CSC 322

SYSTEMS PROGRAMMING

EXCEPTIONAL CONTROL FLOW (2) SIGNALS

(based on chapter 8.5 to 8.6)

Computer Systems: A Programmer’s Perspective 3rd Edition 1

 LINUX PROCESS HIERARCHY

[0]

init [1]

Daemon …
e.g. httpd Login shell Login shell

Child Child Child

Note: you can view the

Grandchild Grandchild hierarchy using the Linux
pstree command

Computer Systems: A Programmer’s Perspective 3rd Edition 2

 AGENDA

oShell
oSignals
oNonlogical Jumps (skipped)

Computer Systems: A Programmer’s Perspective 3rd Edition 3

SHELLS

Computer Systems: A Programmer’s Perspective 3rd Edition 4

 SHELL PROGRAMS
o A shell is an application program that runs programs on behalf of the
user.
• sh Original Unix shell (Stephen Bourne, AT&T Bell Labs, 1977)
• csh/tcsh BSD Unix C shell
• bash “Bourne-Again” Shell (default Linux shell)
int main()
{ Execution is a sequence of
char cmdline[MAXLINE]; /* command line */ read/evaluate steps
while (1) {
/* read */
printf("> ");
fgets(cmdline, MAXLINE, stdin);
if (feof(stdin))
exit(0);

/* evaluate */
eval(cmdline);
}
}
Computer Systems: A Programmer’s Perspective 3rd Edition 5
 SIMPLE SHELL EVAL FUNCTION
o Our example shell correctly waits for and reaps foreground jobs
void eval(char *cmdline)
{
char *argv[MAXARGS]; /* Argument list execve() */
char buf[MAXLINE]; /* Holds modified command line */
int bg; /* Should the job run in bg or fg? */
pid_t pid; /* Process id */

strcpy(buf, cmdline);
bg = parseline(buf, argv);
if (argv[0] == NULL)
return; /* Ignore empty lines */

if (!builtin_command(argv)) {
if ((pid = fork()) == 0) { /* Child runs user job */
if (execve(argv[0], argv, environ) < 0) {
printf("%s: Command not found.\n", argv[0]);
exit(0);
}
}

/* Parent waits for foreground job to terminate */

if (!bg) {
int status;
if (waitpid(pid, &status, 0) < 0)
unix_error("waitfg: waitpid error");
}
else
printf("%d %s", pid, cmdline);
}
return;
}
Computer Systems: A Programmer’s Perspective 3rd Edition 6
 PROBLEM IN THE SHELL EXAMPLE

o The example shell correctly waits for and reaps foreground

jobs
void eval(char *cmdline)
{
char *argv[MAXARGS]; /* Argument list execve() */ • But what about background
char buf[MAXLINE]; /* Holds modified command line */
int bg;
pid_t pid;
/* Should the job run in bg or fg? */
/* Process id */
jobs?
strcpy(buf, cmdline);
bg = parseline(buf, argv);
 Will become zombies when
if (argv[0] == NULL)
return; /* Ignore empty lines */ they terminate
if (!builtin_command(argv)) {
if ((pid = fork()) == 0) { /* Child runs user job */
 Will never be reaped
if (execve(argv[0], argv, environ) < 0) {
printf("%s: Command not found.\n", argv[0]); because shell (typically) will
exit(0);

}
} not terminate
/* Parent waits for foreground job to terminate */
if (!bg) {
 Will create a memory leak
int status;
if (waitpid(pid, &status, 0) < 0)
that could run the kernel out
unix_error("waitfg: waitpid error");
} of memory
else
printf("%d %s", pid, cmdline);
}
return;
}
Computer Systems: A Programmer’s Perspective 3rd Edition 7
 SOLUTION?

o Exceptional control flow (ECF) to the Rescue!

• The kernel will interrupt regular processing to alert us when a
background process completes
• In Unix, the alert mechanism is called a signal

Computer Systems: A Programmer’s Perspective 3rd Edition 8

SIGNALS

Computer Systems: A Programmer’s Perspective 3rd Edition 9

 SIGNALS

oA signal is a small message that notifies a process that an event of

some type has occurred in the system
• Akin to exceptions and interrupts
• Sent from the kernel (sometimes at the request of another process) to a
process
• Signal type is identified by small integer ID’s (1-30)
• Only information in a signal is its ID and the fact that it arrived
ID Name Default Action Corresponding Event
2 SIGINT Terminate User typed ctrl-c
9 SIGKILL Terminate Kill program (cannot override or ignore)
11 SIGSEGV Terminate Segmentation violation
14 SIGALRM Terminate Timer signal
17 SIGCHLD Ignore Child stopped or terminated
From Figure 8.26 in the textbook
Computer Systems: A Programmer’s Perspective 3rd Edition 10
 SIGNAL CONCEPTS: SENDING A SIGNAL

oKernel sends (delivers) a signal to a destination process by

updating some state in the context of the destination process
oKernel sends a signal for one of the following reasons:
• Kernel has detected a system event such as divide-by-zero (SIGFPE) or
the termination of a child process (SIGCHLD)
• Another process has invoked the kill system call to explicitly request
the kernel to send a signal to the destination process

Computer Systems: A Programmer’s Perspective 3rd Edition 11

 SIGNAL CONCEPTS: RECEIVING A SIGNAL

oA destination process receives a signal when it is forced by the

kernel to react in some way to the delivery of the signal
oSome possible ways to react:
• Ignore the signal (do nothing)
• Terminate the process (with optional core dump)
• Catch the signal by executing a user-level function called signal handler
 Akin to a hardware exception handler being called in response to an
asynchronous interrupt:
(1) Signal received (2) Control passes
by process to signal handler
Icurr
Inext (3) Signal
handler runs
(4) Signal handler
returns to
next instruction
Computer Systems: A Programmer’s Perspective 3rd Edition 12
 SIGNAL CONCEPTS: PENDING A SIGNAL

oA signal is pending if sent but not yet received

• There can be at most one pending signal of any particular type
• Important: Signals are not queued
 If a process has a pending signal of type k, then subsequent signals of
type k that are sent to that process are discarded

oKernel maintains pending bit vectors in the context of each

process
• pending: represents the set of pending signals
 Kernel sets bit k in pending when a signal of type k is delivered
 Kernel clears bit k in pending when a signal of type k is received

oA pending signal is received at most once

Computer Systems: A Programmer’s Perspective 3rd Edition 13
 SIGNAL CONCEPTS: BLOCKING A SIGNAL

oA process can block the receipt of certain signals

• Blocked signals can be delivered, but will not be received until the
signal is unblocked

oKernel maintains blocked bit vectors in the context of each

process
• blocked: represents the set of blocked signals
 Can be set and cleared by using the sigprocmask function
 Also referred to as the signal mask.

Computer Systems: A Programmer’s Perspective 3rd Edition 14

 SENDING SIGNALS: PROCESS GROUPS

oEvery process belongs to exactly one process group

pid=10
setpgid(0,0) pgid=10 Shell
will use pid
for pgid.

pid=20 Fore- Back- Back-

pid=32 pid=40
pgid=20 ground ground pgid=32 ground pgid=40
job job #1 job #2

Background Background
process group 32 process group 40
Child Child
• getpgrp()
pid=21 pid=22
pgid=20 pgid=20 Return process group of current process
Foreground • setpgid()
process group 20 Change process group of a process (see
text for details)
Computer Systems: A Programmer’s Perspective 3rd Edition 15
 SENDING SIGNALS WITH /BIN/KILL PROGRAM

o /bin/kill program sends linux> ./forks 16

Child1: pid=24818 pgrp=24817
arbitrary signal to a process or Child2: pid=24819 pgrp=24817
process group linux> ps
PID TTY TIME CMD
• Examples: 24788 pts/2 00:00:00 tcsh
24818 pts/2 00:00:02 forks
 /bin/kill –9 24818 24819 pts/2 00:00:02 forks
 Send SIGKILL to process 24820 pts/2 00:00:00 ps
linux> /bin/kill -9 -24818
24818
linux> ps
PID TTY TIME CMD
24788 pts/2 00:00:00 tcsh
24819 pts/2 00:00:02 forks
24823 pts/2 00:00:00 ps
linux>

