octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Audiorecorder/Player segfault


From: pantxo diribarne
Subject: Re: Audiorecorder/Player segfault
Date: Thu, 25 Jul 2013 18:44:37 +0200

Hi,

I am a bit confused about what happens :
 - compiled without debug symbols, there is no error anymore with "playblocking" , and segfault with "play".
 - with debug symbols the segfault with play doesn't appear anymore (neither in octave nor with gdb running octave).

Now if I insist and try to record again and rebuild a new player :
octave:1> recorder = audiorecorder (44100, 16, 2); # gdb complains about not finding debug symbols in portaudio dylib.
octave:2> record (recorder, 2);
octave:3> player = audioplayer (recorder);
octave:4> play (player)
octave:5> play (player)
octave:6> play (player)  # nothing untill here.
octave:7> record (recorder, 2);
octave:8> player = audioplayer (recorder);
octave:9> play (player)
octave:10>
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x00000001063d2000
[Switching to process 28320 thread 0x910f]
0x0000000105913597 in portaudio_callback ()
(gdb) bt
#0  0x0000000105913597 in portaudio_callback ()
#1  0x0000000103b745e3 in AdaptingOutputOnlyProcess ()
#2  0x0000000103b73752 in PaUtil_EndBufferProcessing ()
#3  0x0000000103b790ed in AudioIOProc ()
#4  0x0000000105007f6f in AUConverterBase::RenderBus ()
#5  0x0000000105007544 in AUBase::DoRenderBus ()
#6  0x00000001050049cd in AUBase::DoRender ()
#7  0x000000010500bc13 in AUHAL::AUIOProc ()
#8  0x00007fff8795b9fa in HALC_ProxyIOContext::IOWorkLoop ()
#9  0x00007fff8795b151 in HALC_ProxyIOContext::IOThreadEntry ()
#10 0x00007fff8795b00c in HALB_IOThread::Entry ()
#11 0x00007fff8eb668bf in _pthread_start ()
#12 0x00007fff8eb69b75 in thread_start ()


Before the crash, I hear both what I recorded the first time and the second time (4 seconds). Is this the intended behavior?

Let me know if I can do further tests, and sorry for not being able to interpret a backtrace.

Pantxo


2013/7/25 Mike Miller <address@hidden>
On Thu, Jul 25, 2013 at 15:15:47 +0300, Vytautas Jancauskas wrote:
> The problem with playblocking should be fixed now. I am not sure what
> causes pantxos segfault though. Can you compile it with -g, run it
> through gdb and post the output of the stack trace maybe?

Cool. I also notice that playblocking is not interruptible with
ctrl-c. Can that signal be interpreted to stop the audio playback
gracefully, as if I had called play followed by stop?

--
mike


reply via email to

[Prev in Thread] Current Thread [Next in Thread]