> Dunno. I don't see these traps in 2.4, which puzzles me. In the end > these reads are responsible for storing the current cards state, so > could this be the reason why X/vt switching fails? Maybe, yes. I need to put my hand on the HW to test myself Ben.