27 Jul
2005
27 Jul
'05
10:25 p.m.
Hi,
I'm going crazy!!!
Last night, after some months playing with Ubuntu, I installed SUSE 9.3
on my amd64 with an Edirol UA-25. I installed the kernel source, patched
it with rt-lsm.patch, built the modules, installed ect.
Added:
/sbin/modprobe -r capability
/sbin/modprobe realtime any=1 mlock=1
to /etc/init.d/boot.local
At this point i got a lot of xruns even with jackd -R -dalsa -p1024 -n4.
On Ubuntu I worked flawlessly with -p256 -3 (17ms).
I recompiled jack from source (configured with /dev/shm as temp dir) and
got the same results.
Then I tried to shutdown some more services that were still on (firewall
and so on).
I restarted jack (via qjackctl) and the xruns were gone. I've pushed
jack settings until I was able to play Zynaddsubfx on my midi keyboard
(Edirol PCR30 - on a different usb port), thru Muse with some plugins
applied to the input channel at:
-p128 -n3 (8,7ms), with no xruns and no audio glitches.
-p64 -n4 (5ms), still with no xruns but with some minimum gliches.
I was so happy that I thought about letting my box run forever, without
shutting it down anymore.
But then I decided to install the drivers for the nvidia card (the
newest from their site). I don't remember if I installed or updated some
rpms at the same time. Rebooted.
Since then I can't get rid of xruns and audio glitches even with a 1024
buffer.
After some googlin I found some old posts on this list about conflicts
with the nvidia driver so I tried to restart without it. Nothing changed.
Please, help me. Because I will not go to sleep until I'll solve this.
Cesare
27 Jul
27 Jul
10:32 p.m.
On Thu, 2005-07-28 at 01:18 +0200, Cesare Marilungo wrote:
Hi,
I'm going crazy!!!
Last night, after some months playing with Ubuntu, I installed SUSE 9.3
on my amd64 with an Edirol UA-25. I installed the kernel source, patched
it with rt-lsm.patch, built the modules, installed ect.
Added:
Were the kernel versions different between the two systems?
USB audio is known to have some problems with older 2.6 kernels.
Lee
11:38 p.m.
Ubuntu had 2.6.10 something with realtime-lsm module.
On Suse I haven't recompiled the kernel. Just the realtime module, with
the patched kernel.
It's the one shipped with Suse 9.3 (2.6.11.4-21.7-default).
I forgot to mention that before than that I also tried with a custom
compiled vanilla 2.4.11.7 with rt-lsm and realtime-preempt patches and
got loads of xruns.
Anyway yesterday I pushed jack settings way to 5ms latency (p128 -n3
with 24bits, 44.100k, 2 ins, 2 outs).
Is it possible to achieve such low latency for USB 1.1 card?
I doubt I had two instances of jack running, changed settings on the
frontmost qjackctl, and listened to the output from the first instance.
I disconnected and reconnected Muse and Zyn various times to change the
buffer size. And I also noticed the degradation of sound when pushing
jack settings so too far.
I haven't had any xrun. Even when connecting apps.
c.
Lee Revell wrote:
On Thu, 2005-07-28 at 01:18 +0200, Cesare Marilungo
wrote:
Hi,
I'm going crazy!!!
Last night, after some months playing with Ubuntu, I installed SUSE 9.3
on my amd64 with an Edirol UA-25. I installed the kernel source, patched
it with rt-lsm.patch, built the modules, installed ect.
Added:
Were the kernel versions different between the two systems?
USB audio is known to have some problems with older 2.6 kernels.
Lee
28 Jul
28 Jul
4:07 a.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
I solved the problem *cough* disabling the ethernet card. Don't know why
it worked randomly yesterday.
It works even with the nvidia accelerated driver. And no more glitches
with -p64 -n4.
5.8ms of total latency with a USB card is better than my old Egosys 2496
(pci) which made Cubase on XP unstable with a buffer less than 512.
Interestingly, the nrpacks parameter of the snd_usb_audio module is set
to 4 (instead of 1 as people suggested on the net).
Thank you, anyway.
c.
Cesare Marilungo wrote:
Ubuntu had 2.6.10 something with realtime-lsm module.
On Suse I haven't recompiled the kernel. Just the realtime module, with
the patched kernel.
