University of Washington

Roadmap Data & addressing

Integers & floats
Machine code & C
C: Java:
x86 assembly
car *c = malloc(sizeof(car)); Car c = new Car(); Procedures & stacks
c->miles = 100; c.setMiles(100);
Arrays & structs
c->gals = 17; c.setGals(17);
float mpg = get_mpg(c); float mpg =
Memory & caches
free(c); c.getMPG(); Processes
Virtual memory
Assembly get_mpg: Memory allocation
language: pushq %rbp Java vs. C
movq %rsp, %rbp
...
popq %rbp
ret
OS:
Machine 0111010000011000
100011010000010000000010
code: 1000100111000010
110000011111101000011111

Computer
system:

Autumn 2013 Processes 1

 University of Washington

Processes – another important abstraction

¢ First some preliminaries
§ Control flow
§ Exceptional control flow
§ Asynchronous exceptions (interrupts)
¢ Processes
§ Creating new processes
§ Fork and wait
§ Zombies

Autumn 2013 Processes 2

 University of Washington

Control Flow
¢ So far, we’ve seen how the flow of control changes as a single
program executes
¢ A CPU executes more than one program at a time though – we
also need to understand how control flows across the many
components of the system

¢ Exceptional control flow is the basic mechanism used for:

§ Transferring control between processes and OS
§ Handling I/O and virtual memory within the OS
§ Implementing multi-process applications like shells and web servers
§ Implementing concurrency

Autumn 2013 Processes 3

 University of Washington

Control Flow
¢ Processors do only one thing:
§ From startup to shutdown, a CPU simply reads and executes
(interprets) a sequence of instructions, one at a time
§ This sequence is the CPU’s control flow (or flow of control)

Physical control flow

<startup>
inst1
inst2
time
inst3
…
instn
<shutdown>
Autumn 2013 Processes 4
 University of Washington

Altering the Control Flow

¢ Up to now: two ways to change control flow:
§ Jumps (conditional and unconditional)
§ Call and return
Both react to changes in program state
¢ Processor also needs to react to changes in system state
§ user hits “Ctrl-C” at the keyboard
§ user clicks on a different application’s window on the screen
§ data arrives from a disk or a network adapter
§ instruction divides by zero
§ system timer expires
¢ Can jumps and procedure calls achieve this?
§ Jumps and calls are not sufficient – the system needs mechanisms for
“exceptional” control flow!

Autumn 2013 Processes 5

 University of Washington

Exceptional Control Flow

¢ Exists at all levels of a computer system
¢ Low level mechanisms
§ Exceptions
§change processor’s in control flow in response to a system event
(i.e., change in system state, user-generated interrupt)
§ Combination of hardware and OS software
¢ Higher level mechanisms
§ Process context switch
§ Signals – you’ll hear about these in CSE451 and CSE466
§ Implemented by either:
§ OS software
§ C language runtime library

Autumn 2013 Processes 6

 University of Washington

Exceptions
¢ An exception is transfer of control to the operating system (OS)
in response to some event (i.e., change in processor state)
User Process OS

event current_instr exception

next_instr exception processing
by exception
handler
• return to current_instr
• return to next_instr
• abort

¢ Examples:
div by 0, page fault, I/O request completes, Ctrl-C
¢ How does the system know where to jump to in the OS?

Autumn 2013 Processes 7

 University of Washington

Interrupt Vectors
Exception
numbers

code for ¢ Each type of event has a

exception handler 0 unique exception number k
Exception code for
Table exception handler 1 ¢ k = index into exception table
0 (a.k.a. interrupt vector)
1 code for
2 exception handler 2
... ¢ Handler k is called each time
n-1 ... exception k occurs

code for
exception handler n-1

basically a jump table for exceptions…

Autumn 2013 Processes 8
 University of Washington

Asynchronous Exceptions (Interrupts)

¢ Caused by events external to the processor
§ Indicated by setting the processor’s interrupt pin(s) (wire into CPU)
§ Handler returns to “next” instruction
¢ Examples:
§ I/O interrupts
§ hitting Ctrl-C on the keyboard
§ clicking a mouse button or tapping a touchscreen
§ arrival of a packet from a network
§ arrival of data from a disk
§ Hard reset interrupt
§ hitting the reset button on front panel
§ Soft reset interrupt
§ hitting Ctrl-Alt-Delete on a PC

