Interrupt latency problems

[Dan Nelson]

In the last episode (May 19), Lars Eggert said:
> Paul Richards wrote:
> > I'm also not sure why a lot more people aren't seeing this, unless
> > my motherboard is odd in sharing the acpi interrupt with a pci
> > slot.
> 
> I tried to see if this is the case on my SMP board, but I don't seem
> to have acpi attached to any interrupt:
> 
> [root at nik: ~] ps -axg | grep irq
>    24  ??  WL     0:00.00  (irq18: mpt0)
>    25  ??  WL     0:19.30  (irq19: em0 mpt1)
>    26  ??  WL     0:00.00  (irq16: xl0 uhci1)
>    27  ??  WL     1:57.07  (irq17: fwohci0 drm0)
>    28  ??  WL     1:09.27  (irq13: em1 ichsmb0)
>    29  ??  WL     0:00.02  (irq20: pcm0)
>    30  ??  WL     0:02.19  (irq2: uhci0 bktr0)
>    31  ??  WL     0:02.84  (irq14: ata0)
>    32  ??  WL     0:00.00  (irq15: ata1)
>    36  ??  WL     0:00.00  (irq6: fdc0)
>    37  ??  WL     0:00.00  (irq1: atkbd0)
> 
> Is this the right way to check for this?

You sure you have acpi loaded?  It shows up for me:

$ ps ax | grep acpi0
   22  ??  WL     0:00.00  (irq9: acpi0)

Maybe the attach message to acpi should print the IRQ it's using.  All
I get in dmesg is "acpi0: <ASUS   CUV4X-D > on motherboard".

-- 
	Dan Nelson
	dnelson at allantgroup.com