Re: Reducing the overhead on the network core

netdev

Re: Reducing the overhead on the network core

To:	Vicente Feito <vicente.feito@xxxxxxxxx>
Subject:	Re: Reducing the overhead on the network core
From:	"David S. Miller" <davem@xxxxxxxxxxxxx>
Date:	Mon, 28 Mar 2005 17:14:36 -0800
Cc:	netdev@xxxxxxxxxxx
In-reply-to:	<200503282144.08870.vicente.feito@xxxxxxxxx>
References:	<200503282144.08870.vicente.feito@xxxxxxxxx>
Sender:	netdev-bounce@xxxxxxxxxxx

We already do this and set the skb->packet_type field to
PACKET_HOST, PACKET_BROADCAST, PACKET_MULTICAST etc.
And the input paths test these values.

We still have to pass the packet into netif_rx() to deal
with network taps that want to see packets that are not
destined for us in any way.

[More with this subject...]

<Prev in Thread]	Current Thread	[Next in Thread>
Reducing the overhead on the network core, Vicente Feito Re: Reducing the overhead on the network core, David S. Miller <=

Previous by Date:	Re: cls_u32 compile failure in current 2.6.12-rc1+BK tree, Thomas Graf
Next by Date:	Re: [PATCH] rtnl_unlock/lock in sch_api.c TRY2, Thomas Graf
Previous by Thread:	Reducing the overhead on the network core, Vicente Feito
Next by Thread:	Re: [PATCH] s390: claw network device driver, Andrew Morton
Indexes:	[Date] [Thread] [Top] [All Lists]