Damjan,

                Thanks for Your answer. One more question is whether 
worker-handoff is production grade code?
                I have run worker-handoff for l2bd use case successfully, 
however, the IPsec use case in transport mode fails.
All the packets are dropped after Ethernet-input graph node is done on the 
second worker core.

Regards
Zhiyong
From: vpp-dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf Of Damjan 
Marion via Lists.Fd.Io
Sent: Wednesday, February 20, 2019 6:23 PM
To: Yang, Zhiyong <zhiyong.y...@intel.com>
Cc: vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] One question about "worker-handoff" graph node.




On 20 Feb 2019, at 09:49, Zhiyong Yang 
<zhiyong.y...@intel.com<mailto:zhiyong.y...@intel.com>> wrote:

Hi VPP guys,
                There exists a “handoff” graph node  in VPP that is named 
“worker-handoff”. From my understanding, it looks to be  acting as a 
distributing packets received role in the multi thread working environment , 
and distribute pkts by hash algo to 1 or more worker threads.  Right?

y, it is basically software RSS....


One question is that which use cases can demostrate how  the graph node is 
applied in VPP.  If have any, please show me.

see "set interface handoff" cli...


I know that nat also use similar handoff  cases. And if NAT is using this 
“worker-handoff” graph node or not. If not, what’s difference between these 
handoff graph nodes?

worker-handoff node does handoff decision based on basic l2-l3-l4 hash, nat 
handoff node does handoff decision based on NAT session locality.



                BTW, I mainly intend to study core to core applications by this 
for multiple-thread working model in VPP. Appreciate your input in advance.

--
Damjan

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12302): https://lists.fd.io/g/vpp-dev/message/12302
Mute This Topic: https://lists.fd.io/mt/29938708/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to