Mark Cochran | 26 Aug 22:08 2014

yum error updating planetccrma-core

I saw an update to planetccrma-core made available the other day, but get this dependency error from yum when I try to update it.

Error: Package: planetccrma-core-2014.08.24-1.fc20.ccrma.x86_64 (planetcore)
           Requires: kernel-x86_64 = 3.17.17-200.rt9.1.fc20.ccrma.rt

Should that dependency be on kernel 3.14.17 perhaps?

Mark Cochran
PlanetCCRMA mailing list
PlanetCCRMA <at>
Yoann LE BARS | 19 Aug 03:42 2014

[PlanetCCRMANews] What is to become concerning real time in Linux kernel

Hello everybody out there!

	As it is my first post to this mailing list, let me introduce myself

	I am a long time musician (I started in sometime like 1985), mostly
using acoustic instruments. I am also a Unix (including GNU/Linux) user
since quite a while (starting in 1997). Of course, when I need a
computer for music, I use Planet CCRMA.

	The reason I am posting today is I have some question about soft
real-time Linux kernel future.

	During Real Time Linux Workshop in November 2013, developers involved
in the real-time patch have indicated the project will be done in 2014,
“one way or another” (<>). In this
workshop, it has also been said that 95 % of the real-time work was
already upstream.

	The question was how will it be done? In considering that the 95 % is
good enough? In getting the rest of the code upstream – which imply an
important effort and several people involved in it? Something between
these two options?

	As SCHED_DEADLINE has been incorporated into Linux 3.14 kernel
(<>), it seems
to me that it has been decided to get some more code upstream. Still,
the question remains: what way will be the project done? Moreover, as I
am using Fedora 20, I can choose between real-time Linux kernel 3.12
from Planet CCRMA and Fedora’s mainstream 3.15 kernel. What about
real-time capabilities of the 3.15 kernel in contrast with the 3.12
real-time kernel?

	Does anyone have any more information about it? As for my part, I
cannot find anything more.


Don Estabrook | 19 Aug 06:32 2014

Fedora 20 3.12.12-300.rt19.1 with Focusrite Saffire Pro40

Hello -

Recently I've been trying to get a Focusrite Saffire Pro40 working on my HP EliteBook 8540w, currently running Fedora 20 with CCRMA RT kernel.

I eventually pieced together enough information from various web sources to get it working, but the issue I have now is that qjackctl reports xruns - sometimes one every couple seconds, sometimes several minutes apart - even with nothing connected through jack.  If I play something using sndfile-jackplay[*], I notice that, around the time of each xrun, there's a 1- to 2-second gap in sound output (on headphone out 1), during which time the player app also freezes.  I've tried setting large values for frames/period and periods/buffer -- up to 1024 and 64, giving a whopping 1.49 seconds of latency at 44.1 kHz.  While the xrun frequency tends to be higher with smaller latencies (< 10 ms say), the variation is pretty small compared to trial-to-trial variation, so I can't say for sure.  Also I've noticed that after the sound returns after a gap, there's some distortion. including something that sounds vaguely like a high-hat cymbal closing softly, about once a second or so.  It gradually fades away after a few seconds.  (Weird, eh?)

The Pro40 was purchased new from a large on-line retailer at the end of 2013.  It's still running the original firmware (nothing I've read so far said conclusively there was any benefit to upgrading), and it's been connected to Focusrite's MixControl (on a Mac) only once, and that was brief - just to see if MixControl recognized it.

