On Sun, 2009-05-31 at 10:13 +0200, Øyvind Harboe wrote: > > Been there, done that. Reread what I wrote: the issue > > is the *board* not wiring it. > > > > The current code to check "is RTCK supported" only > > addresses the JTAG adapter. There's no way for it > > to discern the case which I listed: adapter supports > > it, but board does not. > > So why can't/shouldn't this be handled in the board/target > config file? > > Possibly with some conditional code? > > We could define parameterized config files, e.g.: > > openocd -f interface/foo.cfg -c "set xx 1;set yy 2" -f board/bar.cfg > > If the # of parameter permutations isn't too great we could have > one wrapper config file/combination. > > I don't understand what C can do that tcl can't in this > context and I'd like to keep reset_config minimal... >
It sounds like the problem relates to the fact that reset_config will not work for all combinations of interfaces and targets. All of this discussion makes me think that OpenOCD needs two reset_config commands: one for the interface and one for the target. With knowledge of both, OpenOCD could do the right thing in all cases, as I understand things; however, this may be incorrect for reasons that others can explain. Cheers, Zach _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