Autumn 2013 Processes 9

 University of Washington

Synchronous Exceptions
¢ Caused by events that occur as a result of executing an
instruction:
§ Traps
§Intentional: transfer control to OS to perform some function
§ Examples: system calls, breakpoint traps, special instructions
§ Returns control to “next” instruction
§ Faults
§ Unintentional but possibly recoverable
§ Examples: page faults (recoverable), segment protection faults
(unrecoverable), integer divide-by-zero exceptions (unrecoverable)
§ Either re-executes faulting (“current”) instruction or aborts
§ Aborts
§ Unintentional and unrecoverable
§ Examples: parity error, machine check (hardware failure detected)
§ Aborts current program
Autumn 2013 Processes 10
 University of Washington

Trap Example: Opening File

¢ User calls: open(filename, options)
¢ Function open executes system call instruction int
0804d070 <__libc_open>:
. . .
804d082: cd 80 int $0x80
804d084: 5b pop %ebx
. . .

User Process OS

int exception
pop
open file
returns

¢ OS must find or create file, get it ready for reading or writing

¢ Returns integer file descriptor
Autumn 2013 Processes 11
 University of Washington

Fault Example: Page Fault

int a[1000];
¢ User writes to memory location main ()
{
¢ That portion (page) of user’s memory a[500] = 13;
is currently on disk }

80483b7: c7 05 10 9d 04 08 0d movl $0xd,0x8049d10

User Process OS

exception: page fault

movl
Create page and
returns load into memory

¢ Page handler must load page into physical memory

¢ Returns to faulting instruction: mov is executed again!
¢ Successful on second try
Autumn 2013 Processes 12
 University of Washington

Fault Example: Invalid Memory Reference

int a[1000];
main ()
{
a[5000] = 13;
}

80483b7: c7 05 60 e3 04 08 0d movl $0xd,0x804e360

User Process OS

exception: page fault

movl
detect invalid address
signal process

¢ Page handler detects invalid address

¢ Sends SIGSEGV signal to user process
¢ User process exits with “segmentation fault”
Autumn 2013 Processes 13
 University of Washington

Exception Table IA32 (Excerpt)

Exception Number Description Exception Class

0 Divide error Fault
13 General protection fault Fault
14 Page fault Fault
18 Machine check Abort
32-127 OS-defined Interrupt or trap
128 (0x80) System call Trap
129-255 OS-defined Interrupt or trap

http://download.intel.com/design/processor/manuals/253665.pdf

Autumn 2013 Processes 14

 University of Washington

Summary
¢ Exceptions
§ Events that require non-standard control flow
§ Generated externally (interrupts) or internally (traps and faults)
§ After an exception is handled, one of three things may happen:
§ Re-execute the current instruction
§ Resume execution with the next instruction
§ Abort the process that caused the exception

Autumn 2013 Processes 15

 University of Washington

What is a process?

¢ What is a program? A processor? A process?

Autumn 2013 Processes 16

 University of Washington

What is a process?
¢ Why are we learning about processes?
§ Processes are another abstraction in our computer system – the
process abstraction provides an interface between the program and the
underlying CPU + memory.
¢ What do processes have to do with exceptional control flow
(previous lecture)?
§ Exceptional control flow is the mechanism that the OS uses to enable
multiple processes to run on the same system.

Autumn 2013 Processes 17

 University of Washington

Processes
¢ Definition: A process is an instance of a running program
§ One of the most important ideas in computer science
§ Not the same as “program” or “processor”

¢ Process provides each program with two key abstractions:

§ Logical control flow
§ Each process seems to have exclusive use of the CPU
§ Private virtual address space
§ Each process seems to have exclusive use of main memory

¢ Why are these illusions important?

¢ How are these illusions maintained?
§ Process executions interleaved (multi-tasking)
§ Address spaces managed by virtual memory system – next course topic
Autumn 2013 Processes 18
 University of Washington

Concurrent Processes
¢ Two processes run concurrently (are concurrent) if their
instruction executions (flows) overlap in time
¢ Otherwise, they are sequential
¢ Examples:
§ Concurrent: A & B, A & C
§ Sequential: B & C
Process A Process B Process C

time

Autumn 2013 Processes 19

 University of Washington

User View of Concurrent Processes

¢ Control flows for concurrent processes are physically disjoint
in time
§ CPU only executes instructions for one process at a time
¢ However, we can think of concurrent processes as executing
in parallel

Process A Process B Process C

time

Autumn 2013 Processes 20

 University of Washington

Context Switching
¢ Processes are managed by a shared chunk of OS code
called the kernel
§ Important: the kernel is not a separate process, but rather runs as part
of a user process
¢ Control flow passes from one process to another via a context
switch… (how?)
Process A Process B

user code

kernel code context switch

time user code

kernel code context switch

user code

Autumn 2013 Processes 21

 University of Washington

Creating New Processes & Programs

¢ fork-exec model:
§ fork() creates a copy of the current process
§ execve() replaces the current process’ code & address space with
the code for a different program

¢ fork() and execve() are system calls

§ Note: process creation in Windows is slightly different from Linux’s
fork-exec model

¢ Other system calls for process management:

§ getpid()
§ exit()
§ wait() / waitpid()

Autumn 2013 Processes 22

 University of Washington

fork: Creating New Processes

¢ pid_t fork(void)
§ creates a new process (child process) that is identical to the calling
process (parent process), including all state (memory, registers, etc.)
§ returns 0 to the child process
§ returns child’s process ID (pid) to the parent process
pid_t pid = fork();
if (pid == 0) {
printf("hello from child\n");
} else {
printf("hello from parent\n");
}

¢ fork is unique (and often confusing) because it is called once

but returns twice

Autumn 2013 Processes 23

 University of Washington

Understanding fork
Process n
pid_t pid = fork();
if (pid == 0) {
printf("hello from child\n");
} else {
printf("hello from parent\n");
}

Autumn 2013 Processes 24

 University of Washington

Understanding fork
Process n Child Process m
pid_t pid = fork(); pid_t pid = fork();
if (pid == 0) { if (pid == 0) {
printf("hello from child\n"); printf("hello from child\n");
} else { } else {
printf("hello from parent\n"); printf("hello from parent\n");
} }

Autumn 2013 Processes 25

 University of Washington

Understanding fork
Process n Child Process m
pid_t pid = fork(); pid_t pid = fork();
if (pid == 0) { if (pid == 0) {
printf("hello from child\n"); printf("hello from child\n");
} else { } else {
printf("hello from parent\n"); printf("hello from parent\n");
} }

pid_t pid = fork(); pid_t pid = fork();

if (pid == 0) { if (pid == 0) {
pid = m printf("hello from child\n"); pid = 0 printf("hello from child\n");
} else { } else {
printf("hello from parent\n"); printf("hello from parent\n");
} }

Autumn 2013 Processes 26

 University of Washington

Understanding fork
Process n Child Process m
pid_t pid = fork(); pid_t pid = fork();
if (pid == 0) { if (pid == 0) {
printf("hello from child\n"); printf("hello from child\n");
} else { } else {
printf("hello from parent\n"); printf("hello from parent\n");
} }

pid_t pid = fork(); pid_t pid = fork();

if (pid == 0) { if (pid == 0) {
pid = m printf("hello from child\n"); pid = 0 printf("hello from child\n");
} else { } else {
printf("hello from parent\n"); printf("hello from parent\n");
} }

pid_t pid = fork(); pid_t pid = fork();

if (pid == 0) { if (pid == 0) {
printf("hello from child\n"); printf("hello from child\n");
} else { } else {
printf("hello from parent\n"); printf("hello from parent\n");
} }

hello from parent Which one is first? hello from child

Autumn 2013 Processes 27
 University of Washington

Fork Example
¢ Parent and child both run the same code
§ Distinguish parent from child by return value from fork()
§ Which runs first after the fork() is undefined
¢ Start with same state, but each has a private copy
§ Same variables, same call stack, same file descriptors, same register
contents, same program counter…
void fork1()
{
int x = 1;
pid_t pid = fork();
if (pid == 0) {
printf("Child has x = %d\n", ++x);
} else {
printf("Parent has x = %d\n", --x);
}
printf("Bye from process %d with x = %d\n", getpid(), x);
}
Autumn 2013 Processes 28
 University of Washington

Exec-ing a new program

Stack

Very high-level diagram of what

happens when you run the
Heap command ”ls” in a Linux shell:
Data
Code: /usr/bin/bash
fork():
parent child child
Stack
Stack Stack

exec():

