Linux-audio-user May 2012

linux-audio-user@lists.linuxaudio.org

113 participants
106 discussions

Frequent/regular xruns with jackd2 & puredata (due to multiple cards ?)

by 0 1

Hello, I have 2 soundcards: % cat /proc/asound/cards 0 [PCH ]: HDA-Intel - HDA Intel PCH HDA Intel PCH at 0xf7e00000 irq 42 1 [Pro ]: USB-Audio - FastTrack Pro M-Audio FastTrack Pro at usb-0000:00:1d.0-1.1, full speed I run jackd on the second one of course. When starting puredata, even if I don't set it up to use JACK, I get this error in dmesg: hda-intel: IRQ timing workaround is activated for card #0. Suggest a bigger bdl_pos_adj. --- Jackd, on its side, produces very frequent and regular xruns. I close puredata, and xruns continue to occur: Jack: ALSA XRun wait_status = 0 Jack: JackRequest::Notification Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackPosixMutex::Unlock res = 1 Jack: JackExternalClient::ClientNotify ref = 2 client = qjackctl name = qjackctl notify = 3 Jack: fPollTable i = 2 fd = 12 Jack: JackClient::ClientNotify ref = 2 name = qjackctl notify = 3 Jack: JackClient::kXRunCallback Jack: **** alsa_pcm: xrun of at least 0.032 msecs Jack: fPollTable i = 1 fd = 11 Jack: ALSA XRun wait_status = 0 Jack: JackRequest::Notification Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackPosixMutex::Unlock res = 1 Jack: JackExternalClient::ClientNotify ref = 2 client = qjackctl name = qjackctl notify = 3 Jack: fPollTable i = 2 fd = 12 Jack: JackClient::ClientNotify ref = 2 name = qjackctl notify = 3 Jack: JackClient::kXRunCallback Jack: **** alsa_pcm: xrun of at least 0.008 msecs Jack: fPollTable i = 1 fd = 11 Jack: ALSA XRun wait_status = 0 Jack: JackRequest::Notification Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackPosixMutex::Unlock res = 1 Jack: JackExternalClient::ClientNotify ref = 2 client = qjackctl name = qjackctl notify = 3 Jack: fPollTable i = 2 fd = 12 Jack: JackClient::ClientNotify ref = 2 name = qjackctl notify = 3 Jack: JackClient::kXRunCallback Jack: **** alsa_pcm: xrun of at least 0.009 msecs Jack: fPollTable i = 1 fd = 11 Jack: JackRequest::Notification Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackPosixMutex::Unlock res = 1 Jack: JackExternalClient::ClientNotify ref = 2 client = qjackctl name = qjackctl notify = 3 Jack: ALSA XRun wait_status = 0 Jack: fPollTable i = 2 fd = 12 Jack: JackClient::ClientNotify ref = 2 name = qjackctl notify = 3 Jack: JackClient::kXRunCallback Jack: **** alsa_pcm: xrun of at least 0.010 msecs Jack: fPollTable i = 1 fd = 11 Jack: ALSA XRun wait_status = 0 Jack: JackRequest::Notification Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackPosixMutex::Unlock res = 1 Jack: JackExternalClient::ClientNotify ref = 2 client = qjackctl name = qjackctl notify = 3 Jack: fPollTable i = 2 fd = 12 Jack: JackClient::ClientNotify ref = 2 name = qjackctl notify = 3 Jack: JackClient::kXRunCallback Jack: **** alsa_pcm: xrun of at least 0.005 msecs Jack: fPollTable i = 1 fd = 11 Jack: ALSA XRun wait_status = 0 Jack: JackRequest::Notification Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackPosixMutex::Unlock res = 1 Jack: JackExternalClient::ClientNotify ref = 2 client = qjackctl name = qjackctl notify = 3 Jack: fPollTable i = 2 fd = 12 Jack: JackClient::ClientNotify ref = 2 name = qjackctl notify = 3 Jack: JackClient::kXRunCallback Jack: **** alsa_pcm: xrun of at least 0.051 msecs Jack: fPollTable i = 1 fd = 11 Jack: ALSA XRun wait_status = 0 Jack: JackRequest::Notification Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackPosixMutex::Unlock res = 1 Jack: JackExternalClient::ClientNotify ref = 2 client = qjackctl name = qjackctl notify = 3 Jack: fPollTable i = 2 fd = 12 Jack: JackClient::ClientNotify ref = 2 name = qjackctl notify = 3 Jack: JackClient::kXRunCallback 17:34:43.873 Récupération de désynchronisation (XRUN) (13 sauté). Jack: **** alsa_pcm: xrun of at least 0.027 msecs Jack: fPollTable i = 1 fd = 11 Jack: ALSA XRun wait_status = 0 Jack: JackRequest::Notification Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackPosixMutex::Unlock res = 1 Jack: JackExternalClient::ClientNotify ref = 2 client = qjackctl name = qjackctl notify = 3 Jack: fPollTable i = 2 fd = 12 Jack: JackClient::ClientNotify ref = 2 name = qjackctl notify = 3 Jack: JackClient::kXRunCallback Jack: **** alsa_pcm: xrun of at least 0.008 msecs Jack: fPollTable i = 1 fd = 11 Jack: ALSA XRun wait_status = 0 Jack: JackRequest::Notification Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackEngine::NotifyClient: no callback for event = 3 Jack: JackPosixMutex::Unlock res = 1 Jack: JackExternalClient::ClientNotify ref = 2 client = qjackctl name = qjackctl notify = 3 Jack: fPollTable i = 2 fd = 12 Jack: JackClient::ClientNotify ref = 2 name = qjackctl notify = 3 Jack: JackClient::kXRunCallback Jack: **** alsa_pcm: xrun of at least 0.009 msecs --- When I disable snd-hda-intel, I seem to be able to run jackd & puredata without trouble. Other audio apps don't produce such results, but I guess puredata makes some tests at startup which produce this issue. I'm unsure where to address this issue, but hope this report can be useful in some way, so if anyone has clues about where to investigate ... Cheers, 01 === Other infos: % uname -a Linux ___ 3.2.0-1-rt-amd64 #1 SMP PREEMPT RT Fri Feb 17 06:10:34 UTC 2012 x86_64 GNU/Linux

