PCI1620 with an integrated memory card reader

[Peter Stuge]

On Sat, Jul 28, 2007 at 11:25:39PM +0300, Cristian Onet wrote:
> I have realized that the problem was the way that the PIO data
> transfer was made in pata_pcmcia. It was on 16 bits and it seems
> that the emulated ATA device only handled 8 bit transfers

Good find!


> I would really like to share this with other users of this device
> but I don't really know how. Alan (the author of pata_pcmcia)
> suggested that the modifications would be better in a new module
> (which I named pata_1620).

Yeah, that's a good idea. It would also be a natural home for the
firmware loading.

Might be worth a shot to clean up the driver (I would call it
pata_pci1620) and try to get it included in the kernel proper.


> I hope it will help whoever needs it. For now the process to set up
> the device is very unfriendly but it could be made better.

I haven't checked your wiki page but I've seen plenty of out-of-tree
modules and setup isn't that awkward; just include a Makefile that
builds against /lib/modules/$(uname -r)/build and most would get it
to work out of the box.


> The first issue would be if the firmware could be distributed with
> this Linux driver. This raises some copyright issues... I think.

This is a licensing issue that you would have to discuss with the
firmware author. I doubt TI will allow re-distribution, but I hope
I'm wrong. :)

Also, if they do, I don't see Linux distributing the firmware - that
would have to be done by you.

Anyway, you can always provide instructions and helper utilities to
extract the firmware from the Windows driver which is probably
distributed free of charge by TI already.


//Peter