lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  PHC 
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Tue, 26 Aug 2008 22:56:26 +1000
From:	Herbert Xu <>
To:	Stephen Hemminger <>
Cc:	David Miller <>,,,,
	Jussi Kivilinna <>,
	jamal <>
Subject: Re: [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock().

On Tue, Aug 26, 2008 at 08:50:29AM -0400, Stephen Hemminger wrote:
> Last time I tried TBF(100kbit) { netem(+100ms) } it gave different answers
> than netem(+100ms) { TBF(100kbit) }.  

In what ways were the answers different?

> I would prefer a peek() to the current dequeue/requeue.
> An alternative would be to have netem keep a parallel data structure with
> the time to send for all packets, but that would be assuming the underlying
> qdisc's were work conserving.

The peek() interface isn't really appliacable for netem since the
packet that it's requeueing wasn't dequeued in the first place.

In any case, what I'm trying to say is that netem should really
have its own queue (e.g., just fold tfifo in) to implement the
reordering and delays.

This does not prevent the user from creating children of netem
such as TBF to simulate a network environment where you have
loss/delay/jitter after traffic goes through a shaper.

Visit Openswan at
Email: Herbert Xu ~{PmV>HI~} <>
Home Page:
PGP Key:
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to
More majordomo info at

Powered by blists - more mailing lists