It's the one shipped with Suse 9.3 (2.6.11.4-21.7-default).
I forgot to mention that before than that I also tried with a custom
compiled vanilla 2.4.11.7 with rt-lsm and realtime-preempt patches and
got loads of xruns.
Anyway yesterday I pushed jack settings way to 5ms latency (p128 -n3
with 24bits, 44.100k, 2 ins, 2 outs).
Is it possible to achieve such low latency for USB 1.1 card?
I doubt I had two instances of jack running, changed settings on the
frontmost qjackctl, and listened to the output from the first instance.
I disconnected and reconnected Muse and Zyn various times to change the
buffer size. And I also noticed the degradation of sound when pushing
jack settings so too far.
I haven't had any xrun. Even when connecting apps.
c.
Lee Revell wrote:
On Thu, 2005-07-28 at 01:18 +0200, Cesare
Marilungo wrote:
Hi,
I'm going crazy!!!
Last night, after some months playing with Ubuntu, I installed SUSE 9.3
on my amd64 with an Edirol UA-25. I installed the kernel source, patched
it with rt-lsm.patch, built the modules, installed ect.
Added:
Were the kernel versions different between the two systems?
USB audio is known to have some problems with older 2.6 kernels.
Lee
10:55 a.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Thu, 2005-07-28 at 08:07 +0200, Cesare Marilungo wrote:
I solved the problem *cough* disabling the ethernet
card. Don't know why
it worked randomly yesterday.
I can confirm this problem/solution. UA25 glitchy as heck till eth0 is
disabled. I have a dell inspiron 5100 running gentoo btw.
Makes icecast streaming jackd a bit tough, but at least it works.
It works even with the nvidia accelerated driver. And
no more glitches
with -p64 -n4.
5.8ms of total latency with a USB card is better than my old Egosys 2496
(pci) which made Cubase on XP unstable with a buffer less than 512.
Interestingly, the nrpacks parameter of the snd_usb_audio module is set
to 4 (instead of 1 as people suggested on the net).
Thank you, anyway.
c.
Cesare Marilungo wrote:
Ubuntu had 2.6.10 something with realtime-lsm
module.
On Suse I haven't recompiled the kernel. Just the realtime module, with
the patched kernel.
It's the one shipped with Suse 9.3 (2.6.11.4-21.7-default).
I forgot to mention that before than that I also tried with a custom
compiled vanilla 2.4.11.7 with rt-lsm and realtime-preempt patches and
got loads of xruns.
Anyway yesterday I pushed jack settings way to 5ms latency (p128 -n3
with 24bits, 44.100k, 2 ins, 2 outs).
Is it possible to achieve such low latency for USB 1.1 card?
I doubt I had two instances of jack running, changed settings on the
frontmost qjackctl, and listened to the output from the first instance.
I disconnected and reconnected Muse and Zyn various times to change the
buffer size. And I also noticed the degradation of sound when pushing
jack settings so too far.
I haven't had any xrun. Even when connecting apps.
c.
Lee Revell wrote:
On Thu, 2005-07-28 at 01:18 +0200, Cesare
Marilungo wrote:
Hi,
I'm going crazy!!!
Last night, after some months playing with Ubuntu, I installed SUSE 9.3
on my amd64 with an Edirol UA-25. I installed the kernel source, patched
it with rt-lsm.patch, built the modules, installed ect.
Added:
Were the kernel versions different between the two systems?
USB audio is known to have some problems with older 2.6 kernels.
Lee
5:26 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Thu, 2005-07-28 at 02:59 -0500, Ryan wrote:
On Thu, 2005-07-28 at 08:07 +0200, Cesare Marilungo
wrote:
Are you both using the same network driver/hardware? Does it share an
interrupt with the USB?
Lee
I solved the problem *cough* disabling the
ethernet card. Don't know why
it worked randomly yesterday.
I can confirm this problem/solution. UA25 glitchy as heck till eth0 is
disabled. I have a dell inspiron 5100 running gentoo btw.
Makes icecast streaming jackd a bit tough, but at least it works.
