Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 14 Jun 2013 14:07:44 +0400
From:      Gleb Smirnoff <glebius@FreeBSD.org>
To:        Luigi Rizzo <rizzo@iet.unipi.it>
Cc:        Ermal Lu?i <eri@freebsd.org>, freebsd-net <freebsd-net@freebsd.org>
Subject:   Re: [PATH] ALTQ(9) codel algorithm implementation
Message-ID:  <20130614100744.GS12443@glebius.int.ru>
In-Reply-To: <20130614100828.GA48119@onelab2.iet.unipi.it>
References:  <CAPBZQG0XDmybC5tzchFrcf3TZC7uJ0VcKSX-cVMJ%2BsMZUUqWVg@mail.gmail.com> <20130614095125.GQ12443@FreeBSD.org> <20130614100828.GA48119@onelab2.iet.unipi.it>

next in thread | previous in thread | raw e-mail | index | archive | help
  Luigi,

On Fri, Jun 14, 2013 at 12:08:28PM +0200, Luigi Rizzo wrote:
L> > I'm afraid we can't grow mbuf packet header with 8 bytes just to satisfy
L> > the ALTQ codel algo, which would definitely have a limited usage among
L> > FreeBSD users. Thus, "enqueue_time" should go into mbuf_tags(9) not into
L> > mbuf packet header.
L> 
L> not to take positions one way or the other, but getting and releasing
L> a tag on every packet is going to kill performance.

Does ALTQ care about performance?

L> If i remember well, 2-3 years ago at bsdcan there was discussion
L> (and mention of some pending work, jeffr maybe ?)
L> on providing some leading space in the mbuf so one could put there
L> tags (e.g. ipfw and dummynet ones) without having to allocate them.
L> Not sure where is this.

I even tried to prototype that.

-- 
Totus tuus, Glebius.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20130614100744.GS12443>