strange android bug |
strange android bug |
Simon Sandström
Guest
|
Do you have any more logs? What is it trying to do, open and read a file? Maybe the file doesn't exist?
2014-08-28 16:17 GMT+02:00 Alexey Petruchik:
|
|||||||||||||
|
strange android bug |
Alexey Petruchik
Guest
|
Here is the full log -Â https://gist.github.com/stopiccot/ddafa2081ce458347f90File exists. It crashes randomly from time to time
On Thu, Aug 28, 2014 at 5:27 PM, Simon Sandström wrote:
|
|||||||||||||||
|
strange android bug |
MrOzBarry
|
I'd it only happens randomly, it may be a thread race condition On 28 Aug 2014 10:55, "Alexey Petruchik" wrote:
|
|||||||||||||||||
|
strange android bug |
gabomdq
|
Can you run addr2line and see what line crashes (the process is described in the Android docs) ?Could you be running out of memory?
2014-08-28 11:55 GMT-03:00 Alexey Petruchik:
-- Gabriel. |
|||||||||||||||||
|
strange android bug |
Alexey Petruchik
Guest
|
********** Crash dump: **********
Build fingerprint: 'Android/tate/tate:4.0.3/IML74K/7.4.9_user_4952320:user/release-keys' pid: 28245, tid: 28287  >>> com.deemedya.ChamAmazon <<< signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr deadd00d Stack frame #00  pc 00050a4e  /system/lib/libdvm.so (dvmAbort) Stack frame #01  pc 000556dc  /system/lib/libdvm.so (_Z20dvmDecodeIndirectRefP6ThreadP8_jobject) Stack frame #02  pc 0005865e  /system/lib/libdvm.so Stack frame #03  pc 000b9e00  /data/data/com.deemedya.ChamAmazon/lib/libSDL2.so: Routine Internal_Android_JNI_FileOpen at .../android-sdl/jni/SDL/src/core/android/SDL_android.c:720 Stack frame #04  pc 000ba63c  /data/data/com.deemedya.ChamAmazon/lib/libSDL2.so (Android_JNI_FileOpen): Routine Android_JNI_FileOpen at .../android-sdl/jni/SDL/src/core/android/SDL_android.c:870 Stack frame #05  pc 000db060  /data/data/com.deemedya.ChamAmazon/lib/libSDL2.so (SDL_RWFromFile_REAL): Routine SDL_RWFromFile_REAL at .../android-sdl/jni/SDL/src/file/SDL_rwops.c:499 Stack frame #06  pc 000ca63c  /data/data/com.deemedya.ChamAmazon/lib/libSDL2.so (SDL_RWFromFile): Routine SDL_RWFromFile at .../android-sdl/jni/SDL/src/dynapi/SDL_dynapi_procs.h:386 line 720 for me is: mid = (*mEnv)->GetMethodID(mEnv, (*mEnv)->GetObjectClass(mEnv, context),       "getAssets", "()Landroid/content/res/AssetManager;"); Btw, I'm also suspecting out of memory issues, but is it normal for android OS to just crash in libdvm.so when app is running out of memory istead of killing whole app? On Thu, Aug 28, 2014 at 6:18 PM, Gabriel Jacobo wrote:
|
|||||||||||||||||||
|
strange android bug |
Alexey Petruchik
Guest
|
But I'm not sure if it GetMethodID() that causes a crash or GetObjectClass()
On Thu, Aug 28, 2014 at 7:05 PM, Alexey Petruchik wrote:
|
|||||||||||||||||||||
|
strange android bug |
hardcoredaniel
Guest
|
Hi,
I would not assume that it's an out of memory case you are running into. There are other possibilities for a failure. I suggest the following to narrow down the issue: 1. (if not already in place) Run with CheckJNI enabled: http://android-developers.blogspot.de/2011/07/debugging-android-jni-with-checkjni.html 2. Look at the log messages shortly before the crash. Sometimes Android informs you about the JNI error before it forcibly terminates the VM, as is the case here. 3. Split the line in two, so you can actually check the return value from (*mEnv)->GetObjectClass(mEnv, context) which is of type "jclass", and should not be NULL. It is possible that your JNIEnv* and/or your jobject reference is not valid in the thread that you are calling the function that crashes. Regards, Daniel ---------- Původnà zpráva ---------- Od: Alexey Petruchik Komu: SDL Development List Datum: 28. 8. 2014 19:10:12 Předmět: Re: [SDL] strange android bug
|
|||||||||||||||||||||||
|
strange android bug |
Sam Lantinga
|
Also make sure your copy of SDLActivity.java matches the version of SDL you're using.
On Fri, Aug 29, 2014 at 4:05 AM, hardcoredaniel wrote:
|
|||||||||||||||||||||||||
|
strange android bug |
Alexey Petruchik
Guest
|
It matches but I think adding a check that verifies version in java code would great. I had a headache few times when I forgot to update java class after updating C code.
On Fri, Aug 29, 2014 at 6:03 PM, Sam Lantinga wrote:
|
|||||||||||||||||||||||||||
|
strange android bug |
Sam Lantinga
|
I agree.
On Tue, Sep 2, 2014 at 3:19 AM, Alexey Petruchik wrote:
|
|||||||||||||||||||||||||||||
|