6:07 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Thu, 2005-07-28 at 15:26 -0400, Lee Revell wrote:
On Thu, 2005-07-28 at 02:59 -0500, Ryan wrote:
Ethernet controller: Broadcom Corporation BCM4401 100Base-T (rev 01)
Interrupt: pin A routed to IRQ 7
(b44 module, marked experimental since 2.4.x days)
USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M)
USB UHCI Controller
Interrupt: pin [A, B, C, d] routed to IRQ 11
This is an UHCI controller... I thought the usb kernel stuff was going
to get some attention, something about top-half/bottom-half irq
handling? Afaik that hasn't happened yet.
-ry
On Thu, 2005-07-28 at 08:07 +0200, Cesare
Marilungo wrote:
Are you both using the same network driver/hardware? Does it share an
interrupt with the USB? I solved the problem *cough* disabling the
ethernet card. Don't know why
it worked randomly yesterday.
I can confirm this problem/solution. UA25 glitchy as heck till eth0 is
disabled. I have a dell inspiron 5100 running gentoo btw.
Makes icecast streaming jackd a bit tough, but at least it works.
6:57 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Thu, 2005-07-28 at 10:11 -0500, Ryan wrote:
On Thu, 2005-07-28 at 15:26 -0400, Lee Revell wrote:
Since the interrupt is not shared, you should really not be having this
problem. The b44 module is likely to be the cause. It may be disabling
interrupts globally.
Try running Ingo's RT preempt kernel with all latency/irqs off
debugging enabled.
On Thu, 2005-07-28 at 02:59 -0500, Ryan wrote:
Ethernet controller: Broadcom Corporation BCM4401 100Base-T (rev 01)
Interrupt: pin A routed to IRQ 7
(b44 module, marked experimental since 2.4.x days)
On Thu, 2005-07-28 at 08:07 +0200, Cesare
Marilungo wrote:
Are you both using the same network driver/hardware? Does it share an
interrupt with the USB? I solved the problem *cough* disabling the
ethernet card. Don't know why
it worked randomly yesterday.
I can confirm this problem/solution. UA25 glitchy as heck till eth0 is
disabled. I have a dell inspiron 5100 running gentoo btw.
Makes icecast streaming jackd a bit tough, but at least it works. USB Controller: Intel Corporation 82801DB/DBL/DBM
(ICH4/ICH4-L/ICH4-M)
USB UHCI Controller
Interrupt: pin [A, B, C, d] routed to IRQ 11
This is an UHCI controller... I thought the usb kernel stuff was going
to get some attention, something about top-half/bottom-half irq
handling? Afaik that hasn't happened yet.
2.6.13 should be better.
Lee
10:38 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
Ryan wrote:
On Thu, 2005-07-28 at 15:26 -0400, Lee Revell wrote:
Marvell Yukon Gigabit Ethernet 10/100/1000
I'm in Window$ now( developing in Flash, sigh!), can't check the irqs.
But I remember that I checked for this and the irqs were different.
It's also interesting that I got half the latency switching from
kernel-2.6.10 to 2.6.11. The latter is x86_64, BTW. And I can't get such
low latency on MacOSX or XP (neither with the Egosys PCI card). And this
is a USB card.
c.
On Thu, 2005-07-28 at 02:59 -0500, Ryan wrote:
Ethernet controller: Broadcom Corporation BCM4401 100Base-T (rev 01)
Interrupt: pin A routed to IRQ 7
(b44 module, marked experimental since 2.4.x days)
USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M)
USB UHCI Controller
Interrupt: pin [A, B, C, d] routed to IRQ 11
This is an UHCI controller... I thought the usb kernel stuff was going
to get some attention, something about top-half/bottom-half irq
handling? Afaik that hasn't happened yet.
-ry
On Thu, 2005-07-28 at 08:07 +0200, Cesare
Marilungo wrote:
Are you both using the same network driver/hardware? Does it share an
interrupt with the USB?
I solved the problem *cough* disabling the
ethernet card. Don't know why
it worked randomly yesterday.
I can confirm this problem/solution. UA25 glitchy as heck till eth0 is
disabled. I have a dell inspiron 5100 running gentoo btw.
Makes icecast streaming jackd a bit tough, but at least it works.
11:12 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Fri, Jul 29, 2005 at 02:38:35AM +0200, Cesare wrote:
Ryan wrote:
Is this perchance built in to the motherboard? I'm having issues with an
athlon64 box on an asus a8v with one of these marvell gigabit nics built
in. This is with a digigram mixart 8 aes/ebu card. Jack is reporting
xruns fairly reliable when I switch windows in openbox.
