On Fri, Apr 24, 2009 at 8:44 AM, Michael Schwingen
<rincew...@discworld.dascon.de> wrote:
> Dick Hollenbeck wrote:
>> Why is this code in jtag_add_pathmove()?
>>
>>
>> ??????????????????
>>
>> /* the last state has to be a stable state */
>> if (!tap_is_state_stable(path[num_states - 1]))
>> {
>> LOG_ERROR("BUG: TAP path doesn't finish in a stable state");
>> exit(-1);
>> }
>>
>> ???????????????????
>>
> This is just a guess:
> it might make sense in case you have a JTAG interface that runs with
> continuous clock - in that case, you have to be in a stable state after
> a state move, so that the interface can stay in that state by setting an
> appropriate TMS value.
>
> I don't know if such an interface exists and if this was the idea behind
> that check.
I think it makes sense to allow innovation on the interface side by
creating some wiggling room for the hardware on how it can implement
the JTAG state machine.
At some point OpenOCD has to throw in the towel and say it can't
support some interface hw, because it is just too weird/broken, but
for now I'm favouring keeping the stable state concept.
--
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
