.app.LoadedApk$ReceiverDispatcher$Args.run(LoadedApk.java:768)
E/AndroidRuntime( 8638): at android.os.Handler.handleCallback(Handler.java:725)
E/AndroidRuntime( 8638): at android.os.Handler.dispatchMessage(Handler.java:92)
E/AndroidRuntime( 8638): at android.os.Looper.loop(Looper.java:137)
E/AndroidRuntime( 8638): at android.os.HandlerThread.run(HandlerThread.java:60)
E/AndroidRuntime( 8638): Caused by:
java.lang.ArrayIndexOutOfBoundsException: length=12; index=-1
E/AndroidRuntime( 8638): at
android.text.format.Time.getActualMaximum(Time.java:239)
E/AndroidRuntime( 8638): at
android.net.NetworkPolicyManager.snapToCycleDay(NetworkPolicyManager.java:222)
E/AndroidRuntime( 8638): at
android.net.NetworkPolicyManager.computeLastCycleBoundary(NetworkPolicyManager.java:179)
E/AndroidRuntime( 8638): at
com.android.server.net.NetworkPolicyManagerService.updateNetworkRulesLocked(NetworkPolicyManagerService.java:1029)
E/AndroidRuntime( 8638): at
com.android.server.net.NetworkPolicyManagerService.addNetworkPolicyLocked(NetworkPolicyManagerService.java:1459)
E/AndroidRuntime( 8638): at
com.android.server.net.NetworkPolicyManagerService.ensureActiveMobilePolicyLocked(NetworkPolicyManagerService.java:1137)
E/AndroidRuntime( 8638): at
com.android.server.net.NetworkPolicyManagerService.access$1500(NetworkPolicyManagerService.java:170)
E/AndroidRuntime( 8638): at
com.android.server.net.NetworkPolicyManagerService$12.onReceive(NetworkPolicyManagerService.java:899)
E/AndroidRuntime( 8638): at
android.app.LoadedApk$ReceiverDispatcher$Args.run(LoadedApk.java:758)
E/AndroidRuntime( 8638): ... 4 more

JBQ

On Sun, Dec 23, 2012 at 12:12 PM, Andy Cooper
<andrew.jo...@gmail.com> wrote:
> I just built AOSP for my Verizon Galaxy Nexus (I downloaded and installed
> the VZW Galaxy Nexus binaries as per the instructions on the build page).
> The build went fine. I flashed boot.img, recovery.img, system.img and
> userdata.img onto my phone (I created a zipfile and then used fastboot -w
> update image.zip).
>
> The phone boots fine into JB 4.2.1, but as soon as I enable the LTE/CDMA
> radios, it enters an endless reboot cycle. As soon as it seems like it gets
> 4G connectivity, I get the "android" boot screen and things cycle through
> there endlessly.
>
> I'm assuming that the current master branch works for the VZW Galaxy Nexus.
> Is that correct? Assuming that it is correct, can anyone give me pointers on
> what I've done wrong that is causing this endless boot cycle?

[Jean-Baptiste Queru]

Thanks. I'll try to dig deeper.

As a reminder: it's a good idea to save a full bugreport each time you
see some kind of crash, for later analysis.

JBQ

On Thu, Dec 27, 2012 at 7:08 AM, Andy Cooper
<andrew.jo...@gmail.com> wrote:
> OK, I just installed my build and tested it out. Yes, it's exactly the same
> crash. Here's my logcat output:
>
> W/InputMethodManagerService( 1513): Window already focused, ignoring focus
> gain
> of: com.android.internal.view.IInputMethodClient$Stub$Proxy@422e9370
> attribute=n
> ull, token = android.os.BinderProxy@41de5540
> I/NetworkPolicy( 1513): no policy for active mobile network; generating
> default policy
> W/dalvikvm( 1513): threadid=36: thread exiting with uncaught exception
> (group=0x40c38990)
> E/AndroidRuntime( 1513): *** FATAL EXCEPTION IN SYSTEM PROCESS:
> NetworkPolicy
> E/AndroidRuntime( 1513): java.lang.RuntimeException: Error receiving

> broadcast Intent { act=android.net.conn.CONNECTIVITY_CHANGE_IMMEDIATE
> flg=0x8000010 (has ex

