[PD-dev] [ pure-data-Patches-1942258 ] jack fixes and enhancements

SourceForge.net noreply at sourceforge.net
Sun Jul 25 06:17:11 CEST 2010


Patches item #1942258, was opened at 2008-04-14 10:19
Message generated for change (Comment added) made by millerpuckette
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1942258&group_id=55736

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: puredata
Group: bugfix
>Status: Closed
>Resolution: Duplicate
Priority: 7
Private: No
Submitted By: IOhannes m zmlnig (zmoelnig)
Assigned to: Miller Puckette (millerpuckette)
Summary: jack fixes and enhancements

Initial Comment:
the jack-implementation in Pd has several flaws:
- maximum number of ports = 32 (hardcoded)
- Pd crashes when you request more than 32 channels
- jackd has to run before Pd is started
- Pd doesn't like jackd quitting
- maximum of 2 parallel Pd's connecting to jack
- ...


i believe that the attached patch fixes most of these issues:
- raises the maximum number of ports to 1024 (this is the maximum number of ports qjackctl allows me to select for the jackd)
- when initializing ports, Pd checks whether a port was successfully created; if not it doesn't add more ports but leaves this as the maximum it could get from jackd (no more crashes with high number of ports)
- Pd automatically starts a default jackd if none is running yet
- when the jackd quits while Pd is running, it sets the jack_client to 0, efficiently preventing hangs. jackd is restarted automatically when selected via the audio-settings
- Pd tests the return status when trying to connect to jackd; if it finds that the name is already given it tries alternative names; if initialization failed for other reasons, it just fails.

i have tested this on my debian unstable/testing system that provides libjack-dev 0.109.2-1.1


----------------------------------------------------------------------

>Comment By: Miller Puckette (millerpuckette)
Date: 2010-07-24 21:17

Message:
I believe a later version of this patch got taken.

----------------------------------------------------------------------

Comment By: Rich E (reakin)
Date: 2008-05-23 12:13

Message:
Logged In: YES 
user_id=1586621
Originator: NO

This patch makes pd unusable while running jack in realtime with the
following system:

Ubuntu Gusty, 
r at pal:~$ uname -a
Linux pal 2.6.22-14-rt #1 SMP PREEMPT RT Tue Feb 12 09:57:10 UTC 2008 i686
GNU/Linux

any version of pd-extended autobuilds from April 16th onwards. Pd-vanilla
does not have the same problem.

Exact sound symptons are:
Periodic glitches in audio, which is occasionally corrected when another
application is attached to jack (probably resetting the audio chain), but
the glitches return shortly.  It only occurs when running jack in real-time
mode.

Here is a posting on the pd-list where I was discussing this issue:
http://lists.puredata.info/pipermail/pd-list/2008-05/062449.html

----------------------------------------------------------------------

Comment By: Hans-Christoph Steiner (eighthave)
Date: 2008-04-15 14:29

Message:
Logged In: YES 
user_id=27104
Originator: NO

checked this patch into branches/pd-extended/0.40:

http://pure-data.svn.sourceforge.net/viewvc/pure-data?view=rev&revision=9702

----------------------------------------------------------------------

Comment By: IOhannes m zmlnig (zmoelnig)
Date: 2008-04-14 10:21

Message:
Logged In: YES 
user_id=564396
Originator: YES

i raised priority a bit, as the current implementation makes Pd unusable
with highly multichannel systems (e.g. using 64 channels) and jack

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1942258&group_id=55736



More information about the Pd-dev mailing list