On Sun, Mar 05, 2006 at 02:02:18AM -0800, Florin Andrei wrote: Does anyone know if dual HyperThreaded Xeon's might impact this at all? I've got a user reporting jackd dieing on such hardware with an rme digi96/8 pad. It happens to be sharing and IRQ between these things: 10: 38948682 0 0 0 XT-PIC libata, RME96, uhci_hcd:usb1, eth0 Haven't had a chance to get physical access to the box, yet, to do some thorough sorting of the issue. But, just wondered if this might in anyway be a possible source of problems. -- Eric Dantan Rzewnicki Apprentice Linux Audio Developer and Mostly Harmless Sysadmin (text below this line mandated by management) Technical Operations Division | Radio Free Asia 2025 M Street, NW | Washington, DC 20036 | 202-530-4900 CONFIDENTIAL COMMUNICATION This e-mail message is intended only for the use of the addressee and may contain information that is privileged and confidential. Any unauthorized dissemination, distribution, or copying is strictly prohibited. If you receive this transmission in error, please contact network(a)rfa.org.

On Sun, 05 Mar 2006 02:02:18 -0800 Florin Andrei <florin(a)andrei.myip.org> wrote: Hi, this might work out in the most situations, but of course it depends on what kind of devices share the IRQ. Imagine a situation where the ide controllers and the sound interface are on the same irq. Now, if you do heavy disk I/O there's a very real chance of the audio operation getting disturbed, especially at low latency settings. If the ide controllers and the audio interface were on different IRQ's one could use the irq handler thread priorities of a -rt kernel to practically guarantee xrun free operation even under heavy disk load (simply by giving jack and the audio irq higher prios than the ide controller irqs). With a shared IRQ this option is not availale anymore. So, depending on what you want from your box, non shared IRQs can make sense and can have real advantages. Flo -- Palimm Palimm! http://tapas.affenbande.org