• State Resolved
  • Locked Locked
  • Replies 26 replies
  • Subscribers 103 subscribers
  • Views 18733 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

4AI.1.4 ICS (BLAZE) CTS test - android.mediastress results in Out Of Memory

Keerthi Karanth
Prodigy 230 points

Hi All,

We have a custom platform based on omap4460 , the software on it is ported from 4AI.1.4 , ICS 4.0.4 from TI.

In CTS if 'run cts -p android.mediastress' is given exactly after 54 test cases , 55th would fail with Out Of Memory error.

The attached are the logs

dmesg - 7536.dmesg.txt

logcat - 4454.logcat.txt

If I run the failing test alone i.e. "run cts -c android.mediastress.cts.H264R720pAacLongPlayerTest -m testPlay00" it passes without issue.

If android.mediastress is run as package it always fails

Please note : I have verified this on http://omappedia.org/wiki/Android:_Working_with_pre-built_binaries 4AI.1.4 pre-built for blaze aswell. I see the same "out of memory " issue there too.


Couple of things noticed 

  1. Verified the CTS mediastress code to see if MediaPlayer is released after use. This is done after each clip runs to completion.
  2. pvrsvrinit growing size

cat /d/ion/nonsecure_tiler
client pid size
pvrsrvinit 115 127107072

In the above pvrsvrinit size keeps growing. Is this causing the issue? How to debug this further.

Any pointers to debug/solve the issue would be of great help.

Regards,

Keerthi

  • 0
    Prodigy 235 points

    Hi all,

    I also meet the same problem.

    Have any one know this issue what's happening?

    Best Regards,

    Mike

  • 0 in reply to Mike Hsu
    TI__Genius 10545 points

    I found next 2 patches for similar issue, could you try these patches?

    http://review.omapzoom.org/#/c/21581/3

    http://review.omapzoom.org/#/c/27081/

  • 0 in reply to Manuel Contreras
    Prodigy 235 points

    Hi,

    Unfortunately, I tried these 2 patches but still happen out of memory issue.

    So far, we can't find root cause and keep to investigate.

    If you have any updates or comments, please give your suggestions to help us.

    Thank you!

    Best Regards,

    Mike.

  • 0 in reply to Mike Hsu
    TI__Genius 10545 points

    This is information I found,

    1. for MediaStressTest 2 of the reason why it could fail the common ones are:

    a) it uses some network test patterns that are not distributed in regular release.

    b) sometimes the customized filesystem is not using some feature that is tested and that causes test to fail.

    Maybe it should not be any of the reasons, but I think I needed to mention it.

    2. I found next patches to solve some issues with surfaces and out of memory error.

    http://review.omapzoom.org/#/c/22802/1

    that redirects to proper change https://android-review.googlesource.com/#/c/35040/

    http://review.omapzoom.org/#/c/22803/1

    3. I checked mediastresstests and they are for playback not record, but if it helps you next are some patches for Camera, I haven't check if they are in 4AI1.4 or not.

    http://review.omapzoom.org/#/c/21492/

    http://review.omapzoom.org/#/c/21661/

    http://review.omapzoom.org/#/c/22390/

    http://review.omapzoom.org/#/c/22028/

    http://review.omapzoom.org/#/c/21491/

    http://review.omapzoom.org/#/c/25461/

    4. you may need to check if there are updated for external/webkit about memory leaks, just in case they have any, I found 1 of the 2 that I mentioned before doing a search here, the other one was I found it later and it was making reference to both of them.

    5. for the prints about timeinterpolation, there was a case when the timestamp started not in zero time, this is the patch

    http://review.omapzoom.org/#/c/27350/

    6. I can think in checking full log it could give a clue of what is happening in the background or during the playback, if possible,can you share a full log to check it? if you can do logcat -f /mnt/sdcard/log.txt -n 10000 -r 10 to send it to a file and don't get any line missed. some things to check is if there are some errors for some files that are not found or if some file is giving some error.

