[linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
6 Oct
2004
6 Oct
'04
5:20 p.m.
On Wed, 6 Oct 2004 04:37:17 +0200, Florian Schmidt <mista.tapas(a)gmx.net> wrote:
On Tue, 05 Oct 2004 19:01:30 -0400
Lee Revell <rlrevell(a)joe-job.com> wrote:
Hi,
the kernel does weird things nowadays. I do not get any satisfactory results
with either S8 or S9. I get xruns all over the place [under loads much
lighter than a kernel compilation] although the VP and related settings are
as always. T1 is working smoothly again. I have no idea what caused this
behaviour in S8 and S9. But it feels like a scheduler problem. X11 is much
more jumpy under these kernels and load, too than under S7 and T1
So, i would still expect some weirdnesses in VP enabled kernels, since they
are based on the mm tree [which might hold many surprises in itself].
But anyways, back to Mark.
I talked to Mark on irc and he seems to have had SMP enabled on a UP system
[p4 mobile w/o HT according to mark]. OTOH this worked fine on fedora,
right? Also this is a notebook, maybe power saving features play a role.
One more thing to note. His gentoo system uses a linuxthreads glibc. The
same hardware and kernels are working very nicely on his fedora core system
with CCRMA and VP kernels.
I suspect some weird interplay of a linuxthreads self built glibc and a
2.6.x kernel, especially wrt to SCHED_FIFO scheduling.
One more thing:
The setups between fedora core and gentoo might differ in what default
services run. Do you [Mark] have any powermanagement daemons loaded? cpu
frequency scaling, etc? I know only little about these issues, since i
basically do not use power management.
The two things i would recommend to try would be:
1] Build a very very barebone kernel for, let's say, 386 architecture. no
acpi no apm, no usb [if possible], no smp, local APIC if he likes. Only
those devices really needed, no dri, etc [but of course with the usual set
of VP features enabled].
2] try a gentoo system with a nptl enabled glibc. Either build one, or maybe
get some live cd based on gentoo which fulfills the requirements. See if the
problems persist. Or try some third distribution with nptl libc and slap a
VP kernel on it. If it works fine, your gentoo system is b0rk3n in some way.
flo
p.s. which jack version btw?
Hi all,
OK, Last evening and this morning I have made good headway. It
appears that the main culprit was ACPI support, or some part of it.
With most of the ACPI stuff now disabled in the kernel build I am able
to run jack with no clients attached, as a root or user with the lsm
module modprobed, at p=8, n=2 for over 1 hour with no xruns. If my
calculations are correct this would be a latency of less than 0.5mS!
My current kernel is 2.6.9-rc2-mm4-VP-S7-lsm-UMP-noACPI:
- 2.6.8 from kernel.org
- 2.6.9-rc2 patch
- 2.6.9-rc2-mm4 patch
- voluntary_premption-S7 patch
- realtime-lsm patch
I've built it with most everything I can find that I do not
absolutely need turned off. I'm non-SMP, ACPI mostly gone, and who
knows what else along the way. I'll send the .config file to whoever
might need it. I am then running the sound card and CDROM interrupts
non-threaded and everything else is threaded.
In terms of better testing I now need to concentrate on finding an
application that run without creating xruns within the application
itself. At higher latency values (p=256 for instance) alsaplayer -r -o
jack runs pretty well, but as I move to lower latencies I find that
alsaplayer deterministically creates xruns at the transistion between
every song on a CD. I also tried alsaplayer playing wave files from
disk, but I see xruns as it goes from one file to the next also.
I want to try using chrt to see if alsaplayer is really getting
real-time access, etc. Maybe there are things still not right with the
way I'm running it on my box.
Is there any other simple app that could play a wave file from, for
instance, memory without this sort of problem?
I'll certainly be trying out Ardour as a bigger test once I'm more
confident of the disk access issues in real-time. I think that 1394
support may not be in good shape in this kernel yet.
So, the good news is my Gentoo system is not broken, and that I've
hopefully learned a bit and contributed just a little bit more along
the way. That said I still have a long ways to go.
Cheers,
Mark
The
only 3 xruns happened right when I started the kernel compile.
This is 2.6.9-rc2-mm4-VP-S7
There is absolutely _no way_ that these are kernel-induced. The period
time is 5ms!!! This kernel should _never_ produce an delay of more than
200 usecs. This _has_ to be flaky hardware, or a bug in jackd or in
your clients. What happens if you run jackd with no clients? 
