0
0

I’m running into a problem where the game we’re working on is crashing on the HTC Eris (Android 2.1) when it tries to load the fmodex library. Loading works fine on all other devices we tested (Droid X, Nexus One, Atrix, Incredible, Milestone, Xperia, etc..) but crashes at this point only on the Eris. The point at which it crashes is listed in the debug trace as:

[code:as01b9yx]06-22 16:58:08.048: INFO/DEBUG(54): #00 pc 00033c04 /data/data/com.test/lib/libfmodex.so
06-22 16:58:08.048: INFO/DEBUG(54): #01 lr 80427d98 /data/data/com.test/lib/libfmodex.so[/code:as01b9yx]

Also, when I remove the audio libraries entirely from the code the app operates fine on the Eris. Do you know of any possible fix/workaround for this?

Thank you,

  • You must to post comments
0
0

Could you post the TTY output including the crash call stack running the new version of FMOD?

  • You must to post comments
0
0

I got the exact same errors with the HTC Wildfire

[code:35nknl04]

00 pc 00033c04 /data/data/com.test.app/lib/libfmodex.so

01 lr 80a27d98 /data/data/com.test.app/lib/libfmodex.so

[/code:35nknl04]

It works fine on the HTC Desire HD

Thansk for the feedback

  • You must to post comments
0
0

Turns out I didn’t have all the files properly updated. Version 4.37.00 no longer crashes although it still fails to work, I get an FMOD error saying it’s trying to use a feature that requires hardware support. How can I capture this and disable the audio?

  • You must to post comments
0
0

Yeah, unfortunately we have no fixes for Android in 4.35.06 but thanks for confirming the issue with our latest release.

We are going to pick up an HTC Hero from ebay, I’ll do some testing and get back to you with my findings, hopefully a fix although I am suspicious of another OS related issue.

  • You must to post comments
0
0

Ok..everything is working now!

I added a check in the initialization for FMOD_ERR_NEEDSHARDWARE. If it finds it I disable all calls to FMOD. Now the app works on all our testing devices, only without sound if the hardware doesn’t support it. Not perfect but works well enough for our case. 😀

  • You must to post comments
0
0

Thank you. I just tried updating to FMOD 4.35.06 but still see this issue.

  • You must to post comments
0
0

you could have checked init then reinitialize using FMOD_OUTPUT_NOSOUND as well to allow all commands to succeed, but the way you specified would use less cpu time :)

  • You must to post comments
0
0

So currently the reported non-working phones are: HTC Eris (aka HTC Hero) and HTC myTouch 3G (aka HTC Magic).

From my research all of these devices are ARMv6, so they will be using a different lib to the rest of the phones. However the simulator also uses the same lib as the non-working phones and I don’t get a crash from it. I have also tried running the ARMv5 FMOD lib on all our devices, and it works, however they are all ARMv7 devices. We may have to look into getting an HTC Hero for testing.

  • You must to post comments
0
0

Actually FMOD_OUTPUT_NOSOUND isn’t an option in this case, you do actually need to do what pszNicx has done and disable calls to FMOD. The no sound output mode still processes the audio, just doesn’t output it to speakers, without floating point support our mixer and resampler will not function. Hence you will receive FMOD_ERR_NEEDSHARDWARE regardless of chosen output mode.

  • You must to post comments
0
0

I haven’t had a chance to test this further but we did get an HTC Hero (HVGA & running Android 2.1) and it is also crashing at this point.

Don’t know how this would be related but just an observation that our 2 devices + massimobio’s 1 are all HVGA (480×320) devices.

  • You must to post comments
0
0

I’m using the latest version of FMOD 4.35.05 and the exact same crash happens with the examples (I tried PlaySound).
It works fine on a newer HTC Desire but not on the HTC myTouch 3G which unfortunately is one of the top 10 Android phones currently in use.

  • You must to post comments
0
0

We haven’t seen these crashes before, it looks like more device specific shenanigans.

Are you using the latest version of FMOD? We recently switched back to using the minimal c++ lib to avoid problems with the gnustl lib that manifested like what you are seeing.

Also do the examples function okay on those devices?

  • You must to post comments
0
0

I’m having a similar problem on an HTC mytouch 3G running 2.2.1.
It works fine on a Samsung Galaxy but crashes on the HTC as it tries to load libfmodexL.so

[code:2m8ss180]06-28 13:54:20.439: DEBUG/dalvikvm(920): Trying to load lib /data/data/com.massimobiolcati.irealb/lib/libfmodexL.so 0x43531e38
06-28 13:54:20.639: INFO/DEBUG(54): *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
06-28 13:54:20.649: INFO/DEBUG(54): Build fingerprint: ‘tmobile/opal/sapphire/sapphire:2.2.1/FRG83D/75603:user/release-keys’
06-28 13:54:20.649: INFO/DEBUG(54): pid: 920, tid: 920 >>> com.massimobiolcati.irealb <<<
06-28 13:54:20.649: INFO/DEBUG(54): signal 4 (SIGILL), fault addr 8043a47c
06-28 13:54:20.649: INFO/DEBUG(54): r0 805029b0 r1 804ede98 r2 80502994 r3 804fae48
06-28 13:54:20.649: INFO/DEBUG(54): r4 000ce964 r5 00000000 r6 80502988 r7 00000000
06-28 13:54:20.649: INFO/DEBUG(54): r8 80213b00 r9 0000cca0 10 4104ec28 fp 00000000
06-28 13:54:20.649: INFO/DEBUG(54): ip 00000100 sp bece4700 lr 8042d72c pc 8043a47c cpsr 60000010
06-28 13:54:21.479: INFO/DEBUG(54): #00 pc 0003a47c /data/data/com.massimobiolcati.irealb/lib/libfmodexL.so
06-28 13:54:21.489: INFO/DEBUG(54): #01 lr 8042d72c /data/data/com.massimobiolcati.irealb/lib/libfmodexL.so
06-28 13:54:21.489: INFO/DEBUG(54): code around pc:
06-28 13:54:21.489: INFO/DEBUG(54): 8043a45c e8bd0010 e12fff1e 461c4000 00000000
06-28 13:54:21.489: INFO/DEBUG(54): 8043a46c 3f800000 43b40000 472c4400 000c0f0c
06-28 13:54:21.489: INFO/DEBUG(54): 8043a47c ed9f6a3e eddf7a3e ed9f7a3e eddf6a3e
06-28 13:54:21.489: INFO/DEBUG(54): 8043a48c ed806a1b ed9f6a3d e59f20f4 e52d4004
06-28 13:54:21.489: INFO/DEBUG(54): 8043a49c e08f2002 e3a04001 e1a01000 e5804044
06-28 13:54:21.489: INFO/DEBUG(54): code around lr:
06-28 13:54:21.489: INFO/DEBUG(54): 8042d70c e286200c e2833008 e5862010 e5863008
06-28 13:54:21.499: INFO/DEBUG(54): 8042d71c e586200c e5865014 e2860028 eb003353
06-28 13:54:21.499: INFO/DEBUG(54): 8042d72c e2860f5e ebffd997 e2860fd6 ebffd995
06-28 13:54:21.509: INFO/DEBUG(54): 8042d73c e59f33fc e08f4004 e2867008 e7943003
06-28 13:54:21.509: INFO/DEBUG(54): 8042d74c e2871d36 e2872ed9 e281100c e2822008
06-28 13:54:21.509: INFO/DEBUG(54): stack:
06-28 13:54:21.509: INFO/DEBUG(54): bece46c0 bece47ee [stack]
06-28 13:54:21.509: INFO/DEBUG(54): bece46c4 000000fd
06-28 13:54:21.509: INFO/DEBUG(54): bece46c8 00000000
06-28 13:54:21.519: INFO/DEBUG(54): bece46cc 00000000
06-28 13:54:21.519: INFO/DEBUG(54): bece46d0 00280ea8 [heap]
06-28 13:54:21.519: INFO/DEBUG(54): bece46d4 017bad56
06-28 13:54:21.519: INFO/DEBUG(54): bece46d8 80406a34 /data/data/com.massimobiolcati.irealb/lib/libfmodexL.so
