Re: [Jack-Devel] BUG REPORT : Overflow of XRUNs after a certain?time

On Sat, May 26, 2012 at 10:39:25PM +0100, John Rigg wrote:
> 
> aCOSwt wrote on Fri, 25 May 2012 at 09:22:20 +0200:
> > If the sampling frequency is set to 44100, then the problem occurs after about> 13h 31mn 35s
> > If the sampling frequency is set to 48000, then the problem occurs after about> 12h 25mn 36s
> 
> 
> I can confirm that this behaviour does occur on my system with the same
> sound card.
> 
> It was running at 96kHz and I wasn't timing it accurately, but jackd
> was running perfectly until about 6hr 13min after starting, when it suddenly
> started behaving as described at the start of this thread. That would
> tie in with the above numbers.
> 
> Tomorrow if I get time I'll run the same test with one of my Delta 1010s
> which are also ice1712 based.

The Delta 1010 exhibits the same behaviour. In the past I have run it for
longer than this without the problem occurring. Unfortunately that was a
long time ago (c. 2007) and many changes have been made in the kernel,
ALSA and other software since then. I hadn't encountered this problem
in the meantime as I don't usually run jackd for more than about 8 hours
at a time and stick to 44.1 or 48kHz sample rate.

The fact that people are running jackd for long periods without a problem
with other hardware would suggest that it isn't jackd that's at fault. If
it's a hardware bug it's one which is triggered by some software change
made in the last few years.

Current software here:
Debian testing amd64 (pre multi-arch snapshot)
Standard kernel.org 3.2.16 SMP PREEMPT x86_64 kernel with HZ=1000
alsa-driver-1.0.24, alsa-lib-1.0.24.1
jack 0.121.3

I have an old Debian 5.0 based DAW system on a spare HDD which I'll test
next.

John

1338128260.25074_0.ltw:2, <20120527142420.GA2546 at localhost0 dot localdomain>