On May 19, 2010, at 19:33 , Jean-Michel Pouré - GOOZE wrote:
> On Wed, 2010-05-19 at 18:12 +0200, Juergen Beisert wrote:
>>> Then a single and universal driver (ifdhandler) for pcsc-lite could
>> be
>>> created just as the CCID driver is universal (so-so) for CCID
>> devices.
>> 
>> That could be the goal. 
> 
> I remember a post on the mailing list explaining how OpenCT could be
> integrated in libccid. There was not much discussion about this post. 

pcsc-lite is a framework for smart card reader drivers or ifdhandlers, much 
like OpenCT. There are different classes of drivers, like USB, PCMCIA or SoC. 
I don't see why everything should be jammed into libccid which is a driver for 
a certain class of hardware. Different driver classes should have a different 
driver.

>From source perspective, the "trouble" you're mentioning is the possible need 
>for code duplication for protocol handling, which seems like a pain but can be 
>arranged differently than by creating a hybrid driver.

In addition to the behavior problems Alon has been talking about (sleeping, 
polling)  and API questions (OpenCT != PC/SC) the way to reduce code 
duplication could maybe be to crate a mini-framework for proprietary (USB, SoC, 
Serial) protocols for pcsc-lite.

> This is a post from Andreas Jellinghaus, published on Tue, 20 Apr 2010
> 20:50:56 +0200. Here is a copy:
> 
> ********************************
> * take libccid code
> * isolate the parts that are related to the ccid interface
> * replace them with an interface connecting usb and t=1 protocol
>  in aladdin style (see the source code, it is totaly simple).
> * bonus if the result could be added to libccid as a clean driver
>  implementing several different interfaces (not only ccid).
> 
> ********************************
> 
> Ludovic, this may be useful to describe how libccid could evolve to
> handle t=1 protocol and maybe the work needed to port to libusb 1.0.

Do you want to say that libccid currently does not handle T=1?

Upgrading to libusb 1.0 when appropriate is of course a good thing.


-- 
Martin Paljak
http://martin.paljak.pri.ee
+3725156495


_______________________________________________
opensc-devel mailing list
opensc-devel@lists.opensc-project.org
http://www.opensc-project.org/mailman/listinfo/opensc-devel

Reply via email to