    7. that reminds me, when using logcat and sending it to a file and if you have full cpu load, logcat code will not have enough time to store the logs in the file and it will consume memory until it have time to write it, it is in logcat code in /4AI.1.4/mydroid/system/core/logcat, like you noted in your first post it is PVR that is consuming all memory when this case happens it is a second logcat instance that has all the memory assigned to it, when there are 20 MB free the system starts shutting down applications then it fails later.

    8. if you can capture /proc/meminfo like every second just to check what are the memory movements, there is an application that could help /mydroid/packages/experimental/procstatlog, it is needed to add LOCAL_MODULE_TAGS := optional to Android.mk to compile it in full_blaze-userdebug configuration.

  • 0 in reply to Manuel Contreras
    TI__Genius 10545 points

    Comment,

    I asked if you can share the log but only if it doesn't contains some private information or special logs from internal changes, or something similar that should not be shared.

  • 0 in reply to Manuel Contreras
    Prodigy 235 points

    Hi,

    Thank you for your reply.

    I tried these two patches which list below and out of memory issue is solved:

    http://review.omapzoom.org/#/c/22802/1
    http://review.omapzoom.org/#/c/22803/1

    But meet another fail problems list below:

    04-23 16:32:54 I/014994DA09006019: android.mediastress.cts.MediaRecorderStressTest#testStressCamera FAIL
    Test failed to run to completion. Reason: 'Failed to receive adb shell test output within 600000 ms. Test may have timed out, or adb connection to device became unresponsive'. Check device logcat for details
    04-23 16:33:10 I/014994DA09006019: android.mediastress.cts.MediaRecorderStressTest#testStressCameraSwitchRecorder FAIL
    Test failed to run to completion. Reason: 'Instrumentation run failed due to 'java.lang.RuntimeException''. Check device logcat for details
    04-23 16:33:12 I/014994DA09006019: android.mediastress.cts.MediaRecorderStressTest#testStressRecordVideoAndPlayback FAIL
    java.lang.RuntimeException: start failed.
    at android.media.MediaRecorder.start(Native Method)
    at android.mediastress.cts.MediaRecorderStressTest.testStressRecordVideoAndPlayback(MediaRecorderStressTest.java:395)
    at java.lang.reflect.Method.invokeNative(Native Method)
    at android.test.InstrumentationTestCase.runMethod(InstrumentationTestCase.java:214)
    at android.test.InstrumentationTestCase.runTest(InstrumentationTestCase.java:199)
    at android.test.ActivityInstrumentationTestCase2.runTest(ActivityInstrumentationTestCase2.java:192)
    at android.test.AndroidTestRunner.runTest(AndroidTestRunner.java:169)
    at android.test.AndroidTestRunner.runTest(AndroidTestRunner.java:154)
    at android.test.InstrumentationTestRunner.onStart(InstrumentationTestRunner.java:545)
    at android.app.Instrumentation$InstrumentationThread.run(Instrumentation.java:1551)

    In this moment, we get logcat messages which list below and we found it seems ioctl error in CameraHal. Do you know the root cause of this problems? We also keep to investigate. Thank you for your help.

    I/CameraService(  126): Opening camera 0
    I/CameraHal(  126): camera_device open
    E/CameraHal(  126): (4011e488)   hardware/ti/omap4xxx/camera/OMXCameraAdapter/OMXCameraAdapter.cpp:1827 UseBuffersPreview - Timeout expired on preview buffer registration
    E/CameraHal(  126): (4011e488)   hardware/ti/omap4xxx/camera/OMXCameraAdapter/OMXCameraAdapter.cpp:1840 UseBuffersPreview - Exiting function UseBuffersPreview because of ret -1 eError=0
    E/ion     (  126): ioctl -1073460991 failed with code -1: Bad file number
    E/ion     (  126): ioctl -1073460991 failed with code -1: Bad file number
    E/ion     (  126): ioctl -1073460991 failed with code -1: Bad file number
    E/ion     (  126): ioctl -1073460991 failed with code -1: Bad file number
    E/ion     (  126): ioctl -1073460991 failed with code -1: Bad file number
    E/ion     (  126): ioctl -1073460991 failed with code -1: Bad file number

    Best Regards,

    Mike

  • 0 in reply to Mike Hsu
    Prodigy 230 points

    Hi Mike,

    If -1073460991 is the error you are still getting from ION, then it is running out of memory I guess.

    In the log I have attached 4454.logcat.txt you can see similar prints.

    Further to this, I removed a purgatory list in the SurfaceFlinger after which I did not run into the OOM. Media playback all tests passed and camera only one test failed , that was not due to ION memory alloc failure.

    However, I am not sure what actually caused this mem leak, I suppose removing the purgatory list is not good solution.

    Regards,

    Keerthi

  • 0 in reply to Keerthi Karanth
    TI__Genius 10545 points

    Check the Camera patches in previous post they are to fix a SurfaceFlinger memory leak issue.

  • 0 in reply to Manuel Contreras
    Prodigy 235 points

    Hi ,

    Sorry for reply updates lately because we try to do all day for mediastress test.

    For previous MediaRecorderStressTest of CTS fail issues, thank you for your help and it seems all be solved because we also apply another patch (http://review.omapzoom.org/#/c/21787/5).

    But, we also found another CTS fail issues for android.hardware.

    It will happen fail in different items of android.hardware randomly but will show the same details and log:

    1. details:

    "Failed to receive adb shell test output within 600000 ms. Test may have timed out, or adb connection to device became unresponsive'. Check device logcat for details” and, the next all items will fail and show “Fail to open camera.”

    2. logcat:

    04-25 00:58:18.406  1800  1814 I TestRunner: started: testPreviewCallback(android.hardware.cts.CameraTest)

    ....

    04-25 00:58:20.281  1800  2628 V CameraTest: start loopRun

    04-25 00:58:20.281   125  2544 I CameraService: Opening camera 0

    04-25 00:58:20.281   125  2544 I CameraHal: camera_device open

    04-25 00:58:20.320  1800  2628 V CameraTest: camera is opened

    04-25 00:58:20.484   125   125 E CameraHal: M1 OMX GraphicBuffer ANativeWindowDisplayAdapter::allocateBuffer

    04-25 00:58:20.734   125   125 D DOMX    : hardware/ti/domx/omx_proxy_component/omx_camera/src/omx_proxy_camera.c:214 Camera_SendCommand()        

    04-25 00:58:20.734   125   125 D DOMX    : ERROR:  Error in DCC Init

    04-25 00:58:21.390   125  2631 D CameraHal: PPM: Standby to first shot: Sensor Change completed -  :1064.728 ms :  1366851501397 ms

    04-25 00:58:22.859  1800  1800 V CameraTest: waitForPreviewDone: timeout

    04-25 00:58:22.960   125   197 E CameraHal: (5794f0)   hardware/ti/omap4xxx/camera/BaseCameraAdapter.cpp:1793 setState - Adapter state switch INTIALIZED_STATE Invalid Op! event  = CAMERA_STOP_IMAGE_CAPTURE

    04-25 00:58:22.976   125   197 I CameraService: Destroying camera 0

    04-25 00:58:23.140  1800  1800 V CameraTest: start waiting for looper

    04-25 00:58:23.140  1800  2637 V CameraTest: start loopRun

    04-25 00:58:23.140   125   125 I CameraService: Opening camera 0

    04-25 00:58:23.140   125   125 I CameraHal: camera_device open

    04-25 00:58:23.179  1800  2637 V CameraTest: camera is opened

    04-25 00:58:23.390   125   197 E CameraHal: M1 OMX GraphicBuffer ANativeWindowDisplayAdapter::allocateBuffer

    04-25 00:58:23.750   125   197 D DOMX    : hardware/ti/domx/omx_proxy_component/omx_camera/src/omx_proxy_camera.c:214 Camera_SendCommand()

