Subject: Re: bin/20316: systat interrupt counters
To: Martin Husemann <martin@duskware.de>
From: Johnny Billquist <bqt@update.uu.se>
List: netbsd-bugs
Date: 02/13/2003 00:29:52
On Thu, 13 Feb 2003, Martin Husemann wrote:

> On Wed, Feb 12, 2003 at 06:12:46PM +0000, David Laight wrote:
> 
> > Run systat in a wider window.  Both are actually allowed to be quite long.
> 
> Hmmmm ... "wider window".

:-)

On my VAX, it looks like with with my patch:

    4 users    Load  1.49  1.42  1.27                  Thu Feb 13 00:26:56

          memory totals (in kB)             PAGING   SWAPPING      Interrupts
         real   virtual    free             in  out   in  out       154 total
Active  20040     26088   17744     ops                                 uba0 int
All     31280     37328  211692     pages                               mtc0 int
                                                                     34 uda0 int
Proc:r  d  s  w    Csw   Trp   Sys  Int  Sof   Flt        forks      11 de0 intr
     1    14        45    20   165  145    9    14        fkppw         dmf0 ari
                                                          fksvm         dmf0 axi
  23.0% Sy  14.6% Us   0.0% Ni   1.1% In  61.3% Id        pwait         mtc1 int
|    |    |    |    |    |    |    |    |    |    |       relck         uda1 int
===========>>>>>>>>                                       rlkok       5 soft net
                                                          noram       4 soft clo
Namei         Sys-cache     Proc-cache                    ndcpy     100 clock in
    Calls     hits    %     hits     %                    fltcp
      867      550   63                                 1 zfod
                                                          cow
Discs  ra0  ra1  ra2  ra3  ra4  ra5                    64 fmin
seeks                                                  85 ftarg
xfers   34                                           3252 itarg
bytes  66K                                            460 wired
%busy 69.2                                                pdfre


> IMHO the output in 80 columns is rather useless/hard to decipher on i386
> too. But then there is no real solution, besides leaving out the interrupt
> details in this display.

Which is hardly an improvement. :-)
Anyhow, for me, 80 columns works rather nicely.

	Johnny

Johnny Billquist                  || "I'm on a bus
                                  ||  on a psychedelic trip
email: bqt@update.uu.se           ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol