Lines Matching full:idle
45 uint64_t idle;
61 "\t idle %llu\n",counters[instance].idle);
144 counters[cpu_num].idle = knp->value.ui64;
294 counters is not going to use a comparison to an idle rate. so, the
352 spent in user, idle and kernel. We also have nanoseconds spent
356 idle will also include time spent in interrupt. for netperf's
358 but we certainly do care for idle. the $64B question becomes -
361 we could just subtract interrupt time from idle. that has the
366 time from each of user, kernel and idle in proportion to their
367 numbers. then we sum the corrected user, kernel and idle along
368 with the interrupt time and use that to calculate a new idle
374 more interrupt time than the sum of user, kernel and idle.
383 delta_cpu_counters[i].idle = ending_cpu_counters[i].idle -
384 starting_cpu_counters[i].idle;
398 delta_cpu_counters[i].idle +
419 "\t delta[%d].idle %llu\n",i,delta_cpu_counters[i].idle);
436 (delta_cpu_counters[i].idle * CALC_ACCURACY) / total_cpu_nsec;
470 corrected_cpu_counters[i].idle = delta_cpu_counters[i].idle -
495 corrected_cpu_counters[i].idle +
510 (corrected_cpu_counters[i].idle * CALC_ACCURACY) / total_cpu_nsec;