Heap Heap
Data Data Data
Code: /usr/bin/bash Code: /usr/bin/bash Code: /usr/bin/ls
Autumn 2013 Processes 29
 University of Washington

execve: Loading and Running Programs

Stack bottom
Null-terminated
¢ int execve( env var strings
char *filename,
Null-terminated
char *argv[], cmd line arg strings
char *envp[] unused
) envp[n] == NULL
¢ Loads and runs in current process: envp[n-1]
§ Executable filename …
envp[0]
§ With argument list argv
argv[argc] == NULL
§ And environment variable list envp argv[argc-1]
§ Env. vars: “name=value” strings …
(e.g. “PWD=/homes/iws/bpw”) argv[0]
¢ execve does not return (unless error) Linker vars
envp
¢ Overwrites code, data, and stack argv
§ Keeps pid, open files, a few other items argc
Stack frame for
Autumn 2013 Processes
main Stack top 30
 University of Washington

execve: Example

envp[n] = NULL
envp[n-1] “PWD=/homes/iws/gaetano”
… “PRINTER=ps581”
envp[0] “USER=gaetano”

argv[argc] = NULL
argv[argc-1] “/usr/include”
… “-l”
argv[0] “ls”

Autumn 2012 Exceptional Control and Processes 31

 University of Washington

exit: Ending a process

¢ void exit(int status)
§ Exits a process
§Status code: 0 is used for a normal exit, nonzero for abnormal exit
§ atexit() registers functions to be executed upon exit

void cleanup(void) {
printf("cleaning up\n");
}

void fork6() {
atexit(cleanup);
fork(); function pointer
exit(0);
}

Autumn 2013 Processes 32

 University of Washington

Zombies
¢ Idea
§ When process terminates, it still consumes system resources
Various tables maintained by OS
§
§ Called a “zombie”
§ A living corpse, half alive and half dead

¢ Reaping
§ Performed by parent on terminated child
§ Parent is given exit status information
§ Kernel discards process
¢ What if parent doesn’t reap?
§ If any parent terminates without reaping a child, then child will be
reaped by init process (pid == 1)
§ But in long-running processes we need explicit reaping
§ e.g., shells and servers
Autumn 2013 Processes 33
 University of Washington

wait: Synchronizing with Children

¢ int wait(int *child_status)
§ Suspends current process (i.e. the parent) until one of its children
terminates
§ Return value is the pid of the child process that terminated
§ On successful return, the child process is reaped
§ If child_status != NULL, then the int that it points to will be set
to a status indicating why the child process terminated
§ NULL is a macro for address 0, the null pointer
§ There are special macros for interpreting this status – see wait(2)

¢ If parent process has multiple children, wait() will return

when any of the children terminates
§ waitpid() can be used to wait on a specific child process

Autumn 2013 Processes 34

 University of Washington

wait Example
void fork_wait() {
int child_status;
pid_t child_pid;

if (fork() == 0) { HC Bye
printf("HC: hello from child\n");
} else {
child_pid = wait(&child_status); CT Bye
printf("CT: child %d has terminated\n”,
child_pid);
}
printf("Bye\n");
exit(0);
}

Autumn 2013 Processes 35

 University of Washington

Process management summary

¢ fork gets us two copies of the same process (but fork()
returns different values to the two processes)
¢ execve has a new process substitute itself for the one that
called it
§ Two-process program:
§ First fork()
§ if (pid == 0) { /* child code */ } else { /* parent code */ }
§ Two different programs:
§ First fork()
§ if (pid == 0) { execve() } else { /* parent code */ }
§ Now running two completely different programs
¢ wait / waitpid used to synchronize parent/child execution
and to reap child process
Autumn 2013 Processes 36
 University of Washington

Summary
¢ Processes
§ At any given time, system has multiple active processes
§ Only one can execute at a time, but each process appears to have total
control of the processor
§ OS periodically “context switches” between active processes
§ Implemented using exceptional control flow

¢ Process management
§ fork: one call, two returns
§ exec: one call, usually no return
§ wait or waitpid: synchronization
§ exit: one call, no return

Autumn 2013 Processes 37

 University of Washington

Detailed examples

Autumn 2013 Processes 38

 University of Washington

Fork Example #2
¢ Both parent and child can continue forking
void fork2()
{
printf("L0\n"); Bye
fork(); L1 Bye
printf("L1\n"); Bye
fork();
L0 L1 Bye
printf("Bye\n");
}