06-28 13:54:21.519: INFO/DEBUG(54): bece46dc b00013fd /system/bin/linker
06-28 13:54:21.519: INFO/DEBUG(54): bece46e0 80406a94 /data/data/com.massimobiolcati.irealb/lib/libfmodexL.so
06-28 13:54:21.519: INFO/DEBUG(54): bece46e4 804022e4 /data/data/com.massimobiolcati.irealb/lib/libfmodexL.so
06-28 13:54:21.529: INFO/DEBUG(54): bece46e8 00000000
06-28 13:54:21.529: INFO/DEBUG(54): bece46ec b000c448 /system/bin/linker
06-28 13:54:21.529: INFO/DEBUG(54): bece46f0 804fa004 /data/data/com.massimobiolcati.irealb/lib/libfmodexL.so
06-28 13:54:21.529: INFO/DEBUG(54): bece46f4 00000000
06-28 13:54:21.529: INFO/DEBUG(54): bece46f8 df002777
06-28 13:54:21.529: INFO/DEBUG(54): bece46fc e3a070ad
06-28 13:54:21.529: INFO/DEBUG(54): #00 bece4700 ffffffff
06-28 13:54:21.529: INFO/DEBUG(54): bece4704 bece4720 [stack]
06-28 13:54:21.539: INFO/DEBUG(54): bece4708 00000000
06-28 13:54:21.539: INFO/DEBUG(54): bece470c 00000000
06-28 13:54:21.539: INFO/DEBUG(54): bece4710 ade696b5 /system/lib/libicuuc.so
06-28 13:54:21.539: INFO/DEBUG(54): bece4714 017bad56
06-28 13:54:21.539: INFO/DEBUG(54): bece4718 b0010804
06-28 13:54:21.539: INFO/DEBUG(54): bece471c 8040d6a2 /data/data/com.massimobiolcati.irealb/lib/libfmodexL.so
06-28 13:54:21.539: INFO/DEBUG(54): bece4720 b000c9a0 /system/bin/linker
06-28 13:54:21.539: INFO/DEBUG(54): bece4724 bece4794 [stack]
06-28 13:54:21.549: INFO/DEBUG(54): bece4728 b000c9ac /system/bin/linker
06-28 13:54:21.549: INFO/DEBUG(54): bece472c 00000014
06-28 13:54:21.549: INFO/DEBUG(54): bece4730 00000000
06-28 13:54:21.549: INFO/DEBUG(54): bece4734 b000c448 /system/bin/linker
06-28 13:54:21.549: INFO/DEBUG(54): bece4738 80412608 /data/data/com.massimobiolcati.irealb/lib/libfmodexL.so
06-28 13:54:21.549: INFO/DEBUG(54): bece473c 00000002
06-28 13:54:21.549: INFO/DEBUG(54): bece4740 80406a34 /data/data/com.massimobiolcati.irealb/lib/libfmodexL.so
06-28 13:54:21.549: INFO/DEBUG(54): bece4744 b0001599 /system/bin/linker
06-28 13:54:22.449: INFO/ActivityManager(84): Process com.massimobiolcati.irealb (pid 920) has died.
06-28 13:54:22.449: INFO/WindowManager(84): WIN DEATH: Window{436c1790 com.massimobiolcati.irealb/com.massimobiolcati.irealb.iRealbTab paused=false}
06-28 13:54:22.479: DEBUG/Zygote(56): Process 920 terminated by signal (4)
[/code:2m8ss180]

  • You must to post comments
0
0

Could you link with the logging version of FMOD and provide a dump of the TTY with a full callstack?

  • You must to post comments
0
0

I updated to version 4.37.00. In the revision .txt file it says that it fixes the crash with unsupported devices but I’m still seeing the problem on the Eris.

  • You must to post comments
Showing 15 results
Your Answer

Please first to submit.