Computer Systems: A Programmer’s Perspective 3rd Edition 16

 SENDING SIGNALS WITH /BIN/KILL PROGRAM

o /bin/kill program sends linux> ./forks 16

arbitrary signal to a process or Child1: pid=24818 pgrp=24817
Child2: pid=24819 pgrp=24817
process group linux> ps
PID TTY TIME CMD
• Examples: 24788 pts/2 00:00:00 tcsh
24818 pts/2 00:00:02 forks
 /bin/kill –9 24818 24819 pts/2 00:00:02 forks
 Send SIGKILL to process 24820 pts/2 00:00:00 ps
linux> /bin/kill -9 -24817
24818
linux> ps
 /bin/kill –9 –24817 PID TTY TIME CMD
 Send SIGKILL to every 24788 pts/2 00:00:00 tcsh
24823 pts/2 00:00:00 ps
process in process group 24817 linux>

Computer Systems: A Programmer’s Perspective 3rd Edition 17

 SENDING SIGNALS FROM THE KEYBOARD
o Typing ctrl-c (or ctrl-z) causes the kernel to send a SIGINT
(SIGTSTP) to every job in the foreground process group.
• SIGINT – default action is to terminate each process
• SIGTSTP – default action is to stop (suspend) each process
pid=10
pgid=10 Shell

pid=20 Fore- Back- Back-

pid=32 pid=40
pgid=20 ground ground pgid=32 ground pgid=40
job job #1 job #2
Background Background
process group 32 process group 40
Child Child

pid=21 pid=22
pgid=20 pgid=20

Foreground
process group 20
Computer Systems: A Programmer’s Perspective 3rd Edition 18
 EXAMPLE: CTRL-C AND CTRL-Z
linux> ./forks 17 STAT (process state) Legend:
Child: pid=28108 pgrp=28107
Parent: pid=28107 pgrp=28107 First letter:
<types ctrl-z>
S: sleeping
Suspended
linux> ps w T: stopped
PID TTY STAT TIME COMMAND R: running
27699 pts/8 Ss 0:00 -tcsh
28107 pts/8 T 0:01 ./forks 17 Second letter:
28108 pts/8 T 0:01 ./forks 17 s: session leader
28109 pts/8 R+ 0:00 ps w +: foreground proc group
linux> fg
./forks 17
See “man ps” for more
<types ctrl-c>
linux> ps w details
PID TTY STAT TIME COMMAND
27699 pts/8 Ss 0:00 -tcsh
28110 pts/8 R+ 0:00 ps w

Computer Systems: A Programmer’s Perspective 3rd Edition 19

 SENDING SIGNALS WITH KILL FUNCTION
void fork12()
{
pid_t pid[N];
int i;
int child_status;

for (i = 0; i < N; i++)

if ((pid[i] = fork()) == 0) {
/* Child: Infinite Loop */
while(1)
;
}

for (i = 0; i < N; i++) {

printf("Killing process %d\n", pid[i]);
kill(pid[i], SIGINT);
}

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);
}
}

Computer Systems: A Programmer’s Perspective 3rd Edition 20

 RECEIVING SIGNALS

oSuppose kernel is returning from an exception handler and is

ready to pass control to process p

Process A Process B

user code
kernel code context switch
Time

user code
kernel code context switch

user code

Computer Systems: A Programmer’s Perspective 3rd Edition 21

 RECEIVING SIGNALS
oSuppose kernel is returning from an exception handler and is
ready to pass control to process p
oKernel computes pnb = pending & ~blocked
• The set of pending nonblocked signals for process p
oIf (pnb == 0)
• Pass control to next instruction in the logical flow for p
oElse
• Choose least nonzero bit k in pnb and force process p to receive
signal k
• The receipt of the signal triggers some action by p
• Repeat for all nonzero k in pnb
• Pass control to next instruction in logical flow for p
Computer Systems: A Programmer’s Perspective 3rd Edition 22
 DEFAULT ACTIONS

oEach signal type has a predefined default action, which is one

of:
• The process terminates
• The process stops until restarted by a SIGCONT signal
• The process ignores the signal

Computer Systems: A Programmer’s Perspective 3rd Edition 23

 INSTALLING SIGNAL HANDLERS
o The signal function modifies the default action associated with
the receipt of signal signum:
• handler_t *signal(int signum, handler_t *handler)

o Different values for handler:

