LARTC - Apr 2003 - [patch] tcng tcsim_filter fix for tcsim_plot -t delay

Recently I played with the tcng-9d and found that tcsim_filter script
does produce broken output that can''t be used with tcsim_plot -t delay.
As far as I see, tcsim_plot use remaining part (after "packet length"
field) of input string to identify enqueued/dequeued packets, but
tcsim_filter does not produce it. So here is the pacth for tcsim_filter
to produce unique packet identifier in the last output field based on
simulated skbuff address (that _is unique_ for any enqueued packet in
the simulated qdisc).

Dimitry.

--- tcsim_filter.orig	Fri Sep 20 04:14:21 2002
+++ tcsim_filter	Sun Apr 27 16:14:03 2003
@@ -167,7 +167,7 @@
     }
     $stat{$id}++;
     if (!$count) {
-	print "$1 $id $4\n";
+	print "$1 $id $4 $3\n";
     }
 }

_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

Dimitry V. Ketov wrote:
> Recently I played with the tcng-9d and found that tcsim_filter script
> does produce broken output that can''t be used with tcsim_plot -t
delay.
Aahh ! I dropped that ID in tcng-8x, because I had forgotten that
it was used for the delay calculation, and thought it may just be
some old debugging left-over. (And had the nagging feeling that I
might have overlooked something since ...)

I''ll include your fix in the next release. Thanks a lot !

- Werner

-- 
  _________________________________________________________________________
 / Werner Almesberger, Buenos Aires, Argentina         wa@almesberger.net /
/_http://www.almesberger.net/____________________________________________/
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/