    04-25 00:58:23.750   125   197 D DOMX    : ERROR:  Error in DCC Init

    04-25 00:58:26.765   125   197 E CameraHal: (5794f0)   hardware/ti/omap4xxx/camera/OMXCameraAdapter/OMXCameraAdapter.cpp:1827 UseBuffersPreview - Timeout expired on preview buffer registration

    04-25 00:58:26.765   125   197 E CameraHal: (5794f0)   hardware/ti/omap4xxx/camera/OMXCameraAdapter/OMXCameraAdapter.cpp:1840 UseBuffersPreview - Exiting function UseBuffersPreview because of ret -1 eError=0

    04-25 00:58:26.765   125   197 E ion     : ioctl -1073460991 failed with code -1: Bad file number

    04-25 00:58:26.765   125   197 E ion     : ioctl -1073460991 failed with code -1: Bad file number

    04-25 00:58:26.765   125   197 E ion     : ioctl -1073460991 failed with code -1: Bad file number

    04-25 00:58:26.765   125   197 E ion     : ioctl -1073460991 failed with code -1: Bad file number

    04-25 00:58:26.765   125   197 E ion     : ioctl -1073460991 failed with code -1: Bad file number

    04-25 00:58:26.765   125   197 E ion     : ioctl -1073460991 failed with code -1: Bad file number

    So far, I have no idea.

    Best Regards,

    Mike

  • 0 in reply to Mike Hsu
    TI__Genius 10545 points

    Following the post from Keerthi in this same post, you should try next patches, they are the same mentioned before,

    http://review.omapzoom.org/#/c/21492/ - Camera app: Prevent graphic out of memory error

    http://review.omapzoom.org/#/c/21661/ - Camera app: Properly handle Surface destruction

    http://review.omapzoom.org/#/c/22390/ - OMAPCamera: Hide camera preview when activity not visible

    http://review.omapzoom.org/#/c/22028/ - Camcorder: Speedup video record stop

    http://review.omapzoom.org/#/c/21491/ - Camera app: Prevent OOM during camera sensor change

    http://review.omapzoom.org/#/c/25461/ - Remove malloc of ComponentName that was causing memory leak

    I found next 2 patches that fixes some issue with this test, but first link shows like first option but second one is used for the same issue, maybe trying first each alone then the  two together.

    http://review.omapzoom.org/#/c/27150/1 - Camera: Fix setting VIDEO_MODE on recording

    http://review.omapzoom.org/#/c/27712/1 - CameraHAL: Fix loosing video capture mode after restart preview

  • 0 in reply to Manuel Contreras
    Prodigy 235 points

    Hi,

    I found a patch (http://review.omapzoom.org/#/c/23535/) and try to merge it our branch. For my test result, it seems can solve ion: ioctl -1073460991 issue.

    So I think Keerthi you can try it.

    However, have a little chance will happen fail in testCameraToSurfaceTextureMetadata item of android.hardware and logcat list below:

    04-25 16:00:32 I/014994DA09003015: android.hardware.cts.CameraGLTest#testCameraToSurfaceTextureMetadata FAIL
    junit.framework.AssertionFailedError: Too many frame intervals out of frame rate bounds: 8, limit 4
    at android.hardware.cts.CameraGLTest$5.run(CameraGLTest.java:623)
    at android.hardware.cts.CameraGLTest.runForAllCameras(CameraGLTest.java:325)
    at android.hardware.cts.CameraGLTest.testCameraToSurfaceTextureMetadata(CameraGLTest.java:501)
    at java.lang.reflect.Method.invokeNative(Native Method)
    at android.test.InstrumentationTestCase.runMethod(InstrumentationTestCase.java:214)
    at android.test.InstrumentationTestCase.access$000(InstrumentationTestCase.java:36)
    at android.test.InstrumentationTestCase$2.run(InstrumentationTestCase.java:189)
    at android.app.Instrumentation$SyncRunnable.run(Instrumentation.java:1569)
    at android.os.Handler.handleCallback(Handler.java:605)
    at android.os.Handler.dispatchMessage(Handler.java:92)
    at android.os.Looper.loop(Looper.java:137)
    at android.app.ActivityThread.main(ActivityThread.java:4424)
    at java.lang.reflect.Method.invokeNative(Native Method)
    at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:784)
    at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:551)
    at dalvik.system.NativeStart.main(Native Method)

    For this issue, do you have ever meet it and have some idea to solve it?

