118 lines · plain
1========2CPU load3========4 5Linux exports various bits of information via ``/proc/stat`` and6``/proc/uptime`` that userland tools, such as top(1), use to calculate7the average time system spent in a particular state, for example::8 9 $ iostat10 Linux 2.6.18.3-exp (linmac) 02/20/200711 12 avg-cpu: %user %nice %system %iowait %steal %idle13 10.01 0.00 2.92 5.44 0.00 81.6314 15 ...16 17Here the system thinks that over the default sampling period the18system spent 10.01% of the time doing work in user space, 2.92% in the19kernel, and was overall 81.63% of the time idle.20 21In most cases the ``/proc/stat`` information reflects the reality quite22closely, however due to the nature of how/when the kernel collects23this data sometimes it can not be trusted at all.24 25So how is this information collected? Whenever timer interrupt is26signalled the kernel looks what kind of task was running at this27moment and increments the counter that corresponds to this tasks28kind/state. The problem with this is that the system could have29switched between various states multiple times between two timer30interrupts yet the counter is incremented only for the last state.31 32 33Example34-------35 36If we imagine the system with one task that periodically burns cycles37in the following manner::38 39 time line between two timer interrupts40 |--------------------------------------|41 ^ ^42 |_ something begins working |43 |_ something goes to sleep44 (only to be awaken quite soon)45 46In the above situation the system will be 0% loaded according to the47``/proc/stat`` (since the timer interrupt will always happen when the48system is executing the idle handler), but in reality the load is49closer to 99%.50 51One can imagine many more situations where this behavior of the kernel52will lead to quite erratic information inside ``/proc/stat``::53 54 55 /* gcc -o hog smallhog.c */56 #include <time.h>57 #include <limits.h>58 #include <signal.h>59 #include <sys/time.h>60 #define HIST 1061 62 static volatile sig_atomic_t stop;63 64 static void sighandler(int signr)65 {66 (void) signr;67 stop = 1;68 }69 70 static unsigned long hog (unsigned long niters)71 {72 stop = 0;73 while (!stop && --niters);74 return niters;75 }76 77 int main (void)78 {79 int i;80 struct itimerval it = {81 .it_interval = { .tv_sec = 0, .tv_usec = 1 },82 .it_value = { .tv_sec = 0, .tv_usec = 1 } };83 sigset_t set;84 unsigned long v[HIST];85 double tmp = 0.0;86 unsigned long n;87 signal(SIGALRM, &sighandler);88 setitimer(ITIMER_REAL, &it, NULL);89 90 hog (ULONG_MAX);91 for (i = 0; i < HIST; ++i) v[i] = ULONG_MAX - hog(ULONG_MAX);92 for (i = 0; i < HIST; ++i) tmp += v[i];93 tmp /= HIST;94 n = tmp - (tmp / 3.0);95 96 sigemptyset(&set);97 sigaddset(&set, SIGALRM);98 99 for (;;) {100 hog(n);101 sigwait(&set, &i);102 }103 return 0;104 }105 106 107References108----------109 110- https://lore.kernel.org/r/loom.20070212T063225-663@post.gmane.org111- Documentation/filesystems/proc.rst (1.8)112 113 114Thanks115------116 117Con Kolivas, Pavel Machek118