On Wed, May 20, 2009 at 6:30 PM, Magnus Lundin <lun...@mlu.mine.nu> wrote: > I think that you should first figure out the logic of controlling several > different targets on the same scan chain, with different demands on extra or > no extra RTI, specific numbers of transitions etc, how can they communicate > their differing needs ?
I'm trying to define an API where any default state scheme can be synthesized on top of the JTAG API. This avoids trying to peer into a chrystal ball trying to map out all possible future needs. What I don't understand is what the JTAG layer could possibly contribute here in terms of figuring out what the default state should be. What's even more unclear to me is how any such behaviour could not or should not be synthesized on top of a JTAG API. Certainly a few wrapper fn's around a global variable(be it inside the interface or globally, same thing really), just makes the code more mysterious. -- Øyvind Harboe Embedded software and hardware consulting services http://consulting.zylin.com _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
