1. I/O Management

2. I/O Systems I/O Hardware Application I/O Interface
Kernel I/O Subsystem
Transforming I/O Requests to Hardware Operations
Streams
Performance
I/O Buffer

3. I/O Hardware Incredible variety of I/O devices Common concepts
Port
Bus (daisy chain or shared direct access)
Controller (host adapter)
I/O instructions control devices
Devices have addresses, used by
Direct I/O instructions
Memory-mapped I/O

4. A Typical PC Bus Structure

5. Device I/O Port Locations on PCs (partial)

6. I/O Devices and Controllers
In general, the closer an I/O unit is to the CPU, the faster it can be accessed.
Three categories:
Human Readable (display, keyboard) 80 b/s
Machine Readable (disk) 6 Mb/s
Communication (modem, network) 1 Gb/s
Other Differences:
Data rate, latency, signal propagation, bus arbitration, buffering time
Application - disk requires file system support
Complexity of Control - keyboard driver simpler than disk driver
Unit of Transfer (byte, block)
Data Representation (character set, parity)
Error Conditions

7. Evolution of the I/O Function
CPU directly controls device
Seen in simple microprocessor devices
I/O module (i.e., serial port chip)
CPU uses programmed I/O
Add Interrupts
More efficient – No polling needed
I/O module supports DMA
Device can move a block of data without involving the CPU
I/O Processor
I/O device enhanced with a set of I/O-specific instructions
CPU creates I/O program
I/O Module with memory
Module becomes a computer of its own
Minimal CPU involvement

8. Polling Determines state of device
command-ready
busy
Error
Busy-wait cycle to wait for I/O from device

9. Interrupts CPU Interrupt request line triggered by I/O device
Interrupt handler receives interrupts
Maskable to ignore or delay some interrupts
Interrupt vector to dispatch interrupt to correct handler
Based on priority
Some unmaskable
Interrupt mechanism also used for exceptions

10. Interrupt-Driven I/O Cycle

11. Intel Pentium Processor Event-Vector Table

12. DMA In general, polling is a waste of CPU time
Interrupt processing overhead is greater than polling but without “wasted” cycles
DMA reduces CPU involvement when transferring larger chunks of data
Cycle stealing causes the CPU to execute more slowly
mitigated by cache
smarter DMA controllers
Improvement when path between DMA module and I/O module that does not include the system bus

13. DMA (continued…) Block structure Structure of operations:
Data count (# of bytes/words)
Data register (data to/from device)
Address register (memory location)
Control logic
Structure of operations:
CPU initializes DMA unit
Inform unit if this is a read or write
Set the address of I/O device
Address of memory to read/write
Amount of data to transfer
DMA device performs transfer
Steals memory cycles from CPU
These may occur in the middle of an instruction
DMA device interrupts CPU at end
Data Count
Data Register
Address Register
Control Logic
Data Lines
Address Lines
DMA request
DMA acknowledge
Interrupt
Read
Write

14. DMA Configuration

15. Six Step Process to Perform DMA Transfer

16. I/O Design Objectives Efficiency Generality I/O can be a bottleneck
Swapping is used to increase number of ready processes
Requires I/O activity to do this
Major emphasis is on disk efficiency due to its importance
Generality
Easier to handle devices uniformly
How processes view I/O devices
How O.S. manages I/O devices
Because of differences in devices, it is hard to achieve true generality
Can use a hierarchical, modular approach to hide I/O details
Processes see open/read/write/close for all devices

17. Application I/O Interface
I/O system calls encapsulate device behaviors in generic classes
Device-driver layer hides differences among I/O controllers from kernel
Devices vary in many dimensions
Character-stream or block
Sequential or random-access
Sharable or dedicated
Speed of operation
read-write, read only, or write only

18. A Kernel I/O Structure

19. Characteristics of I/O Devices

20. Block and Character Devices
Block devices include disk drives
Commands include read, write, seek
Raw I/O or file-system access
Memory-mapped file access possible
Character devices include keyboards, mice, serial ports
Commands include get, put
Libraries layered on top allow line editing

21. Network Devices
Varying enough from block and character to have own interface
Unix and Windows NT/9i/2000 include socket interface
Separates network protocol from network operation
Includes select functionality
Approaches vary widely (pipes, FIFOs, streams, queues, mailboxes)

22. Clocks and Timers Provide current time, elapsed time, timer
If programmable interval time used for timings, periodic interrupts
ioctl (on UNIX) covers odd aspects of I/O such as clocks and timers

23. Blocking and Nonblocking I/O
Blocking - process suspended until I/O completed
Easy to use and understand
Insufficient for some needs
Nonblocking - I/O call returns as much as available
User interface, data copy (buffered I/O)
Implemented via multi-threading
Returns quickly with count of bytes read or written
Asynchronous - process runs while I/O executes
Difficult to use
I/O subsystem signals process when I/O completed

