14966 – procps outputting incorrect CPU time/usage values

Bug 14966 - procps outputting incorrect CPU time/usage values

Summary: procps outputting incorrect CPU time/usage values

Status:	RESOLVED FIXED

Alias:	None

Product:	Gentoo Linux
Classification:	Unclassified
Component:	Current packages (show other bugs)
Hardware:	x86 Linux

Importance:	High normal (vote)
Assignee:	Brandon Low (RETIRED)

URL:
Whiteboard:
Keywords:

Depends on:
Blocks:

Reported:	2003-02-02 06:46 UTC by Décio Luiz Gazzoni Filho
Modified:	2003-03-01 15:33 UTC (History)
CC List:	1 user (show)

See Also:
Package list:
Runtime testing required:	---

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.

Description Décio Luiz Gazzoni Filho 2003-02-02 06:46:33 UTC

This is an example output of top on my machine, a dual Athlon 1533 MHz: 
 
  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  Command 
 8641 root      24  -1 26540  25m 1308 R 99.9  2.6  11047:39 gp 
24517 pcarmody  35  10  2564 2112  880 R 99.9  0.2  33231:39 pfgw-linux-2002 
16259 root      39  19 73048  71m  624 R 99.9  7.1  10:43.24 linesiever 
14932 root      39  19 73044  71m  624 R 59.5  7.1  33:42.55 linesiever 
16371 root      16   0   912  912  680 R 52.6  0.1   0:00.63 top 
 
By the way, the machine's current uptime is: 
 
 09:38:39 up 2 days, 21:32,  6 users,  load average: 4.03, 4.10, 4.10 
 
So obviously I couldn't be running processes for 23 days or so as indicated by top. 
 
Also, I'm not sure whether top is supposed to display total CPU usage/number of CPUs (i.e. all 
fields should sum to 100%) or just added CPU usage of both processors (should sum to 200%), 
but either way, this is summing to more than 400%. The command ps suffers from the same 
problems. 
 
I have reasons to believe this is related to the ``Set jiffies for i386 (HZ)'' kernel config option. 
Mine is set at 1000. 

Reproducible: Always
Steps to Reproduce:
1.
2.
3.



Portage 2.0.46-r9 (default-x86-1.4, gcc-3.2.1, glibc-2.3.1-r3) 
================================================================= 
System uname: 2.4.19-gentoo-r10 i686 AMD Athlon(tm) MP 1800+ 
GENTOO_MIRRORS="http://www.ibiblio.org/pub/Linux/distributions/gentoo" 
CONFIG_PROTECT="/etc /var/qmail/control /usr/kde/2/share/config /usr/kde/3.1/share/config 
/usr/kde/3/share/config /usr/share/config" 
CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d" 
PORTDIR="/usr/portage" 
DISTDIR="/usr/portage/distfiles" 
PKGDIR="/usr/portage/packages" 
PORTAGE_TMPDIR="/var/tmp" 
PORTDIR_OVERLAY="" 
USE="x86 oss 3dnow apm avi crypt cups encode gif jpeg libg++ mmx mpeg ncurses nls pdflib 
png qtmt quicktime spell truetype xml2 xmms xv zlib gdbm berkdb slang readline tetex aalib 
svga tcltk java postgres X sdl gpm tcpd pam libwww ssl perl python imlib oggvorbis gtk qt kde 
motif opengl mozilla cdr dvd ipv6 mule sse -gnome -mikmod -arts" 
COMPILER="gcc3" 
CHOST="i686-pc-linux-gnu" 
CFLAGS="-O3 -march=athlon-xp -pipe -fomit-frame-pointer -O3 " 
CXXFLAGS="-O3 -march=athlon-xp -pipe -fomit-frame-pointer -O3 " 
ACCEPT_KEYWORDS="x86 ~x86" 
MAKEOPTS="-j4" 
AUTOCLEAN="true" 
SYNC="rsync://rsync.gentoo.org/gentoo-portage" 
FEATURES="sandbox ccache"

Comment 1 Brandon Low (RETIRED) gentoo-dev

2003-02-02 15:31:16 UTC

Yes, this is due to the HZ value being set to 1000, upgrading to gentoo-sources-2.4.20-r1 will fix it as the variable HZ option has been removed from that patchset for the time being.  There is a newer variable HZ patch which theoretically will work correctly, and I am testing it for future kernels.

Comment 2 Brandon Low (RETIRED) gentoo-dev

2003-03-01 15:33:44 UTC

IF I'm wrong, reopen.