Without this fix, if the host software does not send an ATR, simtrace
waits for this answer in an endless loop of unsuccessfull USB
endpoint write attempts.
In order to record the communication between simtrace and the
mobile phone with a logic analyzer I had to configure the
pins in sniffer mode in order to attach the sniffer clips
to the SIM card slot. I did not manage to connect them to
the connector to the phone flex cable.
The compiler warning:
simtrace/usb.c:553:5: warning: initialization from incompatible pointer type [enabled by default]
&configurationDescriptorSniffer,
^
simtrace/usb.c:553:5: warning: (near initialization for 'configurationDescriptorsArr[0]') [enabled by default]
simtrace/usb.c:554:5: warning: initialization from incompatible pointer type [enabled by default]
&configurationDescriptorCCID,
^
simtrace/usb.c:554:5: warning: (near initialization for 'configurationDescriptorsArr[1]') [enabled by default]
simtrace/usb.c:555:5: warning: initialization from incompatible pointer type [enabled by default]
&configurationDescriptorPhone,
^
simtrace/usb.c:555:5: warning: (near initialization for 'configurationDescriptorsArr[2]') [enabled by default]
simtrace/usb.c:556:5: warning: initialization from incompatible pointer type [enabled by default]
&configurationDescriptorMITM,
^
simtrace/usb.c:556:5: warning: (near initialization for 'configurationDescriptorsArr[3]') [enabled by default]
simtrace/usb.c: In function 'getConfigDesc':
simtrace/usb.c:560:5: warning: return discards 'const' qualifier from pointer target type [enabled by default]
return configurationDescriptorsArr[idx];
^
simtrace/usb.c: At top level:
The BUFLEN is fixed 5 bytes now, which has to be increased.
A timeout has to be implemented, which is smaller than max_waittime.
If the timeout is triggered, the data received in the buffer so far
should be sent to the host computer which is responsible for generating
a response. Without the timeout the max_waittime of the phone would
expire and it would repeat sending the command, so that we do not
have atomary messages anymore.
This is a bug and should be fixed:
The smartcard insertion detection does not work if we uncomment
the check. But on the other hand, pcscd detects a lot of erroneous
smartcard insertion and removal events.