12 years, 6 months

Re: [LAU] rtirq - what does it improve, and how can I measure it?

by Ralf Mardorf

On Mon, 2012-05-07 at 14:52 +0200, Kaj Ailomaa wrote: > The network-manager, cron, and other > things we have heard about, when you get periodic xruns no matter which > latency you use. Can we make all of those problems go away by changing > rtprio for devices? No! There are services that we shouldn't run on an audio production machine. It's hard to say what should be disabled and what doesn't matter. For example, I don't think this counts on my machine regarding to performance $ grep noatime /etc/fstab /dev/sda11 /mnt/music ext3 noatime,defaults 0 2 /dev/sdb12 /home/music ext4 noatime,defaults 0 2 but on another machine noatime might have impact to the performance.

12 years, 6 months

Re: [LAU] What Focusright firewire interfaces are fully working and currently available?

by Edward Diehl

The issue with having to start Jack twice for the Focusrite Saffire Pro24 (and probably Pro40 has been fixed since ffado SVN 2062 and also the mixer functionality improved, so I recommend recompiling ffado with the latest SVN.

12 years, 6 months

[solved] new rtirq release

by Ralf Mardorf

The truth is, that it's not completely solved, there's at least the need to switch the config script when using different kernel versions. RTIRQ STILL IS rtirq - Revision 7, resp. $ wget [snip] --2012-05-05 03:09:00-- First the drawback when using 1 config for both kernels, then 3.2.0-23-lowlatency with good prios and then 3.0.30 (lowlatency) with good prios. Using nvidia-current-295.40 with both kernels. Rebooting seems to be important, stop and start rtirq seems not to work. ## REBOOT spinymouse@precise:~$ uname -a Linux precise 3.2.0-23-lowlatency #31-Ubuntu SMP PREEMPT Wed Apr 11 02:24:03 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux spinymouse@precise:~$ /etc/init.d/rtirq status PID CLS RTPRIO NI PRI %CPU STAT COMMAND 92 FF 90 - 130 0.0 S irq/8-rtc0 85 FF 85 - 125 0.0 S irq/18-ohci_hcd 89 FF 84 - 124 0.0 S irq/18-ohci_hcd 681 FF 83 - 123 0.0 S irq/18-snd_hdsp 1198 FF 82 - 122 0.8 S irq/18-nvidia 678 FF 70 - 110 0.0 S irq/20-snd_ice1 685 FF 69 - 109 0.0 S irq/21-snd_ice1 79 FF 60 - 100 0.0 S irq/19-ehci_hcd 81 FF 60 - 100 0.1 S irq/16-ohci_hcd 83 FF 59 - 99 0.0 S irq/17-ohci_hcd 87 FF 58 - 98 0.0 S irq/17-ohci_hcd 91 FF 55 - 95 0.0 S irq/1-i8042 20 FF 50 - 90 0.0 S irq/9-acpi 68 FF 50 - 90 0.2 S irq/22-ahci 263 FF 50 - 90 0.0 S irq/14-pata_ati 264 FF 50 - 90 0.0 S irq/15-pata_ati 290 FF 50 - 90 0.0 S irq/22-firewire 544 FF 50 - 90 0.0 S