Although it may not be a very useful comparison, I've also used an M-Audio Fasttrack Pro (USB) interface with this same lap-top on and off for months with only an occasional xrun - but that's been mostly on the stock kernel.  With the RT kernel I've yet to see an xrun, although the machine locked up once when I was playing from Ardour3 -- the sound kept playing, but since the UI was unresponsive and I couldn't log in over the network, I resorted to holding the power button down.

ffado-dbus-server throws quite a number of ominous-looking errors at various times, although they don't seem to correlate to the xruns in time.  I can provide output if it would be useful.

ffado-diag output starts with the following:
>  kernel version............ 3.12.12-300.rt19.1.fc20.ccrma.x86_64+rt
>    Preempt (low latency)... False
>    RT patched.............. True
>  old 1394 stack present.... False
>  old 1394 stack loaded..... False
>  old 1394 stack active..... False
>  new 1394 stack present.... True
>  new 1394 stack loaded..... True
>  new 1394 stack active..... True
> . . .
> Hardware...
>   Host controllers:
>46:06.0 FireWire (IEEE 1394) [0c00]: Ricoh Co Ltd R5C832 IEEE 1394 Controller [1180:0832] (rev 06) (prog-if 10 [OHCI])
>        Subsystem: Hewlett-Packard Company Device [103c:1521]
>        Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
>        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
>        Latency: 64 (500ns min, 1000ns max), Cache Line Size: 64 bytes
>        Interrupt: pin A routed to IRQ 20
>        Region 0: Memory at d3101000 (32-bit, non-prefetchable) [size=2K]
>        Capabilities: <access denied>
>        Kernel driver in use: firewire_ohci
>        Kernel modules: firewire_ohci

I can provide the rest of that output if it's useful too.
I'm curious as to why it reports "False" for preempt (low latency) on the CCRMA kernel -- I guess I was thinking that was part of the RT patches, but apparently that isn't the case...

I see FFADO has a new version (2.2.1) out a couple months ago.  I've been thinking about trying that, but what I've read has led me to expect that this interface should work with 2.1.

I'd love to hear what others have experienced.


[*] - I'm pretty sure I built this from source, as I couldn't find it in Fedora or CCRMA repositories.  (I was kind of surprised about that, since these sndfile utilities seem quite useful.)  Anyway the rest of the system is pretty much stock, and up to date a of a couple days ago.
PlanetCCRMA mailing list
PlanetCCRMA <at>
Janina Sajka | 16 Aug 23:06 2014

Something's forcing HDMI as default ALSA device


Something about the Linux kernel appears to me to have changed sometime
after kernel-3.13.10. I'm not certain just exactly when, but all the
kernels Fedora has released since 3.13.10 force the HDMI device as the
default ALSA audio device.

3.13.10 is the last kernel release where my carefully constructed ALSA
audio device ordering identifiers, in /etc/modprobe.d/local.conf,  work as
Any kernel beyond puts the onboard hdmi device as the default ALSA
device, which is simply untenable in my setup. For me it's not even that
I have nothing hanging off my HDMI port, I need the default device for
the Speakup screen reader, and I can't affect that on the Speakup side,
regretably. So, in effect,the system becomes unusable.

I note the CCRMA kernels for Fedora 20 are now also well behind current
Fedora releases and wonder if what I'm seeing is just one part of a
number of changes relating to audio.

Anyone know anything about this? Is there possibly a bug here that
should be tagged sooner rather than later? This is a 3-month old issue





Janina Sajka,	Phone:	+1.443.300.2200
			sip:janina <at>
		Email:	janina <at>

Linux Foundation Fellow
Executive Chair, Accessibility Workgroup:

The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI)
Chair,	Protocols & Formats
	Indie UI
Brian Fay | 15 Aug 16:00 2014

Fedora 20 real time kernel crashes


I'm wondering if anybody else frequently experiences crashes/hangs running the real time kernel on Fedora 20? I'm using the default nouveau driver, and I can't seem to use the kernel reliably.

I'm almost guaranteed to get a crash when waking up from suspend - I'll see a display and I'll be able to move the mouse, but the system will be otherwise unresponsive (no capslock light).

Or, occasionally during a desktop session the screen will freeze and I will be unable to interact at all (audio still seems to play for a while).

Is anybody experiencing similar issues, and are there any thoughts on solutions?

Brian Fay
PlanetCCRMA mailing list
PlanetCCRMA <at>
Tom Poe | 30 Jul 14:00 2014

which Centos5 distro to use?

Hello:  I have a Dell c521 desktop computer that I would like to install 
planetccrma on.  Seems like the Centos5 distro is best suited for this 
purpose, but there are several versions of the Centos5 iso to choose 
from.  Can anyone recommend which one to download, place on a 
thumbdrive, and install on the desktop?
Thanks, much
Nils Tonnätt | 30 Jun 19:05 2014

recent version of linuxsampler


I noticed that the version of linuxsampler in PlanetCCRMA is outdated. 
What is needed to get a more recent version in the repository? I hope 
that some bugs with my samples are already fixed.

Best regards,
Nils Tonnätt
Donald Steven | 30 Jun 18:24 2014


I've installed faust, llvm and a number of  related software, but I 
can't find libfaust.a without which I can't seem to enable the faust 
opcodes in csound6 or csound6-devel.

1. Are the faust opcodes required by csound in order for faust to work?  
(I'm just new to faust and would like to try it)?
2. How do I get cmake to enable the faust opcodes?  I'm using: #cmake 
-DFAUST_LIBRARY=$HOME/lib/faust/libfaust.a ./

The $HOME/lib/faust/libfaust.a is a non-issue.  There is no file 
libfaust.a anywhere on my system.

3. How do I get libfaust.a successfully installed in /usr/lib or 

Many thanks for any help anyone can provide.

Thierry Coduys | 26 Apr 14:36 2014

LoMus 2014 | Submissions was extended to 02 May 2014

Submissions was extended to 02 May 2014

Appel à contribution / Call for participation
Désolé en cas d’envois multiples / Sorry for possible crossposting

LoMus 2014

À la recherche des logiciels libres pour la création sonore et intermédias

Pour sa cinquième édition, LoMus 2014 s’adresse à tous ceux qui développent des logiciels libres musicaux ou qui peuvent contribuer au processus de la création musicale.

Un prix sera remis aux logiciels qui font preuve non seulement d’innovation, mais notamment d’inventivité face aux enjeux actuels de la création musicale.

31 mars 2014 - Appel à soumissions
02 mai 2014 - Date limite de soumission des logiciels
16 mai 2014 - Notification d'acceptation
23 mai 2014 - Remise du prix lors des JIM 2014

LoMus 2014

In search of open-source software for musical and intermedia creation

For its fifth edition, LoMus 2014 invites music and audio open-source software creators to submit original projects that either directly or indirectly contribute to musical creation.

A prize will be awarded to open-source sofware that proves to be not only innovatory but also inventive in the present context of music and audio creation.

Mars 31, 2014 - Call for submissions 
May 02, 2014 - Submission deadline 
May 16, 2014 - Admission notification 
May 23, 2014 - JIM Awards Ceremony 2014
PlanetCCRMA mailing list
PlanetCCRMA <at>
Edgar Berdahl | 26 Apr 00:21 2014

Stompbox Design Workshop at CCRMA (August 4-8)

Dear all,

We are pleased to announce the Stompbox Design workshop happening this summer at Stanford 
University! We will be using the Raspberry Pi.

Looking forward,
Edgar Berdahl and Esteban Maestre


Stompbox Design
CCRMA Summer Workshop at Stanford University

Instructors: Edgar Berdahl and Esteban Maestre
August 4-August 8, 2014
9AM-noon, 1PM-5PM

Come design your own flavor of audio effect at Stanford University&#700;s Stompbox Design workshop. We 
will teach you a new platform for implementing audio effects in hardware that not only enables you to 
re-create sound effects from the past but also promotes the creation of new sound effects. Our 
resources for new media design are also at your fingertips if you would like to re-design the concept of 
a guitar stompbox: for example, put a Wiimote on your guitar and use it to change the sound of the 
effect, or use some LEDs to simulate the glow of vacuum tubes or fire, or even incorporate a micro-sized 
Pico projector into your project! The expanded world of digital audio effects is at your fingertips
the workshop incorporates:
 • Most basic theory of signal processing
 • Overview of Digital Audio Effects and operation of stompboxes
 • Laboratory exercise teaching you how to program your stompbox kit from computer by means of 
basic building blocks in PureData, Faust, or C++
 • Design of human-computer interfaces for projects
 • Overview of open-source hardware and software platforms
 • Invited guest lecture by Jonathan S. Abel
 • Further discussion of open-source hardware and software platforms including Satellite CCRMA, 
Raspberry Pi, Arduino and Pure Data
 • Finally, we will round out the workshop with a roundtable presentation of your projects followed by an 
optional evening jam session for fun.

This workshop is intended for: 
Musicians interested in exploring new possibilities in digital audio effects in a hands-on and technical 
way; Makers, engineers, computer scientists, or product designers interested in exploring artistic
for their talents and collaborating with musicians; and/or anyone looking to gain valuable skills in
audio signal processing and human-computer interfaces, with a focus on invention.

Participants are encouraged (but by no means required) to bring their own laptop computers (with 
Ethernet support) and/or musical instruments. Each participant should bring a pair of headphones, 
preferably although not necessarily with 1/4" (5.08mm) audio connector.

*NOTE: Participants have the option of purchasing a $189 lab kit at the end of the workshop. The kit 
contains Satellite CCRMA featuring Arduino and the Raspberry Pi as well as knobs, buttons, footswitches, 
some other sensors, and an acrylic enclosure. Interested participants can customize the template for the 
enclosure and laser-cut their own enclosure using a mail-order service.

To sign up or for more information, please visit
Bion | 23 Apr 10:09 2014

Re: NVIDIA and RT Kernel (Fedora 20)

Thank you for providing steps for successful configuration, Steve!

On Tue, Apr 22, 2014 at 9:21 PM, Steve Duell <steveduell <at>> wrote:
Wow, guys - you did it!  Finally, after over 2 weeks, I've got an RT kernel working with the proprietary NVIDIA drivers.  Thanks to you, I can now have super low latency AND lots of windows open with no screen corruption.  Well done!

When I tried to do this on my own, I had to bounce back and forth between various sources, sometimes reading long threads and hunting down specific information in each one.  Now, in order to save fellow adventurers on this path from future headaches, I'd like to review the steps I took in one clean, step-by-step procedure:

1) Install Fedora

2) Make downloads more efficient by installing a plugin for yum:

yum install yum-plugin-fastestmirror
yum update

3) Add the RPM Fusion repositories so you can have access to necessary packages outside of official Fedora ones:

su -c 'yum localinstall --nogpgcheck$(rpm -E %fedora).noarch.rpm$(rpm -E %fedora).noarch.rpm'

4) Make sure you have basic software development packages
installed so you can compile stuff: 

yum groupinstall "Development Tools"

5) Uninstall any other type of nvidia and/or noveau driver. Search theses terms in the package manager and remove any installed packages.

6) While still running stock kernel, install the most recent nvidia akmod:

sudo yum install akmod-nvidia.x86_64 xorg-x11-drv-nvidia-libs.x86_64

   (note: replace the x86_64 with i686 if you're running in 32bit.  You'll also need to install the xorg-x11-drv-nvidia-libs.i686 alongside x86_64 if you want to run 32bit opengl apps.)

7) Reboot and verify that the nvidia drivers work with the stock kernel.  It may take a while for it to finish booting, as the driver is compiled from the akmod at boot time. 
Possible troubleshooting:
The akmod installation scripts should disable noveau so if you run into problems, like a blank screen on startup, go to a virtual terminal by hitting CTRL+ALT+F2, login and do a "lsmod |grep noveau" (without the quotes) and see if there's a module called noveau and if there is you have to get rid of it.  If that command shows nothing than do a "lsmod | grep nvidia" and if that doesn't it means the nvidia module either didn't compile or is not loading.

8) Check to see if NVIDIA installed cleanly.  You should have access to 'Applications > Settings > Nvidia X Server Settings' from the application launcher.  Or, if you want to check in the terminal, you can:

lsmod |grep nvidia

which should output something like this:
nvidia               9665481  45
drm                   230882  4 nvidia
i2c_core               28502  2 drm,nvidia

9) Now, you can get ready to install the realtime kernel.  First, add the ccrma repositories:

rpm -Uvh

note: the above command is only for Fedora 20. Commands to add repositories for other versions can be found at:

10) Through the package manager, install "Planet CCRMA core packages (kernel and drivers)".  Take note, as it's installing, whether the kernel is PAE (physical address extension) or not, as you'll need to know for the next step.

11) Install either the kernel-rt-devel or kernel-rtPAE-devel package which matches the CCRMA kernel you just installed.

12) Reboot and give the system some extra time to compile the NVIDIA drivers.  After a while (about 3 minutes in my case), the system should finish booting and you'll have access to the Nvidia X Server Setings as described in step 8.


Good luck.

Thanks again for all the help in getting my install finished.  MUCH appreciated!


On Tue, Apr 22, 2014 at 1:53 PM, Jeremy M Booth <bq20 <at>> wrote:
Also, the kernel and kernel-devel packages installed have to match, so
if CCRMA installed kernel-rtPAE by default and you installed
kernel-rt-devel manually (which is for the non-PAE kernel), the akmod
won't compile.

On Tue, 22 Apr 2014 08:03:25 -0700
Steve Duell <steveduell <at>> wrote:

