[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ms-smartcardddk
Subject:    Re: Is new version of smclib still working correct for chained T=
From:       "Klaus Schutz (Exchange)" <kschutz () EXCHANGE ! MICROSOFT ! COM>
Date:       1998-11-24 16:53:20
[Download RAW message or body]


Johannes,

your observation is correct. The ATR parser of the previous
version of the smart card library did not parse T=1
parameters correctly. This caused in the past that the
smart card library took 'default' timing parameters,
(See ISO 7816-3 9.5.3 Waiting Times) which usually are
longer timeouts than indicated in the ATR of the card.
This means that your driver now times out.

This is because the minimum wait time is now smaller than
you can delay execution on most machines. You should query
the minimum time you can delay execution. Here is an NT example:

minWaitTime = (KeQueryTimeIncrement() / 10000) * 5;

You should not program the serial driver with timeout values
smaller than this minimum wait time.

Let me know if you still have problems.

Klaus U. Schutz
Microsoft

P.S. We could only verify that logo'ed driver still work
with the new version. Unfortunately, the German Geldkarte
is not part of the test set.

-----Original Message-----
From: Johannes Goldynia [mailto:Johannes.Goldynia@PDTS.AT]
Sent: Monday, November 23, 1998 3:15 AM
To: SmartCardDDK@DISCUSS.MICROSOFT.COM
Subject: Is new version of smclib still working correct for chained T=1
mode?


Dear Klaus,

until now I used the file
07.05.98  11:11                 18.560 SMCLIB.SYS
which was a "small" update. My PC/SC drivers (a Windows NT 4.0 - scrCP8T
clone) works well for three months
and passed all PC/SC test cards and also real world apps.

AFTER replacing SMCLIB.SYS above with
10.11.98  11:36                 19.104 SMCLIB.SYS
the "German Geldkarte" (T = 1) stopped working correctly. I have repeated
the PC/SC compliance test of my
PC/SC driver and I got errors using the IBM T=1 testcard:

Test No.  3
   SELECT FILE 0007                                    Passed
   WRITE BINARY 128 Byte(s) Pattern 00h                Passed
   READ  BINARY 128 Byte(s) Pattern 00h                Passed
   WRITE BINARY 128 Byte(s) Pattern FFh              * FAILED
-  IOCTL call failed:
Returned       79H (NTSTATUS c00000b5H)
Expected        0H (NTSTATUS        0H)

So clearly - the timing or the timing information of smclib.sys has been
changed.
Could you give me hints about this changes to fix this behave quickly - or
is the
new smclib.sys broken?

Johannes Goldynia
PDTS Ges.m.b.H
Johannes.Goldynia@pdts.at


-------------------------------------------------------------
   Dr. Johannes Goldynia   eMail: Johannes.Goldynia@pdts.at
   PDTS - Prozessdatentechnik und Systeme
   Gesellschaft fuer industrielle Datenverarbeitung Ges.m.b.H
   A-1070 Wien, Halbgasse 3-5
   Tel.:  +43 (1) 526 17 57-226  Fax.:  +43 (1) 526 17 57-199
   WWW: http://www.pdts.at    WWW: http://www.pdts.at/ChipX
-------------------------------------------------------------

[prev in list] [next in list] [prev in thread] [next in thread]