[ECOS] Re: Fwd: RealTek 8139 ethernet driver

[Eric Doenges]

Gary Thomas wrote:

> Better, but perhaps even more descriptive. For example, the 82559
> driver (arguably not the best driver, but useful) uses something like
>   "struct i82559* p_i82559"
> Then uses of p_i82559-> pretty clearly mean "the pointer to the info
> kept about this 82559 device".

I've now changed it to 'rltk8139_info'.

[ ... ]
>>> * I don't think that the PCI lookup table belongs in the code specific
>>>   to the use/instance.  The PCI code(s) [and thus boards] supported by
>>>   the driver will be the same whether this driver is used by a PC or
>>>   an ARM based target.
>>
>>I agree with you in principle, but since vendor and device ID can be set
>>by the serial eeprom connected to the 8139, I'm not 100% shure about
>>this. There seem to be some vendors out there who don't understand
>>the whole point behind PCI device and vendor IDs and use whatever they
>>fancy.
> 
> 
> That would be an indication of broken hardware (IMHO).  Nonetheless, 
> which devices (IDs) your driver is ready to handle is an attribute of
> the generic driver, not the specific instance, so I still think this 
> table belongs there.

After some soul-searching, I decided to agree with you and changed it.

[ ... ]

> Thanks for your efforts.  I look forward to seeing your changes.

I've just posted them to ecos-patches.
-- 
--------------------------------------------------------------------
|     Eric Doenges              |     DynaPel Laboratories GmbH    |
|     Tel: +49 89 962428 23     |     Fraunhoferstrasse 9/2        |
|     Fax: +49 89 962428 90     |     D - 85737 Ismaning, Germany  |
--------------------------------------------------------------------


-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss