Quoted from Operating Systems: Three Easy Pieces cpu-mechanisms Homework (Measurement) session :
Measuring the cost of a context switch is a little trickier. The lmbench benchmark does so by running two processes on a single CPU, and setting up two UNIX pipes between them; a pipe is just one of many ways processes in a UNIX system can communicate with one another. The first process then issues a write to the first pipe, and waits for a read on the second; upon seeing the first process waiting for something to read from the second pipe, the OS puts the first process in the blocked state, and switches to the other process, which reads from the first pipe and then writes to the second. When the second process tries to read from the first pipe again, it blocks, and thus the back-and-forth cycle of communication continues. By measuring the cost of communicating like this repeatedly, lmbench can make a good estimate of the cost of a context switch.
My question is , does the pipe itself must be accessed by 2 different process ? Or it is just in this case (and also in lmbench tool) it was used by different process in orer to measure context switch ? Here are some code I searched on github:
#define _GNU_SOURCE
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <fcntl.h>
#include <sys/time.h>
#include <sched.h>
int main(int argc, char *argv[]) {
// measure context switch
cpu_set_t set;
CPU_ZERO(&set);
CPU_SET(0, &set);
int first_pipefd[2], second_pipefd[2];
if (pipe(first_pipefd) == -1) {
perror("pipe");
exit(EXIT_FAILURE);
}
if (pipe(second_pipefd) == -1) {
perror("pipe");
exit(EXIT_FAILURE);
}
pid_t cpid = fork();
if (cpid == -1) {
perror("fork");
exit(EXIT_FAILURE);
} else if (cpid == 0) { // child
if (sched_setaffinity(getpid(), sizeof(cpu_set_t), &set) == -1) {
exit(EXIT_FAILURE);
}
for (size_t i = 0; i < nloops; i++) {
read(first_pipefd[0], NULL, 0);
write(second_pipefd[1], NULL, 0);
}
} else { // parent
if (sched_setaffinity(getpid(), sizeof(cpu_set_t), &set) == -1) {
exit(EXIT_FAILURE);
}
gettimeofday(&start, NULL);
for (size_t i = 0; i < nloops; i++) {
write(first_pipefd[1], NULL, 0);
read(second_pipefd[0], NULL, 0);
}
gettimeofday(&end, NULL);
printf("context switch: %f microseconds\n", (float) (end.tv_sec * 1000000 + end.tv_usec - start.tv_sec * 1000000 - start.tv_usec) / nloops);
}
return 0;
}