On Tuesday 13 October 2015 07:29:10 andy pugh wrote:

> On 13 October 2015 at 10:36, Gene Heskett <ghesk...@wdtv.com> wrote:
> > Trying to take advantage of the fact that named vars remain in
> > memory and are accessable from the next program loaded, I added some
> > code to this blanket-chest3.ngc to do exactly that,
>
> It may be that that is the case, but it isn't intentional. ie, this
> falls into the "undefined behaviour" area of G-code.

IMO when it encounters such a reference, it ought to compare it to the 
famous Amigados "deadbeef" string, which is what it used to detect out 
of bounds memory accesses. If its not, then whatever it is is valid.  
But linux has never poisoned memory that way, hasn't had to as it has a 
good memory manager.

> If you want a rather crazy way to do this you could write to G-code
> analogue outputs and read from analogue inputs, and loop them together
> in HAL.

Sounds like it would be usable.  Example?

> A better way would be to add some new persistent _numerical_
> parameters to the vars file. Those should then always exist.

And how would one go about that?  I thought of opening a probe file, but 
I have no clue how to extract the data in a form useful for this.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>

------------------------------------------------------------------------------
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to