LMMS segfaults when using tripleosc with arpeggio

[diizy]

I've had this happen twice now.

Take the triple osc, turn on arpeggio, set gate at a bit over 100 and set the time very low, under 32ms, and the mode to sort.

Then when you play the instrument, at some point LMMS crashes with a segfault. It seems a bit random and unpredictable, though. I haven't found an exact condition that causes this.

[diizy]

I've ran LMMS inside GDB and got the segfaul trapped there. I ran backtrace and it gave me this:

(gdb) backtrace
#0  Mixer::renderNextBuffer (this=0x8475c88) at /home/vesa/tmp/lmms-ui/src/core/Mixer.cpp:657
#1  0x0816fed7 in Mixer::fifoWriter::run (this=0x8a94860) at /home/vesa/tmp/lmms-ui/src/core/Mixer.cpp:1207
#2  0xb7211eb0 in ?? () from /usr/lib/i386-linux-gnu/libQtCore.so.4
#3  0xb7fa4d4c in start_thread () from /lib/i386-linux-gnu/libpthread.so.0
#4  0xb6a29bae in clone () from /lib/i386-linux-gnu/libc.so.6

I'm not very experienced at debugging. Is there anything else I could do to help track down this bug?

[SadaleNet]

I cannot reproduce the bug.
Further info required:

• Approximately, how long does it take for you to reproduce the bug?
• What did gdb say before the line (gdb) backtrace

[diizy]

On 01/25/2014 04:43 AM, wongcc966422 wrote:

I cannot reproduce the problem.
Further data required:

• Approximately, how long does it take for you to reproduce the bug?
• What did gdb say before the line |(gdb) backtrace|

—
Reply to this email directly or view it on GitHub
#128 (comment).

It varies. Try turning the speed all the way to 25ms and then playing
the tripleosc. At some point it should happen. Try also changing the
modulations, put at least one of them on PM, I don't know if that has an
effect...

I don't have gdb open anymore, I'll see later if I can do it again.

[diizy]

Well, I can't seem to reproduce this again myself, either. Maybe it was some kind of weird temporary glitch, maybe even caused by something other than LMMS... impossible to say for now. I'll report here if I catch the bug again.

[tobydox]

Ok, closing this for the time being if you don't mind :-)

[8tab]

Hi,
to catch sporadic issues like this one, you can enable memory dump file
generation when crash occurs.
Having that file, we can debug it offline and hopefully find the root
cause. Ideally, lmms should not crash even if there is an issue with
user settings.
How to enable memory dump in Linux:
http://www.redhat.com/archives/fedora-list/2006-May/msg00760.html.
How to enable memory dump in Windows:
http://msdn.microsoft.com/en-us/library/windows/desktop/bb787181(v=vs.85).aspx.

On 25.01.2014 17:03, Vesa V wrote:

Well, I can't seem to reproduce this again myself, either. Maybe it
was some kind of weird temporary glitch, maybe even caused by
something other than LMMS... impossible to say for now. I'll report
here if I catch the bug again.

—
Reply to this email directly or view it on GitHub
#128 (comment).