Re: [PATCH v2] Input: drivers/joystick: use parallel port device model

[Dmitry Torokhov <dmitry.torokhov@gmail.com>]

On Thu, Aug 13, 2015 at 07:16:14PM +0530, Sudip Mukherjee wrote:
> On Sat, Aug 08, 2015 at 04:05:15PM +0530, Sudip Mukherjee wrote:
> > On Tue, Aug 04, 2015 at 07:55:51PM +0530, Sudip Mukherjee wrote:
> > > Modify db9 driver to use the new Parallel Port device model.
> > > 
> > > Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> > > ---
> > > 
> > > v2: an extra check was removed. db9_base could not be removed. Already
> > > mailed you the reasons, reference is at:
> > > https://lkml.org/lkml/2015/8/3/403
> > 
> > Hi Dmitry,
> > I was telling you on the other thread that to remove db9_base we will
> > need to modify the parport code and then it has to be tested by everyone
> > again and that will also change the old behaviour of parport. The v2 to
> > which I am replying now is still having that db9_base.
> > 
> > Please see the following patch, this is the minimum change required to
> > update db9 driver. Personally I donot like this approach as in this
> > method if a parallel port is hotplugged after this driver is loaded then
> > it will not work. Please let me know your views on the posted v2 and
> > also on the following proposal.
> 
> Hi Dmitry,
> Did you have a chance to look at this?

Right, so I agree that we should allow for hotplug and so the v2 is in
that sense is better than v3.

I still would prefer for the core to keep track of relationship between
devices and drivers, but if you would prefer to postpone implementing it
that woudl be fine.

> Pali Rohár has the hardware and he is ready to test it.

Pali, could you give the v2 version of the patch a spin
(https://patchwork.kernel.org/patch/6940321/)?

Thanks.

-- 
Dmitry