Virtual Memory - CS 61, Exercises of Operating Systems

Download Virtual Memory - CS 61 and more Exercises Operating Systems in PDF only on Docsity!

Virtual Memory

CS61, Lecture 15

Prof. Stephen Chong

October 20, 2011

Stephen Chong, Harvard University

Announcements

• Midterm review session: Monday Oct 24
  • 5:30pm to 7pm, 60 Oxford St. room 330
  • Large and small group interaction 2

Stephen Chong, Harvard University (^4)

Today

• Running multiple programs at once
• Virtual memory
  • Address spaces
  • Benefits
  • Physical memory as a cache for virtual memory
  • Page tables
  • Page hits and page faults
  • Virtual memory as a tool for memory management
  • Virtual memory as a tool for memory protection
  • Address translation
  • Table Lookaside Buffer (TLB)
  • Example

Stephen Chong, Harvard University (^5)

Running multiple programs at once

• So far, we have discussed memory management for a single

program running on a computer.

• Most modern computers run multiple programs simultaneously.
  • This is called multiprogramming or multitasking. I think I have my own CPU So do I!

Stephen Chong, Harvard University (^7)

Virtual memory

• OS also gives every program illusion

of having its own memory!

• Each program running on the machine

is called a process.

• Each process has a process ID (a number) and an owner – the user ID running the process.
• (^) UNIX command ps prints out info about the running processes.
• Each process has its own address

space

• Contents of that process' memory, visible only to that process.
• 4GB on 32 bit machine, ~16 TB on 64 bit machine OS memory Read/write segment .data, .bss Read-only segment .text, .rodata unused Shared libraries Heap (used by malloc) User stack 0x Process address space 0xffffffff

Stephen Chong, Harvard University Problem 1: How Does Everything Fit? Physical main memory: Few Gigabytes

8 OS memory Read/write segment .data, .bss Read-only segment .text, .rodata unused Shared libraries Heap (used by malloc) User stack 4GB to 16 EB virtual memory

Stephen Chong, Harvard University

Problem 3: How To Protect

Physical main memory Process i Process j 10

Stephen Chong, Harvard University

Problem 4: How To Share

Physical main memory Process i Process j 11

Stephen Chong, Harvard University

Aside: indirection

13

David Wheeler

1927– Computer Science Pioneer World’s first PhD in Computer Science All problems in computer science can be solved by another level of indirection... Except for the problem of too many layers of indirection.

Stephen Chong, Harvard University

Address Spaces

• Linear address space:
  • Ordered set of contiguous non-negative integer addresses: {0, 1, 2, 3 … }
• Virtual address space:
  • Set of N = 2 n virtual addresses {0, 1, 2, 3, …, N-1}
• Physical address space:
  • Set of M = 2 m physical addresses {0, 1, 2, 3, …, M-1}
• Clean distinction between data (bytes) and their attributes (addresses)
• Each object can now have multiple addresses
• Every byte in main memory: one physical address, one (or more) virtual addresses 14

Stephen Chong, Harvard University

A System Using Virtual Addressing

• Used in all modern desktops, laptops, workstations
• One of the great ideas in computer science MMU Physical address (PA) Data word Virtual address (VA) CPU Chip 16 0: 1: M-1: Main memory 2: 3: 4: 5: 6: 7: 8: ... CPU

Stephen Chong, Harvard University

Benefits of Virtual Memory (VM)

• Efficient use of limited main memory (RAM)
  • Use RAM as a cache for the parts of a virtual address space
    • some non-cached parts stored on disk
    • some (unallocated) non-cached parts stored nowhere
  • Keep only active areas of virtual address space in memory
    • transfer data back and forth as needed
• Simplifies memory management for programmers
  • Each process gets the same full, private linear address space
• Isolates address spaces
  • One process can’t interfere with another’s memory
    • because they operate in different address spaces
  • User process cannot access privileged information
    • different sections of address spaces have different permissions 17

Stephen Chong, Harvard University

VM as a Tool for Caching

• Virtual memory: array of N = 2 n contiguous bytes
• think of the array (allocated part) as being stored on disk
• Physical main memory (DRAM) = cache for allocated virtual memory
• (^) Blocks are called pages ; size = 2 p PP 2 m-p - Physical memory Empty Empty Uncached VP 0 VP 1 VP 2 n-p

Virtual memory Unallocated Cached Uncached Unallocated Cached Uncached PP 0 PP 1 Empty Cached 0 2 n

2 m

0 Virtual pages (VP's) stored on disk Physical pages (PP's) cached in DRAM 19

Stephen Chong, Harvard University (^20) Memory Core 1 Regs Unified L Cache L d-cache L i-cache L Cache Core 2 Regs Unified L Cache L d-cache L i-cache Core 3 Regs Unified L Cache L d-cache L i-cache Core 4 Regs Unified L Cache L d-cache L i-cache

Memory hierarchy: Intel Core i

Processor Disk ~1ns ~3ns ~12ns ~60ns ~8ms 60 × ~100 000×