C - Waiting For Multiple Threads to Terminate - c

I am trying to wait for all the threads to terminate before the main() process terminates. Here is what I have so far:
void* mapperFunction()
{
printf("Hello\n");
return NULL;
}
int main()
{
int i; // Used in "for" loops.
int N = 3;
pthread_t* mapperThreads = (pthread_t*) malloc(sizeof(pthread_t) * N);
for ( i = 0; i < N; i++)
{ // Creates all the mapper threads.
pthread_create( &mapperThreads[N], NULL, mapperFunction, NULL);
}
for ( i = 0; i < N; i++)
{ // Waits for all the mapper threads to terminate.
pthread_join( mapperThreads[N],NULL);
}
return 0;
}
I get three different outputs when I run this code;
1- Hello\n
2- Helle\nHello\n
3- Hello\nHello\nHello\n
It looks like the main() process does not always wait for all threads to terminate. What am I doing wrong?

You want &mapperThreads[i] instead of &mapperThreads[N] in each case.

Maybe pthread_barrier_wait is what you're looking for ?
Link

Related

why does Pthread Signal from main, hangs the code?

The requirement for the sample application creating using pthread is given below:
A single thread is created from main using Pthread
Inside thread, Mutex is locked, a counter counts the value and while loop is incremented, While is set to the max count of 10.
after while loop finishes, the mutex is unlocked.
The above requirement I have tried implementing using pthread
Code is shown below:
#include <pthread.h>
#include <stdio.h>
pthread_mutex_t count_mutex;
pthread_cond_t count_threshold_cv;
int samples = 10;
int count = 0;
struct example
{
int i;
int a;
};
void *inc_x(void *x_void_ptr)
{
pthread_mutex_lock(&count_mutex);
printf("Thread is locked \n");
while(count < samples)
{
printf("inside While loop \n");
struct example *E2_ptr;
E2_ptr = (struct example *)x_void_ptr;
printf("inside thread count = %d\n",count);
E2_ptr->a = count;
E2_ptr->i = (count + 1);
count ++;
//pthread_cond_wait(&count_threshold_cv, &count_mutex);
}
pthread_mutex_unlock(&count_mutex);
printf ( "\n Test Successful for Thread\n");
pthread_exit(NULL);
}
int main()
{
int x = 100, y = 0,i = 0;
struct example *E1_ptr;
E1_ptr->a = 0;
E1_ptr->i = 0;
printf("Before\t E1_ptr->a = %d\t, E1_ptr->i = %d\n",E1_ptr->a,E1_ptr->i);
pthread_t inc_x_thread;
if(pthread_create(&inc_x_thread, NULL, inc_x, E1_ptr))
{
printf("Error creating thread\n");
}
if(pthread_join(inc_x_thread, NULL))
{
printf("Error joining thread\n");
}
for(i = 0; i<(samples-1); i++)
{
if(pthread_cond_signal(&count_threshold_cv))
{
printf("Error Signaling thread at sample = %d\n",i);
}
}
printf("after\t E1_ptr->a = %d\t, E1_ptr->i = %d\n",E1_ptr->a,E1_ptr->i);
pthread_mutex_destroy(&count_mutex);
pthread_cond_destroy(&count_threshold_cv);
pthread_exit (NULL);
return 0;
}
Doubt:
In the above code, the thread executes its function properly and exits.
Once the condition is applied, i.e. the below shown code is uncommented then,
pthread_cond_wait(&count_threshold_cv, &count_mutex);
Then the thread is stopped after 1st iteration of while loop as expected.
The signal is generated from main by the code shown below:
for(i = 0; i<(samples-1); i++)
{
if(pthread_cond_signal(&count_threshold_cv))
{
printf("Error Signaling thread at sample = %d\n",i);
}
}
observed that the signal is never sent.
Can someone please guide me, where am I going wrong. I'm a newbie to Pthreads.
Thanks in advance.
count_mutex and count_threshold_cv are not initialized, add:
int main()
{
pthread_mutex_init(&count_mutex, NULL);
pthread_cond_init(&count_threshold_cv, NULL);
//...
E1_ptr is not initialized.
There are many ways to solve it:
You can call malloc to allocate memory:
struct example *E1_ptr = malloc(sizeof(struct example));
E1_ptr->a = 0;
E1_ptr->i = 0;
or holds pointer to local variable:
struct example ex;
struct example *E1_ptr = &ex; //malloc(sizeof(struct example));
E1_ptr->a = 0;
E1_ptr->i = 0;
or
struct example ex;
ex.a = 0;
ex.i = 0;
then create thread with pthread_create(&inc_x_thread, NULL, inc_x, &ex)
pthread_cond_signal function does not wait. If a thread is blocked by condition variable pthread_cond_signal function unblocks this thread, otherwise returns immediately without waiting and does nothing. So your for loop with 10 iterations is executed as soon as possible, without any waiting for pthread_cond_wait is called.
So can rewrite your for loop to infinite loop, calling pthread_cond_signal repeatedly.
if(pthread_create(&inc_x_thread, NULL, inc_x, E1_ptr)) {
printf("Error creating thread\n");
}
while(1) { // INFINITE LOOP
if(pthread_cond_signal(&count_threshold_cv)) {
printf("Error Signaling thread at sample = %d\n",i);
}
if (taskDone) // testing global flag, if 1 break
break; // it means inc_x thread has ended
}
if(pthread_join(inc_x_thread, NULL)) { // it was pointed out in comment
printf("Error joining thread\n"); // you need to join at the end of main function
}
taskDone is global int, with 0 as default value. It is set to 1 before pthread_exit is called in inc_x function. Setting/checking taskDone should be wrapped with some synchronization mechanism, for example by adding new mutex or use count_mutex.

