On Fri, Oct 10, 2014 at 11:36:39AM +0530, Shashwat Srivastava wrote:
> We are working on the implementation on eviction mechanism on basis of 
> importance as per openflow specs-1.4. We have added importance parameter 
> in the flow entry. And are going through various commands (replace-flows, 
> diff-flows) for enhancement.
> 
> We wanted to confirm about the "ovs-ofctl diff-flows br0 <Filename>" 
> command gives the output with both signs (plus + & minus -) for flows 
> which are on bridge and for flows on the file as per the case.

I don't understand that sentence.

> When we are mentioning openflow13 in the command as in "ovs-ofctl -O 
> openflow13 diff-flows br0 <Filename>", then the output received contain 
> only the flows written in the file with a plus or minus sign i.e. it is 
> dumping only the flows of the file and not compairing with the flows on 
> the switch which it should ideally do.

I don't fully understand that sentence, either, but it sounds like you
are reporting a bug.  I appreciate bug fixes.

> Query 1: What is the significance by mentioning openflow13 in the 
> ovs-ofctl -O openflow13 diff-flows br0 <Filename>.

This instructs ovs-ofctl to use OpenFlow 1.3 to talk to the switch.

> Query 2: When we are adding a flow using default standard (that is not 
> including importance in the flow) and adding the flow in the file having 
> importance, then using "ovs-ofclt diff-flows br0 <filename>" we are 
> getting difference in each flow as the file contains importance and the 
> flow in switch does not. This is because we are not adding importance 
> field in the structure ofp10_flow_mod, as discussed with you earlier. 
> So the question is after implementing these changes how to make the test 
> case for the same, meaning as the output is having difference so shall 
> compare the output with the ones we are receiving after doing changes of 
> importance.

If the flows in one place have importance (other than the default
importance) and those in another place do not, then I would expect
every flow to differ.
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev

Reply via email to