lkml.org 
[lkml]   [1999]   [Apr]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectWrong timing in Linux 2.2.6 with smp?
I noticed that some timings seem to be wrong in Linux 2.2.6 with SMP
compiled in.

<michael@rakete>:~>date ; sleep 10 ; date
Mon Apr 19 12:38:42 CEST 1999
Mon Apr 19 12:38:52 CEST 1999

These commands needed 55 seconds to finish! The clock is also going
wrong.

<michael@rakete>:~>uptime ; date
12:45pm up 49 min, 3 users, load average: 1.80, 1.36, 0.63
Mon Apr 19 12:45:35 CEST 1999

Now it is really 16:20 MEST. But I don't know if the computer clock was
OK on boot and if the given value (12:45 CEST) is MEST (Median European
Summer Time) or if only the BIOS clock is MEST and this value was
calculated wrong. If you need reliable information about the clock, ask
me and I will make some experiments to find out.


These things seem to be independent from the system load.
XOSVIEW sometimes shows me a cpu load of > 100%. These values are always
a multiple of 10.
When I load the system for testpurposes, I open some shells and type in:
<michael@michael>:~>while true; do true ; done
On my 486 I get a CPU=100% and LOAD=1.

On my Dual Pentium II with SMP-kernel with 4 shells doing such "work"
(to assure that the cpus are really 100% loaded) I get a load between
2.8 and 3.2 and CPU0=550% and CPU1=550%.
Strange, isn't it?
There seems to be a magic factor of 5.5

The sceduling seems to be slow, too. Sometimes the system needs some
time if you press ctrl-c to cancel a command.

If you get a beep (console or X), it sounds longer than usual.
If I set the beep in kpanel to 1000ms length, it lasts 11 seconds.


<michael@rakete>:~>uname -a
Linux rakete 2.2.6 #7 SMP Sun Apr 18 18:33:05 CEST 1999 i686 unknown
<michael@rakete>:~>

<michael@rakete>:~>cat work.cc
void main()
{

int a=0;
while (1) {

a=a+2;

}

}
<michael@rakete>:~>
Two processes of these program running give me a load of 2.0 if running
some time.
4 of them give me a load of 3.0

Greetings, Michael

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:51    [W:0.042 / U:0.168 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site