Linux-audio-dev June 2004

linux-audio-dev@lists.linuxaudio.org

99 participants
93 discussions

by Juhana Sadeharju

>From: Paul Davis <paul(a)linuxaudiosystems.com> >>> >>>> you cannot modify the graph in JACK while the graph is being used to >>>> process audio. you do not know how long the graph modification will >>>> take if you try to do it (for example) right after you're done with >>>> one process cycle. the only sure way to do this is to use lock free >>>> parallel programming techniques. >> >>Would anyone please explane details of these techniques? > >No. Google for "herlihy lock free". Dozens of references. Papers range >from about 12 years ago to this year. I did mean the details on the graph modification. Up so far I have not understood what is the problem in modifying the graphs in the engine. It sounded like you solved the problem using lock-free par prog techniques. By reading the solution I could catch up the problem. ;-) Want explane what kind of modifications you mean? Why they take so much time? Juhana

19 years, 10 months

[linux-audio-dev] OT: priority inversion on Mars

by Lee Revell

Apologies if everyone here has seen this, it's not really linux or audio related, but I think people will find it interesting and somewhat relevant: http://research.microsoft.com/~mbj/Mars_Pathfinder/Mars_Pathfinder.html Lee

19 years, 10 months

[linux-audio-dev] lock-free data structures

by Joshua Haberman

Greetings, all. I have not been subscribed to this list for several months now, but someone pointed me to a link to Paul's message that I quote below. I realized that now is a good time to check back in, see what everyone is up to, and let you know what I'm up to. A few days ago Paul posted: > the only sure way to do [live graph modification] is to use lock free > parallel programming techniques. this is still a highly experimental > area of computer science, and i have yet to see more than a glimmer of > a technique that would actually work in a complex real time system > like JACK. Right now I am very interested in lock-free data structures, and I am working with Ross Bencina (of PortAudio and AudioMulch) to implement them in an open source library. See [0] and [1]. It turns out there are several practical algorithms being published in this field, most notably by Maged Michael from IBM research. Ross and I have been in contact with him for advice and clarification about his papers, which are available on his web site [2]. We have some working code, but since all this stuff is highly architecture-specific we need to do more research. A significant part of the library will probably end up being in assembly. For me, this is part of a bigger effort: rewriting Audacity's back-end, and making it a GUI-independent library. We're tentatively calling this library "Mezzo" -- more about Mezzo's goals and basic API is at [3]. Mezzo's data representation scheme is just a refactoring of what Audacity uses now, but the real-time system will be completely redesigned to be similar to JACK's model: several nodes with input ports and output ports that implement callbacks. The whole graph will be run from the audio api's callback. Getting back to lock-free data structures, my intention is not to use locks anywhere in this real-time system. Since I haven't written the whole thing yet I'm not sure if this is completely possible, but here is my general strategy: 1. signal graph is constructed in main thread 2. once audio starts, the audio thread runs the graph once for every callback 3. if the main thread wants to change the graph while it's running, it does any heavy lifting (memory allocation, etc), then sends a message to the audio thread with a lock-free queue (ex: "add this node and connect it to this other node"). 4. buffers are passed between the disk thread and the audio thread using lock-free queues 5. in the real-time thread, buffers are allocated from and returned to a lock-free freelist. For example, if you have a node that generates a sine wave, it needs to allocate a buffer when it runs to put its data in. Instead of calling malloc, it asks for a free buffer from the freelist. The real-time code is still in its beginning stages, but I have gotten it to play data from disk by sending it from a disk thread to an audio thread. You can see the code in our SVN repository. [4] I'm also keeping a log of my activity and my thought process on my web site [5]. I am interested in any reactions people may have to this work. I drew a lot of the inspiration for Mezzo's real-time system from JACK and other things I have read on this list. Regards, Josh Haberman [0] http://www.audiomulch.com/~rossb/code/lockfree/ [1] http://www.audiomulch.com/~rossb/code/lockfree/spec.txt [2] http://www.research.ibm.com/people/m/michael/pubs.htm [3] http://www.audacityteam.org/wiki/index.pl?Mezzo [4] http://mezzo.homelinux.net/svn/ [5] http://www.reverberate.org/log/

19 years, 10 months

[linux-audio-dev] Multithreaded programming for a poll model?

by Mario Lang

Hi. I am currently trying to figure out how to restructure the code of my little yatm tool to be flexible and extensible in the future. Basically, I'd like to be able to become a JACK client at some point in the future. Since JACK uses a poll model, I have to majorly restructure my code. I think what I need is the following: Three threads, one for the UI, one for the MPEG/Vorbis/Speex decoder, and one for audio output (to either libao or JACK). I'll also need a buffer between the decoder and the audio thread, and some way to signal the decoder thread that more audio data is needed. Additionally, I'd like to have the ability to make the buffer in between remember the last n seconds of audio data and make it possible to jump back in time <=n seconds during playback. I've looked at the lock-free ringbuffer of JACK, but I am not sure if this is really what I need here. I am very new to linux audio programming, and especially to pthreads (wrote my first thread using program two days ago). I'd be very happy if some of you experts here could give me a hand here and point me into the right direction. Maybe there is already code out there which does roughly that which I can look at? Or maybe you can suggest which mechanisms of pthreads I should be using for this kind of scenario. -- CYa, Mario | Debian Developer <URL:http://debian.org/> | Get my public key via finger mlang(a)db.debian.org | 1024D/7FC1A0854909BCCDBE6C102DDFFC022A6B113E44

19 years, 10 months

[linux-audio-dev] Linux soundapps Euro site back on-line

by Dave Phillips

Greetings: The European mirror of the Linux soundapps site is back online with a new address: http://linuxsound.atnet.at Please update your bookmarks. This site is now in sync with the US and Japanese sites. Best regards, dp

19 years, 10 months

[linux-audio-dev] jack in a live situation

by Dave Griffiths

Hi all, A question regarding jack use in a live setting. As far as I can tell the soft mode only works with non realtime jack. What should I do if I want realtime performance, but a really forgiving server, so dropouts never cause the server to timeout. It's more important to me to have continuous rather than glitch free audio (I never ever want everything to shut down) If the answer is ultimately to write better jack clients - thats Ok ;) cheers, dave

19 years, 10 months

[linux-audio-dev] Project Ideas

by michael tewner

OK. I have permission to get one of these programs moving with some constraints (appended). WHich do you guys think would be most useful (publishable :-) )? 1. The OpenGL Spectrogram implemented into Steve Harris's meterbridge 2. A 31-BAND (or more?) graphic EQ These would both be implemented into JACK My advisor requested these specs: 1- This should be run on any platform including PC platform. <Yeah, right. Perhaps I"ll just implement the code, and wrap it with JACK and WIN-API with preprocessor commands....Grrr...> 2- The interface would be build with a Tcl/Tk script which is compatible with SGI/ Aplle and PC platform. <I'm just not sure he knows what he's talking about...Anyway, I think he just wants an interface that will work on linux and windows...again, see #1> 3- The FFT algorithm should be run in parallel and optimize at least for two platforms. <I guess once we're splitting up the spectrum, this shouldn't be a problem...BTW,the advisor teaches an MPI course. This idea probably came from this...I'm not sure what he means by "2 platforms."> 4- We can use a Wavelet Transformation inside of FFT and to parallize it. <*shrug* I'm gonna hafta do some research. wikipedia says it's like a FFT running in O(N) instead of O(N log N): http://en.wikipedia.org/wiki/Wavelet_transform > I can recomend against some of these. I just think he wants me to do some extra work... Granted, without these recomendations, I'm not looking at the suggested 500 Hours of work... This project is starting to look big and clunky: 1. Implementing both windows and linux 2. Parrellizing the code? I don't mind writing multithreaded or MPI... it just looks like overkill.... 3. Can JACK compile on IRIX? That's his OS of choice (Being his only *nix experience...) and, well, SGI's are fun! Perhaps I can convince him to put linux on his old O2. Anyway, bottom line, it'll be fun, and I don't mind spending my time on this, if ANY part of it will be of use.... ...I'de just hate to kill Steve's code, though. Anyway, suggestions? Thanks y'all! -Mike