I'll try disabling the nic tomorrow and see if that changes things at
all here.
On Thu, 2005-07-28 at 15:26 -0400, Lee Revell
wrote:
Marvell Yukon Gigabit Ethernet 10/100/1000 On Thu, 2005-07-28 at 02:59 -0500, Ryan wrote:
Ethernet controller: Broadcom Corporation BCM4401 100Base-T
(rev 01)
Interrupt: pin A routed to IRQ 7
(b44 module, marked experimental since 2.4.x days)
USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M)
USB UHCI Controller
Interrupt: pin [A, B, C, d] routed to IRQ 11
This is an UHCI controller... I thought the usb kernel stuff was going
to get some attention, something about top-half/bottom-half irq
handling? Afaik that hasn't happened yet.
-ry On Thu, 2005-07-28 at 08:07 +0200, Cesare
Marilungo wrote:
>I solved the problem *cough* disabling the ethernet card. Don't know why
>it worked randomly yesterday.
I can confirm this problem/solution. UA25 glitchy as heck till eth0 is
disabled. I have a dell inspiron 5100 running gentoo btw.
Makes icecast streaming jackd a bit tough, but at least it works.
Are you both using
the same network driver/hardware? Does it share an
interrupt with the USB? I'm in Window$ now( developing in Flash, sigh!),
can't check the irqs.
But I remember that I checked for this and the irqs were different.
It's also interesting that I got half the latency switching from
kernel-2.6.10 to 2.6.11. The latter is x86_64, BTW. And I can't get such
low latency on MacOSX or XP (neither with the Egosys PCI card). And this
is a USB card.
athlon 64 3200+ here. though at the moment the kernel is compiled with
CONFIG_MK7=y. I'll try MK8 tomorrow as well.
this is linux 2.6.12.3 on debian sarge.
--
Eric Dantan Rzewnicki | Systems Administrator
Technical Operations Division | Radio Free Asia
2025 M Street, NW | Washington, DC 20036 | 202-530-4900
CONFIDENTIAL COMMUNICATION
This e-mail message is intended only for the use of the addressee and
may contain information that is privileged and confidential. Any
unauthorized dissemination, distribution, or copying is strictly
prohibited. If you receive this transmission in error, please contact
network(a)rfa.org.
11:48 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
Eric Dantan Rzewnicki wrote:
On Fri, Jul 29, 2005 at 02:38:35AM +0200, Cesare
wrote:
Yep. asus k8v-x with athlon 64 3000+.
No xruns after disabling the ethernet interface. Kernel is the one
shipped in Suse. I think it is fair to compile just the realtime module
with the patch applied to source. I still have the capability module but
don't load it.
With the nvidia driver some xruns appear at -p64 -n4 but just in some
critical occasions.
c.
Ryan wrote:
Is this perchance built in to the motherboard? I'm having issues with an
athlon64 box on an asus a8v with one of these marvell gigabit nics built
in. This is with a digigram mixart 8 aes/ebu card. Jack is reporting
xruns fairly reliable when I switch windows in openbox.
I'll try disabling the nic tomorrow and see if that changes things at
all here.
On Thu, 2005-07-28 at 15:26 -0400, Lee Revell
wrote:
Marvell Yukon Gigabit Ethernet 10/100/1000
On Thu, 2005-07-28 at 02:59 -0500, Ryan wrote:
>On Thu, 2005-07-28 at 08:07 +0200, Cesare Marilungo wrote:
>
>
>>I solved the problem *cough* disabling the ethernet card. Don't know why
>>it worked randomly yesterday.
>>
>>
>I can confirm this problem/solution. UA25 glitchy as heck till eth0 is
>disabled. I have a dell inspiron 5100 running gentoo btw.
>Makes icecast streaming jackd a bit tough, but at least it works.
>
>
Are you both using the same network driver/hardware? Does it share an
interrupt with the USB?
Ethernet controller: Broadcom Corporation BCM4401 100Base-T (rev 01)
Interrupt: pin A routed to IRQ 7
(b44 module, marked experimental since 2.4.x days)
USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M)
USB UHCI Controller
Interrupt: pin [A, B, C, d] routed to IRQ 11
This is an UHCI controller... I thought the usb kernel stuff was going
to get some attention, something about top-half/bottom-half irq
handling? Afaik that hasn't happened yet.