Pass argument to multiple threads

New to C, I was reading here about how to properly pass an argument to a thread. What about if this argument needed to be passed to multiple threads? Where/how should I use free()? Say:
void *foo(void *i) {
int a = *((int *) i);
while(1){
printf("foo running \n");
sleep(1);
}
}
void *bar(void *i) {
int a = *((int *) i);
while(1){
printf("bar running \n");
sleep(1);
}
}
int main() {
pthread_t threads[2];
int i;
for (i = 0; i < 2; i++ ) {
int *arg = malloc(sizeof(*arg));
if ( arg == NULL ) {
fprintf(stderr, "Couldn't allocate memory for thread arg.\n");
exit(1);
}
*arg = i;
pthread_create(&threads[0], NULL, foo, arg);
pthread_create(&threads[1], NULL, bar, arg);
}
for (i = 0; i < 2; i++){
pthread_join(threads[i],NULL);
}
return 0;
}
Is calling free(arg); in main after spawning the threads the same thing / safe?
If all threads needs the exact same argument, and they are not modifying the argument, there's no need to allocate it dynamically at all, instead just declare it as a variable at function scope in the main function. If there's no dynamic allocation, there's no need to free it.
On the other hand if you need separate arguments in a loop like you do, you need to keep track of all arguments, for example using an array:
// Rest of program...
#define NUMBER_OF_ITERATIONS 2
int main(void)
{
int args[NUMBER_OF_ITERATIONS];
pthread_t threads[NUMBER_OF_ITERATIONS][2];
// Create threads
for (unsigned i = 0; i < NUMBER_OF_ITERATIONS; ++i)
{
args[i] = i;
pthread_create(&threads[i][0], NULL, foo, &args[i]);
pthread_create(&threads[i][1], NULL, bar, &args[i]);
}
// Wait for threads to finish
for (unsigned i = 0; i < NUMBER_OF_ITERATIONS; ++i)
{
pthread_join(threads[i][0]);
pthread_join(threads[i][1]);
}
return 0;
}
The program above also solves another problem you have, when you create a total of four threads but only join two.
You need to make sure both threads have finished before calling free(arg).
That means you can do it after you call pthread_join on both threads.
for (i = 0; i < 2; i++){
pthread_join(threads[i],NULL);
}
free(args);

How to make main thread wait for all child threads finish?

I intend to fire 2 threads in the main thread, and the main thread should wait till all the 2 child threads finish, this is how I do it.
void *routine(void *arg)
{
sleep(3);
}
int main()
{
for (int i = 0; i < 2; i++) {
pthread_t tid;
pthread_create(&tid, NULL, routine, NULL);
pthread_join(&tid, NULL); //This function will block main thread, right?
}
}
In the above code, pthread_join indeed makes main thread wait for the child threads, but the problem is, the second thread won't be created untill the first one finishes. This is not what I want.
What I want is, the 2 threads get created immediatly in the main thread, and then main thread waits for them to finish. Seems like pthread_join cannot do the trick, can it?
I thought, maybe via a semaphore I can do the job, but any other way?
int main()
{
pthread_t tid[2];
for (int i = 0; i < 2; i++) {
pthread_create(&tid[i], NULL, routine, NULL);
}
for (int i = 0; i < 2; i++)
pthread_join(tid[i], NULL);
return 0;
}
First create all the threads, then join all of them:
pthread_t tid[2];
/// create all threads
for (int i = 0; i < 2; i++) {
pthread_create(&tid[i], NULL, routine, NULL);
}
/// wait all threads by joining them
for (int i = 0; i < 2; i++) {
pthread_join(tid[i], NULL);
}
Alternatively, have some pthread_attr_t variable, use pthread_attr_init(3) then pthread_attr_setdetachedstate(3)
on it, then pass its address to pthread_create(3) second argument. Thos would create the threads in detached state. Or use pthread_detach as explained in Jxh's answer.
Remember to read some good Pthread tutorial. You may want to use mutexes and condition variables.
You could use frameworks wrapping them, e.g. Qt or POCO (in C++), or read a good C++ book and use C++ threads.
Conceptually, threads have each their call stack and are related to continuations. They are "heavy".
Consider some agent-oriented programming approach: as a rule of thumb, you don't want to have a lot of threads (e.g. 20 threads on a 10 core processor is reasonable, 200 threads won't be unless a lot of them are sleeping or waiting) and and do want threads to synchronize using mutex and condition variables and communicate and/or synchronize with other threads quite often (several times per second). See also poll(2), fifo(7), unix(7), sem_overview(7) with shm_overview(7) as another way of communicating between threads. In general, avoid using signal(7) with threads (read signal-safety(7)...), and use dlopen(3) with caution (probably only in the main thread).
A pragmatical approach would be to have most of your threads running some event loop (using poll(2), pselect(2), perhaps eventfd(2), signalfd(2), ....), perhaps communicating using pipe(7) or unix(7) sockets. See also socket(7).
Don't forget to document (on paper) the communication protocols between threads. For a theoretical approach, read books about π-calculus and be aware of Rice's theorem : debugging concurrent programs is difficult.
You could start the threads detached, and not worry about joining.
for (int i = 0; i < 2; i++) {
pthread_t tid;
pthread_create(&tid, NULL, routine, NULL);
pthread_detach(tid);
}
pthread_exit(0);
Or, alternatively, you can have the thread that dies report back to the main thread who it is, so that the threads are joined in the order they exited, rather than in the order you created them in.
void *routine(void *arg)
{
int *fds = (int *)arg;
pthread_t t = pthread_self();
usleep((rand()/(1.0 + RAND_MAX)) * 1000000);
write(fds[1], &t, sizeof(t));
}
int main()
{
int fds[2];
srand(time(0));
pipe(fds);
for (int i = 0; i < 2; i++) {
pthread_t tid;
pthread_create(&tid, NULL, routine, fds);
printf("created: %llu\n", (unsigned long long)tid);
}
for (int i = 0; i < 2; i++) {
pthread_t tid;
read(fds[0], &tid, sizeof(tid));
printf("joining: %llu\n", (unsigned long long)tid);
pthread_join(tid, 0);
}
pthread_exit(0);
}
#include<stdio.h>
#include<pthread.h>
int icnt = 0; //in non_bss data segment
pthread_mutex_t lock; //lock variable created stored into bss data segment
void *Thread_count(void* args) //syncronization
{
pthread_mutex_lock(&lock); //lock aquire
icnt++;
for(int x = 1; x <= icnt; x++)
{
printf("Hello from Thread_count : %d \n",icnt);
}
printf("\n");
pthread_mutex_unlock(&lock); //release lock
pthread_exit(NULL); //exit from child thread
}
int main()
{
pthread_t threads[4]; //created array of {unsigned long int}
int status = 0;
//creating threads in loop
for(int i = 1; i <= sizeof(threads)/sizeof(threads[0]); i++)
{
pthread_create(&threads[i], NULL, &Thread_count, NULL);
}
//waiting for threads in loop
for(int j = 1; j <= sizeof(threads)/sizeof(threads[0]); j++)
{
pthread_join(threads[j], &status);
printf("Thread number : %d <--> Thread status : %d\n",j, status);
}
pthread_exit(0); //end of main thread
}

Using of shared variable by 10 pthreads

The problem is in following:
I want to write a short program that creates 10 threads and each prints a tread "id" that is passed to thread function by pointer.
Full code of the program is below:
#include <pthread.h>
#include <stdio.h>
#include <stdlib.h>
struct params {
pthread_mutex_t mutex;
int id;
};
typedef struct params params_t;
void* hello(void* arg){
int id;
pthread_mutex_lock(&(*(params_t*)(arg)).mutex);
id = (*(params_t*)(arg)).id;
pthread_mutex_unlock(&(*(params_t*)(arg)).mutex);
printf("Hello from %d\n", id);
}
int main() {
pthread_t threads[10];
params_t params;
pthread_mutex_init (&params.mutex , NULL);
int i;
for(i = 0; i < 10; i++) {
params.id = i;
if(pthread_create(&threads[i], NULL, hello, &params));
}
for(i = 0; i < 10; i++) {
pthread_join(threads[i], NULL);
}
return 0;
}
The supposed output is (not necessary in this order):
Hello from 0
....
Hello from 9
Actual result is:
Hello from 2
Hello from 3
Hello from 3
Hello from 4
Hello from 5
Hello from 6
Hello from 8
Hello from 9
Hello from 9
Hello from 9
I tried to place mutex in different places in hello() function, but it didn't help.
How should I implement thread sync?
EDIT: Supposed result is not necessary 0...9 it can be any combination of these numbers, but each one should appear only one time.
The problem lies in the below code:
for(i = 0; i < 10; i++)
{
params.id = i;
if(pthread_create(&threads[i], NULL, hello, &params));
}
Your params.id value keeps getting updated in the main thread, whereas you are passing the same pointer to all the threads.
Please create seperate memory for params by dynamically allocating it and pass it to different threads to solve the problem.
EDIT1:
Your usage of mutex to protect is also an incorrect idea. Though your mutex if used in main while setting the id also, may make the updation mutually exclusive, but you may not get your desired output. Instead of getting values from 0 .. 9 in different threads, you may get all 9s or still multiple threads may print same values.
So, using thread synchronization is not such a good idea for the output which you are expecting. If you still need to use one param variable between all threads and get output as 0 to 9 from each of the threads, better move the pthread_join into the first loop. This will ensure that each thread gets created, prints the value and then returns before the main spawns the next thread. In this case, you don't need the mutex also.
EDIT2:
As for the updated question, where it is asked that it is not necessary to print the numbers 0..9 in a sequence, the printing can be random, but only once, the problem still remains the same more or less.
Now, let's say, the value of params.id is first 0 and thread 0 got created, now, thread 0 must print it before it is updated in the main thread, else, when thread 0 accessess it, the value of params.id would have become 1 and you will never get your unique set of values. So, how to ensure that thread 0 prints it before it is updated in main, Two ways for it:
Ensure thread 0 completes execution and printing before main updates
the value
Use condition variables & signalling to ensure that main thread waits
for thread 0 to complete printing before it updates the value (Refer
to Arjun's answer below for more details)
In my honest opinion, you have selected the wrong problem for learning synchronization & shared memory. You can try this with some good problems like "Producer-Consumer", where you really need synchronization for things to work.
There are two problems:
A. You're using a lock but main is unaware of this lock.
B. A lock is not enough in this case. What you would want is for threads to cooperate by signalling each other (because you want main to not increment the variable until a thread says that it is done printing it). You can use a pthread_cond_t to achieve this (Look here to learn more about this). This boils down to the following code (basically, I added an appropriate usage of pthread_cond_t to your code, and a bunch of comments explaining what is going on):
#include <pthread.h>
#include <stdio.h>
#include <stdlib.h>
struct params {
pthread_mutex_t mutex;
pthread_cond_t done;
int id;
};
typedef struct params params_t;
void* hello(void* arg){
int id;
/* Lock. */
pthread_mutex_lock(&(*(params_t*)(arg)).mutex);
/* Work. */
id = (*(params_t*)(arg)).id;
printf("Hello from %d\n", id);
/* Unlock and signal completion. */
pthread_mutex_unlock(&(*(params_t*)(arg)).mutex);
pthread_cond_signal (&(*(params_t*)(arg)).done);
/* After signalling `main`, the thread could actually
go on to do more work in parallel. */
}
int main() {
pthread_t threads[10];
params_t params;
pthread_mutex_init (&params.mutex , NULL);
pthread_cond_init (&params.done, NULL);
/* Obtain a lock on the parameter. */
pthread_mutex_lock (&params.mutex);
int i;
for(i = 0; i < 10; i++) {
/* Change the parameter (I own it). */
params.id = i;
/* Spawn a thread. */
pthread_create(&threads[i], NULL, hello, &params);
/* Give up the lock, wait till thread is 'done',
then reacquire the lock. */
pthread_cond_wait (&params.done, &params.mutex);
}
for(i = 0; i < 10; i++) {
pthread_join(threads[i], NULL);
}
/* Destroy all synchronization primitives. */
pthread_mutex_destroy (&params.mutex);
pthread_cond_destroy (&params.done);
return 0;
}
I see that the example you are trying is a toy program to probably learn about the POSIX thread library. In the real world, as we all know this can be done much faster without even using threads. But you already know this.
The problem is that you are modifying the params.id "unprotected" in main. This modification in main also needs to be mutex protected. You could protect this access by localizing this by creating getId() and setId() functions that would lock the mutex and protect access to the id, as follows. This will most likely still give the problem reported, since depending on when the thread calls getData() it will have one value or another. So to solve this, you could add an incrementId() function and call it from the hello() function.
struct params {
pthread_mutex_t mutex;
int id;
};
typedef struct params params_t;
int getId(params_t *p)
{
int id;
pthread_mutex_lock(&(p->mutex));
id = p->id;
pthread_mutex_unlock(&(p->mutex));
return id;
}
void setId(params_t *p, int val)
{
pthread_mutex_lock(&(p->mutex));
p->id = val;
pthread_mutex_unlock(&(p->mutex));
}
void incrementId(params_t *p)
{
    pthread_mutex_lock(&(p->mutex));
    p->id++;
    pthread_mutex_unlock(&(p->mutex));
}
void* hello(void* arg){
params_t *p = (params_t*)(arg);
incrementId(p);
int id = getId(p);
// This could possibly be quite messy since it
// could print the data for multiple threads at once
printf("Hello from %d\n", id);
}
int main() {
pthread_t threads[10];
params_t params;
params.id = 0;
pthread_mutex_init (&params.mutex , NULL);
int i;
for(i = 0; i < 10; i++) {
if(pthread_create(&threads[i], NULL, hello, &params));
}
for(i = 0; i < 10; i++) {
pthread_join(threads[i], NULL);
}
return 0;
}
A better way to get a unique thread id would be to define the hello method as follows:
void* hello(void* arg){
pthread_t threadId = pthread_self();
printf("Hello from %d\n", threadId);
}
And to avoid the problem with all threads trying to print at once, you could do the following:
void* hello(void* arg){
params_t *p = (params_t*)(arg);
