You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We can save some time on pongs by storing the root address at each node. Then we'd only have to travel to our oldest blossom ancestor to figure out what our total internal weight is before replying with a recommendation. Obviously this means that the root will have to be modified by various operations, but I don't think that will be very difficult. See #22 for additional ideas -- we could potentially pong right away if we're pistil-less, and skip the adjoin-root message altogether.
NB: Not sure if there are any live/deadlock considerations here. Upon changing this, a child could send a reply pong even if the root is starting to lock (before that wouldn't happen because all pongs exit through the root). I don't think there is a problem, but just wanted to point it out.
The text was updated successfully, but these errors were encountered:
We can save some time on pongs by storing the root address at each node. Then we'd only have to travel to our oldest blossom ancestor to figure out what our total internal weight is before replying with a recommendation. Obviously this means that the root will have to be modified by various operations, but I don't think that will be very difficult. See #22 for additional ideas -- we could potentially pong right away if we're pistil-less, and skip the adjoin-root message altogether.
NB: Not sure if there are any live/deadlock considerations here. Upon changing this, a child could send a reply pong even if the root is starting to lock (before that wouldn't happen because all pongs exit through the root). I don't think there is a problem, but just wanted to point it out.
The text was updated successfully, but these errors were encountered: