Suppose I had the following code:
pid_t pid0, pid1;
pid0 = fork();
if (!pid0) {
pid1 = fork();
if (!pid1) { // p2
/* do something */
return;
} else { // p1
wait(NULL);
printf("p1 done\n");
return;
}
}
// p0
wait(NULL):
printf("p0 done\n");
Is the order of the print statements deterministic, or will it be up to the will of the CPU? In practice, it seems to be the same every time, but the docs for wait() make it seem like it should be random.
EDIT: I've thought about it some more, and I'm wondering if it's always in that order because p0 doesn't have p2 as a child process; that's p1's child. So "p0 done" won't print until p1 finishes waiting for its child, p2. So, I suppose the real question is does wait() wait for children's children, or just processes one "generation" away
Process 0 will always wait for process 1, because that's the only child it has.
And process 1 will always wait for process 2, because that's the only child it has.
Process 0 won't return from wait until process 1 exits, which means that process 1 has already printed its message, because it does that before it exits.
So in this case process 0 cannot continue until after the message has been printed.
Related
I have this piece of code, maybe I'm missing something:
const int NPROCESSES = 32;
pid_t pids[128];
for (int i = 0; i < NPROCESSES;i ++) {
pids[i] = fork();
if (!pids[i]) {
/*... other code ...*/
exit(0);
}
}
for (int i = 0; i < NPROCESSES; i++)
waitpid(pids[i], 0, 0);
The program should launch 32 processes and wait until all processes are terminated. But sometimes the program get blocked with child zombie processes.
Where am I wrong?
Using:
waitpid(pids[i], 0, 0);
you specify an exact order in which the parent will reap its children: it will be the same order as they were created.
As a result, if one of the children blocks or delays for any reason, and some other child which was created later has already finished (and called exit()), the latter will remain in a zombie state, until the parent reaps the former child first.
So if, for example, the process created in the first iteration of the loop needs 1min to complete and the rest 31 processes are completed in 1sec, you're be able to observe 31 zombie processes waiting to be reaped by their parent, who (the parent) will wait to reap that one delayed process first.
To change this behavior, the parent can use:
waitpid(-1, NULL, 0);
instead. A value equal to -1 in the first argument of waitpid() means it will reap any of the child processes, quoting from man 2 waitpid:
The value of pid can be:
< -1
meaning wait for any child process whose process group ID is equal to
the absolute value of pid.
-1
meaning wait for any child process.
0
meaning wait for any child process whose process group ID is equal to
that of the calling process.
> 0
meaning wait for the child whose process ID is equal to the value of
pid.
Alternatively, you can just use:
wait(NULL);
which is the same as waitpid(-1, NULL, 0).
I have this code that requires a parent to fork 3 children.
How do you know (and) where to put the "wait()" statement to kill
zombie processes?
What is the command to view zombie processes if you have Linux
virtual box?
main(){
pid_t child;
printf("-----------------------------------\n");
about("Parent");
printf("Now .. Forking !!\n");
child = fork();
int i=0;
for (i=0; i<3; i++){
if (child < 0) {
perror ("Unable to fork");
break;
}
else if (child == 0){
printf ("creating child #%d\n", (i+1));
about ("Child");
break;
}
else{
child = fork();
}
}
}
void about(char * msg){
pid_t me;
pid_t oldone;
me = getpid();
oldone = getppid();
printf("***[%s] PID = %d PPID = %d.\n", msg, me, oldone);
}
How do you know (and) where to put the "wait()" statement to kill
zombie processes?
If your parent spawns only a small, fixed number of children; does not care when or whether they stop, resume, or finish; and itself exits quickly, then you do not need to use wait() or waitpid() to clean up the child processes. The init process (pid 1) takes responsibility for orphaned child processes, and will clean them up when they finish.
Under any other circumstances, however, you must wait() for child processes. Doing so frees up resources, ensures that the child has finished, and allows you to obtain the child's exit status. Via waitpid() you can also be notified when a child is stopped or resumed by a signal, if you so wish.
As for where to perform the wait,
You must ensure that only the parent wait()s.
You should wait at or before the earliest point where you need the child to have finished (but not before forking), OR
if you don't care when or whether the child finishes, but you need to clean up resources, then you can periodically call waitpid(-1, NULL, WNOHANG) to collect a zombie child if there is one, without blocking if there isn't any.
