* [LARTC] Inspection of SFQ work
@ 2001-12-20 22:02 Michael T. Babcock
2001-12-20 22:14 ` bert hubert
0 siblings, 1 reply; 2+ messages in thread
From: Michael T. Babcock @ 2001-12-20 22:02 UTC (permalink / raw)
To: lartc
Is there any way / patch allowing inspection of decisions made by various
qdiscs? I'm particularly interested in SFQ and RED but I'm sure following
the logic flow in CBQ would be enlightening as well ;-).
To the point, something that logs all the decisions made;
[timestamp] queued packet [header] [hashval]
[timestamp] dequeued packet [header] [hashval]
[timestamp] dequeued packet [header] [hashval]
... I'm sure it would slow things down, but there are some occasions where
turning this kind of thing on is 'smart'.
Just wondering
--
Michael T. Babcock
CTO, FibreSpeed Ltd. (Hosting, Security, Consultation, Database, etc)
http://www.fibrespeed.net/~mbabcock/
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://ds9a.nl/lartc/
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [LARTC] Inspection of SFQ work
2001-12-20 22:02 [LARTC] Inspection of SFQ work Michael T. Babcock
@ 2001-12-20 22:14 ` bert hubert
0 siblings, 0 replies; 2+ messages in thread
From: bert hubert @ 2001-12-20 22:14 UTC (permalink / raw)
To: lartc
On Thu, Dec 20, 2001 at 05:02:52PM -0500, Michael T. Babcock wrote:
> Is there any way / patch allowing inspection of decisions made by various
> qdiscs? I'm particularly interested in SFQ and RED but I'm sure following
> the logic flow in CBQ would be enlightening as well ;-).
I've thought about it. There is no easy single entrypoint to do this. Each
qdisc would have its own debugging statements. The good news is that this is
easy to add, just insert 'printk' statements in relevant places.
Kernel policy however is that no 'explanatory' statements are allowed in the
mainline kernel, just as there is no kernel debugger in the mainline kernel.
> ... I'm sure it would slow things down, but there are some occasions where
> turning this kind of thing on is 'smart'.
I'm not sure, but perhaps IBM 'dprobes' might be suited for this work.
Regards,
bert
--
http://www.PowerDNS.com Versatile DNS Software & Services
http://www.tk the dot in .tk
Netherlabs BV / Rent-a-Nerd.nl - Nerd Available -
Linux Advanced Routing & Traffic Control: http://ds9a.nl/lartc
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://ds9a.nl/lartc/
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2001-12-20 22:14 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-12-20 22:02 [LARTC] Inspection of SFQ work Michael T. Babcock
2001-12-20 22:14 ` bert hubert
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.