Leo, I am sorry. Yes I mean IO-APIC. So basically the output of lspci -v are the same as cat /proc/interrupts.
It is a riser, I will check on that. So here's my questions then. If APIC routes the IRQs to 1-15 for real world use....can you safely have two devices on, say, 14? APIC will assign one to maybe 23 and one to 20. But are they really both on 15 with a potential for conflict? On 2/11/07, Leo Ann Boon <[EMAIL PROTECTED]> wrote:
Matt wrote: > Leo, > Yes I did read this. And I have ACPI turned on. Unfortunately lspci > -vb still is showing devices sharing IRQs. You mean IO-APIC? ACPI is a different beast altogether. lspci -vb and lspci -v should show different results on a proper IO-APIC system. lspci -vb shows what the card thinks it's using. On a XT-PIC system, what it thinks and what it's assigned should be the same. On an IO-APIC system, the interrupts are routed through secondary APICs that can assign additional hardware (local) interrupts (>15) to each card. But, for real-mode compatibility sake, the motherboard is required to route the new interrupts to IRQs 1-15. Hence the different value in lspci -vb. Is the 2950 using a riser? You might want to check if there are jumpers on the riser. Some brain dead risers actually share the IRQ lines unless you change the jumpers. Frankly, I really hope that Digium will change the PCI controller on their cards. I think it's buggy that's why it doesn't work properly with some IO-APIC system. If Sangoma can do it, I don't see why not. Leo _______________________________________________ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
_______________________________________________ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