In particular, you must not wait() (unconditionally) immediately after fork()ing because parent and child run the same code. You must use the return value of fork() to determine whether you are in the child (return value == 0), or in the parent (any other return value). Furthermore, the parent must wait() only if forking was successful, in which case fork() returns the child's pid, which is always greater than zero. A return value less than zero indicates failure to fork.
Your program doesn't really need to wait() because it spawns exactly four (not three) children, then exits. However, if you wanted the parent to have at most one live child at any time, then you could write it like this:
int main() {
pid_t child;
int i;
printf("-----------------------------------\n");
about("Parent");
for (i = 0; i < 3; i++) {
printf("Now .. Forking !!\n");
child = fork();
if (child < 0) {
perror ("Unable to fork");
break;
} else if (child == 0) {
printf ("In child #%d\n", (i+1));
about ("Child");
break;
} else {
/* in parent */
if (waitpid(child, NULL, 0) < 0) {
perror("Failed to collect child process");
break;
}
}
}
return 0;
}
If the parent exits before one or more of its children, which can happen if it does not wait, then the child will thereafter see its parent process being pid 1.
Others have already answered how to get a zombie process list via th ps command. You may also be able to see zombies via top. With your original code you are unlikely to catch a glimpse of zombies, however, because the parent process exits very quickly, and init will then clean up the zombies it leaves behind.
How do you know (and) where to put the "wait()" statement to kill
zombie processes?
You can use wait() anywhere in the parent process, and when the child process terminates it'll be removed from the system. Where to put it is up to you, in your specific case you probably want to put it immediately after the child = fork(); line so that the parent process won't resume its execution until its child has exited.
What is the command to view zombie processes if you have Linux virtual box?
You can use the ps aux command to view all processes in the system (including zombie processes), and the STAT column will be equal to Z if the process is a zombie. An example output would be:
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
daniel 1000 0.0 0.0 0 0 ?? Z 17:15 0:00 command
How do you know (and) where to put the "wait()" statement to kill
zombie processes?
You can register a signal handler for SIGCHLD that sets a global volatile sig_atomic_t flag = 0 variable to 1. Then, at some convenient place in your program, test whether flag is set to 1, and, if so, set it back to 0 and afterwards (for otherwise you might miss a signal) call waitpid(-1, NULL, WNOHANG) in a loop until it tells you that no more processes are to be waited for. Note that the signal will interrupt system calls with EINTR, which is a good condition to check for the value of flag. If you use an indefinitely blocking system call like select(), you might want to specify a timeout after which you check for flag, since otherwise you might miss a signal that was raised after your last waitpid() call but before entering the indefinitely blocking system call. An alternative to this kludge is to use pselect().
Use:
ps -e -opid,ppid,pgid,stat,etime,cmd | grep defunct
to see your zombies, also the ppid and pgid to see the parent ID and process group ID. The etime to see the elapsed (cpu) time your zombie has been alive. The parent ID is useful to send custom signals to the parent process.
If the parent process is right coded to catch and handle the SIGCHLD signal, and to what expected (i.e., wait/reap the zombies), then you can submit:
kill -CHLD <parent_pid>
to tell the parent to reap all their zombies.
What's the right way to tell a specific process to sleep?
I don't fully understand how to control the different processes I create..
I'm trying to make two processes that sleep for 2 and 3 seconds. When process 1 has slept for 2 seconds, and process 2 is still sleeping I want process 3 to start sleeping. But how do I tell a process to sleep? Or don't I?
int main(void)
{
pid_t p1 = fork();
pid_t p2 = fork();
pid_t p3 = fork();
//make p1 sleep(2) and p2 sleep(3)
waitpid(p1, NULL, 0); //waiting for p1 to terminate
//make p3 sleep(2);
}
As you can see I don't understand how to handle processes, or what they really are. I kind of see them as objects, but I'm guessing that's wrong.. I've tried reading some stuff about it, but they are all over 9000 pages PDF's.. A simple explanation of what I should see them as would be appreciated. And yes, this is school-material, but no, it's not an assignment.
