[ardour-users] [Fwd: Re: Is 64 bit dual core supported and stable yet?]

Fernando Lopez-Lezcano nando at ccrma.Stanford.EDU
Sat Dec 3 13:43:01 PST 2005


On Sat, 2005-12-03 at 10:49 -0500, Joel White wrote:
> Fernando Lopez-Lezcano wrote:
> >On Fri, 2005-12-02 at 11:20 -0500, George Chan wrote:
> >>I finally decided on what system to build, that included a dual core
> >>Athlon.  
> >>
> >>Then I received this email.
> >>
> >>So, I guess, that for now, it would be better to have a Dual CPU Single
> >>Core Opteron system?
> >>
> >>What do you think?
> >
> >The latest patch for realtime preemption from Ingo Molnar does not use
> >the TSC for internal timing in the kernel in the case of the dual core
> >Athlon X2 cpus (2.6.14-rt12). So the kernel problems do not happen (in
> >my case I experienced sporadic key repeat problems and random temporary
> >activations of the screensaver). 
> >
> >A patch is needed for Jack so that it does not use the TSC for timing
> >with those cpus, some stuff has been posted to the jackit-devel list but
> >no definite patch is yet on CVS. 
> >  
> >
> I am currently running stock 2.6.14 with the 3800+ X2 (that's me, 
> below).  If I set the smp_affinity of my sound card to one processor, 
> the "delay of xx usecs exceeds.." messages virtually disappear.  Without 
> that, they pop up regularly, with increasing delay times.

Yup, that's the problem of drifting TSC's. Hmmm, most probably you want
to pin the jack processes to one processor, not the soundcard. 

> As a test, I just got through recording 55+ min of 8 tracks (-r 44100 -p 
> 64 -n 2, duplex) before ardour halted with an xrun and "delay.." error.  

I sounds to me that -p64 is too low for the stock kernel in any case. 
-- Fernando

> Is it likely that this is still due to the problems mentioned above and 
> in the jackit-devel list?  I don't see any other reason for this 
> particular xrun/delay (I wasn't doing anything to stress the system).  
> Any ideas would be appreciated.  (I tried running with acpi=off, as well 
> as building the kernel without power management as Paul suggested to me, 
> but the system hung up, so I'm back to acpi).





More information about the Ardour-Users mailing list