Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Crash on 1-tick loops #3879

Closed
gi0e5b06 opened this issue Oct 13, 2017 · 8 comments
Closed

Crash on 1-tick loops #3879

gi0e5b06 opened this issue Oct 13, 2017 · 8 comments

Comments

@gi0e5b06
Copy link

gi0e5b06 commented Oct 13, 2017

LMMS crashes on playing 1-tick loops, but 2-ticks loops are fine.
(rc4 master alsa)
memory20171013.zip

@qnebra
Copy link

qnebra commented Oct 13, 2017

In my case lmms holds this 1-tick loop fine, very glitchy but no crash at all. But I have compiled version with merged bunch of PRs.

@gi0e5b06
Copy link
Author

You tried the new file? If yes, then that means it is not a logical mistake...
(btw, which OS and audio?)
Not an out of memory mistake.
Maybe a memory mistake (Segmentation fault (core dumped)) thread-related?
For me, it is not glitchy (cpu usage is low because only the instruments that don't use a lot of cpu are active) but it crashes after half a second. No problem with the 2-tick loop.

@qnebra
Copy link

qnebra commented Oct 13, 2017

I tried this one you provide in first post of this thread.
(Ubuntu 17.10, lmms use SDL2 and system using pulseaudio)

@gi0e5b06
Copy link
Author

gi0e5b06 commented Oct 14, 2017

I tried the other audio drivers, same problem.
It seems @PhysSong has the same problem (or at least it looks like) #3881 (8th comment).

Hmm. I have applied several patches from stable (but nothing directly related to this problem). I have fully recompiled everything and now it doesn't crash...

@PhysSong
Copy link
Member

I got a crash and the backtrace looks like one in #777. If there are no evidences which show this isn't #777, this should be closed as a duplicate.

@PhysSong
Copy link
Member

You may want to take a look at #2539.

@PhysSong
Copy link
Member

@gi0e5b06 Have you got any backtrace(s)? If so, please post it. It will help a lot.

@gi0e5b06
Copy link
Author

Thanks @qnebra and @PhysSong, it helps a lot. It was indeed #777 (May 28, 2014).
I changed the size of the queue and it doesn't crash anymore:
MixerWorkerThread.h:

#define JOB_QUEUE_SIZE 2048`

I also added this:
MixerWorkerThread::JobQueue:

		if( (int) m_queueSize >= JOB_QUEUE_SIZE-16 )
			qFatal("MixerWorkerThread::JobQueue full (%d)",JOB_QUEUE_SIZE);

I haven't had time to fix the problem but I will. It just needs a better queue.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants