Ce contenu n'est pas disponible dans la langue sélectionnée.
4.3.3. Determining Time Spent in Kernel and User Space
This section illustrates how to determine the amount of time any given thread is spending in either kernel or user-space.
thread-times.stp
#! /usr/bin/env stap
probe perf.sw.cpu_clock!, timer.profile {
// NB: To avoid contention on SMP machines, no global scalars/arrays used,
// only contention-free statistics aggregates.
tid=tid(); e=execname()
if (!user_mode())
kticks[e,tid] <<< 1
else
uticks[e,tid] <<< 1
ticks <<< 1
tids[e,tid] <<< 1
}
global uticks, kticks, ticks
global tids
probe timer.s(5), end {
allticks = @count(ticks)
printf ("%16s %5s %7s %7s (of %d ticks)\n",
"comm", "tid", "%user", "%kernel", allticks)
foreach ([e,tid] in tids- limit 20) {
uscaled = @count(uticks[e,tid])*10000/allticks
kscaled = @count(kticks[e,tid])*10000/allticks
printf ("%16s %5d %3d.%02d%% %3d.%02d%%\n",
e, tid, uscaled/100, uscaled%100, kscaled/100, kscaled%100)
}
printf("\n")
delete uticks
delete kticks
delete ticks
delete tids
}
thread-times.stp lists the top 20 processes currently taking up CPU time within a 5-second sample, along with the total number of CPU ticks made during the sample. The output of this script also notes the percentage of CPU time each process used, as well as whether that time was spent in kernel space or user space.
Example 4.13, “thread-times.stp Sample Output” contains a 5-second sample of the output for thread-times.stp:
Example 4.13. thread-times.stp Sample Output
tid %user %kernel (of 20002 ticks)
0 0.00% 87.88%
32169 5.24% 0.03%
9815 3.33% 0.36%
9859 0.95% 0.00%
3611 0.56% 0.12%
9861 0.62% 0.01%
11106 0.37% 0.02%
32167 0.08% 0.08%
3897 0.01% 0.08%
3800 0.03% 0.00%
2886 0.02% 0.00%
3243 0.00% 0.01%
3862 0.01% 0.00%
3782 0.00% 0.00%
21767 0.00% 0.00%
2522 0.00% 0.00%
3883 0.00% 0.00%
3775 0.00% 0.00%
3943 0.00% 0.00%
3873 0.00% 0.00%