• SIG_IGN: ignore signals of type signum
• SIG_DFL: revert to the default action on receipt of signals of type
signum
• Otherwise, handler is the address of a user-level signal handler
 Called when process receives signal of type signum
 Referred to as “installing” the handler
 Executing handler is called “catching” or “handling” the signal
 When the handler executes its return statement, control passes back to
instruction in the control flow of the process that was interrupted by receipt of
the signal
Computer Systems: A Programmer’s Perspective 3rd Edition 24
 SIGNAL HANDLING EXAMPLE
void sigint_handler(int sig) /* SIGINT handler */
{
printf("So you think you can stop the bomb with ctrl-c, do you?\n");
sleep(2);
printf("Well...");
fflush(stdout);
sleep(1);
printf("OK. :-)\n");
exit(0);
}

int main()
{
/* Install the SIGINT handler */
if (signal(SIGINT, sigint_handler) == SIG_ERR)
unix_error("signal error");

/* Wait for the receipt of a signal */

pause();

return 0;
}
Computer Systems: A Programmer’s Perspective 3rd Edition 25
 SIGNALS HANDLERS AS CONCURRENT FLOWS

oA signal handler is a separate logical flow (not process) that

runs concurrently with the main program
Process A Process A Process B

while (1); handler () {

…
}
Time

Computer Systems: A Programmer’s Perspective 3rd Edition 26

SIGNALS HANDLERS AS CONCURRENT FLOWS

Process A Process B

user code (main)

Icurr
kernel code context switch

user code (main)

Time

kernel code context switch

user code (handler)

kernel code
Inext
user code (main)

Computer Systems: A Programmer’s Perspective 3rd Edition 27

 NESTED SIGNAL HANDLERS

oHandlers can be interrupted by other handlers.

Main program Handler S Handler T

(2) Control passes

(1) Program Icurr to handler S
catches signal s (4) Control passes
(3) Program to handler T
(7) Main program Inext catches signal t
resumes
(5) Handler T
(6) Handler S
returns to
returns to
handler S
main program

Computer Systems: A Programmer’s Perspective 3rd Edition 28

 BLOCKING AND UNBLOCKING SIGNALS

oImplicit blocking mechanism

• Kernel blocks any pending signals of type currently being handled.
e.g., A SIGINT handler can’t be interrupted by another SIGINT

oExplicit blocking and unblocking mechanism

• sigprocmask function
oSupporting functions
• sigemptyset – Create empty set
• sigfillset – Add every signal number to set
• sigaddset – Add signal number to set
• sigdelset – Delete signal number from set

Computer Systems: A Programmer’s Perspective 3rd Edition 29

 TEMPORARILY BLOCKING SIGNALS

sigset_t mask, prev_mask;

Sigemptyset(&mask);
Sigaddset(&mask, SIGINT);

/* Block SIGINT and save previous blocked set */

Sigprocmask(SIG_BLOCK, &mask, &prev_mask);
…
/* Code region that will not be interrupted by SIGINT */
…
/* Restore previous blocked set, unblocking SIGINT */
Sigprocmask(SIG_SETMASK, &prev_mask, NULL);
…

Computer Systems: A Programmer’s Perspective 3rd Edition 30

 SAFE SIGNAL HANDLING

oHandlers are tricky because they are concurrent with main

program and share the same global data structures.
• Shared data structures can become corrupted.
• We’ll explore concurrency issues later in the term.
oFor now here are some guidelines to help you avoid trouble.

Computer Systems: A Programmer’s Perspective 3rd Edition 31

 GUIDELINES FOR SAFE HANDLERS
o G0: Keep your handlers as simple as possible
e.g., Set a global flag and return
o G1: Call only async-signal-safe functions in your handlers
• printf, sprintf, malloc, and exit are not safe!
o G2: Save and restore errno on entry and exit
• So that other handlers don’t overwrite your value of errno
o G3: Protect accesses to shared data structures by temporarily blocking all
signals.
• To prevent possible corruption
o G4: Declare global variables as volatile
• To prevent compiler from storing them in a register
o G5: Declare global flags as volatile sig_atomic_t
• flag: variable that is only read or written (e.g. flag = 1, not flag++)
• Flag declared this way does not need to be protected like other globals
Computer Systems: A Programmer’s Perspective 3rd Edition 32
 ASYNC-SIGNAL-SAFETY

oFunction is async-signal-safe if either reentrant (e.g., all

variables stored on stack frame) or non-interruptible by
signals.
oPosix guarantees 117 functions to be async-signal-safe
• Source: “man 7 signal”
• Popular functions on the list:
 _exit, write, wait, waitpid, sleep, kill
• Popular functions that are not on the list:
 printf, sprintf, malloc, exit
 Unfortunate fact: write is the only async-signal-safe output function

