RFC: making the aiptek.c tablet driver xf86-input-wacom compatible

RFC: making the aiptek.c tablet driver xf86-input-wacom compatible

[Stefan Brüns]

Hi everyone,

I have started to modify the aiptek tablet driver to be compatible with the 
xf86-input-wacom driver.

Motivation:
First, as wacom dominates the market, most configuration frontends are for 
wacom, desktop integration only exists for wacom, and most application 
developers only test with wacom tablets.
Second, the xf86-input-aiptek driver seems to be unmaintained, at least 
opensuse has dropped it from distribution, and xf86-input-evdev is no 
sufficient replacement.

So for me there are two possibilities, either replicating all the work done 
for the wacom driver in the xorg stack and above, or just making the the 
aiptek driver mimic the output of the wacom kernel driver - I have taken the 
second option.

Currently, I am cleaning up my changes, and I am investigating some bugs in 
the original driver parsing the reports from the device.

I have tested my Aiptek 6000U with the xf86-input-wacom driver, and it is 
working fine in Krita, Inkscape and The Gimp, configuring the tablet via
kcm_wacomtablet (KDE configuration module).

If anyone is interested, I will send my patches to this list after the 
cleanup.

Regards,

Stefan


-- 
Stefan Brüns  /  Bergstraße 21  /  52062 Aachen
phone: +49 241 53809034     mobile: +49 151 50412019
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: RFC: making the aiptek.c tablet driver xf86-input-wacom compatible

[Benjamin Tissoires]

Hi Stephan,

On Wed, Jun 26, 2013 at 12:02 AM, Stefan Brüns
<stefan.bruens@rwth-aachen.de> wrote:
> Hi everyone,
>
> I have started to modify the aiptek tablet driver to be compatible with the
> xf86-input-wacom driver.
>
> Motivation:
> First, as wacom dominates the market, most configuration frontends are for
> wacom, desktop integration only exists for wacom, and most application
> developers only test with wacom tablets.
> Second, the xf86-input-aiptek driver seems to be unmaintained, at least
> opensuse has dropped it from distribution, and xf86-input-evdev is no
> sufficient replacement.

Yes the fact is that xf86-input-aiptek is unmaintained due to lack of
testers with the device. xf86-input-evdev is not a sufficient
replacement because it aims at forwarding the raw events from the
kernel without any processing. For tablets, the de facto natural way
is to use the wacom driver, which should be called now "tablet" but is
still called "wacom" for historical reasons.

>
> So for me there are two possibilities, either replicating all the work done
> for the wacom driver in the xorg stack and above, or just making the the
> aiptek driver mimic the output of the wacom kernel driver - I have taken the
> second option.

thanks for choosing the second solution :) I think Peter will join me
on this thank you.
I just gave a quick look at the xf86-input-aiptek sources, and there
is nothing which prevents you to port this device to the wacom driver.

>
> Currently, I am cleaning up my changes, and I am investigating some bugs in
> the original driver parsing the reports from the device.

The device seems to be a HID declared one. I'd be curious to have a
look at the report descriptors to know if the parsing is really
necessary. The reports descriptor can be retrieved by calling "lsusb
-v" when the device is not bound to its usb driver.

>
> I have tested my Aiptek 6000U with the xf86-input-wacom driver, and it is
> working fine in Krita, Inkscape and The Gimp, configuring the tablet via
> kcm_wacomtablet (KDE configuration module).

great!

>
> If anyone is interested, I will send my patches to this list after the
> cleanup.

I think we are. And I think you will also get some advantages in
pushing those patches upstream so you will not have to manually
maintain your own branch :)

Cheers,
Benjamin
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: RFC: making the aiptek.c tablet driver xf86-input-wacom compatible

[Peter Hutterer]

On Wed, Jun 26, 2013 at 10:17:30AM +0200, Benjamin Tissoires wrote:
> Hi Stephan,
> 
> On Wed, Jun 26, 2013 at 12:02 AM, Stefan Brüns
> <stefan.bruens@rwth-aachen.de> wrote:
> > Hi everyone,
> >
> > I have started to modify the aiptek tablet driver to be compatible with the
> > xf86-input-wacom driver.
> >
> > Motivation:
> > First, as wacom dominates the market, most configuration frontends are for
> > wacom, desktop integration only exists for wacom, and most application
> > developers only test with wacom tablets.
> > Second, the xf86-input-aiptek driver seems to be unmaintained, at least
> > opensuse has dropped it from distribution, and xf86-input-evdev is no
> > sufficient replacement.
> 
> Yes the fact is that xf86-input-aiptek is unmaintained due to lack of
> testers with the device. xf86-input-evdev is not a sufficient
> replacement because it aims at forwarding the raw events from the
> kernel without any processing. For tablets, the de facto natural way
> is to use the wacom driver, which should be called now "tablet" but is
> still called "wacom" for historical reasons.

fwiw, wacom is still very much centered around wacom devices. it will work
from a technical perspective with any tablet but there may be the odd thing
here and there that may look different. we'll have to fix that up as we go
in the X driver, but there are already non-wacom tablets that work fine.

specifically, test your tablet with libwacom as well (or the GNOME stack) as
this covers parts of the device-specific client stack for tablets.
 
> > So for me there are two possibilities, either replicating all the work done
> > for the wacom driver in the xorg stack and above, or just making the the
> > aiptek driver mimic the output of the wacom kernel driver - I have taken the
> > second option.
> 
> thanks for choosing the second solution :) I think Peter will join me
> on this thank you.

yes, indeed. less xorg drivers is the future :)

Cheers,
   Peter

> I just gave a quick look at the xf86-input-aiptek sources, and there
> is nothing which prevents you to port this device to the wacom driver.
> 
> >
> > Currently, I am cleaning up my changes, and I am investigating some bugs in
> > the original driver parsing the reports from the device.
> 
> The device seems to be a HID declared one. I'd be curious to have a
> look at the report descriptors to know if the parsing is really
> necessary. The reports descriptor can be retrieved by calling "lsusb
> -v" when the device is not bound to its usb driver.
> 
> >
> > I have tested my Aiptek 6000U with the xf86-input-wacom driver, and it is
> > working fine in Krita, Inkscape and The Gimp, configuring the tablet via
> > kcm_wacomtablet (KDE configuration module).
> 
> great!
> 
> >
> > If anyone is interested, I will send my patches to this list after the
> > cleanup.
> 
> I think we are. And I think you will also get some advantages in
> pushing those patches upstream so you will not have to manually
> maintain your own branch :)
> 
> Cheers,
> Benjamin
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html