I made a deliberate choice to not include that, mostly due to post size/complexity reasons. I decided that the focus of the piece was going to be on the task of effectively managing the memory you have in the face of calls to malloc/free. I decided against talking about any environment-specific information (brk, mmap, cache locality, multithreading, etc.)

May not have been the right call, but it kept the length fairly reasonable and I think it gives people enough of a nudge to dip their toes into the playground if they have time.

If you check the HTML comments on the page you'll find quite a lot of cut content. I wanted to do deep dives on real-world malloc implementations, but it ended up being very difficult to cover without making the post take over an hour to read.