Re: [LAU] Too many xruns

On 5 September 2012 at 21:41, Jeremy Jongepier <jeremy@email-addr-hidden> wrote:

> On 09/05/2012 09:24 PM, Kevin Cosgrove wrote:
> >
> > Xrunning system says:
> >
> > 0 [PCH ]: HDA-Intel - HDA Intel PCH
> > HDA Intel PCH at 0xf7e20000 irq 70
> > 1 [M1010 ]: ICE1712 - M Audio Delta 1010
> > M Audio Delta 1010 at 0xa040, irq 16
> > 2 [NVidia ]: HDA-Intel - HDA NVidia
> > HDA NVidia at 0xf7080000 irq 17
> >
> > Fine system says:
> >
> > 0 [Intel ]: HDA-Intel - HDA Intel
> > HDA Intel at 0xfbff8000 irq 30
> > 1 [M1010 ]: ICE1712 - M Audio Delta 1010
> > M Audio Delta 1010 at 0xec00, irq 17
> >
> > I'm using the Delta 1010 cards on each system with Jack & Ardour.
> >
>
> Hello Kevin, thanks for the results!
>
> Ok, that looks good, so it's about a Delta 1010 card which uses the
> snd_ice1712 kernel module.
>
> >> cat /proc/interrupts
> > 16: 56322 0 0 0 0 0 0 0 IO-APIC-fasteoi ahci, ehci_hcd:usb1, firewire_ohci, snd_ice1712, nvidia
>
> There you go, looks kinda busy there. Looks a bit like the situation on
> my notebook:
>
> 16: 320312 30 IO-APIC-fasteoi uhci_hcd:usb3, mmc0,
> firewire_ohci, jmb38x_ms:slot0, eth1, nvidia
>
> But I manage to use to run a FireWire audio card just fine on this
> notebook so there's still hope for your Delta 1010 ;)
>
> >
> > Fine system says:
> > 17: 0 0 0 0 IO-APIC-fasteoi ICE1712

I temporarily moved my PCI Delta 1010 card into a PCIe slot, via a
PCIe to PCI adapter card. I *really* don't like how that works out
mechanically. For one, the S/PDIF connectors end up covered up by
the case. The mechanical support for the card is not very good
either.

But, all the xruns are now gone. That would tend to support the
theory that this is an interrupt crowding issue. ;-)

On this board I have no freedom to change the interrupt assignment.
The video card I have now takes two slots worth of space, though it
only plugs into one slot. That means that I can't move it to a
better location. I might buy a smaller video card to get around
that. How much GPU do I need for Linux audio after all?

My next experiment will be to try the CCRMA rtirq stuff to see if
changing the kernel servicing of interrupts will get me good enough
performance that I can just leave the hardware where it was.

Cheerio....

--
Kevin
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/listinfo/linux-audio-user