    Thank you for your help.

    Best Regards,

    Mike

  • 0 in reply to Mike Hsu
    Prodigy 230 points

    Hi Mike

    With the help of these patches on 4AI.1.4 ICS

    http://review.omapzoom.org/#/c/22802/1
    http://review.omapzoom.org/#/c/22803/1

    the out of memory problem was solved. Manuel thanks for pointing to patches.

    I am not getting any OOM in camera stress test.

    Regards,

    Keerthi

  • 0 in reply to Keerthi Karanth
    TI__Genius 10545 points

    Keerthi,

    No problem, may I ask if are you facing the issue with testCameraToSurfaceTextureMetadata?

  • 0 in reply to Manuel Contreras
    Prodigy 235 points

    Hi Manuel, Keerthi,

    Thank you for your help.

    Unfortunately, I put some devices for doing all days CTS test, we still found the below problem we discussed. So my previous suggestion seems not the real solution..

    04-25 00:58:26.765   125   197 E CameraHal: (5794f0)   hardware/ti/omap4xxx/camera/OMXCameraAdapter/OMXCameraAdapter.cpp:1840 UseBuffersPreview - Exiting function UseBuffersPreview because of ret -1 eError=0

    04-25 00:58:26.765   125   197 E ion     : ioctl -1073460991 failed with code -1: Bad file number

    Besides, I still meet the issue with testCameraToSurfaceTextureMetadata.

    The two issues are so confuse to me and will block our CTS test.

    We keep to investigate and find some patch could solve these issues or not.

    Best Regards,

    Mike

  • 0 in reply to Mike Hsu
    Prodigy 230 points

    Hi Mike,Manuel

    No, I do not see failure in testCameraToSurfaceTextureMetadata


    Regards,

    Keerthi

  • 0 in reply to Keerthi Karanth
    Prodigy 235 points

    Hi Keerthi,

    Do you have apply some related patches about camera, surface or fix the testCameraToSurfaceTextureMetadata item?


    Best Regards,

    Mike

  • 0 in reply to Mike Hsu
    TI__Genius 10545 points

    For what I investigated I cannot find a specific patch for this testCameraToSurfaceTextureMetadata issue.

    I was thinking on 2 possible solutions, one is debug the issue and the second is to isolate the patches that made it work in one of the setups.

    No action path from my side by now since previous post's answer should provide required information.

  • 0 in reply to Manuel Contreras
    Prodigy 235 points

    Do you know what happen about this and how to solve this? It will cause next all CTS items fail.

    From logcat message, I guess it seems the CTS app is crashing because no memory enough.

