headers
Jorge Suárez de Lis | 25 Apr 02:38
Gravatar

Driver for Upek Eikon model no. TCRD4C4 (aka. Eikon II)

Hello,

I've managed to get this device working with libfprint 0.1.0 pre2, with
a modified version of the upekts driver.

There is one huge problem here: this device has the same ID as the
upeksonly supported device, so:

 * we can't simply add support for this device without conflicting with
the upeksonly driver
 * if you just want to make it work, apply the patch and compile without
upeksoly support (or edit upeksonly.c and change the ids).

Please notice this is a hackish modification by guesses based on the USB
traffic gathered from both the original upekts driver and the privative
tool provided by upek for linux. It works for me, but it may not work
for you. Also, with this patch I'm ditching the support for the original
device supported by this driver.

You can find the patch here:
http://sharing.jorgesuarezdelis.name/2010/upekts-upek_eikon_tcr4d.diff

Some thoughts about the IDs problem? What's next? Should I rename the
driver? Or it would be better to change the upekts driver to support
both devices?

Sincerely,

--

-- 
Jorge Suárez de Lis <yo <at> jorgesuarezdelis.name>
(Continue reading)

Permalink | Reply |
headers
Jorge Suárez de Lis | 28 Apr 14:57
Gravatar

Re: Driver for Upek Eikon model no. TCRD4C4 (aka. Eikon II)

Here's an updated patch.

In this patch, instead of replacing the upekts driver, I created a new
driver called upek2e.

The driver is not compiled by default, because of the problems with the
upeksonly driver sharing IDs with this driver. To compile this driver,
you must include it in the --with-drivers option when configurying:

By example: ./configure --with-drivers="upek2e upekts"

Why a new driver? 

* The verification is exactly the same as the upekts one. Copy-paste.
* There are some changes in the enrollment, and surely more changes will
follow since this was only tried on two device variants (and there are
still some issues on both of them).
* The initialization part is almost completely different.

With this driver, I want to create a sandbox to test better the
targeting devices, without messing up with the original upekts driver.
Since we're maintaining the same code structure as the upekts driver,
when this code becomes more tested and solid we'll be able to
incorporate back the changes in a future --if we still think that is a
good idea at that time.

Link: 
http://sharing.jorgesuarezdelis.name/2010/libfprint-0.1.0~pre2
+upeke2driver.diff
(Continue reading)

Permalink | Reply |
headers
Jorge Suárez de Lis | 28 Apr 15:08
Gravatar

Re: Driver for Upek Eikon model no. TCRD4C4 (aka. Eikon II)

Sorry by the confussion. The driver is called upek2e.

The configure example should be:
./configure --with-drivers="upeke2 upekts"

--

-- 
Jorge Suárez de Lis <yo <at> jorgesuarezdelis.name>

_______________________________________________
fprint mailing list
fprint <at> reactivated.net
http://lists.reactivated.net/mailman/listinfo/fprint
Permalink | Reply |
headers
Daniel Drake | 29 Apr 03:47
Favicon

Re: Driver for Upek Eikon model no. TCRD4C4 (aka. Eikon II)

El 24/04/10 21:38, Jorge Suárez de Lis escribió:
> Some thoughts about the IDs problem? What's next? Should I rename the
> driver? Or it would be better to change the upekts driver to support
> both devices?

My suggestion would be to modify the driver architecture so that 2 
drivers can claim the same ID. If one of them fails to initialize 
(perhaps with -ENODEV) then the next one is tried.

Daniel
Permalink | Reply |

Gmane