#2030 NORM Trial-3: Refresh approximate memory footprint representation

[Zarro Boogs per Child]

#2030: Refresh approximate memory footprint representation
-------------------------------+--------------------------------------------
  Reporter:  Eben              |       Owner:  marco  
      Type:  defect            |      Status:  new    
  Priority:  normal            |   Milestone:  Trial-3
 Component:  interface-design  |     Version:         
Resolution:                    |    Keywords:         
  Verified:  0                 |  
-------------------------------+--------------------------------------------
Comment (by danw):

 Replying to [comment:10 Eben]:
 > I think there's an edge case that this script should, and might not yet
 handle.  The case I want to prevent is one where there are less than
 MAX_ACTIVITIES in the ring, and the extra_space < MIN_ACTIVITY_WEDGE_SIZE.
 This would result in what appeared to be a not-quite-full ring, yet they
 would not be able to launch any new activities.

 No, if there's enough memory to start that activity, then it would just
 cause the other activity wedges that are larger than
 MIN_ACTIVITY_WEDGE_SIZE to shrink a little bit to make room for another
 activity of size MIN_ACTIVITY_WEDGE_SIZE in the ring. (The fact that
 wedges have to shrink sometimes is unavoidable given the minimum wedge
 size restriction.) And if there's *not* enough memory to start that
 activity, that's not going to correlate with "extra_space <
 MIN_ACTIVITY_WEDGE_SIZE"; some activities will take less than
 MIN_ACTIVITY_WEDGE_SIZE memory, and some will take more.

 At any rate, if the free memory wedge is smaller than
 MIN_ACTIVITY_WEDGE_SIZE, then either (a) almost all RAM is used up, in
 which case our ability to predict whether or not another activity can be
 launched is limited, or (b) the user is running nearly MAX_ACTIVITIES
 activities, in which case our ability to *express* our prediction of
 whether or not another activity can be launched is limited, because all of
 the wedges will be almost exactly the same size. So I don't think there's
 a solution here that will provide better information to the user, but if
 you think having a tiny free memory wedge is inaesthetic, then we could
 either make it disappear when it gets below MIN_ACTIVITY_WEDGE_SIZE, as
 you suggested, or else make it refuse to shrink below that size, like the
 activity wedges do.

-- 
Ticket URL: <http://dev.laptop.org/ticket/2030#comment:15>
One Laptop Per Child <http://laptop.org/>