    pthread_mutex_lock(&(p->mutex));
p->id++;
int id = p->id;
printf("Hello from %d\n", id);
    pthread_mutex_unlock(&(p->mutex));
}
Easiest way to get the desired output would be to modify your main function as follows:
int main() {
pthread_t threads[10];
params_t params;
pthread_mutex_init (&params.mutex , NULL);
int i;
for(i = 0; i < 10; i++) {
params.id = i;
if(pthread_create(&threads[i], NULL, hello, &params));
pthread_join(threads[i], NULL); //wait for thread to finish
}
/*for(i = 0; i < 10; i++) {
pthread_join(threads[i], NULL);
}*/
return 0;
}
Output would be:
Hello from 0
...
Hello from 9
EDIT: Here's the synchronization for the corrected question:
#include <pthread.h>
#include <stdio.h>
#include <stdlib.h>
struct params {
pthread_mutex_t* mutex;
int id;
};
typedef struct params params_t;
void* hello(void* arg){
int id = 0;
params_t* params = (params_t*)arg;
if(params != 0)
{
id = params->id;
delete params;
params = 0;
}
printf("Hello from %d\n", id);
}
int main() {
pthread_t threads[10];
params_t* params = 0;
pthread_mutex_t main_mutex;
pthread_mutex_init (&main_mutex , NULL);
int i;
for(i = 0; i < 10; i++) {
params = new params_t(); //create copy of the id to pass to each thread -> each thread will have it's own copy of the id
params->id = i;
params->mutex = &main_mutex;
if(pthread_create(&threads[i], NULL, hello, params));
}
for(i = 0; i < 10; i++) {
pthread_join(threads[i], NULL);
}
return 0;
}
Each thread must have it's own copy of the id so that the other threads do not modify the id before it is printed.
I'm just putting this one here to provide another solution to this problem - this one does not involve mutexes - no synchronization - no conditionals, etc.
The main dfference is that we are using pthread_detach to automatically release the thread's resources upon completion.
#include <pthread.h>
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#define NUMTHREADS 10
typedef struct params {
int id;
} params_t;
void* hello(void *arg)
{
params_t *p = (params_t*)arg;
int status;
status = pthread_detach(pthread_self());
if (status !=0 )
{
printf("detaching thread\n");
abort();
}
printf("Hello from %d\n", p->id);
free(p);
return NULL;
}
int main()
{
pthread_t thread;
params_t *par;
int i, status;
for (i=0; i<NUMTHREADS; i++)
{
par = (params_t*)malloc(sizeof(params_t));
if (par == NULL)
{
printf("allocating params_t");
abort();
}
par->id = i;
status = pthread_create(&thread, NULL, hello, par);
if (status != 0)
exit(1);
}
/* DO some more work ...*/
sleep(3);
exit(0);
}

Rejoining threads in C

In my main function, I spawn j threads which all compute the same task in parallel -- and then I want to wait for them to finish before exiting.
int main(...) {
// ...
int threads = 6;
pthread_t* thread = malloc(sizeof(pthread_t)*threads);
for(i = 0; i < threads; i++) {
struct thread_param *tp;
tp = malloc(sizeof(*tp));
// ...
int ret = pthread_create(&thread[i], NULL, &control, (void*)tp);
if(ret != 0) {
printf ("Create pthread error!\n");
exit (1);
}
}
for (j = 0; j < threads; j++) {
printf("JOINING THREAD: %i\n", j);
pthread_join( &thread[j], NULL);
}
exit(0);
}
However, nothing waits. Main just exits without ever completing the threaded tasks. Am I missing something?
hey, try pthread_join( thread[j], NULL); i think the problem is with types. I checked docs:
int pthread_join(pthread_t thread, void **value_ptr);
and thread is p_thread*, and thread[j] is p_thread, while &thread[j] is p_thread*, which is invalid. There just might internal error happen.
Edit: Yeah, I am very positive with that, pthread_t is basically int, so pthread_t* is accepted, it is just invalid thread handle, so pthread_join fails internally.

Resources