irq/7-parport0 929 FF 50 - 90 0.0 S irq/43-eth0 3 TS - 0 19 0.2 S ksoftirqd/0 9 TS - 0 19 0.2 S ksoftirqd/1 spinymouse@precise:~$ grep RTIRQ_NAME_LIST /etc/default/rtirq RTIRQ_NAME_LIST="rtc hdspm ICE1712 snd_hdsp snd_ice1 snd usb i8042" ## REBOOT spinymouse@precise:~$ uname -a Linux precise 3.2.0-23-lowlatency #31-Ubuntu SMP PREEMPT Wed Apr 11 02:24:03 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux spinymouse@precise:~$ /etc/init.d/rtirq status PID CLS RTPRIO NI PRI %CPU STAT COMMAND 92 FF 90 - 130 0.0 S irq/8-rtc0 730 FF 85 - 125 0.0 S irq/18-snd_hdsp 712 FF 80 - 120 0.0 S irq/20-snd_ice1 750 FF 79 - 119 0.0 S irq/21-snd_ice1 79 FF 70 - 110 0.0 S irq/19-ehci_hcd 81 FF 70 - 110 0.1 S irq/16-ohci_hcd 83 FF 69 - 109 0.0 S irq/17-ohci_hcd 87 FF 68 - 108 0.0 S irq/17-ohci_hcd 91 FF 65 - 105 0.0 S irq/1-i8042 20 FF 50 - 90 0.0 S irq/9-acpi 68 FF 50 - 90 0.3 S irq/22-ahci 85 FF 50 - 90 0.0 S irq/18-ohci_hcd 89 FF 50 - 90 0.0 S irq/18-ohci_hcd 267 FF 50 - 90 0.0 S irq/14-pata_ati 268 FF 50 - 90 0.0 S irq/15-pata_ati 289 FF 50 - 90 0.0 S irq/22-firewire 507 FF 50 - 90 0.0 S irq/7-parport0 945 FF 50 - 90 0.0 S irq/43-eth0 1225 FF 50 - 90 0.7 S irq/18-nvidia 3 TS - 0 19 0.2 S ksoftirqd/0 9 TS - 0 19 0.2 R ksoftirqd/1 spinymouse@precise:~$ grep RTIRQ_NAME_LIST /etc/default/rtirq #RTIRQ_NAME_LIST="rtc hdspm ICE1712 snd_hdsp snd_ice1 snd usb i8042" RTIRQ_NAME_LIST="rtc snd_hdsp snd_ice1 snd usb i8042" ## REBOOT spinymouse@precise:~$ uname -a Linux precise 3.0.30 #1 SMP PREEMPT Sun May 6 22:44:54 CEST 2012 x86_64 x86_64 x86_64 GNU/Linux spinymouse@precise:~$ /etc/init.d/rtirq status PID CLS RTPRIO NI PRI %CPU STAT COMMAND 91 FF 90 - 130 0.0 S irq/8-rtc0 760 FF 85 - 125 0.0 S irq/18-hdspm 767 FF 80 - 120 0.0 S irq/20-ICE1712 785 FF 79 - 119 0.0 S irq/21-ICE1712 78 FF 60 - 100 0.0 S irq/19-ehci_hcd 80 FF 60 - 100 0.2 S irq/16-ohci_hcd 82 FF 59 - 99 0.0 S irq/17-ohci_hcd 86 FF 58 - 98 0.0 S irq/17-ohci_hcd 90 FF 55 - 95 0.0 S irq/1-i8042 18 FF 50 - 90 0.0 S irq/9-acpi 67 FF 50 - 90 0.2 S irq/22-ahci 84 FF 50 - 90 0.0 S irq/18-ohci_hcd 88 FF 50 - 90 0.0 S irq/18-ohci_hcd 253 FF 50 - 90 0.0 S irq/14-pata_ati 256 FF 50 - 90 0.0 S irq/15-pata_ati 289 FF 50 - 90 0.0 S irq/22-firewire 577 FF 50 - 90 0.0 S irq/7-parport0 1004 FF 50 - 90 0.0 S irq/43-eth0 1175 FF 50 - 90 0.7 S irq/18-nvidia 3 TS - 0 19 0.2 S ksoftirqd/0 9 TS - 0 19 0.2 S ksoftirqd/1 spinymouse@precise:~$ grep RTIRQ_NAME_LIST /etc/default/rtirq RTIRQ_NAME_LIST="rtc hdspm ICE1712 snd_hdsp snd_ice1 snd usb i8042" #RTIRQ_NAME_LIST="rtc snd_hdsp snd_ice1 snd usb i8042" Regards, Ralf

