oxmc/libgtop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Do the math correctly to avoid loss of precision for rtime.

Browse Source 
I've noticed this with system-monitor where all processes got
a perfectly round CPU time to the second.
This commit is contained in:
Benoit Dejean
2015-07-18 15:08:08 +02:00
parent 6acebf416d
commit 3bd720ad81
1 changed files with 1 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
sysdeps/freebsd/proctime.c
View File

@@ -80,7 +80,6 @@ glibtop_get_proc_time_p (glibtop *server, glibtop_proc_time *buf,
glibtop_suid_leave (server);
buf->rtime = pinfo [0].ki_runtime * 1e-6;
len = sizeof (ci);
if (sysctlbyname ("kern.clockrate", &ci, &len, NULL, 0)) {
@@ -90,10 +89,9 @@ glibtop_get_proc_time_p (glibtop *server, glibtop_proc_time *buf,
}
buf->frequency = (ci.stathz ? ci.stathz : ci.hz);
buf->rtime = pinfo [0].ki_runtime * buf->frequency / 1000000;
buf->flags = _glibtop_sysdeps_proc_time;
buf->rtime *= buf->frequency;
if ((pinfo [0].ki_flag & PS_INMEM)) {
buf->utime = (pinfo [0].ki_runtime * 1e-6) * buf->frequency;
buf->stime = tv2sec (pinfo [0].ki_rusage.ru_stime) * buf->frequency;