On Sunday 20 December 2009, Øyvind Harboe wrote:
> On Sun, Dec 20, 2009 at 10:12 PM, Catalin Patulea <[email protected]> wrote:
> > Quick note about usb_blaster_reset:
> >
> > usb_blaster is in much the same situation as ft2232. There are no
> > defined bits for TRST/SRST in the USB Blaster pinout, so each
> > interface-board combination will implement that differently. I'm
> > leaning toward deferring that part until someone actually needs the
> > functionality.
> 
> The interface is going to be near useless for a lot of development
> without it.

That's why we now allow "reset-assert" handlers ... which
are currently used only by ARM11 and Cortex-A8.  (I've got
an ARM7/ARM9 patch pending but haven't yet made time to
test it out.)

There's no need to require TRST ... just issue five clocks
with TMS high, and you've done a JTAG reset.

And there are lots of systems that are designed to be fully
debugged without SRST.  We don't yet support that mode as
well as we should, but "reset-assert" handlers solve a key
problem.  (Among other things ... that SRST resets a lot
more than just one target, so that complex boards really
ought to be avoiding its use.)

- Dave


_______________________________________________
Openocd-development mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/openocd-development

Reply via email to