> All we need to do is accept the SVF names when
> we're reading SVF files.  Are you saying that
> the current code does not do that??
When I was implementing SVF support, the tap_name was not compatible.
As I know, the tap_name has changed for some times, so I'm not sure
whether it will change again. Rather than follow the change, I
implement a privite tap_name.

2009/10/24, David Brownell <[email protected]>:
> On Friday 23 October 2009, simon qian wrote:
>> If using tap_state_by_name, it MUST be totally compatible with the SVF
>> specification(previous version is not match, so I add a private version).
>> Otherwise, a change to the global name of state will affact svf
>> implementation.
>
> All we need to do is accept the SVF names when
> we're reading SVF files.  Are you saying that
> the current code does not do that??
>
> Changing state names is not something to be done
> lightly in any case; it'd break more than SVF.
> The comment by the name table highlights that.
>
> Note that patch #2 moves away from that notion
> of "the" global name of a state ... we'll only
> use one name for output (almost exclusively for
> debug), but can accept multiple names on input.
>
> Me, I'd be content with only using SVF names.
> But there may be people with other preferences
> We could _now_ abet such models, not that I'd
> encourage it.  :)
>
> - Dave
>
>
>


-- 
Best Regards, SimonQian
http://www.SimonQian.com
_______________________________________________
Openocd-development mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/openocd-development

Reply via email to