6 Oct
6 Oct
5:57 p.m.
Mark Knecht <markknecht(a)gmail.com> writes:
In terms of better testing I now need to
concentrate on finding an
application that run without creating xruns within the application
itself. At higher latency values (p=256 for instance) alsaplayer -r -o
jack runs pretty well, but as I move to lower latencies I find that
alsaplayer deterministically creates xruns at the transistion between
every song on a CD. I also tried alsaplayer playing wave files from
disk, but I see xruns as it goes from one file to the next also.
I want to try using chrt to see if alsaplayer is really getting
real-time access, etc. Maybe there are things still not right with the
way I'm running it on my box.
I sometimes see problems with alsaplayer at low latencies, too.
Generally below -p128 or so.
Is there any other simple app that could play a
wave file from, for
instance, memory without this sort of problem?
I've had good luck with hydrogen. Play one of the demo drum patterns.
So, the good news is my Gentoo system is not
broken, and that I've
hopefully learned a bit and contributed just a little bit more along
the way. That said I still have a long ways to go.
Good work.
--
joq
6:31 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
Jack O'Quin wrote:
I had Ardour on the machine already so I fired it up. It's running fine
playing 'Drums on Demand' loops which I very quickly imported into a new
Ardour session. No xruns in the last 20 minutes, although this time I'm
using QJC which doesn't seem to let me start any faster than -p64. I can
live with sub 3mS. (for now...)
All this with updatedb running in the background. This could make me
heady, if it holds up. Much testing needed.
And to you to for doing the realtim-lsm stuff. One thing I am seeing
which I don't quite understand is with realtime-lsm loaded it seems that
apps default to somewhere other than my home account the first time I
open a file menu. They are either going to / or /root. Not sure if
that's by design or some sort of limitation, but it seems a bit off to me.
Thanks,
Mark
Mark Knecht <markknecht(a)gmail.com> writes:
So -p8 -n2 qualifies as 'low latencies', right? ;-)
In terms of better testing I now need to
concentrate on finding an
application that run without creating xruns within the application
itself. At higher latency values (p=256 for instance) alsaplayer -r -o
jack runs pretty well, but as I move to lower latencies I find that
alsaplayer deterministically creates xruns at the transistion between
every song on a CD. I also tried alsaplayer playing wave files from
disk, but I see xruns as it goes from one file to the next also.
I want to try using chrt to see if alsaplayer is really getting
real-time access, etc. Maybe there are things still not right with the
way I'm running it on my box.
I sometimes see problems with alsaplayer at low latencies, too.
Generally below -p128 or so. Is there any other simple app that could play a
wave file from, for
instance, memory without this sort of problem?
I've had good luck with hydrogen. Play one of the demo drum patterns. So, the good news is my Gentoo system is not
broken, and that I've
hopefully learned a bit and contributed just a little bit more along
the way. That said I still have a long ways to go.
Good work.
9:33 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
Mark Knecht <mknecht(a)controlnet.com> writes:
And to you to for doing the realtim-lsm stuff. One
thing I am seeing
which I don't quite understand is with realtime-lsm loaded it seems
that apps default to somewhere other than my home account the first
time I open a file menu. They are either going to / or /root. Not sure
if that's by design or some sort of limitation, but it seems a bit off
to me.
The LSM should not be affecting this in any way.
More likely, you're logging in differently and the applications are
starting with different values of $HOME in their environment. On most
distributions /root is the home directory of the superuser. You
really don't want to be using that for audio data.
It matters whether you log in as `root' or as yourself. You should
generally log in as `mknecht', and use `sudo' to run individual
commands requiring root privileges.
--
joq
7:27 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
On Wed, Oct 06, 2004 at 12:01:43PM -0700, Mark Knecht wrote:
Hi all,
OK, Last evening and this morning I have made good headway. It
appears that the main culprit was ACPI support, or some part of it.
With most of the ACPI stuff now disabled in the kernel build I am able
to run jack with no clients attached, as a root or user with the lsm
module modprobed, at p=8, n=2 for over 1 hour with no xruns. If my
calculations are correct this would be a latency of less than 0.5mS!
p=8??? good lord. I'd like to see you try that with any real apps :-)
I'm happy if I can get 128, would like to get down to 64 reliably...
