Re: [PATCH 1/1] ACPI: pci_irq, add PRT_ quirk for IBM Bartolo

[Jiri Slaby <jirislaby@gmail.com>]

On 07/20/2010 11:29 AM, Jiri Slaby wrote:
> On 07/19/2010 09:19 PM, Robert Hancock wrote:
>> On Mon, Jul 19, 2010 at 10:29 AM, Jiri Slaby <jslaby@suse.cz> wrote:
>>>>> I still no point in comparing this to Windows' setup. We can't find out
>>>>> whether it is quirked or better (without some bug) handled there.
>>>>
>>>> Well, you can see if Windows shows IRQ 10 or 11 for that device..
>>>
>>> But how can I find out which link it is routed to in Windows? Without
>>> that information the number is meaningless, no?
>>
>> If you look at the pattern of which IRQs are shared by what devices in
>> Linux and compare it to Windows you can get a good idea. Normally the
>> assignment of devices to interrupt lines is hard-wired on the
>> motherboard and doesn't change.
> 
> Ok, thanks for the hint.
> 
> What we've found out is that it works on 2.6.27 (with slightly changed
> configuration).

No, this was a false alarm. It never worked with acpi irq routing on
older kernels in this HW configuration.

So, to sum up:
1) acpi routing enabled (no kernel parameter) => ports 4+5 defunct.
ports 4+5+6+7 are all on irq 11

2) acpi routing disabled (acpi=noirq) => all ports working, 4+5 on irq
10, 6+7 on irq 11

3) with the quirk [1] and acpi routing enabled => all ports working,
ports 4+5 on irq 10, 6+7 on irq 11

4) in windows => 4+5+6+7 are all on irq 9 and the ports are all working.

Any ideas what this means? Especially point 4)?

[1] http://lkml.org/lkml/2010/6/27/85

thanks,
-- 
js