<br><br><div class="gmail_quote">On Sun, Feb 1, 2009 at 3:08 AM, <span dir="ltr"><<a href="mailto:hollunder@gmx.at">hollunder@gmx.at</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div><div></div><div class="Wj3C7c">On Sun, 01 Feb 2009 04:42:40 +0100<br>
Robin Gareus <<a href="mailto:robin@gareus.org">robin@gareus.org</a>> wrote:<br>
<br>
> -----BEGIN PGP SIGNED MESSAGE-----<br>
> Hash: SHA1<br>
><br>
> Fernando Lopez-Lezcano wrote:<br>
> > On Fri, 2009-01-30 at 03:23 +0100, <a href="mailto:torbenh@gmx.de">torbenh@gmx.de</a> wrote:<br>
> >> On Thu, Jan 29, 2009 at 03:41:48PM -0800, Fernando Lopez-Lezcano<br>
> >> wrote:<br>
> >>> On Tue, 2009-01-27 at 14:06 +0100, Peder Hedlund wrote:<br>
> >>>> Quoting Ken Restivo <<a href="mailto:ken@restivo.org">ken@restivo.org</a>>:<br>
> >>>><br>
> >>>>> And here is the next installment in the saga of trying to get<br>
> >>>>> Ingo RT going on my Asus EEE.<br>
> >>>>><br>
> >>>>> I successfully built and ran the 2.6.26.8-rt12 with the<br>
> >>>>> alsa_seq patch. It ran.<br>
> >>>>><br>
> >>>>> The problem is that neither the Ethernet (atl1e) or wireless<br>
> >>>>> (rt2860sta) work. So I pretty much had to reboot back out of<br>
> >>>>> it immediately.<br>
> >>>> I've been running the standard kernel from openSUSE 11.0 on my<br>
> >>>> Athlon 2000+ and can get down to at least 5.3ms latency on an<br>
> >>>> Audiophile 2496 using the limits.conf "trick".<br>
> >>>><br>
> >>>> Do people really need lower latencies for music purposes or are<br>
> >>>> we just thinking "well, I needed the RT patch three years ago; I<br>
> >>>> ain't stopping now" ?<br>
> >>> It depends on your usage (this question seems to come up every<br>
> >>> couple of months lately). The current kernels are much better in<br>
> >>> low latency applications than three years ago. They are usable if<br>
> >>> you don't require "low" latencies (64 or 128 x 2). What you get<br>
> >>> also strongly depends on the hardware mix you have.<br>
> >>><br>
> >>> If you want to use 64 or 128 frame periods (or less) you probably<br>
> >>> will need at rt patched kernel in most cases. Then again if an<br>
> >>> occasional xrun is not a problem then you would be fine with the<br>
> >>> stock kernel.<br>
> >> i am running with -p64 -n3 on an intel-hda with 2.6.28<br>
> >> of course internal cards have the greatest potential for<br>
> >> lowlatencies. so this might be unfair, compared to pci.<br>
> ><br>
> > Hmmm, I'm not sure, the load on pci itself by a soundcard should be<br>
> > nothing really hard. What would the internal card use? Would not<br>
> > that be pci or pci express anyway?<br>
> ><br>
> surely they are.<br>
><br>
> $ lspci | grep Audio<br>
> 00:1b.0 Audio device: Intel Corporation 82801G (ICH7 Family) High<br>
> Definition Audio Controller (rev 02)<br>
><br>
> The RT patch does two things:<br>
> It allows to prioritize interrupts and it [almost] guarantees<br>
> real-time scheduling for a dedicated process or thread.<br>
><br>
> While the soundcard is low bandwith on the PCI bus, IRQ prio may still<br>
> be required to override HDD and [sometimes] graphics I/O; at least<br>
> when playing or recording many tracks. NTL, you can get a perfect<br>
> x-run free system without the RT patch; you can just not rely on it<br>
> to be as x-run free as a RT patched kernel ;)<br>
><br>
> >> and i havent really seen xruns which i could not relate to some<br>
> >> programm which wasnt RT-safe, and i am compiling stuff most of the<br>
> >> day... though perhaps i am not pushing the DSP load hard enough.<br>
> >><br>
> >> i did not even turn preemptible RCU on.<br>
> >> the latency measurement instrumentation is also in 2.6.28 btw.<br>
> ><br>
> > Well, that's very good news then!<br>
> ><br>
> > I think the last time I tried to use a non-preempt was 2.6.27.x<br>
> > (maybe, I would have to double check). Playing 24 channels in<br>
> > ardour would result in xruns, not very often but they would happen,<br>
> > this is with 128x2 on an RME hdsp card runing on a quad core intel<br>
> > system. I should try again with the latest available.<br>
><br>
> I just booted into a vanilla 2.6.28.2 #1 SMP PREEMPT<br>
><br>
> right, there's no realtime patch, yet running jackd at 64 * 3 @48kSPS<br>
> on a HDA - ardour2 with 12 tracks, a couple of LADSPA effects and<br>
> jamin (lots of CPU!) - there's no xruns yet!<br>
><br>
> I'll be back in the studio in two weeks from now to test it with USB<br>
> and 1394 devices. With <=2.6.24 kernels those were always working more<br>
> reliably that the HDA so I don't expect problems there.<br>
><br>
> BTW. with 2.6.28 I needed to<br>
><br>
> `echo -1 > /proc/sys/kernel/sched_rt_runtime_us`<br>
><br>
> or edit /etc/sysctl.conf and add<br>
> sys.kernel.sched_rt_runtime_us = -1<br>
><br>
> before JACK was able to acquire real-time privileges.<br>
<br>
</div></div>May I ask what that does? This value is 950000 on my system.<br>
I think jack has rt privileges here on 2.6.28.2 but I'm not too certain.<br>
I don't even know how to check that reliably.<br></blockquote><div><br>jackd would not start with -R if realtime permission can not be granted. I think you would know if there was a problem ;)<br>I have to say I don't know either what this option is you are setting, Robin. Is that the granularity of the timer (in u-sec's)? Setting to -1 seems to suggest that the highest possible value is used, although I'm only wildly guessing. Would be great to know!<br>
<br> cheers, <br><br>karsten<br></div></div>