12 years, 6 months

What Focusright firewire interfaces are fully working and currently available?

by drew Roberts

So I am looking to buy a new multi channel audio interface. Looking at firewire. Of the ones I might be interested in, it seems the supported ones are discontinued and the available ones are not fully supported. Does anyone have any useful input? Looking mainly at focusrite but will look at others. I want high channel count possibilities. 8+ preferably 24+ down the road. all the best, drew

12 years, 6 months

Re: [LAU] rt prio with kernel 3.x, issue to use the vesa driver

by Ralf Mardorf

On Sun, 2012-05-06 at 16:16 +0200, Ralf Mardorf wrote: > Did for anyone using a 3.0 rt or 3.x rt work to patch the kernel or to > patch the nvidia driver to get the rt working with the nvidia driver? > Both methods didn't work for me. To ensure that a patch might be buggy, > I edited manually. Broken English ... a buggy patch can't be the cause, since I edited manually several times.

12 years, 6 months

Re: [LAU] [Rosegarden-user] rosegarden ubuntu 12.04

by david

On 05/05/2012 11:25 AM, bart deruyter wrote: > Gnome 3 would be a good alternative too, if it wouldn't hang so often > and reduce jack performance so much. I work on an i7 with 16 GB of ram > and a quadro600 nvidia, so in my opinion that even should be overkill > for audio production, but on Gnome3 based systems like gnome-shell and > unity it seems not to be. Gnome3 appears to be a major resource hog. I've been told that Gnome3 desktop uses 1GB RAM without running anything else. For comparison, XFCE uses less than 256MB. My desktop machine has quad-core AMD Phenom II processor, 12GB RAM and some NVidia card that isn't supported by Nouveau. It works very well with RG and other audio and graphics stuff I do with it. It's running the XFCE environment. Some audio distros I know of use XFCE or LXDE because both KDE4 and Gnome3 have gotten too big. KDE in particular fires up a whole bunch of KDE system daemons that suck up processor cycles, check hardware status (a good way to trigger extraneous interrupts), index the files in your home folder, etc. Lots of functionality that you might not want to have interfering with audio work. > The other issue remains though, why does rosegarden make such a mess of > jack when it exits? I sometimes even have to reboot in order to get jack > running again after I stop rosegarden. Hmmmm. What version of Rosegarden? Do you have RG start JACK, or do you start JACK first, then RG? And do you have the problem if you don't use Ladish? > 2012/5/5 david > > On 05/04/2012 10:18 PM, bart deruyter wrote: > > Hi all, > > > > I've noticed an old issue with rosegarden that seem not to get > fixed, at > > least, I don't know how to fix it or find a workaround for it, > and a new > > issue on ubuntu 12.04. > > > > The old issue is one concerning jack. When exiting rosegarden jack > > crashes in a strange way. It keeps running, but there is no in- nor > > output anymore, in other words : no audio. I'm sure that cannot be a > > wanted result of exiting rosegarden, it should only exit > rosegarden and > > not block sound for all the other sound-releated applications. This > > always happens when using ladish. > > I've never encountered that problem, but I don't use Ladish. > > > A new issue is a gui-issue. In Ubuntu 12.04 after a while the > menu when > > right-clickingdoes not show up anymore. This happens so far in > > unity-mode. I didn't test it in gnome-shell yet. So, I don't know > if it > > is a ubuntu related issue or a rosegarden issue. In KDE it seems > to work > > fine. > > My wife uses JPilot on her netbook, with the Unity interface. JPilot > frequently quits responding to clicks under Unity (especially if it's > maximized, but also when it's not maximized). JPilot works fine under > KDE3 & 4, LXDE, XFCE and FluxBox. So I think it's an Ubuntu Unity > interface problem. > > I dislike the Unity interface, so my suggestion would be to replace it > with one of the above. -- David gnome(a)hawaii.rr.com authenticity, honesty, community http://clanjones.org/david/ http://dancing-treefrog.deviantart.com/