--
Paul Winkler
http://www.slinkp.com
7:48 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
On Wed, 2004-10-06 at 14:09, Paul Winkler wrote:
On Wed, Oct 06, 2004 at 12:01:43PM -0700, Mark Knecht
wrote:
If he tries to go any further he'll have to hack the hardware, most
probably that's getting close to the minimum transfer size on the pci
buss, right?
-- Fernando
Hi all,
OK, Last evening and this morning I have made good headway. It
appears that the main culprit was ACPI support, or some part of it.
With most of the ACPI stuff now disabled in the kernel build I am able
to run jack with no clients attached, as a root or user with the lsm
module modprobed, at p=8, n=2 for over 1 hour with no xruns. If my
calculations are correct this would be a latency of less than 0.5mS!
p=8??? good lord. I'd like to see you try that with any real apps :-)
I'm happy if I can get 128, would like to get down to 64 reliably...
8:01 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
Fernando Pablo Lopez-Lezcano wrote:
On Wed, 2004-10-06 at 14:09, Paul Winkler wrote:
At -p4 I just get messages about the Jack watchdog timer timing out.
This is all getting pretty silly down that low. No reason I know of to
do it other then to test the limits of these new kernels.
On Wed, Oct 06, 2004 at 12:01:43PM -0700, Mark
Knecht wrote:
If he tries to go any further he'll have to hack the hardware, most
probably that's getting close to the minimum transfer size on the pci
buss, right?
-- Fernando
Hi all,
OK, Last evening and this morning I have made good headway. It
appears that the main culprit was ACPI support, or some part of it.
With most of the ACPI stuff now disabled in the kernel build I am able
to run jack with no clients attached, as a root or user with the lsm
module modprobed, at p=8, n=2 for over 1 hour with no xruns. If my
calculations are correct this would be a latency of less than 0.5mS!
p=8??? good lord. I'd like to see you try that with any real apps :-)
I'm happy if I can get 128, would like to get down to 64 reliably...
7:48 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
Paul Winkler wrote:
On Wed, Oct 06, 2004 at 12:01:43PM -0700, Mark Knecht
wrote:
Well, it's not much of a test yet, but I've been running 24-bit drum
loops in Ardour for the last 90 minutes from my internal hard drive at
-p64, -n2. I can go to the console, change workspaces in fluxbox, do
Evolution and have both Mozilla and Firefox open. No xruns. Using a
consistent 16% of CPU and 170K-205K of disk bandwidth.
It's very early yet. There is very little disk activity, but Ardour,
Jack, and QJC seem happy so far.
I'm sure this will come back and burn me in the next few hours. We'll
see. I'm about to fire up jack_vst and Battery to see what happens there.
- Mark
Hi all,
OK, Last evening and this morning I have made good headway. It
appears that the main culprit was ACPI support, or some part of it.
With most of the ACPI stuff now disabled in the kernel build I am able
to run jack with no clients attached, as a root or user with the lsm
module modprobed, at p=8, n=2 for over 1 hour with no xruns. If my
calculations are correct this would be a latency of less than 0.5mS!
p=8??? good lord. I'd like to see you try that with any real apps :-)
I'm happy if I can get 128, would like to get down to 64 reliably...
7 Oct
7 Oct
2:18 a.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
Last Wednesday 06 October 2004 20:01, Mark Knecht was like:
as I move to lower latencies I find that
alsaplayer deterministically creates xruns at the transistion between
every song on a CD.
Even if the CD is recorded DAO?
gnome-cd should be ok.
cheers
tim hall
12:56 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
tim hall wrote:
Last Wednesday 06 October 2004 20:01, Mark Knecht was
like:
Is gnome-cd a Jack app? If so I'll look into building it.
As a follow up to gettting xruns, I'm now running Native Instruments
Battery under jack_fst & wine-20040505. It's been running for 2 hours
with no xruns.
Most interesting, and applicable to another thread, Battery/jack_ft seem
to be able to attach reliably to JAck without creating any xruns even at
very low latencies like -p32.
- Mark
as I move to lower latencies I find that
alsaplayer deterministically creates xruns at the transistion between
every song on a CD.
Even if the CD is recorded DAO?
gnome-cd should be ok.
cheers
tim hall
1:56 p.m.
New subject: [linux-audio-user] Press material on Linux Audio
Hello,
we are currently collecting press material on Linux Audio for the web page
of the next Linux Audio Conference: http://www.zkm.de:81/lac/press.shtml
In addition to conference reports we will add an item "articles on Linux
Audio". Any link / (licensed) PDF on that subject is welcome. Language
need not necessarily be english.
Matthias
--
Dr. Matthias Nagorni
SuSE Linux AG
Maxfeldstr. 5 phone: +49 911 74053375
D - 90409 Nuernberg fax : +49 911 74053483
25 Oct
25 Oct
1:06 p.m.
New subject: [linux-audio-user] Press material on Linux Audio
Hi Matthias:
You probably know that I've been writing a monthly column about Linux
audio for the Linux Journal On-line. Feel free to link to any or all of
that material (if it qualifies as press material).
Best regards,
dp
Matthias Nagorni wrote:
Hello,
we are currently collecting press material on Linux Audio for the web page
of the next Linux Audio Conference: http://www.zkm.de:81/lac/press.shtml
In addition to conference reports we will add an item "articles on Linux
Audio". Any link / (licensed) PDF on that subject is welcome. Language
need not necessarily be english.
Matthias
7 Oct
7 Oct
4:59 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
Last Thursday 07 October 2004 15:38, Mark Knecht was like:
Is gnome-cd a Jack app? If so I'll look into
building it.
Ah, no. It doesn't present any connections to Jackd, although for
some reason
does work when Jackd is running. I tend not to play CDs directly through
Jackd. Apologies for the noise factor.
cheers
tim hall
5:43 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
On Thu, 2004-10-07 at 14:38, tim hall wrote:
Last Thursday 07 October 2004 15:38, Mark Knecht was
like:
This is the expected behavior if your soundcard does hardware mixing.
Lee
Is gnome-cd a Jack app? If so I'll look into
building it.
Ah, no. It doesn't present any connections to Jackd, although for
some reason
does work when Jackd is running.
5:51 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
tim hall wrote:
Last Thursday 07 October 2004 15:38, Mark Knecht was
like:
No problem at all. In fact, thanks for chiming in. I like that.
Still looking for other Jack-enabled CD players to test out. If anyone
knows of any please speak up. I looked on the Jack applications page.
The only app that shows CD support is alspalayer.
- Mark
Is gnome-cd a Jack app? If so I'll look into
building it.
Ah, no. It doesn't present any connections to Jackd, although for some reason
does work when Jackd is running. I tend not to play CDs directly through
Jackd. Apologies for the noise factor.
cheers
tim hall
6:43 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
tim hall wrote:
Xmms has its own jack plugin and can even read CD's cdda2wav style
(over the IDE bus) with the "cdread" plugin (downloaded separately).
Loading the disc is a little goofy, do "Add-->URL" and enter
"/dev/cdrom", or whichever device you use.
Kevin
Last Thursday 07 October 2004 15:38, Mark Knecht
was like:
No problem at all. In fact, thanks for chiming in. I like that.
Still looking for other Jack-enabled CD players to test out. If anyone
knows of any please speak up. I looked on the Jack applications page.
The only app that shows CD support is alspalayer.
- Mark
7:20 p.m.
New subject: [linux-audio-user] Re: [Jackit-devel] What is a 'delay exceeds' vs. and xrun - BIG MISSES - AND SOLUTIONS!
Kevin Ernste wrote:
Thanks. This does hook up without causing xruns at -p32 and plays an ogg
file from disk cleanly. Cannot get the playlist working yet, but this
will be helpful. Thanks.
At -p8 or 16 I do get a few xruns while playing so I think this is good.
It uses something called 'bio2jack'...
- Mark
tim hall
wrote:
Xmms has its own jack plugin and can even read CD's cdda2wav style
(over the IDE bus) with the "cdread" plugin (downloaded separately).
Loading the disc is a little goofy, do "Add-->URL" and enter
"/dev/cdrom", or whichever device you use.
Kevin
Last Thursday 07 October 2004 15:38, Mark Knecht
was like:
No problem at all. In fact, thanks for chiming in. I like that.
Still looking for other Jack-enabled CD players to test out. If anyone
knows of any please speak up. I looked on the Jack applications page.
The only app that shows CD support is alspalayer.
- Mark
7362
days inactive
7381
days old
linux-audio-user@lists.linuxaudio.org
16 comments
10 participants
participants (10)
-
Dave Phillips -
Fernando Pablo Lopez-Lezcano -
Jack O'Quin -
Kevin Ernste -
Lee Revell -
Mark Knecht -
Mark Knecht -
Matthias Nagorni -
Paul Winkler -
tim hall