mixxx 1.6.2+3 seg faults with jack + realtime

[seamus1089]

Hi,

When I load beta2 or beta 3 as a regular user running jack in realtime mode I get the following malloc related crash:

Debug: Starting up...
Debug: SoundManager::SoundManager()
Debug: SampleRate 44100
Debug: Latency 16
Debug: SoundManager::queryDevices()
Debug: SoundManager::clearDeviceList()
Debug: SoundManager::closeDevices()
Debug: type signal
Debug: type marks
Debug: type signal
Debug: type marks
Debug: Loading playlists and library tracks from XML...
Debug: Track::readXML "/home/seamus/.mixxxtrack.xml"
Debug: Break
Debug: playlist name Default
Debug: Constructed LibraryScanner!!!
Debug: FIXME: Need to tell the m_pPlaylistListModel to refresh in src/track.cpp on line: 1120
Debug: Trying to add 7017 songs to the library playlist
Debug: Adjusting column widths: tracktable width = 582 1% of that is: 5.82 FIXME: this should be done when initalizing the skin.
Debug: Shrinking Title/Comment for small screen...
Debug: selectedAPI is: "ALSA"
Debug: SoundManager::getDeviceList
Debug: SoundManager::getDeviceList
Debug: SoundManager::getDeviceList
Debug: Warning: Creation of the midi queue failed. Operation not permitted
Warning: Requested MIDI mapping file "Akai MPD24.midi.xml" not found; defaulting to "Evolution_Xsession.midi.xml"
Debug: PowerMate: write(): Bad file descriptor
Debug: PowerMate: write(): Bad file descriptor
Debug: m_pHercules init: 0x84221a0
Warning: QThread::start: Thread creation error:
Debug: Midi OK (Workaround not required)
Debug: setupMappings( "/usr/share/mixxx/midi/Evolution_Xsession.midi.xml" )
Debug: loadSettings: 1 0 "SlowFade"
Debug: slotApply crossfader: 1 "SlowFade"
Debug: BpmSchemes::readXML "/home/seamus/.mixxxbpmscheme.xml"
Debug: SoundManager::setupDevices()
Debug: Xwax Vinyl control starting with a sample rate of: 44100
Debug: Building timecode lookup tables...
Allocating 2097152 slots (8192Kb) for 20 bit timecode (Serato 2nd Ed., side A)
malloc: Cannot allocate memory
Warning: QThread::start: Thread creation error:
Debug: Created new VinylControlXwax!
Debug: Xwax Vinyl control starting with a sample rate of: 44100
Debug: Building timecode lookup tables...
Allocating 2097152 slots (8192Kb) for 20 bit timecode (Serato 2nd Ed., side A)
malloc: Cannot allocate memory
Warning: QThread::start: Thread creation error:
Debug: Created new VinylControlXwax!
Debug: SoundDevicePortAudio::open() "0, system"
Debug: m_dSampleRate 44100
Debug: iLatencyMSec: 16
Debug: output channels: 2 | input channels: 0
Debug: iLatencySamples: 1412
Debug: iLatencyMSec: 16
Debug: Opening stream with id 0
Debug: Opened PortAudio stream successfully... starting
Debug: PortAudio: Started stream successfully
Debug: iNumDevicesOpenedForOutput: 1
Debug: iNumDevicesOpenedForInput: 0
Segmentation fault

With realtime disabled in jack, or with "no memory locking" enabled mixxx runs fine. I have ~200MB as the upper limit for memory locking, so im not running out either.
(Other audio programs are running ok btw)

System info:
Using Ubuntu Studio 8.04

$uname -a
>Linux seamus-laptop 2.6.24-19-rt #1 SMP PREEMPT RT Wed Jun 18 16:35:41 UTC 2008 i686 GNU/Linux

$ lsb_release -a
>No LSB modules are available.
>Distributor ID: Ubuntu
>Description: Ubuntu 8.04.1
>Release: 8.04
>Codename: hardy

jack settings:
$ jackd --version
> jackd version 0.109.2 tmpdir /dev/shm protocol 22
$ cat ~/.jackdrc
> /usr/bin/jackd -R -P89 -dalsa -dhw:0 -r44100 -p128 -n3 -S

Any other info required let me know.

[Albert Santoni (gamegod)]

You're definitely out of memory, or JACK has somehow managed to usurp Mixxx's ability to allocate memory. This isn't a bug in Mixxx, but I wish I could help, but I honestly don't know what this memory locking stuff is about. It's definitely causing a problem though.

[seamus1089 (s-holden1)]

Ok, thanks for the advice. Ill repost in mixxx forum, attempt to have a look on the JACK side of things, and let people know if I find a solution.

[seamus1089 (s-holden1)]

Just to confirm for the benefit of anyone googling etc, albert was quite right. It was just insufficient memory, upping the memory lock limit to 500mb fixed the problem.
(Thanks Albert)

Seamus

[Swiftb0y (swiftb0y)]

Mixxx now uses GitHub for bug tracking. This bug has been migrated to:
https://github.com/mixxxdj/mixxx/issues/4984