On Monday 01 March 2010, Øyvind Harboe wrote: > >> I'm telling the hardware to clock out a tms sequence and I'm > >> telling it what state it is in afterwards, a simple and robust > >> design that lets software figure out weird special rules. > > > > Right. The top level passes that state down. My question is > > why the low level driver would care about the state ... since > > the software stack in between the "single step" command (or > > whatever) is already tracking that. > > I don't quite follow you. Are you suggesting that the driver > should reach up to the higher level fn's. Wouldn't that > be a layering violation?
Nope. That state is part of the framework interface. On the other hand, you're still not answering my question about why the low level driver(s) would care about that state. They have one job: "respond to requests from upper levels". It's the upper levels who are responsible for ensuring that only valid requests get passed down. (Leaving no reason for the low level code to care about TAP states. - Dave _______________________________________________ Openocd-development mailing list [email protected] https://lists.berlios.de/mailman/listinfo/openocd-development
