On Saturday 02 February 2008 22:41:03 chromatic wrote: > Second, we need a way to know the proper initialization order of PMC > vtables such that we handle all parents before all children. Making this > work for core PMCs is fairly easy. Making it work for dynpmcs is a little > more difficult, but still doable.
After more reflection, this is only necessary for dynpmcs. Core PMCs all get stuffed into the same libparrot, so all of the necessary symbols are visible. That's probably easier, as dynpmc groups get generated together. I suspect that this strategy would be somewhat faster than the current strategy even for core PMCs, but I don't have code to prove it yet. -- c