24. Kernel I/O Subsystem Scheduling
Some I/O request ordering via per-device queue
Some OSs try fairness
Buffering - store data in memory while transferring between devices
To cope with device speed mismatch
To cope with device transfer size mismatch
To maintain “copy semantics”

25. Sun Enterprise 6000 Device-Transfer Rates

26. Kernel I/O Subsystem Caching - fast memory holding copy of data
Always just a copy
Key to performance
Spooling - hold output for a device
If device can serve only one request at a time
i.e., Printing
Device reservation - provides exclusive access to a device
System calls for allocation and deallocation
Watch out for deadlock

27. Error Handling
OS can recover from disk read, device unavailable, transient write failures
Most return an error number or code when I/O request fails
System error logs hold problem reports

28. Kernel Data Structures
Kernel keeps state info for I/O components, including open file tables, network connections, character device state
Many, many complex data structures to track buffers, memory allocation, “dirty” blocks
Some use object-oriented methods and message passing to implement I/O

29. UNIX I/O Kernel Structure

30. I/O Requests to Hardware Operations
Consider reading a file from disk for a process:
Determine device holding file
Translate name to device representation
Physically read data from disk into buffer
Make data available to requesting process
Return control to process

31. Life Cycle of An I/O Request

32. STREAMS - STREAM head interfaces with the user process
STREAM – a full-duplex communication channel between a user-level process and a device
A STREAM consists of:
- STREAM head interfaces with the user process
- driver end interfaces with the device - zero or more STREAM modules between them.
Each module contains a read queue and a write queue
Message passing is used to communicate between queues

33. The STREAMS Structure

34. Performance I/O a major factor in system performance:
Demands CPU to execute device driver, kernel I/O code
Context switches due to interrupts
Data copying
Network traffic especially stressful

35. Intercomputer Communications

36. Improving Performance
Reduce number of context switches
Reduce data copying
Reduce interrupts by using large transfers, smart controllers, polling
Use DMA
Balance CPU, memory, bus, and I/O performance for highest throughput

37. Device-Functionality Progression

38. Logical Structure Hierarchical structure Simple structure
Divide levels by
complexity
time scale
level of abstraction
Low levels may work at nanosecond time frames
Simple structure

39. I/O Organization
Logical I/O – Think of the device as a logical resource
Directory Management
Convert file names to identifiers that reference a file
File System
Deal with logical structure and open, read, write, also permissions
Physical Organization
Convert logical references to storage addresses
Device I/O – Convert operations and data into I/O instruction sequences
May use buffering to improve speed
Scheduling and Control – Actual queuing and scheduling of operations
This level handles interrupts, status

40. I/O Buffering No buffering Single buffering Double buffering
Operating System
User Process
I/O Device In
Single buffering
Operating System
User Process
I/O Device In
Move
Double buffering
I/O Device In
Move
Operating System
User Process
Circular buffering
Operating System
User Process
I/O Device In
Move

41. I/O Buffering I/O operations from user memory space creates problems
The pages holding the data must remain in main memory
Limits choices O.S. can make
Cannot completely swap out process, or deadlock may result
Process waiting for I/O to complete
I/O waiting for process to be swapped in
May want to read in advance
May want to delay writes
Combine writes when updating disk
Buffering Methods:

42. I/O Buffering Block-oriented Stream-oriented
information is stored in fixed sized blocks
transfers are made a block at a time
used for disks and tapes
Stream-oriented
transfer information as a stream of bytes
used for terminals, printers, communication ports, mouse, and most other devices that are not secondary storage

43. Single Buffering
I/O device transfers data to a system buffer, then O.S. copies data to user
As soon as transfer is completed, try to read the next block in advance
Hopefully that block will be used next
User process can be working on one block while the next is being read
Block transfer time: max[C,T]+M
C = time to compute, T = time to do I/O, M = time to transfer to buffer
Note: with no buffering, time is C+T
O.S. must track which buffers are assigned to user processes
Don’t want to swap to a device if we are waiting on that device for I/O
Able to swap processes without interfering with I/O operations
For character-I/O, can use buffer for byte buffering or line buffering - Producer/Consumer type problem

44. Buffering Double Buffering Circular Buffering Summary
Uses two system buffers, read into one buffer, copy data to user from the other
Block transfer time: max[C,T]
max[C+M,T]? – Account for move time
Can do transfer while next I/O starts
Circular Buffering
Similar to double buffering, but have multiple system buffers
Helps handle I/O bursts
Summary
Buffering helps smooth out peaks in I/O use
Doesn’t improve sustained I/O rate
Generally helps in multiprogramming systems