why packet is duplicated dev_queue_xmit

To:	netdev@xxxxxxxxxxx
Subject:	why packet is duplicated dev_queue_xmit_nit?
From:	linux lover <linux.lover2004@xxxxxxxxx>
Date:	Thu, 23 Dec 2004 22:28:16 +0530
Domainkey-signature:	a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:mime-version:content-type:content-transfer-encoding; b=aup8t7Ohvd2fUXvquRcA5bVMe3GO4xMHfPZg72sYw1mLt2/Tjzac5Xb8bl7EopDsXiNm6nmybXUolMng0J8i1arxNGnIIeHFIbLAzB366vajR5WwnA+9tmhmDrfal98mbEjIYOfDiV7qG6sbmF7+wgBOGscpP4aJxULNPaRI0Bs=
Reply-to:	linux lover <linux.lover2004@xxxxxxxxx>
Sender:	netdev-bounce@xxxxxxxxxxx

Hello all,
     I want to know what is use of dev_queue_xmit_nit function in
dev.c file? Also why it calls following statement
                      ptype->func(skb2, skb->dev, ptype);
Also why skb2 is created by cloning skb? Acually i trace TCP packets
and found that control goes from neigh_resolve_output to directly
dev_queue_xmit_nit and then to HW driver 8139too.c? I want to know why
it not goes from dev_queue_xmit? I place printk statments and found
after dev_queue_xmit_nit control moves to network interface driver?
Help to understand the control  packet path.
Thanks in advance.
linux_lover

<Prev in Thread]	Current Thread	[Next in Thread>
why packet is duplicated dev_queue_xmit_nit?, linux lover <=

Previous by Date:	Re: LLTX and netif_stop_queue, Patrick McHardy
Next by Date:	Re: LLTX and netif_stop_queue, Eric Lemoine
Previous by Thread:	device driver problem, cranium2003
Next by Thread:	[RFC] Extended Generic Packet Classifier, Thomas Graf
Indexes:	[Date] [Thread] [Top] [All Lists]

why packet is duplicated dev_queue_xmit_nit?