> tras) } in com.android.server.net.NetworkPolicyManagerService$12@415b1eb8
> E/AndroidRuntime( 1513): at
> android.app.LoadedApk$ReceiverDispatcher$Args.run(LoadedApk.java:768)
> E/AndroidRuntime( 1513): at
> android.os.Handler.handleCallback(Handler.java:725)
> E/AndroidRuntime( 1513): at
> android.os.Handler.dispatchMessage(Handler.java:92)
> E/AndroidRuntime( 1513): at android.os.Looper.loop(Looper.java:137)
> E/AndroidRuntime( 1513): at
> android.os.HandlerThread.run(HandlerThread.java:60)
> E/AndroidRuntime( 1513): Caused by:
> java.lang.ArrayIndexOutOfBoundsException: length=12; index=-1
> E/AndroidRuntime( 1513): at
> android.text.format.Time.getActualMaximum(Time.java:239)
> E/AndroidRuntime( 1513): at
> android.net.NetworkPolicyManager.snapToCycleDay(NetworkPolicyManager.java:222)
> E/AndroidRuntime( 1513): at
> android.net.NetworkPolicyManager.computeLastCycleBoundary(NetworkPolicyManager.java:179)
> E/AndroidRuntime( 1513): at
> com.android.server.net.NetworkPolicyManagerService.updateNetworkRulesLocked(NetworkPolicyManagerService.java:1029)
> E/AndroidRuntime( 1513): at
> com.android.server.net.NetworkPolicyManagerService.addNetworkPolicyLocked(NetworkPolicyManagerService.java:1459)
> E/AndroidRuntime( 1513): at
> com.android.server.net.NetworkPolicyManagerService.ensureActiveMobilePolicyLocked(NetworkPolicyManagerService.java:1137)
> E/AndroidRuntime( 1513): at
> com.android.server.net.NetworkPolicyManagerService.access$1500(NetworkPolicyManagerService.java:170)
> E/AndroidRuntime( 1513): at
> com.android.server.net.NetworkPolicyManagerService$12.onReceive(NetworkPolicyManagerService.java:899)
> E/AndroidRuntime( 1513): at
> android.app.LoadedApk$ReceiverDispatcher$Args.run(LoadedApk.java:758)
> E/AndroidRuntime( 1513): ... 4 more
> I/Process ( 1513): Sending signal. PID: 1513 SIG: 9
>
> It looks like the data usage settings in "Settings > Data Usage" are causing
> the problem. It appears that if I turn off the data usage, I can get the
> system to come up and stay up. When I look at Data Usage periods, I get "Jan
> 1 - -1" as the period. That seems to correspond to the IndexOutOfBounds
> exception in the stack trace above.
>
>
>
>
>
>
>
> On Wednesday, December 26, 2012 12:45:32 PM UTC-5, Jean-Baptiste Queru

[Jean-Baptiste Queru]

My tests suggest that the crash only happens in the master branch
(bleeding edge, etc...), and the tagged 4.2.1 release doesn't have
that issue.

I've got to assume that a contribution made after 4.2.x was branched
introduced a bug.

JBQ

[Jean-Baptiste Queru]

I've tested on maguro as well (the GSM Galaxy Nexus), and the exact
same crash happens in the AOSP master, as soon as the data connection
comes up. That means that the crash isn't specific to the CDMA/LTE
code (it didn't look that way from the stack trace anyway).

I can't test on other devices (toro and maguro are the only devices
that get a cell connection in the AOSP master), but it's likely that
the issue affects all devices.

JBQ

[FZE]

I pull out the files after factory reset running in Galaxy nexus and pushed it to AOSP 4.1.2 in Galaxy nexus to the mentioned location. Still it fails to load GPS. Any suggestion?

On Wednesday, July 11, 2012 10:13:49 AM UTC-6, jdkoreclipse wrote:

JBQ:

This is actually an odd issue, because not having props SHOULDNT break the build as badly as it has (IE, the lack google now files causing deep sleep issues). However, it makes sense that the  deep sleep issue is just in AOSP and not in retail builds, because as you know, Google Apps and Proprietaries arent in AOSP, but they do make their way into retail builds.

Also, for the person who asked about camera and gps, just take these files out of ICS and put them in your build (i recognize your name from the forums):

- /vendor/etc/sirfgps.conf

- /vendor/lib/hw/gps.omap4.so

- /vendor/firmware/ducati-m3..bin

On Wednesday, July 11, 2012 11:54:23 AM UTC-4, Jean-Baptiste Queru wrote: