Re: [Lmuse-user] Re: Re: Can't compile current CVS

[Robert J. <rj...@sp...>]

Hello again,

> >
> > Are they really sigsegv as you wrote or does it crash differently? Most
> > common apart from sigsegv are lockups that cause one of the watchdogs to
> > trigger.
>
> I'm not sure about that. I hope the debug-messages from muse will help you
> a little bit to answer this question.
>
> torben@x3m-rechner:~$ muse -D
> Start euid: 1000 ruid: 1000, Now euid 1000
> init Jack Audio
> init Jack Audio: register device
> init Jack Audio: register device
> init Jack Audio: register client
> JACK: sample rate changed: 48000
> global lib:   </usr/local/lib/muse>
> global share: </usr/local/share/muse>
> muse home:    </home/torben>
> project dir:  <./>
> scan ladspa plugin dir </usr/local/lib/muse/plugins>
> scan ladspa plugin dir </usr/lib/ladspa>
> scan ladspa plugin dir </usr/local/lib/ladspa>
> QtLibraryPath:
>   </usr/local/lib/muse/qtplugins>
>   </usr/lib/kde3/plugins/>
>   </home/torben/.kde/lib/kde3/plugins/>
>   </usr/lib/qt3/plugins>
>   </usr/local/bin>
> load instrument definitions from </usr/local/share/muse/instruments>
> READ IDF /usr/local/share/muse/instruments/Roland_FantomXR.idf
> READ IDF /usr/local/share/muse/instruments/Roland-XP30.idf
> READ IDF /usr/local/share/muse/instruments/Roland-SCD70.idf
> READ IDF /usr/local/share/muse/instruments/Access_Virus.idf
> READ IDF /usr/local/share/muse/instruments/emuproteus2000.idf
> READ IDF /usr/local/share/muse/instruments/Yamaha-PSR530.idf
> READ IDF /usr/local/share/muse/instruments/Yamaha-PSR275.idf
> READ IDF /usr/local/share/muse/instruments/Alesis-QS-78R.idf
> READ IDF /usr/local/share/muse/instruments/Yamaha-S90.idf
> READ IDF /usr/local/share/muse/instruments/AlesisQS6.idf
> READ IDF /usr/local/share/muse/instruments/ns5r.idf
> READ IDF /usr/local/share/muse/instruments/xg.idf
> READ IDF /usr/local/share/muse/instruments/Roland_SRX-09.idf
> READ IDF /usr/local/share/muse/instruments/MC303.idf
> READ IDF /usr/local/share/muse/instruments/MC505.idf
> READ IDF /usr/local/share/muse/instruments/Roland-E28.idf
> READ IDF /usr/local/share/muse/instruments/gs.idf
> READ IDF /usr/local/share/muse/instruments/gm.idf
> READ IDF /usr/local/share/muse/instruments/Waldorf_Microwave-I.idf
> READ IDF /usr/local/share/muse/instruments/Hammond_XB-1.idf
> READ IDF /usr/local/share/muse/instruments/Roland_SRX-02.idf
> READ IDF /usr/local/share/muse/instruments/Yamaha-P50m.idf
> READ IDF /usr/local/share/muse/instruments/ZynAdd-1_4.idf
> initMidiAlsa
> ALSA port add: <Midi Through Port-0>, 62:0 flags 3 0x63
> ALSA port add: <M Audio Delta 1010LT MIDI>, 64:0 flags 3 0x7f
> ALSA port add: <qjackctl>, 128:0 flags 1 0xc2
> Start thread Midi with priority 80
> Start thread Disc with priority 0
> QObject::connect: No such signal PartCanvas::horizontalScroll(int)
> QObject::connect:  (sender name:   'unnamed')
> QObject::connect:  (receiver name: 'unnamed')
> Arranger::configChanged - no bitmap!
> searching for software synthesizer in </usr/local/lib/muse/synthi>
> 7 soft synth found
> starting with default template
> Arranger::configChanged - no bitmap!
> AlsaTimer::setTimerTicks(): requested freq 1024 Hz too high for timer (max
> is 1000)
>   freq stays at 1000 Hz
> watchdog set to SCHED_FIFO priority 99
> Thread <Midi> set to SCHED_FIFO priority 80
> Thread <Disc> set to SCHED_OTHER priority 0
> JACK: graph changed
> JACK: graph changed
> JACK: graph changed
> unknown kbAccel 0x1023
> JACK: graph changed
> JACK: graph changed
> fluidsynth sampleRate 48000
> AudioNode::setRecordFlag1: create internal
> file /home/torben/Muse-Songs/rec103.wav
> fluidsynth sampleRate 48000
> Speicherzugriffsfehler
>
>
> "Speicherzugriffsfehler" means something like "Segmentation fault".

Yes, I see. Though I still think this is related.

>
> > What jack-buffers do you use?
>
> 64 Frames/Period

Quite small yes. I bet it works better if you try and start the project with 
bigger buffers.
>
> > There was a problem with thread syncronization while loading projects. I
> > experiemented lately with this and using 32-frames I could reliably lock
> > up muse everytime, since pre4 it does not anymore.
>
> I compiled pre4 yesterday and it worked only 10 seconds. It did not matter
> what I did within these 10 seconds... Muse ended only with
> "Speicherzugriffsfehler". So I installed pre3 again. (tar.gz from Homepage,
> not from CVS, thats why I have problems with compiling the source)

pre4 had a rather annoying bug that could lead to stuff like this. It used RTC 
even though RTC was not available. I can well imagine that leading to 
crashes.

>
> > > > Loading big soundfonts is known to cause xruns, but there are other
> > > > ways to accomodate for this.
> > >
> > > But sometimes these xruns doesn't matter.
> >
> > Indeed. I could not care less if it xruns during loading. Actually the
> > only time I really really care is during recording. Others may have
> > different opinions.
>
> I'll raise the timeout from 500ms to 1000 and have a look if this solves
> the problem..

As far as I remember that option is in microseconds (god knows why), set it to 
atleast 100000.

>
> > What I meant to write was rather that muse does not meet it's
> > jack-deadline. Not exactly the same, and since MusE is kicked out it is
> > quite a bit more disturbing. Since pre4 muse should be handle being
> > kicked out.
> > Still it's probably a good idea to adjust jack to allow for "late"
> > clients. e.g.
> > jackd -timeout 100000
>
> Yes, we'll see.
>
> > Regards,
> > Robert
>
> Thank you again,

No problem!

Regards,
Robert


-- 
http://spamatica.se/musicsite/