I've been testing the latest firmware (git master) to see how patches from Min Xu made it better and tracing of fast sims (lost bytes, broken tracing, etc.) seems to be a lot improved.

I have found one bug - somehow simtrace_hdr makes it into the APDU payloads, I've been trying to find the root cause of this but have not managed yet - seems like the FW sends 2 messages but they arrive as a single message into the host software thus the header is considered APDU payload.

I've reported the issue here: https://osmocom.org/issues/2614 along with all the tracing/investigation I've performed so far - anybody got further ideas how this might happen?

Other than this and the painful building of the firmware in today's world it seems tracing modern phones is not such a pain as it used to be, yay!

Lukas