patch for acpi_dock

[Nate Lawson]

I agree, I was just wondering if he found some dock device that behaved 
differently than the standard says.  Not that there *ever* has been one 
of those before.  ;-)

Moore, Robert wrote:
> Yes, if _STA is not present, device is assumed to be "present" and
> "functional".
> 
> Bob
> 
> 
>> -----Original Message-----
>> From: owner-freebsd-acpi at freebsd.org [mailto:owner-freebsd-
>> acpi at freebsd.org] On Behalf Of Nate Lawson
>> Sent: Thursday, June 22, 2006 11:15 AM
>> To: iwasaki at freebsd.org
>> Cc: acpi at freebsd.org
>> Subject: patch for acpi_dock
>>
>> Iwasaki-san,
>>
>> I've done some minor cleanups in acpi_dock, please make sure it still
>> works for you as I realize my docking station (T23) is handled by SMI
>> even though an acpi_dock0 device appears.
>>
>> Attached is a patch that improves it a little also.  The main changes
>> are getting rid of the global acpi_dock_status and changing _STA
> behavior.
>> For the first one, it seems the goal was to prevent duplicate
> attachment
>> of docking station devices and duplicate notifies.  Duplicate
> attachment
>> should never happen since newbus probe/attach should prevent reprobe
>> after we return 0.  Is that not the case for you?  For the second one,
>> it seems the locking and ordering on the taskq should be sufficient
> that
>> even if multiple notifies come in, we will handle them sequentially.
> Is
>> there something I'm missing here?
>>
>> For _STA behavior, I added a check for the _STA method not being
>> present.  I think even for dock devices, if the method is not present,
>> the device is always there as part of the docking station.  What do
> you
>> think?
>>
>> Thanks,
>> --
>> Nate


-- 
Nate