Hi Alexei, I just committed the fix in ticket #1195 (http://trac.pjsip.org/repos/ticket/1195). Thanks for the feedback. Best regards, Ming On Thu, Jan 13, 2011 at 7:52 AM, Alexei Kuznetsov <eofster at gmail.com> wrote: > Hello Ming, > > I can confirm that adding just one line > > pj_bzero(stream->play_thread_desc, sizeof(pj_thread_desc)); > > at 651 inside output_renderer() just before pj_thread_register() fixes the problem. What should I do next? Can it be considered as a solution or we did it just to test something? > > Alexei > > > > On 20 Dec, 2010, at 4:43, Ming wrote: > >> Hi Alexei, >> >> You can try adding the instruction: >> ? ?pj_bzero(stream->rec_thread_desc, sizeof(pj_thread_desc)); >> or >> ? ?pj_bzero(stream->play_thread_desc, sizeof(pj_thread_desc)); >> >> before each call of pj_thread_register(). Please let me know the >> result since I can't try it here. >> >> Best regards, >> Ming >> >> On Sun, Dec 19, 2010 at 10:57 PM, Alexei Kuznetsov <eofster at gmail.com> wrote: >>> On 17 Dec, 2010, at 16:48, Ming wrote: >>> >>>> For the crash, unless you can somehow reproduce it using pjsua, we >>>> can't help you much since I can't reproduce it on my machine. >>> >>> For me it's 100% reproducible in my app. What can I do to fix this or to help you to fix this? Under what circumstances can this assertion be hit? I don't do anything special with pjsua-lib from my app, I don't make any pjsip function calls from secondary threads. >>> >>> I made a debug build of pjsip. Here's the backtrace when assertion was hit. >>> >>> Assertion failed: (thread->signature1 != 0xDEAFBEEF || thread->signature2 != 0xDEADC0DE || (thread->thread == pthread_self())), function pj_thread_register, file ../src/pj/os_core_unix.c, line 405. >>> Program received signal: ??SIGABRT?. >>> Data Formatters temporarily unavailable, will re-try after a 'continue'. (Not safe to call dlopen at this time.) >>> (gdb) bt >>> #0 ?0x00007fff8871dc75 in __abort () >>> #1 ?0x00007fff8871dcd9 in abort () >>> #2 ?0x00007fff8870ac90 in __assert_rtn () >>> #3 ?0x000000010017334b in pj_thread_register (cstr_thread_name=0x1001bd238 "coreaudio", desc=0x1028adb08, ptr_thread=0x1028add08) at ../src/pj/os_core_unix.c:403 >>> #4 ?0x00000001000e776f in output_renderer (inRefCon=0x1028ad828, ioActionFlags=0x1027f326c, inTimeStamp=0x101eb4d00, inBusNumber=0, inNumberFrames=471, ioData=0x115cf8930) at ../src/pjmedia-audiodev/coreaudio_dev.c:651 >>> #5 ?0x00000001025915d0 in dyld_stub_sprintf () >>> #6 ?0x0000000102591693 in dyld_stub_sprintf () >>> #7 ?0x00007fff851bac2f in AudioConverterChain::CallInputProc () >>> #8 ?0x00007fff851ba7da in AudioConverterChain::FillBufferFromInputProc () >>> #9 ?0x00007fff851ba75b in BufferedAudioConverter::GetInputBytes () >>> #10 0x00007fff851ba61f in CBRConverter::RenderOutput () >>> #11 0x00007fff851ba39f in BufferedAudioConverter::FillBuffer () >>> #12 0x00007fff851ba739 in BufferedAudioConverter::GetInputBytes () >>> #13 0x00007fff851fd647 in Resampler2Wrapper::RenderOutput () >>> #14 0x00007fff851ba39f in BufferedAudioConverter::FillBuffer () >>> #15 0x00007fff851ba739 in BufferedAudioConverter::GetInputBytes () >>> #16 0x00007fff851ba61f in CBRConverter::RenderOutput () >>> #17 0x00007fff851ba39f in BufferedAudioConverter::FillBuffer () >>> #18 0x00007fff851ba510 in AudioConverterChain::RenderOutput () >>> #19 0x00007fff851ba39f in BufferedAudioConverter::FillBuffer () >>> #20 0x00007fff851ba11b in AudioConverterFillComplexBuffer () >>> #21 0x000000010259174d in dyld_stub_sprintf () >>> #22 0x00000001025912e0 in dyld_stub_sprintf () >>> #23 0x000000010258f523 in dyld_stub_sprintf () >>> #24 0x000000010258e870 in dyld_stub_sprintf () >>> #25 0x0000000102593ceb in AUGenericOutputEntry () >>> #26 0x00007fff8216032d in HP_IOProc::Call () >>> #27 0x00007fff821600ff in IOA_Device::CallIOProcs () >>> #28 0x00007fff8215ff35 in HP_IOThread::PerformIO () >>> #29 0x00007fff8215df44 in HP_IOThread::WorkLoop () >>> #30 0x00007fff8215d817 in HP_IOThread::ThreadEntry () >>> #31 0x00007fff8215d745 in CAPThread::Entry () >>> #32 0x00007fff88668536 in _pthread_start () >>> #33 0x00007fff886683e9 in thread_start () >>> >>> Alexei > > > _______________________________________________ > Visit our blog: http://blog.pjsip.org > > pjsip mailing list > pjsip at lists.pjsip.org > http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org > >
