public inbox for linux-newbie@vger.kernel.org
 help / color / mirror / Atom feed
* Route lookup in ip_finish_output2
@ 2013-02-11  8:54 Prashant Batra
  0 siblings, 0 replies; only message in thread
From: Prashant Batra @ 2013-02-11  8:54 UTC (permalink / raw)
  To: linux-newbie

Hi Folks,

I am trying to do a small hack into the kernel just for some 
experimentation wherein I need to do a ipv4 route lookup from 
ip_finish_output2, after the packet is received from outside and has got 
forwarded.

The route lookup is done using kernel API "ip_route_output_flow" with 
the flow having the dst address.
This seems to be working fine with only exception that at very high 
loads, I see certain CPU hikes and the profiling results show that the 
same API comes out to be the culprit.

What I understand is that even to take the forwarding decision the 
kernel ip_rcv should have done the route lookup to decide that the 
packet needs to be forwarded, then why am I seeing a CPU hike in my case.

The second route lookup is done as the original packet has got 
encapsulated within a new IP header having a different destination than 
the original packet.


Thanks,
Prashant
--
To unsubscribe from this list: send the line "unsubscribe linux-newbie" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.linux-learn.org/faqs

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2013-02-11  8:54 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-02-11  8:54 Route lookup in ip_finish_output2 Prashant Batra

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox