Le 29 avril 2011 18:30, Jean-Michel Pouré - GOOZE <[email protected]> a écrit :
> Dear all,
>
> The Feitian R-301 and especially the R-310 are recent ccid readers,
> which can connect up to 600.000 bps.
>
> Let's take the example of the Feitian R310:
>
> => libccid log:
>
> ccid_usb.c:954:get_data_rates() declared: 10752 bps
> ..
> ifdhandler.c:1883:find_baud_rate() Card baud rate: 125000
> ifdhandler.c:1892:find_baud_rate() Reader can do: 10752
> ifdhandler.c:826:IFDHSetProtocolParameters() Reader does not support
> 125000 bauds
> towitoko/atr.c:329:ATR_GetDefaultProtocol() default protocol: T=1
>
>
> => The problem stems from .parse, which returns:
> ***
>  bVoltageSupport: 0x07
>   5.0V
>   3.0V
>   1.8V
>  dwProtocols: 0x0000 0x0003
>   T=0
>   T=1
>  dwDefaultClock: 4.000 MHz
>  dwMaximumClock: 8.000 MHz
>  bNumClockSupported: 0 (will use whatever is returned)
>   Support 4000 kHz
>  dwDataRate: 10752 bps
>  dwMaxDataRate: 600000 bps
>  bNumDataRatesSupported: 0 (will use whatever is returned)
>   Support 10752 bps
>  dwMaxIFSD: 248
>  dwSynchProtocols: 0x00000000
>  dwMechanical: 0x00000000
>   No special characteristics
> ***
>
> The R-310 reader does not seem to advertise all possible speeds. Maybe
> there is a way to retrieve all speeds, maybe not.
>
> Why not try to establish a communication with the card at the speed of
> 125000 or other common speeds? I feel that the current algorithm tries
> the minimum speed dwDataRate (10752 bp) when it should try also a speed
> compatible with the card (125000) and less than dwMaxDataRate (600000
> bps).
>
> This would allow several readers to discuss at high speeds, like it is
> done under Windows (which is sometimes 10 times faster than GNU/Linux
> when reading smartcards).
>
> Or Would there be other methods to detect speeds?

This is a (known) limitation of my CCID driver. Very few readers are
affected by this limitation.
Feel free to provide a validated patch.

Bye

-- 
 Dr. Ludovic Rousseau

_______________________________________________
Muscle mailing list
[email protected]
http://lists.drizzle.com/mailman/listinfo/muscle

Reply via email to