-ry
I'm in
Window$ now( developing in Flash, sigh!), can't check the irqs.
But I remember that I checked for this and the irqs were different.
It's also interesting that I got half the latency switching from
kernel-2.6.10 to 2.6.11. The latter is x86_64, BTW. And I can't get such
low latency on MacOSX or XP (neither with the Egosys PCI card). And this
is a USB card.
athlon 64 3200+ here. though at the moment the kernel is compiled with
CONFIG_MK7=y. I'll try MK8 tomorrow as well.
this is linux 2.6.12.3 on debian sarge.
29 Jul
29 Jul
12:04 a.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Fri, 2005-07-29 at 03:48 +0200, Cesare wrote:
No xruns after disabling the ethernet interface.
Kernel is the one
shipped in Suse.
Those xruns are probably doe to non-preemptible network softirqs in the
mainline kernel and would go away with the realtime-preempt patch.
Lee
1:11 a.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
Lee Revell wrote:
On Fri, 2005-07-29 at 03:48 +0200, Cesare wrote:
I've tried also with a vanilla 2.6.11.7 with both the realtime preempt
and the realtime-lsm patch. And got even more xruns.
But maybe I haven't disabled some daemon or something else.
c.
No xruns after disabling the ethernet interface.
Kernel is the one
shipped in Suse.
Those xruns are probably doe to non-preemptible network softirqs in the
mainline kernel and would go away with the realtime-preempt patch.
Lee
1:53 a.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Fri, 2005-07-29 at 05:11 +0200, Cesare wrote:
I've tried also with a vanilla 2.6.11.7 with both
the realtime
preempt
and the realtime-lsm patch. And got even more xruns.
But maybe I haven't disabled some daemon or something else.
Please verify that it's not an NPTL issue, by running JACK and your
clients with LD_ASSUME_KERNEL=2.4.19.
Lee
1:54 a.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Fri, 2005-07-29 at 05:11 +0200, Cesare wrote:
I've tried also with a vanilla 2.6.11.7 with both
the realtime preempt
and the realtime-lsm patch. And got even more xruns.
But maybe I haven't disabled some daemon or something else.
If you still get xruns with the realtime preempt kernel then enable the
latency tracing options and see what is in /proc/latency_trace.
Lee
2:28 a.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
Lee Revell wrote:
On Fri, 2005-07-29 at 05:11 +0200, Cesare wrote:
I prefer not to patch and recompile the kernel for this installation, I
need at least a stable linux on my hd to show to my friends when they
come to my house/studio :-) . I'll experiment with a fresh install on
another disk partition. I can cope with having ethernet disabled. I
don't need it when I'm making music. But I'll try to have both the
ethernet and jack without xruns for testing purposes.
Thanks again.
c.
I've tried also with a vanilla 2.6.11.7 with
both the realtime preempt
and the realtime-lsm patch. And got even more xruns.
But maybe I haven't disabled some daemon or something else.
If you still get xruns with the realtime preempt kernel then enable the
latency tracing options and see what is in /proc/latency_trace.
Lee
7:52 a.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Fri, 29 Jul 2005 05:11:22 +0200
Cesare <cesare(a)poeticstudios.com> wrote:
I've tried also with a vanilla 2.6.11.7 with both
the realtime preempt
and the realtime-lsm patch. And got even more xruns.
But maybe I haven't disabled some daemon or something else.
Hmm,
a realtime preempt kernel alone will not nessecarily do any good. You
need to tune the irq handlers. If your network irq handler has a higher
RT prio than the soundcard irq handler, xruns are preprogrammed.
If you run RP make sure that:
a] the soundcard irq has the highest prio in the system (i.e. around 90)
b] all other irq's are left near their default (around 50)
c] jackd's prio is inbetween (i use 70 with great results)
If you then still get xruns, the hw is to blame [imho]. Or maybe a bug
in RP patches. Use the latency diagnosis tools that the RP kernel
provides to find out what part of the kernel produces the delays.
