lkml.org 
[lkml]   [2006]   [Nov]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: 2.6.19-rc6-rt8: alsa xruns
From
Date
On Tue, 2006-11-28 at 13:04 -0800, Fernando Lopez-Lezcano wrote:
> On Tue, 2006-11-28 at 12:37 -0800, Fernando Lopez-Lezcano wrote:
> > On Tue, 2006-11-28 at 21:09 +0100, Ingo Molnar wrote:
> > > * Fernando Lopez-Lezcano <nando@ccrma.Stanford.EDU> wrote:
> > >
> > > > Hi, I'm trying out the latest -rt patch and getting alsa xruns when
> > > > using jackd and jack clients. This is a sample from the output of
> > > > qjackctl / jackd (jack 0.102.25, qjackctl 0.2.21):
> > >
> > > > ( japa-4096 |#0): new 17 us maximum-latency wakeup.
> > > > ( beagled-3412 |#1): new 19 us maximum-latency wakeup.
> > > > ( IRQ 18-1081 |#1): new 26 us maximum-latency wakeup.
> > > > ( snd-4040 |#1): new 1107 us maximum-latency wakeup.
> > > > ( japa-4096 |#0): new 1445 us maximum-latency wakeup.
> > > > ( japa-4096 |#0): new 2110 us maximum-latency wakeup.
> > > > ( qjackctl-4038 |#1): new 2328 us maximum-latency wakeup.
> > > > ( japa-4096 |#0): new 2548 us maximum-latency wakeup.
> > > > ( IRQ 18-1081 |#0): new 10291 us maximum-latency wakeup.
> > >
> > > hm, lets fix this. Could you enable tracing (on the yum rpm) via:
> > >
> > > echo 1 > /proc/sys/kernel/trace_enabled
> > >
> > > does /proc/latency_trace have any meaningful events included for such a
> > > long delay? If not then it would be nice to rebuild the kernel with
> > > CONFIG_LATENCY_TRACING - and in any case my previous suggestion holds
> > > too: booting with maxcpus=1 to reproduce the latencies will give easier
> > > to interpret latency traces.
> >
> > Sorry, it looks like it is an smp issue. Booting with maxcpus=1 reduces
> > the xrun reports significantly (only three so far but very short, in the
> > range of 0.029 to 0.041 ms). The long ones seem to have gone away, so
> > far...
>
> Strange, I rebooted smp and I'm not seeing the very long xruns I was
> seeing before, only short ones as reported above. Here are some traces,
> but nothing that makes sense I think.
>
> I'll turn off the machine and cold boot it...)

No difference, actually it looks like the regression re-regresses if I
enable the trace... Arghhh.

Toggling /proc/sys/kernel/trace_enabled makes the long xruns reported by
jack come and go.

-- Fernando


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2006-11-28 22:39    [W:0.062 / U:0.696 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site