Hello Yutaka,
I have no news from you since my lastest mail.
You may have found a real problem but I need to understand your problem
before I can apply any changes.
Can you answer the questions of my previous mail, please?
Thanks a lot
Ludovic Rousseau a écrit :
On Mon, Oct 27, 2008 at 9:34 AM, NIIBE Yutaka <gni...@fsij.org> wrote:
Hi, there.
Hello,
This is not a Debian bug but an upstream bug. You are lucky since I am
also the upstream author.
I think that current implementation for ICCD version A does not work
well. While I am implementing ICCD device, I found some problems.
(Note that I don't have any ICCD version A device other than my own
development device. So, my understanding/interpretation of protocol
would be wrong.)
Here is a patch to discuss issues.
(1) Function CCID_Receive() should change the value pointed by rx_length.
Can you show a trace exhibiting a problem?
(2) Function CmdXfrBlockCHAR_T0() should handle length of received
packet correctly.
Can you show a trace exhibiting a problem?
(3) It is better for Function CmdXfrBlockCHAR_T0() to check status
before calling CCID_Transmit().
Why not.
I only have ONE device using ICCD type A. So my implementation may not
be correct for your device.
Can you:
- split your 3 changes in 3 distinct patches
- try to convince me the code is wrong/bogus by showing a pcscd trace
exhibiting the problem. See [1]
Thanks
[1] http://pcsclite.alioth.debian.org/ccid.html#support
--
Dr. Ludovic Rousseau
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