Autumn 2013 Processes 39

 University of Washington

Fork Example #3
¢ Both parent and child can continue forking
void fork3() Bye
{
L2 Bye
printf("L0\n");
fork(); Bye
printf("L1\n"); L1 L2 Bye
fork(); Bye
printf("L2\n"); L2 Bye
fork();
Bye
printf("Bye\n");
} L0 L1 L2 Bye

Autumn 2013 Processes 40

 University of Washington

Fork Example #4
¢ Both parent and child can continue forking
void fork4()
{
printf("L0\n");
if (fork() != 0) {
printf("L1\n"); Bye
if (fork() != 0) {
printf("L2\n"); Bye
fork(); Bye
} L0 L1 L2 Bye
}
printf("Bye\n");
}

Autumn 2013 Processes 41

 University of Washington

Fork Example #5
¢ Both parent and child can continue forking
void fork5()
{
printf("L0\n");
if (fork() == 0) {
printf("L1\n"); Bye
if (fork() == 0) { L2 Bye
printf("L2\n");
L1 Bye
fork();
} L0 Bye
}
printf("Bye\n");
}

Autumn 2013 Processes 42

 University of Washington

Zombie void fork7()

{
if (fork() == 0) {
Example /* Child */
printf("Terminating Child, PID = %d\n",
getpid());
exit(0);
} else {
printf("Running Parent, PID = %d\n",
linux> ./forks 7 & getpid());
[1] 6639 while (1)
Running Parent, PID = 6639 ; /* Infinite loop */
Terminating Child, PID = 6640 }
}
linux> ps
PID TTY TIME CMD
6585 ttyp9 00:00:00 tcsh
6639 ttyp9 00:00:03 forks
¢ ps shows child process as
6640 ttyp9 00:00:00 forks <defunct> “defunct”
6641 ttyp9 00:00:00 ps
linux> kill 6639
[1] Terminated ¢ Killing parent allows child to be
linux> ps reaped by init
PID TTY TIME CMD
6585 ttyp9 00:00:00 tcsh
6642 ttyp9 00:00:00 ps

Autumn 2013 Processes 43

 University of Washington

void fork8()

Non-terminating {
if (fork() == 0) {
/* Child */
Child Example printf("Running Child, PID = %d\n",
getpid());
while (1)
; /* Infinite loop */
} else {
printf("Terminating Parent, PID = %d\n",
getpid());
exit(0);
}
}
linux> ./forks 8
Terminating Parent, PID = 6675
Running Child, PID = 6676 ¢ Child process still active even
linux> ps though parent has terminated
PID TTY TIME CMD
6585 ttyp9 00:00:00 tcsh
6676 ttyp9 00:00:06 forks ¢ Must kill explicitly, or else will keep
6677 ttyp9 00:00:00 ps running indefinitely
linux> kill 6676
linux> ps
PID TTY TIME CMD
6585 ttyp9 00:00:00 tcsh
6678 ttyp9 00:00:00 ps
Autumn 2013 Processes 44
 University of Washington

wait() Example
¢ If multiple children completed, will take in arbitrary order
¢ Can use macros WIFEXITED and WEXITSTATUS to get information about exit
status
void fork10()
{
pid_t pid[N];
int i;
int child_status;
for (i = 0; i < N; i++)
if ((pid[i] = fork()) == 0)
exit(100+i); /* Child */
for (i = 0; i < N; i++) {
pid_t wpid = wait(&child_status);
if (WIFEXITED(child_status))
printf("Child %d terminated with exit status %d\n",
wpid, WEXITSTATUS(child_status));
else
printf("Child %d terminated abnormally\n", wpid);
}
}
Autumn 2013 Processes 45
 University of Washington

waitpid(): Waiting for a Specific Process

¢ waitpid(pid, &status, options)
§ suspends current process until specific process terminates
§ various options (that we won’t talk about)
void fork11()
{
pid_t pid[N];
int i;
int child_status;
for (i = 0; i < N; i++)
if ((pid[i] = fork()) == 0)
exit(100+i); /* Child */
for (i = 0; i < N; i++) {
pid_t wpid = waitpid(pid[i], &child_status, 0);
if (WIFEXITED(child_status))
printf("Child %d terminated with exit status %d\n",
wpid, WEXITSTATUS(child_status));
else
printf("Child %d terminated abnormally\n", wpid);
}
Autumn 2013 Processes 46