Frequent binder error and crash on android/imx6

[oussama]

I'm using android 4.4.2/imx6q. I have frequent binder error with frequent random crash (null pointer crash usually) when plying with android GUI.

Here is an example of the binder error:

[  137.361335] binder: release proc 3628, transaction 25361, not freed

[  137.367610] binder: release proc 3628, transaction 25388, not freed

[  162.268018] binder: 3182:3182 transaction failed 29189, size 92-0

[  174.459282] binder: release 2297:2345 transaction 30308 in, still active

[  174.466001] binder: send failed reply for transaction 30308 to 3057:3075

[  174.474214] binder: release 2297:3257 transaction 30299 in, still active

[  174.483290] binder: send failed reply for transaction 30299 to 3870:3870

[  174.490148] binder: release proc 2297, transaction 30309, not freed

[  174.496757] request_suspend_state: wakeup (0->0) at 174258127689 (1970-01-02 00:29:57.974331687 UTC)

[  174.529938] init: untracked pid 3505 exited

[  174.559633] init: untracked pid 3689 exited

[  174.589377] init: untracked pid 3482 exited

[  174.593865] init: untracked pid 3373 exited

[  174.598194] init: untracked pid 3410 exited

[  174.608484] init: untracked pid 3430 exited

[  174.616773] init: untracked pid 3229 exited

[  174.622210] init: untracked pid 3316 exited

[  174.628353] init: untracked pid 3445 exited

[  174.634000] init: untracked pid 3182 exited

[  174.639448] init: untracked pid 3212 exited

[  174.645194] init: untracked pid 3195 exited

[  174.653668] init: untracked pid 3113 exited

[  174.659857] init: untracked pid 3262 exited

[  174.665214] init: untracked pid 3279 exited

[  174.670170] init: untracked pid 3462 exited

[  174.674735] init: untracked pid 3538 exited

[  174.679930] init: untracked pid 3057 exited

[  174.759346] init: untracked pid 3870 exited

[  174.839368] alarm_release: clear alarm, pending 0

[  174.844092] alarm_release: clear alarm, pending 0

[  175.003312] binder: 3057: binder_alloc_buf, no vma

[  175.008343] binder: 4008:4104 transaction failed 29201, size 124-4

[  175.015018] binder: 3057: binder_alloc_buf, no vma

[  175.019899] binder: 4008:4104 transaction failed 29201, size 124-4

[oussama]

The issue is solved. The errors were caused by DRAM issue.

[hominjeon]

Hi. oussama.

I had an issue similar to you.

I wonder how you solved the DRAM issue.

[nicolasbugnariu]

Hello, 
Wondering, is there an answer to the DRAM issue?  How was it solved?  Thank you

[SergioSolis]

Hello Oussama.

The GUI works without problems rigth?, but you still get the error correct?

There could be a lot of reasons for this, but probably android init is trying to initialize a specific service and it is failing.

I would recommend you to check the logcat (it most probably be a huge file) and you could find the program that is constantly crashing.

Have a great day,
Sergio

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[oussama]

Hello Sergio,

Each time the GUI crash and restart, there are fatal signal like signal 11, singal 7.

Example:

F/libc    ( 2283): invalid address or address of corrupt block 0x2b357b54 passed to dlfree

F/libc    ( 2283): Fatal signal 7 (SIGBUS) at 0x00000000 (code=128), thread 2321 (Binder_2)

I/DEBUG   ( 2282): *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***

I/DEBUG   ( 2282): Build fingerprint: 'Freescale/xxxx/xxxx:4.4.2/0.0.1/20150102:eng/dev-keys'

I/DEBUG   ( 2282): Revision: '405525'

I/DEBUG   ( 2282): pid: 2283, tid: 2321, name: Binder_2  >>> /system/bin/surfaceflinger <<<

I/DEBUG   ( 2282): signal 7 (SIGBUS), code 128 (SI_KERNEL), fault addr 00000000

I/DEBUG   ( 2282): Abort message: 'invalid address or address of corrupt block 0x2b357b54 passed to dlfree'

OR

F/libc    ( 3781): Fatal signal 11 (SIGSEGV) at 0x00000400 (code=1), thread 4073 (android.browser)

I/DEBUG   ( 3896): *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***

I/DEBUG   ( 3896): Build fingerprint: 'Freescale/xxxx/xxxx:4.4.2/0.0.1/20150102:eng/dev-keys'

I/DEBUG   ( 3896): Revision: '405525'

I/DEBUG   ( 3896): pid: 3781, tid: 4073, name: android.browser  >>> com.android.browser <<<

I/DEBUG   ( 3896): signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 00000400

OR

F/libc    ( 2274): invalid address or address of corrupt block 0x34e95b54 passed to dlfree

F/libc    ( 2274): Fatal signal 7 (SIGBUS) at 0x00000000 (code=128), thread 3231 (Binder_3)

I/DEBUG   ( 2273): *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***

I/DEBUG   ( 2273): Build fingerprint: 'Freescale/xxxx/xxxx:4.4.2/0.0.1/20150102:eng/dev-keys'

I/DEBUG   ( 2273): Revision: '405525'

I/DEBUG   ( 2273): pid: 2274, tid: 3231, name: Binder_3  >>> /system/bin/surfaceflinger <<<

I/DEBUG   ( 2273): signal 7 (SIGBUS), code 128 (SI_KERNEL), fault addr 00000000

I/DEBUG   ( 2273): Abort message: 'invalid address or address of corrupt block 0x34e95b54 passed to dlfree'

OR

F/libc    ( 2277): Fatal signal 11 (SIGSEGV) at 0x0000041c (code=1), thread 3053 (Binder_2)

I/DEBUG   ( 2273): *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***

I/DEBUG   ( 2273): Build fingerprint: 'Freescale/xxxx/xxxx:4.4.2/0.0.1/20150102:eng/dev-keys'

I/DEBUG   ( 2273): Revision: '405525'

I/DEBUG   ( 2273): pid: 2277, tid: 3053, name: Binder_2  >>> /system/bin/mediaserver <<<

I/DEBUG   ( 2273): signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 0000041c

Any clue who cause this bad memory access on the kernel and userspace?

Have a Great Day!

Oussama

[SergioSolis]

Have you seen this behaviour with any other FSL BSP you have built?