Computer Systems: A Programmer’s Perspective 3rd Edition 33

 SAFELY GENERATING FORMATTED OUTPUT

oUse the reentrant SIO (Safe I/O library) from csapp.c in your
handlers.
• ssize_t sio_puts(char s[]) /* Put string */
• ssize_t sio_putl(long v) /* Put long */
• void sio_error(char s[]) /* Put msg & exit */
void sigint_handler(int sig) /* Safe SIGINT handler */
{
Sio_puts("So you think you can stop the bomb with ctrl-c, do you?\n");
sleep(2);
Sio_puts("Well...");
sleep(1);
Sio_puts("OK. :-)\n");
_exit(0);
}

Computer Systems: A Programmer’s Perspective 3rd Edition 34

int ccount = 0;
CORRECT SIGNAL
void child_handler(int sig) {
int olderrno = errno;
HANDLING
pid_t pid;
if ((pid = wait(NULL)) < 0)
oPending signals are not queued
Sio_error("wait error");
ccount--; • For each signal type, one bit
Sio_puts("Handler reaped child "); indicates whether or not signal is
Sio_putl((long)pid);
Sio_puts(" \n"); pending…
sleep(1);
errno = olderrno; • …thus at most one pending signal of
}
any particular type.
void fork14() {
pid_t pid[N]; o You can’t use signals to count
int i;
ccount = N; events, such as children
Signal(SIGCHLD, child_handler); terminating.
for (i = 0; i < N; i++) {
if ((pid[i] = Fork()) == 0) { linux> ./forks 14
Sleep(1); Handler reaped child 23240
exit(0); /* Child exits */
Handler reaped child 23241
}
}
while (ccount > 0) /* Parent spins */
;
} Computer Systems: A Programmer’s Perspective 3rd Edition 35
 CORRECT SIGNAL HANDLING

oMust wait for all terminated child processes

• Put wait in a loop to reap all terminated children
void child_handler2(int sig)
{
int olderrno = errno;
pid_t pid;
while ((pid = wait(NULL)) > 0) {
ccount--;
Sio_puts("Handler reaped child ");
Sio_putl((long)pid);
linux> ./forks 15
Sio_puts(" \n"); Handler reaped child 23246
} Handler reaped child 23247
if (errno != ECHILD) Handler reaped child 23248
Sio_error("wait error"); Handler reaped child 23249
errno = olderrno; Handler reaped child 23250
}
linux>

Computer Systems: A Programmer’s Perspective 3rd Edition 36

 PORTABLE SIGNAL HANDLING
oDifferent versions of Unix can have different signal handling
semantics
• Some older systems restore action to default after catching signal
• Some interrupted system calls can return with errno == EINTR
• Some systems don’t block signals of the type being handled
oSolution: sigaction
handler_t *Signal(int signum, handler_t *handler)
{
struct sigaction action, old_action;

action.sa_handler = handler;
sigemptyset(&action.sa_mask); /* Block sigs of type being handled */
action.sa_flags = SA_RESTART; /* Restart syscalls if possible */

if (sigaction(signum, &action, &old_action) < 0)

unix_error("Signal error");
return (old_action.sa_handler);
}
Computer Systems: A Programmer’s Perspective 3rd Edition 37
 SYNCHRONIZING FLOW TO AVOID RACES

oSimple shell with a subtle synchronization error because it

assumes parent runs before child.
int main(int argc, char **argv)
{
int pid;
sigset_t mask_all, prev_all;

Sigfillset(&mask_all);
Signal(SIGCHLD, handler);
initjobs(); /* Initialize the job list */

while (1) {
if ((pid = Fork()) == 0) { /* Child */
Execve("/bin/date", argv, NULL);
}
Sigprocmask(SIG_BLOCK, &mask_all, &prev_all); /* Parent */
addjob(pid); /* Add the child to the job list */
Sigprocmask(SIG_SETMASK, &prev_all, NULL);
}
exit(0);
}

Computer Systems: A Programmer’s Perspective 3rd Edition 38

 SYNCHRONIZING FLOW TO AVOID RACES

oSIGCHILD handler for a simple shell

void handler(int sig)
{
int olderrno = errno;
sigset_t mask_all, prev_all;
pid_t pid;

Sigfillset(&mask_all);
while ((pid = waitpid(-1, NULL, 0)) > 0) { /* Reap child */
Sigprocmask(SIG_BLOCK, &mask_all, &prev_all);
deletejob(pid); /* Delete the child from the job list */
Sigprocmask(SIG_SETMASK, &prev_all, NULL);
}
if (errno != ECHILD)
Sio_error("waitpid error");
errno = olderrno;
}

Computer Systems: A Programmer’s Perspective 3rd Edition 39

 CORRECTED SHELL PROGRAM WITHOUT RACE
int main(int argc, char **argv)
{
int pid;
sigset_t mask_all, mask_one, prev_one;

Sigfillset(&mask_all);
Sigemptyset(&mask_one);
Sigaddset(&mask_one, SIGCHLD);
Signal(SIGCHLD, handler);
initjobs(); /* Initialize the job list */

while (1) {
Sigprocmask(SIG_BLOCK, &mask_one, &prev_one); /* Block SIGCHLD */
if ((pid = Fork()) == 0) { /* Child process */
Sigprocmask(SIG_SETMASK, &prev_one, NULL); /* Unblock SIGCHLD */
Execve("/bin/date", argv, NULL);
}
Sigprocmask(SIG_BLOCK, &mask_all, NULL); /* Parent process */
addjob(pid); /* Add the child to the job list */
Sigprocmask(SIG_SETMASK, &prev_one, NULL); /* Unblock SIGCHLD */
}
exit(0);
}
Computer Systems: A Programmer’s Perspective 3rd Edition 40
 EXPLICITLY WAITING FOR SIGNALS

o Handlers for program explicitly waiting for SIGCHLD to arrive.

volatile sig_atomic_t pid;

void sigchld_handler(int s)
{
int olderrno = errno;
pid = Waitpid(-1, NULL, 0); /* Main is waiting for nonzero pid */
errno = olderrno;
}

void sigint_handler(int s)
{
}

Computer Systems: A Programmer’s Perspective 3rd Edition 41

 EXPLICITLY WAITING FOR SIGNALS
int main(int argc, char **argv) {
sigset_t mask, prev; Similar to a shell waiting
Signal(SIGCHLD, sigchld_handler); for a foreground job to
Signal(SIGINT, sigint_handler); terminate.
Sigemptyset(&mask);
Sigaddset(&mask, SIGCHLD);

while (1) {
Sigprocmask(SIG_BLOCK, &mask, &prev); /* Block SIGCHLD */
if (Fork() == 0) /* Child */
exit(0);
/* Parent */
pid = 0;
Sigprocmask(SIG_SETMASK, &prev, NULL); /* Unblock SIGCHLD */

/* Wait for SIGCHLD to be received (wasteful!) */

while (!pid)
;
/* Do some work after receiving SIGCHLD */
printf(".");
}
exit(0);
}
Computer Systems: A Programmer’s Perspective 3rd Edition 42
 WAITING FOR SIGNALS WITH SIGSUSPEND

oProgram is correct, but very wasteful

oOther options:
while (!pid) /* Race! */ while (!pid) /* Too slow! */
pause(); sleep(1);

oSolution: sigsuspend
int sigsuspend(const sigset_t *mask)
• Equivalent to atomic (uninterruptable) version of:
sigprocmask(SIG_BLOCK, &mask, &prev);
pause();
sigprocmask(SIG_SETMASK, &prev, NULL);

Computer Systems: A Programmer’s Perspective 3rd Edition 43

 WAITING FOR SIGNALS WITH SIGSUSPEND
int main(int argc, char **argv) {
sigset_t mask, prev;
Signal(SIGCHLD, sigchld_handler);
Signal(SIGINT, sigint_handler);
Sigemptyset(&mask);
Sigaddset(&mask, SIGCHLD);

while (1) {
Sigprocmask(SIG_BLOCK, &mask, &prev); /* Block SIGCHLD */
if (Fork() == 0) /* Child */
exit(0);

/* Wait for SIGCHLD to be received */

pid = 0;
while (!pid)
Sigsuspend(&prev);

/* Optionally unblock SIGCHLD */

Sigprocmask(SIG_SETMASK, &prev, NULL);
/* Do some work after receiving SIGCHLD */
printf(".");
}
exit(0);
}
Computer Systems: A Programmer’s Perspective 3rd Edition 44
NONLOGICAL JUMPS

Computer Systems: A Programmer’s Perspective 3rd Edition 45

 NONLOCAL JUMPS: SETJMP/LONGJMP

oPowerful (but dangerous) user-level mechanism for transferring

control to an arbitrary location
• Controlled to way to break the procedure call / return discipline
• Useful for error recovery and signal handling
oint setjmp(jmp_buf j)
• Must be called before longjmp
• Identifies a return site for a subsequent longjmp
• Called once, returns one or more times
oImplementation:
• Remember where you are by storing the current register context, stack
pointer, and PC value in jmp_buf
• Return 0

Computer Systems: A Programmer’s Perspective 3rd Edition 46

 SEGJMP/LONGJMP
void longjmp(jmp_buf j, int i)
• Meaning:
 return from the setjmp remembered by jump buffer j again ...
 … this time returning i instead of 0
• Called after setjmp
• Called once, but never returns
longjmp Implementation:
• Restore register context (stack pointer, base pointer, PC value) from
jump buffer j
• Set %eax (the return value) to i
• Jump to the location indicated by the PC stored in jump buf j

Computer Systems: A Programmer’s Perspective 3rd Edition 47

 EXAMPLE SETJMP/LONGJMP

oGoal: return directly to original caller from a deeply-nested

function
/* Deeply nested function foo */
void foo(void)
{
if (error1)
longjmp(buf, 1);
bar();
}

void bar(void)
{
if (error2)
longjmp(buf, 2);
}

Computer Systems: A Programmer’s Perspective 3rd Edition 48

 EXAMPLE SETJMP/LONGJMP
jmp_buf buf;

int error1 = 0;
int error2 = 1;

void foo(void), bar(void);

int main()
{
switch(setjmp(buf)) {
case 0:
foo();
break;
case 1:
printf("Detected an error1 condition in foo\n");
break;
case 2:
printf("Detected an error2 condition in foo\n");
break;
default:
printf("Unknown error condition in foo\n");
}
exit(0);
}
Computer Systems: A Programmer’s Perspective 3rd Edition 49
 LIMITATIONS OF NONLOGICAL JUMPS

oWorks within stack discipline

• Can only long jump to environment of function that has been called but
not yet completed
jmp_buf env; Before After
longjmp longjmp
P1()
{ env P1 P1
if (setjmp(env)) {
/* Long Jump to here */
} else {
P2
P2();
} P2
}
P2
P2()
{ . . . P2(); . . . P3(); } P3
P3()
{
longjmp(env, 1);
}
Computer Systems: A Programmer’s Perspective 3rd Edition 50
 LIMITATIONS OF LONG JUMPS

oWorks within stack discipline

• Can only long jump to environment of function that has been called but
not yet completed
jmp_buf env; P1
env
P1() P2
{
P2(); P3(); At setjmp
}
P1
P2()
{ env
if (setjmp(env)) { X P2
/* Long Jump to here */ P2 returns
}
}
P1
P3()
env
{ X P3
longjmp(env, 1);
} At longjmp
Computer Systems: A Programmer’s Perspective 3rd Edition 51
 PUTTING IT ALL TOGETHER: A PROGRAM
THAT RESTARTS ITSELF WHEN CTRL-C’D
#include "csapp.h"

sigjmp_buf buf;

void handler(int sig)

{
siglongjmp(buf, 1);
}

int main()
{ linux> ./restart
if (!sigsetjmp(buf, 1)) { starting
Signal(SIGINT, handler); processing...
Sio_puts("starting\n"); processing...
} processing...
else restarting
Sio_puts("restarting\n"); Ctrl-c
processing...
while(1) { processing...
Sleep(1); restarting Ctrl-c
Sio_puts("processing...\n"); processing...
} processing...
exit(0); /* Control never reaches here */ processing...
}
Computer Systems: A Programmer’s Perspective 3rd Edition 52
 SUMMARY

oSignals provide process-level exception handling

• Can generate from user programs
• Can define effect by declaring signal handler
• Be very careful when writing signal handlers
oNonlocal jumps provide exceptional control flow within process
• Within constraints of stack discipline

Computer Systems: A Programmer’s Perspective 3rd Edition 53