Re: CVS commit: src/sys/dev/pci

["Garrett D'Amore" <garrett_damore%tadpole.com@localhost>]

Jared D. McNeill wrote:
> On 23-Feb-06, at 3:47 PM, Tim Rightnour wrote:
>
>>
>> On 23-Feb-2006 Jason Thorpe wrote:
>>> What's wrong with autri specifically matching only the device class
>>> it cares about?
>>
>> I'm ok with that.. but does anyone know for sure what class of device
>> the autri
>> claims to be?   I don't have one.. so I'm not sure..
>>
>> If someone can tell me the class, my plan would be to change autri to
>> return 1
>> if found, 2 if found and class is correct,  and have if_pcn return 2
>> if it
>> finds one and the class is network.
>
> This might not be a big deal, but I'd just like to point out that in
> the event that someone boots a kernel that includes the autri driver
> but not the pcn driver, with this solution the autri driver will still
> try to attach.
>
> I wouldn't expect this to cause any major problems.. just thought I'd
> mention it incase someone thinks otherwise.
Yes, and it seems a lot cleaner to just include the check for the class
code and *reject* the attachment for a mismatched device than to use the
ordering logic.  Just my 2c.

    -- Garrett
>
> Cheers,
> Jared


-- 
Garrett D'Amore, Principal Software Engineer
Tadpole Computer / Computing Technologies Division,
General Dynamics C4 Systems
http://www.tadpolecomputer.com/
Phone: 951 325-2134  Fax: 951 325-2191