Attention is currently required from: laforge.
1 comment:
Patchset:
I'm curious about your thoughts of rbtree vs. hash table. […]
I really have no hard thoughts regarding one or another here, I could have used a hashtable too. Each has its pros and cons. Since hashtable means allocating more memory and we may want to add more lookup data structures for other MI types in the future (IMSI string, etc.) I thought rbtree was the easiest to replicate. Also because I already had code around from previous patches.
I think simply rbtree is good enough (TM).
It's not like we are moving millions of objects so the difference shouldn't be that much, and the time the analysis that would require is probably not worth it.
To view, visit change 31875. To unsubscribe, or for help writing mail filters, visit settings.