Re: Memory fragmentation?

> I don't think this would be the best general purpose solution.  Most
> large allocations (think image size allocations) end up outside of the
> main heap area by most allocators already.  We can do something like
> this in smaller areas, but I think us getting a better understanding
> of lifetimes of grouped objects that we can pool together and free
> together will result in much bigger wins.  We also need to simply
> reduce the number of allocations.

OK, thanks Stuart.  We had a separate discussion about this on IRC.
Some thoughts for those following along at home.  Our very smart
friend Zach Brown pointed out to us that the right way to think about
these allocations is to not think about them in terms of size, but
instead think of them in terms of lifetime.  Stuart eludes to that but
doesn't call it out specifically.  I thought it was worth mentioning
so that people will think about it differently.