19 years, 10 months

Re: [linux-audio-dev] Is ladspa actually la-dsp-a? Is JACK the ultimate solution?

by eviltwin69＠cableone.net

Hey, I wouldn't mind working on the graphics, I just don't know where to start or who else is working on it. Jan On Tue, 8 Jun 2004 09:18 , Steve Harris <S.W.Harris(a)ecs.soton.ac.uk> sent: >On Tue, Jun 08, 2004 at 11:45:53 +0200, Marek Peteraj wrote: >> VST plugins tend to be rather complex, offering tons of features and >> eyecandish GUIs, while LADSPAs usually offer limited functionality, no >> GUI at all(hosts usually provide simple ones to control the parameters). >> But what's interesting is that each LADSPA plugin usually implements >> exactly one type of DSP technique, for example, an oscillator, or a >> delay. This basically leads to a situation where a certain DSP technique >> is 'isolated' in a separate plugin. > >I think thats down to two factors (and its not a good thing) > >1) LADSPA developers are few in number and short in time. The basics are a > good place to start. > >2) The lack of a UI standard makes complex plugins a bit pointless. > >There are a few counter examples (e.g. my VyNil plugin wraps a lot of >different bits), and infact if you look in many LADSPA plugins you will >see theres really more going on than there appears to be. > >[OT] - my canned plugin writing experience - all generalisations and IMHO > of course > > Time breakdown: 10% writing code, 10% maths and optimising, 80% tweaking > and tuning. > > Mapping the controls 1:1 with DSP parameters makes plugins crap - people > say they want that if you ask them, but they dont mean it ;) > > Fewer controls is better. > > Affordance, appearance and usability has as much affect on the perceived > sound quality as the DSP code (posivly and negativly). Some of this can > be achieved without a custom UI. > >You mentioned JAMin - true that does use LADSPA plugins - but of the total >ammount of code the LADSPA plugins are a tiny fraction. I just reused them >because I hate fixing bugs in two places :) > >[OOT] I used to think that a UI spec for LADSPA (to make it competetive >with VST) was a technological problem. I now thinks its a manpower issue >(I think Paul Davis pointed this out a couple of years ago :). Games >develpment has moved to the point where the graphics work is more >expensive than the software development, and I bet its not far off in >plugin / eyecandy app development. We have no, or almost no, graphics >people here. > >There are plenty of graphics people working on Free Software projects, but >they all seem to be working on games projects. What a waste. I guess >drawing goblins is more fun than sliders and LEDs. Who knew? ;) >

19 years, 10 months

[linux-audio-dev] Re: Graph reordering

by Juhana Sadeharju

>From: Alfons Adriaensen <fons.adriaensen(a)alcatel.be> > >On Fri, Jun 11, 2004 at 08:37:09AM -0400, Paul Davis wrote: > >> you cannot modify the graph in JACK while the graph is being used to >> process audio. you do not know how long the graph modification will >> take if you try to do it (for example) right after you're done with >> one process cycle. the only sure way to do this is to use lock free >> parallel programming techniques. Would anyone please explane details of these techniques? Maybe it is something I know already but it is not clear. >The 'heavy state' is AFAICS not the real problem. >Heavy things can be put in place before the actual graph reordering >is done or removed after. This does not need to be done in the RT >thread. For speeding up this approach one can also use caches for the items. The engine butler can initialize module structures, delay lines, etc. at its spare time. These initialized items are placed to cache. When the engine needs an initialized item, it gets the item instantly from the cache. (No patent pending.) Juhana

19 years, 10 months

[linux-audio-dev] GThread vs. pthread

by Pete Bessman

Does anybody have any opinion on which threading system is superior? I've been using glib for a lot of things, but for whatever reason I'm hesitant about using it for threading if the only benefit it will provide is consistency (I'm guessing it's just a wrapper for pthread anyway). [pb]

19 years, 10 months

← Newer
1
2
3
4
5
6
7
8
9
10
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Linux-audio-dev June 2004