On Wed, Jun 20, 2012 at 11:27:00PM +0200, Simon Budig wrote:

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Henrik

On 06/20/2012 08:40 PM, Henrik Rydberg wrote:

quoted

quoted

+/* + * Part of this code is inspired from a first version of a 
driver for + * this chip sent by Simon Budig to the linux-input 
mailing list + */

Replacing the above with the standard GPL header with both of your
 names ought to resolve any eventual issues, no?

Not sure. Olivier looked at an ancient version of my code, missing
multiple review spins and tests I did in the meantime with Dmitry and
Anatolij.

I obviously would prefer to see my driver in the kernel. Of course I
am open to improvement suggestions, but I am not really enthusiastic
about reviewing a driver that basically ditches most of my work.

quoted

quoted

+ft5x06_setting(threshold, W_REGISTER_THRESHOLD, 20, 80); 
+ft5x06_setting(gain, W_REGISTER_GAIN, 0, 31); 
+ft5x06_setting(offset, W_REGISTER_OFFSET, 0, 31);

Question: will the driver work without fiddling with these 
settings? If so, are they really necessary? Maybe some part could 
be moved to debugfs instead?

Well, these are there in my driver as well (plus the "report rate"
setting of the sensor).

There are some parameters (gain/threshold/offset) which need tweaking
depending on the panel glass thickness and other device-specific
things. At least in the development phase it is necessary to tweak
with these to figure out good settings.

Dunno if sysfs is the best place for this, but I suspect that this is
a parameter that needs to be available at runtime as well, moving it
e.g. to platform data might not cut it.

Anyway. I have reviewed the changes Dmitry suggested and will test
them on real hardware tomorrow.

That would be great as I already reviewed your version. If there is
something missing from your version that is present in Olivier, I woudl
prefer getting an incremental patch.

Thanks.

-- 
Dmitry