Date: Wed, 19 May 2004 23:34:46 -0400 (EDT) From: Mikhail Teterin <mi@corbulon.video-collage.com> To: current@FreeBSD.org Subject: a scheduling question Message-ID: <200405200334.i4K3YlGU027751@corbulon.video-collage.com>
next in thread | raw e-mail | index | archive | help
Here is a top's snapshot from a dual CPU machine. Two lame encoders compete for the first CPU, while the total idle time is 35.6%. Why is that? Because they are nice? Is niceness really supposed to allow for wasted CPU? Thanks! 5.2-CURRENT #0: Wed May 12 -mi last pid: 947; load averages: 1.69, 1.27, 1.25 up 1+23:07:07 23:30:41 144 processes: 3 running, 138 sleeping, 1 stopped, 2 zombie CPU states: 4.9% user, 51.2% nice, 7.0% system, 1.4% interrupt, 35.6% idle Mem: 461M Active, 235M Inact, 191M Wired, 48M Cache, 111M Buf, 62M Free Swap: 2048M Total, 288K Used, 2047M Free PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND 842 mi 139 10 3216K 2056K RUN 0 0:32 46.09% 46.09% lame 454 mi 139 10 3216K 2056K RUN 0 1:11 44.53% 44.53% lame 476 mi -8 10 5920K 5108K cbwait 1 0:08 7.03% 7.03% cdparanoi 411 mi -8 10 6664K 5952K cbwait 0 0:06 4.69% 4.69% cdparanoi 687 root 77 0 209M 162M select 1 210:44 3.12% 3.12% XFree86 743 root 95 18 3456K 1844K select 1 112:34 3.12% 3.12% fam 32537 mi 77 0 90424K 43896K select 1 64:32 3.12% 3.12% kdeinit [...]
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200405200334.i4K3YlGU027751>