#Why memory comes first
In a young kernel, memory management is less a subsystem and more a source of truth. Every subsequent decision inherits its constraints.
#Layering
The working model separates early boot allocation from the long-lived allocator surface. This keeps platform bring-up concerns from leaking into general kernel services.
Design principle
Ownership rules are defined before performance tuning, because unclear ownership multiplies future cost.
#Transition cost
The most delicate handoff happens between temporary boot mappings and the stable address-space model that later services depend on.
#Practical result
The immediate result is a cleaner substrate for task state, driver buffers, and interrupt-safe paths.