On 28/10/05, [EMAIL PROTECTED]
<[EMAIL PROTECTED]> wrote:
> Dear all,

Hello,

> During performance measurements using the read binary command (256B),
> I noticed differences between windows (100ms) and linux (500ms) execution
> time.
>
> I figured out that the reason for this is a SCardReconnect call which
> forces the
> pcsc deamon to sleep for 400ms:
>
> winscard.c - SCard[Re|Dis|C]onnect:
> ----------------------------------------------------------------------------------------
>
> SYS_USleep(PCSCLITE_STATUS_POLL_RATE + 10);
>
> return SCARD_S_SUCCESS;
>
> ----------------------------------------------------------------------------------------
>
> I think this passive waiting for status conveyance decreases
> performance dramatically.
>
> What is the exact reason for sleeping such a long time?

This is related to the way pcscd and libpcsclite talks to each-other.

> Wouldn't it be better to notify the client in the case of a status change?

Yes. The communication protocol between pcscd and libpcsclite could be
largely improved.

I just commited a TODO file [1] including the feature you requested.
Do you think that Omnikey could sponsor some of the features from the
todo list?

Regards,

[1] http://svn.debian.org/wsvn/pcsclite/trunk/PCSC/TODO?op=file&rev=0&sc=0

--
 Dr. Ludovic Rousseau
 For private mail use [EMAIL PROTECTED] and not "big brother" Google

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

Reply via email to