kernel memory management

virtual memory

as the name suggests. not actual memory. used when more memory is needed than is physically available. uses storage from harddrives or any other storage device available
slower than actual ram. allows multitasking and program isolation

moves the data thats not needed that much from the ram to storage in blocks called pages. this process is called swapping

pages: fixed size blocks of virtual memory used to divide a process’s address space
page frames: fixed size blocks of physical memory (ram) that hold the pages
page size = page frame size
typical size = 4KB (4096 bytes). some systems support multiple page sizes

a hardware component in the cpu
translates virtual addresses to physical ones in the ram or storage using page tables, page table base register (PTBR)s and translation lookaside buffer (TLB)s

primarily stored in physical ram within kernel space of the OS’s virtual memory
each process has its own page table
isnt part of the MMU itself
used to translate virtual addresses to physical ones. essentially acting as a map
each entry in the page table is called PTE for short. they contain info on virtual pages like
- physical frame number of the page location
- status bits indicating validity/modification(dirty)/accessed
- protection bits specifying access perms like rwx
hierarchical multi level page tables exists to let us handle large memory address spaces

a cache for recent translations from the page table. speeds up memory access.
is tipically small. storing 64-512 entries
when the cpu needs to access memory it first checks the TLB
- if it finds a transtation in it its a “TLB hit” and if not its a “TLB miss” and proceeds to consult the page table
if data is written. its first written to the cache and then eventually written into the underlying storage device