Start with man fork which is slightly shorter than 9000 pages. The main thing is that successful fork returns two times: it returns 0 to the child process and the child's PID to the parent process. It's typically used like this:
pid_t pid = fork();
if (pid<0) { /* error while forking */
};
if (!pid) { /* code for child */
play();
whine();
sleep();
exit(0);
} else { /* code for parent */
grunt();
waitpid(...);
}
You don't normally tell the child process to do this and that, you just add code doing it to the appropriate if branch.
In your example, if all forks are successful, you end up with 8 processes:
First fork creates a new process, p1 gets 0 in the new process and some pid in the parent.
Second fork is called both in the original parent and in the child, adding 2 processes to the picture.
p2 gets 0 in all "grandchildren" and two different pids in 2 processes existing before step 2.
Third fork is called in four different processes, adding four other processes to the picture.
You could send it a SIGSTOP and then a SIGCONT I think.
kill(p1, SIGSTOP);
Alternatively and more sanely, since you're only forking and thus have complete controll over the code, you could handle the paths:
if (in_child_1)
sleep(..);
As a side note, in your code more processes are created than you expect. The thing is once p1 is created it starts executing from that point, in parallel with its parent. And so on.
Can someone please explain to me what does fork() != 0 mean? From what I understand I think it means if fork is not false? Or if fork is true then.... I don't understand how Fork() can be true or false, seeing that it just creates a copy of a process into a parent and child. Also if a program where to say if (Fork() == 0) what would that mean?
#include "csapp.h"
int main(void)
{
int x = 3;
if (Fork() != 0)
printf("x=%d\n", ++x);
printf("x=%d\n", --x);
exit(0);
}
fork() returns -1 if it fails, and if it succeeds, it returns the forked child's pid in the parent, and 0 in the child. So if (fork() != 0) tests whether it's the parent process.
From man fork
Return Value
On success, the PID of the child process is returned in the parent,
and 0 is returned in the child. On failure, -1 is returned in the
parent, no child process is created, and errno is set appropriately.
Assuming success, fork returns twice: once in the parent, and once in the child.
OK, I did the OP a disservice: I don't know where csapp.h comes from, but if it's this one then it isn't doing you any favours. I guess it is a thin wrapper on POSIX (eg. around fork()), but maybe works on other platforms too?
Because you mentioned fork() before Fork(), I assumed the latter was a typo, whereas it's actually a library function.
If you had been using fork() directly, it would be reasonable to expect you to check the manpage.
Since you're using a Fork() function provided by some library, that library really ought to document it, and doesn't seem to.
Standard (non csapp) usage is:
pid_t child = fork();
if (child == -1) {
printf("fork failed - %d - %s\n", errno, strerror(errno));
exit(-1);
}
if (child) {
printf("I have a child with pid %d, so I must be the parent!\n", child);
} else {
printf("I don't have a child ... so I must be the child!\n")
}
exit(0);
Let's try explaining it differently... When the function starts there's 1 process, this process has a int x = 3
Once you hit this line of code:
if (fork() != 0)
Now, assuming the fork() worked, we have two processes. They both have the same execution space, they both are going to run the same code (to a point), but the child will get its own copy of x to play with.
fork() will return a 0 to the child process, so from the child processes' prospective, the rest of the function is this:
printf(("x=%d\n", --x);
exit(0);
The parent process on the other hand, will get a valid value back from the fork() command, thus it will execute:
printf("x=%d\n", ++x);
printf("x=%d\n", --x);
exit(0);
What the output at this point will be is anyone's guess... You can't tell if parent or child will run first
But if we assume the parent hits the ++x is the next operation then the output is:
x=4
x=3
x=2
As both parent and child will hit the --x. (the parent's x was 4 after the ++x, will be 3 at the end. The child's x was 3, will be 2 at the end)
From fork() manual:
Upon successful completion, fork() returns a value of 0 to the child process and returns the process ID of the child process to the parent
process. Otherwise, a value of -1 is returned to the parent process, no child process is created, and the global variable errno is set to indi-
cate the error.
After the code execution you have two execution threads. Into the if statement you have the parent process' thread and in else statement you have the child process' thread.
if ( fork() ) {
printf("I am the parent!\n");
} else {
printf("I am the child\n");
}
EDIT
For clarification purpose: fork starts a process, which has a thread, memory and may have other resources. I tried (it seems like that without success) to emphasize the flows of execution by adding the "thread" word.
However, by no means, one can say that "parent" relates to "thread" in "parent process' thread".
