Subject: Re: multiple Tx queues
To: Erik E. Fair <firstname.lastname@example.org>
From: Jason Thorpe <email@example.com>
Date: 01/07/2004 17:47:02
Content-Type: text/plain; charset=US-ASCII; format=flowed
On Jan 7, 2004, at 5:31 PM, Erik E. Fair wrote:
> So, you're suggesting a driver-specific queueing mechanism based on
> TOS, rather than a more general packet queueing framework?
No. I suggested using packet tags (in a generic way) to identify the
type of service that a packet requires that the driver can use, if the
driver supports it.
Let's look at it another way...
Right now, we can support alternate packet queueing schemes using ALTQ.
Ok, fine. That allows packets to be re-ordered within the single
ifqueue that the interface driver gets to look at.
However, the point David was making is that that's not necessarily
sufficient. On wireless networks, there is even bandwidth reserved
between beacons for certain types of traffic. Even if this is the
first packet in the ifqueue that the driver sees, the driver still
needs to know if it should transmit the packet in a way that will use
that reserved bandwidth.
-- Jason R. Thorpe <firstname.lastname@example.org>
content-type: application/pgp-signature; x-mac-type=70674453;
content-description: This is a digitally signed message part
content-disposition: inline; filename=PGP.sig
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (Darwin)
-----END PGP SIGNATURE-----