Regards,
Flo
--
Palimm Palimm!
http://affenbande.org/~tapas/
2:44 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Fri, Jul 29, 2005 at 11:52:37AM +0200, Florian Schmidt wrote:
On Fri, 29 Jul 2005 05:11:22 +0200
Cesare <cesare(a)poeticstudios.com> wrote:
Hi Flo,
Is the wiki in a decently up to date state on this stuff?
I'll be away next week, but starting August 8th, I _will_ follow through
on chasing this down for the boxes I've got here.
For now, I'm going to disable the NICs and put in eepros.
--
Eric Dantan Rzewnicki | Systems Administrator
Technical Operations Division | Radio Free Asia
2025 M Street, NW | Washington, DC 20036 | 202-530-4900
CONFIDENTIAL COMMUNICATION
This e-mail message is intended only for the use of the addressee and
may contain information that is privileged and confidential. Any
unauthorized dissemination, distribution, or copying is strictly
prohibited. If you receive this transmission in error, please contact
network(a)rfa.org.
I've tried also with a vanilla 2.6.11.7 with
both the realtime preempt
and the realtime-lsm patch. And got even more xruns.
But maybe I haven't disabled some daemon or something else.
Hmm,
a realtime preempt kernel alone will not nessecarily do any good. You
need to tune the irq handlers. If your network irq handler has a higher
RT prio than the soundcard irq handler, xruns are preprogrammed.
If you run RP make sure that:
a] the soundcard irq has the highest prio in the system (i.e. around 90)
b] all other irq's are left near their default (around 50)
c] jackd's prio is inbetween (i use 70 with great results)
If you then still get xruns, the hw is to blame [imho]. Or maybe a bug
in RP patches. Use the latency diagnosis tools that the RP kernel
provides to find out what part of the kernel produces the delays.
3:40 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Fri, 2005-07-29 at 12:43 -0400, Eric Dantan Rzewnicki wrote:
Is the wiki in a decently up to date state on this
stuff?
I think all the steps he mentioned are in the wiki. But it's a bad idea
to expect any realtime preempt related site to be accurate. It changes
too fast. If you set out to document every single feature, by the time
you finished, half the doc would be obsolete. Then by the time you get
up to speed on the changes since you started the docs, 75% of your doc
would now be obsolete. Etc. Better off just spending your time testing
and writing code, to get the damn thing finished faster.
Once realtime preempt is in the Linus kernel, then it might be a good
idea to document it, because at that point they are at least committed
to keeping the userspace API stable.
Lee
4:03 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Fri, Jul 29, 2005 at 01:40:36PM -0400, Lee Revell wrote:
On Fri, 2005-07-29 at 12:43 -0400, Eric Dantan
Rzewnicki wrote:
Ok. understood regarding documentation. Can I rely on you (and perhaps
Florian) to get me up to speed on the state of things when I get back on
the 8th? i.e. "just ask here" ...?
I really appreciate all the efforts you and Florian and others have put
into pressing these issues with the kernel developers and helping others
here to test. I've tried a number of times since 2.6 came out to
participate in testing the pre-empt stuff, but mostly on my own time at
home. Often I just didn't have time to follow through as I would have
very much liked to be able. Now I'm actually _tasked_ with working on
this stuff for work, so I think I can justifiably commit significant
hours to it.
--
Eric Dantan Rzewnicki | Systems Administrator
Technical Operations Division | Radio Free Asia
2025 M Street, NW | Washington, DC 20036 | 202-530-4900
CONFIDENTIAL COMMUNICATION
This e-mail message is intended only for the use of the addressee and
may contain information that is privileged and confidential. Any
unauthorized dissemination, distribution, or copying is strictly
prohibited. If you receive this transmission in error, please contact
network(a)rfa.org.
Is the wiki in a decently up to date state on
this stuff?
I think all the steps he mentioned are in the wiki. But it's a bad
idea
to expect any realtime preempt related site to be accurate. It changes
too fast. If you set out to document every single feature, by the time
you finished, half the doc would be obsolete. Then by the time you get
up to speed on the changes since you started the docs, 75% of your doc
would now be obsolete. Etc. Better off just spending your time testing
and writing code, to get the damn thing finished faster.
Once realtime preempt is in the Linus kernel, then it might be a good
idea to document it, because at that point they are at least committed
to keeping the userspace API stable.
