This paper is a very nice overview of how the Linux kernel allocates memory: http://halobates.de/memorywaste.pdf
To summarise it asks the BIOS for memory and manually allocates out of until it can set up a simple "bootmem" allocator. Once it's finished allocator it transitions that to a "page allocator", which in turn backs various "slab allocators" and userspace allocators.
Memory is typically freed on demand, and "mempools" ensure there's enough available for *critical* operations like swapping pages.
@alcinnz interesting :)
I never thought about memory allocation much, until a few months back, when I read another pretty good article but from a higher level programming perspective.
It starts out about how ruby memory allocation may be unnecessarily wasteful but actually ends up explaining a lot about how multiple different memory allocators interact.
Might be of interest to programmers who never "need" to think about memory allocation.
https://www.joyfulbikeshedding.com/blog/2019-03-14-what-causes-ruby-memory-bloat.html
