11:47 <DocScrutinizer05> ((<wpwrak> maybe a fresh pair of eyes will find it in minutes ...)) Link, and instructions what's the problem to look for, please!

11:49 <DocScrutinizer05> whitequark: are you meanwhile ready to emigrate before the frontiers close and arms are pointing direction west?

12:41 <wpwrak> DocScrutinizer05: it's a hard piece of work. the question is: why is my implementation of SWD unreliable ? my implementation is here: https://gitorious.org/anelok/anelok/source/swdlib

12:41 <wpwrak> there are some links to resources in https://gitorious.org/anelok/anelok/source/swdlib/swdlib.c

12:42 <wpwrak> note that there is more out there, and some of the material is contradictory. so the first step is to develop consistent conceptd

12:59 <DocScrutinizer05> sdwlib is conecting to the anelok "debug port" to program the chip? From ben?

14:53 <wpwrak> yes. ben -> ubb -> anelok -> "debug port" of the kinetis

14:53 <wpwrak> i think most arms nowadays have some form of swd

15:05 <wpwrak> swd then talks to the debug port or the AP. on the AP you have one entity that does a few flash things and another one that gives access to the AHB (system bus). so you basically have: 1) a wire protocol to move data to the debug interface. 2) a protocol to talk to the debug port. 3) a protocol to talk to the AHB port. 4) a protocol (which is also used by software) to command flash operation.

15:07 <wpwrak> plus a few bits and pieces, e.g. the protocol on the MDM-AP, which is the critter that does the additional flash operations, such as checking security status, commanding a mass erase, etc.