blob: f0e7086b2919e0629da5fe00cf3a9e976d28b1d9 [file] [log] [blame]
/**
* @file
* This test burns some CPU cycles in user space and kernel space in order to
* get some non-zero values returned by times(2).
*/
#ifdef HAVE_CONFIG_H
# include "config.h"
#endif
#include <sched.h>
#include <stdio.h>
#include <time.h>
#include <unistd.h>
#include <sys/syscall.h>
#include <sys/times.h>
#include <sys/wait.h>
enum {
NUM_USER_ITERS = 1000000,
PARENT_CPUTIME_LIMIT_NSEC = 200000000,
CHILD_CPUTIME_LIMIT_NSEC = 300000000
};
int
main (void)
{
struct timespec ts;
volatile int dummy;
int i;
pid_t pid = fork();
if (pid < 0)
return 77;
const long cputime_limit =
pid ? PARENT_CPUTIME_LIMIT_NSEC : CHILD_CPUTIME_LIMIT_NSEC;
/* Enjoying my user time */
while (clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &ts) == 0) {
if (ts.tv_sec || ts.tv_nsec >= cputime_limit)
break;
for (i = 0; i < NUM_USER_ITERS; ++i)
++dummy;
}
/* Enjoying my system time */
while (clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &ts) == 0) {
if (ts.tv_sec || ts.tv_nsec >= cputime_limit * 2)
break;
sched_yield();
}
if (pid == 0) {
return 0;
} else {
wait(NULL);
}
struct tms tbuf;
unsigned long long llres;
/*
* On systems where user's and kernel's long types are the same,
* prefer direct times syscall over libc's times function because
* the latter is more prone to return value truncation.
*/
#if !defined __NR_times \
|| defined LINUX_MIPSN32 \
|| defined __x86_64__ && defined __ILP32__
clock_t res = times(&tbuf);
if ((clock_t) -1 == res)
return 77;
if (sizeof(res) < sizeof(unsigned long long))
llres = (unsigned long) res;
else
llres = res;
#else
long res = syscall(__NR_times, &tbuf);
if (-1L == res)
return 77;
else
llres = (unsigned long) res;
#endif
printf("times({tms_utime=%llu, tms_stime=%llu, ",
(unsigned long long) tbuf.tms_utime,
(unsigned long long) tbuf.tms_stime);
printf("tms_cutime=%llu, tms_cstime=%llu}) = %llu\n",
(unsigned long long) tbuf.tms_cutime,
(unsigned long long) tbuf.tms_cstime,
llres);
puts("+++ exited with 0 +++");
return 0;
}