    45517:I/TestRunner(17926): started: testCallback(android.media.cts.MediaPlayerTest)
    45518-I/ActivityThread(17926): Pub android.content.cts.dummyprovider: android.content.cts.DummyProvider
    45519-E/ASFDummyExtractor(  125): isASFParserAvailable
    45520-W/ASFDummyExtractor(  125): ASF parser is not available
    45521-I/ActivityManager(  227): START {act=android.intent.action.MAIN flg=0x10000000 cmp=com.android.cts.stub/android.media.cts.MediaStubActivity} from pid 17926
    45522-I/InputReader(  227): Reconfiguring input devices.  changes=0x00000004
    45523-I/InputReader(  227): Device reconfigured: id=5, name='Zinitix_BT4x2', surface size is now 272x480, mode is 1
    45524-I/ActivityManager(  227): Config changed: {1.0 0mcc0mnc en_US layoutdir=0 sw272dp w362dp h614dp nrml long port finger -keyb/v/h -nav/h s.337}
    45525-D/dalvikvm(  802): GC_CONCURRENT freed 413K, 10% free 6223K/6855K, paused 2ms+2ms
    45526-D/OpenGLRenderer(15603): Flushing caches (mode 0)
    45527-D/OpenGLRenderer(15603): Flushing caches (mode 0)
    45528-D/OpenGLRenderer(  502): Flushing caches (mode 0)
    45529-D/dalvikvm(  227): GC_CONCURRENT freed 1011K, 58% free 10459K/24583K, paused 2ms+8ms
    45530-D/OpenGLRenderer(  502): Flushing caches (mode 0)
    45531-W/WindowManagerService.mKeyguardTokenWatcher(  227): cleaning up leaked reference: cts
    45532-D/OpenGLRenderer(15603): Flushing caches (mode 1)
    45533-I/libEGL  (17926): Processor type: omap4460
    45534-D/libEGL  (17926): loaded /system/lib/egl/libGLES_android.so
    45535-D/dalvikvm(  502): GC_EXPLICIT freed 235K, 34% free 6174K/9287K, paused 6ms+5ms
    45536-D/libEGL  (17926): loaded /vendor/lib/egl/libEGL_POWERVR_SGX540_120.so
    45537-D/libEGL  (17926): loaded /vendor/lib/egl/libGLESv1_CM_POWERVR_SGX540_120.so
    45538-D/libEGL  (17926): loaded /vendor/lib/egl/libGLESv2_POWERVR_SGX540_120.so
    45539-D/OpenGLRenderer(17926): Enabling debug mode 0
    45540-I/ActivityManager(  227): Displayed com.android.cts.stub/android.media.cts.MediaStubActivity: +446ms
    45541-W/InputManagerService(  227): Got RemoteException sending setActive(false) notification to pid 17646 uid 10064
    45542-D/dalvikvm(  227): GC_EXPLICIT freed 231K, 59% free 10313K/24583K, paused 6ms+4ms
    45543-D/dalvikvm( 7302): GC_CONCURRENT freed 415K, 9% free 6016K/6599K, paused 3ms+4ms
    45544-D/        (  118): DescriptorTransport::scheduleReopen() IS REOPENING!!!
    45545-D/dalvikvm( 7302): GC_CONCURRENT freed 414K, 9% free 6016K/6599K, paused 3ms+3ms
    45546-D/dalvikvm( 7302): GC_CONCURRENT freed 414K, 9% free 6016K/6599K, paused 3ms+4ms
    45547-I/EventLogService(  802): Aggregate from 1367561132949 (log), 1367561132949 (data)
    45548-W/ThrottleService(  227): unable to find stats for iface rmnet0
    45549-D/dalvikvm(  227): GC_CONCURRENT freed 971K, 58% free 10407K/24583K, paused 5ms+5ms
    45550-D/skia    (18029): SkGraphics::Init() - BLTsville (CPU) dlopen success
    45551-D/AndroidRuntime(18029):
    45552-D/AndroidRuntime(18029): >>>>>> AndroidRuntime START com.android.internal.os.RuntimeInit <<<<<<
    45553-D/AndroidRuntime(18029): CheckJNI is OFF
    45554-D/AndroidRuntime(18029): Calling main entry com.android.commands.pm.Pm
    45555-D/AndroidRuntime(18029): Shutting down VM
    45556-D/dalvikvm(18029): GC_CONCURRENT freed 97K, 83% free 459K/2560K, paused 0ms+1ms
    45557-D/skia    (18029): SkGraphics::Term() dlclose() BLTsville (CPU)
    45558-D/skia    (18053): SkGraphics::Init() - BLTsville (CPU) dlopen success
    45559-D/AndroidRuntime(18053):
    45560-D/AndroidRuntime(18053): >>>>>> AndroidRuntime START com.android.internal.os.RuntimeInit <<<<<<
    45561-D/AndroidRuntime(18053): CheckJNI is OFF
