Re: patch for the checkpointing paper

["Kamal R. Prasad" <kamalpr@xxxxxxxxx>]

--- Matthew Dillon <dillon@xxxxxxxxxxxxxxxxxxxx>
wrote:
[snip]
>     not want to be restored and to track state.  The
> patch set cannot not be
>     committed in its current form.
> 
Matt,

 I do *not* wish to submit this patch/feature to
dragonflybsd src tree i.e. I have abandoned that
effort. What Im doing here is to hand out the
unmodified code changes, so that users can look at how
the concepts outlined in the paper have been
implemented by me. If someone wants to modify the code
and integrate it with the dragonflybsd src tree -that
is fine with me, as long as I am not involved
in/responsible for -the integration.
 Also, the patch is missing a couple of lines.
in 
---------------------
case CKPT_LONGJMP:
   p->prev = p->curr;
   p->curr = respt;
---------------------

regards
-kamal


------------------------------------------------------------
Kamal R. Prasad
UNIX systems consultant 
kamalp@xxxxxxx

In theory, there is no difference between theory and practice. In practice, there is:-).
------------------------------------------------------------


	
		
__________________________________ 
Celebrate Yahoo!'s 10th Birthday! 
Yahoo! Netrospective: 100 Moments of the Web 
http://birthday.yahoo.com/netrospective/