12 years, 6 months

Re: [LAU] rt prio with kernel 3.x, issue to use the vesa driver

by Emanuel Rumpf

a list more appropriate for this trash: linux-tuning : http://lists.linuxaudio.org/listinfo/

12 years, 6 months

rt prio with kernel 3.x, issue to use the vesa driver

by Ralf Mardorf

That for a shared IRQ the graphics will become an unwanted high prio doesn't happen with a kernel 2.6.31-rt19 + NVIDIA 195.36.15 on Suse 11.2 on my machine. As already reported, it also doesn't happen with the kernels 3.0.23-avl-7-pae (threadirqs) and 3.0.23-rt40 on current AV Linux, where still the nv driver is available and used on my machine. IIUC it's definitive related to the kernel, I anyway will test the vesa driver with Ubuntu Studio Precise, but can' get it working. I modified the xorg.conf I use with the nvida driver, startup ends with tty, no X. Does anybody notice what I might have missed? Latest modification: Section "ServerLayout" Identifier "X.org Configured" Screen 0 "Screen0" 0 0 InputDevice "Mouse0" "CorePointer" InputDevice "Keyboard0" "CoreKeyboard" EndSection Section "Module" #Load "glx" #Load "dbe" #Load "dri2" #Load "extmod" #Load "record" EndSection Section "InputDevice" Identifier "Keyboard0" Driver "kbd" Option "XkbRules" "xorg" Option "XkbModel" "pc105" Option "XkbLayout" "de" Option "XkbVariant" "ro" Option "XkbOptions" "lv3:ralt_switch" EndSection Section "InputDevice" Identifier "Mouse0" Driver "mouse" Option "Protocol" "auto" Option "Device" "/dev/input/mice" Option "ZAxisMapping" "4 5 6 7" EndSection Section "Monitor" Identifier "Monitor0" VendorName "Monitor Vendor" ModelName "Monitor Model" DisplaySize 305 230 HorizSync 29-98 VertRefresh 50-120 #modeline "1152x864" 128.42 1152 1232 1360 1568 864 865 868 910 #Gamma 1.0 EndSection Section "Device" Identifier "Card0" Driver "vesa" #Option "Coolbits" "1" #Option "AddARGBGLXVisuals" "true" #Option "TripleBuffer" "false" #VendorName "nVidia Corporation" #BoardName "G72 [GeForce 7300 SE/7200 GS]" BusID "PCI:1:0:0" EndSection Section "Screen" Identifier "Screen0" Device "Card0" Monitor "Monitor0" SubSection "Display" #Viewport 0 0 #Depth 24 #Modes "1152x864" #Virtual 3840 1200 EndSubSection EndSection #Section "Extensions" # Option "Composite" "enable" #EndSection Regards, Ralf PS: If there won't be an rtirq script able to handle it or any other way to do it, I hadn't time to test until now, I'll simply will use a kernel <= 3.0 instead > 3.0.

12 years, 6 months

Open Audio Programming Tutorials

by Harry van Haaren

Hey All, After some very inspiring conversations at the LAC, I have decided to renew the efforts to document linux audio programming for beginners. I feel that although there's a lot of really useful tutorials out there, but there's still a lack of easy accessible introductory audio programming. Particularly topics such as threading, and thread synchronization are particularly difficult to learn, or even find relevant, easy to read code about. Announcing: Open Audio Programming Tutorials! This is a documentation effort, not of any particular library or tool, just "Linux Audio Programming" in general. Feel free to check the code posted, feedback on it, fork it and send me merge requests, whatever :) Currently there's 6 different tutorials, all C++ with GTKmm for user interfaces. Intentions are to add more as time permits! -Harry PS: Sending to Linux-Audio-User list in case there's people who want to try start programming, but haven't subscribed to the Linux-Audio-Developers list (yet). Replies to Linux-Audio-Developers please :)

12 years, 6 months

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-user May 2012