    45562-D/AndroidRuntime(18053): Calling main entry com.android.commands.pm.Pm
    45563-D/AndroidRuntime(18053): Shutting down VM
    45564-D/dalvikvm(18053): GC_CONCURRENT freed 97K, 83% free 459K/2560K, paused 0ms+1ms
    45565-D/skia    (18053): SkGraphics::Term() dlclose() BLTsville (CPU)
    45566-D/skia    (18074): SkGraphics::Init() - BLTsville (CPU) dlopen success
    45567-D/AndroidRuntime(18074):
    45568-D/AndroidRuntime(18074): >>>>>> AndroidRuntime START com.android.internal.os.RuntimeInit <<<<<<
    45569-D/AndroidRuntime(18074): CheckJNI is OFF
    45570-D/AndroidRuntime(18074): Calling main entry com.android.commands.am.Am
    45571-I/ActivityManager(  227): Force stopping package com.android.cts.stub uid=10064
    45572-I/ActivityManager(  227): Killing proc 17926:com.android.cts.stub/10064: force stop
    45573-W/ActivityManager(  227): Force removing ActivityRecord{41178798 com.android.cts.stub/android.media.cts.MediaStubActivity}: app died, no saved state
    45574:W/ActivityManager(  227): Crash of app com.android.cts.stub running instrumentation ComponentInfo{com.android.cts.media/android.test.InstrumentationCtsTestRunner}
    45575-I/InputReader(  227): Reconfiguring input devices.  changes=0x00000004
    45576-I/InputReader(  227): Device reconfigured: id=5, name='Zinitix_BT4x2', surface size is now 272x480, mode is 1
    45577-I/ActivityManager(  227): Force stopping package com.android.cts.stub uid=10064
    45578-V/PhoneStatusBar(  502): setLightsOn(true)
    45579-I/ActivityManager(  227): Config changed: {1.0 0mcc0mnc en_US layoutdir=0 sw272dp w640dp h337dp nrml long land finger -keyb/v/h -nav/h s.338}
    45580-D/OpenGLRenderer(  502): Flushing caches (mode 0)
    45581-D/OpenGLRenderer(  502): Flushing caches (mode 0)
    45582-I/ActivityManager(  227): Start proc com.android.cts.stub for added application com.android.cts.stub: pid=18084 uid=10064 gids={3003, 1006, 1015, 1007}
    45583-W/InputDispatcher(  227): channel '411a2208 com.android.cts.stub/android.media.cts.MediaStubActivity (server)' ~ Consumer closed input channel or an error occurred.  events=0x8
    45584-E/InputDispatcher(  227): channel '411a2208 com.android.cts.stub/android.media.cts.MediaStubActivity (server)' ~ Channel is unrecoverably broken and will be disposed!
    45585-W/InputDispatcher(  227): Attempted to unregister already unregistered input channel '411a2208 com.android.cts.stub/android.media.cts.MediaStubActivity (server)'
    45586-I/WindowManager(  227): WINDOW DIED Window{411a2208 com.android.cts.stub/android.media.cts.MediaStubActivity paused=false}
    45587-W/WindowManager(  227): Force-removing child win Window{410ee1d0 SurfaceView paused=false} from container Window{411a2208 com.android.cts.stub/android.media.cts.MediaStubActivity paused=false}
    45588-W/WindowManager(  227): Force-removing child win Window{413bc830 SurfaceView paused=false} from container Window{411a2208 com.android.cts.stub/android.media.cts.MediaStubActivity paused=false}
    45589-W/WindowManager(  227): Force-removing child win Window{411c2580 SurfaceView paused=false} from container Window{411a2208 com.android.cts.stub/android.media.cts.MediaStubActivity paused=false}
    45590:I/InstrumentationCtsTestRunner(18084): Disabling keyguard
    45591-I/ActivityThread(18084): Pub ctstest: android.content.cts.MockContentProvider
    45592-I/ActivityThread(18084): Pub android.content.cts.dummyprovider: android.content.cts.DummyProvider
    45593:D/InstrumentationCtsTestRunner(18084): Total memory  : 6168544
    45594:D/InstrumentationCtsTestRunner(18084): Used memory   : 6018152
    45595:D/InstrumentationCtsTestRunner(18084): Free memory   : 150392
    45596:D/InstrumentationCtsTestRunner(18084): Now executing : android.media.cts.MediaPlayerTest

