Hmmm.  I think that the aria driver should probably be disabled in
GENERIC, since I had a look at my probe routine, and I think that
a little more logic is really necessary in there.  Probably something
was lost when I bus_space(9)'ified it.

Anyway, I don't think that I'll have time before the final code
freeze to have a proper look at the logic in there to make the
probe a bit more robust.

Sorry.

On 924345565 seconds since the Beginning of the UNIX epoch
Matthias Drochner wrote:
>
>
>markus@core.de said:
>> aria0 at isa0 port 0x290-0x297 irq 10panic: intr_etablish: irq 10
>> can't share level-triggered with edge-triggered stopped in at
>> _Debugger 0x4: leaver
>
>>  Does that help? 
>
>I missed the beginning of this thread, so forgive if I repeat
>something already said.
>
>This panic typically happens if an interrupt was assigned
>to a PCI device by the BIOS and then later an ISA device
>is attached which claims the same IRQ.
>So there are 2 posibilities:
>a) You have indeed an "aria" sound board at IRQ10. Then the
>   BIOS should be told that this interrupt is not free for
>   PCI/PnP use - there is a setup page for this stuff.
>b) You don't have an "aria"; the "aria" driver probe gives a false
>   positive. This would be a driver bug, and if this cannot
>   be fixed the driver should be disabled in GENERIC.
>
>best regards
>Matthias

 == Roland Dowdeswell
 == http://www.imrryr.org/~elric/