[ovs-dev] Megaflows Question

[openvswitcher]

>
>If the OpenFlow table changes, the kernel flow cache has to be
>reexamined and sometimes updated.

You mean the revalidate thread ? I read the source and I thought the thread would only delete the flow in kernel
when the timeout. So I don't think the kernel flow will be updated if I add the rule for '9.8.8.8' unless the flow for 
'8.8.8.8' timeout which will cause new flow-miss.
Could you tell me more about it?







At 2015-09-09 23:46:12, "Ben Pfaff" <blp at nicira.com> wrote:
>On Wed, Sep 09, 2015 at 02:34:43PM +0800, openvswitcher wrote:
>> But I don't know why the 'xxxxxx0x.xxxxxxxx.xxxxxxxx.xxxxxxxx' is calculated as the result.
>> Why is it not "xxxxxxxx.xxxx0xxx.xxxxxxxx.xxxxxxxx" which can also be used to differ the
>> 00001010.00000000.(=10.0) from 00001000.00001000(=8.8) ?
>
>The algorithm is described in our NSDI 2015 paper:
>https://www.usenix.org/conference/nsdi15/technical-sessions/presentation/pfaff
>
>> And another problem is:
>>  in the situation '0.0.0.0/2.0.0.0'  is used, if I add another flow 'nw_dst=9(bit=00001001).8.8.8 action output:200'
>> then the actions will be not same with 'nw_dst=8.8.8.8' but the flow installed in kernel will make the packet 
>> of which the ip destination is 9.8.8.8 not be upcalled to userspace by flow-miss, it will use the flow for '8.8.8.8'
>>  in kernel. Is this right?
>
>If the OpenFlow table changes, the kernel flow cache has to be
>reexamined and sometimes updated.