small doubt about fair-scheduler patch
Hi,
Im using your fair scheduler patch (2.4.19-fairsched), applied on top of 2.4.20-rmap15a+read_latency2, and everything seems great. However, I have one small doubt, ps output seems a little strange, some processes appear to be out of order. To illustrate, here's the output of ps. Im wondering, is this normal?
bash-2.0.5b# ps -eo pid,ppid,user,command
PID PPID USER COMMAND
3 1 root [ksoftirqd_CPU0]
2 1 root [keventd]
5 1 root [kscand]
6 1 root [bdflush]
8 1 root [kreiserfsd]
410 1 root /sbin/dhcpcd eth0
551 1 root [eth1]
613 1 root klogd -x
627 1 root sshd
1 0 root init [3]
682 627 root sshd
685 684 kurt -bash
760 685 root su -
799 1 root xinetd
604 1 root syslogd -m 0
820 1 snmp /usr/sbin/snmpd
841 1 root logger -t safe_mysqld
840 1 root /bin/sh /usr/bin/safe_mysqld
870 840 mysql /usr/sbin/mysqld
4 1 root [kswapd]
894 885 httpd /usr/sbin/httpd -DSSL
893 885 httpd /usr/sbin/httpd -DSSL
885 1 root /usr/sbin/httpd -DSSL
898 885 httpd /usr/sbin/httpd -DSSL
680 1 root portsentry -audp
681 1 root portsentry -atcp
665 1 root crond
7 1 root [kupdated]
761 760 root -bash
684 682 kurt sshd
1003 761 root ps -eo pid,ppid,user,commandWithout the fairsched patch, ps output is normal, eg, init is always the first process listed, all kernel threads appear at the top (notice kswapd and kupdated near the bottom), and processes with lower pids never appear after others with higher (eg, portsentry is 680 and was started before httpd which is 898, yet it still appears after it). Im just wondering, is this purely aesthetically or is there something fishy?
Regards,
//kj
__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page from News and Sport to Email and Music Charts
http://uk.my.yahoo.com
Quoted 7 lines · Kurt Johnson
Im using your fair scheduler patch (2.4.19-fairsched),
Without the fairsched patch, ps output is normal, eg, init is always the first process listed,
Im just wondering, is this
purely aesthetically or is there something fishy?
The fairsched patch reorders processes on the tasklist so that the processes of a user all get CPU time alternately.
It's just a cosmetic issue, caused by the fact that procfs walks that same list to display all the tasks and now that list can get reordered ...
regards,
Rik
Signature · 3 lines
Bravely reimplemented by the knights who say "NIH". http://www.surriel.com/ http://guru.conectiva.com/ Current spamtrap: <a href=mailto:"[email protected]">[email protected]</a>