  • 0 in reply to Mike Hsu
    TI__Genius 10545 points

    I have to say, it should not be this but it is one thing to check.

    I am trying to find a pattern in the fail but I can only think to ask if you followed the instructions to disable Keyguard (Security/none) found in Compatibility Test Suite User Manual? see "4.3 Setting up your device" section. I remember that by general code there should be a check to see if Keyguard is enabled or not then if it is disable it.

    http://source.android.com/compatibility/android-cts-manual-r6.pdf

    In you log it is present "Disabling Keyguard" and another message with similar text about Keyguard.

    And I found the media patterns at the bottom part of next link, that is great.

    http://source.android.com/compatibility/cts-intro.html

  • 0 in reply to Manuel Contreras
    TI__Genius 10545 points

    Mike,

    I received feedback about the testCameraToSurfaceTextureMetadata error.

    In file CtsHardwareTestCases.xml it has defined a timeout for this test of testCameraToSurfaceTextureMetadata, like the number of camera preview sets, it is not possible to pass it using 20 minutes timeout, try changing it to 80 minutes, this for CTS 4.2 R4. In recent CTS this timeout is not present, I don't have information if error is present or not.

    Thanks.

  • 0 in reply to Manuel Contreras
    Prodigy 235 points

    Hi ,


    Thank you for keeping investigation about the testCameraToSurfaceTextureMetadata error.

    We are using CTS 4.0.3 R3 and the version doesn't have timeout flags for this test of testCameraToSurfaceTextureMetadata in file CtsHardwareTestCases.xml.

    Do you know, is it acceptable to add 80 minutes timeout flag for testCameraToSurfaceTextureMetadata by myself in CTS 4.0.3 R3?


    Besides, when running full CTS test, have a rarely chance will happen below timeout message in mediastress. Is this a normal situation or do you have ideas?

    Test failed to run to completion. Reason: 'Failed to receive adb shell test output within 600000 ms. Test may have timed out, or adb connection to device became unresponsive'.


    Thanks.

    Best regards,

    Mike

  • 0 in reply to Mike Hsu
    Prodigy 235 points

    In CTS 4.2 R4, I found the version also add timeout flags in CtsMediaStressTestCases.xml. I'm not sure is it related to my previous problem which timeout for mediastress?

    <TestCase name="H263QcifLongPlayerTest"><Test name="testPlay00" timeout="11"/></TestCase>

  • 0 in reply to Mike Hsu
    TI__Genius 10545 points

    This is code snippet they shared

    The CtsHardwareTestCases.xml with path android-cts/repository/testcases has code
      </TestCase>
    <TestCase name="CameraGLTest">
    <Test name="testSetPreviewTexturePreviewCallback" />
    <Test name="testSetPreviewTextureBothCallbacks" />
    <Test name="testSetPreviewTextureTextureCallback" />
    <Test name="testCameraToSurfaceTextureMetadata" timeout="    20" />

    they mentioned recent releases don't have this timeout value, modifying it to 80 made a 60 minutes test pass.

  • 0 in reply to Manuel Contreras
    Prodigy 235 points

    Thank you for your directions.

    We can't found that timeout value in CtsHardwareTestCases.xml in CTS 4.0.3_R3 version.

    Do you know is it acceptable to add the timeout value by myself in CTS 4.0.3_R3 version?

  • 0 in reply to Mike Hsu
    TI__Genius 10545 points

    I don't know how Google handles it, if modifying the timeout works you may need to check with them in their forums.

  • 0 in reply to Manuel Contreras
    Prodigy 10 points

    Try  cts-tf >   run cts --help-all

    Possible it helps you.