Of course, my answer could be improved but I think there are already enough good answers here.
Fork returns 0 for the child process and the process id of the child to the parent process. Hence commonly code has if(fork){ }else code. Which implies that the code inside the if is going to be executed only in a parent.
The better way to deal with it is
pid = fork()
if(pid){
// I am parent. Let us do something that only the parent has to do
}else{
// I am child. Let us do something only the child has to do
}
// This code is common to both
The child pid may be useful to wait upon later or to detach from the parent.
I recommend replacing the if with a switch because there are 3 possible results:
#include <sys/types.h>
#include <unistd.h>
pid_t pid;
switch ((pid = fork ())) {
case -1: /* error creating child. */
break;
case 0: /* I am the child process. */
break;
default: /* I am the parent process. */
break;
}
On success, the PID of the child
process is returned in the parent’s
thread of execution, and a 0 is
returned in the child’s thread of execution.
p = fork();
I'm confused at its manual page,is p equal to 0 or PID?
I'm not sure how the manual can be any clearer! fork() creates a new process, so you now have two identical processes. To distinguish between them, the return value of fork() differs. In the original process, you get the PID of the child process. In the child process, you get 0.
So a canonical use is as follows:
p = fork();
if (0 == p)
{
// We're the child process
}
else if (p > 0)
{
// We're the parent process
}
else
{
// We're the parent process, but child couldn't be created
}
p = fork();
/* assume no errors */
/* you now have two */
/* programs running */
--------------------
if (p > 0) { | if (p == 0) {
printf("parent\n"); | printf("child\n");
... | ...
Processes are structured in a directed tree where you only know your single-parent (getppid()). In short, fork() returns -1 on error like many other system functions, non-zero value is useful for initiator of the fork call (the parent) to know its new-child pid.
Nothing is as good as example:
/* fork/getpid test */
#include <sys/types.h>
#include <unistd.h> /* fork(), getpid() */
#include <stdio.h>
int main(int argc, char* argv[])
{
int pid;
printf("Entry point: my pid is %d, parent pid is %d\n",
getpid(), getppid());
pid = fork();
if (pid == 0) {
printf("Child: my pid is %d, parent pid is %d\n",
getpid(), getppid());
}
else if (pid > 0) {
printf("Parent: my pid is %d, parent pid is %d, my child pid is %d\n",
getpid(), getppid(), pid);
}
else {
printf("Parent: oops! can not create a child (my pid is %d)\n",
getpid());
}
return 0;
}
And the result (bash is pid 2249, in this case):
Entry point: my pid is 16051, parent pid is 2249
Parent: my pid is 16051, parent pid is 2249, my child pid is 16052
Child: my pid is 16052, parent pid is 16051
If you need to share some resources (files, parent pid, etc.) between parent and child, look at clone() (for GNU C library, and maybe others)
Once fork is executed, you have two processes. The call returns different values to each process.
If you do something like this
int f;
f = fork();
if (f == 0) {
printf("I am the child\n");
} else {
printf("I am the parent and the childs pid is %d\n",f);
}
You will see both the messages printed. They're being printed by two separate processes. This is they way you can differentiate between the two processes created.
This is the cool part. It's equal to BOTH.
Well, not really. But once fork returns, you now have two copies of your program running! Two processes. You can sort of think of them as alternate universes. In one, the return value is 0. In the other, it's the ID of the new process!
Usually you will have something like this:
p = fork();
if (p == 0){
printf("I am a child process!\n");
//Do child things
}
else {
printf("I am the parent process! Child is number %d\n", p);
//Do parenty things
}
In this case, both strings will get printed, but by different processes!
fork() is invoked in the parent process. Then a child process is spawned. By the time the child process spawns, fork() has finished its execution.
At this point, fork() is ready to return, but it returns a different value depending on whether it's in the parent or child. In the child process, it returns 0, and in the parent process/thread, it returns the child's process ID.
Fork creates a duplicate process and a new process context. When it returns a 0 value it means that a child process is running, but when it returns another value that means a parent process is running. We usually use wait statement so that a child process completes and parent process starts executing.
I think that it works like this:
when pid = fork(), the code should be executed two times, one is in current process, one is in child process.
So it explains why if/else both execute.
And the order is, first current process, and then execute the child.