4:19 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Fri, 2005-07-29 at 14:03 -0400, Eric Dantan Rzewnicki wrote:
Ok. understood regarding documentation. Can I rely on
you (and perhaps
Florian) to get me up to speed on the state of things when I get back on
the 8th? i.e. "just ask here" ...?
Sure.
FWIW I think a lot of the remaining issues are in the USB layer or the
low level drivers, because basically all PCI devices seem to be able to
operate xrun free down to the hardware limits these days.
2.6.13 is supposed to fix some of these problems. Others may be fixed
in ALSA CVS.
Also FWIW, my Windows using friends say that XP SP2 basically broke a
lot of firewire audio interfaces, so we're not alone in having these
problems...
Lee
2:37 p.m.
New subject: [linux-audio-user] jack xruns on SUSE 9.3 with Edirol UA25 card - SOLVED
On Fri, Jul 29, 2005 at 03:48:03AM +0200, Cesare wrote:
Eric Dantan Rzewnicki wrote:
Yep. asus k8v-x with athlon 64 3000+.
Ok. thanks for confirming.
Marvell
Yukon Gigabit Ethernet 10/100/1000
Is this perchance built in to the motherboard?
I'm having issues with an
athlon64 box on an asus a8v with one of these marvell gigabit nics built
in. This is with a digigram mixart 8 aes/ebu card. Jack is reporting
xruns fairly reliable when I switch windows in openbox.
I'll try disabling the nic tomorrow and see if that changes things at
all here. No xruns after disabling the ethernet interface.
Kernel is the one
shipped in Suse. I think it is fair to compile just the realtime module
with the patch applied to source. I still have the capability module but
don't load it.
With the nvidia driver some xruns appear at -p64 -n4 but just in some
critical occasions.
>>I'm in Window$ now( developing in Flash, sigh!), can't check the irqs.
>>But I remember that I checked for this and the irqs were different.
>>It's also interesting that I got half the latency switching from
>>kernel-2.6.10 to 2.6.11. The latter is x86_64, BTW. And I can't get such
>>low latency on MacOSX or XP (neither with the Egosys PCI card). And this
>>is a USB card.
>athlon 64 3200+ here. though at the moment the kernel is compiled with
>CONFIG_MK7=y. I'll try MK8 tomorrow as well.
>this is linux 2.6.12.3 on debian sarge.
--
Eric Dantan Rzewnicki | Systems Administrator
Technical Operations Division | Radio Free Asia
2025 M Street, NW | Washington, DC 20036 | 202-530-4900
CONFIDENTIAL COMMUNICATION
This e-mail message is intended only for the use of the addressee and
may contain information that is privileged and confidential. Any
unauthorized dissemination, distribution, or copying is strictly
prohibited. If you receive this transmission in error, please contact
network(a)rfa.org.
28 Jul
28 Jul
3:51 p.m.
Cesare Marilungo wrote:
Ubuntu had 2.6.10 something with realtime-lsm module.
On Suse I haven't recompiled the kernel. Just the realtime module, with
the patched kernel.
It's the one shipped with Suse 9.3 (2.6.11.4-21.7-default).
Hi Cesare,
for the one who tried already a few things, and didn't succeed: Which
patches did you apply, where can I find the working instructions
(besides the rest of this thread)? Have a RME Multiface (PCMCIA), and
'would like to get rid of the xruns.
Regards,
Georg
10:43 p.m.
Georg Rudolph wrote:
Cesare Marilungo wrote:
http://sourceforge.net/mailarchive/forum.php?thread_id=7634192&forum_id…
c.
Ubuntu had 2.6.10 something with realtime-lsm
module.
On Suse I haven't recompiled the kernel. Just the realtime module, with
the patched kernel.
It's the one shipped with Suse 9.3 (2.6.11.4-21.7-default).
Hi Cesare,
for the one who tried already a few things, and didn't succeed: Which
patches did you apply, where can I find the working instructions
(besides the rest of this thread)? Have a RME Multiface (PCMCIA), and
'would like to get rid of the xruns.
Regards,
Georg
7059
days inactive
7061
days old
linux-audio-user@lists.linuxaudio.org
23 comments
7 participants
participants (7)
-
Cesare -
Cesare Marilungo
-
Eric Dantan Rzewnicki -
Florian Schmidt -
Georg Rudolph -
Lee Revell -
Ryan