> Jeremy,
> Thanks indeed for the detailed steps.  Here's how it went:
> 1. make sure you have basic software development packages
> installed so you can compile stuff
> (wasn't sure what I needed, so I just installed gcc c++ package)
> 2.  install the kernel-devel packages for stock kernel
> (no problem)
> 3. uninstall any other type of nvidia driver and/or noveau
> (I did this through the package manager - appeared to do the job)
> 4. install nvidia akmod while running stock kernel
> (check)
> 5. reboot and make sure akmod working first with stock kernel
> (Nvidia drivers built successfully & working great on the stock
> kernel.)
> 6. once it does then install the CCRMA stuff/realtime kernel but make
> sure you also install the kernel-rt-devel* BEFORE you reboot.
> (done)
> Upon rebooting into the CCRMA kernel, the bootloader animation made it
> almost to the end before the system started running through a
> checklist.  I saw in that list that it was attempting to build from a
> kmod package. Other entries appeared after that, and then the system
> hung on the message "starting Avahi DNS something...".  I waited 5
> minutes with no activity, then went to the virtual terminal and ran
> the grep commands.  They returned no nouveau or nvidia modules.  So I
> guess it's as you say - the nvidia drivers for this kernel didn't
> compile.
> I have some idea about why this may be:
> - Even though I attempted to install the regular i686 kernel package,
> I noticed that the package manager gave me a PAE version.  I don't
> think I need this, as I'm pretty sure I don't have 4 gigs of RAM. Is
> it possible to force the installation of the non-PAE version through
> the terminal?
> - Maybe there is another development package I need besides the gcc
> one?
> Anyway, thanks again for the help - you've helped me get farther
> along in the process than I've been able to so far on my own.  Much
> appreciated!
> -Steve
> On Mon, Apr 21, 2014 at 6:30 PM, Jeremy M Booth <bq20 <at>>
> wrote:
> > If it helps to keep track of things to reinstall that's fine but
> > technically you shouldn't have to do that.  All the akmod is,
> > is the proprietary nvidia driver repackaged, so unless your machine
> > is over a decade old it shouldn't take more than 3 or 4 minutes.
> > Some akmods compile right after you install them so I guess just pay
> > attention to what's going on before you reboot as well. Also every
> > time you install a new kernel it will have to recompile itself,
> > which it does automatically. You need the kernel devel packages
> > first, because the akmod needs them to compile. Otherwise I'm not
> > sure whether it warns you about the dependencies or not.  I don't
> > see why any spin would hurt so long as it's still fedora.  So if
> > you're installing from scratch this is what I'd do:
> >
> > first to make downloads more efficient:
> >
> > yum install yum-plugin-fastestmirror
> > yum update
> >
> > 1. make sure you have basic software development packages
> > installed so you can compile stuff
> > 2.  install the kernel-devel packages for stock kernel
> > 3. uninstall any other type of nvidia driver and/or noveau
> > 4. install nvidia akmod while running stock kernel
> > 5. reboot and make sure akmod working first with stock kernel
> > 6. once it does then install the CCRMA stuff/realtime kernel but
> > make sure you also install the kernel-rt-devel* BEFORE you reboot
> >
> > Possible troubleshooting:
> > The akmod installation scripts should disable noveau so if you run
> > into problems, like a blank screen on startup, go to a virtual
> > terminal by hitting CTRL+ALT+F2, login and do a "lsmod |grep
> > noveau" (without the quotes) and see if there's a module called
> > noveau and if there is you have to get rid of it.  If that command
> > shows nothing than do a "lsmod | grep nvidia"
> > and if that doesn't it means the nvidia module either didn't compile
> > or is not loading.
> >
> > This setup definitely works though as I've been using the
> > fedora/ccrma/akmod combination for several releases now, but like
> > you said the information on the web is very spotty feel free to
> > report back with any problems.
> >
> > On Mon, 21 Apr 2014 16:41:51 -0700
> > Steve Duell <steveduell <at>> wrote:
> >
> > > Thank, Jeremy!  I'll reinstall F20 from scratch and follow your
> > > steps. Just a few more questions:
> > >
> > >
> > > Should I install the akmod or the kernel-rt-devel package first?
> > > Or does it matter?  I gather that as long as they're both present
> > > at reboot, everything should come out fine.
> > >
> > > You mention a delay on first boot after installing the akmod.
> > > Can you estimate about how long it should take?  It's possible
> > > that the first time I tried this, I just gave up too soon.
> > >
> > > Also, is there any harm in using the CCRMA kernel with the Fedora
> > > Jam spin?  After weeks of trying to get various distros to play
> > > nice with my hardware, I'm to the point where I'd gladly accept
> > > one with most of the workhorse audio utilities pre-packaged for
> > > my convenience.
> > >
> > > Thanks again for the prompt reply.
> > >
> > > -Steve
> > >
> > >
> > > On Mon, Apr 21, 2014 at 11:21 AM, Jeremy M Booth
> > > <bq20 <at>> wrote:
> > >
> > > > Also make sure you have the kernel-rt-devel* package installed
> > > > for the akmods to compile.
> > > >
> > > >
> > > >
> > > > On Sun, 20 Apr 2014 21:29:43 -0700
> > > > Steve Duell <steveduell <at>> wrote:
> > > >
> > > > > Hey Folks,
> > > > >
> > > > > I'm new to linux, and have gone through several distributions
> > > > > over the last few weeks, looking for the right one.  I've
> > > > > tried and am very happy with Fedora 19 & 20 (both Xfce and
> > > > > LXDE editions). I've had no problems installing the RT
> > > > > kernel, but running the nouveau driver for graphics doesn't
> > > > > work well with my NVIDIA Geforce 8 chip. Once enough windows
> > > > > are open, or enough images load on a webpage, the display
> > > > > gets corrupted and I can't continue to use the computer
> > > > > without a reboot.
> > > > >
> > > > > I tried installing the proprietary drivers, which work fine
> > > > > on a stock kernel, but the CCRMA RT kernel will not boot
> > > > > after the nvidia drivers are installed.
> > > > >
> > > > > I've googled the issue pretty deeply, and I see lots of
> > > > > conflicting information about the exact steps to get the
> > > > > nvidia driver working with F20.  I'm hoping that one of the
> > > > > power-users on this list has already surmounted this issue
> > > > > and can walk a newbie through it, step-by-step.
> > > > >
> > > > > I'm perfectly happy to start all over with a clean install of
> > > > > fedora (any version) if it will make life easier for me or
> > > > > you.
> > > > >
> > > > > Thanks a ton,
> > > > >
> > > > > Steve
> > > >
> > > >
> >
> >

PlanetCCRMA mailing list
PlanetCCRMA <at>

PlanetCCRMA mailing list
PlanetCCRMA <at>